| { | |
| "best_metric": 0.7005918622016907, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-2/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": 0.7674039006233215, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1716, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.6510034799575806, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.0331, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.6953703761100769, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1488, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.0946769714355469, | |
| "eval_runtime": 24.375, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.6922580003738403, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.0495, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.6589558720588684, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.0749, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.5799499154090881, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0165, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9406652450561523, | |
| "eval_runtime": 24.3572, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3963906466960907, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8875, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.27924686670303345, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.9195, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.30913951992988586, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.8171, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 0.8598155379295349, | |
| "eval_runtime": 24.3523, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.33593547344207764, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8588, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.3248369097709656, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7836, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2768213152885437, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.7886, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.7891804575920105, | |
| "eval_runtime": 24.3631, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.2458682656288147, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.7524, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.23660539090633392, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.7586, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.753984272480011, | |
| "eval_runtime": 24.3522, | |
| "eval_samples_per_second": 0.986, | |
| "eval_steps_per_second": 0.986, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.21786364912986755, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.6762, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.23146505653858185, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7117, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.2580190896987915, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.6893, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.7249401211738586, | |
| "eval_runtime": 24.3561, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.20970988273620605, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.6552, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.23407581448554993, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.6272, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.24965035915374756, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.6344, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.7065951228141785, | |
| "eval_runtime": 24.3589, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.2688140869140625, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.6011, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.24507106840610504, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.5704, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.29102611541748047, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.5774, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7005918622016907, | |
| "eval_runtime": 24.3547, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.2889251112937927, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.5231, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.3169316053390503, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.5214, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.7153194546699524, | |
| "eval_runtime": 24.3541, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.38805821537971497, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.5063, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.3671804964542389, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.4467, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.4408819079399109, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.4418, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7320634722709656, | |
| "eval_runtime": 24.3563, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.43601396679878235, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.4178, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.4430766701698303, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.3666, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.7093108892440796, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.3485, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.8033430576324463, | |
| "eval_runtime": 24.358, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.6464765071868896, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.2912, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.8268999457359314, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.2803, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.6604214906692505, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.2374, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8848142027854919, | |
| "eval_runtime": 24.3559, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.654966413974762, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.1974, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.9196223616600037, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1445, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 1.002466082572937, | |
| "eval_runtime": 24.3624, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.5715006589889526, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.1426, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.6640578508377075, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.1189, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 1.080757975578308, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.075, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.309071660041809, | |
| "eval_runtime": 24.3569, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.5289912223815918, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.0488, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.5767114758491516, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0553, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.7736446857452393, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.0347, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 1.3889188766479492, | |
| "eval_runtime": 24.3582, | |
| "eval_samples_per_second": 0.985, | |
| "eval_steps_per_second": 0.985, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "step": 82, | |
| "total_flos": 1.2546788251454669e+17, | |
| "train_loss": 0.558828686977305, | |
| "train_runtime": 4503.0164, | |
| "train_samples_per_second": 0.977, | |
| "train_steps_per_second": 0.056 | |
| } | |
| ], | |
| "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": 1.2546788251454669e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |