| { | |
| "best_metric": 0.5313496280552603, | |
| "best_model_checkpoint": "timesformer-base-finetuned-k400-finetuned-ElderReact-anger-balanced/checkpoint-75", | |
| "epoch": 9.0625, | |
| "eval_steps": 500, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 10.033487319946289, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.7267, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 10.82132339477539, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6923, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_accuracy": 0.46014877789585545, | |
| "eval_loss": 0.8338257670402527, | |
| "eval_runtime": 193.5219, | |
| "eval_samples_per_second": 4.862, | |
| "eval_steps_per_second": 0.305, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 6.4887261390686035, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 0.5405, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 4.2266340255737305, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.5519, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 38.938472747802734, | |
| "learning_rate": 4.3981481481481486e-05, | |
| "loss": 0.6436, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.5058448459086079, | |
| "eval_loss": 0.9187208414077759, | |
| "eval_runtime": 186.1328, | |
| "eval_samples_per_second": 5.056, | |
| "eval_steps_per_second": 0.317, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 5.287866115570068, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.4336, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 8.11617374420166, | |
| "learning_rate": 3.935185185185186e-05, | |
| "loss": 0.5464, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_accuracy": 0.5313496280552603, | |
| "eval_loss": 0.8164978623390198, | |
| "eval_runtime": 165.1643, | |
| "eval_samples_per_second": 5.697, | |
| "eval_steps_per_second": 0.357, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.02, | |
| "grad_norm": 9.028487205505371, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.5459, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "grad_norm": 4.760679721832275, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.4503, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 13.96630573272705, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.4147, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "eval_accuracy": 0.5228480340063762, | |
| "eval_loss": 0.9949731230735779, | |
| "eval_runtime": 166.6223, | |
| "eval_samples_per_second": 5.648, | |
| "eval_steps_per_second": 0.354, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "grad_norm": 3.802431344985962, | |
| "learning_rate": 3.0092592592592593e-05, | |
| "loss": 0.3409, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.08, | |
| "grad_norm": 9.562992095947266, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4483, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "eval_accuracy": 0.5079702444208289, | |
| "eval_loss": 1.064165711402893, | |
| "eval_runtime": 194.4496, | |
| "eval_samples_per_second": 4.839, | |
| "eval_steps_per_second": 0.303, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "grad_norm": 8.446785926818848, | |
| "learning_rate": 2.5462962962962965e-05, | |
| "loss": 0.6922, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "grad_norm": 3.8589024543762207, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.3718, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "grad_norm": 5.9254536628723145, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.3565, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_accuracy": 0.5143464399574921, | |
| "eval_loss": 0.8610547780990601, | |
| "eval_runtime": 184.6388, | |
| "eval_samples_per_second": 5.096, | |
| "eval_steps_per_second": 0.32, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.04, | |
| "grad_norm": 5.898458003997803, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.3232, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 6.08, | |
| "grad_norm": 3.0727579593658447, | |
| "learning_rate": 1.6203703703703704e-05, | |
| "loss": 0.3114, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.1, | |
| "eval_accuracy": 0.5185972369819342, | |
| "eval_loss": 0.9594563245773315, | |
| "eval_runtime": 185.3195, | |
| "eval_samples_per_second": 5.078, | |
| "eval_steps_per_second": 0.318, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "grad_norm": 1.4301713705062866, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.253, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.06, | |
| "grad_norm": 6.7790045738220215, | |
| "learning_rate": 1.1574074074074075e-05, | |
| "loss": 0.3144, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "grad_norm": 26.213571548461914, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.3074, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.1, | |
| "eval_accuracy": 0.5196599362380446, | |
| "eval_loss": 1.0241553783416748, | |
| "eval_runtime": 191.958, | |
| "eval_samples_per_second": 4.902, | |
| "eval_steps_per_second": 0.307, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "grad_norm": 3.681013822555542, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.2552, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 8.08, | |
| "grad_norm": 6.130499839782715, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.2888, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.1, | |
| "eval_accuracy": 0.5069075451647184, | |
| "eval_loss": 1.1212139129638672, | |
| "eval_runtime": 182.5202, | |
| "eval_samples_per_second": 5.156, | |
| "eval_steps_per_second": 0.323, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "grad_norm": 4.048097133636475, | |
| "learning_rate": 2.3148148148148148e-06, | |
| "loss": 0.2812, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "grad_norm": 2.3711907863616943, | |
| "learning_rate": 0.0, | |
| "loss": 0.2609, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_accuracy": 0.5111583421891605, | |
| "eval_loss": 1.111896276473999, | |
| "eval_runtime": 168.1115, | |
| "eval_samples_per_second": 5.597, | |
| "eval_steps_per_second": 0.351, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "step": 240, | |
| "total_flos": 3.246125779066061e+18, | |
| "train_loss": 0.43130043546358743, | |
| "train_runtime": 2651.285, | |
| "train_samples_per_second": 1.448, | |
| "train_steps_per_second": 0.091 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_accuracy": 0.5011600928074246, | |
| "eval_loss": 0.8047753572463989, | |
| "eval_runtime": 160.408, | |
| "eval_samples_per_second": 5.374, | |
| "eval_steps_per_second": 0.337, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_accuracy": 0.5011600928074246, | |
| "eval_loss": 0.8047754168510437, | |
| "eval_runtime": 160.5206, | |
| "eval_samples_per_second": 5.37, | |
| "eval_steps_per_second": 0.336, | |
| "step": 240 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 9223372036854775807, | |
| "save_steps": 500, | |
| "total_flos": 3.246125779066061e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |