| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.11355034065102196, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001514004542013626, | |
| "grad_norm": 0.6870254278182983, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7026, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001514004542013626, | |
| "eval_loss": 0.6710895299911499, | |
| "eval_runtime": 127.6316, | |
| "eval_samples_per_second": 4.356, | |
| "eval_steps_per_second": 0.548, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.003028009084027252, | |
| "grad_norm": 1.0985076427459717, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6838, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.004542013626040878, | |
| "grad_norm": 0.6782929301261902, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6833, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.006056018168054504, | |
| "grad_norm": 0.6256589293479919, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7221, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00757002271006813, | |
| "grad_norm": 0.7078294157981873, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6957, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.009084027252081756, | |
| "grad_norm": 0.6197799444198608, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7096, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010598031794095382, | |
| "grad_norm": 0.6056140065193176, | |
| "learning_rate": 7e-05, | |
| "loss": 0.65, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.012112036336109008, | |
| "grad_norm": 0.9818187952041626, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5974, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.013626040878122634, | |
| "grad_norm": 0.7567906379699707, | |
| "learning_rate": 9e-05, | |
| "loss": 0.6949, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.013626040878122634, | |
| "eval_loss": 0.6097605228424072, | |
| "eval_runtime": 127.5185, | |
| "eval_samples_per_second": 4.36, | |
| "eval_steps_per_second": 0.549, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01514004542013626, | |
| "grad_norm": 0.9188516736030579, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6127, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.016654049962149888, | |
| "grad_norm": 0.8129972815513611, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.6875, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.018168054504163512, | |
| "grad_norm": 0.9102829694747925, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6385, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.01968205904617714, | |
| "grad_norm": 0.9624842405319214, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.667, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.021196063588190765, | |
| "grad_norm": 0.7224256992340088, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.6147, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.022710068130204392, | |
| "grad_norm": 0.6421948671340942, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.7355, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.024224072672218017, | |
| "grad_norm": 0.6296693682670593, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.5936, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.025738077214231644, | |
| "grad_norm": 0.529050886631012, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.6936, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.02725208175624527, | |
| "grad_norm": 0.6707794070243835, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.5679, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02725208175624527, | |
| "eval_loss": 0.5738157033920288, | |
| "eval_runtime": 127.5235, | |
| "eval_samples_per_second": 4.36, | |
| "eval_steps_per_second": 0.549, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.028766086298258896, | |
| "grad_norm": 0.6477593779563904, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.5251, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.03028009084027252, | |
| "grad_norm": 0.5631066560745239, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.5807, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03179409538228615, | |
| "grad_norm": 0.6563144326210022, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.5969, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.033308099924299776, | |
| "grad_norm": 0.5367952585220337, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.7125, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0348221044663134, | |
| "grad_norm": 0.5025123953819275, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.7012, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.036336109008327025, | |
| "grad_norm": 0.5811274647712708, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.5697, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03785011355034065, | |
| "grad_norm": 0.5691922903060913, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.674, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03936411809235428, | |
| "grad_norm": 0.5442315340042114, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.542, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0408781226343679, | |
| "grad_norm": 0.49935996532440186, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.5723, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0408781226343679, | |
| "eval_loss": 0.5580048561096191, | |
| "eval_runtime": 127.335, | |
| "eval_samples_per_second": 4.366, | |
| "eval_steps_per_second": 0.55, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.04239212717638153, | |
| "grad_norm": 0.43072766065597534, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.6508, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.04390613171839516, | |
| "grad_norm": 0.559459388256073, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.5142, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.045420136260408785, | |
| "grad_norm": 0.5589441061019897, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.5605, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.046934140802422405, | |
| "grad_norm": 0.46951043605804443, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.6581, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.04844814534443603, | |
| "grad_norm": 0.5119128227233887, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.5879, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04996214988644966, | |
| "grad_norm": 0.680155873298645, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.5249, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.05147615442846329, | |
| "grad_norm": 0.5114449858665466, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.5143, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.05299015897047691, | |
| "grad_norm": 0.43590912222862244, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.5876, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.05450416351249054, | |
| "grad_norm": 0.4947156310081482, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6025, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05450416351249054, | |
| "eval_loss": 0.5477901697158813, | |
| "eval_runtime": 127.3887, | |
| "eval_samples_per_second": 4.365, | |
| "eval_steps_per_second": 0.549, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.056018168054504165, | |
| "grad_norm": 0.5622116923332214, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.5587, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.05753217259651779, | |
| "grad_norm": 0.442304790019989, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6293, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.059046177138531414, | |
| "grad_norm": 0.5162755846977234, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.6347, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.06056018168054504, | |
| "grad_norm": 0.4614929258823395, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.5356, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.06207418622255867, | |
| "grad_norm": 0.4985050559043884, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.5409, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.0635881907645723, | |
| "grad_norm": 0.5837581753730774, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.4478, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.06510219530658592, | |
| "grad_norm": 0.4671187996864319, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.6622, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.06661619984859955, | |
| "grad_norm": 0.5558192729949951, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.6513, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.06813020439061317, | |
| "grad_norm": 0.5534436106681824, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.649, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06813020439061317, | |
| "eval_loss": 0.5416960716247559, | |
| "eval_runtime": 127.3858, | |
| "eval_samples_per_second": 4.365, | |
| "eval_steps_per_second": 0.55, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0696442089326268, | |
| "grad_norm": 0.5023759603500366, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.6837, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.07115821347464042, | |
| "grad_norm": 0.4820226728916168, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.6636, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.07267221801665405, | |
| "grad_norm": 0.5207632184028625, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5335, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.07418622255866768, | |
| "grad_norm": 0.45377227663993835, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.6229, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0757002271006813, | |
| "grad_norm": 0.46908336877822876, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.699, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07721423164269493, | |
| "grad_norm": 0.5295423269271851, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.5422, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07872823618470856, | |
| "grad_norm": 0.4357456862926483, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.5737, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.08024224072672217, | |
| "grad_norm": 0.5066563487052917, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.571, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.0817562452687358, | |
| "grad_norm": 0.449705570936203, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.6728, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.0817562452687358, | |
| "eval_loss": 0.5378050208091736, | |
| "eval_runtime": 127.4206, | |
| "eval_samples_per_second": 4.364, | |
| "eval_steps_per_second": 0.549, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.08327024981074943, | |
| "grad_norm": 0.4846964180469513, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6263, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.08478425435276306, | |
| "grad_norm": 0.3770967423915863, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.5541, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.08629825889477669, | |
| "grad_norm": 0.49300625920295715, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.4627, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08781226343679031, | |
| "grad_norm": 0.4052840769290924, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.6452, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.08932626797880394, | |
| "grad_norm": 0.4715135097503662, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.6527, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.09084027252081757, | |
| "grad_norm": 0.48373106122016907, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.4634, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.09235427706283118, | |
| "grad_norm": 0.4589581787586212, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.6037, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.09386828160484481, | |
| "grad_norm": 0.49407216906547546, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.5858, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.09538228614685844, | |
| "grad_norm": 0.49783697724342346, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.4889, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09538228614685844, | |
| "eval_loss": 0.5348504185676575, | |
| "eval_runtime": 127.4674, | |
| "eval_samples_per_second": 4.362, | |
| "eval_steps_per_second": 0.549, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.09689629068887207, | |
| "grad_norm": 0.4428796172142029, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.5492, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.0984102952308857, | |
| "grad_norm": 0.49797767400741577, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.4911, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.09992429977289932, | |
| "grad_norm": 0.47284457087516785, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.6434, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.10143830431491295, | |
| "grad_norm": 0.42389413714408875, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.6123, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.10295230885692658, | |
| "grad_norm": 0.4539527893066406, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.542, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.10446631339894019, | |
| "grad_norm": 0.48071274161338806, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.6658, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.10598031794095382, | |
| "grad_norm": 0.4266805648803711, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.4129, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.10749432248296745, | |
| "grad_norm": 0.4307482838630676, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.5453, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.10900832702498107, | |
| "grad_norm": 0.4067230820655823, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.6219, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10900832702498107, | |
| "eval_loss": 0.5327683091163635, | |
| "eval_runtime": 127.4093, | |
| "eval_samples_per_second": 4.364, | |
| "eval_steps_per_second": 0.549, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.1105223315669947, | |
| "grad_norm": 0.5064144730567932, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.4994, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.11203633610900833, | |
| "grad_norm": 0.44154641032218933, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.6057, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.11355034065102196, | |
| "grad_norm": 0.4402220845222473, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.5487, | |
| "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": 1.0571893790736384e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |