| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9996261682242991, | |
| "eval_steps": 250, | |
| "global_step": 1337, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05009345794392523, | |
| "grad_norm": 74.71154022216797, | |
| "learning_rate": 4.925373134328357e-07, | |
| "loss": 2.8802, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.10018691588785046, | |
| "grad_norm": 51.97969055175781, | |
| "learning_rate": 9.925373134328357e-07, | |
| "loss": 2.3089, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 0.1502803738317757, | |
| "grad_norm": 8.622843742370605, | |
| "learning_rate": 9.451371571072319e-07, | |
| "loss": 1.1023, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 0.18691588785046728, | |
| "eval_loss": 0.8559619188308716, | |
| "eval_runtime": 67.5638, | |
| "eval_samples_per_second": 7.4, | |
| "eval_steps_per_second": 3.7, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.20037383177570092, | |
| "grad_norm": 8.973918914794922, | |
| "learning_rate": 8.894430590191188e-07, | |
| "loss": 0.8727, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 0.2504672897196262, | |
| "grad_norm": 14.043181419372559, | |
| "learning_rate": 8.337489609310058e-07, | |
| "loss": 0.6746, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.3005607476635514, | |
| "grad_norm": 13.3187255859375, | |
| "learning_rate": 7.780548628428927e-07, | |
| "loss": 0.5801, | |
| "step": 402 | |
| }, | |
| { | |
| "epoch": 0.3506542056074766, | |
| "grad_norm": 9.319698333740234, | |
| "learning_rate": 7.223607647547797e-07, | |
| "loss": 0.4428, | |
| "step": 469 | |
| }, | |
| { | |
| "epoch": 0.37383177570093457, | |
| "eval_loss": 0.438012033700943, | |
| "eval_runtime": 67.9433, | |
| "eval_samples_per_second": 7.359, | |
| "eval_steps_per_second": 3.68, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.40074766355140184, | |
| "grad_norm": 6.293191909790039, | |
| "learning_rate": 6.666666666666666e-07, | |
| "loss": 0.4403, | |
| "step": 536 | |
| }, | |
| { | |
| "epoch": 0.4508411214953271, | |
| "grad_norm": 4.233675479888916, | |
| "learning_rate": 6.109725685785536e-07, | |
| "loss": 0.3645, | |
| "step": 603 | |
| }, | |
| { | |
| "epoch": 0.5009345794392523, | |
| "grad_norm": 11.243965148925781, | |
| "learning_rate": 5.552784704904405e-07, | |
| "loss": 0.2658, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 0.5510280373831775, | |
| "grad_norm": 6.386570453643799, | |
| "learning_rate": 4.995843724023275e-07, | |
| "loss": 0.2573, | |
| "step": 737 | |
| }, | |
| { | |
| "epoch": 0.5607476635514018, | |
| "eval_loss": 0.254865825176239, | |
| "eval_runtime": 67.6778, | |
| "eval_samples_per_second": 7.388, | |
| "eval_steps_per_second": 3.694, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6011214953271028, | |
| "grad_norm": 9.038969993591309, | |
| "learning_rate": 4.438902743142144e-07, | |
| "loss": 0.2482, | |
| "step": 804 | |
| }, | |
| { | |
| "epoch": 0.6512149532710281, | |
| "grad_norm": 7.054247856140137, | |
| "learning_rate": 3.881961762261014e-07, | |
| "loss": 0.2215, | |
| "step": 871 | |
| }, | |
| { | |
| "epoch": 0.7013084112149532, | |
| "grad_norm": 4.841005325317383, | |
| "learning_rate": 3.3250207813798835e-07, | |
| "loss": 0.2055, | |
| "step": 938 | |
| }, | |
| { | |
| "epoch": 0.7476635514018691, | |
| "eval_loss": 0.19729501008987427, | |
| "eval_runtime": 67.9831, | |
| "eval_samples_per_second": 7.355, | |
| "eval_steps_per_second": 3.677, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.7514018691588785, | |
| "grad_norm": 3.596928358078003, | |
| "learning_rate": 2.7680798004987534e-07, | |
| "loss": 0.1745, | |
| "step": 1005 | |
| }, | |
| { | |
| "epoch": 0.8014953271028037, | |
| "grad_norm": 3.6744558811187744, | |
| "learning_rate": 2.2111388196176226e-07, | |
| "loss": 0.1734, | |
| "step": 1072 | |
| }, | |
| { | |
| "epoch": 0.851588785046729, | |
| "grad_norm": 8.033069610595703, | |
| "learning_rate": 1.6541978387364923e-07, | |
| "loss": 0.1929, | |
| "step": 1139 | |
| }, | |
| { | |
| "epoch": 0.9016822429906542, | |
| "grad_norm": 2.648909091949463, | |
| "learning_rate": 1.0972568578553615e-07, | |
| "loss": 0.1766, | |
| "step": 1206 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "eval_loss": 0.18509575724601746, | |
| "eval_runtime": 67.0808, | |
| "eval_samples_per_second": 7.454, | |
| "eval_steps_per_second": 3.727, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.9517757009345794, | |
| "grad_norm": 4.325321197509766, | |
| "learning_rate": 5.403158769742311e-08, | |
| "loss": 0.1678, | |
| "step": 1273 | |
| }, | |
| { | |
| "epoch": 0.9996261682242991, | |
| "step": 1337, | |
| "total_flos": 1.0258269216649574e+17, | |
| "train_loss": 0.5975192953778811, | |
| "train_runtime": 4072.514, | |
| "train_samples_per_second": 2.627, | |
| "train_steps_per_second": 0.328 | |
| } | |
| ], | |
| "logging_steps": 67, | |
| "max_steps": 1337, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.0258269216649574e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |