| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.4645760743321719, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004645760743321719, | |
| "eval_loss": 0.6413776278495789, | |
| "eval_runtime": 26.814, | |
| "eval_samples_per_second": 13.538, | |
| "eval_steps_per_second": 1.716, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.013937282229965157, | |
| "grad_norm": 0.7640010118484497, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6593, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.027874564459930314, | |
| "grad_norm": 0.6851508617401123, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6301, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.041811846689895474, | |
| "grad_norm": 0.46766743063926697, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5405, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.041811846689895474, | |
| "eval_loss": 0.4691576659679413, | |
| "eval_runtime": 27.1079, | |
| "eval_samples_per_second": 13.391, | |
| "eval_steps_per_second": 1.697, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.05574912891986063, | |
| "grad_norm": 0.4390910267829895, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.4779, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06968641114982578, | |
| "grad_norm": 0.41035574674606323, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.427, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08362369337979095, | |
| "grad_norm": 0.2806294560432434, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.4068, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.08362369337979095, | |
| "eval_loss": 0.3835928440093994, | |
| "eval_runtime": 27.168, | |
| "eval_samples_per_second": 13.361, | |
| "eval_steps_per_second": 1.693, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0975609756097561, | |
| "grad_norm": 0.22390054166316986, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.3653, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11149825783972125, | |
| "grad_norm": 0.21830645203590393, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.3572, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1254355400696864, | |
| "grad_norm": 0.2081187516450882, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.3729, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.1254355400696864, | |
| "eval_loss": 0.358090877532959, | |
| "eval_runtime": 27.1856, | |
| "eval_samples_per_second": 13.353, | |
| "eval_steps_per_second": 1.692, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.13937282229965156, | |
| "grad_norm": 0.21237684786319733, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.3412, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.15331010452961671, | |
| "grad_norm": 0.2204415649175644, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.3651, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.1672473867595819, | |
| "grad_norm": 0.2142229676246643, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.367, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.1672473867595819, | |
| "eval_loss": 0.34969642758369446, | |
| "eval_runtime": 27.2204, | |
| "eval_samples_per_second": 13.336, | |
| "eval_steps_per_second": 1.69, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.18118466898954705, | |
| "grad_norm": 0.16911321878433228, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.3524, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.1951219512195122, | |
| "grad_norm": 0.20093809068202972, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.3611, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.20905923344947736, | |
| "grad_norm": 0.19687530398368835, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.3725, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.20905923344947736, | |
| "eval_loss": 0.34502366185188293, | |
| "eval_runtime": 27.193, | |
| "eval_samples_per_second": 13.349, | |
| "eval_steps_per_second": 1.692, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.2229965156794425, | |
| "grad_norm": 0.19324517250061035, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.341, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.23693379790940766, | |
| "grad_norm": 0.182427778840065, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.3379, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.2508710801393728, | |
| "grad_norm": 0.20922890305519104, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.3596, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.2508710801393728, | |
| "eval_loss": 0.3411136269569397, | |
| "eval_runtime": 27.223, | |
| "eval_samples_per_second": 13.334, | |
| "eval_steps_per_second": 1.69, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.26480836236933797, | |
| "grad_norm": 0.19875657558441162, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.3421, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.2787456445993031, | |
| "grad_norm": 0.18166325986385345, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.3452, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.2926829268292683, | |
| "grad_norm": 0.19043953716754913, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.3667, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.2926829268292683, | |
| "eval_loss": 0.3384956419467926, | |
| "eval_runtime": 27.1995, | |
| "eval_samples_per_second": 13.346, | |
| "eval_steps_per_second": 1.691, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.30662020905923343, | |
| "grad_norm": 0.17853577435016632, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.3372, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.3205574912891986, | |
| "grad_norm": 0.1952916830778122, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.3604, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.3344947735191638, | |
| "grad_norm": 0.17483319342136383, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.3153, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.3344947735191638, | |
| "eval_loss": 0.33674171566963196, | |
| "eval_runtime": 27.2031, | |
| "eval_samples_per_second": 13.344, | |
| "eval_steps_per_second": 1.691, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.34843205574912894, | |
| "grad_norm": 0.19136692583560944, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.3436, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.3623693379790941, | |
| "grad_norm": 0.1997017115354538, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.3509, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.37630662020905925, | |
| "grad_norm": 0.22049453854560852, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.374, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.37630662020905925, | |
| "eval_loss": 0.3358399271965027, | |
| "eval_runtime": 27.1971, | |
| "eval_samples_per_second": 13.347, | |
| "eval_steps_per_second": 1.691, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.3902439024390244, | |
| "grad_norm": 0.196411594748497, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.3244, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.40418118466898956, | |
| "grad_norm": 0.18862295150756836, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.3651, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.4181184668989547, | |
| "grad_norm": 0.20165245234966278, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.3375, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.4181184668989547, | |
| "eval_loss": 0.3353938162326813, | |
| "eval_runtime": 27.1964, | |
| "eval_samples_per_second": 13.347, | |
| "eval_steps_per_second": 1.691, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.43205574912891986, | |
| "grad_norm": 0.19487881660461426, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.3416, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.445993031358885, | |
| "grad_norm": 0.17371226847171783, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.3358, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.45993031358885017, | |
| "grad_norm": 0.1921817660331726, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.3434, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.45993031358885017, | |
| "eval_loss": 0.33536291122436523, | |
| "eval_runtime": 27.1965, | |
| "eval_samples_per_second": 13.347, | |
| "eval_steps_per_second": 1.691, | |
| "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": 1.5542538590591386e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |