| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 18.761904761904763, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.6349206349206349, | |
| "grad_norm": 1.5354194641113281, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 1.3494, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.253968253968254, | |
| "grad_norm": 0.43494924902915955, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.7194, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.8888888888888888, | |
| "grad_norm": 0.5042936205863953, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6417, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.507936507936508, | |
| "grad_norm": 0.37082698941230774, | |
| "learning_rate": 1.9932383577419432e-05, | |
| "loss": 0.5639, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.126984126984127, | |
| "grad_norm": 0.3500129282474518, | |
| "learning_rate": 1.973044870579824e-05, | |
| "loss": 0.5159, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.761904761904762, | |
| "grad_norm": 0.47998425364494324, | |
| "learning_rate": 1.9396926207859085e-05, | |
| "loss": 0.4783, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.380952380952381, | |
| "grad_norm": 0.7525663375854492, | |
| "learning_rate": 1.8936326403234125e-05, | |
| "loss": 0.4346, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.5472385287284851, | |
| "learning_rate": 1.8354878114129368e-05, | |
| "loss": 0.4163, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.634920634920634, | |
| "grad_norm": 0.8724488615989685, | |
| "learning_rate": 1.766044443118978e-05, | |
| "loss": 0.3745, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.253968253968254, | |
| "grad_norm": 0.5862165093421936, | |
| "learning_rate": 1.686241637868734e-05, | |
| "loss": 0.3326, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 6.888888888888889, | |
| "grad_norm": 0.9971191883087158, | |
| "learning_rate": 1.5971585917027864e-05, | |
| "loss": 0.2989, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.507936507936508, | |
| "grad_norm": 0.8492709398269653, | |
| "learning_rate": 1.5000000000000002e-05, | |
| "loss": 0.2443, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.126984126984127, | |
| "grad_norm": 0.9441637992858887, | |
| "learning_rate": 1.396079766039157e-05, | |
| "loss": 0.2203, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 8.761904761904763, | |
| "grad_norm": 0.9310017228126526, | |
| "learning_rate": 1.2868032327110904e-05, | |
| "loss": 0.1696, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.380952380952381, | |
| "grad_norm": 0.9161068201065063, | |
| "learning_rate": 1.1736481776669307e-05, | |
| "loss": 0.1336, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.643371045589447, | |
| "learning_rate": 1.0581448289104759e-05, | |
| "loss": 0.1025, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 10.634920634920634, | |
| "grad_norm": 1.1491143703460693, | |
| "learning_rate": 9.418551710895243e-06, | |
| "loss": 0.0676, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 11.253968253968253, | |
| "grad_norm": 0.5557671189308167, | |
| "learning_rate": 8.263518223330698e-06, | |
| "loss": 0.0616, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 11.88888888888889, | |
| "grad_norm": 0.5206998586654663, | |
| "learning_rate": 7.131967672889101e-06, | |
| "loss": 0.045, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 12.507936507936508, | |
| "grad_norm": 0.356916218996048, | |
| "learning_rate": 6.039202339608432e-06, | |
| "loss": 0.0269, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 13.126984126984127, | |
| "grad_norm": 0.2968302369117737, | |
| "learning_rate": 5.000000000000003e-06, | |
| "loss": 0.0192, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 13.761904761904763, | |
| "grad_norm": 0.15373599529266357, | |
| "learning_rate": 4.028414082972141e-06, | |
| "loss": 0.0146, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 14.380952380952381, | |
| "grad_norm": 0.1345711648464203, | |
| "learning_rate": 3.1375836213126653e-06, | |
| "loss": 0.0115, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.11057666689157486, | |
| "learning_rate": 2.339555568810221e-06, | |
| "loss": 0.0092, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 15.634920634920634, | |
| "grad_norm": 0.08674044162034988, | |
| "learning_rate": 1.6451218858706374e-06, | |
| "loss": 0.0081, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 16.253968253968253, | |
| "grad_norm": 0.08988489955663681, | |
| "learning_rate": 1.0636735967658785e-06, | |
| "loss": 0.0076, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 16.88888888888889, | |
| "grad_norm": 0.07731083780527115, | |
| "learning_rate": 6.030737921409169e-07, | |
| "loss": 0.0077, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 17.507936507936506, | |
| "grad_norm": 0.08936144411563873, | |
| "learning_rate": 2.6955129420176193e-07, | |
| "loss": 0.0065, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 18.126984126984127, | |
| "grad_norm": 0.11082036793231964, | |
| "learning_rate": 6.761642258056977e-08, | |
| "loss": 0.0073, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 18.761904761904763, | |
| "grad_norm": 0.13867439329624176, | |
| "learning_rate": 0.0, | |
| "loss": 0.0069, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.416508510620877e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |