| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5102040816326531, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00510204081632653, | |
| "eval_loss": 6.939616680145264, | |
| "eval_runtime": 1.0293, | |
| "eval_samples_per_second": 160.296, | |
| "eval_steps_per_second": 40.803, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.015306122448979591, | |
| "grad_norm": 0.4716786742210388, | |
| "learning_rate": 3e-05, | |
| "loss": 6.9394, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.030612244897959183, | |
| "grad_norm": 0.49812349677085876, | |
| "learning_rate": 6e-05, | |
| "loss": 6.9356, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.04591836734693878, | |
| "grad_norm": 0.5289084911346436, | |
| "learning_rate": 9e-05, | |
| "loss": 6.933, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.04591836734693878, | |
| "eval_loss": 6.927786350250244, | |
| "eval_runtime": 0.2695, | |
| "eval_samples_per_second": 612.282, | |
| "eval_steps_per_second": 155.854, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.061224489795918366, | |
| "grad_norm": 0.4804115891456604, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 6.9266, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.07653061224489796, | |
| "grad_norm": 0.5177415013313293, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 6.9163, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.09183673469387756, | |
| "grad_norm": 0.5697883367538452, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 6.9044, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09183673469387756, | |
| "eval_loss": 6.897478103637695, | |
| "eval_runtime": 0.271, | |
| "eval_samples_per_second": 608.897, | |
| "eval_steps_per_second": 154.992, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.10714285714285714, | |
| "grad_norm": 0.5388941168785095, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 6.8956, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.12244897959183673, | |
| "grad_norm": 0.5204723477363586, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 6.885, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1377551020408163, | |
| "grad_norm": 0.4605754613876343, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 6.8733, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1377551020408163, | |
| "eval_loss": 6.866656303405762, | |
| "eval_runtime": 0.2715, | |
| "eval_samples_per_second": 607.839, | |
| "eval_steps_per_second": 154.723, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.15306122448979592, | |
| "grad_norm": 0.4325270652770996, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 6.8621, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.1683673469387755, | |
| "grad_norm": 0.3710213303565979, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 6.8547, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1836734693877551, | |
| "grad_norm": 0.321429967880249, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 6.8474, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1836734693877551, | |
| "eval_loss": 6.846508979797363, | |
| "eval_runtime": 0.2698, | |
| "eval_samples_per_second": 611.451, | |
| "eval_steps_per_second": 155.642, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1989795918367347, | |
| "grad_norm": 0.28444114327430725, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 6.8474, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.21428571428571427, | |
| "grad_norm": 0.28362318873405457, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 6.8428, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.22959183673469388, | |
| "grad_norm": 0.23609378933906555, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 6.8412, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.22959183673469388, | |
| "eval_loss": 6.8371405601501465, | |
| "eval_runtime": 0.2731, | |
| "eval_samples_per_second": 604.065, | |
| "eval_steps_per_second": 153.762, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.24489795918367346, | |
| "grad_norm": 0.25101956725120544, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 6.8369, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.2602040816326531, | |
| "grad_norm": 0.21246355772018433, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 6.8376, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2755102040816326, | |
| "grad_norm": 0.20616650581359863, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 6.8328, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2755102040816326, | |
| "eval_loss": 6.832831859588623, | |
| "eval_runtime": 0.2703, | |
| "eval_samples_per_second": 610.543, | |
| "eval_steps_per_second": 155.411, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.29081632653061223, | |
| "grad_norm": 0.22721412777900696, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 6.8316, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.30612244897959184, | |
| "grad_norm": 0.2554455101490021, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 6.8269, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.32142857142857145, | |
| "grad_norm": 0.1824101209640503, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 6.8329, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.32142857142857145, | |
| "eval_loss": 6.830348014831543, | |
| "eval_runtime": 0.2768, | |
| "eval_samples_per_second": 596.042, | |
| "eval_steps_per_second": 151.72, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.336734693877551, | |
| "grad_norm": 0.21916939318180084, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 6.829, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3520408163265306, | |
| "grad_norm": 0.23701028525829315, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 6.8313, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3673469387755102, | |
| "grad_norm": 0.18295414745807648, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 6.8278, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3673469387755102, | |
| "eval_loss": 6.82876443862915, | |
| "eval_runtime": 0.2784, | |
| "eval_samples_per_second": 592.608, | |
| "eval_steps_per_second": 150.846, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3826530612244898, | |
| "grad_norm": 0.20278629660606384, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 6.8285, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3979591836734694, | |
| "grad_norm": 0.19427503645420074, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 6.8339, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.413265306122449, | |
| "grad_norm": 0.2242947667837143, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 6.8312, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.413265306122449, | |
| "eval_loss": 6.827848434448242, | |
| "eval_runtime": 0.2684, | |
| "eval_samples_per_second": 614.721, | |
| "eval_steps_per_second": 156.474, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.42857142857142855, | |
| "grad_norm": 0.19890524446964264, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 6.8276, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.44387755102040816, | |
| "grad_norm": 0.23790700733661652, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 6.8308, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.45918367346938777, | |
| "grad_norm": 0.22041282057762146, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 6.8293, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.45918367346938777, | |
| "eval_loss": 6.827453136444092, | |
| "eval_runtime": 0.2686, | |
| "eval_samples_per_second": 614.27, | |
| "eval_steps_per_second": 156.36, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4744897959183674, | |
| "grad_norm": 0.20826943218708038, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 6.8322, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.4897959183673469, | |
| "grad_norm": 0.20993828773498535, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 6.8318, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.5051020408163265, | |
| "grad_norm": 0.21039879322052002, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 6.8296, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.5051020408163265, | |
| "eval_loss": 6.827375411987305, | |
| "eval_runtime": 0.2813, | |
| "eval_samples_per_second": 586.647, | |
| "eval_steps_per_second": 149.328, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 961128038400.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |