| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9949238578680203, | |
| "eval_steps": 500, | |
| "global_step": 98, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.10152284263959391, | |
| "grad_norm": 0.2823072373867035, | |
| "learning_rate": 3.9948625730435075e-05, | |
| "loss": 2.8273, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.20304568527918782, | |
| "grad_norm": 0.2618003189563751, | |
| "learning_rate": 3.969770236712935e-05, | |
| "loss": 2.6708, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.30456852791878175, | |
| "grad_norm": 0.24902400374412537, | |
| "learning_rate": 3.9240711835383766e-05, | |
| "loss": 2.495, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.40609137055837563, | |
| "grad_norm": 0.21406474709510803, | |
| "learning_rate": 3.858297520699212e-05, | |
| "loss": 2.3206, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5076142131979695, | |
| "grad_norm": 0.24804013967514038, | |
| "learning_rate": 3.7732150985966965e-05, | |
| "loss": 2.1775, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6091370558375635, | |
| "grad_norm": 0.28641432523727417, | |
| "learning_rate": 3.6698145935029794e-05, | |
| "loss": 1.9521, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7106598984771574, | |
| "grad_norm": 0.3936513066291809, | |
| "learning_rate": 3.5492999723996456e-05, | |
| "loss": 1.7139, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8121827411167513, | |
| "grad_norm": 0.2143162190914154, | |
| "learning_rate": 3.41307447431802e-05, | |
| "loss": 1.4854, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9137055837563451, | |
| "grad_norm": 0.08941923081874847, | |
| "learning_rate": 3.262724271410661e-05, | |
| "loss": 1.3727, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.162752609153024e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |