| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.007050860204945004, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 9.401146939926671e-05, | |
| "grad_norm": 1.1071391105651855, | |
| "learning_rate": 1e-05, | |
| "loss": 3.1216, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 9.401146939926671e-05, | |
| "eval_loss": 2.95656681060791, | |
| "eval_runtime": 253.1318, | |
| "eval_samples_per_second": 35.389, | |
| "eval_steps_per_second": 4.425, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00018802293879853342, | |
| "grad_norm": 0.9681172370910645, | |
| "learning_rate": 2e-05, | |
| "loss": 2.9395, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0002820344081978001, | |
| "grad_norm": 0.9034966230392456, | |
| "learning_rate": 3e-05, | |
| "loss": 2.6474, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00037604587759706683, | |
| "grad_norm": 1.0967257022857666, | |
| "learning_rate": 4e-05, | |
| "loss": 2.9769, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00047005734699633354, | |
| "grad_norm": 0.9553994536399841, | |
| "learning_rate": 5e-05, | |
| "loss": 2.889, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0005640688163956002, | |
| "grad_norm": 1.050720453262329, | |
| "learning_rate": 6e-05, | |
| "loss": 2.8748, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.000658080285794867, | |
| "grad_norm": 1.073960781097412, | |
| "learning_rate": 7e-05, | |
| "loss": 2.8886, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0007520917551941337, | |
| "grad_norm": 1.1246281862258911, | |
| "learning_rate": 8e-05, | |
| "loss": 3.0083, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0008461032245934004, | |
| "grad_norm": 1.4751750230789185, | |
| "learning_rate": 9e-05, | |
| "loss": 3.011, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0008461032245934004, | |
| "eval_loss": 2.774919033050537, | |
| "eval_runtime": 252.8918, | |
| "eval_samples_per_second": 35.422, | |
| "eval_steps_per_second": 4.429, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0009401146939926671, | |
| "grad_norm": 1.148061752319336, | |
| "learning_rate": 0.0001, | |
| "loss": 2.506, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.001034126163391934, | |
| "grad_norm": 1.3042597770690918, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 2.517, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0011281376327912005, | |
| "grad_norm": 1.1282496452331543, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 2.5106, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0012221491021904673, | |
| "grad_norm": 1.2557356357574463, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 2.3613, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.001316160571589734, | |
| "grad_norm": 1.1485098600387573, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 2.3087, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0014101720409890007, | |
| "grad_norm": 1.3429392576217651, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 2.293, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0015041835103882673, | |
| "grad_norm": 1.2680690288543701, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 2.0638, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0015981949797875341, | |
| "grad_norm": 1.4249025583267212, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 2.19, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0016922064491868007, | |
| "grad_norm": 1.2500929832458496, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 2.0502, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0016922064491868007, | |
| "eval_loss": 1.9111806154251099, | |
| "eval_runtime": 252.9419, | |
| "eval_samples_per_second": 35.415, | |
| "eval_steps_per_second": 4.428, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0017862179185860676, | |
| "grad_norm": 1.1099755764007568, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 1.8685, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0018802293879853342, | |
| "grad_norm": 1.2947331666946411, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.8453, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0019742408573846008, | |
| "grad_norm": 1.0603139400482178, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 1.7496, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.002068252326783868, | |
| "grad_norm": 1.1734318733215332, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 1.6085, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0021622637961831344, | |
| "grad_norm": 1.0922938585281372, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 1.5754, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.002256275265582401, | |
| "grad_norm": 1.184679627418518, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.6059, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0023502867349816676, | |
| "grad_norm": 1.039542555809021, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 1.3529, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0024442982043809346, | |
| "grad_norm": 1.1066913604736328, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 1.4648, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.002538309673780201, | |
| "grad_norm": 1.1527578830718994, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 1.2047, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002538309673780201, | |
| "eval_loss": 1.295229196548462, | |
| "eval_runtime": 252.4175, | |
| "eval_samples_per_second": 35.489, | |
| "eval_steps_per_second": 4.437, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.002632321143179468, | |
| "grad_norm": 1.4122720956802368, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 1.4598, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0027263326125787344, | |
| "grad_norm": 1.1315748691558838, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 1.2364, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0028203440819780014, | |
| "grad_norm": 1.089930772781372, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 1.2123, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.002914355551377268, | |
| "grad_norm": 1.0383858680725098, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 1.1399, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0030083670207765346, | |
| "grad_norm": 0.978026807308197, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.9636, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.0031023784901758012, | |
| "grad_norm": 0.9167242050170898, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.8764, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0031963899595750683, | |
| "grad_norm": 1.2081202268600464, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 1.0168, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.003290401428974335, | |
| "grad_norm": 1.3978822231292725, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.9602, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0033844128983736015, | |
| "grad_norm": 1.0774189233779907, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.8812, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0033844128983736015, | |
| "eval_loss": 0.842465341091156, | |
| "eval_runtime": 252.46, | |
| "eval_samples_per_second": 35.483, | |
| "eval_steps_per_second": 4.436, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.003478424367772868, | |
| "grad_norm": 1.4241242408752441, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.8564, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.003572435837172135, | |
| "grad_norm": 1.0194820165634155, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.7914, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0036664473065714017, | |
| "grad_norm": 0.9767679572105408, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.8322, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0037604587759706683, | |
| "grad_norm": 1.0500026941299438, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.7714, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0038544702453699353, | |
| "grad_norm": 0.8121858239173889, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.6804, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0039484817147692015, | |
| "grad_norm": 0.8457009792327881, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.6827, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.004042493184168469, | |
| "grad_norm": 0.7305456399917603, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.6149, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.004136504653567736, | |
| "grad_norm": 0.9492170810699463, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.5956, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.004230516122967002, | |
| "grad_norm": 0.8870730996131897, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.7184, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004230516122967002, | |
| "eval_loss": 0.650706946849823, | |
| "eval_runtime": 252.9778, | |
| "eval_samples_per_second": 35.41, | |
| "eval_steps_per_second": 4.427, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.004324527592366269, | |
| "grad_norm": 0.9316329956054688, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5832, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.004418539061765535, | |
| "grad_norm": 0.7315248250961304, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.5629, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.004512550531164802, | |
| "grad_norm": 0.9830067753791809, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.7129, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.004606562000564069, | |
| "grad_norm": 1.0499166250228882, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.5431, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.004700573469963335, | |
| "grad_norm": 0.7427704930305481, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.5814, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.004794584939362603, | |
| "grad_norm": 1.1322435140609741, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.6718, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.004888596408761869, | |
| "grad_norm": 0.7453857064247131, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.5561, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.004982607878161136, | |
| "grad_norm": 0.928943932056427, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.5619, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.005076619347560402, | |
| "grad_norm": 0.9922681450843811, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.5967, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005076619347560402, | |
| "eval_loss": 0.563852846622467, | |
| "eval_runtime": 253.4318, | |
| "eval_samples_per_second": 35.347, | |
| "eval_steps_per_second": 4.419, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.005170630816959669, | |
| "grad_norm": 0.7717866897583008, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4669, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.005264642286358936, | |
| "grad_norm": 0.9942904710769653, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.6044, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.005358653755758202, | |
| "grad_norm": 0.8181668519973755, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.5087, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.005452665225157469, | |
| "grad_norm": 0.8513317108154297, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.5598, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.005546676694556736, | |
| "grad_norm": 0.7018467783927917, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.5638, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.005640688163956003, | |
| "grad_norm": 0.938843309879303, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.5768, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0057346996333552695, | |
| "grad_norm": 0.9042795300483704, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.5493, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.005828711102754536, | |
| "grad_norm": 0.6332735419273376, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.531, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.005922722572153803, | |
| "grad_norm": 1.0177136659622192, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.5125, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.005922722572153803, | |
| "eval_loss": 0.5256929397583008, | |
| "eval_runtime": 252.2686, | |
| "eval_samples_per_second": 35.51, | |
| "eval_steps_per_second": 4.44, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.006016734041553069, | |
| "grad_norm": 0.7307918667793274, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.4858, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.006110745510952336, | |
| "grad_norm": 1.128337025642395, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.6508, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.0062047569803516025, | |
| "grad_norm": 0.6984878182411194, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.5505, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.00629876844975087, | |
| "grad_norm": 0.873836874961853, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.5086, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.0063927799191501366, | |
| "grad_norm": 0.9610751867294312, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.5957, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.006486791388549403, | |
| "grad_norm": 0.841046154499054, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.5448, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.00658080285794867, | |
| "grad_norm": 0.7103133201599121, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.4729, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.006674814327347936, | |
| "grad_norm": 0.8467883467674255, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.5291, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.006768825796747203, | |
| "grad_norm": 0.8004826903343201, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.4299, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.006768825796747203, | |
| "eval_loss": 0.4963763654232025, | |
| "eval_runtime": 252.6181, | |
| "eval_samples_per_second": 35.461, | |
| "eval_steps_per_second": 4.434, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0068628372661464696, | |
| "grad_norm": 0.7845374941825867, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.4789, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.006956848735545736, | |
| "grad_norm": 0.7107625007629395, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.4897, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.007050860204945004, | |
| "grad_norm": 0.8745075464248657, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.461, | |
| "step": 75 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7720265790259200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |