| { | |
| "best_metric": 1.4920570850372314, | |
| "best_model_checkpoint": "./models_trained/sft_full_df_2/checkpoint-6696", | |
| "epoch": 2.9989922741014445, | |
| "eval_steps": 2232, | |
| "global_step": 6696, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08017019370731161, | |
| "grad_norm": 9.433825492858887, | |
| "learning_rate": 1.0022396416573349e-06, | |
| "loss": 2.5707, | |
| "step": 179 | |
| }, | |
| { | |
| "epoch": 0.16034038741462323, | |
| "grad_norm": 5.339526176452637, | |
| "learning_rate": 2.0044792833146697e-06, | |
| "loss": 1.9288, | |
| "step": 358 | |
| }, | |
| { | |
| "epoch": 0.24051058112193482, | |
| "grad_norm": 6.231087684631348, | |
| "learning_rate": 3.0067189249720046e-06, | |
| "loss": 1.6422, | |
| "step": 537 | |
| }, | |
| { | |
| "epoch": 0.32068077482924645, | |
| "grad_norm": 6.053775787353516, | |
| "learning_rate": 4.0089585666293395e-06, | |
| "loss": 1.6095, | |
| "step": 716 | |
| }, | |
| { | |
| "epoch": 0.40085096853655805, | |
| "grad_norm": 6.309902667999268, | |
| "learning_rate": 4.999999235639976e-06, | |
| "loss": 1.5876, | |
| "step": 895 | |
| }, | |
| { | |
| "epoch": 0.48102116224386965, | |
| "grad_norm": 6.86934757232666, | |
| "learning_rate": 4.993742312315323e-06, | |
| "loss": 1.5839, | |
| "step": 1074 | |
| }, | |
| { | |
| "epoch": 0.5611913559511813, | |
| "grad_norm": 5.88812255859375, | |
| "learning_rate": 4.975275594860625e-06, | |
| "loss": 1.5634, | |
| "step": 1253 | |
| }, | |
| { | |
| "epoch": 0.6413615496584929, | |
| "grad_norm": 6.350925445556641, | |
| "learning_rate": 4.944689499521886e-06, | |
| "loss": 1.5406, | |
| "step": 1432 | |
| }, | |
| { | |
| "epoch": 0.7215317433658045, | |
| "grad_norm": 6.42539119720459, | |
| "learning_rate": 4.902133781117591e-06, | |
| "loss": 1.5544, | |
| "step": 1611 | |
| }, | |
| { | |
| "epoch": 0.8017019370731161, | |
| "grad_norm": 6.217959880828857, | |
| "learning_rate": 4.847816799813184e-06, | |
| "loss": 1.5301, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 0.8818721307804277, | |
| "grad_norm": 6.820368766784668, | |
| "learning_rate": 4.782004500953626e-06, | |
| "loss": 1.5179, | |
| "step": 1969 | |
| }, | |
| { | |
| "epoch": 0.9620423244877393, | |
| "grad_norm": 7.019912242889404, | |
| "learning_rate": 4.705019112948941e-06, | |
| "loss": 1.5084, | |
| "step": 2148 | |
| }, | |
| { | |
| "epoch": 0.9996640913671482, | |
| "eval_loss": 1.5233653783798218, | |
| "eval_runtime": 72.6117, | |
| "eval_samples_per_second": 27.337, | |
| "eval_steps_per_second": 13.675, | |
| "step": 2232 | |
| }, | |
| { | |
| "epoch": 1.042212518195051, | |
| "grad_norm": 6.613691806793213, | |
| "learning_rate": 4.617237569588121e-06, | |
| "loss": 1.5067, | |
| "step": 2327 | |
| }, | |
| { | |
| "epoch": 1.1223827119023626, | |
| "grad_norm": 7.442265510559082, | |
| "learning_rate": 4.519089664506044e-06, | |
| "loss": 1.4983, | |
| "step": 2506 | |
| }, | |
| { | |
| "epoch": 1.2025529056096742, | |
| "grad_norm": 9.082566261291504, | |
| "learning_rate": 4.411055946839413e-06, | |
| "loss": 1.4935, | |
| "step": 2685 | |
| }, | |
| { | |
| "epoch": 1.2827230993169858, | |
| "grad_norm": 7.871016979217529, | |
| "learning_rate": 4.293665368374987e-06, | |
| "loss": 1.4923, | |
| "step": 2864 | |
| }, | |
| { | |
| "epoch": 1.3628932930242974, | |
| "grad_norm": 7.989074230194092, | |
| "learning_rate": 4.167492693710046e-06, | |
| "loss": 1.4794, | |
| "step": 3043 | |
| }, | |
| { | |
| "epoch": 1.443063486731609, | |
| "grad_norm": 7.531464576721191, | |
| "learning_rate": 4.033155686105407e-06, | |
| "loss": 1.4967, | |
| "step": 3222 | |
| }, | |
| { | |
| "epoch": 1.5232336804389206, | |
| "grad_norm": 7.842589378356934, | |
| "learning_rate": 3.8913120828095415e-06, | |
| "loss": 1.4583, | |
| "step": 3401 | |
| }, | |
| { | |
| "epoch": 1.6034038741462322, | |
| "grad_norm": 8.500748634338379, | |
| "learning_rate": 3.7426563746631257e-06, | |
| "loss": 1.4787, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 1.6835740678535438, | |
| "grad_norm": 9.060755729675293, | |
| "learning_rate": 3.587916405751636e-06, | |
| "loss": 1.4766, | |
| "step": 3759 | |
| }, | |
| { | |
| "epoch": 1.7637442615608554, | |
| "grad_norm": 8.827229499816895, | |
| "learning_rate": 3.4278498097546904e-06, | |
| "loss": 1.472, | |
| "step": 3938 | |
| }, | |
| { | |
| "epoch": 1.843914455268167, | |
| "grad_norm": 8.073822975158691, | |
| "learning_rate": 3.2632403004403746e-06, | |
| "loss": 1.4988, | |
| "step": 4117 | |
| }, | |
| { | |
| "epoch": 1.9240846489754788, | |
| "grad_norm": 9.162446022033691, | |
| "learning_rate": 3.0948938344669414e-06, | |
| "loss": 1.4583, | |
| "step": 4296 | |
| }, | |
| { | |
| "epoch": 1.9993281827342964, | |
| "eval_loss": 1.496500849723816, | |
| "eval_runtime": 72.6334, | |
| "eval_samples_per_second": 27.329, | |
| "eval_steps_per_second": 13.671, | |
| "step": 4464 | |
| }, | |
| { | |
| "epoch": 2.00425484268279, | |
| "grad_norm": 9.242565155029297, | |
| "learning_rate": 2.9236346652794664e-06, | |
| "loss": 1.4732, | |
| "step": 4475 | |
| }, | |
| { | |
| "epoch": 2.084425036390102, | |
| "grad_norm": 9.86392593383789, | |
| "learning_rate": 2.750301307422268e-06, | |
| "loss": 1.4263, | |
| "step": 4654 | |
| }, | |
| { | |
| "epoch": 2.1645952300974134, | |
| "grad_norm": 10.472150802612305, | |
| "learning_rate": 2.575742431026521e-06, | |
| "loss": 1.4263, | |
| "step": 4833 | |
| }, | |
| { | |
| "epoch": 2.244765423804725, | |
| "grad_norm": 11.642035484313965, | |
| "learning_rate": 2.40081270657435e-06, | |
| "loss": 1.4318, | |
| "step": 5012 | |
| }, | |
| { | |
| "epoch": 2.3249356175120366, | |
| "grad_norm": 11.944063186645508, | |
| "learning_rate": 2.226368620284175e-06, | |
| "loss": 1.4432, | |
| "step": 5191 | |
| }, | |
| { | |
| "epoch": 2.4051058112193484, | |
| "grad_norm": 10.567192077636719, | |
| "learning_rate": 2.0532642806058894e-06, | |
| "loss": 1.4353, | |
| "step": 5370 | |
| }, | |
| { | |
| "epoch": 2.48527600492666, | |
| "grad_norm": 13.025165557861328, | |
| "learning_rate": 1.8823472363580105e-06, | |
| "loss": 1.4327, | |
| "step": 5549 | |
| }, | |
| { | |
| "epoch": 2.5654461986339716, | |
| "grad_norm": 9.754578590393066, | |
| "learning_rate": 1.714454326981919e-06, | |
| "loss": 1.4214, | |
| "step": 5728 | |
| }, | |
| { | |
| "epoch": 2.6456163923412834, | |
| "grad_norm": 9.988022804260254, | |
| "learning_rate": 1.5504075852310582e-06, | |
| "loss": 1.4299, | |
| "step": 5907 | |
| }, | |
| { | |
| "epoch": 2.725786586048595, | |
| "grad_norm": 9.87160587310791, | |
| "learning_rate": 1.3910102123562535e-06, | |
| "loss": 1.4144, | |
| "step": 6086 | |
| }, | |
| { | |
| "epoch": 2.805956779755906, | |
| "grad_norm": 9.950671195983887, | |
| "learning_rate": 1.2370426454933122e-06, | |
| "loss": 1.4077, | |
| "step": 6265 | |
| }, | |
| { | |
| "epoch": 2.886126973463218, | |
| "grad_norm": 11.326949119567871, | |
| "learning_rate": 1.0892587365076916e-06, | |
| "loss": 1.4311, | |
| "step": 6444 | |
| }, | |
| { | |
| "epoch": 2.96629716717053, | |
| "grad_norm": 11.63015365600586, | |
| "learning_rate": 9.483820610052311e-07, | |
| "loss": 1.4271, | |
| "step": 6623 | |
| }, | |
| { | |
| "epoch": 2.9989922741014445, | |
| "eval_loss": 1.4920570850372314, | |
| "eval_runtime": 72.6497, | |
| "eval_samples_per_second": 27.323, | |
| "eval_steps_per_second": 13.668, | |
| "step": 6696 | |
| } | |
| ], | |
| "logging_steps": 179, | |
| "max_steps": 8928, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 2232, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.021394676596736e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |