| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "global_step": 17200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 4.854651162790698e-05, | |
| "loss": 3.572, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.3753416538238525, | |
| "eval_runtime": 146.7285, | |
| "eval_samples_per_second": 163.69, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 4.709302325581396e-05, | |
| "loss": 3.4691, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 4.563953488372093e-05, | |
| "loss": 3.2751, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 3.3520445823669434, | |
| "eval_runtime": 147.0097, | |
| "eval_samples_per_second": 163.377, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.418604651162791e-05, | |
| "loss": 3.1479, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "learning_rate": 4.2732558139534885e-05, | |
| "loss": 3.0201, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 3.3326475620269775, | |
| "eval_runtime": 146.9025, | |
| "eval_samples_per_second": 163.496, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 4.127906976744187e-05, | |
| "loss": 2.8872, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 3.355100154876709, | |
| "eval_runtime": 146.8556, | |
| "eval_samples_per_second": 163.548, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 3.9825581395348835e-05, | |
| "loss": 2.8373, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 3.837209302325582e-05, | |
| "loss": 2.6775, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 3.35534405708313, | |
| "eval_runtime": 146.787, | |
| "eval_samples_per_second": 163.625, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 5.23, | |
| "learning_rate": 3.691860465116279e-05, | |
| "loss": 2.6133, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 3.5465116279069774e-05, | |
| "loss": 2.5406, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 3.336427927017212, | |
| "eval_runtime": 146.5342, | |
| "eval_samples_per_second": 163.907, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 3.401162790697674e-05, | |
| "loss": 2.4318, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 3.2558139534883724e-05, | |
| "loss": 2.4171, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 3.332899808883667, | |
| "eval_runtime": 146.7159, | |
| "eval_samples_per_second": 163.704, | |
| "step": 6020 | |
| }, | |
| { | |
| "epoch": 7.56, | |
| "learning_rate": 3.11046511627907e-05, | |
| "loss": 2.267, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 3.3342535495758057, | |
| "eval_runtime": 146.2923, | |
| "eval_samples_per_second": 164.178, | |
| "step": 6880 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 2.9651162790697678e-05, | |
| "loss": 2.232, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 8.72, | |
| "learning_rate": 2.8197674418604653e-05, | |
| "loss": 2.1545, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 3.3147542476654053, | |
| "eval_runtime": 147.8917, | |
| "eval_samples_per_second": 162.403, | |
| "step": 7740 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.674418604651163e-05, | |
| "loss": 2.1127, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "learning_rate": 2.5290697674418607e-05, | |
| "loss": 2.064, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 3.3158812522888184, | |
| "eval_runtime": 147.2862, | |
| "eval_samples_per_second": 163.07, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 2.3837209302325582e-05, | |
| "loss": 1.9663, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 3.3465089797973633, | |
| "eval_runtime": 148.1203, | |
| "eval_samples_per_second": 162.152, | |
| "step": 9460 | |
| }, | |
| { | |
| "epoch": 11.05, | |
| "learning_rate": 2.238372093023256e-05, | |
| "loss": 1.9353, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 2.0930232558139536e-05, | |
| "loss": 1.8572, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 3.2874529361724854, | |
| "eval_runtime": 147.3935, | |
| "eval_samples_per_second": 162.952, | |
| "step": 10320 | |
| }, | |
| { | |
| "epoch": 12.21, | |
| "learning_rate": 1.9476744186046514e-05, | |
| "loss": 1.8373, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 1.802325581395349e-05, | |
| "loss": 1.7866, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 3.321674108505249, | |
| "eval_runtime": 146.9637, | |
| "eval_samples_per_second": 163.428, | |
| "step": 11180 | |
| }, | |
| { | |
| "epoch": 13.37, | |
| "learning_rate": 1.6569767441860464e-05, | |
| "loss": 1.7253, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 1.5116279069767441e-05, | |
| "loss": 1.719, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 3.2760589122772217, | |
| "eval_runtime": 145.6136, | |
| "eval_samples_per_second": 164.943, | |
| "step": 12040 | |
| }, | |
| { | |
| "epoch": 14.53, | |
| "learning_rate": 1.3662790697674418e-05, | |
| "loss": 1.6293, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 3.302276611328125, | |
| "eval_runtime": 146.2216, | |
| "eval_samples_per_second": 164.258, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 1.2209302325581395e-05, | |
| "loss": 1.6339, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 15.7, | |
| "learning_rate": 1.0755813953488372e-05, | |
| "loss": 1.5656, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 3.2689430713653564, | |
| "eval_runtime": 145.5127, | |
| "eval_samples_per_second": 165.058, | |
| "step": 13760 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 9.302325581395349e-06, | |
| "loss": 1.5503, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 16.86, | |
| "learning_rate": 7.848837209302325e-06, | |
| "loss": 1.5344, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 3.237142562866211, | |
| "eval_runtime": 146.4021, | |
| "eval_samples_per_second": 164.055, | |
| "step": 14620 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 6.395348837209303e-06, | |
| "loss": 1.4974, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 3.209892749786377, | |
| "eval_runtime": 147.2158, | |
| "eval_samples_per_second": 163.148, | |
| "step": 15480 | |
| }, | |
| { | |
| "epoch": 18.02, | |
| "learning_rate": 4.941860465116279e-06, | |
| "loss": 1.4739, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 3.488372093023256e-06, | |
| "loss": 1.4574, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 3.2351009845733643, | |
| "eval_runtime": 146.7384, | |
| "eval_samples_per_second": 163.679, | |
| "step": 16340 | |
| }, | |
| { | |
| "epoch": 19.19, | |
| "learning_rate": 2.0348837209302328e-06, | |
| "loss": 1.4283, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 5.813953488372093e-07, | |
| "loss": 1.4204, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 3.2112162113189697, | |
| "eval_runtime": 147.2288, | |
| "eval_samples_per_second": 163.134, | |
| "step": 17200 | |
| } | |
| ], | |
| "max_steps": 17200, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.8827072512e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |