| { |
| "best_global_step": 229, |
| "best_metric": 0.3871820569038391, |
| "best_model_checkpoint": "lora_qwen7b_python_ab_v1/checkpoint-229", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 687, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.043835616438356165, |
| "grad_norm": 0.38687002658843994, |
| "learning_rate": 3.913043478260869e-06, |
| "loss": 1.0194, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.08767123287671233, |
| "grad_norm": 0.46047139167785645, |
| "learning_rate": 8.260869565217392e-06, |
| "loss": 1.0168, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.13150684931506848, |
| "grad_norm": 0.9523305296897888, |
| "learning_rate": 1.2608695652173912e-05, |
| "loss": 0.931, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.17534246575342466, |
| "grad_norm": 0.5194224119186401, |
| "learning_rate": 1.6956521739130433e-05, |
| "loss": 0.8556, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.2191780821917808, |
| "grad_norm": 0.6766444444656372, |
| "learning_rate": 2.1304347826086958e-05, |
| "loss": 0.6603, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.26301369863013696, |
| "grad_norm": 0.3464139699935913, |
| "learning_rate": 2.565217391304348e-05, |
| "loss": 0.5166, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.30684931506849317, |
| "grad_norm": 0.2768701910972595, |
| "learning_rate": 3e-05, |
| "loss": 0.4596, |
| "step": 70 |
| }, |
| { |
| "epoch": 0.3506849315068493, |
| "grad_norm": 0.1779787391424179, |
| "learning_rate": 2.9514563106796115e-05, |
| "loss": 0.4121, |
| "step": 80 |
| }, |
| { |
| "epoch": 0.39452054794520547, |
| "grad_norm": 0.18716883659362793, |
| "learning_rate": 2.9029126213592237e-05, |
| "loss": 0.3962, |
| "step": 90 |
| }, |
| { |
| "epoch": 0.4383561643835616, |
| "grad_norm": 0.16104164719581604, |
| "learning_rate": 2.854368932038835e-05, |
| "loss": 0.367, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.4821917808219178, |
| "grad_norm": 0.15184655785560608, |
| "learning_rate": 2.8058252427184466e-05, |
| "loss": 0.355, |
| "step": 110 |
| }, |
| { |
| "epoch": 0.5260273972602739, |
| "grad_norm": 0.14750339090824127, |
| "learning_rate": 2.757281553398058e-05, |
| "loss": 0.3417, |
| "step": 120 |
| }, |
| { |
| "epoch": 0.5698630136986301, |
| "grad_norm": 0.1846708357334137, |
| "learning_rate": 2.7087378640776702e-05, |
| "loss": 0.3327, |
| "step": 130 |
| }, |
| { |
| "epoch": 0.6136986301369863, |
| "grad_norm": 0.18890278041362762, |
| "learning_rate": 2.6601941747572816e-05, |
| "loss": 0.3282, |
| "step": 140 |
| }, |
| { |
| "epoch": 0.6575342465753424, |
| "grad_norm": 0.2321624755859375, |
| "learning_rate": 2.611650485436893e-05, |
| "loss": 0.3235, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.7013698630136986, |
| "grad_norm": 0.20538988709449768, |
| "learning_rate": 2.563106796116505e-05, |
| "loss": 0.3125, |
| "step": 160 |
| }, |
| { |
| "epoch": 0.7452054794520548, |
| "grad_norm": 0.2320716381072998, |
| "learning_rate": 2.5145631067961167e-05, |
| "loss": 0.3107, |
| "step": 170 |
| }, |
| { |
| "epoch": 0.7890410958904109, |
| "grad_norm": 0.24439719319343567, |
| "learning_rate": 2.466019417475728e-05, |
| "loss": 0.2904, |
| "step": 180 |
| }, |
| { |
| "epoch": 0.8328767123287671, |
| "grad_norm": 0.24646572768688202, |
| "learning_rate": 2.41747572815534e-05, |
| "loss": 0.2879, |
| "step": 190 |
| }, |
| { |
| "epoch": 0.8767123287671232, |
| "grad_norm": 0.27469298243522644, |
| "learning_rate": 2.3689320388349514e-05, |
| "loss": 0.2917, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.9205479452054794, |
| "grad_norm": 0.3135608434677124, |
| "learning_rate": 2.3203883495145632e-05, |
| "loss": 0.266, |
| "step": 210 |
| }, |
| { |
| "epoch": 0.9643835616438357, |
| "grad_norm": 0.31775224208831787, |
| "learning_rate": 2.2718446601941746e-05, |
| "loss": 0.255, |
| "step": 220 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.3871820569038391, |
| "eval_runtime": 77.3712, |
| "eval_samples_per_second": 15.949, |
| "eval_steps_per_second": 7.975, |
| "step": 229 |
| }, |
| { |
| "epoch": 1.0043835616438357, |
| "grad_norm": 0.3090009391307831, |
| "learning_rate": 2.2233009708737864e-05, |
| "loss": 0.2568, |
| "step": 230 |
| }, |
| { |
| "epoch": 1.0482191780821917, |
| "grad_norm": 0.3228119909763336, |
| "learning_rate": 2.1747572815533982e-05, |
| "loss": 0.2369, |
| "step": 240 |
| }, |
| { |
| "epoch": 1.0920547945205479, |
| "grad_norm": 0.4035201370716095, |
| "learning_rate": 2.1262135922330097e-05, |
| "loss": 0.2175, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.135890410958904, |
| "grad_norm": 0.3466794192790985, |
| "learning_rate": 2.0776699029126215e-05, |
| "loss": 0.2308, |
| "step": 260 |
| }, |
| { |
| "epoch": 1.1797260273972603, |
| "grad_norm": 0.3883700966835022, |
| "learning_rate": 2.029126213592233e-05, |
| "loss": 0.2235, |
| "step": 270 |
| }, |
| { |
| "epoch": 1.2235616438356165, |
| "grad_norm": 0.3835398256778717, |
| "learning_rate": 1.9805825242718447e-05, |
| "loss": 0.2121, |
| "step": 280 |
| }, |
| { |
| "epoch": 1.2673972602739725, |
| "grad_norm": 0.3837865889072418, |
| "learning_rate": 1.9320388349514565e-05, |
| "loss": 0.2124, |
| "step": 290 |
| }, |
| { |
| "epoch": 1.3112328767123287, |
| "grad_norm": 0.3975505828857422, |
| "learning_rate": 1.883495145631068e-05, |
| "loss": 0.1805, |
| "step": 300 |
| }, |
| { |
| "epoch": 1.355068493150685, |
| "grad_norm": 0.5245018005371094, |
| "learning_rate": 1.8349514563106795e-05, |
| "loss": 0.1922, |
| "step": 310 |
| }, |
| { |
| "epoch": 1.398904109589041, |
| "grad_norm": 0.5155487656593323, |
| "learning_rate": 1.7864077669902913e-05, |
| "loss": 0.1862, |
| "step": 320 |
| }, |
| { |
| "epoch": 1.4427397260273973, |
| "grad_norm": 0.47153836488723755, |
| "learning_rate": 1.737864077669903e-05, |
| "loss": 0.1988, |
| "step": 330 |
| }, |
| { |
| "epoch": 1.4865753424657533, |
| "grad_norm": 0.5374466180801392, |
| "learning_rate": 1.6893203883495145e-05, |
| "loss": 0.1808, |
| "step": 340 |
| }, |
| { |
| "epoch": 1.5304109589041097, |
| "grad_norm": 0.6311222910881042, |
| "learning_rate": 1.6407766990291263e-05, |
| "loss": 0.1769, |
| "step": 350 |
| }, |
| { |
| "epoch": 1.5742465753424657, |
| "grad_norm": 0.45171788334846497, |
| "learning_rate": 1.592233009708738e-05, |
| "loss": 0.1851, |
| "step": 360 |
| }, |
| { |
| "epoch": 1.618082191780822, |
| "grad_norm": 0.6755343675613403, |
| "learning_rate": 1.5436893203883496e-05, |
| "loss": 0.1651, |
| "step": 370 |
| }, |
| { |
| "epoch": 1.6619178082191781, |
| "grad_norm": 0.5747941136360168, |
| "learning_rate": 1.4951456310679612e-05, |
| "loss": 0.1752, |
| "step": 380 |
| }, |
| { |
| "epoch": 1.705753424657534, |
| "grad_norm": 0.5249078869819641, |
| "learning_rate": 1.4466019417475728e-05, |
| "loss": 0.163, |
| "step": 390 |
| }, |
| { |
| "epoch": 1.7495890410958905, |
| "grad_norm": 0.6157597303390503, |
| "learning_rate": 1.3980582524271846e-05, |
| "loss": 0.1721, |
| "step": 400 |
| }, |
| { |
| "epoch": 1.7934246575342465, |
| "grad_norm": 0.5300430059432983, |
| "learning_rate": 1.349514563106796e-05, |
| "loss": 0.1611, |
| "step": 410 |
| }, |
| { |
| "epoch": 1.8372602739726027, |
| "grad_norm": 0.6655022501945496, |
| "learning_rate": 1.3009708737864079e-05, |
| "loss": 0.1669, |
| "step": 420 |
| }, |
| { |
| "epoch": 1.881095890410959, |
| "grad_norm": 0.6132180690765381, |
| "learning_rate": 1.2524271844660193e-05, |
| "loss": 0.1687, |
| "step": 430 |
| }, |
| { |
| "epoch": 1.924931506849315, |
| "grad_norm": 0.47309410572052, |
| "learning_rate": 1.2038834951456311e-05, |
| "loss": 0.1583, |
| "step": 440 |
| }, |
| { |
| "epoch": 1.9687671232876713, |
| "grad_norm": 0.647148072719574, |
| "learning_rate": 1.1553398058252427e-05, |
| "loss": 0.161, |
| "step": 450 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.421628475189209, |
| "eval_runtime": 49.8056, |
| "eval_samples_per_second": 24.776, |
| "eval_steps_per_second": 12.388, |
| "step": 458 |
| }, |
| { |
| "epoch": 2.0087671232876714, |
| "grad_norm": 0.5745662450790405, |
| "learning_rate": 1.1067961165048544e-05, |
| "loss": 0.1393, |
| "step": 460 |
| }, |
| { |
| "epoch": 2.0526027397260274, |
| "grad_norm": 0.3223569393157959, |
| "learning_rate": 1.058252427184466e-05, |
| "loss": 0.148, |
| "step": 470 |
| }, |
| { |
| "epoch": 2.0964383561643833, |
| "grad_norm": 0.48584961891174316, |
| "learning_rate": 1.0097087378640776e-05, |
| "loss": 0.1291, |
| "step": 480 |
| }, |
| { |
| "epoch": 2.1402739726027398, |
| "grad_norm": 0.48411062359809875, |
| "learning_rate": 9.611650485436894e-06, |
| "loss": 0.1398, |
| "step": 490 |
| }, |
| { |
| "epoch": 2.1841095890410958, |
| "grad_norm": 0.4757746458053589, |
| "learning_rate": 9.12621359223301e-06, |
| "loss": 0.1497, |
| "step": 500 |
| }, |
| { |
| "epoch": 2.227945205479452, |
| "grad_norm": 0.5401110649108887, |
| "learning_rate": 8.640776699029127e-06, |
| "loss": 0.1413, |
| "step": 510 |
| }, |
| { |
| "epoch": 2.271780821917808, |
| "grad_norm": 0.584707498550415, |
| "learning_rate": 8.155339805825243e-06, |
| "loss": 0.1442, |
| "step": 520 |
| }, |
| { |
| "epoch": 2.315616438356164, |
| "grad_norm": 0.3924754858016968, |
| "learning_rate": 7.66990291262136e-06, |
| "loss": 0.1336, |
| "step": 530 |
| }, |
| { |
| "epoch": 2.3594520547945206, |
| "grad_norm": 0.7305352091789246, |
| "learning_rate": 7.1844660194174755e-06, |
| "loss": 0.1313, |
| "step": 540 |
| }, |
| { |
| "epoch": 2.4032876712328766, |
| "grad_norm": 0.40309375524520874, |
| "learning_rate": 6.699029126213593e-06, |
| "loss": 0.1253, |
| "step": 550 |
| }, |
| { |
| "epoch": 2.447123287671233, |
| "grad_norm": 0.4579068720340729, |
| "learning_rate": 6.213592233009709e-06, |
| "loss": 0.1268, |
| "step": 560 |
| }, |
| { |
| "epoch": 2.490958904109589, |
| "grad_norm": 0.3836471140384674, |
| "learning_rate": 5.728155339805826e-06, |
| "loss": 0.1306, |
| "step": 570 |
| }, |
| { |
| "epoch": 2.534794520547945, |
| "grad_norm": 0.4680565297603607, |
| "learning_rate": 5.242718446601942e-06, |
| "loss": 0.1444, |
| "step": 580 |
| }, |
| { |
| "epoch": 2.5786301369863014, |
| "grad_norm": 0.5456928610801697, |
| "learning_rate": 4.7572815533980585e-06, |
| "loss": 0.1554, |
| "step": 590 |
| }, |
| { |
| "epoch": 2.6224657534246574, |
| "grad_norm": 0.4823114275932312, |
| "learning_rate": 4.271844660194175e-06, |
| "loss": 0.1371, |
| "step": 600 |
| }, |
| { |
| "epoch": 2.666301369863014, |
| "grad_norm": 0.6588716506958008, |
| "learning_rate": 3.7864077669902915e-06, |
| "loss": 0.1251, |
| "step": 610 |
| }, |
| { |
| "epoch": 2.71013698630137, |
| "grad_norm": 0.39495226740837097, |
| "learning_rate": 3.3009708737864078e-06, |
| "loss": 0.1368, |
| "step": 620 |
| }, |
| { |
| "epoch": 2.7539726027397258, |
| "grad_norm": 0.6559995412826538, |
| "learning_rate": 2.815533980582524e-06, |
| "loss": 0.1292, |
| "step": 630 |
| }, |
| { |
| "epoch": 2.797808219178082, |
| "grad_norm": 0.4756312668323517, |
| "learning_rate": 2.3300970873786407e-06, |
| "loss": 0.1252, |
| "step": 640 |
| }, |
| { |
| "epoch": 2.8416438356164386, |
| "grad_norm": 0.5348303318023682, |
| "learning_rate": 1.8446601941747572e-06, |
| "loss": 0.1165, |
| "step": 650 |
| }, |
| { |
| "epoch": 2.8854794520547946, |
| "grad_norm": 0.2880028188228607, |
| "learning_rate": 1.359223300970874e-06, |
| "loss": 0.1412, |
| "step": 660 |
| }, |
| { |
| "epoch": 2.9293150684931506, |
| "grad_norm": 0.5411520004272461, |
| "learning_rate": 8.737864077669904e-07, |
| "loss": 0.1376, |
| "step": 670 |
| }, |
| { |
| "epoch": 2.9731506849315066, |
| "grad_norm": 0.4965899586677551, |
| "learning_rate": 3.883495145631068e-07, |
| "loss": 0.1335, |
| "step": 680 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.43176329135894775, |
| "eval_runtime": 144.0381, |
| "eval_samples_per_second": 8.567, |
| "eval_steps_per_second": 4.284, |
| "step": 687 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 687, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4.290963516451369e+17, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
