| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.09467455621301775, | |
| "eval_steps": 20, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001183431952662722, | |
| "eval_loss": 3.7497785091400146, | |
| "eval_runtime": 2.512, | |
| "eval_samples_per_second": 597.919, | |
| "eval_steps_per_second": 37.42, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011834319526627219, | |
| "grad_norm": Infinity, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.4365, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "grad_norm": Infinity, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 3.552, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "eval_loss": 3.752192735671997, | |
| "eval_runtime": 2.5179, | |
| "eval_samples_per_second": 596.534, | |
| "eval_steps_per_second": 37.333, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03550295857988166, | |
| "grad_norm": Infinity, | |
| "learning_rate": 4.8e-05, | |
| "loss": 3.6338, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "grad_norm": Infinity, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 3.7366, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "eval_loss": 3.750141143798828, | |
| "eval_runtime": 2.4739, | |
| "eval_samples_per_second": 607.148, | |
| "eval_steps_per_second": 37.997, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": Infinity, | |
| "learning_rate": 8e-05, | |
| "loss": 3.7361, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "grad_norm": Infinity, | |
| "learning_rate": 9.6e-05, | |
| "loss": 3.5694, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "eval_loss": 3.730466842651367, | |
| "eval_runtime": 2.4847, | |
| "eval_samples_per_second": 604.506, | |
| "eval_steps_per_second": 37.832, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08284023668639054, | |
| "grad_norm": Infinity, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 3.9619, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "grad_norm": Infinity, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 3.5414, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "eval_loss": 3.7425360679626465, | |
| "eval_runtime": 3.6241, | |
| "eval_samples_per_second": 414.453, | |
| "eval_steps_per_second": 25.938, | |
| "step": 80 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2171445471870976.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |