| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.03568242640499554, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007136485280999108, | |
| "eval_loss": 0.648069441318512, | |
| "eval_runtime": 25.3297, | |
| "eval_samples_per_second": 23.293, | |
| "eval_steps_per_second": 11.646, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.007136485280999108, | |
| "grad_norm": 0.7408186197280884, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.644, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.009277430865298841, | |
| "eval_loss": 0.5619068145751953, | |
| "eval_runtime": 25.3319, | |
| "eval_samples_per_second": 23.291, | |
| "eval_steps_per_second": 11.645, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014272970561998216, | |
| "grad_norm": 0.6046487092971802, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.548, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.018554861730597682, | |
| "eval_loss": 0.5528106093406677, | |
| "eval_runtime": 25.4487, | |
| "eval_samples_per_second": 23.184, | |
| "eval_steps_per_second": 11.592, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.021409455842997322, | |
| "grad_norm": 0.5992456674575806, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.5277, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.02783229259589652, | |
| "eval_loss": 0.5509209036827087, | |
| "eval_runtime": 25.3803, | |
| "eval_samples_per_second": 23.246, | |
| "eval_steps_per_second": 11.623, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.028545941123996433, | |
| "grad_norm": 0.6934235095977783, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.5933, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03568242640499554, | |
| "grad_norm": 0.6936048269271851, | |
| "learning_rate": 0.0, | |
| "loss": 0.4805, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.17054349246464e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |