| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9655172413793104, | |
| "eval_steps": 6, | |
| "global_step": 21, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04597701149425287, | |
| "grad_norm": 2.537923574447632, | |
| "learning_rate": 0.0001, | |
| "loss": 1.4151, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04597701149425287, | |
| "eval_loss": 0.6549283862113953, | |
| "eval_runtime": 59.8262, | |
| "eval_samples_per_second": 3.31, | |
| "eval_steps_per_second": 0.418, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09195402298850575, | |
| "grad_norm": 2.5883562564849854, | |
| "learning_rate": 0.0002, | |
| "loss": 1.3772, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.13793103448275862, | |
| "grad_norm": 2.2641923427581787, | |
| "learning_rate": 0.0001996917333733128, | |
| "loss": 1.3153, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.1839080459770115, | |
| "grad_norm": 5.304536819458008, | |
| "learning_rate": 0.00019876883405951377, | |
| "loss": 1.1035, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.22988505747126436, | |
| "grad_norm": 1.074016809463501, | |
| "learning_rate": 0.00019723699203976766, | |
| "loss": 0.9059, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "grad_norm": 0.8219733834266663, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.8678, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.27586206896551724, | |
| "eval_loss": 0.6351205706596375, | |
| "eval_runtime": 59.9835, | |
| "eval_samples_per_second": 3.301, | |
| "eval_steps_per_second": 0.417, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3218390804597701, | |
| "grad_norm": 1.478880524635315, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.8782, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.367816091954023, | |
| "grad_norm": 0.7757106423377991, | |
| "learning_rate": 0.0001891006524188368, | |
| "loss": 0.7993, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.41379310344827586, | |
| "grad_norm": 0.5512004494667053, | |
| "learning_rate": 0.00018526401643540922, | |
| "loss": 0.7126, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.45977011494252873, | |
| "grad_norm": 1.5733872652053833, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.685, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5057471264367817, | |
| "grad_norm": 0.34443506598472595, | |
| "learning_rate": 0.0001760405965600031, | |
| "loss": 0.653, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "grad_norm": 0.5645097494125366, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.6408, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5517241379310345, | |
| "eval_loss": 0.6028693318367004, | |
| "eval_runtime": 59.4134, | |
| "eval_samples_per_second": 3.333, | |
| "eval_steps_per_second": 0.421, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.5977011494252874, | |
| "grad_norm": 0.5011037588119507, | |
| "learning_rate": 0.00016494480483301836, | |
| "loss": 0.6326, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.6436781609195402, | |
| "grad_norm": 0.2876424789428711, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.6937, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.6896551724137931, | |
| "grad_norm": 0.47212648391723633, | |
| "learning_rate": 0.0001522498564715949, | |
| "loss": 0.6298, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.735632183908046, | |
| "grad_norm": 0.6836615800857544, | |
| "learning_rate": 0.00014539904997395468, | |
| "loss": 0.5569, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.7816091954022989, | |
| "grad_norm": 0.26638293266296387, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.6057, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "grad_norm": 0.5527795553207397, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.6546, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8275862068965517, | |
| "eval_loss": 0.5754240155220032, | |
| "eval_runtime": 59.3448, | |
| "eval_samples_per_second": 3.336, | |
| "eval_steps_per_second": 0.421, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.8735632183908046, | |
| "grad_norm": 0.3253306746482849, | |
| "learning_rate": 0.00012334453638559057, | |
| "loss": 0.5521, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.9195402298850575, | |
| "grad_norm": 0.17234736680984497, | |
| "learning_rate": 0.0001156434465040231, | |
| "loss": 0.5802, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9655172413793104, | |
| "grad_norm": 0.2543017864227295, | |
| "learning_rate": 0.0001078459095727845, | |
| "loss": 0.5579, | |
| "step": 21 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 42, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 21, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.1881844810396467e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |