| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.06780810306831667, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013561620613663332, | |
| "grad_norm": 2.4592158794403076, | |
| "learning_rate": 5e-05, | |
| "loss": 1.8825, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0013561620613663332, | |
| "eval_loss": 1.859857439994812, | |
| "eval_runtime": 269.352, | |
| "eval_samples_per_second": 18.444, | |
| "eval_steps_per_second": 4.611, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0027123241227326664, | |
| "grad_norm": 2.3916006088256836, | |
| "learning_rate": 0.0001, | |
| "loss": 1.8278, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0040684861840989996, | |
| "grad_norm": 1.9038305282592773, | |
| "learning_rate": 9.989294616193017e-05, | |
| "loss": 1.7998, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.005424648245465333, | |
| "grad_norm": 1.2749255895614624, | |
| "learning_rate": 9.957224306869053e-05, | |
| "loss": 1.5731, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.006780810306831667, | |
| "grad_norm": 1.2263180017471313, | |
| "learning_rate": 9.903926402016153e-05, | |
| "loss": 1.4569, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.008136972368197999, | |
| "grad_norm": 1.0318810939788818, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 1.3807, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.009493134429564332, | |
| "grad_norm": 0.7358406782150269, | |
| "learning_rate": 9.73465064747553e-05, | |
| "loss": 1.3654, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.010849296490930666, | |
| "grad_norm": 0.7399661540985107, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.3542, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012205458552296999, | |
| "grad_norm": 0.5511712431907654, | |
| "learning_rate": 9.484363707663442e-05, | |
| "loss": 1.2937, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013561620613663334, | |
| "grad_norm": 0.5158834457397461, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.311, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.014917782675029667, | |
| "grad_norm": 0.46730250120162964, | |
| "learning_rate": 9.157348061512727e-05, | |
| "loss": 1.3127, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.016273944736395998, | |
| "grad_norm": 0.467517226934433, | |
| "learning_rate": 8.966766701456177e-05, | |
| "loss": 1.2825, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01763010679776233, | |
| "grad_norm": 0.44724756479263306, | |
| "learning_rate": 8.759199037394887e-05, | |
| "loss": 1.3309, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.01763010679776233, | |
| "eval_loss": 1.2728127241134644, | |
| "eval_runtime": 267.2784, | |
| "eval_samples_per_second": 18.587, | |
| "eval_steps_per_second": 4.647, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.018986268859128665, | |
| "grad_norm": 0.39181774854660034, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 1.2826, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.020342430920494998, | |
| "grad_norm": 0.4049271047115326, | |
| "learning_rate": 8.296729075500344e-05, | |
| "loss": 1.2844, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.02169859298186133, | |
| "grad_norm": 0.40139493346214294, | |
| "learning_rate": 8.043807145043604e-05, | |
| "loss": 1.2874, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.023054755043227664, | |
| "grad_norm": 0.36409640312194824, | |
| "learning_rate": 7.777851165098012e-05, | |
| "loss": 1.3006, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.024410917104593997, | |
| "grad_norm": 0.36486950516700745, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.2763, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02576707916596033, | |
| "grad_norm": 0.39716753363609314, | |
| "learning_rate": 7.211443451095007e-05, | |
| "loss": 1.2574, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.027123241227326667, | |
| "grad_norm": 0.34689292311668396, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 1.2535, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.028479403288693, | |
| "grad_norm": 0.33380135893821716, | |
| "learning_rate": 6.607197326515808e-05, | |
| "loss": 1.2424, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.029835565350059334, | |
| "grad_norm": 0.35058942437171936, | |
| "learning_rate": 6.294095225512603e-05, | |
| "loss": 1.2841, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.031191727411425667, | |
| "grad_norm": 0.3344472646713257, | |
| "learning_rate": 5.9754516100806423e-05, | |
| "loss": 1.2546, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.032547889472791997, | |
| "grad_norm": 0.32480135560035706, | |
| "learning_rate": 5.6526309611002594e-05, | |
| "loss": 1.206, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03390405153415833, | |
| "grad_norm": 0.3209456205368042, | |
| "learning_rate": 5.327015646150716e-05, | |
| "loss": 1.2352, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03526021359552466, | |
| "grad_norm": 0.315898060798645, | |
| "learning_rate": 5e-05, | |
| "loss": 1.2431, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.03526021359552466, | |
| "eval_loss": 1.2289891242980957, | |
| "eval_runtime": 267.3219, | |
| "eval_samples_per_second": 18.584, | |
| "eval_steps_per_second": 4.646, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.036616375656891, | |
| "grad_norm": 0.3312748372554779, | |
| "learning_rate": 4.6729843538492847e-05, | |
| "loss": 1.241, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.03797253771825733, | |
| "grad_norm": 0.3222319781780243, | |
| "learning_rate": 4.347369038899744e-05, | |
| "loss": 1.2303, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.039328699779623666, | |
| "grad_norm": 0.3243453800678253, | |
| "learning_rate": 4.0245483899193595e-05, | |
| "loss": 1.227, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.040684861840989996, | |
| "grad_norm": 0.32615312933921814, | |
| "learning_rate": 3.705904774487396e-05, | |
| "loss": 1.2461, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.04204102390235633, | |
| "grad_norm": 0.3400299847126007, | |
| "learning_rate": 3.392802673484193e-05, | |
| "loss": 1.2265, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04339718596372266, | |
| "grad_norm": 0.33165961503982544, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 1.2531, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.044753348025089, | |
| "grad_norm": 0.30722764134407043, | |
| "learning_rate": 2.7885565489049946e-05, | |
| "loss": 1.2222, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.04610951008645533, | |
| "grad_norm": 0.31792157888412476, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 1.2555, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.047465672147821665, | |
| "grad_norm": 0.308610737323761, | |
| "learning_rate": 2.2221488349019903e-05, | |
| "loss": 1.2574, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.048821834209187995, | |
| "grad_norm": 0.3103911876678467, | |
| "learning_rate": 1.9561928549563968e-05, | |
| "loss": 1.2055, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05017799627055433, | |
| "grad_norm": 0.33152303099632263, | |
| "learning_rate": 1.703270924499656e-05, | |
| "loss": 1.2281, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05153415833192066, | |
| "grad_norm": 0.3187631368637085, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 1.2171, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.052890320393287, | |
| "grad_norm": 0.3220902681350708, | |
| "learning_rate": 1.2408009626051137e-05, | |
| "loss": 1.2111, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.052890320393287, | |
| "eval_loss": 1.2134052515029907, | |
| "eval_runtime": 267.2229, | |
| "eval_samples_per_second": 18.591, | |
| "eval_steps_per_second": 4.648, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.054246482454653334, | |
| "grad_norm": 0.31619793176651, | |
| "learning_rate": 1.0332332985438248e-05, | |
| "loss": 1.239, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.055602644516019664, | |
| "grad_norm": 0.33402693271636963, | |
| "learning_rate": 8.426519384872733e-06, | |
| "loss": 1.2204, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.056958806577386, | |
| "grad_norm": 0.3335304260253906, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 1.1981, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.05831496863875233, | |
| "grad_norm": 0.3046887218952179, | |
| "learning_rate": 5.156362923365588e-06, | |
| "loss": 1.2376, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.05967113070011867, | |
| "grad_norm": 0.3035743534564972, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 1.2419, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.061027292761485, | |
| "grad_norm": 0.3044282793998718, | |
| "learning_rate": 2.653493525244721e-06, | |
| "loss": 1.2429, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.062383454822851334, | |
| "grad_norm": 0.3118952512741089, | |
| "learning_rate": 1.70370868554659e-06, | |
| "loss": 1.2389, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.06373961688421767, | |
| "grad_norm": 0.3790152370929718, | |
| "learning_rate": 9.607359798384785e-07, | |
| "loss": 1.2521, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.06509577894558399, | |
| "grad_norm": 0.3012060523033142, | |
| "learning_rate": 4.277569313094809e-07, | |
| "loss": 1.2164, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06645194100695033, | |
| "grad_norm": 0.3109649419784546, | |
| "learning_rate": 1.0705383806982606e-07, | |
| "loss": 1.2321, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.06780810306831667, | |
| "grad_norm": 0.32313668727874756, | |
| "learning_rate": 0.0, | |
| "loss": 1.2133, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.21695466389504e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |