| { | |
| "best_metric": 0.9936628643852978, | |
| "best_model_checkpoint": "neunit-ks-ertong-chengren/checkpoint-275", | |
| "epoch": 4.954954954954955, | |
| "global_step": 275, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.0714285714285714e-05, | |
| "loss": 0.6808, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 0.6099, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 2.9757085020242918e-05, | |
| "loss": 0.4493, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 2.854251012145749e-05, | |
| "loss": 0.239, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 2.7327935222672065e-05, | |
| "loss": 0.1371, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.9797211660329531, | |
| "eval_loss": 0.09274197369813919, | |
| "eval_runtime": 47.6497, | |
| "eval_samples_per_second": 16.558, | |
| "eval_steps_per_second": 0.525, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 2.611336032388664e-05, | |
| "loss": 0.0725, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 2.4898785425101213e-05, | |
| "loss": 0.0496, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 2.368421052631579e-05, | |
| "loss": 0.0505, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 2.2469635627530364e-05, | |
| "loss": 0.0474, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 2.125506072874494e-05, | |
| "loss": 0.0319, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 2.0040485829959516e-05, | |
| "loss": 0.0421, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9898605830164765, | |
| "eval_loss": 0.04827561601996422, | |
| "eval_runtime": 47.1329, | |
| "eval_samples_per_second": 16.74, | |
| "eval_steps_per_second": 0.53, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 1.882591093117409e-05, | |
| "loss": 0.0408, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 1.7611336032388663e-05, | |
| "loss": 0.0268, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 1.639676113360324e-05, | |
| "loss": 0.0201, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 1.5182186234817814e-05, | |
| "loss": 0.0132, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 1.396761133603239e-05, | |
| "loss": 0.0163, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.9885931558935361, | |
| "eval_loss": 0.04729734733700752, | |
| "eval_runtime": 47.1403, | |
| "eval_samples_per_second": 16.737, | |
| "eval_steps_per_second": 0.53, | |
| "step": 166 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 1.2753036437246964e-05, | |
| "loss": 0.0137, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 1.153846153846154e-05, | |
| "loss": 0.021, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 1.0323886639676113e-05, | |
| "loss": 0.0184, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 9.109311740890687e-06, | |
| "loss": 0.0223, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 7.894736842105263e-06, | |
| "loss": 0.008, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 6.680161943319838e-06, | |
| "loss": 0.0101, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9898605830164765, | |
| "eval_loss": 0.03389865159988403, | |
| "eval_runtime": 47.0992, | |
| "eval_samples_per_second": 16.752, | |
| "eval_steps_per_second": 0.531, | |
| "step": 222 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 5.465587044534413e-06, | |
| "loss": 0.0097, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 4.251012145748988e-06, | |
| "loss": 0.0097, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.0364372469635627e-06, | |
| "loss": 0.0068, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 1.8218623481781377e-06, | |
| "loss": 0.0123, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 6.072874493927125e-07, | |
| "loss": 0.0089, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "eval_accuracy": 0.9936628643852978, | |
| "eval_loss": 0.02500770427286625, | |
| "eval_runtime": 47.1174, | |
| "eval_samples_per_second": 16.745, | |
| "eval_steps_per_second": 0.531, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "step": 275, | |
| "total_flos": 6.3848207630592e+17, | |
| "train_loss": 0.09714971287684007, | |
| "train_runtime": 1154.3714, | |
| "train_samples_per_second": 30.731, | |
| "train_steps_per_second": 0.238 | |
| } | |
| ], | |
| "max_steps": 275, | |
| "num_train_epochs": 5, | |
| "total_flos": 6.3848207630592e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |