| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0638297872340425, | |
| "eval_steps": 3, | |
| "global_step": 36, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0851063829787234, | |
| "eval_loss": 3.537301540374756, | |
| "eval_runtime": 0.8155, | |
| "eval_samples_per_second": 24.526, | |
| "eval_steps_per_second": 3.679, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "grad_norm": 4.053133964538574, | |
| "learning_rate": 3e-05, | |
| "loss": 3.9013, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "eval_loss": 3.498924732208252, | |
| "eval_runtime": 0.4147, | |
| "eval_samples_per_second": 48.228, | |
| "eval_steps_per_second": 7.234, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "grad_norm": 3.631814479827881, | |
| "learning_rate": 6e-05, | |
| "loss": 3.5773, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "eval_loss": 3.226794481277466, | |
| "eval_runtime": 0.4126, | |
| "eval_samples_per_second": 48.473, | |
| "eval_steps_per_second": 7.271, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "grad_norm": 3.4452314376831055, | |
| "learning_rate": 9e-05, | |
| "loss": 3.432, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "eval_loss": 3.062047243118286, | |
| "eval_runtime": 0.416, | |
| "eval_samples_per_second": 48.077, | |
| "eval_steps_per_second": 7.211, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.0212765957446808, | |
| "grad_norm": 4.869143962860107, | |
| "learning_rate": 9.85470908713026e-05, | |
| "loss": 3.4585, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.0212765957446808, | |
| "eval_loss": 2.945756196975708, | |
| "eval_runtime": 0.4153, | |
| "eval_samples_per_second": 48.158, | |
| "eval_steps_per_second": 7.224, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "grad_norm": 3.5793509483337402, | |
| "learning_rate": 9.114919329468282e-05, | |
| "loss": 2.8128, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.2765957446808511, | |
| "eval_loss": 2.8750696182250977, | |
| "eval_runtime": 0.4152, | |
| "eval_samples_per_second": 48.165, | |
| "eval_steps_per_second": 7.225, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.5319148936170213, | |
| "grad_norm": 3.459306001663208, | |
| "learning_rate": 7.840323733655778e-05, | |
| "loss": 2.6305, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.5319148936170213, | |
| "eval_loss": 2.800814151763916, | |
| "eval_runtime": 0.4153, | |
| "eval_samples_per_second": 48.16, | |
| "eval_steps_per_second": 7.224, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.7872340425531914, | |
| "grad_norm": 2.895676851272583, | |
| "learning_rate": 6.19657832143779e-05, | |
| "loss": 2.6033, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.7872340425531914, | |
| "eval_loss": 2.7672295570373535, | |
| "eval_runtime": 0.4158, | |
| "eval_samples_per_second": 48.098, | |
| "eval_steps_per_second": 7.215, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 2.0425531914893615, | |
| "grad_norm": 3.6542279720306396, | |
| "learning_rate": 4.397316598723385e-05, | |
| "loss": 2.8825, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.0425531914893615, | |
| "eval_loss": 2.7445950508117676, | |
| "eval_runtime": 0.4188, | |
| "eval_samples_per_second": 47.757, | |
| "eval_steps_per_second": 7.164, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.297872340425532, | |
| "grad_norm": 2.8489627838134766, | |
| "learning_rate": 2.6763841397811573e-05, | |
| "loss": 2.2496, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.297872340425532, | |
| "eval_loss": 2.7331249713897705, | |
| "eval_runtime": 0.4223, | |
| "eval_samples_per_second": 47.356, | |
| "eval_steps_per_second": 7.103, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "grad_norm": 2.887648820877075, | |
| "learning_rate": 1.257446259144494e-05, | |
| "loss": 2.4327, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.5531914893617023, | |
| "eval_loss": 2.729001998901367, | |
| "eval_runtime": 0.4168, | |
| "eval_samples_per_second": 47.983, | |
| "eval_steps_per_second": 7.197, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.8085106382978724, | |
| "grad_norm": 2.3609039783477783, | |
| "learning_rate": 3.249187865729264e-06, | |
| "loss": 2.3787, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 2.8085106382978724, | |
| "eval_loss": 2.7300968170166016, | |
| "eval_runtime": 0.4182, | |
| "eval_samples_per_second": 47.827, | |
| "eval_steps_per_second": 7.174, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.0638297872340425, | |
| "grad_norm": 2.7310924530029297, | |
| "learning_rate": 0.0, | |
| "loss": 2.3388, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.0638297872340425, | |
| "eval_loss": 2.7303249835968018, | |
| "eval_runtime": 0.4199, | |
| "eval_samples_per_second": 47.631, | |
| "eval_steps_per_second": 7.145, | |
| "step": 36 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 36, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7206952438333440.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |