| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.8, | |
| "eval_steps": 22, | |
| "global_step": 110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4888888888888889, | |
| "grad_norm": 0.9318796396255493, | |
| "learning_rate": 0.00018857142857142857, | |
| "loss": 2.4418, | |
| "mean_token_accuracy": 0.5003274670194231, | |
| "step": 11, | |
| "timestamp_in_seconds": 1740058570.7495685 | |
| }, | |
| { | |
| "epoch": 0.9777777777777777, | |
| "grad_norm": 0.8860308527946472, | |
| "learning_rate": 0.00016761904761904763, | |
| "loss": 1.9864, | |
| "mean_token_accuracy": 0.5491961038627919, | |
| "step": 22, | |
| "timestamp_in_seconds": 1740058588.2654634 | |
| }, | |
| { | |
| "contract_score": 0.5170776833233711, | |
| "epoch": 0.9777777777777777, | |
| "eval_loss": 2.119361162185669, | |
| "eval_mean_token_accuracy": 0.5128270433443488, | |
| "eval_runtime": 0.524, | |
| "eval_samples_per_second": 38.169, | |
| "eval_steps_per_second": 5.725, | |
| "step": 22, | |
| "timestamp_in_seconds": 1740058617.4410768 | |
| }, | |
| { | |
| "epoch": 1.4444444444444444, | |
| "grad_norm": 0.8875806927680969, | |
| "learning_rate": 0.00014666666666666666, | |
| "loss": 1.7433, | |
| "mean_token_accuracy": 0.5918375980378316, | |
| "step": 33, | |
| "timestamp_in_seconds": 1740058634.3158836 | |
| }, | |
| { | |
| "epoch": 1.9333333333333333, | |
| "grad_norm": 1.0898572206497192, | |
| "learning_rate": 0.00012571428571428572, | |
| "loss": 1.8075, | |
| "mean_token_accuracy": 0.5785033192624862, | |
| "step": 44, | |
| "timestamp_in_seconds": 1740058651.833792 | |
| }, | |
| { | |
| "contract_score": 0.5216960497800869, | |
| "epoch": 1.9333333333333333, | |
| "eval_loss": 2.1378540992736816, | |
| "eval_mean_token_accuracy": 0.4983097909781991, | |
| "eval_runtime": 0.5221, | |
| "eval_samples_per_second": 38.303, | |
| "eval_steps_per_second": 5.745, | |
| "step": 44, | |
| "timestamp_in_seconds": 1740058680.6121447 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.2661283016204834, | |
| "learning_rate": 0.00010476190476190477, | |
| "loss": 1.4937, | |
| "mean_token_accuracy": 0.6320650446268925, | |
| "step": 55, | |
| "timestamp_in_seconds": 1740058697.4347959 | |
| }, | |
| { | |
| "epoch": 2.888888888888889, | |
| "grad_norm": 1.691213607788086, | |
| "learning_rate": 8.380952380952382e-05, | |
| "loss": 1.409, | |
| "mean_token_accuracy": 0.6488769100266372, | |
| "step": 66, | |
| "timestamp_in_seconds": 1740058714.934673 | |
| }, | |
| { | |
| "contract_score": 0.5374702803883722, | |
| "epoch": 2.888888888888889, | |
| "eval_loss": 2.2828011512756348, | |
| "eval_mean_token_accuracy": 0.4983097909781991, | |
| "eval_runtime": 0.522, | |
| "eval_samples_per_second": 38.312, | |
| "eval_steps_per_second": 5.747, | |
| "step": 66, | |
| "timestamp_in_seconds": 1740058746.603091 | |
| }, | |
| { | |
| "epoch": 3.3555555555555556, | |
| "grad_norm": 2.212834119796753, | |
| "learning_rate": 6.285714285714286e-05, | |
| "loss": 1.0352, | |
| "mean_token_accuracy": 0.7221250653501003, | |
| "step": 77, | |
| "timestamp_in_seconds": 1740058763.4824302 | |
| }, | |
| { | |
| "epoch": 3.8444444444444446, | |
| "grad_norm": 2.533773899078369, | |
| "learning_rate": 4.190476190476191e-05, | |
| "loss": 0.9292, | |
| "mean_token_accuracy": 0.7455536097074219, | |
| "step": 88, | |
| "timestamp_in_seconds": 1740058781.1095395 | |
| }, | |
| { | |
| "contract_score": 0.5145041195676524, | |
| "epoch": 3.8444444444444446, | |
| "eval_loss": 2.5186023712158203, | |
| "eval_mean_token_accuracy": 0.4868165874886475, | |
| "eval_runtime": 0.5208, | |
| "eval_samples_per_second": 38.406, | |
| "eval_steps_per_second": 5.761, | |
| "step": 88, | |
| "timestamp_in_seconds": 1740058807.5439005 | |
| }, | |
| { | |
| "epoch": 4.311111111111111, | |
| "grad_norm": 2.1104252338409424, | |
| "learning_rate": 2.0952380952380954e-05, | |
| "loss": 0.8056, | |
| "mean_token_accuracy": 0.7859161157789674, | |
| "step": 99, | |
| "timestamp_in_seconds": 1740058824.42298 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 2.379983901977539, | |
| "learning_rate": 0.0, | |
| "loss": 0.644, | |
| "mean_token_accuracy": 0.8199670519498358, | |
| "step": 110, | |
| "timestamp_in_seconds": 1740058842.0086482 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "eval_loss": 2.7059178352355957, | |
| "eval_mean_token_accuracy": 0.4916755852569576, | |
| "eval_runtime": 0.5201, | |
| "eval_samples_per_second": 38.456, | |
| "eval_steps_per_second": 5.768, | |
| "step": 110, | |
| "timestamp_in_seconds": 1740058842.5604758 | |
| } | |
| ], | |
| "logging_steps": 11, | |
| "max_steps": 110, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 22, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5949225070012416.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |