| { | |
| "best_metric": 0.8890439637124913, | |
| "best_model_checkpoint": "videomae-base-finetuned-ElderReact-Disgust/checkpoint-77", | |
| "epoch": 4.189473684210526, | |
| "eval_steps": 500, | |
| "global_step": 380, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 10.353791236877441, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.744, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 2.988565444946289, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.5971, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 6.786588191986084, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.6241, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 7.078255653381348, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.6023, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 3.006227493286133, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.45, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 5.826990127563477, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.683, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 2.513857126235962, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.5255, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_f1": 0.8890439637124913, | |
| "eval_loss": 0.49672603607177734, | |
| "eval_runtime": 1611.6383, | |
| "eval_samples_per_second": 0.494, | |
| "eval_steps_per_second": 0.062, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 1.385485053062439, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.4968, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 2.920797109603882, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.5231, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.5780858993530273, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.628, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 4.24209451675415, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.5087, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 2.839043140411377, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.4976, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 1.7627954483032227, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.5793, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 1.3475885391235352, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.5746, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 7.263377666473389, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.5276, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_f1": 0.8556701030927835, | |
| "eval_loss": 0.5537058711051941, | |
| "eval_runtime": 1589.9409, | |
| "eval_samples_per_second": 0.501, | |
| "eval_steps_per_second": 0.063, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 2.271644353866577, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.5934, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 3.0725769996643066, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.5932, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 2.309803009033203, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.5627, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.2822645902633667, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4367, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 2.9160258769989014, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.4411, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "grad_norm": 1.275068998336792, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.6641, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.17, | |
| "grad_norm": 3.5662224292755127, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.4247, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 8.418479919433594, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.5456, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "eval_f1": 0.2175732217573222, | |
| "eval_loss": 0.9158822298049927, | |
| "eval_runtime": 1437.2422, | |
| "eval_samples_per_second": 0.554, | |
| "eval_steps_per_second": 0.07, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 2.4661779403686523, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.5765, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "grad_norm": 2.6517739295959473, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.6186, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "grad_norm": 4.40958833694458, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.5523, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 2.175184726715088, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.4407, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.13, | |
| "grad_norm": 2.439981698989868, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.4845, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "grad_norm": 2.0729973316192627, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.5261, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.18, | |
| "grad_norm": 1.7637895345687866, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.4651, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "eval_f1": 0.8877980364656382, | |
| "eval_loss": 0.528616726398468, | |
| "eval_runtime": 1491.2351, | |
| "eval_samples_per_second": 0.534, | |
| "eval_steps_per_second": 0.067, | |
| "step": 308 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 10.06191349029541, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.4746, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "grad_norm": 2.493375539779663, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.5077, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 6.715433597564697, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.4665, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 2.7564079761505127, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.6276, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "grad_norm": 4.325897693634033, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.4483, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "grad_norm": 3.6840827465057373, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.4864, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 4.16, | |
| "grad_norm": 1.9301488399505615, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.4279, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "grad_norm": 4.3399176597595215, | |
| "learning_rate": 0.0, | |
| "loss": 0.3905, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.8068269976726145, | |
| "eval_loss": 0.5771048665046692, | |
| "eval_runtime": 1395.6854, | |
| "eval_samples_per_second": 0.57, | |
| "eval_steps_per_second": 0.072, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "step": 380, | |
| "total_flos": 3.783052093636215e+18, | |
| "train_loss": 0.534650657051488, | |
| "train_runtime": 19096.7535, | |
| "train_samples_per_second": 0.159, | |
| "train_steps_per_second": 0.02 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.9130434782608695, | |
| "eval_loss": 0.4410730004310608, | |
| "eval_runtime": 1236.8508, | |
| "eval_samples_per_second": 0.606, | |
| "eval_steps_per_second": 0.076, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "eval_f1": 0.9130434782608695, | |
| "eval_loss": 0.4410730302333832, | |
| "eval_runtime": 615.7027, | |
| "eval_samples_per_second": 1.218, | |
| "eval_steps_per_second": 0.153, | |
| "step": 380 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 380, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.783052093636215e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |