| { | |
| "best_metric": 0.8506781458854675, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_120-samples_config-1_full_auto/checkpoint-55", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 132, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 1.012304663658142, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 1.171, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 1.2253605127334595, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 1.322, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.9963738322257996, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 1.2139, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.9595217704772949, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 1.2163, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.8450391888618469, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 1.2429, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.6564331650733948, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.1614, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.1416277885437012, | |
| "eval_runtime": 25.879, | |
| "eval_samples_per_second": 0.927, | |
| "eval_steps_per_second": 0.927, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.6047136187553406, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.1924, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 0.39012518525123596, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 1.0928, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.4803689122200012, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 1.0944, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 0.4926101565361023, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 1.1072, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.4342550039291382, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.0556, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.39658549427986145, | |
| "learning_rate": 4e-05, | |
| "loss": 0.9734, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0177873373031616, | |
| "eval_runtime": 25.8714, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.3215712010860443, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 1.0057, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.5286137461662292, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.9539, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.4327259063720703, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.9102, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.3998343050479889, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.9345, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.37148231267929077, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.8716, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.891235888004303, | |
| "eval_runtime": 25.871, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 0.31582313776016235, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.8644, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.28333473205566406, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.8479, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 0.30229219794273376, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.8432, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.26926714181900024, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.7941, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 0.36835384368896484, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.7592, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.26915186643600464, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8725, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8626171946525574, | |
| "eval_runtime": 25.8722, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 0.3232079744338989, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.7644, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.30384936928749084, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.8133, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 0.27904725074768066, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.7727, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.3493795394897461, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.7659, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 0.27498477697372437, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.8125, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8506781458854675, | |
| "eval_runtime": 25.8728, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.28064048290252686, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.7575, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 0.33129581809043884, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.7589, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.37331491708755493, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.7, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 0.3793705999851227, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.7142, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.4419674575328827, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.7132, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.44069525599479675, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7541, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.8532574772834778, | |
| "eval_runtime": 25.8729, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.5028610229492188, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.6384, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 0.6297198534011841, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.6797, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.4972052276134491, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.6112, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 0.4890666604042053, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.6494, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.6373401284217834, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.6871, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.8666830062866211, | |
| "eval_runtime": 25.8692, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 0.5935012102127075, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.7017, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.6022512316703796, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.5229, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 0.8682543635368347, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.5421, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.6757321357727051, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.5795, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 0.6471613645553589, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.5569, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.7226406335830688, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.6206, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9129067063331604, | |
| "eval_runtime": 25.8689, | |
| "eval_samples_per_second": 0.928, | |
| "eval_steps_per_second": 0.928, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.777737557888031, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.5258, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 1.1797261238098145, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.4784, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 0.8039947748184204, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.5046, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.7812055349349976, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.4289, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 0.839874267578125, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.4919, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.9646734595298767, | |
| "eval_runtime": 25.886, | |
| "eval_samples_per_second": 0.927, | |
| "eval_steps_per_second": 0.927, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.7887756824493408, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.4812, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 1.1427494287490845, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.3731, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 1.0812352895736694, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.3769, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 1.0405755043029785, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.4414, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.982853889465332, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.3949, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.2930797338485718, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.381, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.0133259296417236, | |
| "eval_runtime": 25.8852, | |
| "eval_samples_per_second": 0.927, | |
| "eval_steps_per_second": 0.927, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 1.0800763368606567, | |
| "learning_rate": 9.676378356149734e-05, | |
| "loss": 0.3013, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 1.1959099769592285, | |
| "learning_rate": 9.653540160603956e-05, | |
| "loss": 0.3115, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 1.054177165031433, | |
| "learning_rate": 9.629952196931901e-05, | |
| "loss": 0.306, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 1.0009381771087646, | |
| "learning_rate": 9.60561826557425e-05, | |
| "loss": 0.3264, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 1.032883882522583, | |
| "learning_rate": 9.580542287160348e-05, | |
| "loss": 0.3491, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.0941095352172852, | |
| "eval_runtime": 25.8919, | |
| "eval_samples_per_second": 0.927, | |
| "eval_steps_per_second": 0.927, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.090909090909092, | |
| "grad_norm": 0.9598305225372314, | |
| "learning_rate": 9.554728301876526e-05, | |
| "loss": 0.2806, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 1.169908046722412, | |
| "learning_rate": 9.528180468815155e-05, | |
| "loss": 0.2436, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 11.454545454545455, | |
| "grad_norm": 1.0633145570755005, | |
| "learning_rate": 9.50090306530454e-05, | |
| "loss": 0.2585, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.9604322910308838, | |
| "learning_rate": 9.472900486219769e-05, | |
| "loss": 0.2903, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 11.818181818181818, | |
| "grad_norm": 1.161165714263916, | |
| "learning_rate": 9.444177243274618e-05, | |
| "loss": 0.2438, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.2847615480422974, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2613, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.119102954864502, | |
| "eval_runtime": 25.8889, | |
| "eval_samples_per_second": 0.927, | |
| "eval_steps_per_second": 0.927, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 132, | |
| "total_flos": 1.0340910033482547e+17, | |
| "train_loss": 0.6881949709671916, | |
| "train_runtime": 3843.3104, | |
| "train_samples_per_second": 1.145, | |
| "train_steps_per_second": 0.143 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 550, | |
| "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": 1.0340910033482547e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |