| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9818579918673755, | |
| "eval_steps": 10000, | |
| "global_step": 198, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.889162063598633, | |
| "learning_rate": 9.494949494949495e-05, | |
| "loss": 1.2812, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.6296908855438232, | |
| "learning_rate": 8.98989898989899e-05, | |
| "loss": 0.4083, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.7294905185699463, | |
| "learning_rate": 8.484848484848486e-05, | |
| "loss": 0.3107, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.2312090396881104, | |
| "learning_rate": 7.97979797979798e-05, | |
| "loss": 0.2749, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.5575577020645142, | |
| "learning_rate": 7.474747474747475e-05, | |
| "loss": 0.2595, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.0925426483154297, | |
| "learning_rate": 6.96969696969697e-05, | |
| "loss": 0.2348, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.147430658340454, | |
| "learning_rate": 6.464646464646466e-05, | |
| "loss": 0.2328, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 1.0835628509521484, | |
| "learning_rate": 5.959595959595959e-05, | |
| "loss": 0.2262, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.1361318826675415, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.2039, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.2124133110046387, | |
| "learning_rate": 4.94949494949495e-05, | |
| "loss": 0.1923, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.1303669214248657, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.1269, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 0.9401446580886841, | |
| "learning_rate": 3.939393939393939e-05, | |
| "loss": 0.1166, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 0.707255482673645, | |
| "learning_rate": 3.434343434343435e-05, | |
| "loss": 0.1152, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 0.9547311663627625, | |
| "learning_rate": 2.9292929292929294e-05, | |
| "loss": 0.1036, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 0.9109894037246704, | |
| "learning_rate": 2.4242424242424244e-05, | |
| "loss": 0.1132, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 0.951465368270874, | |
| "learning_rate": 1.919191919191919e-05, | |
| "loss": 0.1012, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 0.6391630172729492, | |
| "learning_rate": 1.4141414141414141e-05, | |
| "loss": 0.0948, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 1.319863200187683, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.0906, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 0.5226757526397705, | |
| "learning_rate": 4.040404040404041e-06, | |
| "loss": 0.0996, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "step": 198, | |
| "total_flos": 1077876168327168.0, | |
| "train_loss": 0.23560525100640575, | |
| "train_runtime": 756.2534, | |
| "train_samples_per_second": 33.819, | |
| "train_steps_per_second": 0.262 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 198, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 500, | |
| "total_flos": 1077876168327168.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |