| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 6.564102564102564, | |
| "eval_steps": 500, | |
| "global_step": 32, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.41025641025641024, | |
| "grad_norm": 1.4377572536468506, | |
| "learning_rate": 0.0001875, | |
| "loss": 9.492, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.8205128205128205, | |
| "grad_norm": 2.2069075107574463, | |
| "learning_rate": 0.000175, | |
| "loss": 8.8969, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 2.2332770824432373, | |
| "learning_rate": 0.00016250000000000002, | |
| "loss": 8.1134, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.641025641025641, | |
| "grad_norm": 2.1227211952209473, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 7.5957, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.051282051282051, | |
| "grad_norm": 1.8074111938476562, | |
| "learning_rate": 0.0001375, | |
| "loss": 6.9518, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.6330405473709106, | |
| "learning_rate": 0.000125, | |
| "loss": 6.7793, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.871794871794872, | |
| "grad_norm": 1.6258740425109863, | |
| "learning_rate": 0.00011250000000000001, | |
| "loss": 6.6454, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 3.282051282051282, | |
| "grad_norm": 1.0058015584945679, | |
| "learning_rate": 0.0001, | |
| "loss": 6.6708, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 1.2760761976242065, | |
| "learning_rate": 8.75e-05, | |
| "loss": 6.4539, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 4.102564102564102, | |
| "grad_norm": 1.0890202522277832, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 6.56, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.512820512820513, | |
| "grad_norm": 0.9972789287567139, | |
| "learning_rate": 6.25e-05, | |
| "loss": 6.4305, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 0.7041062712669373, | |
| "learning_rate": 5e-05, | |
| "loss": 6.4812, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 0.6713021397590637, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 6.2827, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 5.743589743589744, | |
| "grad_norm": 0.7198984026908875, | |
| "learning_rate": 2.5e-05, | |
| "loss": 6.3623, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 0.8625803589820862, | |
| "learning_rate": 1.25e-05, | |
| "loss": 6.3825, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.564102564102564, | |
| "grad_norm": 0.4870496690273285, | |
| "learning_rate": 0.0, | |
| "loss": 6.4128, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.564102564102564, | |
| "step": 32, | |
| "total_flos": 177335876845380.0, | |
| "train_loss": 7.031960189342499, | |
| "train_runtime": 130.3681, | |
| "train_samples_per_second": 4.725, | |
| "train_steps_per_second": 0.245 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 32, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 177335876845380.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |