| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.00013320111888939868, | |
| "eval_steps": 3, | |
| "global_step": 9, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.4800124321044296e-05, | |
| "grad_norm": 6.357696533203125, | |
| "learning_rate": 2e-05, | |
| "loss": 6.1877, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.4800124321044296e-05, | |
| "eval_loss": 7.481135845184326, | |
| "eval_runtime": 836.1259, | |
| "eval_samples_per_second": 34.026, | |
| "eval_steps_per_second": 17.013, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 2.9600248642088592e-05, | |
| "grad_norm": 4.821768283843994, | |
| "learning_rate": 4e-05, | |
| "loss": 7.0639, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 4.440037296313289e-05, | |
| "grad_norm": 6.758498191833496, | |
| "learning_rate": 6e-05, | |
| "loss": 8.0063, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 4.440037296313289e-05, | |
| "eval_loss": 7.439722537994385, | |
| "eval_runtime": 834.9323, | |
| "eval_samples_per_second": 34.075, | |
| "eval_steps_per_second": 17.037, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 5.9200497284177184e-05, | |
| "grad_norm": 5.206923007965088, | |
| "learning_rate": 8e-05, | |
| "loss": 7.2314, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 7.400062160522149e-05, | |
| "grad_norm": 8.670096397399902, | |
| "learning_rate": 0.0001, | |
| "loss": 7.2969, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 8.880074592626579e-05, | |
| "grad_norm": 7.160818099975586, | |
| "learning_rate": 0.00012, | |
| "loss": 7.756, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 8.880074592626579e-05, | |
| "eval_loss": 6.845438480377197, | |
| "eval_runtime": 834.7376, | |
| "eval_samples_per_second": 34.083, | |
| "eval_steps_per_second": 17.041, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00010360087024731008, | |
| "grad_norm": 4.990078926086426, | |
| "learning_rate": 0.00014, | |
| "loss": 6.4076, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.00011840099456835437, | |
| "grad_norm": 7.9193572998046875, | |
| "learning_rate": 0.00016, | |
| "loss": 6.4641, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00013320111888939868, | |
| "grad_norm": 8.150678634643555, | |
| "learning_rate": 0.00018, | |
| "loss": 6.7354, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.00013320111888939868, | |
| "eval_loss": 5.326913356781006, | |
| "eval_runtime": 842.0517, | |
| "eval_samples_per_second": 33.787, | |
| "eval_steps_per_second": 16.893, | |
| "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": 216490963894272.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |