| { | |
| "best_metric": 0.8230898976325989, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_120-samples_config-1_full_auto/checkpoint-44", | |
| "epoch": 11.0, | |
| "eval_steps": 500, | |
| "global_step": 121, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09090909090909091, | |
| "grad_norm": 2.0714240074157715, | |
| "learning_rate": 1.818181818181818e-06, | |
| "loss": 1.7193, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 2.1086008548736572, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 1.7033, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 2.060307264328003, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 1.805, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 2.4383561611175537, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 1.7449, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 1.7405235767364502, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 1.6484, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 1.68947434425354, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 1.6872, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.5145670175552368, | |
| "eval_runtime": 9.9666, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.5627727508544922, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 1.5685, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.2727272727272727, | |
| "grad_norm": 1.1418180465698242, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 1.4788, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 1.1541528701782227, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 1.3362, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.6363636363636362, | |
| "grad_norm": 1.3468446731567383, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 1.3399, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.7159044742584229, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 1.2106, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.143359661102295, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0353, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.0329313278198242, | |
| "eval_runtime": 9.9628, | |
| "eval_samples_per_second": 2.409, | |
| "eval_steps_per_second": 2.409, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.1660093069076538, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.9525, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 1.0319758653640747, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.9479, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.9095191955566406, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.8791, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 0.8936154842376709, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.8663, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.8487228751182556, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.8377, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8574712872505188, | |
| "eval_runtime": 9.967, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 3.090909090909091, | |
| "grad_norm": 0.7327587008476257, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.8285, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.7707822918891907, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.7542, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 3.4545454545454546, | |
| "grad_norm": 0.7735487222671509, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.7502, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.8239573240280151, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.7624, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.8181818181818183, | |
| "grad_norm": 1.01390540599823, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.7512, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.9193002581596375, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7346, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8230898976325989, | |
| "eval_runtime": 9.9652, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 4.181818181818182, | |
| "grad_norm": 0.9260618090629578, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.6508, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.9957122206687927, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.6485, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 1.1015777587890625, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.6417, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.9873030185699463, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.6238, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "grad_norm": 1.1196095943450928, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.5798, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8247186541557312, | |
| "eval_runtime": 9.9633, | |
| "eval_samples_per_second": 2.409, | |
| "eval_steps_per_second": 2.409, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 1.0865353345870972, | |
| "learning_rate": 9.999899300364532e-05, | |
| "loss": 0.5368, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 5.2727272727272725, | |
| "grad_norm": 1.1556649208068848, | |
| "learning_rate": 9.99909372761763e-05, | |
| "loss": 0.4762, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 1.3103891611099243, | |
| "learning_rate": 9.997482711915927e-05, | |
| "loss": 0.4255, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 5.636363636363637, | |
| "grad_norm": 1.5533417463302612, | |
| "learning_rate": 9.99506651282272e-05, | |
| "loss": 0.4607, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 1.5097439289093018, | |
| "learning_rate": 9.991845519630678e-05, | |
| "loss": 0.4211, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.3186739683151245, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.4162, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.8889004588127136, | |
| "eval_runtime": 9.9645, | |
| "eval_samples_per_second": 2.409, | |
| "eval_steps_per_second": 2.409, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 1.118858814239502, | |
| "learning_rate": 9.982991356370404e-05, | |
| "loss": 0.3094, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 1.5540125370025635, | |
| "learning_rate": 9.977359612865423e-05, | |
| "loss": 0.3042, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 1.762613296508789, | |
| "learning_rate": 9.970925928158274e-05, | |
| "loss": 0.3044, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 6.7272727272727275, | |
| "grad_norm": 1.974087119102478, | |
| "learning_rate": 9.963691338830044e-05, | |
| "loss": 0.3062, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 1.864396095275879, | |
| "learning_rate": 9.955657010501806e-05, | |
| "loss": 0.2904, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 1.0323710441589355, | |
| "eval_runtime": 9.9657, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 7.090909090909091, | |
| "grad_norm": 1.2859554290771484, | |
| "learning_rate": 9.946824237646824e-05, | |
| "loss": 0.2488, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 1.0327765941619873, | |
| "learning_rate": 9.937194443381972e-05, | |
| "loss": 0.2182, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 7.454545454545454, | |
| "grad_norm": 1.2168482542037964, | |
| "learning_rate": 9.926769179238466e-05, | |
| "loss": 0.2006, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 1.4469397068023682, | |
| "learning_rate": 9.915550124911866e-05, | |
| "loss": 0.1995, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 7.818181818181818, | |
| "grad_norm": 1.7654582262039185, | |
| "learning_rate": 9.903539087991462e-05, | |
| "loss": 0.2129, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.216096878051758, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.2039, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.125268578529358, | |
| "eval_runtime": 9.9644, | |
| "eval_samples_per_second": 2.409, | |
| "eval_steps_per_second": 2.409, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 1.1632194519042969, | |
| "learning_rate": 9.877148934427037e-05, | |
| "loss": 0.1461, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 1.42656409740448, | |
| "learning_rate": 9.862774069706346e-05, | |
| "loss": 0.1465, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 8.545454545454545, | |
| "grad_norm": 1.3725488185882568, | |
| "learning_rate": 9.847615725553456e-05, | |
| "loss": 0.1449, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 1.2858436107635498, | |
| "learning_rate": 9.831676344247342e-05, | |
| "loss": 0.1565, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "grad_norm": 1.4211639165878296, | |
| "learning_rate": 9.814958493905963e-05, | |
| "loss": 0.155, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 1.1651321649551392, | |
| "eval_runtime": 9.9649, | |
| "eval_samples_per_second": 2.408, | |
| "eval_steps_per_second": 2.408, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 1.1477361917495728, | |
| "learning_rate": 9.797464868072488e-05, | |
| "loss": 0.136, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 9.272727272727273, | |
| "grad_norm": 1.401991367340088, | |
| "learning_rate": 9.779198285281325e-05, | |
| "loss": 0.121, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 1.5960593223571777, | |
| "learning_rate": 9.760161688604008e-05, | |
| "loss": 0.1198, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 9.636363636363637, | |
| "grad_norm": 1.5943067073822021, | |
| "learning_rate": 9.740358145174998e-05, | |
| "loss": 0.1204, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 1.374107003211975, | |
| "learning_rate": 9.719790845697533e-05, | |
| "loss": 0.1326, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.3806663751602173, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1233, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.1566959619522095, | |
| "eval_runtime": 9.9735, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 2.406, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.864838719367981, | |
| "learning_rate": 9.676378356149734e-05, | |
| "loss": 0.097, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 10.363636363636363, | |
| "grad_norm": 0.8773109912872314, | |
| "learning_rate": 9.653540160603956e-05, | |
| "loss": 0.0921, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 1.2778701782226562, | |
| "learning_rate": 9.629952196931901e-05, | |
| "loss": 0.105, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 10.727272727272727, | |
| "grad_norm": 1.9501029253005981, | |
| "learning_rate": 9.60561826557425e-05, | |
| "loss": 0.1133, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 1.3584322929382324, | |
| "learning_rate": 9.580542287160348e-05, | |
| "loss": 0.1064, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 1.243621826171875, | |
| "eval_runtime": 9.9736, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 2.406, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "step": 121, | |
| "total_flos": 1.5952182134177792e+16, | |
| "train_loss": 0.6013495232448105, | |
| "train_runtime": 1176.0591, | |
| "train_samples_per_second": 3.741, | |
| "train_steps_per_second": 0.468 | |
| } | |
| ], | |
| "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": 1.5952182134177792e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |