| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 199.0, | |
| "learning_rate": 0.0001, | |
| "loss": 51.5906, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 16.52898597717285, | |
| "eval_runtime": 0.3649, | |
| "eval_samples_per_second": 27.405, | |
| "eval_steps_per_second": 2.74, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 14.1665678024292, | |
| "eval_runtime": 0.2332, | |
| "eval_samples_per_second": 42.89, | |
| "eval_steps_per_second": 4.289, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 40.75, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 38.4458, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 13.090660095214844, | |
| "eval_runtime": 0.2346, | |
| "eval_samples_per_second": 42.63, | |
| "eval_steps_per_second": 4.263, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 11.630752563476562, | |
| "eval_runtime": 0.2341, | |
| "eval_samples_per_second": 42.709, | |
| "eval_steps_per_second": 4.271, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 12.9375, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 23.9261, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 10.357588768005371, | |
| "eval_runtime": 0.2331, | |
| "eval_samples_per_second": 42.901, | |
| "eval_steps_per_second": 4.29, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 9.484556198120117, | |
| "eval_runtime": 0.2457, | |
| "eval_samples_per_second": 40.7, | |
| "eval_steps_per_second": 4.07, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 9.030795097351074, | |
| "eval_runtime": 0.2351, | |
| "eval_samples_per_second": 42.541, | |
| "eval_steps_per_second": 4.254, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 5.65625, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 20.7948, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 8.803549766540527, | |
| "eval_runtime": 0.2348, | |
| "eval_samples_per_second": 42.593, | |
| "eval_steps_per_second": 4.259, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 8.740669250488281, | |
| "eval_runtime": 0.2369, | |
| "eval_samples_per_second": 42.218, | |
| "eval_steps_per_second": 4.222, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 4.65625, | |
| "learning_rate": 0.0, | |
| "loss": 20.2787, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 8.724044799804688, | |
| "eval_runtime": 0.2339, | |
| "eval_samples_per_second": 42.762, | |
| "eval_steps_per_second": 4.276, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 20, | |
| "total_flos": 6.098355666236211e+16, | |
| "train_loss": 26.5185827255249, | |
| "train_runtime": 51.1971, | |
| "train_samples_per_second": 20.314, | |
| "train_steps_per_second": 0.391 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "total_flos": 6.098355666236211e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |