| { | |
| "best_metric": 0.8896821741133119, | |
| "best_model_checkpoint": "model_ajustado/checkpoint-3258", | |
| "epoch": 3.0, | |
| "global_step": 3258, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 9.801104972375691e-06, | |
| "loss": 0.6336, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 9.602209944751382e-06, | |
| "loss": 0.445, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 9.403314917127072e-06, | |
| "loss": 0.4139, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 9.204419889502764e-06, | |
| "loss": 0.3952, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.005524861878454e-06, | |
| "loss": 0.374, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.806629834254145e-06, | |
| "loss": 0.3476, | |
| "step": 648 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 8.607734806629835e-06, | |
| "loss": 0.3492, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 8.408839779005525e-06, | |
| "loss": 0.3277, | |
| "step": 864 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 8.209944751381216e-06, | |
| "loss": 0.3479, | |
| "step": 972 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 8.011049723756906e-06, | |
| "loss": 0.3279, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8813910640257946, | |
| "eval_loss": 0.31820762157440186, | |
| "eval_runtime": 24.2259, | |
| "eval_samples_per_second": 179.23, | |
| "eval_steps_per_second": 11.228, | |
| "step": 1086 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "learning_rate": 7.812154696132598e-06, | |
| "loss": 0.2958, | |
| "step": 1188 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 7.613259668508289e-06, | |
| "loss": 0.2934, | |
| "step": 1296 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 7.414364640883979e-06, | |
| "loss": 0.2638, | |
| "step": 1404 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "learning_rate": 7.215469613259669e-06, | |
| "loss": 0.2794, | |
| "step": 1512 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 7.01657458563536e-06, | |
| "loss": 0.258, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "learning_rate": 6.81767955801105e-06, | |
| "loss": 0.2662, | |
| "step": 1728 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 6.6187845303867405e-06, | |
| "loss": 0.3038, | |
| "step": 1836 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 6.419889502762432e-06, | |
| "loss": 0.2594, | |
| "step": 1944 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 6.220994475138122e-06, | |
| "loss": 0.2475, | |
| "step": 2052 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.0220994475138124e-06, | |
| "loss": 0.2922, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8871487793643482, | |
| "eval_loss": 0.2970804274082184, | |
| "eval_runtime": 24.4656, | |
| "eval_samples_per_second": 177.473, | |
| "eval_steps_per_second": 11.118, | |
| "step": 2172 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 5.823204419889504e-06, | |
| "loss": 0.2174, | |
| "step": 2268 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 5.624309392265194e-06, | |
| "loss": 0.2346, | |
| "step": 2376 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 5.425414364640884e-06, | |
| "loss": 0.2061, | |
| "step": 2484 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 5.226519337016575e-06, | |
| "loss": 0.2441, | |
| "step": 2592 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 5.027624309392266e-06, | |
| "loss": 0.2284, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.59, | |
| "learning_rate": 4.828729281767956e-06, | |
| "loss": 0.2147, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 4.629834254143647e-06, | |
| "loss": 0.2296, | |
| "step": 2916 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 4.430939226519337e-06, | |
| "loss": 0.2489, | |
| "step": 3024 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 4.2320441988950274e-06, | |
| "loss": 0.2199, | |
| "step": 3132 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 4.033149171270719e-06, | |
| "loss": 0.2178, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8896821741133119, | |
| "eval_loss": 0.3544089198112488, | |
| "eval_runtime": 24.4186, | |
| "eval_samples_per_second": 177.816, | |
| "eval_steps_per_second": 11.139, | |
| "step": 3258 | |
| } | |
| ], | |
| "max_steps": 5430, | |
| "num_train_epochs": 5, | |
| "total_flos": 2646085391685036.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |