| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.968888888888889, | |
| "eval_steps": 112, | |
| "global_step": 448, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49777777777777776, | |
| "grad_norm": 0.6881284713745117, | |
| "learning_rate": 0.00018162162162162164, | |
| "loss": 2.141, | |
| "mean_token_accuracy": 0.5292911021872896, | |
| "step": 56, | |
| "timestamp_in_seconds": 1740062763.3345187 | |
| }, | |
| { | |
| "epoch": 0.9955555555555555, | |
| "grad_norm": 0.7973398566246033, | |
| "learning_rate": 0.00016144144144144144, | |
| "loss": 2.0305, | |
| "mean_token_accuracy": 0.5420126888514543, | |
| "step": 112, | |
| "timestamp_in_seconds": 1740062853.1339858 | |
| }, | |
| { | |
| "contract_score": 0.5277858586659635, | |
| "epoch": 0.9955555555555555, | |
| "eval_loss": 2.1109883785247803, | |
| "eval_mean_token_accuracy": 0.5238924493600638, | |
| "eval_runtime": 2.3876, | |
| "eval_samples_per_second": 41.883, | |
| "eval_steps_per_second": 5.445, | |
| "step": 112, | |
| "timestamp_in_seconds": 1740062970.6124947 | |
| }, | |
| { | |
| "epoch": 1.488888888888889, | |
| "grad_norm": 1.147162914276123, | |
| "learning_rate": 0.00014126126126126127, | |
| "loss": 1.7827, | |
| "mean_token_accuracy": 0.5765386033928739, | |
| "step": 168, | |
| "timestamp_in_seconds": 1740063059.592407 | |
| }, | |
| { | |
| "epoch": 1.9866666666666668, | |
| "grad_norm": 1.4091215133666992, | |
| "learning_rate": 0.00012108108108108109, | |
| "loss": 1.7063, | |
| "mean_token_accuracy": 0.5918067827979663, | |
| "step": 224, | |
| "timestamp_in_seconds": 1740063149.7022772 | |
| }, | |
| { | |
| "contract_score": 0.5348455982731602, | |
| "epoch": 1.9866666666666668, | |
| "eval_loss": 2.1476526260375977, | |
| "eval_mean_token_accuracy": 0.5196792704495521, | |
| "eval_runtime": 2.3928, | |
| "eval_samples_per_second": 41.792, | |
| "eval_steps_per_second": 5.433, | |
| "step": 224, | |
| "timestamp_in_seconds": 1740063269.0954084 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 2.3702051639556885, | |
| "learning_rate": 0.00010090090090090089, | |
| "loss": 1.2626, | |
| "mean_token_accuracy": 0.677473354366674, | |
| "step": 280, | |
| "timestamp_in_seconds": 1740063358.0768616 | |
| }, | |
| { | |
| "epoch": 2.977777777777778, | |
| "grad_norm": 2.522280693054199, | |
| "learning_rate": 8.072072072072072e-05, | |
| "loss": 1.1613, | |
| "mean_token_accuracy": 0.6870262543456257, | |
| "step": 336, | |
| "timestamp_in_seconds": 1740063447.9438603 | |
| }, | |
| { | |
| "contract_score": 0.5358325677018221, | |
| "epoch": 2.977777777777778, | |
| "eval_loss": 2.443709373474121, | |
| "eval_mean_token_accuracy": 0.49963776708283747, | |
| "eval_runtime": 2.3818, | |
| "eval_samples_per_second": 41.984, | |
| "eval_steps_per_second": 5.458, | |
| "step": 336, | |
| "timestamp_in_seconds": 1740063559.142597 | |
| }, | |
| { | |
| "epoch": 3.471111111111111, | |
| "grad_norm": 2.7066662311553955, | |
| "learning_rate": 6.0540540540540543e-05, | |
| "loss": 0.766, | |
| "mean_token_accuracy": 0.7945459790707263, | |
| "step": 392, | |
| "timestamp_in_seconds": 1740063648.3416014 | |
| }, | |
| { | |
| "epoch": 3.968888888888889, | |
| "grad_norm": 3.577619791030884, | |
| "learning_rate": 4.036036036036036e-05, | |
| "loss": 0.7079, | |
| "mean_token_accuracy": 0.8051769671117456, | |
| "step": 448, | |
| "timestamp_in_seconds": 1740063738.5429106 | |
| }, | |
| { | |
| "epoch": 3.968888888888889, | |
| "eval_loss": 2.8902642726898193, | |
| "eval_mean_token_accuracy": 0.4819183297035222, | |
| "eval_runtime": 2.3896, | |
| "eval_samples_per_second": 41.848, | |
| "eval_steps_per_second": 5.44, | |
| "step": 448, | |
| "timestamp_in_seconds": 1740063740.977169 | |
| } | |
| ], | |
| "logging_steps": 56, | |
| "max_steps": 560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 112, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.4238183188596736e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |