| { | |
| "best_metric": 0.7681158185005188, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_120-samples_config-2_full_auto/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": 1.0387680530548096, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1483, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.9039549231529236, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.1119, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.9227668642997742, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1522, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.096885085105896, | |
| "eval_runtime": 30.7969, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.815257728099823, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.1159, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.6035822629928589, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.0559, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.40278160572052, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0455, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.9818471074104309, | |
| "eval_runtime": 30.7916, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.30725058913230896, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9521, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.4131484925746918, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.9493, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.40404313802719116, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.9291, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 0.9177868962287903, | |
| "eval_runtime": 30.7921, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.42260661721229553, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8981, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.41342228651046753, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8381, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.26879212260246277, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8133, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8174063563346863, | |
| "eval_runtime": 30.7914, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.2632295787334442, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.7748, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.263170450925827, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.754, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.7906758189201355, | |
| "eval_runtime": 30.7913, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.2127697467803955, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.7556, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.23889917135238647, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7311, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.1890866905450821, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.7232, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.776813805103302, | |
| "eval_runtime": 30.7928, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.21611766517162323, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.712, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.18974632024765015, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.6962, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.19970326125621796, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.7137, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.7700424194335938, | |
| "eval_runtime": 30.7945, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.20971913635730743, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.6995, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.19011849164962769, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.6598, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.20270422101020813, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.6783, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7681158185005188, | |
| "eval_runtime": 30.7897, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.20463965833187103, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.6463, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.24679303169250488, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.6309, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.7696793675422668, | |
| "eval_runtime": 30.7921, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.25649893283843994, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6479, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.26783517003059387, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.589, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.26946693658828735, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.6237, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.7780429720878601, | |
| "eval_runtime": 30.7889, | |
| "eval_samples_per_second": 0.78, | |
| "eval_steps_per_second": 0.78, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.2977355420589447, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.5656, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.34951236844062805, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.5501, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.3267185389995575, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.5633, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.793101966381073, | |
| "eval_runtime": 30.7909, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.35570961236953735, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.5285, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.46553680300712585, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.5233, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.5251120924949646, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.4943, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8164987564086914, | |
| "eval_runtime": 30.7945, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.5660581588745117, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.4675, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.5252504944801331, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.4499, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 0.8475565314292908, | |
| "eval_runtime": 30.793, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.5200802087783813, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.4346, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.7916308045387268, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.4485, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.728557288646698, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.4037, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.9135186076164246, | |
| "eval_runtime": 30.7912, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.6481806635856628, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.3598, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.646964430809021, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.3885, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.7271377444267273, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.3067, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 0.9351425766944885, | |
| "eval_runtime": 30.7925, | |
| "eval_samples_per_second": 0.779, | |
| "eval_steps_per_second": 0.779, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "step": 82, | |
| "total_flos": 1.611878944032686e+17, | |
| "train_loss": 0.694953751818436, | |
| "train_runtime": 5838.2254, | |
| "train_samples_per_second": 0.754, | |
| "train_steps_per_second": 0.043 | |
| } | |
| ], | |
| "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": 1.611878944032686e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |