| { | |
| "best_metric": 1.8321382999420166, | |
| "best_model_checkpoint": "output/lyapis-trubetskoy/checkpoint-72", | |
| "epoch": 3.0, | |
| "global_step": 72, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 0.00012302403914397873, | |
| "loss": 2.4696, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 8.635498649403293e-05, | |
| "loss": 2.2437, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 4.2347916539754844e-05, | |
| "loss": 1.9644, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 9.190657300387505e-06, | |
| "loss": 1.9388, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.9219719171524048, | |
| "eval_runtime": 1.5376, | |
| "eval_samples_per_second": 22.112, | |
| "eval_steps_per_second": 3.252, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 5.868825097566078e-07, | |
| "loss": 2.1206, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 2.009247481060283e-05, | |
| "loss": 1.9512, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 5.964590321370446e-05, | |
| "loss": 2.0299, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 0.00010290000000000001, | |
| "loss": 2.1005, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.00013197813593027427, | |
| "loss": 1.852, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.857775330543518, | |
| "eval_runtime": 1.5473, | |
| "eval_samples_per_second": 21.973, | |
| "eval_steps_per_second": 3.231, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.00013486251168343008, | |
| "loss": 1.8459, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 0.00011036103402999826, | |
| "loss": 1.7933, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 6.860000000000001e-05, | |
| "loss": 1.9168, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 2.683896597000177e-05, | |
| "loss": 1.9094, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.337488316569914e-06, | |
| "loss": 1.7401, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.8321382999420166, | |
| "eval_runtime": 1.5433, | |
| "eval_samples_per_second": 22.031, | |
| "eval_steps_per_second": 3.24, | |
| "step": 72 | |
| } | |
| ], | |
| "max_steps": 72, | |
| "num_train_epochs": 3, | |
| "total_flos": 75252105216000.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |