| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05454700417625501, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027273502088127504, | |
| "grad_norm": 15.259760856628418, | |
| "learning_rate": 1e-05, | |
| "loss": 2.9208, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0027273502088127504, | |
| "eval_loss": 2.921485662460327, | |
| "eval_runtime": 219.1663, | |
| "eval_samples_per_second": 11.27, | |
| "eval_steps_per_second": 2.82, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005454700417625501, | |
| "grad_norm": 15.232994079589844, | |
| "learning_rate": 2e-05, | |
| "loss": 2.8955, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.008182050626438252, | |
| "grad_norm": 15.136183738708496, | |
| "learning_rate": 3e-05, | |
| "loss": 2.852, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010909400835251002, | |
| "grad_norm": 16.66099739074707, | |
| "learning_rate": 4e-05, | |
| "loss": 2.7422, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013636751044063752, | |
| "grad_norm": 18.048038482666016, | |
| "learning_rate": 5e-05, | |
| "loss": 2.2015, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.013636751044063752, | |
| "eval_loss": 1.4312995672225952, | |
| "eval_runtime": 223.4476, | |
| "eval_samples_per_second": 11.054, | |
| "eval_steps_per_second": 2.766, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.016364101252876503, | |
| "grad_norm": 14.389070510864258, | |
| "learning_rate": 6e-05, | |
| "loss": 1.421, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.01909145146168925, | |
| "grad_norm": 8.234234809875488, | |
| "learning_rate": 7e-05, | |
| "loss": 0.6896, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.021818801670502003, | |
| "grad_norm": 4.542180061340332, | |
| "learning_rate": 8e-05, | |
| "loss": 0.3594, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.02454615187931475, | |
| "grad_norm": 3.838514804840088, | |
| "learning_rate": 9e-05, | |
| "loss": 0.2131, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027273502088127503, | |
| "grad_norm": 1.6445430517196655, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1487, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.027273502088127503, | |
| "eval_loss": 0.26161929965019226, | |
| "eval_runtime": 223.4522, | |
| "eval_samples_per_second": 11.054, | |
| "eval_steps_per_second": 2.766, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.030000852296940255, | |
| "grad_norm": 5.722675800323486, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.2533, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03272820250575301, | |
| "grad_norm": 2.8063879013061523, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1486, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.035455552714565755, | |
| "grad_norm": 1.7891703844070435, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.1449, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0381829029233785, | |
| "grad_norm": 2.760188579559326, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.1632, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.04091025313219126, | |
| "grad_norm": 0.9324430227279663, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1375, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04091025313219126, | |
| "eval_loss": 0.21234437823295593, | |
| "eval_runtime": 219.4952, | |
| "eval_samples_per_second": 11.253, | |
| "eval_steps_per_second": 2.816, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04363760334100401, | |
| "grad_norm": 3.428504705429077, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.197, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.046364953549816755, | |
| "grad_norm": 2.601438283920288, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.1705, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0490923037586295, | |
| "grad_norm": 2.918519973754883, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.1809, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05181965396744226, | |
| "grad_norm": 1.0848146677017212, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.1384, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05454700417625501, | |
| "grad_norm": 0.06631074100732803, | |
| "learning_rate": 0.0, | |
| "loss": 0.1314, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05454700417625501, | |
| "eval_loss": 0.13964733481407166, | |
| "eval_runtime": 219.3967, | |
| "eval_samples_per_second": 11.258, | |
| "eval_steps_per_second": 2.817, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.182135069160243e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |