| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4024144869215292, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008048289738430584, | |
| "eval_loss": 0.38133421540260315, | |
| "eval_runtime": 33.5879, | |
| "eval_samples_per_second": 6.222, | |
| "eval_steps_per_second": 0.804, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02414486921529175, | |
| "grad_norm": 0.7796753644943237, | |
| "learning_rate": 3e-05, | |
| "loss": 0.325, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04024144869215292, | |
| "eval_loss": 0.21984921395778656, | |
| "eval_runtime": 32.9714, | |
| "eval_samples_per_second": 6.339, | |
| "eval_steps_per_second": 0.819, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0482897384305835, | |
| "grad_norm": 0.5462653040885925, | |
| "learning_rate": 6e-05, | |
| "loss": 0.2524, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07243460764587525, | |
| "grad_norm": 0.2789907455444336, | |
| "learning_rate": 9e-05, | |
| "loss": 0.094, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08048289738430583, | |
| "eval_loss": 0.04736727476119995, | |
| "eval_runtime": 32.9992, | |
| "eval_samples_per_second": 6.333, | |
| "eval_steps_per_second": 0.818, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.096579476861167, | |
| "grad_norm": 0.2077365219593048, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.0431, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.12072434607645875, | |
| "grad_norm": 0.2362423539161682, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.0303, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.12072434607645875, | |
| "eval_loss": 0.022640502080321312, | |
| "eval_runtime": 32.9869, | |
| "eval_samples_per_second": 6.336, | |
| "eval_steps_per_second": 0.819, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1448692152917505, | |
| "grad_norm": 0.11892444640398026, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0165, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.16096579476861167, | |
| "eval_loss": 0.013070314191281796, | |
| "eval_runtime": 32.9925, | |
| "eval_samples_per_second": 6.335, | |
| "eval_steps_per_second": 0.818, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16901408450704225, | |
| "grad_norm": 0.07342974841594696, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.0177, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.193158953722334, | |
| "grad_norm": 0.1093919426202774, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.013, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2012072434607646, | |
| "eval_loss": 0.010123283602297306, | |
| "eval_runtime": 33.0079, | |
| "eval_samples_per_second": 6.332, | |
| "eval_steps_per_second": 0.818, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.21730382293762576, | |
| "grad_norm": 0.07795149087905884, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.0092, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.2414486921529175, | |
| "grad_norm": 0.0602896623313427, | |
| "learning_rate": 5e-05, | |
| "loss": 0.009, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2414486921529175, | |
| "eval_loss": 0.008096303790807724, | |
| "eval_runtime": 33.0663, | |
| "eval_samples_per_second": 6.321, | |
| "eval_steps_per_second": 0.817, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2655935613682093, | |
| "grad_norm": 0.037129033356904984, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.0068, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.28169014084507044, | |
| "eval_loss": 0.007025540340691805, | |
| "eval_runtime": 33.0263, | |
| "eval_samples_per_second": 6.328, | |
| "eval_steps_per_second": 0.818, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.289738430583501, | |
| "grad_norm": 0.05079665035009384, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0047, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.31388329979879276, | |
| "grad_norm": 0.053295738995075226, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.0069, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.32193158953722334, | |
| "eval_loss": 0.006381278857588768, | |
| "eval_runtime": 33.0314, | |
| "eval_samples_per_second": 6.327, | |
| "eval_steps_per_second": 0.817, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3380281690140845, | |
| "grad_norm": 0.06008715555071831, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0047, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.36217303822937624, | |
| "grad_norm": 0.05889170616865158, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0063, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.36217303822937624, | |
| "eval_loss": 0.0061826580204069614, | |
| "eval_runtime": 33.0232, | |
| "eval_samples_per_second": 6.329, | |
| "eval_steps_per_second": 0.818, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.386317907444668, | |
| "grad_norm": 0.04322628304362297, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0043, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4024144869215292, | |
| "eval_loss": 0.006113050039857626, | |
| "eval_runtime": 33.0447, | |
| "eval_samples_per_second": 6.325, | |
| "eval_steps_per_second": 0.817, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 50, | |
| "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": 1.497199726905262e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |