| { | |
| "best_metric": 0.11746824532747269, | |
| "best_model_checkpoint": "./sft-qwen25/checkpoint-40", | |
| "epoch": 0.49382716049382713, | |
| "eval_steps": 10, | |
| "global_step": 40, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.012345679012345678, | |
| "grad_norm": 107.62647247314453, | |
| "learning_rate": 5e-06, | |
| "loss": 2.9284, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "grad_norm": 101.32521057128906, | |
| "learning_rate": 1e-05, | |
| "loss": 2.7586, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.037037037037037035, | |
| "grad_norm": 74.68183135986328, | |
| "learning_rate": 9.99604698613651e-06, | |
| "loss": 1.9898, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04938271604938271, | |
| "grad_norm": 60.77379608154297, | |
| "learning_rate": 9.98419419507348e-06, | |
| "loss": 1.1595, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.06172839506172839, | |
| "grad_norm": 27.641386032104492, | |
| "learning_rate": 9.964460368509868e-06, | |
| "loss": 0.2989, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 7.500521659851074, | |
| "learning_rate": 9.936876709681668e-06, | |
| "loss": 0.1825, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.08641975308641975, | |
| "grad_norm": 17.12440299987793, | |
| "learning_rate": 9.901486834023182e-06, | |
| "loss": 0.2373, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.09876543209876543, | |
| "grad_norm": 9.882857322692871, | |
| "learning_rate": 9.85834670020205e-06, | |
| "loss": 0.1815, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 6.776613235473633, | |
| "learning_rate": 9.807524521637103e-06, | |
| "loss": 0.1521, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.12345679012345678, | |
| "grad_norm": 2.03835391998291, | |
| "learning_rate": 9.749100658638914e-06, | |
| "loss": 0.1381, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.12345679012345678, | |
| "eval_loss": 0.13423670828342438, | |
| "eval_runtime": 13.0005, | |
| "eval_samples_per_second": 78.305, | |
| "eval_steps_per_second": 19.615, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13580246913580246, | |
| "grad_norm": 5.280986309051514, | |
| "learning_rate": 9.68316749134364e-06, | |
| "loss": 0.1284, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.14814814814814814, | |
| "grad_norm": 3.772414445877075, | |
| "learning_rate": 9.609829273641034e-06, | |
| "loss": 0.1387, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.16049382716049382, | |
| "grad_norm": 5.155209541320801, | |
| "learning_rate": 9.529201968327618e-06, | |
| "loss": 0.1207, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1728395061728395, | |
| "grad_norm": 5.389438629150391, | |
| "learning_rate": 9.44141306374566e-06, | |
| "loss": 0.1437, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.18518518518518517, | |
| "grad_norm": 3.371304750442505, | |
| "learning_rate": 9.346601372197914e-06, | |
| "loss": 0.1422, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.19753086419753085, | |
| "grad_norm": 4.07261323928833, | |
| "learning_rate": 9.244916810456822e-06, | |
| "loss": 0.134, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.20987654320987653, | |
| "grad_norm": 2.3795979022979736, | |
| "learning_rate": 9.136520162715288e-06, | |
| "loss": 0.121, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 2.3384926319122314, | |
| "learning_rate": 9.021582826353825e-06, | |
| "loss": 0.121, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2345679012345679, | |
| "grad_norm": 4.116921901702881, | |
| "learning_rate": 8.900286540926062e-06, | |
| "loss": 0.1275, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.24691358024691357, | |
| "grad_norm": 3.9383344650268555, | |
| "learning_rate": 8.772823100791152e-06, | |
| "loss": 0.1495, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.24691358024691357, | |
| "eval_loss": 0.12288570404052734, | |
| "eval_runtime": 13.029, | |
| "eval_samples_per_second": 78.133, | |
| "eval_steps_per_second": 19.572, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.25925925925925924, | |
| "grad_norm": 2.939857244491577, | |
| "learning_rate": 8.639394051847472e-06, | |
| "loss": 0.1113, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.2716049382716049, | |
| "grad_norm": 4.183154582977295, | |
| "learning_rate": 8.500210372847128e-06, | |
| "loss": 0.1299, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2839506172839506, | |
| "grad_norm": 2.1485512256622314, | |
| "learning_rate": 8.355492141795185e-06, | |
| "loss": 0.1256, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.2962962962962963, | |
| "grad_norm": 3.5548007488250732, | |
| "learning_rate": 8.2054681879611e-06, | |
| "loss": 0.1219, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.30864197530864196, | |
| "grad_norm": 2.197593927383423, | |
| "learning_rate": 8.050375730052622e-06, | |
| "loss": 0.0967, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.32098765432098764, | |
| "grad_norm": 4.257010459899902, | |
| "learning_rate": 7.890460001124242e-06, | |
| "loss": 0.1451, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 2.651317834854126, | |
| "learning_rate": 7.725973860813338e-06, | |
| "loss": 0.1102, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.345679012345679, | |
| "grad_norm": 3.145437717437744, | |
| "learning_rate": 7.5571773955171124e-06, | |
| "loss": 0.1251, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.35802469135802467, | |
| "grad_norm": 3.316354751586914, | |
| "learning_rate": 7.3843375071425315e-06, | |
| "loss": 0.1132, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "grad_norm": 3.7446980476379395, | |
| "learning_rate": 7.2077274910795605e-06, | |
| "loss": 0.1215, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.37037037037037035, | |
| "eval_loss": 0.12460627406835556, | |
| "eval_runtime": 13.0813, | |
| "eval_samples_per_second": 77.821, | |
| "eval_steps_per_second": 19.493, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.38271604938271603, | |
| "grad_norm": 4.762835502624512, | |
| "learning_rate": 7.02762660406497e-06, | |
| "loss": 0.1338, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.3950617283950617, | |
| "grad_norm": 4.5091328620910645, | |
| "learning_rate": 6.844319622620039e-06, | |
| "loss": 0.1314, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.4074074074074074, | |
| "grad_norm": 3.881032705307007, | |
| "learning_rate": 6.65809639276034e-06, | |
| "loss": 0.13, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.41975308641975306, | |
| "grad_norm": 1.9894459247589111, | |
| "learning_rate": 6.469251371689606e-06, | |
| "loss": 0.1087, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.43209876543209874, | |
| "grad_norm": 2.829587459564209, | |
| "learning_rate": 6.278083162202374e-06, | |
| "loss": 0.1249, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 2.045034885406494, | |
| "learning_rate": 6.084894040531591e-06, | |
| "loss": 0.116, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.4567901234567901, | |
| "grad_norm": 3.9003093242645264, | |
| "learning_rate": 5.8899894783877536e-06, | |
| "loss": 0.1235, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.4691358024691358, | |
| "grad_norm": 3.978607177734375, | |
| "learning_rate": 5.693677659945343e-06, | |
| "loss": 0.1309, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.48148148148148145, | |
| "grad_norm": 1.7156412601470947, | |
| "learning_rate": 5.496268994540309e-06, | |
| "loss": 0.1113, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "grad_norm": 2.837052345275879, | |
| "learning_rate": 5.2980756258491e-06, | |
| "loss": 0.1354, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.49382716049382713, | |
| "eval_loss": 0.11746824532747269, | |
| "eval_runtime": 13.1912, | |
| "eval_samples_per_second": 77.172, | |
| "eval_steps_per_second": 19.331, | |
| "step": 40 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 81, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.560561014145024e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |