| { | |
| "best_metric": 0.6101364522417154, | |
| "best_model_checkpoint": "brand-safety-model\\checkpoint-260", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.6331872940063477, | |
| "learning_rate": 2.7692307692307694e-05, | |
| "loss": 3.319, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.42300194931773877, | |
| "eval_loss": 2.7462074756622314, | |
| "eval_runtime": 1.5522, | |
| "eval_samples_per_second": 330.495, | |
| "eval_steps_per_second": 10.952, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 3.3225834369659424, | |
| "learning_rate": 2.5384615384615386e-05, | |
| "loss": 2.6272, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5009746588693957, | |
| "eval_loss": 2.07950758934021, | |
| "eval_runtime": 1.523, | |
| "eval_samples_per_second": 336.824, | |
| "eval_steps_per_second": 11.162, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 3.2619457244873047, | |
| "learning_rate": 2.307692307692308e-05, | |
| "loss": 2.137, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5769980506822612, | |
| "eval_loss": 1.6682791709899902, | |
| "eval_runtime": 1.5108, | |
| "eval_samples_per_second": 339.548, | |
| "eval_steps_per_second": 11.252, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 3.448869228363037, | |
| "learning_rate": 2.076923076923077e-05, | |
| "loss": 1.7515, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 3.639822483062744, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.469, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6101364522417154, | |
| "eval_loss": 1.472076654434204, | |
| "eval_runtime": 1.5018, | |
| "eval_samples_per_second": 341.601, | |
| "eval_steps_per_second": 11.32, | |
| "step": 260 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 325268996251104.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |