| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.997356828193833, | |
| "global_step": 2830, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 9.293286219081273e-05, | |
| "loss": 1.0928, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_accuracy": 0.47887325286865234, | |
| "eval_loss": 1.5795212984085083, | |
| "eval_runtime": 38.2173, | |
| "eval_samples_per_second": 14.862, | |
| "eval_steps_per_second": 7.431, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 8.593639575971731e-05, | |
| "loss": 0.7645, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "eval_accuracy": 0.7869718074798584, | |
| "eval_loss": 0.742794930934906, | |
| "eval_runtime": 37.531, | |
| "eval_samples_per_second": 15.134, | |
| "eval_steps_per_second": 7.567, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 7.886925795053004e-05, | |
| "loss": 0.7006, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "eval_accuracy": 0.8362675905227661, | |
| "eval_loss": 0.6150082349777222, | |
| "eval_runtime": 40.2645, | |
| "eval_samples_per_second": 14.107, | |
| "eval_steps_per_second": 7.053, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 7.180212014134276e-05, | |
| "loss": 0.5091, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.8309859037399292, | |
| "eval_loss": 0.6766383051872253, | |
| "eval_runtime": 37.4365, | |
| "eval_samples_per_second": 15.172, | |
| "eval_steps_per_second": 7.586, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 6.473498233215549e-05, | |
| "loss": 0.4137, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "eval_accuracy": 0.8644366264343262, | |
| "eval_loss": 0.6185892820358276, | |
| "eval_runtime": 40.2147, | |
| "eval_samples_per_second": 14.124, | |
| "eval_steps_per_second": 7.062, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "learning_rate": 5.7667844522968195e-05, | |
| "loss": 0.3657, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.24, | |
| "eval_accuracy": 0.8978873491287231, | |
| "eval_loss": 0.5504735112190247, | |
| "eval_runtime": 37.4129, | |
| "eval_samples_per_second": 15.182, | |
| "eval_steps_per_second": 7.591, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "learning_rate": 5.0636042402826856e-05, | |
| "loss": 0.3252, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_accuracy": 0.8838028311729431, | |
| "eval_loss": 0.5585792660713196, | |
| "eval_runtime": 39.9534, | |
| "eval_samples_per_second": 14.217, | |
| "eval_steps_per_second": 7.108, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 4.356890459363958e-05, | |
| "loss": 0.2616, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "eval_accuracy": 0.8820422291755676, | |
| "eval_loss": 0.5848411321640015, | |
| "eval_runtime": 37.6048, | |
| "eval_samples_per_second": 15.104, | |
| "eval_steps_per_second": 7.552, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "learning_rate": 3.6501766784452293e-05, | |
| "loss": 0.2373, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 6.36, | |
| "eval_accuracy": 0.8978873491287231, | |
| "eval_loss": 0.596436619758606, | |
| "eval_runtime": 38.595, | |
| "eval_samples_per_second": 14.717, | |
| "eval_steps_per_second": 7.358, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "learning_rate": 2.9434628975265022e-05, | |
| "loss": 0.1699, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.07, | |
| "eval_accuracy": 0.9031690359115601, | |
| "eval_loss": 0.5169617533683777, | |
| "eval_runtime": 40.3197, | |
| "eval_samples_per_second": 14.087, | |
| "eval_steps_per_second": 7.044, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "learning_rate": 2.236749116607774e-05, | |
| "loss": 0.1467, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 7.77, | |
| "eval_accuracy": 0.9066901206970215, | |
| "eval_loss": 0.5568466186523438, | |
| "eval_runtime": 37.5356, | |
| "eval_samples_per_second": 15.132, | |
| "eval_steps_per_second": 7.566, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "learning_rate": 1.530035335689046e-05, | |
| "loss": 0.0975, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 8.48, | |
| "eval_accuracy": 0.9049295783042908, | |
| "eval_loss": 0.5153928995132446, | |
| "eval_runtime": 40.2253, | |
| "eval_samples_per_second": 14.12, | |
| "eval_steps_per_second": 7.06, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "learning_rate": 8.233215547703181e-06, | |
| "loss": 0.0589, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.19, | |
| "eval_accuracy": 0.922535240650177, | |
| "eval_loss": 0.48342233896255493, | |
| "eval_runtime": 37.4958, | |
| "eval_samples_per_second": 15.148, | |
| "eval_steps_per_second": 7.574, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "learning_rate": 1.1660777385159012e-06, | |
| "loss": 0.0775, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 9.89, | |
| "eval_accuracy": 0.922535240650177, | |
| "eval_loss": 0.47951266169548035, | |
| "eval_runtime": 40.6487, | |
| "eval_samples_per_second": 13.973, | |
| "eval_steps_per_second": 6.987, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2830, | |
| "total_flos": 1.204556515540952e+18, | |
| "train_loss": 0.36904500712354277, | |
| "train_runtime": 3573.2389, | |
| "train_samples_per_second": 6.353, | |
| "train_steps_per_second": 0.792 | |
| } | |
| ], | |
| "max_steps": 2830, | |
| "num_train_epochs": 10, | |
| "total_flos": 1.204556515540952e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |