| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.704530087897228, | |
| "eval_steps": 500, | |
| "global_step": 4000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3380662609871535, | |
| "grad_norm": 1.1497818231582642, | |
| "learning_rate": 1.776876267748479e-05, | |
| "loss": 1.475, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3380662609871535, | |
| "eval_loss": 0.4233010411262512, | |
| "eval_runtime": 4.5459, | |
| "eval_samples_per_second": 289.274, | |
| "eval_steps_per_second": 36.297, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.676132521974307, | |
| "grad_norm": 0.609937846660614, | |
| "learning_rate": 1.55149876042371e-05, | |
| "loss": 0.4585, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.676132521974307, | |
| "eval_loss": 0.3847877085208893, | |
| "eval_runtime": 4.0513, | |
| "eval_samples_per_second": 324.584, | |
| "eval_steps_per_second": 40.727, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0141987829614605, | |
| "grad_norm": 0.7841311097145081, | |
| "learning_rate": 1.3261212530989409e-05, | |
| "loss": 0.4269, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.0141987829614605, | |
| "eval_loss": 0.3729918599128723, | |
| "eval_runtime": 4.4571, | |
| "eval_samples_per_second": 295.035, | |
| "eval_steps_per_second": 37.02, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.352265043948614, | |
| "grad_norm": 0.8554628491401672, | |
| "learning_rate": 1.100743745774172e-05, | |
| "loss": 0.4106, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.352265043948614, | |
| "eval_loss": 0.36659201979637146, | |
| "eval_runtime": 4.8023, | |
| "eval_samples_per_second": 273.828, | |
| "eval_steps_per_second": 34.359, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.6903313049357673, | |
| "grad_norm": 0.7967121005058289, | |
| "learning_rate": 8.753662384494028e-06, | |
| "loss": 0.4016, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.6903313049357673, | |
| "eval_loss": 0.36214107275009155, | |
| "eval_runtime": 4.099, | |
| "eval_samples_per_second": 320.808, | |
| "eval_steps_per_second": 40.253, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.028397565922921, | |
| "grad_norm": 0.5886064171791077, | |
| "learning_rate": 6.499887311246338e-06, | |
| "loss": 0.3963, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.028397565922921, | |
| "eval_loss": 0.35916823148727417, | |
| "eval_runtime": 4.5892, | |
| "eval_samples_per_second": 286.545, | |
| "eval_steps_per_second": 35.954, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.366463826910074, | |
| "grad_norm": 0.5553242564201355, | |
| "learning_rate": 4.246112237998648e-06, | |
| "loss": 0.3905, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.366463826910074, | |
| "eval_loss": 0.3574349880218506, | |
| "eval_runtime": 4.1299, | |
| "eval_samples_per_second": 318.41, | |
| "eval_steps_per_second": 39.953, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.704530087897228, | |
| "grad_norm": 3.7581846714019775, | |
| "learning_rate": 1.992337164750958e-06, | |
| "loss": 0.3889, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.704530087897228, | |
| "eval_loss": 0.35647767782211304, | |
| "eval_runtime": 5.0273, | |
| "eval_samples_per_second": 261.573, | |
| "eval_steps_per_second": 32.821, | |
| "step": 4000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 4437, | |
| "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": 1082599069974528.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |