| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.506649778340722, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01013299556681444, | |
| "grad_norm": 2.299077033996582, | |
| "learning_rate": 1e-05, | |
| "loss": 3.7257, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01013299556681444, | |
| "eval_loss": 3.523265838623047, | |
| "eval_runtime": 98.4632, | |
| "eval_samples_per_second": 1.696, | |
| "eval_steps_per_second": 0.853, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02026599113362888, | |
| "grad_norm": 2.1457676887512207, | |
| "learning_rate": 2e-05, | |
| "loss": 3.3609, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.030398986700443317, | |
| "grad_norm": 2.0729289054870605, | |
| "learning_rate": 3e-05, | |
| "loss": 3.2415, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.04053198226725776, | |
| "grad_norm": 2.152341365814209, | |
| "learning_rate": 4e-05, | |
| "loss": 3.0918, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0506649778340722, | |
| "grad_norm": 2.4330685138702393, | |
| "learning_rate": 5e-05, | |
| "loss": 3.1871, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.060797973400886635, | |
| "grad_norm": 2.830127000808716, | |
| "learning_rate": 6e-05, | |
| "loss": 3.007, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.07093096896770108, | |
| "grad_norm": 2.7601826190948486, | |
| "learning_rate": 7e-05, | |
| "loss": 2.7736, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.08106396453451552, | |
| "grad_norm": 3.2441565990448, | |
| "learning_rate": 8e-05, | |
| "loss": 2.668, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.09119696010132995, | |
| "grad_norm": 3.553675651550293, | |
| "learning_rate": 9e-05, | |
| "loss": 2.3073, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1013299556681444, | |
| "grad_norm": 3.3283772468566895, | |
| "learning_rate": 0.0001, | |
| "loss": 2.017, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11146295123495883, | |
| "grad_norm": 3.136801242828369, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 1.7767, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.12159594680177327, | |
| "grad_norm": 2.8050150871276855, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 1.563, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1317289423685877, | |
| "grad_norm": 2.3837151527404785, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 1.1287, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.1317289423685877, | |
| "eval_loss": 1.2567861080169678, | |
| "eval_runtime": 3.9408, | |
| "eval_samples_per_second": 42.377, | |
| "eval_steps_per_second": 21.315, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.14186193793540217, | |
| "grad_norm": 2.615652084350586, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.1747, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.1519949335022166, | |
| "grad_norm": 3.228269338607788, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 1.1195, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.16212792906903103, | |
| "grad_norm": 3.2835893630981445, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.7324, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.17226092463584547, | |
| "grad_norm": 3.4717910289764404, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 0.4257, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.1823939202026599, | |
| "grad_norm": 1.8152432441711426, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.2427, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.19252691576947434, | |
| "grad_norm": 1.5609540939331055, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.1929, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.2026599113362888, | |
| "grad_norm": 1.3951321840286255, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.0777, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.21279290690310323, | |
| "grad_norm": 1.6223750114440918, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.061, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.22292590246991767, | |
| "grad_norm": 0.5461512804031372, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0276, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.2330588980367321, | |
| "grad_norm": 2.7356250286102295, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.258, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.24319189360354654, | |
| "grad_norm": 1.6690704822540283, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.0822, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.253324889170361, | |
| "grad_norm": 1.9990379810333252, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.1948, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2634578847371754, | |
| "grad_norm": 1.84889817237854, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0953, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.2634578847371754, | |
| "eval_loss": 0.1430395096540451, | |
| "eval_runtime": 3.934, | |
| "eval_samples_per_second": 42.45, | |
| "eval_steps_per_second": 21.352, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.27359088030398987, | |
| "grad_norm": 2.240417718887329, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.2279, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.28372387587080433, | |
| "grad_norm": 1.6692705154418945, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.1103, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.29385687143761874, | |
| "grad_norm": 1.1334484815597534, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.107, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.3039898670044332, | |
| "grad_norm": 1.8195340633392334, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0558, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3141228625712476, | |
| "grad_norm": 1.8854806423187256, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.3479, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.32425585813806207, | |
| "grad_norm": 2.1316421031951904, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 0.1326, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.3343888537048765, | |
| "grad_norm": 2.0381174087524414, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.3254, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.34452184927169094, | |
| "grad_norm": 1.3313711881637573, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.1059, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.3546548448385054, | |
| "grad_norm": 0.8397698998451233, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.0336, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3647878404053198, | |
| "grad_norm": 1.6760809421539307, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.0713, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.37492083597213427, | |
| "grad_norm": 0.7013064622879028, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.0364, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3850538315389487, | |
| "grad_norm": 0.7092528939247131, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.1465, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.39518682710576314, | |
| "grad_norm": 0.9758922457695007, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.1129, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.39518682710576314, | |
| "eval_loss": 0.10418615490198135, | |
| "eval_runtime": 3.9348, | |
| "eval_samples_per_second": 42.441, | |
| "eval_steps_per_second": 21.348, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.4053198226725776, | |
| "grad_norm": 1.2004624605178833, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.1043, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.415452818239392, | |
| "grad_norm": 0.6554136872291565, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.059, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.42558581380620647, | |
| "grad_norm": 2.516598701477051, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.2211, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.4357188093730209, | |
| "grad_norm": 0.694494366645813, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.0519, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.44585180493983534, | |
| "grad_norm": 0.9121596217155457, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.0693, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.4559848005066498, | |
| "grad_norm": 0.816668689250946, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.0688, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.4661177960734642, | |
| "grad_norm": 0.7210935354232788, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0747, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.47625079164027867, | |
| "grad_norm": 0.6753107309341431, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.0283, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.4863837872070931, | |
| "grad_norm": 0.4177698791027069, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.0249, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.49651678277390754, | |
| "grad_norm": 1.1643296480178833, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.1529, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.506649778340722, | |
| "grad_norm": 1.1032949686050415, | |
| "learning_rate": 0.0, | |
| "loss": 0.0554, | |
| "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": 3.2956458467328e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |