diff options
| author | YurenHao0426 <Blackhao0426@gmail.com> | 2026-08-29 12:26:11 -0500 |
|---|---|---|
| committer | YurenHao0426 <Blackhao0426@gmail.com> | 2026-08-29 12:26:17 -0500 |
| commit | 903f77bbf11b20011abc70645b06ccc0f669bfdf (patch) | |
| tree | ca4d27bfc51ee2e034f0b4eba5f914583490f6b5 /results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json | |
| parent | 2b9caf80d9a1054ec585a03718fc37106f8c7aae (diff) | |
results: record raw KP traffic scaling collapse
Diffstat (limited to 'results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json')
| -rw-r--r-- | results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json | 7472 |
1 files changed, 7472 insertions, 0 deletions
diff --git a/results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json b/results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json new file mode 100644 index 0000000..8d14aaf --- /dev/null +++ b/results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json @@ -0,0 +1,7472 @@ +{ + "architecture": { + "adaptive_apical_parameters": 606208, + "adaptive_feedback_parameters": 461440, + "base_width": 16, + "blocks_per_stage": 5, + "bn_eps": 1e-05, + "bn_momentum": 0.1, + "depth": 32, + "family": "CIFAR 6n+2 ResNet, option-A shortcuts", + "fixed_feedback_parameters": 0, + "fixed_traffic_coefficients": 303104, + "forward_parameters": 464154, + "hidden_shapes": [ + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 16, + 32, + 32 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 32, + 16, + 16 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ], + [ + 64, + 8, + 8 + ] + ], + "normalization": "batchnorm", + "predictor_parameters": 606208, + "residual_scale": 1.0, + "vectorizer_mode": null, + "vectorizer_parameters": 0 + }, + "args": { + "a_scale": 1.0, + "a_warmup_steps": 0, + "alignment_probe": 32, + "apical_calibration_mode": "unit_targets", + "apical_seed": null, + "augment_train": 1, + "batch_size": 128, + "bn_eps": 1e-05, + "bn_momentum": 0.1, + "data_dir": "/home/yurenh2/sdrn/data", + "depth": 32, + "device": "cuda", + "epochs": 200, + "eta_A": 0.01, + "eta_P": 0.1, + "eval_every": 1, + "eval_split": "validation", + "learn_P": 1, + "loader_seed": 0, + "lr": 0.1, + "lr_gamma": 0.1, + "lr_milestones": "100,150", + "lr_schedule": "step", + "max_steps": 0, + "mirror_batch_size": 1, + "mirror_eta": 0.1, + "mirror_every": 16, + "mirror_noise_std": 1.0, + "mirror_seed": 3000, + "mirror_warmup_steps": 0, + "mode": "kp_traffic", + "momentum": 0.9, + "neutral_projection": 1, + "normalization": "batchnorm", + "nuisance_scale": 0.0, + "out": "/scratch/yurenh2/sdil-raw-kp-a6000-r1/results/kp_raw_traffic_scaling/r1/raw-kp-traffic4-r1-d32.json", + "output_lr": 0.1, + "pert_directions": 1, + "pert_every": 4, + "pert_sigma": 0.01, + "perturb_seed": 1000, + "predictor_every": 0, + "predictor_mode": "closed_form", + "predictor_warmup_steps": 1, + "residual_scale": null, + "seed": 0, + "split_seed": 2027, + "traffic_calibration_examples": 64, + "traffic_ratio": 4.0, + "traffic_rule": "raw", + "traffic_seed": 5000, + "train_limit": 0, + "use_residual": 1, + "val_examples": 5000, + "vectorizer_mode": "spatial_template", + "warmup_epochs": 5, + "weight_decay": 0.0001, + "weight_scale": 1.0, + "width": 16 + }, + "calibration_metric_space": "reciprocal_local_activity_products_with_mixed_apical_traffic", + "counters": { + "apical_warmup_examples": 0, + "calibration_event_examples": 0, + "causal_scalar_observations": 0, + "logical_batch_loss_queries": 0, + "mirror_conv_examples": 0, + "mirror_events": 0, + "mirror_readout_examples": 0, + "neutral_projection_examples": 9000000, + "ordinary_examples": 9000000, + "per_example_loss_terms": 0, + "perturbation_events": 0, + "perturbation_forward_examples": 0, + "predictor_update_examples": 64, + "predictor_warmup_examples": 0, + "traffic_audit_examples": 0, + "traffic_calibration_examples": 64 + }, + "diagnostics": { + "early_third_mean": NaN, + "feedback_forward_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "feedback_forward_norm_ratio": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "feedback_forward_relative_error": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "innovation_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "instruction_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "matched_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "max_norm_match_direction_error": NaN, + "max_norm_match_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "neutral_projection": { + "instruction_observations": 0, + "max_absolute_correction_slope": 0.0, + "max_absolute_post_projection_soma_slope": 0.0, + "max_absolute_pre_projection_soma_slope": 0.0, + "max_positive_post_projection_soma_slope": 0.0, + "min_post_projection_soma_slope": 0.0, + "observations": 32, + "post_projection_traffic_rms_ratio": NaN, + "pre_projection_traffic_rms_ratio": NaN + }, + "normalization_state": "training_batch_stats_without_running_update", + "predictor_traffic_residual_rms_ratio": NaN, + "raw_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "teaching_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "traffic_instruction_rms_ratio": NaN, + "used_negative_gradient_cosine": [ + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN, + NaN + ], + "wall_s": 0.14020562171936035 + }, + "epochs": [ + { + "epoch": 1, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.02, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.125, + "maximum_absolute_post_projection_soma_slope": 0.0017428398132324219, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": 3.027403427875087e-08, + "maximum_pre_projection_traffic_rms_ratio": 0.006192635978101392, + "minimum_observations": 72 + }, + "output_lr": 0.02, + "step": 352, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 2, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.04, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.04, + "step": 704, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 3, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.06000000000000001, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.06000000000000001, + "step": 1056, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 4, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.08, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.08, + "step": 1408, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 5, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 1760, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 6, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 2112, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 7, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 2464, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 8, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 2816, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 9, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 3168, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 10, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 3520, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 11, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 3872, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 12, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 4224, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 13, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 4576, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 14, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 4928, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 15, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 5280, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 16, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 5632, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 17, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 5984, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 18, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 6336, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 19, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 6688, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 20, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 7040, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 21, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 7392, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 22, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 7744, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 23, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 8096, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 24, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 8448, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 25, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 8800, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 26, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 9152, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 27, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 9504, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 28, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 9856, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 29, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 10208, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 30, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 10560, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 31, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 10912, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 32, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 11264, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 33, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 11616, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 34, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 11968, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 35, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 12320, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 36, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 12672, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 37, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 13024, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 38, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 13376, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 39, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 13728, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 40, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 14080, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 41, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 14432, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 42, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 14784, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 43, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 15136, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 44, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 15488, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 45, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 15840, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 46, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 16192, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 47, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 16544, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 48, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 16896, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 49, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 17248, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 50, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 17600, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 51, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 17952, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 52, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 18304, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 53, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 18656, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 54, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 19008, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 55, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 19360, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 56, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 19712, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 57, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 20064, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 58, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 20416, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 59, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 20768, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 60, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 21120, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 61, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 21472, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 62, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 21824, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 63, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 22176, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 64, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 22528, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 65, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 22880, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 66, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 23232, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 67, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 23584, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 68, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 23936, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 69, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 24288, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 70, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 24640, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 71, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 24992, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 72, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 25344, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 73, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 25696, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 74, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 26048, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 75, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 26400, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 76, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 26752, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 77, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 27104, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 78, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 27456, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 79, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 27808, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 80, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 28160, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 81, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 28512, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 82, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 28864, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 83, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 29216, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 84, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 29568, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 85, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 29920, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 86, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 30272, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 87, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 30624, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 88, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 30976, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 89, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 31328, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 90, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 31680, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 91, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 32032, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 92, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 32384, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 93, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 32736, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 94, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 33088, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 95, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 33440, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 96, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 33792, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 97, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 34144, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 98, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 34496, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 99, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 34848, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 100, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.1, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.1, + "step": 35200, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 101, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 35552, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 102, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 35904, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 103, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 36256, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 104, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 36608, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 105, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 36960, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 106, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 37312, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 107, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 37664, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 108, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 38016, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 109, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 38368, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 110, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 38720, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 111, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 39072, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 112, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 39424, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 113, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 39776, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 114, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 40128, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 115, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 40480, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 116, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 40832, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 117, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 41184, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 118, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 41536, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 119, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 41888, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 120, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 42240, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 121, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 42592, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 122, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 42944, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 123, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 43296, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 124, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 43648, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 125, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 44000, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 126, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 44352, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 127, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 44704, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 128, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 45056, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 129, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 45408, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 130, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 45760, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 131, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 46112, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 132, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 46464, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 133, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 46816, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 134, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 47168, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 135, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 47520, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 136, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 47872, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 137, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 48224, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 138, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 48576, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 139, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 48928, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 140, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 49280, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 141, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 49632, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 142, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 49984, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 143, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 50336, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 144, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 50688, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 145, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 51040, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 146, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 51392, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 147, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 51744, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 148, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 52096, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 149, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 52448, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 150, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.010000000000000002, + "step": 52800, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 151, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 53152, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 152, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 53504, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 153, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 53856, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 154, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 54208, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 155, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 54560, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 156, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 54912, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 157, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 55264, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 158, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 55616, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 159, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 55968, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 160, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 56320, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 161, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 56672, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 162, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 57024, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 163, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 57376, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 164, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 57728, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 165, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 58080, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 166, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 58432, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 167, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 58784, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 168, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 59136, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 169, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 59488, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 170, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 59840, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 171, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 60192, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 172, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 60544, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 173, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 60896, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 174, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 61248, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 175, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 61600, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 176, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 61952, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 177, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 62304, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 178, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 62656, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 179, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 63008, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 180, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 63360, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 181, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 63712, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 182, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 64064, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 183, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 64416, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 184, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 64768, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 185, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 65120, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 186, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 65472, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 187, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 65824, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 188, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 66176, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 189, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 66528, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 190, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 66880, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 191, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 67232, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 192, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 67584, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 193, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 67936, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 194, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 68288, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 195, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 68640, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 196, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 68992, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 197, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 69344, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 198, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 69696, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 199, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 70048, + "train_examples": 45000, + "train_loss": NaN + }, + { + "epoch": 200, + "eval_accuracy": 0.1, + "eval_loss": NaN, + "eval_split": "validation", + "feedback_tracking": { + "max_feedback_forward_relative_error": NaN, + "mean_feedback_forward_cosine": NaN, + "mean_feedback_forward_relative_error": NaN, + "min_feedback_forward_cosine": NaN + }, + "lr": 0.0010000000000000002, + "mixed_apical": { + "innovation_rms": NaN, + "instruction_rms": NaN, + "raw_apical_rms": NaN, + "teaching_rms": NaN, + "traffic_rms": NaN + }, + "neutral_projection": { + "instruction_observations": 0, + "maximum_absolute_correction_slope": 0.0, + "maximum_absolute_post_projection_soma_slope": 0.0, + "maximum_observations": 128, + "maximum_post_projection_traffic_rms_ratio": NaN, + "maximum_pre_projection_traffic_rms_ratio": NaN, + "minimum_observations": 72 + }, + "output_lr": 0.0010000000000000002, + "step": 70400, + "train_examples": 45000, + "train_loss": NaN + } + ], + "evaluation_protocol": { + "test_evaluations": 0, + "test_used_for_selection": false, + "validation_evaluations": 201 + }, + "final": { + "accuracy": 0.1, + "epoch": 200, + "evaluation_split": "validation", + "finite": false, + "loss": NaN, + "step": 70400 + }, + "hardware": { + "cuda_device_name": "NVIDIA RTX A6000", + "cuda_visible_devices": "3", + "device": "cuda", + "device_total_memory_bytes": 51041271808, + "peak_memory_allocated_bytes": 2328666112, + "peak_memory_reserved_bytes": 3525312512, + "torch_version": "2.10.0+cu128" + }, + "predictor_warmup": { + "closed_form_fit": { + "max_absolute_residual_soma_slope": 6.300626154143174e-08, + "max_applied_stability_margin": 0.0, + "max_positive_residual_soma_slope": 3.633729050989132e-08, + "min_residual_soma_slope": -6.300626154143174e-08, + "mse": 5.190107862186738e-18, + "observations": 64, + "residual_traffic_rms_ratio": 7.191502243931076e-08, + "stability_margin": 0.0 + }, + "examples": 64, + "instruction_present": false, + "mode": "closed_form", + "post_warmup_traffic_residual_rms_ratio": 7.191502243931076e-08, + "reuses_traffic_calibration_forward": true, + "steps": 1, + "task_loader_state_restored": true + }, + "protocol_family": "oral_a_cifar_local_resnet_development", + "provenance": { + "git_commit": "2b9caf80d9a1054ec585a03718fc37106f8c7aae", + "git_tracked_dirty": false + }, + "schema_version": 1, + "split": { + "cifar_source_files": [ + { + "bytes": 31035704, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/data_batch_1", + "sha256": "54636561a3ce25bd3e19253c6b0d8538147b0ae398331ac4a2d86c6d987368cd" + }, + { + "bytes": 31035320, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/data_batch_2", + "sha256": "766b2cef9fbc745cf056b3152224f7cf77163b330ea9a15f9392beb8b89bc5a8" + }, + { + "bytes": 31035999, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/data_batch_3", + "sha256": "0f00d98ebfb30b3ec0ad19f9756dc2630b89003e10525f5e148445e82aa6a1f9" + }, + { + "bytes": 31035696, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/data_batch_4", + "sha256": "3f7bb240661948b8f4d53e36ec720d8306f5668bd0071dcb4e6c947f78e9682b" + }, + { + "bytes": 31035623, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/data_batch_5", + "sha256": "d91802434d8376bbaeeadf58a737e3a1b12ac839077e931237e0dcd43adcb154" + }, + { + "bytes": 31035526, + "path": "/home/yurenh2/sdrn/data/cifar-10-batches-py/test_batch", + "sha256": "f53d8d457504f7cff4ea9e021afcf0e0ad8e24a91f3fc42091b8adef61157831" + } + ], + "dataset": "cifar10", + "evaluation_split": "validation", + "input_layout": "NCHW", + "input_shape": [ + 3, + 32, + 32 + ], + "loader_seed": 0, + "normalization_mean": [ + 0.49140000343322754, + 0.4821999967098236, + 0.4465000033378601 + ], + "normalization_std": [ + 0.24699999392032623, + 0.2434999942779541, + 0.26159998774528503 + ], + "split_from_training_only": true, + "split_seed": 2027, + "test_examples": 10000, + "train_examples": 45000, + "training_augmentation": "random_crop_32_padding_4_zero_then_horizontal_flip_p0.5", + "validation_class_counts": { + "0": 500, + "1": 500, + "2": 500, + "3": 500, + "4": 500, + "5": 500, + "6": 500, + "7": 500, + "8": 500, + "9": 500 + }, + "validation_examples": 5000, + "validation_index_sha256": "8328b206a97c420e49e54e3eca4abe3274c4756b084355784ea3fb8059e4515b" + }, + "timing": { + "apical_warmup_wall_s": 0.0, + "evaluation_wall_s": 32.23249626159668, + "mirror_warmup_wall_s": 0.0, + "predictor_warmup_wall_s": 0.03771519660949707, + "timing_excludes_data_loading_hashing_and_model_construction": true, + "total_timed_wall_s": 4973.364273548126, + "train_wall_s": 4939.011425495148 + }, + "traffic_calibration": { + "data_source": "first unaugmented development-training examples", + "examples": 64, + "instruction_rms": [ + 0.02028265967965126, + 0.015888990834355354, + 0.009587029926478863, + 0.010533088818192482, + 0.006491520442068577, + 0.007582552265375853, + 0.004749298561364412, + 0.00565798394382, + 0.003715394763275981, + 0.004712278954684734, + 0.0028127566911280155, + 0.005799341015517712, + 0.004073382820934057, + 0.004938906989991665, + 0.0032066938001662493, + 0.004145961254835129, + 0.0026535307988524437, + 0.0034991821739822626, + 0.0022640512324869633, + 0.0030579636804759502, + 0.0020110998302698135, + 0.004513302817940712, + 0.0031777797266840935, + 0.004072114359587431, + 0.0027934671379625797, + 0.003359902184456587, + 0.0025724514853209257, + 0.0031278119422495365, + 0.0024165017530322075, + 0.0028539455961436033, + 0.0024009260814636946 + ], + "realized_traffic_instruction_rms_ratio": [ + 4.0, + 4.0, + 4.0, + 4.0, + 4.000000476837158, + 3.999999761581421, + 4.0, + 4.0, + 4.000000476837158, + 4.0, + 4.000000476837158, + 4.0, + 4.0, + 4.0, + 4.0, + 4.0, + 4.0, + 4.0, + 4.0, + 4.0, + 3.999999523162842, + 4.0, + 4.0, + 4.0, + 4.0, + 3.999999761581421, + 4.0, + 3.999999761581421, + 4.0, + 4.0, + 3.999999523162842 + ], + "target_ratio": 4.0, + "traffic_gain": [ + 0.10581493377685547, + 0.0858360305428505, + 0.03321487456560135, + 0.05464927479624748, + 0.017319533973932266, + 0.04140952602028847, + 0.010942637920379639, + 0.029760098084807396, + 0.007737674750387669, + 0.025834038853645325, + 0.0053402637131512165, + 0.030604055151343346, + 0.009308629669249058, + 0.026025518774986267, + 0.006474803201854229, + 0.021557938307523727, + 0.004869470838457346, + 0.01855599135160446, + 0.0038137801457196474, + 0.015923958271741867, + 0.0030991104431450367, + 0.024264555424451828, + 0.006372333038598299, + 0.021969744935631752, + 0.005072691012173891, + 0.01743658073246479, + 0.00426547322422266, + 0.017027076333761215, + 0.003758358070626855, + 0.015189743600785732, + 0.003484738990664482 + ], + "unscaled_traffic_rms": [ + 0.7667220234870911, + 0.7404345870018005, + 1.1545466184616089, + 0.7709590792655945, + 1.499236822128296, + 0.7324451804161072, + 1.7360708713531494, + 0.760479211807251, + 1.920677661895752, + 0.7296232581138611, + 2.106829881668091, + 0.7579833269119263, + 1.7503684759140015, + 0.7590867877006531, + 1.9810293912887573, + 0.7692685723304749, + 2.1797282695770264, + 0.7542970180511475, + 2.374600648880005, + 0.7681416273117065, + 2.595712423324585, + 0.7440157532691956, + 1.9947354793548584, + 0.7414040565490723, + 2.202749729156494, + 0.7707708477973938, + 2.4123480319976807, + 0.7347854375839233, + 2.5718696117401123, + 0.7515454292297363, + 2.755932092666626 + ], + "uses_validation_endpoint": false + }, + "work": { + "apical_macs_per_example": 68420224, + "causal_scalar_observations": 0, + "components": { + "apical_projection_macs": 615786394894336, + "apical_regression_macs": 0, + "bp_reverse_macs_estimate": 0, + "kp_reciprocal_correlation_macs": 615782016000000, + "local_weight_correlation_macs": 619763328000000, + "mirror_feedback_prediction_macs": 0, + "mirror_local_correlation_macs": 0, + "mirror_response_macs": 0, + "ordinary_forward_macs": 619763328000000, + "perturbation_forward_macs": 0, + "warmup_clean_forward_macs": 4407205888 + }, + "definition": "multiply-accumulates in conv/linear maps; one local weight correlation equals one forward-weight MAC count; BP reverse is estimated as one weight-gradient plus one activation-gradient convolution per forward convolution; mixed-traffic/predictor elementwise arithmetic is reported as a conservative operation estimate separately and is not folded into MACs", + "elementwise_operations_estimate": 65471084756992, + "forward_macs_per_example": 68862592, + "logical_batch_loss_queries": 0, + "mirror_probe_examples": 0, + "mirror_readout_probe_examples": 0, + "neutral_projection_observations": 9000000, + "per_example_cross_entropy_terms": 0, + "total_clean_forward_examples": 9000064, + "total_forward_equivalent_examples": 9000064, + "total_macs_estimate": 2471099474100224 + } +} |