| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 2000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.76945424079895, | |
| "learning_rate": 9.940874595551404e-05, | |
| "loss": 1.2118, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.4828996658325195, | |
| "learning_rate": 9.760113134728384e-05, | |
| "loss": 0.8364, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.7079859972000122, | |
| "learning_rate": 9.46214189061859e-05, | |
| "loss": 0.6766, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.6615920066833496, | |
| "learning_rate": 9.054297904161868e-05, | |
| "loss": 0.545, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.234976768493652, | |
| "learning_rate": 8.54662364786137e-05, | |
| "loss": 0.4208, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 14.309597969055176, | |
| "learning_rate": 7.951619746781474e-05, | |
| "loss": 0.3242, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 6.556349277496338, | |
| "learning_rate": 7.283937171671498e-05, | |
| "loss": 0.2409, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 5.399975776672363, | |
| "learning_rate": 6.560016483442075e-05, | |
| "loss": 0.1779, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 4.709582328796387, | |
| "learning_rate": 5.797683011992432e-05, | |
| "loss": 0.1373, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.5215970277786255, | |
| "learning_rate": 5.0157079374293983e-05, | |
| "loss": 0.1079, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 8.067852973937988, | |
| "learning_rate": 4.233346081315196e-05, | |
| "loss": 0.0849, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.95754873752594, | |
| "learning_rate": 3.469861789057497e-05, | |
| "loss": 0.0693, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.9738472104072571, | |
| "learning_rate": 2.7440545777907746e-05, | |
| "loss": 0.0583, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.98776775598526, | |
| "learning_rate": 2.0737962298724512e-05, | |
| "loss": 0.0505, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 1.5931845903396606, | |
| "learning_rate": 1.4755907302881927e-05, | |
| "loss": 0.0448, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 2.104052782058716, | |
| "learning_rate": 9.641678837679985e-06, | |
| "loss": 0.0412, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.45166563987731934, | |
| "learning_rate": 5.521206181083111e-06, | |
| "loss": 0.0386, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 0.7560405135154724, | |
| "learning_rate": 2.4959490449614096e-06, | |
| "loss": 0.0368, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.8283025026321411, | |
| "learning_rate": 6.403993002590425e-07, | |
| "loss": 0.0358, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.7078251242637634, | |
| "learning_rate": 2.4673990708934393e-10, | |
| "loss": 0.0353, | |
| "step": 2000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.56828324077568e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |