| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10121836925960637, | |
| "eval_steps": 9, | |
| "global_step": 27, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0037488284910965324, | |
| "eval_loss": 2.055473566055298, | |
| "eval_runtime": 12.7703, | |
| "eval_samples_per_second": 35.238, | |
| "eval_steps_per_second": 4.463, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011246485473289597, | |
| "grad_norm": 0.1467806100845337, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.9063, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.022492970946579195, | |
| "grad_norm": 0.18374566733837128, | |
| "learning_rate": 3e-05, | |
| "loss": 1.9823, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.033739456419868794, | |
| "grad_norm": 0.2081272304058075, | |
| "learning_rate": 4.5e-05, | |
| "loss": 2.1133, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.033739456419868794, | |
| "eval_loss": 2.050758123397827, | |
| "eval_runtime": 12.8585, | |
| "eval_samples_per_second": 34.996, | |
| "eval_steps_per_second": 4.433, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04498594189315839, | |
| "grad_norm": 0.2559334337711334, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 2.0606, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.056232427366447985, | |
| "grad_norm": 0.15697269141674042, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 2.0357, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.06747891283973759, | |
| "grad_norm": 0.16153453290462494, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 2.005, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.06747891283973759, | |
| "eval_loss": 2.0186078548431396, | |
| "eval_runtime": 12.902, | |
| "eval_samples_per_second": 34.878, | |
| "eval_steps_per_second": 4.418, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.07872539831302718, | |
| "grad_norm": 0.15399610996246338, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 1.9246, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.08997188378631678, | |
| "grad_norm": 0.18686576187610626, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 1.9773, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.10121836925960637, | |
| "grad_norm": 0.1838475912809372, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 2.0225, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.10121836925960637, | |
| "eval_loss": 1.988869071006775, | |
| "eval_runtime": 12.9477, | |
| "eval_samples_per_second": 34.755, | |
| "eval_steps_per_second": 4.402, | |
| "step": 27 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8726412244549632.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |