| { | |
| "best_metric": 0.9769736842105263, | |
| "best_model_checkpoint": "ai_vs_real_image_detection/checkpoint-140", | |
| "epoch": 9.565217391304348, | |
| "eval_steps": 10, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 67.0071029663086, | |
| "learning_rate": 5e-06, | |
| "loss": 2.1934, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 110.92020416259766, | |
| "learning_rate": 1e-05, | |
| "loss": 0.6235, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 145.63641357421875, | |
| "learning_rate": 1.5e-05, | |
| "loss": 2.5218, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 220.67752075195312, | |
| "learning_rate": 2e-05, | |
| "loss": 3.0083, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 73.61774444580078, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.9029, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 85.73609924316406, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2997, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 135.3199920654297, | |
| "learning_rate": 3.5e-05, | |
| "loss": 2.2808, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 127.65010833740234, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0529, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 110.29368591308594, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.2478, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 23.880586624145508, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7569, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.7927631578947368, | |
| "eval_loss": 0.545388400554657, | |
| "eval_runtime": 6.0123, | |
| "eval_samples_per_second": 50.563, | |
| "eval_steps_per_second": 1.663, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 30.196847915649414, | |
| "learning_rate": 4.9772727272727275e-05, | |
| "loss": 0.5346, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 16.50008201599121, | |
| "learning_rate": 4.9545454545454553e-05, | |
| "loss": 0.5966, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 26.836137771606445, | |
| "learning_rate": 4.931818181818182e-05, | |
| "loss": 0.2751, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 20.927852630615234, | |
| "learning_rate": 4.909090909090909e-05, | |
| "loss": 0.4556, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 25.444683074951172, | |
| "learning_rate": 4.886363636363637e-05, | |
| "loss": 0.2919, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 11.910268783569336, | |
| "learning_rate": 4.863636363636364e-05, | |
| "loss": 0.2462, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 14.978479385375977, | |
| "learning_rate": 4.840909090909091e-05, | |
| "loss": 0.2663, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 13.504115104675293, | |
| "learning_rate": 4.8181818181818186e-05, | |
| "loss": 0.3365, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 16.242828369140625, | |
| "learning_rate": 4.795454545454546e-05, | |
| "loss": 0.4131, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 13.582551002502441, | |
| "learning_rate": 4.772727272727273e-05, | |
| "loss": 0.3778, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.9506578947368421, | |
| "eval_loss": 0.14110171794891357, | |
| "eval_runtime": 6.0499, | |
| "eval_samples_per_second": 50.249, | |
| "eval_steps_per_second": 1.653, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 15.637821197509766, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.4292, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 38.1275749206543, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.2089, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.508777618408203, | |
| "learning_rate": 4.704545454545455e-05, | |
| "loss": 0.0328, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 7.219000339508057, | |
| "learning_rate": 4.681818181818182e-05, | |
| "loss": 0.1145, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 12.110613822937012, | |
| "learning_rate": 4.659090909090909e-05, | |
| "loss": 0.1894, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 4.768282890319824, | |
| "learning_rate": 4.636363636363636e-05, | |
| "loss": 0.0846, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 6.834845542907715, | |
| "learning_rate": 4.6136363636363635e-05, | |
| "loss": 0.1962, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 2.7093985080718994, | |
| "learning_rate": 4.5909090909090914e-05, | |
| "loss": 0.0618, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 5.430306434631348, | |
| "learning_rate": 4.5681818181818186e-05, | |
| "loss": 0.07, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 13.695512771606445, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.1126, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "eval_accuracy": 0.9506578947368421, | |
| "eval_loss": 0.12380365282297134, | |
| "eval_runtime": 6.1201, | |
| "eval_samples_per_second": 49.672, | |
| "eval_steps_per_second": 1.634, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 6.926172733306885, | |
| "learning_rate": 4.522727272727273e-05, | |
| "loss": 0.0518, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 9.466341972351074, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1033, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 17.950077056884766, | |
| "learning_rate": 4.4772727272727275e-05, | |
| "loss": 0.1797, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 34.07773971557617, | |
| "learning_rate": 4.454545454545455e-05, | |
| "loss": 0.5779, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 6.071528911590576, | |
| "learning_rate": 4.431818181818182e-05, | |
| "loss": 0.051, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 5.799258232116699, | |
| "learning_rate": 4.409090909090909e-05, | |
| "loss": 0.0435, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 18.01268768310547, | |
| "learning_rate": 4.386363636363637e-05, | |
| "loss": 0.1435, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 7.667330741882324, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0928, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 7.548737525939941, | |
| "learning_rate": 4.340909090909091e-05, | |
| "loss": 0.057, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 15.859737396240234, | |
| "learning_rate": 4.318181818181819e-05, | |
| "loss": 0.151, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.9671052631578947, | |
| "eval_loss": 0.08715644478797913, | |
| "eval_runtime": 6.0686, | |
| "eval_samples_per_second": 50.094, | |
| "eval_steps_per_second": 1.648, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 0.6900173425674438, | |
| "learning_rate": 4.295454545454546e-05, | |
| "loss": 0.0061, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 25.680091857910156, | |
| "learning_rate": 4.2727272727272724e-05, | |
| "loss": 0.1879, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 13.87479305267334, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.0599, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 10.368620872497559, | |
| "learning_rate": 4.2272727272727275e-05, | |
| "loss": 0.0233, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 10.640165328979492, | |
| "learning_rate": 4.204545454545455e-05, | |
| "loss": 0.1061, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 100.82930755615234, | |
| "learning_rate": 4.181818181818182e-05, | |
| "loss": 0.3247, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 13.85188102722168, | |
| "learning_rate": 4.159090909090909e-05, | |
| "loss": 0.0916, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 22.93225860595703, | |
| "learning_rate": 4.1363636363636364e-05, | |
| "loss": 0.1722, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 25.919754028320312, | |
| "learning_rate": 4.113636363636364e-05, | |
| "loss": 0.1576, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 11.392365455627441, | |
| "learning_rate": 4.0909090909090915e-05, | |
| "loss": 0.0464, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "eval_accuracy": 0.9638157894736842, | |
| "eval_loss": 0.0998014286160469, | |
| "eval_runtime": 6.1785, | |
| "eval_samples_per_second": 49.203, | |
| "eval_steps_per_second": 1.619, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.22, | |
| "grad_norm": 14.013245582580566, | |
| "learning_rate": 4.068181818181818e-05, | |
| "loss": 0.0405, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.25485870242118835, | |
| "learning_rate": 4.045454545454546e-05, | |
| "loss": 0.0013, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 5.155923843383789, | |
| "learning_rate": 4.022727272727273e-05, | |
| "loss": 0.0179, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "grad_norm": 93.75792694091797, | |
| "learning_rate": 4e-05, | |
| "loss": 0.321, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "grad_norm": 14.470589637756348, | |
| "learning_rate": 3.9772727272727275e-05, | |
| "loss": 0.0336, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 16.80307388305664, | |
| "learning_rate": 3.954545454545455e-05, | |
| "loss": 0.1538, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 15.844971656799316, | |
| "learning_rate": 3.931818181818182e-05, | |
| "loss": 0.0616, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "grad_norm": 10.336301803588867, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 0.0428, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "grad_norm": 35.54351043701172, | |
| "learning_rate": 3.8863636363636364e-05, | |
| "loss": 0.0874, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 2.9273905754089355, | |
| "learning_rate": 3.8636363636363636e-05, | |
| "loss": 0.0154, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "eval_accuracy": 0.9572368421052632, | |
| "eval_loss": 0.10613898187875748, | |
| "eval_runtime": 6.2452, | |
| "eval_samples_per_second": 48.677, | |
| "eval_steps_per_second": 1.601, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "grad_norm": 1.2449793815612793, | |
| "learning_rate": 3.840909090909091e-05, | |
| "loss": 0.0048, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 4.352685928344727, | |
| "learning_rate": 3.818181818181819e-05, | |
| "loss": 0.0103, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 24.665245056152344, | |
| "learning_rate": 3.795454545454545e-05, | |
| "loss": 0.0784, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "grad_norm": 3.592439651489258, | |
| "learning_rate": 3.7727272727272725e-05, | |
| "loss": 0.0184, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "grad_norm": 4.012323379516602, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0116, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.87, | |
| "grad_norm": 3.2316153049468994, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 0.0325, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.8774870038032532, | |
| "learning_rate": 3.704545454545455e-05, | |
| "loss": 0.0057, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 14.955449104309082, | |
| "learning_rate": 3.681818181818182e-05, | |
| "loss": 0.1373, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 50.09528350830078, | |
| "learning_rate": 3.659090909090909e-05, | |
| "loss": 0.153, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 6.00044584274292, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0166, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "eval_accuracy": 0.9539473684210527, | |
| "eval_loss": 0.12729084491729736, | |
| "eval_runtime": 6.4657, | |
| "eval_samples_per_second": 47.017, | |
| "eval_steps_per_second": 1.547, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 2.869675636291504, | |
| "learning_rate": 3.613636363636364e-05, | |
| "loss": 0.0055, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 4.115776062011719, | |
| "learning_rate": 3.590909090909091e-05, | |
| "loss": 0.0262, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "grad_norm": 4.357750415802002, | |
| "learning_rate": 3.568181818181818e-05, | |
| "loss": 0.0128, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 3.22, | |
| "grad_norm": 2.3195817470550537, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 0.0103, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "grad_norm": 2.9432053565979004, | |
| "learning_rate": 3.522727272727273e-05, | |
| "loss": 0.0052, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 1.4674235582351685, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.005, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.35, | |
| "grad_norm": 0.4246145188808441, | |
| "learning_rate": 3.4772727272727276e-05, | |
| "loss": 0.0009, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 52.41901397705078, | |
| "learning_rate": 3.454545454545455e-05, | |
| "loss": 0.3822, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.43, | |
| "grad_norm": 44.01215362548828, | |
| "learning_rate": 3.431818181818182e-05, | |
| "loss": 0.1986, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "grad_norm": 36.397560119628906, | |
| "learning_rate": 3.409090909090909e-05, | |
| "loss": 0.2452, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "eval_accuracy": 0.9539473684210527, | |
| "eval_loss": 0.15055274963378906, | |
| "eval_runtime": 6.3479, | |
| "eval_samples_per_second": 47.89, | |
| "eval_steps_per_second": 1.575, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 21.1817626953125, | |
| "learning_rate": 3.3863636363636364e-05, | |
| "loss": 0.3099, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 3.57, | |
| "grad_norm": 20.235713958740234, | |
| "learning_rate": 3.3636363636363636e-05, | |
| "loss": 0.0407, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 3.61, | |
| "grad_norm": 0.6778246760368347, | |
| "learning_rate": 3.3409090909090915e-05, | |
| "loss": 0.0031, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 3.65, | |
| "grad_norm": 2.1587185859680176, | |
| "learning_rate": 3.318181818181819e-05, | |
| "loss": 0.0079, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 1.5716959238052368, | |
| "learning_rate": 3.295454545454545e-05, | |
| "loss": 0.003, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.74, | |
| "grad_norm": 3.423281669616699, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.0058, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "grad_norm": 1.5846844911575317, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.0033, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 3.83, | |
| "grad_norm": 0.27800872921943665, | |
| "learning_rate": 3.2272727272727276e-05, | |
| "loss": 0.0009, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "grad_norm": 14.288936614990234, | |
| "learning_rate": 3.204545454545455e-05, | |
| "loss": 0.1006, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "grad_norm": 4.256282806396484, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 0.0265, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.91, | |
| "eval_accuracy": 0.8552631578947368, | |
| "eval_loss": 0.5866047739982605, | |
| "eval_runtime": 6.3891, | |
| "eval_samples_per_second": 47.581, | |
| "eval_steps_per_second": 1.565, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "grad_norm": 25.256847381591797, | |
| "learning_rate": 3.159090909090909e-05, | |
| "loss": 0.0574, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.033596429973840714, | |
| "learning_rate": 3.1363636363636365e-05, | |
| "loss": 0.0001, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 19.746484756469727, | |
| "learning_rate": 3.113636363636364e-05, | |
| "loss": 0.0874, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 1.464240312576294, | |
| "learning_rate": 3.090909090909091e-05, | |
| "loss": 0.002, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "grad_norm": 0.09598905593156815, | |
| "learning_rate": 3.068181818181818e-05, | |
| "loss": 0.0004, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.17, | |
| "grad_norm": 0.7159618735313416, | |
| "learning_rate": 3.0454545454545456e-05, | |
| "loss": 0.0016, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 4.22, | |
| "grad_norm": 6.904734134674072, | |
| "learning_rate": 3.0227272727272725e-05, | |
| "loss": 0.0242, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "grad_norm": 5.715118885040283, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0319, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 2.873875856399536, | |
| "learning_rate": 2.9772727272727273e-05, | |
| "loss": 0.0061, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "grad_norm": 21.81316375732422, | |
| "learning_rate": 2.954545454545455e-05, | |
| "loss": 0.2385, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "eval_accuracy": 0.9638157894736842, | |
| "eval_loss": 0.14010807871818542, | |
| "eval_runtime": 6.5543, | |
| "eval_samples_per_second": 46.382, | |
| "eval_steps_per_second": 1.526, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "grad_norm": 11.216484069824219, | |
| "learning_rate": 2.9318181818181817e-05, | |
| "loss": 0.0667, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "grad_norm": 2.7263998985290527, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.0055, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "grad_norm": 2.389941453933716, | |
| "learning_rate": 2.8863636363636365e-05, | |
| "loss": 0.0067, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 10.113710403442383, | |
| "learning_rate": 2.863636363636364e-05, | |
| "loss": 0.0483, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 4.57, | |
| "grad_norm": 2.801983594894409, | |
| "learning_rate": 2.8409090909090912e-05, | |
| "loss": 0.0063, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 4.61, | |
| "grad_norm": 40.15452575683594, | |
| "learning_rate": 2.818181818181818e-05, | |
| "loss": 0.1347, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "grad_norm": 29.40395736694336, | |
| "learning_rate": 2.7954545454545457e-05, | |
| "loss": 0.3017, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 6.391096591949463, | |
| "learning_rate": 2.772727272727273e-05, | |
| "loss": 0.0121, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "grad_norm": 5.7057342529296875, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0088, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "grad_norm": 12.210604667663574, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.0378, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.78, | |
| "eval_accuracy": 0.9572368421052632, | |
| "eval_loss": 0.14174394309520721, | |
| "eval_runtime": 6.4084, | |
| "eval_samples_per_second": 47.437, | |
| "eval_steps_per_second": 1.56, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.83, | |
| "grad_norm": 0.04858306050300598, | |
| "learning_rate": 2.7045454545454545e-05, | |
| "loss": 0.0002, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "grad_norm": 0.39052003622055054, | |
| "learning_rate": 2.681818181818182e-05, | |
| "loss": 0.001, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "grad_norm": 4.1193084716796875, | |
| "learning_rate": 2.6590909090909093e-05, | |
| "loss": 0.0109, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "grad_norm": 1.6979681253433228, | |
| "learning_rate": 2.636363636363636e-05, | |
| "loss": 0.0084, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.00023737615265417844, | |
| "learning_rate": 2.6136363636363637e-05, | |
| "loss": 0.0, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 11.732938766479492, | |
| "learning_rate": 2.590909090909091e-05, | |
| "loss": 0.0498, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 3.312427043914795, | |
| "learning_rate": 2.5681818181818185e-05, | |
| "loss": 0.0073, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 5.13, | |
| "grad_norm": 0.23251380026340485, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.0004, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "grad_norm": 0.2869594395160675, | |
| "learning_rate": 2.5227272727272726e-05, | |
| "loss": 0.0006, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "grad_norm": 2.0357539653778076, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0031, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "eval_accuracy": 0.9605263157894737, | |
| "eval_loss": 0.1753275841474533, | |
| "eval_runtime": 6.4879, | |
| "eval_samples_per_second": 46.857, | |
| "eval_steps_per_second": 1.541, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "grad_norm": 10.356167793273926, | |
| "learning_rate": 2.4772727272727277e-05, | |
| "loss": 0.1893, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "grad_norm": 0.02243722230195999, | |
| "learning_rate": 2.4545454545454545e-05, | |
| "loss": 0.0001, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 5.35, | |
| "grad_norm": 1.1501038074493408, | |
| "learning_rate": 2.431818181818182e-05, | |
| "loss": 0.0032, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 5.39, | |
| "grad_norm": 10.073415756225586, | |
| "learning_rate": 2.4090909090909093e-05, | |
| "loss": 0.0374, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 5.43, | |
| "grad_norm": 0.24568603932857513, | |
| "learning_rate": 2.3863636363636365e-05, | |
| "loss": 0.0003, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 5.48, | |
| "grad_norm": 0.036038391292095184, | |
| "learning_rate": 2.3636363636363637e-05, | |
| "loss": 0.0002, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 5.52, | |
| "grad_norm": 0.05611402168869972, | |
| "learning_rate": 2.340909090909091e-05, | |
| "loss": 0.0001, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 5.57, | |
| "grad_norm": 8.793002128601074, | |
| "learning_rate": 2.318181818181818e-05, | |
| "loss": 0.0368, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "grad_norm": 0.7039275169372559, | |
| "learning_rate": 2.2954545454545457e-05, | |
| "loss": 0.0023, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 0.2844730615615845, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.0011, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_accuracy": 0.9473684210526315, | |
| "eval_loss": 0.22137421369552612, | |
| "eval_runtime": 6.4167, | |
| "eval_samples_per_second": 47.377, | |
| "eval_steps_per_second": 1.558, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.7, | |
| "grad_norm": 0.3269338309764862, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0011, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 5.74, | |
| "grad_norm": 0.37626028060913086, | |
| "learning_rate": 2.2272727272727274e-05, | |
| "loss": 0.0007, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "grad_norm": 5.648035526275635, | |
| "learning_rate": 2.2045454545454546e-05, | |
| "loss": 0.011, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 5.83, | |
| "grad_norm": 6.630437850952148, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0181, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 5.87, | |
| "grad_norm": 0.35264378786087036, | |
| "learning_rate": 2.1590909090909093e-05, | |
| "loss": 0.0005, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "grad_norm": 3.137737274169922, | |
| "learning_rate": 2.1363636363636362e-05, | |
| "loss": 0.0029, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 0.02204626239836216, | |
| "learning_rate": 2.1136363636363638e-05, | |
| "loss": 0.0001, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.006504398304969072, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 0.0, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 2.309501886367798, | |
| "learning_rate": 2.0681818181818182e-05, | |
| "loss": 0.0042, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 0.41132238507270813, | |
| "learning_rate": 2.0454545454545457e-05, | |
| "loss": 0.001, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "eval_accuracy": 0.9769736842105263, | |
| "eval_loss": 0.09925788640975952, | |
| "eval_runtime": 6.5113, | |
| "eval_samples_per_second": 46.688, | |
| "eval_steps_per_second": 1.536, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.13, | |
| "grad_norm": 0.004451736807823181, | |
| "learning_rate": 2.022727272727273e-05, | |
| "loss": 0.0, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "grad_norm": 0.6769189238548279, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0018, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "grad_norm": 1.85093355178833, | |
| "learning_rate": 1.9772727272727274e-05, | |
| "loss": 0.005, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 6.26, | |
| "grad_norm": 7.151664733886719, | |
| "learning_rate": 1.9545454545454546e-05, | |
| "loss": 0.0124, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 6.3, | |
| "grad_norm": 2.592489004135132, | |
| "learning_rate": 1.9318181818181818e-05, | |
| "loss": 0.0048, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "grad_norm": 0.04051464423537254, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 0.0001, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 6.39, | |
| "grad_norm": 19.51251983642578, | |
| "learning_rate": 1.8863636363636362e-05, | |
| "loss": 0.0214, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 6.43, | |
| "grad_norm": 0.008436283096671104, | |
| "learning_rate": 1.8636363636363638e-05, | |
| "loss": 0.0, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 6.48, | |
| "grad_norm": 0.11793872714042664, | |
| "learning_rate": 1.840909090909091e-05, | |
| "loss": 0.0005, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "grad_norm": 0.016041185706853867, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0001, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.52, | |
| "eval_accuracy": 0.9736842105263158, | |
| "eval_loss": 0.1299658864736557, | |
| "eval_runtime": 6.5416, | |
| "eval_samples_per_second": 46.472, | |
| "eval_steps_per_second": 1.529, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.57, | |
| "grad_norm": 0.034806687384843826, | |
| "learning_rate": 1.7954545454545454e-05, | |
| "loss": 0.0002, | |
| "step": 151 | |
| }, | |
| { | |
| "epoch": 6.61, | |
| "grad_norm": 0.06806668639183044, | |
| "learning_rate": 1.772727272727273e-05, | |
| "loss": 0.0002, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 6.65, | |
| "grad_norm": 0.03514229133725166, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0001, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 6.7, | |
| "grad_norm": 0.04649640619754791, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 0.0002, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 6.74, | |
| "grad_norm": 0.031281836330890656, | |
| "learning_rate": 1.7045454545454546e-05, | |
| "loss": 0.0002, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 6.78, | |
| "grad_norm": 0.040468715131282806, | |
| "learning_rate": 1.6818181818181818e-05, | |
| "loss": 0.0001, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "grad_norm": 0.11787349730730057, | |
| "learning_rate": 1.6590909090909094e-05, | |
| "loss": 0.0005, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 6.87, | |
| "grad_norm": 0.09801948070526123, | |
| "learning_rate": 1.6363636363636366e-05, | |
| "loss": 0.0002, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "grad_norm": 0.025944946333765984, | |
| "learning_rate": 1.6136363636363638e-05, | |
| "loss": 0.0001, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "grad_norm": 0.21682406961917877, | |
| "learning_rate": 1.590909090909091e-05, | |
| "loss": 0.0006, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "eval_accuracy": 0.9605263157894737, | |
| "eval_loss": 0.16923262178897858, | |
| "eval_runtime": 6.5686, | |
| "eval_samples_per_second": 46.281, | |
| "eval_steps_per_second": 1.522, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.19760921597480774, | |
| "learning_rate": 1.5681818181818182e-05, | |
| "loss": 0.0003, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 3.799219846725464, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 0.0072, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "grad_norm": 0.028768137097358704, | |
| "learning_rate": 1.5227272727272728e-05, | |
| "loss": 0.0002, | |
| "step": 163 | |
| }, | |
| { | |
| "epoch": 7.13, | |
| "grad_norm": 0.8019386529922485, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0019, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 7.17, | |
| "grad_norm": 0.017596999183297157, | |
| "learning_rate": 1.4772727272727274e-05, | |
| "loss": 0.0, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 7.22, | |
| "grad_norm": 0.03721166402101517, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0001, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 7.26, | |
| "grad_norm": 1.2334744930267334, | |
| "learning_rate": 1.431818181818182e-05, | |
| "loss": 0.0022, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 7.3, | |
| "grad_norm": 1.070383071899414, | |
| "learning_rate": 1.409090909090909e-05, | |
| "loss": 0.0037, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 7.35, | |
| "grad_norm": 0.0028092290740460157, | |
| "learning_rate": 1.3863636363636364e-05, | |
| "loss": 0.0, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "grad_norm": 36.44822692871094, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.1166, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.39, | |
| "eval_accuracy": 0.9703947368421053, | |
| "eval_loss": 0.123698890209198, | |
| "eval_runtime": 6.8146, | |
| "eval_samples_per_second": 44.61, | |
| "eval_steps_per_second": 1.467, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "grad_norm": 0.0008054708014242351, | |
| "learning_rate": 1.340909090909091e-05, | |
| "loss": 0.0, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 7.48, | |
| "grad_norm": 0.001217369339428842, | |
| "learning_rate": 1.318181818181818e-05, | |
| "loss": 0.0, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 7.52, | |
| "grad_norm": 0.026522669941186905, | |
| "learning_rate": 1.2954545454545455e-05, | |
| "loss": 0.0002, | |
| "step": 173 | |
| }, | |
| { | |
| "epoch": 7.57, | |
| "grad_norm": 0.0015209235716611147, | |
| "learning_rate": 1.2727272727272727e-05, | |
| "loss": 0.0, | |
| "step": 174 | |
| }, | |
| { | |
| "epoch": 7.61, | |
| "grad_norm": 0.014780080877244473, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0001, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "grad_norm": 0.01029743067920208, | |
| "learning_rate": 1.2272727272727273e-05, | |
| "loss": 0.0, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 7.7, | |
| "grad_norm": 0.025874657556414604, | |
| "learning_rate": 1.2045454545454547e-05, | |
| "loss": 0.0002, | |
| "step": 177 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "grad_norm": 0.025635922327637672, | |
| "learning_rate": 1.1818181818181819e-05, | |
| "loss": 0.0001, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "grad_norm": 0.0031894599087536335, | |
| "learning_rate": 1.159090909090909e-05, | |
| "loss": 0.0, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "grad_norm": 0.03724868968129158, | |
| "learning_rate": 1.1363636363636365e-05, | |
| "loss": 0.0002, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "eval_accuracy": 0.9671052631578947, | |
| "eval_loss": 0.14251494407653809, | |
| "eval_runtime": 6.5105, | |
| "eval_samples_per_second": 46.694, | |
| "eval_steps_per_second": 1.536, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.87, | |
| "grad_norm": 0.006170081906020641, | |
| "learning_rate": 1.1136363636363637e-05, | |
| "loss": 0.0, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "grad_norm": 0.042060792446136475, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0001, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 7.96, | |
| "grad_norm": 0.02992098405957222, | |
| "learning_rate": 1.0681818181818181e-05, | |
| "loss": 0.0001, | |
| "step": 183 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 61.35695266723633, | |
| "learning_rate": 1.0454545454545455e-05, | |
| "loss": 0.1738, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "grad_norm": 0.13377808034420013, | |
| "learning_rate": 1.0227272727272729e-05, | |
| "loss": 0.0003, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "grad_norm": 0.03412250056862831, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0001, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 8.13, | |
| "grad_norm": 0.0030899515841156244, | |
| "learning_rate": 9.772727272727273e-06, | |
| "loss": 0.0, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 8.17, | |
| "grad_norm": 0.011750655248761177, | |
| "learning_rate": 9.545454545454547e-06, | |
| "loss": 0.0, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 8.22, | |
| "grad_norm": 0.054395273327827454, | |
| "learning_rate": 9.318181818181819e-06, | |
| "loss": 0.0001, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "grad_norm": 0.0035029698628932238, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.26, | |
| "eval_accuracy": 0.9703947368421053, | |
| "eval_loss": 0.1182524561882019, | |
| "eval_runtime": 6.7162, | |
| "eval_samples_per_second": 45.264, | |
| "eval_steps_per_second": 1.489, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.3, | |
| "grad_norm": 4.122413158416748, | |
| "learning_rate": 8.863636363636365e-06, | |
| "loss": 0.0225, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "grad_norm": 0.0010495582828298211, | |
| "learning_rate": 8.636363636363637e-06, | |
| "loss": 0.0, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 8.39, | |
| "grad_norm": 1.0246374607086182, | |
| "learning_rate": 8.409090909090909e-06, | |
| "loss": 0.0014, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 8.43, | |
| "grad_norm": 0.004268502350896597, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.0, | |
| "step": 194 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "grad_norm": 0.03880951553583145, | |
| "learning_rate": 7.954545454545455e-06, | |
| "loss": 0.0001, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 8.52, | |
| "grad_norm": 0.021236296743154526, | |
| "learning_rate": 7.727272727272727e-06, | |
| "loss": 0.0, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 8.57, | |
| "grad_norm": 0.05536981672048569, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.0002, | |
| "step": 197 | |
| }, | |
| { | |
| "epoch": 8.61, | |
| "grad_norm": 0.14701944589614868, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.0002, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 8.65, | |
| "grad_norm": 1.7185938358306885, | |
| "learning_rate": 7.045454545454545e-06, | |
| "loss": 0.0013, | |
| "step": 199 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "grad_norm": 0.3261477053165436, | |
| "learning_rate": 6.818181818181818e-06, | |
| "loss": 0.0005, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.7, | |
| "eval_accuracy": 0.9671052631578947, | |
| "eval_loss": 0.13489210605621338, | |
| "eval_runtime": 6.5689, | |
| "eval_samples_per_second": 46.278, | |
| "eval_steps_per_second": 1.522, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.74, | |
| "grad_norm": 0.0023177217226475477, | |
| "learning_rate": 6.59090909090909e-06, | |
| "loss": 0.0, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "grad_norm": 0.03938745707273483, | |
| "learning_rate": 6.363636363636363e-06, | |
| "loss": 0.0001, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 8.83, | |
| "grad_norm": 0.009258226491510868, | |
| "learning_rate": 6.136363636363636e-06, | |
| "loss": 0.0, | |
| "step": 203 | |
| }, | |
| { | |
| "epoch": 8.87, | |
| "grad_norm": 0.019902734085917473, | |
| "learning_rate": 5.909090909090909e-06, | |
| "loss": 0.0001, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "grad_norm": 0.0311611145734787, | |
| "learning_rate": 5.681818181818182e-06, | |
| "loss": 0.0001, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "grad_norm": 30.388660430908203, | |
| "learning_rate": 5.4545454545454545e-06, | |
| "loss": 0.0178, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.0003517134173307568, | |
| "learning_rate": 5.2272727272727274e-06, | |
| "loss": 0.0, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.04, | |
| "grad_norm": 11.453664779663086, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0121, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "grad_norm": 0.0006242034723982215, | |
| "learning_rate": 4.772727272727273e-06, | |
| "loss": 0.0, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "grad_norm": 1.1630523204803467, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.002, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.13, | |
| "eval_accuracy": 0.9703947368421053, | |
| "eval_loss": 0.13131581246852875, | |
| "eval_runtime": 6.4558, | |
| "eval_samples_per_second": 47.089, | |
| "eval_steps_per_second": 1.549, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.17, | |
| "grad_norm": 0.011197259649634361, | |
| "learning_rate": 4.3181818181818185e-06, | |
| "loss": 0.0, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 9.22, | |
| "grad_norm": 0.125521719455719, | |
| "learning_rate": 4.0909090909090915e-06, | |
| "loss": 0.0002, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 9.26, | |
| "grad_norm": 0.017533954232931137, | |
| "learning_rate": 3.863636363636364e-06, | |
| "loss": 0.0001, | |
| "step": 213 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "grad_norm": 0.012250928208231926, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "grad_norm": 0.003785262815654278, | |
| "learning_rate": 3.409090909090909e-06, | |
| "loss": 0.0, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 9.39, | |
| "grad_norm": 0.004202316515147686, | |
| "learning_rate": 3.1818181818181817e-06, | |
| "loss": 0.0, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 9.43, | |
| "grad_norm": 0.10783527046442032, | |
| "learning_rate": 2.9545454545454547e-06, | |
| "loss": 0.0004, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 9.48, | |
| "grad_norm": 0.08085985481739044, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.0002, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "grad_norm": 0.03627089783549309, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0001, | |
| "step": 219 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "grad_norm": 3.906628370285034, | |
| "learning_rate": 2.2727272727272728e-06, | |
| "loss": 0.006, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.57, | |
| "eval_accuracy": 0.9736842105263158, | |
| "eval_loss": 0.12350568920373917, | |
| "eval_runtime": 6.4757, | |
| "eval_samples_per_second": 46.945, | |
| "eval_steps_per_second": 1.544, | |
| "step": 220 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 230, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 10, | |
| "total_flos": 5.318096316030812e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |