| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.006564551422319475, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007293946024799417, | |
| "grad_norm": 1.4022825956344604, | |
| "learning_rate": 2e-05, | |
| "loss": 2.325, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0007293946024799417, | |
| "eval_loss": 2.170013189315796, | |
| "eval_runtime": 28.6088, | |
| "eval_samples_per_second": 20.204, | |
| "eval_steps_per_second": 10.102, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0014587892049598833, | |
| "grad_norm": 1.3102909326553345, | |
| "learning_rate": 4e-05, | |
| "loss": 2.1401, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.002188183807439825, | |
| "grad_norm": 1.7629603147506714, | |
| "learning_rate": 6e-05, | |
| "loss": 2.3028, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.002188183807439825, | |
| "eval_loss": 2.1548879146575928, | |
| "eval_runtime": 27.827, | |
| "eval_samples_per_second": 20.771, | |
| "eval_steps_per_second": 10.386, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0029175784099197666, | |
| "grad_norm": 1.6993635892868042, | |
| "learning_rate": 8e-05, | |
| "loss": 2.3704, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0036469730123997084, | |
| "grad_norm": 1.0507618188858032, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1938, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00437636761487965, | |
| "grad_norm": 1.165635347366333, | |
| "learning_rate": 0.00012, | |
| "loss": 1.9088, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00437636761487965, | |
| "eval_loss": 2.006481647491455, | |
| "eval_runtime": 27.9404, | |
| "eval_samples_per_second": 20.687, | |
| "eval_steps_per_second": 10.343, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.005105762217359592, | |
| "grad_norm": 2.2356066703796387, | |
| "learning_rate": 0.00014, | |
| "loss": 1.9656, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.005835156819839533, | |
| "grad_norm": 0.6858603358268738, | |
| "learning_rate": 0.00016, | |
| "loss": 2.278, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.006564551422319475, | |
| "grad_norm": 0.5601918697357178, | |
| "learning_rate": 0.00018, | |
| "loss": 1.8191, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006564551422319475, | |
| "eval_loss": 1.8506412506103516, | |
| "eval_runtime": 27.8918, | |
| "eval_samples_per_second": 20.723, | |
| "eval_steps_per_second": 10.361, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2968265781411840.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |