{ "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": 10.0, "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_volume_cf_batch4_k4.json" }, "initial_acc": 0.5703125, "initial_tok_acc": 0.8521412037037037, "steps": [ { "step": 0, "sup_loss": 0.5222936757968037, "rf_loss": 0.0006888913339935243, "total_loss": 0.5291825892531543, "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": 0.391090969390571, "rf_loss": 0.00011522181739564985, "total_loss": 0.3922431875936313, "lyap1_mean": 0.010208295658230782, "lyap1_max": 0.06430927664041519, "lyap_volume_mean": -0.00033163721673190594, "lyap_volume_max": 0.04146828502416611, "frac_above_star": 0.25 }, { "step": 2, "sup_loss": 0.5387712348536822, "rf_loss": 0.0005317514296621084, "total_loss": 0.5440887491503033, "lyap1_mean": 0.054024145007133484, "lyap1_max": 0.06291214376688004, "lyap_volume_mean": 0.041455041617155075, "lyap_volume_max": 0.04952701926231384, "frac_above_star": 1.0 }, { "step": 3, "sup_loss": 0.5480737313519946, "rf_loss": 0.0002225781645392999, "total_loss": 0.5502995129100761, "lyap1_mean": 0.035990431904792786, "lyap1_max": 0.0537957139313221, "lyap_volume_mean": 0.02514621987938881, "lyap_volume_max": 0.04844913259148598, "frac_above_star": 0.5 }, { "step": 4, "sup_loss": 0.4902087767179898, "rf_loss": 0.0002845990238711238, "total_loss": 0.49305476695670103, "lyap1_mean": 0.012163210660219193, "lyap1_max": 0.07159461081027985, "lyap_volume_mean": -5.186721682548523e-05, "lyap_volume_max": 0.05374012514948845, "frac_above_star": 0.25 }, { "step": 5, "sup_loss": 0.6701527516399279, "rf_loss": 0.00026014778995886445, "total_loss": 0.6727542294231013, "lyap1_mean": 0.03536625951528549, "lyap1_max": 0.06755589693784714, "lyap_volume_mean": 0.02238558977842331, "lyap_volume_max": 0.0470651350915432, "frac_above_star": 0.75 }, { "step": 6, "sup_loss": 0.4908993595959489, "rf_loss": 0.0002350080176256597, "total_loss": 0.4932494398886208, "lyap1_mean": 0.008177475072443485, "lyap1_max": 0.07038463652133942, "lyap_volume_mean": 0.0025268220342695713, "lyap_volume_max": 0.04867817088961601, "frac_above_star": 0.5 }, { "step": 7, "sup_loss": 0.48578304891429425, "rf_loss": 0.00018165030633099377, "total_loss": 0.4875995519193965, "lyap1_mean": 0.01930863782763481, "lyap1_max": 0.05495723709464073, "lyap_volume_mean": 0.011218505911529064, "lyap_volume_max": 0.04498312994837761, "frac_above_star": 0.5 }, { "step": 8, "sup_loss": 0.34791163689936155, "rf_loss": 0.0003737200750038028, "total_loss": 0.3516488376493996, "lyap1_mean": 0.0033230241388082504, "lyap1_max": 0.05606033653020859, "lyap_volume_mean": -0.00013478659093379974, "lyap_volume_max": 0.05866368114948273, "frac_above_star": 0.25 }, { "step": 9, "sup_loss": 0.45022874145111974, "rf_loss": 0.0007103827665559947, "total_loss": 0.457332569233095, "lyap1_mean": 0.03756653890013695, "lyap1_max": 0.07181493192911148, "lyap_volume_mean": 0.03013727441430092, "lyap_volume_max": 0.06584680080413818, "frac_above_star": 0.75 } ], "evals": [ { "step": 0, "acc": 0.5703125, "tok_acc": 0.8521412037037037 }, { "step": 10, "acc": 0.578125, "tok_acc": 0.8606288580246914 }, { "step": 10, "acc": 0.578125, "tok_acc": 0.8606288580246914 } ], "final_acc": 0.578125, "final_tok_acc": 0.8606288580246914 }