| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "global_step": 4950, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.22303473949432373, | |
| "eval_loss": 3.9924213886260986, | |
| "eval_runtime": 3.5667, | |
| "eval_samples_per_second": 306.729, | |
| "eval_steps_per_second": 38.411, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "learning_rate": 8.98989898989899e-06, | |
| "loss": 4.3795, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.3957952558994293, | |
| "eval_loss": 3.1812195777893066, | |
| "eval_runtime": 3.5362, | |
| "eval_samples_per_second": 309.369, | |
| "eval_steps_per_second": 38.742, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5511882901191711, | |
| "eval_loss": 2.5903849601745605, | |
| "eval_runtime": 3.5231, | |
| "eval_samples_per_second": 310.519, | |
| "eval_steps_per_second": 38.886, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 3.03, | |
| "learning_rate": 7.97979797979798e-06, | |
| "loss": 3.2046, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6407678127288818, | |
| "eval_loss": 2.1536314487457275, | |
| "eval_runtime": 3.5444, | |
| "eval_samples_per_second": 308.659, | |
| "eval_steps_per_second": 38.653, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 4.55, | |
| "learning_rate": 6.969696969696971e-06, | |
| "loss": 2.3683, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7129799127578735, | |
| "eval_loss": 1.8079293966293335, | |
| "eval_runtime": 3.555, | |
| "eval_samples_per_second": 307.739, | |
| "eval_steps_per_second": 38.538, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7687385678291321, | |
| "eval_loss": 1.5419940948486328, | |
| "eval_runtime": 3.5056, | |
| "eval_samples_per_second": 312.075, | |
| "eval_steps_per_second": 39.081, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 6.06, | |
| "learning_rate": 5.95959595959596e-06, | |
| "loss": 1.8065, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.8053016662597656, | |
| "eval_loss": 1.3433690071105957, | |
| "eval_runtime": 3.4889, | |
| "eval_samples_per_second": 313.564, | |
| "eval_steps_per_second": 39.267, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 7.58, | |
| "learning_rate": 4.94949494949495e-06, | |
| "loss": 1.373, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.825411319732666, | |
| "eval_loss": 1.1881896257400513, | |
| "eval_runtime": 3.5056, | |
| "eval_samples_per_second": 312.068, | |
| "eval_steps_per_second": 39.08, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8400365710258484, | |
| "eval_loss": 1.0700345039367676, | |
| "eval_runtime": 3.4951, | |
| "eval_samples_per_second": 313.014, | |
| "eval_steps_per_second": 39.198, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.93939393939394e-06, | |
| "loss": 1.0931, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.851005494594574, | |
| "eval_loss": 0.9790602326393127, | |
| "eval_runtime": 3.4892, | |
| "eval_samples_per_second": 313.542, | |
| "eval_steps_per_second": 39.264, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 10.61, | |
| "learning_rate": 2.9292929292929295e-06, | |
| "loss": 0.8714, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.8619744181632996, | |
| "eval_loss": 0.9201710224151611, | |
| "eval_runtime": 3.4883, | |
| "eval_samples_per_second": 313.616, | |
| "eval_steps_per_second": 39.274, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.868372917175293, | |
| "eval_loss": 0.8640827536582947, | |
| "eval_runtime": 3.5075, | |
| "eval_samples_per_second": 311.902, | |
| "eval_steps_per_second": 39.059, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 12.12, | |
| "learning_rate": 1.9191919191919192e-06, | |
| "loss": 0.7428, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.8747714757919312, | |
| "eval_loss": 0.8372448086738586, | |
| "eval_runtime": 3.5209, | |
| "eval_samples_per_second": 310.718, | |
| "eval_steps_per_second": 38.911, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 13.64, | |
| "learning_rate": 9.090909090909091e-07, | |
| "loss": 0.6531, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8765996098518372, | |
| "eval_loss": 0.8168175220489502, | |
| "eval_runtime": 3.5294, | |
| "eval_samples_per_second": 309.965, | |
| "eval_steps_per_second": 38.816, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.8756855726242065, | |
| "eval_loss": 0.810522198677063, | |
| "eval_runtime": 3.5261, | |
| "eval_samples_per_second": 310.256, | |
| "eval_steps_per_second": 38.853, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 4950, | |
| "total_flos": 1.043199019279872e+16, | |
| "train_loss": 1.72243186873619, | |
| "train_runtime": 2003.3488, | |
| "train_samples_per_second": 79.038, | |
| "train_steps_per_second": 2.471 | |
| } | |
| ], | |
| "max_steps": 4950, | |
| "num_train_epochs": 15, | |
| "total_flos": 1.043199019279872e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |