| { | |
| "best_metric": 0.6670854271356784, | |
| "best_model_checkpoint": "videomae-base-finetuned-ElderReact-Surprise11/checkpoint-195", | |
| "epoch": 9.076315789473684, | |
| "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": 3.594555377960205, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.7595, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 3.358711004257202, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.6991, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 2.800182342529297, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.7047, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.5891959798994975, | |
| "eval_loss": 0.6541748046875, | |
| "eval_runtime": 222.5887, | |
| "eval_samples_per_second": 3.576, | |
| "eval_steps_per_second": 0.225, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 5.824236869812012, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.6805, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 6.393433094024658, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.6454, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 3.4522626399993896, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.6746, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 3.465517520904541, | |
| "learning_rate": 4.5321637426900585e-05, | |
| "loss": 0.7111, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.585427135678392, | |
| "eval_loss": 0.6550820469856262, | |
| "eval_runtime": 189.1721, | |
| "eval_samples_per_second": 4.208, | |
| "eval_steps_per_second": 0.264, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 1.5881491899490356, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6761, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.0569772720336914, | |
| "learning_rate": 4.239766081871345e-05, | |
| "loss": 0.657, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 4.930370330810547, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.6329, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 4.686254024505615, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 0.6839, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.6507537688442211, | |
| "eval_loss": 0.6406037211418152, | |
| "eval_runtime": 197.8228, | |
| "eval_samples_per_second": 4.024, | |
| "eval_steps_per_second": 0.253, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "grad_norm": 2.4547111988067627, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.7011, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "grad_norm": 9.182774543762207, | |
| "learning_rate": 3.654970760233918e-05, | |
| "loss": 0.6745, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 6.363346099853516, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.6615, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.09, | |
| "grad_norm": 2.053642988204956, | |
| "learning_rate": 3.362573099415205e-05, | |
| "loss": 0.6872, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.6306532663316583, | |
| "eval_loss": 0.631442129611969, | |
| "eval_runtime": 192.6912, | |
| "eval_samples_per_second": 4.131, | |
| "eval_steps_per_second": 0.259, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 4.01, | |
| "grad_norm": 2.378765821456909, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.6168, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 4.013543605804443, | |
| "learning_rate": 3.0701754385964913e-05, | |
| "loss": 0.6156, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.06, | |
| "grad_norm": 3.605048656463623, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.6783, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.09, | |
| "grad_norm": 2.6724507808685303, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6678, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.6670854271356784, | |
| "eval_loss": 0.6236329674720764, | |
| "eval_runtime": 195.2419, | |
| "eval_samples_per_second": 4.077, | |
| "eval_steps_per_second": 0.256, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.01, | |
| "grad_norm": 6.503556251525879, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.6341, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.04, | |
| "grad_norm": 1.863840103149414, | |
| "learning_rate": 2.485380116959064e-05, | |
| "loss": 0.6398, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.07, | |
| "grad_norm": 3.606435537338257, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.6361, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "grad_norm": 5.716346740722656, | |
| "learning_rate": 2.1929824561403507e-05, | |
| "loss": 0.5806, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.6482412060301508, | |
| "eval_loss": 0.6078271865844727, | |
| "eval_runtime": 193.0001, | |
| "eval_samples_per_second": 4.124, | |
| "eval_steps_per_second": 0.259, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 6.02, | |
| "grad_norm": 2.487823486328125, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.6345, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 4.1022629737854, | |
| "learning_rate": 1.9005847953216373e-05, | |
| "loss": 0.6229, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.07, | |
| "grad_norm": 2.2805275917053223, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.6239, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "grad_norm": 2.4343156814575195, | |
| "learning_rate": 1.608187134502924e-05, | |
| "loss": 0.5697, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.5979899497487438, | |
| "eval_loss": 0.6825251579284668, | |
| "eval_runtime": 189.4608, | |
| "eval_samples_per_second": 4.201, | |
| "eval_steps_per_second": 0.264, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 3.1624414920806885, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.5657, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "grad_norm": 4.5414958000183105, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 0.6419, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "grad_norm": 4.060510158538818, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.5491, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 4.565443992614746, | |
| "learning_rate": 1.023391812865497e-05, | |
| "loss": 0.6127, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.6005025125628141, | |
| "eval_loss": 0.6445818543434143, | |
| "eval_runtime": 190.0918, | |
| "eval_samples_per_second": 4.187, | |
| "eval_steps_per_second": 0.263, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 8.02, | |
| "grad_norm": 2.4785315990448, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.6422, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.05, | |
| "grad_norm": 4.520650386810303, | |
| "learning_rate": 7.3099415204678366e-06, | |
| "loss": 0.562, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.07, | |
| "grad_norm": 2.4147799015045166, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.5909, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "grad_norm": 5.0589919090271, | |
| "learning_rate": 4.3859649122807014e-06, | |
| "loss": 0.5645, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.6306532663316583, | |
| "eval_loss": 0.6240276098251343, | |
| "eval_runtime": 195.3851, | |
| "eval_samples_per_second": 4.074, | |
| "eval_steps_per_second": 0.256, | |
| "step": 351 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 3.9938855171203613, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.5564, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.05, | |
| "grad_norm": 5.353033065795898, | |
| "learning_rate": 1.4619883040935671e-06, | |
| "loss": 0.5628, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "grad_norm": 2.4078638553619385, | |
| "learning_rate": 0.0, | |
| "loss": 0.549, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.6180904522613065, | |
| "eval_loss": 0.6290447115898132, | |
| "eval_runtime": 190.6021, | |
| "eval_samples_per_second": 4.176, | |
| "eval_steps_per_second": 0.262, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "step": 380, | |
| "total_flos": 7.475141472240992e+18, | |
| "train_loss": 0.6359516570442602, | |
| "train_runtime": 3680.9423, | |
| "train_samples_per_second": 1.652, | |
| "train_steps_per_second": 0.103 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.652, | |
| "eval_loss": 0.638110339641571, | |
| "eval_runtime": 181.9234, | |
| "eval_samples_per_second": 4.123, | |
| "eval_steps_per_second": 0.258, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.08, | |
| "eval_accuracy": 0.652, | |
| "eval_loss": 0.6381102800369263, | |
| "eval_runtime": 177.1077, | |
| "eval_samples_per_second": 4.235, | |
| "eval_steps_per_second": 0.265, | |
| "step": 380 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 380, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 7.475141472240992e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |