| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9101251422070534, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 0.26715654134750366, | |
| "learning_rate": 4.9992855064046754e-05, | |
| "loss": 2.6697, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.4067687392234802, | |
| "learning_rate": 4.997142434019578e-05, | |
| "loss": 2.5369, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 0.44472697377204895, | |
| "learning_rate": 4.9935720078139045e-05, | |
| "loss": 2.5661, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 0.45230674743652344, | |
| "learning_rate": 4.988576268624979e-05, | |
| "loss": 2.4824, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 0.4477585554122925, | |
| "learning_rate": 4.982158071991725e-05, | |
| "loss": 2.3343, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 0.3630908131599426, | |
| "learning_rate": 4.974321086522453e-05, | |
| "loss": 2.4377, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 0.3237389028072357, | |
| "learning_rate": 4.9650697917979025e-05, | |
| "loss": 2.4114, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.3014233708381653, | |
| "learning_rate": 4.954409475810737e-05, | |
| "loss": 2.2636, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 0.3054388761520386, | |
| "learning_rate": 4.942346231942955e-05, | |
| "loss": 2.2758, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 0.2940792143344879, | |
| "learning_rate": 4.92888695548294e-05, | |
| "loss": 2.3207, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 0.25064757466316223, | |
| "learning_rate": 4.9140393396841565e-05, | |
| "loss": 2.2209, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 0.3227023780345917, | |
| "learning_rate": 4.89781187136772e-05, | |
| "loss": 2.2328, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.312673419713974, | |
| "learning_rate": 4.880213826071375e-05, | |
| "loss": 2.2737, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 0.26930904388427734, | |
| "learning_rate": 4.861255262747643e-05, | |
| "loss": 2.2686, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 0.2343609631061554, | |
| "learning_rate": 4.8409470180141827e-05, | |
| "loss": 2.2661, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 0.3088403642177582, | |
| "learning_rate": 4.8193006999596294e-05, | |
| "loss": 2.191, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 0.3336387574672699, | |
| "learning_rate": 4.796328681508473e-05, | |
| "loss": 2.2106, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 0.28899675607681274, | |
| "learning_rate": 4.7720440933487575e-05, | |
| "loss": 2.2347, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 0.30223432183265686, | |
| "learning_rate": 4.746460816426647e-05, | |
| "loss": 2.2307, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 0.2692021429538727, | |
| "learning_rate": 4.7195934740121485e-05, | |
| "loss": 2.1503, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 0.3296695053577423, | |
| "learning_rate": 4.6914574233405236e-05, | |
| "loss": 2.2145, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.28714171051979065, | |
| "learning_rate": 4.662068746834176e-05, | |
| "loss": 2.1163, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 0.35001716017723083, | |
| "learning_rate": 4.6314442429100155e-05, | |
| "loss": 2.1868, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.32721835374832153, | |
| "learning_rate": 4.599601416377575e-05, | |
| "loss": 2.1865, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 0.30708980560302734, | |
| "learning_rate": 4.566558468433344e-05, | |
| "loss": 2.2035, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 0.3122975528240204, | |
| "learning_rate": 4.532334286257064e-05, | |
| "loss": 2.1762, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 0.3531278669834137, | |
| "learning_rate": 4.496948432215913e-05, | |
| "loss": 2.2452, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 0.3190854489803314, | |
| "learning_rate": 4.460421132682751e-05, | |
| "loss": 2.2267, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 0.29605475068092346, | |
| "learning_rate": 4.4227732664748365e-05, | |
| "loss": 2.2548, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 0.37863361835479736, | |
| "learning_rate": 4.384026352919595e-05, | |
| "loss": 2.2053, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 0.3235403895378113, | |
| "learning_rate": 4.344202539554285e-05, | |
| "loss": 2.193, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.3326057195663452, | |
| "learning_rate": 4.3033245894665814e-05, | |
| "loss": 2.2349, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 0.3827252686023712, | |
| "learning_rate": 4.261415868283304e-05, | |
| "loss": 2.1247, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 0.34777992963790894, | |
| "learning_rate": 4.218500330814753e-05, | |
| "loss": 2.1555, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 0.3294101059436798, | |
| "learning_rate": 4.174602507362258e-05, | |
| "loss": 2.1771, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 0.3454132676124573, | |
| "learning_rate": 4.1297474896967814e-05, | |
| "loss": 2.1616, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 0.3628969192504883, | |
| "learning_rate": 4.083960916716597e-05, | |
| "loss": 2.1681, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 0.31756460666656494, | |
| "learning_rate": 4.0372689597922215e-05, | |
| "loss": 2.146, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 0.41087692975997925, | |
| "learning_rate": 3.989698307806995e-05, | |
| "loss": 2.2185, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 0.3311336636543274, | |
| "learning_rate": 3.941276151901853e-05, | |
| "loss": 2.0976, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 657, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 6.836880289234944e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |