{ "args": { "ckpt_root": "/home/yurenh2/rrm/trm/checkpoints/Sudoku-extreme-1k-aug-1000-ACT-torch/pretrain_mlp_t_sudoku_singleGPU", "ckpt_name": "step_26041", "n_steps": 10, "batch_size": 4, "lr": 1e-05, "alpha_rf": 0.0, "lambda_star": 0.02, "rf_mode": "volume_cf", "k_lyap": 4, "lyap_act_steps": 4, "seed": 42, "eval_every": 10, "eval_n": 128, "eval_batch_size": 32, "out": "smoke_trm_baseline_batch4_k4.json" }, "initial_acc": 0.5703125, "initial_tok_acc": 0.8521412037037037, "steps": [ { "step": 0, "sup_loss": 0.5224550515927421, "rf_loss": 0.0, "total_loss": 0.5224550515927421, "lyap1_mean": 0.029093526303768158, "lyap1_max": 0.08353869616985321, "lyap_volume_mean": 0.015920955687761307, "lyap_volume_max": 0.06817857921123505, "frac_above_star": 0.5 }, { "step": 1, "sup_loss": 0.2893205619936239, "rf_loss": 0.0, "total_loss": 0.2893205619936239, "lyap1_mean": 0.016572164371609688, "lyap1_max": 0.0947149395942688, "lyap_volume_mean": 0.004541127942502499, "lyap_volume_max": 0.06567265093326569, "frac_above_star": 0.5 }, { "step": 2, "sup_loss": 0.533731442617599, "rf_loss": 0.0, "total_loss": 0.533731442617599, "lyap1_mean": 0.0415254682302475, "lyap1_max": 0.06299518048763275, "lyap_volume_mean": 0.02515004761517048, "lyap_volume_max": 0.0505189523100853, "frac_above_star": 0.75 }, { "step": 3, "sup_loss": 0.6825523199828674, "rf_loss": 0.0, "total_loss": 0.6825523199828674, "lyap1_mean": 0.042442794889211655, "lyap1_max": 0.09923867136240005, "lyap_volume_mean": 0.027183376252651215, "lyap_volume_max": 0.06509065628051758, "frac_above_star": 0.5 }, { "step": 4, "sup_loss": 0.46650786503362784, "rf_loss": 0.0, "total_loss": 0.46650786503362784, "lyap1_mean": 0.01466364972293377, "lyap1_max": 0.10311321169137955, "lyap_volume_mean": 6.554648280143738e-05, "lyap_volume_max": 0.07338342815637589, "frac_above_star": 0.25 }, { "step": 5, "sup_loss": 0.6539231951722436, "rf_loss": 0.0, "total_loss": 0.6539231951722436, "lyap1_mean": 0.03335200995206833, "lyap1_max": 0.07353045791387558, "lyap_volume_mean": 0.02371128648519516, "lyap_volume_max": 0.05123281478881836, "frac_above_star": 0.75 }, { "step": 6, "sup_loss": 0.4977401191610872, "rf_loss": 0.0, "total_loss": 0.4977401191610872, "lyap1_mean": 0.01054409984499216, "lyap1_max": 0.05166617035865784, "lyap_volume_mean": 0.0034741926938295364, "lyap_volume_max": 0.04562399536371231, "frac_above_star": 0.5 }, { "step": 7, "sup_loss": 0.4983922783865375, "rf_loss": 0.0, "total_loss": 0.4983922783865375, "lyap1_mean": 0.01408039964735508, "lyap1_max": 0.048263903707265854, "lyap_volume_mean": 0.007988786324858665, "lyap_volume_max": 0.04666714742779732, "frac_above_star": 0.5 }, { "step": 8, "sup_loss": 0.3797426061189834, "rf_loss": 0.0, "total_loss": 0.3797426061189834, "lyap1_mean": 0.016823258250951767, "lyap1_max": 0.1099422350525856, "lyap_volume_mean": 0.005312556400895119, "lyap_volume_max": 0.07368713617324829, "frac_above_star": 0.25 }, { "step": 9, "sup_loss": 0.4811805004085315, "rf_loss": 0.0, "total_loss": 0.4811805004085315, "lyap1_mean": 0.0443924255669117, "lyap1_max": 0.08661376684904099, "lyap_volume_mean": 0.03681604564189911, "lyap_volume_max": 0.07089020311832428, "frac_above_star": 0.75 } ], "evals": [ { "step": 0, "acc": 0.5703125, "tok_acc": 0.8521412037037037 }, { "step": 10, "acc": 0.5625, "tok_acc": 0.8549382716049383 }, { "step": 10, "acc": 0.5625, "tok_acc": 0.8549382716049383 } ], "final_acc": 0.5625, "final_tok_acc": 0.8549382716049383 }