| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 5.649717514124294, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 0.47637438774108887, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 1.5071, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 0.6510860919952393, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 1.3602, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.7494458556175232, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 1.2589, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 0.9085477590560913, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 1.1859, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "grad_norm": 1.0386964082717896, | |
| "learning_rate": 0.0001, | |
| "loss": 1.093, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "grad_norm": 1.1297998428344727, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.9995, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "grad_norm": 1.2726491689682007, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.9538, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.52, | |
| "grad_norm": 1.2885305881500244, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.854, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "grad_norm": 1.14137864112854, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.8395, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "grad_norm": 1.1960450410842896, | |
| "learning_rate": 0.0, | |
| "loss": 0.806, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 1.480881862385664e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |