| { | |
| "best_metric": 0.9918838103374626, | |
| "best_model_checkpoint": "resnet-50-finetuned-dog-vs-cat/checkpoint-329", | |
| "epoch": 2.986342943854325, | |
| "eval_steps": 500, | |
| "global_step": 492, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 0.5951987504959106, | |
| "learning_rate": 1e-05, | |
| "loss": 0.693, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 0.743243932723999, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6912, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 0.6900031566619873, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6871, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 0.7302324175834656, | |
| "learning_rate": 4e-05, | |
| "loss": 0.68, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 0.9748005270957947, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6664, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.131984829902649, | |
| "learning_rate": 4.8868778280542986e-05, | |
| "loss": 0.6408, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.0235670804977417, | |
| "learning_rate": 4.7737556561085976e-05, | |
| "loss": 0.617, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 1.1163190603256226, | |
| "learning_rate": 4.660633484162896e-05, | |
| "loss": 0.5807, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.282690405845642, | |
| "learning_rate": 4.547511312217195e-05, | |
| "loss": 0.5438, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.068588137626648, | |
| "learning_rate": 4.434389140271493e-05, | |
| "loss": 0.502, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 0.9256152510643005, | |
| "learning_rate": 4.321266968325792e-05, | |
| "loss": 0.4701, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 0.9594171643257141, | |
| "learning_rate": 4.2081447963800907e-05, | |
| "loss": 0.447, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 1.0906646251678467, | |
| "learning_rate": 4.095022624434389e-05, | |
| "loss": 0.4238, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 0.9434269070625305, | |
| "learning_rate": 3.981900452488688e-05, | |
| "loss": 0.4026, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 1.15977144241333, | |
| "learning_rate": 3.868778280542987e-05, | |
| "loss": 0.3807, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 0.9563286304473877, | |
| "learning_rate": 3.7556561085972854e-05, | |
| "loss": 0.3357, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9867577958137548, | |
| "eval_loss": 0.22545380890369415, | |
| "eval_runtime": 27.2631, | |
| "eval_samples_per_second": 85.867, | |
| "eval_steps_per_second": 2.714, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.0790256261825562, | |
| "learning_rate": 3.642533936651584e-05, | |
| "loss": 0.3166, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 0.9720374345779419, | |
| "learning_rate": 3.529411764705883e-05, | |
| "loss": 0.2961, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 1.2068029642105103, | |
| "learning_rate": 3.416289592760181e-05, | |
| "loss": 0.2753, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 1.0407794713974, | |
| "learning_rate": 3.3031674208144794e-05, | |
| "loss": 0.2631, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 0.9938662052154541, | |
| "learning_rate": 3.1900452488687784e-05, | |
| "loss": 0.2427, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.8150984048843384, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.2163, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.7988576889038086, | |
| "learning_rate": 2.9638009049773758e-05, | |
| "loss": 0.216, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.1607186794281006, | |
| "learning_rate": 2.850678733031674e-05, | |
| "loss": 0.2086, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.0582228899002075, | |
| "learning_rate": 2.737556561085973e-05, | |
| "loss": 0.1857, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 0.8427866697311401, | |
| "learning_rate": 2.6244343891402718e-05, | |
| "loss": 0.1808, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.8713784217834473, | |
| "learning_rate": 2.51131221719457e-05, | |
| "loss": 0.1689, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.8434356451034546, | |
| "learning_rate": 2.3981900452488688e-05, | |
| "loss": 0.1732, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 0.7676334977149963, | |
| "learning_rate": 2.2850678733031675e-05, | |
| "loss": 0.1678, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 0.6907545328140259, | |
| "learning_rate": 2.1719457013574662e-05, | |
| "loss": 0.1624, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 0.7248879075050354, | |
| "learning_rate": 2.058823529411765e-05, | |
| "loss": 0.1613, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 1.4968321323394775, | |
| "learning_rate": 1.9457013574660635e-05, | |
| "loss": 0.1683, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9918838103374626, | |
| "eval_loss": 0.05769222229719162, | |
| "eval_runtime": 27.9863, | |
| "eval_samples_per_second": 83.648, | |
| "eval_steps_per_second": 2.644, | |
| "step": 329 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.3721855878829956, | |
| "learning_rate": 1.832579185520362e-05, | |
| "loss": 0.1635, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.0472960472106934, | |
| "learning_rate": 1.7194570135746606e-05, | |
| "loss": 0.1576, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 0.9706544280052185, | |
| "learning_rate": 1.6063348416289596e-05, | |
| "loss": 0.157, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 0.6168745756149292, | |
| "learning_rate": 1.493212669683258e-05, | |
| "loss": 0.1535, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "grad_norm": 2.201444149017334, | |
| "learning_rate": 1.3800904977375568e-05, | |
| "loss": 0.1494, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "grad_norm": 0.6486875414848328, | |
| "learning_rate": 1.2669683257918553e-05, | |
| "loss": 0.1502, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 0.7056523561477661, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 0.1446, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "grad_norm": 1.3912076950073242, | |
| "learning_rate": 1.0407239819004526e-05, | |
| "loss": 0.1428, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "grad_norm": 1.0273276567459106, | |
| "learning_rate": 9.276018099547511e-06, | |
| "loss": 0.144, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "grad_norm": 1.1831059455871582, | |
| "learning_rate": 8.144796380090498e-06, | |
| "loss": 0.1572, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "grad_norm": 1.5577871799468994, | |
| "learning_rate": 7.013574660633485e-06, | |
| "loss": 0.1419, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.67, | |
| "grad_norm": 1.2978945970535278, | |
| "learning_rate": 5.882352941176471e-06, | |
| "loss": 0.1562, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "grad_norm": 0.5621709823608398, | |
| "learning_rate": 4.751131221719457e-06, | |
| "loss": 0.1352, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 1.0484684705734253, | |
| "learning_rate": 3.619909502262444e-06, | |
| "loss": 0.143, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "grad_norm": 0.9431272149085999, | |
| "learning_rate": 2.48868778280543e-06, | |
| "loss": 0.1517, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "grad_norm": 0.8055468201637268, | |
| "learning_rate": 1.3574660633484164e-06, | |
| "loss": 0.1455, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "grad_norm": 1.1355047225952148, | |
| "learning_rate": 2.2624434389140275e-07, | |
| "loss": 0.1448, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9918838103374626, | |
| "eval_loss": 0.04604041948914528, | |
| "eval_runtime": 27.4499, | |
| "eval_samples_per_second": 85.283, | |
| "eval_steps_per_second": 2.696, | |
| "step": 492 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "step": 492, | |
| "total_flos": 1.336513820941394e+18, | |
| "train_loss": 0.2994473668617931, | |
| "train_runtime": 953.1608, | |
| "train_samples_per_second": 66.313, | |
| "train_steps_per_second": 0.516 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 492, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.336513820941394e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |