| { | |
| "best_metric": 0.48496538400650024, | |
| "best_model_checkpoint": "mikhail-panzo/fil_b32_le5_s4000/checkpoint-500", | |
| "epoch": 44.44444444444444, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 4.288165092468262, | |
| "learning_rate": 2.5000000000000004e-07, | |
| "loss": 0.7989, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 2.4493463039398193, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 0.7701, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 1.3798907995224, | |
| "learning_rate": 7.5e-07, | |
| "loss": 0.7374, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 17.77777777777778, | |
| "grad_norm": 2.6173877716064453, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 0.7101, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 22.22222222222222, | |
| "grad_norm": 1.3756701946258545, | |
| "learning_rate": 1.25e-06, | |
| "loss": 0.6887, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 1.0456843376159668, | |
| "learning_rate": 1.5e-06, | |
| "loss": 0.672, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 31.11111111111111, | |
| "grad_norm": 0.9518398642539978, | |
| "learning_rate": 1.75e-06, | |
| "loss": 0.6583, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 35.55555555555556, | |
| "grad_norm": 1.305978775024414, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 0.6258, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.8857623338699341, | |
| "learning_rate": 2.25e-06, | |
| "loss": 0.5835, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 44.44444444444444, | |
| "grad_norm": 1.052182912826538, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.5539, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 44.44444444444444, | |
| "eval_loss": 0.48496538400650024, | |
| "eval_runtime": 9.4574, | |
| "eval_samples_per_second": 16.812, | |
| "eval_steps_per_second": 2.115, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 4000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 364, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.3753846024576416e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |