| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.770992366412214, | |
| "eval_steps": 500, | |
| "global_step": 320, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.3053435114503817, | |
| "grad_norm": 1.0634765625, | |
| "learning_rate": 0.0001995184726672197, | |
| "loss": 1.8094, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6106870229007634, | |
| "grad_norm": 0.9521484375, | |
| "learning_rate": 0.00019807852804032305, | |
| "loss": 1.1712, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.916030534351145, | |
| "grad_norm": 2.720703125, | |
| "learning_rate": 0.0001956940335732209, | |
| "loss": 1.0862, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2213740458015268, | |
| "grad_norm": 2.3515625, | |
| "learning_rate": 0.0001923879532511287, | |
| "loss": 0.9789, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.5267175572519083, | |
| "grad_norm": 0.46435546875, | |
| "learning_rate": 0.0001881921264348355, | |
| "loss": 0.8284, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.83206106870229, | |
| "grad_norm": 2.533203125, | |
| "learning_rate": 0.00018314696123025454, | |
| "loss": 0.809, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.1374045801526718, | |
| "grad_norm": 0.8310546875, | |
| "learning_rate": 0.0001773010453362737, | |
| "loss": 0.7111, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.4427480916030535, | |
| "grad_norm": 4.42578125, | |
| "learning_rate": 0.00017071067811865476, | |
| "loss": 0.7286, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.7480916030534353, | |
| "grad_norm": 0.70947265625, | |
| "learning_rate": 0.00016343932841636456, | |
| "loss": 0.7382, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.053435114503817, | |
| "grad_norm": 1.1298828125, | |
| "learning_rate": 0.00015555702330196023, | |
| "loss": 0.7349, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.3587786259541983, | |
| "grad_norm": 2.6484375, | |
| "learning_rate": 0.0001471396736825998, | |
| "loss": 0.7027, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.66412213740458, | |
| "grad_norm": 0.407958984375, | |
| "learning_rate": 0.000138268343236509, | |
| "loss": 0.707, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.969465648854962, | |
| "grad_norm": 0.427001953125, | |
| "learning_rate": 0.00012902846772544624, | |
| "loss": 0.6915, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.2748091603053435, | |
| "grad_norm": 0.34228515625, | |
| "learning_rate": 0.00011950903220161285, | |
| "loss": 0.6953, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.580152671755725, | |
| "grad_norm": 0.489990234375, | |
| "learning_rate": 0.0001098017140329561, | |
| "loss": 0.6831, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.885496183206107, | |
| "grad_norm": 0.38525390625, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6754, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.190839694656488, | |
| "grad_norm": 0.389892578125, | |
| "learning_rate": 9.019828596704394e-05, | |
| "loss": 0.6688, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.4961832061068705, | |
| "grad_norm": 0.467529296875, | |
| "learning_rate": 8.049096779838719e-05, | |
| "loss": 0.668, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.801526717557252, | |
| "grad_norm": 0.39794921875, | |
| "learning_rate": 7.097153227455379e-05, | |
| "loss": 0.6651, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.106870229007634, | |
| "grad_norm": 0.52294921875, | |
| "learning_rate": 6.173165676349103e-05, | |
| "loss": 0.662, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 6.412213740458015, | |
| "grad_norm": 0.37109375, | |
| "learning_rate": 5.286032631740023e-05, | |
| "loss": 0.6603, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.717557251908397, | |
| "grad_norm": 0.354248046875, | |
| "learning_rate": 4.444297669803981e-05, | |
| "loss": 0.6439, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 7.022900763358779, | |
| "grad_norm": 0.43017578125, | |
| "learning_rate": 3.6560671583635467e-05, | |
| "loss": 0.6696, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 7.32824427480916, | |
| "grad_norm": 0.37109375, | |
| "learning_rate": 2.9289321881345254e-05, | |
| "loss": 0.6603, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.633587786259542, | |
| "grad_norm": 0.4580078125, | |
| "learning_rate": 2.26989546637263e-05, | |
| "loss": 0.6418, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.938931297709924, | |
| "grad_norm": 0.40185546875, | |
| "learning_rate": 1.6853038769745467e-05, | |
| "loss": 0.6404, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 8.244274809160306, | |
| "grad_norm": 0.391357421875, | |
| "learning_rate": 1.1807873565164506e-05, | |
| "loss": 0.6574, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 8.549618320610687, | |
| "grad_norm": 0.383544921875, | |
| "learning_rate": 7.612046748871327e-06, | |
| "loss": 0.6405, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.854961832061068, | |
| "grad_norm": 0.358642578125, | |
| "learning_rate": 4.305966426779118e-06, | |
| "loss": 0.6382, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 9.16030534351145, | |
| "grad_norm": 0.365234375, | |
| "learning_rate": 1.921471959676957e-06, | |
| "loss": 0.6499, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 9.465648854961833, | |
| "grad_norm": 0.40673828125, | |
| "learning_rate": 4.815273327803182e-07, | |
| "loss": 0.6562, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.770992366412214, | |
| "grad_norm": 0.364013671875, | |
| "learning_rate": 0.0, | |
| "loss": 0.6581, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.770992366412214, | |
| "step": 320, | |
| "total_flos": 1.561516427968512e+16, | |
| "train_loss": 0.7572343468666076, | |
| "train_runtime": 291.1608, | |
| "train_samples_per_second": 4.499, | |
| "train_steps_per_second": 1.099 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 320, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.561516427968512e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |