| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.018361581920903956, | |
| "eval_steps": 10, | |
| "global_step": 39, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00047080979284369113, | |
| "eval_loss": 4.613402843475342, | |
| "eval_runtime": 95.6249, | |
| "eval_samples_per_second": 9.359, | |
| "eval_steps_per_second": 4.685, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0023540489642184556, | |
| "grad_norm": 3.0230181217193604, | |
| "learning_rate": 5e-05, | |
| "loss": 4.9196, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.004708097928436911, | |
| "grad_norm": 1.0691684484481812, | |
| "learning_rate": 0.0001, | |
| "loss": 4.2913, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004708097928436911, | |
| "eval_loss": 4.355589866638184, | |
| "eval_runtime": 95.3377, | |
| "eval_samples_per_second": 9.388, | |
| "eval_steps_per_second": 4.699, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.007062146892655367, | |
| "grad_norm": 0.8888987302780151, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 3.6859, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.009416195856873822, | |
| "grad_norm": 1.3303650617599487, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 3.9099, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009416195856873822, | |
| "eval_loss": 3.8060638904571533, | |
| "eval_runtime": 95.3447, | |
| "eval_samples_per_second": 9.387, | |
| "eval_steps_per_second": 4.699, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.011770244821092278, | |
| "grad_norm": 1.7675096988677979, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 3.6147, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.014124293785310734, | |
| "grad_norm": 1.2738316059112549, | |
| "learning_rate": 5e-05, | |
| "loss": 3.7694, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.014124293785310734, | |
| "eval_loss": 3.6177690029144287, | |
| "eval_runtime": 95.6706, | |
| "eval_samples_per_second": 9.355, | |
| "eval_steps_per_second": 4.683, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.01647834274952919, | |
| "grad_norm": 1.5537434816360474, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 3.6585, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.286248505278464e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |