| { | |
| "best_metric": 0.9906542056074766, | |
| "best_model_checkpoint": "finetuned-for-YogaPosesv4/checkpoint-800", | |
| "epoch": 12.0, | |
| "eval_steps": 100, | |
| "global_step": 1368, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08771929824561403, | |
| "grad_norm": 7.6294169425964355, | |
| "learning_rate": 0.00019853801169590645, | |
| "loss": 0.1835, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 2.819319009780884, | |
| "learning_rate": 0.00019707602339181287, | |
| "loss": 0.1356, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2631578947368421, | |
| "grad_norm": 7.1533966064453125, | |
| "learning_rate": 0.0001956140350877193, | |
| "loss": 0.0432, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 6.63047456741333, | |
| "learning_rate": 0.00019415204678362573, | |
| "loss": 0.1341, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "grad_norm": 7.353053092956543, | |
| "learning_rate": 0.00019269005847953217, | |
| "loss": 0.0872, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 0.5175597071647644, | |
| "learning_rate": 0.0001912280701754386, | |
| "loss": 0.0915, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.6140350877192983, | |
| "grad_norm": 2.7669591903686523, | |
| "learning_rate": 0.00018976608187134503, | |
| "loss": 0.0663, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 15.684842109680176, | |
| "learning_rate": 0.00018830409356725147, | |
| "loss": 0.0644, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7894736842105263, | |
| "grad_norm": 12.486333847045898, | |
| "learning_rate": 0.00018684210526315792, | |
| "loss": 0.1148, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "grad_norm": 2.681043863296509, | |
| "learning_rate": 0.00018538011695906433, | |
| "loss": 0.0434, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "eval_accuracy": 0.9595015576323987, | |
| "eval_loss": 0.1421448141336441, | |
| "eval_runtime": 35.5517, | |
| "eval_samples_per_second": 9.029, | |
| "eval_steps_per_second": 1.153, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9649122807017544, | |
| "grad_norm": 0.4669760465621948, | |
| "learning_rate": 0.00018391812865497077, | |
| "loss": 0.1035, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 1.2481253147125244, | |
| "learning_rate": 0.0001824561403508772, | |
| "loss": 0.0242, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1403508771929824, | |
| "grad_norm": 0.747638463973999, | |
| "learning_rate": 0.00018099415204678363, | |
| "loss": 0.0911, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2280701754385965, | |
| "grad_norm": 4.2482428550720215, | |
| "learning_rate": 0.00017953216374269005, | |
| "loss": 0.0243, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3157894736842106, | |
| "grad_norm": 14.569940567016602, | |
| "learning_rate": 0.0001780701754385965, | |
| "loss": 0.0906, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 6.877868175506592, | |
| "learning_rate": 0.00017660818713450294, | |
| "loss": 0.0318, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.4912280701754386, | |
| "grad_norm": 0.7561317086219788, | |
| "learning_rate": 0.00017514619883040938, | |
| "loss": 0.0707, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 3.7642083168029785, | |
| "learning_rate": 0.0001736842105263158, | |
| "loss": 0.0514, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.7942452430725098, | |
| "learning_rate": 0.00017222222222222224, | |
| "loss": 0.0531, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 0.05782110244035721, | |
| "learning_rate": 0.00017076023391812865, | |
| "loss": 0.033, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "eval_accuracy": 0.9875389408099688, | |
| "eval_loss": 0.0879807248711586, | |
| "eval_runtime": 40.2074, | |
| "eval_samples_per_second": 7.984, | |
| "eval_steps_per_second": 1.02, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8421052631578947, | |
| "grad_norm": 13.322481155395508, | |
| "learning_rate": 0.0001692982456140351, | |
| "loss": 0.1105, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9298245614035088, | |
| "grad_norm": 0.9878503084182739, | |
| "learning_rate": 0.0001678362573099415, | |
| "loss": 0.0581, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.017543859649123, | |
| "grad_norm": 3.3777120113372803, | |
| "learning_rate": 0.00016637426900584796, | |
| "loss": 0.0391, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 0.03219028189778328, | |
| "learning_rate": 0.0001649122807017544, | |
| "loss": 0.0413, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.192982456140351, | |
| "grad_norm": 11.16067123413086, | |
| "learning_rate": 0.00016345029239766084, | |
| "loss": 0.0693, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.280701754385965, | |
| "grad_norm": 17.428421020507812, | |
| "learning_rate": 0.00016198830409356726, | |
| "loss": 0.1285, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3684210526315788, | |
| "grad_norm": 3.1800684928894043, | |
| "learning_rate": 0.0001605263157894737, | |
| "loss": 0.0185, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.456140350877193, | |
| "grad_norm": 0.18285800516605377, | |
| "learning_rate": 0.00015906432748538012, | |
| "loss": 0.0343, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.543859649122807, | |
| "grad_norm": 4.973541259765625, | |
| "learning_rate": 0.00015760233918128656, | |
| "loss": 0.053, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 0.5851675271987915, | |
| "learning_rate": 0.00015614035087719297, | |
| "loss": 0.084, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "eval_accuracy": 0.9844236760124611, | |
| "eval_loss": 0.09186869114637375, | |
| "eval_runtime": 39.6899, | |
| "eval_samples_per_second": 8.088, | |
| "eval_steps_per_second": 1.033, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.719298245614035, | |
| "grad_norm": 0.42921513319015503, | |
| "learning_rate": 0.00015467836257309942, | |
| "loss": 0.0589, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "grad_norm": 14.734365463256836, | |
| "learning_rate": 0.00015321637426900586, | |
| "loss": 0.0975, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.8947368421052633, | |
| "grad_norm": 3.1166839599609375, | |
| "learning_rate": 0.0001517543859649123, | |
| "loss": 0.0401, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.982456140350877, | |
| "grad_norm": 3.258279800415039, | |
| "learning_rate": 0.00015029239766081872, | |
| "loss": 0.0549, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.0701754385964914, | |
| "grad_norm": 0.8289499878883362, | |
| "learning_rate": 0.00014883040935672514, | |
| "loss": 0.0592, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 0.16162537038326263, | |
| "learning_rate": 0.00014736842105263158, | |
| "loss": 0.1327, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.245614035087719, | |
| "grad_norm": 3.4843053817749023, | |
| "learning_rate": 0.00014590643274853802, | |
| "loss": 0.0478, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 0.20168577134609222, | |
| "learning_rate": 0.00014444444444444444, | |
| "loss": 0.0901, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.4210526315789473, | |
| "grad_norm": 3.0230636596679688, | |
| "learning_rate": 0.00014298245614035088, | |
| "loss": 0.0749, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.5087719298245617, | |
| "grad_norm": 4.313819408416748, | |
| "learning_rate": 0.00014152046783625732, | |
| "loss": 0.0655, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.5087719298245617, | |
| "eval_accuracy": 0.9875389408099688, | |
| "eval_loss": 0.09323806315660477, | |
| "eval_runtime": 37.6349, | |
| "eval_samples_per_second": 8.529, | |
| "eval_steps_per_second": 1.089, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.5964912280701755, | |
| "grad_norm": 0.0746302455663681, | |
| "learning_rate": 0.00014005847953216377, | |
| "loss": 0.0508, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.6842105263157894, | |
| "grad_norm": 12.243117332458496, | |
| "learning_rate": 0.00013859649122807018, | |
| "loss": 0.0327, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.7719298245614032, | |
| "grad_norm": 0.5793654918670654, | |
| "learning_rate": 0.0001371345029239766, | |
| "loss": 0.05, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.8596491228070176, | |
| "grad_norm": 0.30418357253074646, | |
| "learning_rate": 0.00013567251461988304, | |
| "loss": 0.0135, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.9473684210526314, | |
| "grad_norm": 0.7474086284637451, | |
| "learning_rate": 0.00013421052631578948, | |
| "loss": 0.0464, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.035087719298246, | |
| "grad_norm": 3.3401076793670654, | |
| "learning_rate": 0.0001327485380116959, | |
| "loss": 0.181, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.12280701754386, | |
| "grad_norm": 3.931779623031616, | |
| "learning_rate": 0.00013128654970760234, | |
| "loss": 0.0092, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 0.41829147934913635, | |
| "learning_rate": 0.0001298245614035088, | |
| "loss": 0.0177, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.298245614035087, | |
| "grad_norm": 0.22863353788852692, | |
| "learning_rate": 0.00012836257309941523, | |
| "loss": 0.0183, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "grad_norm": 0.01852666214108467, | |
| "learning_rate": 0.00012690058479532165, | |
| "loss": 0.0126, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "eval_accuracy": 0.9875389408099688, | |
| "eval_loss": 0.06964406371116638, | |
| "eval_runtime": 36.8852, | |
| "eval_samples_per_second": 8.703, | |
| "eval_steps_per_second": 1.112, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.473684210526316, | |
| "grad_norm": 1.5573415756225586, | |
| "learning_rate": 0.00012543859649122806, | |
| "loss": 0.0668, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.56140350877193, | |
| "grad_norm": 0.009869000874459743, | |
| "learning_rate": 0.0001239766081871345, | |
| "loss": 0.0137, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.649122807017544, | |
| "grad_norm": 0.20358864963054657, | |
| "learning_rate": 0.00012251461988304095, | |
| "loss": 0.0164, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.7368421052631575, | |
| "grad_norm": 0.5513414144515991, | |
| "learning_rate": 0.00012105263157894738, | |
| "loss": 0.028, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.824561403508772, | |
| "grad_norm": 3.767604351043701, | |
| "learning_rate": 0.0001195906432748538, | |
| "loss": 0.016, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.912280701754386, | |
| "grad_norm": 0.5676788091659546, | |
| "learning_rate": 0.00011812865497076025, | |
| "loss": 0.037, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 94.76567077636719, | |
| "learning_rate": 0.00011666666666666668, | |
| "loss": 0.2472, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.087719298245614, | |
| "grad_norm": 0.013286423869431019, | |
| "learning_rate": 0.00011520467836257311, | |
| "loss": 0.0449, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.175438596491228, | |
| "grad_norm": 2.3555777072906494, | |
| "learning_rate": 0.00011374269005847952, | |
| "loss": 0.1002, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.2631578947368425, | |
| "grad_norm": 21.486955642700195, | |
| "learning_rate": 0.00011228070175438597, | |
| "loss": 0.0487, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.2631578947368425, | |
| "eval_accuracy": 0.9719626168224299, | |
| "eval_loss": 0.08467654883861542, | |
| "eval_runtime": 35.5793, | |
| "eval_samples_per_second": 9.022, | |
| "eval_steps_per_second": 1.152, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.350877192982456, | |
| "grad_norm": 3.456000328063965, | |
| "learning_rate": 0.0001108187134502924, | |
| "loss": 0.0451, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.43859649122807, | |
| "grad_norm": 0.21063438057899475, | |
| "learning_rate": 0.00010935672514619884, | |
| "loss": 0.0016, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.526315789473684, | |
| "grad_norm": 0.007295363582670689, | |
| "learning_rate": 0.00010789473684210527, | |
| "loss": 0.0298, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.614035087719298, | |
| "grad_norm": 1.68939208984375, | |
| "learning_rate": 0.00010643274853801171, | |
| "loss": 0.0175, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.701754385964913, | |
| "grad_norm": 0.7690061330795288, | |
| "learning_rate": 0.00010497076023391814, | |
| "loss": 0.0565, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.7894736842105265, | |
| "grad_norm": 0.004914429504424334, | |
| "learning_rate": 0.00010350877192982457, | |
| "loss": 0.0151, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.87719298245614, | |
| "grad_norm": 2.762733221054077, | |
| "learning_rate": 0.00010204678362573099, | |
| "loss": 0.0428, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 5.964912280701754, | |
| "grad_norm": 1.9332022666931152, | |
| "learning_rate": 0.00010058479532163743, | |
| "loss": 0.0335, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.052631578947368, | |
| "grad_norm": 0.0930035263299942, | |
| "learning_rate": 9.912280701754386e-05, | |
| "loss": 0.0342, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.140350877192983, | |
| "grad_norm": 0.03953389823436737, | |
| "learning_rate": 9.76608187134503e-05, | |
| "loss": 0.0114, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.140350877192983, | |
| "eval_accuracy": 0.9813084112149533, | |
| "eval_loss": 0.11030136793851852, | |
| "eval_runtime": 35.4585, | |
| "eval_samples_per_second": 9.053, | |
| "eval_steps_per_second": 1.156, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.228070175438597, | |
| "grad_norm": 0.06557751446962357, | |
| "learning_rate": 9.619883040935673e-05, | |
| "loss": 0.0089, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 11.056671142578125, | |
| "learning_rate": 9.473684210526316e-05, | |
| "loss": 0.0294, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.4035087719298245, | |
| "grad_norm": 0.26337969303131104, | |
| "learning_rate": 9.327485380116959e-05, | |
| "loss": 0.0213, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.491228070175438, | |
| "grad_norm": 0.15492522716522217, | |
| "learning_rate": 9.181286549707603e-05, | |
| "loss": 0.0469, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.578947368421053, | |
| "grad_norm": 0.013370133936405182, | |
| "learning_rate": 9.035087719298246e-05, | |
| "loss": 0.0004, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.10022695362567902, | |
| "learning_rate": 8.888888888888889e-05, | |
| "loss": 0.0255, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.754385964912281, | |
| "grad_norm": 0.0027779792435467243, | |
| "learning_rate": 8.742690058479532e-05, | |
| "loss": 0.0358, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.842105263157895, | |
| "grad_norm": 2.183762788772583, | |
| "learning_rate": 8.596491228070177e-05, | |
| "loss": 0.0171, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.9298245614035086, | |
| "grad_norm": 2.652369737625122, | |
| "learning_rate": 8.45029239766082e-05, | |
| "loss": 0.0813, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.017543859649122, | |
| "grad_norm": 0.02325315773487091, | |
| "learning_rate": 8.304093567251462e-05, | |
| "loss": 0.0377, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.017543859649122, | |
| "eval_accuracy": 0.9906542056074766, | |
| "eval_loss": 0.07433710992336273, | |
| "eval_runtime": 36.845, | |
| "eval_samples_per_second": 8.712, | |
| "eval_steps_per_second": 1.113, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.105263157894737, | |
| "grad_norm": 0.002300787018612027, | |
| "learning_rate": 8.157894736842105e-05, | |
| "loss": 0.0111, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.192982456140351, | |
| "grad_norm": 0.8302111625671387, | |
| "learning_rate": 8.01169590643275e-05, | |
| "loss": 0.0242, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 7.280701754385965, | |
| "grad_norm": 0.0038752101827412844, | |
| "learning_rate": 7.865497076023393e-05, | |
| "loss": 0.0003, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 7.368421052631579, | |
| "grad_norm": 0.19968080520629883, | |
| "learning_rate": 7.719298245614036e-05, | |
| "loss": 0.0124, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 7.456140350877193, | |
| "grad_norm": 0.9523430466651917, | |
| "learning_rate": 7.573099415204679e-05, | |
| "loss": 0.0279, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.543859649122807, | |
| "grad_norm": 1.5014656782150269, | |
| "learning_rate": 7.426900584795321e-05, | |
| "loss": 0.0167, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.631578947368421, | |
| "grad_norm": 10.451250076293945, | |
| "learning_rate": 7.280701754385966e-05, | |
| "loss": 0.0142, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 7.719298245614035, | |
| "grad_norm": 2.1814494132995605, | |
| "learning_rate": 7.134502923976609e-05, | |
| "loss": 0.0469, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.807017543859649, | |
| "grad_norm": 2.4280495643615723, | |
| "learning_rate": 6.988304093567252e-05, | |
| "loss": 0.0096, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 7.894736842105263, | |
| "grad_norm": 0.14394675195217133, | |
| "learning_rate": 6.842105263157895e-05, | |
| "loss": 0.062, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.894736842105263, | |
| "eval_accuracy": 0.9781931464174455, | |
| "eval_loss": 0.16418205201625824, | |
| "eval_runtime": 37.5674, | |
| "eval_samples_per_second": 8.545, | |
| "eval_steps_per_second": 1.091, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.982456140350877, | |
| "grad_norm": 0.20802760124206543, | |
| "learning_rate": 6.695906432748539e-05, | |
| "loss": 0.0651, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 8.070175438596491, | |
| "grad_norm": 0.10575652867555618, | |
| "learning_rate": 6.549707602339182e-05, | |
| "loss": 0.0115, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.157894736842104, | |
| "grad_norm": 0.8286623954772949, | |
| "learning_rate": 6.403508771929825e-05, | |
| "loss": 0.0087, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 8.24561403508772, | |
| "grad_norm": 0.39614301919937134, | |
| "learning_rate": 6.257309941520468e-05, | |
| "loss": 0.0198, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.0022884588688611984, | |
| "learning_rate": 6.111111111111112e-05, | |
| "loss": 0.0006, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 0.06943880021572113, | |
| "learning_rate": 5.9649122807017544e-05, | |
| "loss": 0.0138, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 8.508771929824562, | |
| "grad_norm": 0.62754225730896, | |
| "learning_rate": 5.818713450292398e-05, | |
| "loss": 0.004, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 8.596491228070175, | |
| "grad_norm": 0.0023998187389224768, | |
| "learning_rate": 5.6725146198830416e-05, | |
| "loss": 0.0039, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 8.68421052631579, | |
| "grad_norm": 0.12702779471874237, | |
| "learning_rate": 5.526315789473685e-05, | |
| "loss": 0.0277, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 4.152740955352783, | |
| "learning_rate": 5.3801169590643275e-05, | |
| "loss": 0.0025, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "eval_accuracy": 0.9875389408099688, | |
| "eval_loss": 0.05983072146773338, | |
| "eval_runtime": 37.137, | |
| "eval_samples_per_second": 8.644, | |
| "eval_steps_per_second": 1.104, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.859649122807017, | |
| "grad_norm": 0.05080701410770416, | |
| "learning_rate": 5.233918128654971e-05, | |
| "loss": 0.1165, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 8.947368421052632, | |
| "grad_norm": 0.007482455112040043, | |
| "learning_rate": 5.087719298245615e-05, | |
| "loss": 0.0076, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 9.035087719298245, | |
| "grad_norm": 4.2491483327466995e-05, | |
| "learning_rate": 4.941520467836258e-05, | |
| "loss": 0.0035, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 9.12280701754386, | |
| "grad_norm": 0.017159271985292435, | |
| "learning_rate": 4.7953216374269006e-05, | |
| "loss": 0.0157, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 9.210526315789474, | |
| "grad_norm": 0.1872653365135193, | |
| "learning_rate": 4.649122807017544e-05, | |
| "loss": 0.0113, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 9.298245614035087, | |
| "grad_norm": 0.000704402569681406, | |
| "learning_rate": 4.502923976608187e-05, | |
| "loss": 0.0003, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 9.385964912280702, | |
| "grad_norm": 0.938255250453949, | |
| "learning_rate": 4.356725146198831e-05, | |
| "loss": 0.015, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 9.473684210526315, | |
| "grad_norm": 0.03299758583307266, | |
| "learning_rate": 4.210526315789474e-05, | |
| "loss": 0.0333, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.56140350877193, | |
| "grad_norm": 3.074659824371338, | |
| "learning_rate": 4.0643274853801174e-05, | |
| "loss": 0.0159, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 9.649122807017545, | |
| "grad_norm": 0.00028988588019274175, | |
| "learning_rate": 3.9181286549707604e-05, | |
| "loss": 0.0041, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.649122807017545, | |
| "eval_accuracy": 0.9813084112149533, | |
| "eval_loss": 0.12797725200653076, | |
| "eval_runtime": 37.1154, | |
| "eval_samples_per_second": 8.649, | |
| "eval_steps_per_second": 1.105, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.736842105263158, | |
| "grad_norm": 0.0019729495979845524, | |
| "learning_rate": 3.771929824561404e-05, | |
| "loss": 0.0362, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 9.824561403508772, | |
| "grad_norm": 0.0025722056161612272, | |
| "learning_rate": 3.625730994152047e-05, | |
| "loss": 0.0009, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 9.912280701754385, | |
| "grad_norm": 1.9475336074829102, | |
| "learning_rate": 3.4795321637426905e-05, | |
| "loss": 0.0368, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.05613500997424126, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.032, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 10.087719298245615, | |
| "grad_norm": 0.031211694702506065, | |
| "learning_rate": 3.187134502923977e-05, | |
| "loss": 0.0146, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.175438596491228, | |
| "grad_norm": 1.126404881477356, | |
| "learning_rate": 3.0409356725146197e-05, | |
| "loss": 0.0035, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 10.263157894736842, | |
| "grad_norm": 0.0027087063062936068, | |
| "learning_rate": 2.8947368421052634e-05, | |
| "loss": 0.0031, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 10.350877192982455, | |
| "grad_norm": 0.012115487828850746, | |
| "learning_rate": 2.7485380116959063e-05, | |
| "loss": 0.0152, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 10.43859649122807, | |
| "grad_norm": 0.0014923892449587584, | |
| "learning_rate": 2.60233918128655e-05, | |
| "loss": 0.0298, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 10.526315789473685, | |
| "grad_norm": 10.097140312194824, | |
| "learning_rate": 2.456140350877193e-05, | |
| "loss": 0.0305, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.526315789473685, | |
| "eval_accuracy": 0.9813084112149533, | |
| "eval_loss": 0.0920475497841835, | |
| "eval_runtime": 38.0741, | |
| "eval_samples_per_second": 8.431, | |
| "eval_steps_per_second": 1.077, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.614035087719298, | |
| "grad_norm": 0.04899158701300621, | |
| "learning_rate": 2.309941520467836e-05, | |
| "loss": 0.0228, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 10.701754385964913, | |
| "grad_norm": 1.95464289188385, | |
| "learning_rate": 2.1637426900584794e-05, | |
| "loss": 0.0053, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.789473684210526, | |
| "grad_norm": 0.011684381403028965, | |
| "learning_rate": 2.0175438596491227e-05, | |
| "loss": 0.0057, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 10.87719298245614, | |
| "grad_norm": 0.004223628900945187, | |
| "learning_rate": 1.871345029239766e-05, | |
| "loss": 0.0025, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 10.964912280701755, | |
| "grad_norm": 0.018841328099370003, | |
| "learning_rate": 1.7251461988304093e-05, | |
| "loss": 0.0029, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 11.052631578947368, | |
| "grad_norm": 0.0007397038862109184, | |
| "learning_rate": 1.5789473684210526e-05, | |
| "loss": 0.1863, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 11.140350877192983, | |
| "grad_norm": 0.12742939591407776, | |
| "learning_rate": 1.4327485380116959e-05, | |
| "loss": 0.0001, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 11.228070175438596, | |
| "grad_norm": 0.05407753214240074, | |
| "learning_rate": 1.2865497076023392e-05, | |
| "loss": 0.0001, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 11.31578947368421, | |
| "grad_norm": 1.0473092794418335, | |
| "learning_rate": 1.1403508771929824e-05, | |
| "loss": 0.0458, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 11.403508771929825, | |
| "grad_norm": 1.6566870212554932, | |
| "learning_rate": 9.941520467836257e-06, | |
| "loss": 0.0148, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.403508771929825, | |
| "eval_accuracy": 0.9875389408099688, | |
| "eval_loss": 0.12092220783233643, | |
| "eval_runtime": 36.0207, | |
| "eval_samples_per_second": 8.912, | |
| "eval_steps_per_second": 1.138, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.491228070175438, | |
| "grad_norm": 0.0020481087267398834, | |
| "learning_rate": 8.47953216374269e-06, | |
| "loss": 0.0019, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 11.578947368421053, | |
| "grad_norm": 0.064646415412426, | |
| "learning_rate": 7.017543859649123e-06, | |
| "loss": 0.0064, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 11.666666666666666, | |
| "grad_norm": 0.0019295840756967664, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0377, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 11.75438596491228, | |
| "grad_norm": 0.25071534514427185, | |
| "learning_rate": 4.093567251461989e-06, | |
| "loss": 0.0035, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 11.842105263157894, | |
| "grad_norm": 0.48753437399864197, | |
| "learning_rate": 2.631578947368421e-06, | |
| "loss": 0.0237, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 11.929824561403509, | |
| "grad_norm": 0.01427147351205349, | |
| "learning_rate": 1.1695906432748538e-06, | |
| "loss": 0.0029, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 1368, | |
| "total_flos": 6.93495297202176e+16, | |
| "train_loss": 0.044889834785714136, | |
| "train_runtime": 3145.9622, | |
| "train_samples_per_second": 6.916, | |
| "train_steps_per_second": 0.435 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1368, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 12, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.93495297202176e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |