| { | |
| "best_metric": 0.7439252336448599, | |
| "best_model_checkpoint": "sew-ft-fake-detection/checkpoint-301", | |
| "epoch": 9.850746268656717, | |
| "eval_steps": 500, | |
| "global_step": 330, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.29850746268656714, | |
| "grad_norm": 1.730605125427246, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.6755, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5970149253731343, | |
| "grad_norm": 0.684485673904419, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.6506, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8955223880597015, | |
| "grad_norm": 0.7370636463165283, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.6344, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9850746268656716, | |
| "eval_accuracy": 0.6504672897196262, | |
| "eval_loss": 0.6394948959350586, | |
| "eval_runtime": 30.1661, | |
| "eval_samples_per_second": 35.47, | |
| "eval_steps_per_second": 2.221, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.1940298507462686, | |
| "grad_norm": 1.093192458152771, | |
| "learning_rate": 2.9292929292929294e-05, | |
| "loss": 0.635, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.4925373134328357, | |
| "grad_norm": 0.6133630275726318, | |
| "learning_rate": 2.8282828282828285e-05, | |
| "loss": 0.6023, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.7910447761194028, | |
| "grad_norm": 0.9280603528022766, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.6157, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5214953271028038, | |
| "eval_loss": 0.6703016757965088, | |
| "eval_runtime": 34.578, | |
| "eval_samples_per_second": 30.945, | |
| "eval_steps_per_second": 1.938, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 2.08955223880597, | |
| "grad_norm": 0.9497929215431213, | |
| "learning_rate": 2.6262626262626265e-05, | |
| "loss": 0.5853, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.388059701492537, | |
| "grad_norm": 1.5462061166763306, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.5659, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.6865671641791042, | |
| "grad_norm": 1.5534844398498535, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": 0.5512, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "grad_norm": 1.4848833084106445, | |
| "learning_rate": 2.3232323232323235e-05, | |
| "loss": 0.5572, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.9850746268656714, | |
| "eval_accuracy": 0.7130841121495327, | |
| "eval_loss": 0.561539351940155, | |
| "eval_runtime": 36.5879, | |
| "eval_samples_per_second": 29.245, | |
| "eval_steps_per_second": 1.831, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.283582089552239, | |
| "grad_norm": 2.040893316268921, | |
| "learning_rate": 2.222222222222222e-05, | |
| "loss": 0.5456, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.582089552238806, | |
| "grad_norm": 1.084596037864685, | |
| "learning_rate": 2.121212121212121e-05, | |
| "loss": 0.5214, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.8805970149253732, | |
| "grad_norm": 1.547035574913025, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.4968, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.694392523364486, | |
| "eval_loss": 0.8148747086524963, | |
| "eval_runtime": 39.5102, | |
| "eval_samples_per_second": 27.082, | |
| "eval_steps_per_second": 1.696, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 4.17910447761194, | |
| "grad_norm": 1.8584659099578857, | |
| "learning_rate": 1.919191919191919e-05, | |
| "loss": 0.4943, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.477611940298507, | |
| "grad_norm": 2.004117727279663, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.5122, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.776119402985074, | |
| "grad_norm": 1.3472633361816406, | |
| "learning_rate": 1.717171717171717e-05, | |
| "loss": 0.4988, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.985074626865671, | |
| "eval_accuracy": 0.7037383177570093, | |
| "eval_loss": 0.8099448680877686, | |
| "eval_runtime": 35.842, | |
| "eval_samples_per_second": 29.853, | |
| "eval_steps_per_second": 1.869, | |
| "step": 167 | |
| }, | |
| { | |
| "epoch": 5.074626865671641, | |
| "grad_norm": 1.908387303352356, | |
| "learning_rate": 1.616161616161616e-05, | |
| "loss": 0.4812, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.373134328358209, | |
| "grad_norm": 2.6505331993103027, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.4671, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.6716417910447765, | |
| "grad_norm": 2.50083327293396, | |
| "learning_rate": 1.4141414141414143e-05, | |
| "loss": 0.4538, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.970149253731344, | |
| "grad_norm": 1.8395198583602905, | |
| "learning_rate": 1.3131313131313132e-05, | |
| "loss": 0.4756, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7102803738317757, | |
| "eval_loss": 0.8318859934806824, | |
| "eval_runtime": 36.3806, | |
| "eval_samples_per_second": 29.411, | |
| "eval_steps_per_second": 1.842, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 6.268656716417911, | |
| "grad_norm": 1.486412763595581, | |
| "learning_rate": 1.2121212121212122e-05, | |
| "loss": 0.4576, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.567164179104478, | |
| "grad_norm": 1.7824214696884155, | |
| "learning_rate": 1.111111111111111e-05, | |
| "loss": 0.4462, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.865671641791045, | |
| "grad_norm": 2.2163608074188232, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.4334, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.985074626865671, | |
| "eval_accuracy": 0.7336448598130841, | |
| "eval_loss": 0.7316926717758179, | |
| "eval_runtime": 38.478, | |
| "eval_samples_per_second": 27.808, | |
| "eval_steps_per_second": 1.741, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 7.164179104477612, | |
| "grad_norm": 2.735869884490967, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.4686, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.462686567164179, | |
| "grad_norm": 2.0291202068328857, | |
| "learning_rate": 8.08080808080808e-06, | |
| "loss": 0.4604, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.7611940298507465, | |
| "grad_norm": 1.6389554738998413, | |
| "learning_rate": 7.070707070707071e-06, | |
| "loss": 0.4321, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7383177570093458, | |
| "eval_loss": 0.6548246145248413, | |
| "eval_runtime": 40.7344, | |
| "eval_samples_per_second": 26.268, | |
| "eval_steps_per_second": 1.645, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 8.059701492537313, | |
| "grad_norm": 1.9984620809555054, | |
| "learning_rate": 6.060606060606061e-06, | |
| "loss": 0.4594, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.35820895522388, | |
| "grad_norm": 1.9911682605743408, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.4236, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.656716417910447, | |
| "grad_norm": 2.0912842750549316, | |
| "learning_rate": 4.04040404040404e-06, | |
| "loss": 0.4326, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.955223880597014, | |
| "grad_norm": 2.4890761375427246, | |
| "learning_rate": 3.0303030303030305e-06, | |
| "loss": 0.4436, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.985074626865671, | |
| "eval_accuracy": 0.7439252336448599, | |
| "eval_loss": 0.623187780380249, | |
| "eval_runtime": 33.9508, | |
| "eval_samples_per_second": 31.516, | |
| "eval_steps_per_second": 1.973, | |
| "step": 301 | |
| }, | |
| { | |
| "epoch": 9.253731343283581, | |
| "grad_norm": 2.2211296558380127, | |
| "learning_rate": 2.02020202020202e-06, | |
| "loss": 0.438, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.552238805970148, | |
| "grad_norm": 1.8036954402923584, | |
| "learning_rate": 1.01010101010101e-06, | |
| "loss": 0.4104, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.850746268656717, | |
| "grad_norm": 1.8852014541625977, | |
| "learning_rate": 0.0, | |
| "loss": 0.4493, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.850746268656717, | |
| "eval_accuracy": 0.7439252336448599, | |
| "eval_loss": 0.6278349757194519, | |
| "eval_runtime": 37.2552, | |
| "eval_samples_per_second": 28.721, | |
| "eval_steps_per_second": 1.798, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.850746268656717, | |
| "step": 330, | |
| "total_flos": 7.07452399466208e+17, | |
| "train_loss": 0.5083336223255505, | |
| "train_runtime": 1835.4297, | |
| "train_samples_per_second": 23.302, | |
| "train_steps_per_second": 0.18 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 330, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.07452399466208e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |