| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.872727272727273, | |
| "eval_steps": 500, | |
| "global_step": 30, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14545454545454545, | |
| "grad_norm": 25.54774211851432, | |
| "learning_rate": 4.986304738420684e-06, | |
| "loss": 1.0453, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 3.20058486867644, | |
| "learning_rate": 4.665063509461098e-06, | |
| "loss": 0.4619, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.5818181818181818, | |
| "grad_norm": 0.8956385642815118, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.3321, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.4363636363636365, | |
| "grad_norm": 0.7893148037461633, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.27, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 3.290909090909091, | |
| "grad_norm": 0.7352460096742335, | |
| "learning_rate": 1.2500000000000007e-06, | |
| "loss": 0.2191, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.1454545454545455, | |
| "grad_norm": 0.6985314968887066, | |
| "learning_rate": 3.3493649053890325e-07, | |
| "loss": 0.1778, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.872727272727273, | |
| "grad_norm": 0.6697082521139333, | |
| "learning_rate": 0.0, | |
| "loss": 0.1349, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.872727272727273, | |
| "step": 30, | |
| "total_flos": 2.456711189401436e+17, | |
| "train_loss": 0.2854084233442942, | |
| "train_runtime": 3135.0611, | |
| "train_samples_per_second": 1.397, | |
| "train_steps_per_second": 0.01 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 30, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.456711189401436e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |