| { | |
| "best_metric": 0.7533459966277345, | |
| "best_model_checkpoint": "ckpt/tuned_unli/checkpoint-6940", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 6940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 11.59940242767334, | |
| "learning_rate": 4.855907780979828e-06, | |
| "loss": 0.4912, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 4.5099921226501465, | |
| "learning_rate": 4.711815561959655e-06, | |
| "loss": 0.4898, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 4.301983833312988, | |
| "learning_rate": 4.567723342939482e-06, | |
| "loss": 0.4803, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 8.535087585449219, | |
| "learning_rate": 4.423631123919308e-06, | |
| "loss": 0.486, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 6.304492473602295, | |
| "learning_rate": 4.279538904899136e-06, | |
| "loss": 0.4725, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 5.865860939025879, | |
| "learning_rate": 4.135446685878963e-06, | |
| "loss": 0.4738, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5043371915817261, | |
| "eval_mse": 0.05504356324672699, | |
| "eval_pearson": 0.7481511352198802, | |
| "eval_runtime": 11.1028, | |
| "eval_samples_per_second": 273.805, | |
| "eval_spearman": 0.7563540381881723, | |
| "eval_steps_per_second": 17.113, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 9.249129295349121, | |
| "learning_rate": 3.99135446685879e-06, | |
| "loss": 0.4737, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 5.8351874351501465, | |
| "learning_rate": 3.847262247838617e-06, | |
| "loss": 0.4416, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 5.798779010772705, | |
| "learning_rate": 3.703170028818444e-06, | |
| "loss": 0.4457, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 6.772558212280273, | |
| "learning_rate": 3.5590778097982716e-06, | |
| "loss": 0.4448, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 12.370868682861328, | |
| "learning_rate": 3.4149855907780984e-06, | |
| "loss": 0.4499, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 7.551795482635498, | |
| "learning_rate": 3.2708933717579256e-06, | |
| "loss": 0.4441, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 3.4603450298309326, | |
| "learning_rate": 3.1268011527377524e-06, | |
| "loss": 0.4466, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.5080664157867432, | |
| "eval_mse": 0.055369943380355835, | |
| "eval_pearson": 0.7533459966277345, | |
| "eval_runtime": 10.7962, | |
| "eval_samples_per_second": 281.582, | |
| "eval_spearman": 0.7597360368405089, | |
| "eval_steps_per_second": 17.599, | |
| "step": 6940 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 17350, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.0217853882348156e+16, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |