| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0022637238256932655, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00025152486952147396, | |
| "grad_norm": 0.19798757135868073, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8211, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00025152486952147396, | |
| "eval_loss": NaN, | |
| "eval_runtime": 57.5657, | |
| "eval_samples_per_second": 29.097, | |
| "eval_steps_per_second": 14.557, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005030497390429479, | |
| "grad_norm": 0.19033122062683105, | |
| "learning_rate": 4e-05, | |
| "loss": 0.6942, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0007545746085644218, | |
| "grad_norm": 0.2211087942123413, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9675, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0007545746085644218, | |
| "eval_loss": NaN, | |
| "eval_runtime": 56.1471, | |
| "eval_samples_per_second": 29.832, | |
| "eval_steps_per_second": 14.925, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0010060994780858958, | |
| "grad_norm": 0.23829740285873413, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8838, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0012576243476073697, | |
| "grad_norm": 0.18251283466815948, | |
| "learning_rate": 0.0001, | |
| "loss": 0.4454, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0015091492171288435, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00012, | |
| "loss": 0.703, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0015091492171288435, | |
| "eval_loss": NaN, | |
| "eval_runtime": 56.3421, | |
| "eval_samples_per_second": 29.729, | |
| "eval_steps_per_second": 14.873, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0017606740866503176, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.00014, | |
| "loss": 0.8471, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0020121989561717917, | |
| "grad_norm": 0.2415139526128769, | |
| "learning_rate": 0.00016, | |
| "loss": 0.7513, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0022637238256932655, | |
| "grad_norm": 0.21635624766349792, | |
| "learning_rate": 0.00018, | |
| "loss": 0.5943, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0022637238256932655, | |
| "eval_loss": NaN, | |
| "eval_runtime": 56.2065, | |
| "eval_samples_per_second": 29.801, | |
| "eval_steps_per_second": 14.909, | |
| "step": 9 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 10, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 3, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1252301482229760.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |