| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 48, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.06477732793522267, |
| "grad_norm": 2.6904048919677734, |
| "learning_rate": 3.333333333333333e-07, |
| "loss": 0.7875139713287354, |
| "step": 1, |
| "token_acc": 0.7841123294890244 |
| }, |
| { |
| "epoch": 0.32388663967611336, |
| "grad_norm": 2.6191625595092773, |
| "learning_rate": 9.95134034370785e-07, |
| "loss": 0.7787209749221802, |
| "step": 5, |
| "token_acc": 0.7859186270144402 |
| }, |
| { |
| "epoch": 0.6477732793522267, |
| "grad_norm": 2.2350027561187744, |
| "learning_rate": 9.414737964294634e-07, |
| "loss": 0.7690708637237549, |
| "step": 10, |
| "token_acc": 0.7875346756409188 |
| }, |
| { |
| "epoch": 0.97165991902834, |
| "grad_norm": 1.7701807022094727, |
| "learning_rate": 8.34565303179429e-07, |
| "loss": 0.7466525554656982, |
| "step": 15, |
| "token_acc": 0.7900878692322302 |
| }, |
| { |
| "epoch": 1.2591093117408907, |
| "grad_norm": 1.4035495519638062, |
| "learning_rate": 6.87303296707956e-07, |
| "loss": 0.7205144882202148, |
| "step": 20, |
| "token_acc": 0.7961060457644802 |
| }, |
| { |
| "epoch": 1.582995951417004, |
| "grad_norm": 1.2812139987945557, |
| "learning_rate": 5.174497483512505e-07, |
| "loss": 0.6808315277099609, |
| "step": 25, |
| "token_acc": 0.8017477767899973 |
| }, |
| { |
| "epoch": 1.9068825910931175, |
| "grad_norm": 1.1584919691085815, |
| "learning_rate": 3.454915028125263e-07, |
| "loss": 0.680242919921875, |
| "step": 30, |
| "token_acc": 0.8010927420223759 |
| }, |
| { |
| "epoch": 2.194331983805668, |
| "grad_norm": 0.9851439595222473, |
| "learning_rate": 1.9216926233717084e-07, |
| "loss": 0.6618224143981933, |
| "step": 35, |
| "token_acc": 0.8085923445672631 |
| }, |
| { |
| "epoch": 2.5182186234817814, |
| "grad_norm": 0.9299372434616089, |
| "learning_rate": 7.597595192178702e-08, |
| "loss": 0.662947416305542, |
| "step": 40, |
| "token_acc": 0.8051387469024985 |
| }, |
| { |
| "epoch": 2.8421052631578947, |
| "grad_norm": 0.9523605108261108, |
| "learning_rate": 1.0926199633097154e-08, |
| "loss": 0.6646108627319336, |
| "step": 45, |
| "token_acc": 0.8049214857101155 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 48, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "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": 5.667590557631775e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
