| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.29282576866764276, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.005856515373352855, | |
| "eval_loss": 8.009540557861328, | |
| "eval_runtime": 18.1404, | |
| "eval_samples_per_second": 15.876, | |
| "eval_steps_per_second": 1.985, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.017569546120058566, | |
| "grad_norm": 24.97199821472168, | |
| "learning_rate": 3e-05, | |
| "loss": 31.8204, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.029282576866764276, | |
| "eval_loss": 7.016048431396484, | |
| "eval_runtime": 18.3934, | |
| "eval_samples_per_second": 15.658, | |
| "eval_steps_per_second": 1.957, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.03513909224011713, | |
| "grad_norm": 23.506786346435547, | |
| "learning_rate": 6e-05, | |
| "loss": 29.8118, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0527086383601757, | |
| "grad_norm": 23.666690826416016, | |
| "learning_rate": 9e-05, | |
| "loss": 21.4367, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05856515373352855, | |
| "eval_loss": 3.3852756023406982, | |
| "eval_runtime": 18.4101, | |
| "eval_samples_per_second": 15.644, | |
| "eval_steps_per_second": 1.955, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07027818448023426, | |
| "grad_norm": 30.18821144104004, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 14.3256, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.08784773060029283, | |
| "grad_norm": 22.153850555419922, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 7.8936, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08784773060029283, | |
| "eval_loss": 1.4027457237243652, | |
| "eval_runtime": 18.4902, | |
| "eval_samples_per_second": 15.576, | |
| "eval_steps_per_second": 1.947, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.1054172767203514, | |
| "grad_norm": 18.735877990722656, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4.3689, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.1171303074670571, | |
| "eval_loss": 0.8026209473609924, | |
| "eval_runtime": 18.4677, | |
| "eval_samples_per_second": 15.595, | |
| "eval_steps_per_second": 1.949, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.12298682284040996, | |
| "grad_norm": 11.218101501464844, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.945, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.14055636896046853, | |
| "grad_norm": 13.951614379882812, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.5484, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.14641288433382138, | |
| "eval_loss": 0.6135609149932861, | |
| "eval_runtime": 18.4602, | |
| "eval_samples_per_second": 15.601, | |
| "eval_steps_per_second": 1.95, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.1581259150805271, | |
| "grad_norm": 10.735950469970703, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.6784, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.17569546120058566, | |
| "grad_norm": 8.114375114440918, | |
| "learning_rate": 5e-05, | |
| "loss": 2.319, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.17569546120058566, | |
| "eval_loss": 0.48306477069854736, | |
| "eval_runtime": 18.4788, | |
| "eval_samples_per_second": 15.585, | |
| "eval_steps_per_second": 1.948, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.19326500732064422, | |
| "grad_norm": 9.957014083862305, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.2084, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.20497803806734993, | |
| "eval_loss": 0.4224419891834259, | |
| "eval_runtime": 18.4611, | |
| "eval_samples_per_second": 15.6, | |
| "eval_steps_per_second": 1.95, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2108345534407028, | |
| "grad_norm": 8.859992980957031, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 1.908, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.22840409956076135, | |
| "grad_norm": 11.509817123413086, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 1.5966, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.2342606149341142, | |
| "eval_loss": 0.36830881237983704, | |
| "eval_runtime": 18.4993, | |
| "eval_samples_per_second": 15.568, | |
| "eval_steps_per_second": 1.946, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.24597364568081992, | |
| "grad_norm": 10.164499282836914, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 1.5064, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.2635431918008785, | |
| "grad_norm": 5.043495178222656, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.0295, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2635431918008785, | |
| "eval_loss": 0.33903414011001587, | |
| "eval_runtime": 18.4737, | |
| "eval_samples_per_second": 15.59, | |
| "eval_steps_per_second": 1.949, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.28111273792093705, | |
| "grad_norm": 6.2125043869018555, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 1.0625, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.29282576866764276, | |
| "eval_loss": 0.33487504720687866, | |
| "eval_runtime": 18.5458, | |
| "eval_samples_per_second": 15.529, | |
| "eval_steps_per_second": 1.941, | |
| "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": 6.65254329581568e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |