| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.053016453382084, | |
| "eval_steps": 5, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05850091407678245, | |
| "grad_norm": 7.276617050170898, | |
| "learning_rate": 1e-05, | |
| "loss": 8.7408, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05850091407678245, | |
| "eval_loss": 9.801617622375488, | |
| "eval_runtime": 17.0417, | |
| "eval_samples_per_second": 6.807, | |
| "eval_steps_per_second": 1.702, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1170018281535649, | |
| "grad_norm": 7.246295928955078, | |
| "learning_rate": 2e-05, | |
| "loss": 8.7566, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.17550274223034734, | |
| "grad_norm": 6.676323413848877, | |
| "learning_rate": 3e-05, | |
| "loss": 8.5029, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2340036563071298, | |
| "grad_norm": 6.782390117645264, | |
| "learning_rate": 4e-05, | |
| "loss": 8.6189, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.29250457038391225, | |
| "grad_norm": 6.849842071533203, | |
| "learning_rate": 5e-05, | |
| "loss": 8.7124, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.29250457038391225, | |
| "eval_loss": 9.541211128234863, | |
| "eval_runtime": 16.353, | |
| "eval_samples_per_second": 7.093, | |
| "eval_steps_per_second": 1.773, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3510054844606947, | |
| "grad_norm": 6.971000671386719, | |
| "learning_rate": 6e-05, | |
| "loss": 8.2878, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.40950639853747717, | |
| "grad_norm": 6.006114482879639, | |
| "learning_rate": 7e-05, | |
| "loss": 7.6127, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4680073126142596, | |
| "grad_norm": 6.785839557647705, | |
| "learning_rate": 8e-05, | |
| "loss": 7.419, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.526508226691042, | |
| "grad_norm": 6.747194766998291, | |
| "learning_rate": 9e-05, | |
| "loss": 6.9854, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5850091407678245, | |
| "grad_norm": 5.7017059326171875, | |
| "learning_rate": 0.0001, | |
| "loss": 6.3435, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5850091407678245, | |
| "eval_loss": 6.967709541320801, | |
| "eval_runtime": 16.3531, | |
| "eval_samples_per_second": 7.093, | |
| "eval_steps_per_second": 1.773, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.643510054844607, | |
| "grad_norm": 6.990315914154053, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 6.2191, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.7020109689213894, | |
| "grad_norm": 6.611320495605469, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 4.9522, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.7605118829981719, | |
| "grad_norm": 8.069223403930664, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 4.3585, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.8190127970749543, | |
| "grad_norm": 7.039530277252197, | |
| "learning_rate": 5e-05, | |
| "loss": 3.7174, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.8775137111517367, | |
| "grad_norm": 9.83570384979248, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 3.8227, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8775137111517367, | |
| "eval_loss": 4.3420915603637695, | |
| "eval_runtime": 16.3441, | |
| "eval_samples_per_second": 7.097, | |
| "eval_steps_per_second": 1.774, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.9360146252285192, | |
| "grad_norm": 10.739554405212402, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 3.776, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.9945155393053017, | |
| "grad_norm": 9.82874870300293, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 3.9549, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.053016453382084, | |
| "grad_norm": 26.742786407470703, | |
| "learning_rate": 0.0, | |
| "loss": 8.1481, | |
| "step": 18 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.881887547799634e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |