| { | |
| "best_metric": 0.8600020408630371, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_120-samples_config-2_full/checkpoint-44", | |
| "epoch": 14.909090909090908, | |
| "eval_steps": 500, | |
| "global_step": 82, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.1166044473648071, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.2833, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.9945052266120911, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2321, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.9880269765853882, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.2758, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.2130223512649536, | |
| "eval_runtime": 25.7562, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.8734469413757324, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.2327, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.6308353543281555, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.173, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.4346662163734436, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1641, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0929126739501953, | |
| "eval_runtime": 25.7611, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3353748917579651, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0642, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.34078848361968994, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.0489, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.37743261456489563, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.0378, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 1.0190138816833496, | |
| "eval_runtime": 25.7573, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.4174298048019409, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.0076, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.5215668082237244, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9352, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.3867717385292053, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.9062, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9140139222145081, | |
| "eval_runtime": 25.7634, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.27111002802848816, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8699, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.2489100992679596, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.8437, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.8844981789588928, | |
| "eval_runtime": 25.7575, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.20925898849964142, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.8479, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.2360294610261917, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8252, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.2191130369901657, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.8116, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.8703708648681641, | |
| "eval_runtime": 25.7543, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.20331333577632904, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.8011, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.20767617225646973, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.7846, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.19857938587665558, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.7944, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.8623221516609192, | |
| "eval_runtime": 25.7577, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.2330530434846878, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7857, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.23159313201904297, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.7435, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.21888864040374756, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.7673, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8600020408630371, | |
| "eval_runtime": 25.7573, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.20804190635681152, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.732, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.2797938883304596, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.7099, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.8654985427856445, | |
| "eval_runtime": 25.7554, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.3057672381401062, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.7284, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.29697948694229126, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.6651, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.30097565054893494, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.7036, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.8722507953643799, | |
| "eval_runtime": 25.749, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.3619317412376404, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.6348, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.5063188672065735, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.6178, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.45339635014533997, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.6352, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.8909294605255127, | |
| "eval_runtime": 25.7525, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.5648482441902161, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.594, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.4616495370864868, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.5885, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.6866025924682617, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.5565, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.9371545910835266, | |
| "eval_runtime": 25.7566, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.6438589096069336, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.5179, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.7238674163818359, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.4968, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 0.953558623790741, | |
| "eval_runtime": 25.7578, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.609680712223053, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.4806, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.858428955078125, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.4862, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.7900117039680481, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.4323, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.9886630177497864, | |
| "eval_runtime": 25.7579, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.9413170218467712, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.3838, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 1.3710968494415283, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.4198, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 1.2227569818496704, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.3236, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 1.038561463356018, | |
| "eval_runtime": 25.7618, | |
| "eval_samples_per_second": 0.932, | |
| "eval_steps_per_second": 0.932, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "step": 82, | |
| "total_flos": 1.282096954825769e+17, | |
| "train_loss": 0.7752329968824619, | |
| "train_runtime": 4771.3457, | |
| "train_samples_per_second": 0.922, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 250, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.282096954825769e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |