| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.012656091798852515, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016874789065136686, | |
| "grad_norm": 0.6673353910446167, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1828, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016874789065136686, | |
| "eval_loss": 1.133446216583252, | |
| "eval_runtime": 670.9071, | |
| "eval_samples_per_second": 7.438, | |
| "eval_steps_per_second": 0.93, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003374957813027337, | |
| "grad_norm": 0.6421517133712769, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1352, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0005062436719541006, | |
| "grad_norm": 0.8541758060455322, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9903, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006749915626054674, | |
| "grad_norm": 0.4865100383758545, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1104, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0008437394532568343, | |
| "grad_norm": 0.5594805479049683, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9953, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0010124873439082012, | |
| "grad_norm": 0.5936105251312256, | |
| "learning_rate": 6e-05, | |
| "loss": 1.019, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001181235234559568, | |
| "grad_norm": 0.7241169214248657, | |
| "learning_rate": 7e-05, | |
| "loss": 1.0544, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0013499831252109348, | |
| "grad_norm": 0.8012577295303345, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0349, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0015187310158623017, | |
| "grad_norm": 0.6014957427978516, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8921, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0015187310158623017, | |
| "eval_loss": 0.9576449990272522, | |
| "eval_runtime": 672.5267, | |
| "eval_samples_per_second": 7.42, | |
| "eval_steps_per_second": 0.928, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016874789065136687, | |
| "grad_norm": 0.3161397874355316, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0526, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0018562267971650355, | |
| "grad_norm": 0.3254038393497467, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.9626, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0020249746878164025, | |
| "grad_norm": 0.29402822256088257, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8483, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0021937225784677693, | |
| "grad_norm": 0.3435806930065155, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.9679, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.002362470469119136, | |
| "grad_norm": 0.35293349623680115, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.9465, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.002531218359770503, | |
| "grad_norm": 0.3628636598587036, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.832, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0026999662504218697, | |
| "grad_norm": 0.354307621717453, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7619, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0028687141410732365, | |
| "grad_norm": 0.40063461661338806, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.8526, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0030374620317246033, | |
| "grad_norm": 0.3604790270328522, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.8343, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0030374620317246033, | |
| "eval_loss": 0.8436176180839539, | |
| "eval_runtime": 672.8891, | |
| "eval_samples_per_second": 7.416, | |
| "eval_steps_per_second": 0.927, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00320620992237597, | |
| "grad_norm": 0.32435283064842224, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8905, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0033749578130273373, | |
| "grad_norm": 0.29493603110313416, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.8299, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.003543705703678704, | |
| "grad_norm": 0.28454816341400146, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.8245, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.003712453594330071, | |
| "grad_norm": 0.30943214893341064, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.8342, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0038812014849814377, | |
| "grad_norm": 0.2970975637435913, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.854, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.004049949375632805, | |
| "grad_norm": 0.32714584469795227, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8385, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.004218697266284171, | |
| "grad_norm": 0.30870357155799866, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.715, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.004387445156935539, | |
| "grad_norm": 0.2743527591228485, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.8243, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.004556193047586905, | |
| "grad_norm": 0.2194703072309494, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.8237, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.004556193047586905, | |
| "eval_loss": 0.7984665632247925, | |
| "eval_runtime": 673.0127, | |
| "eval_samples_per_second": 7.414, | |
| "eval_steps_per_second": 0.927, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.004724940938238272, | |
| "grad_norm": 0.2173057645559311, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.6946, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0048936888288896386, | |
| "grad_norm": 0.26414012908935547, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.9683, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.005062436719541006, | |
| "grad_norm": 0.19070331752300262, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.7274, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.005231184610192373, | |
| "grad_norm": 0.22258605062961578, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.6563, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.005399932500843739, | |
| "grad_norm": 0.16770727932453156, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.7469, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.005568680391495107, | |
| "grad_norm": 0.2346215546131134, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.7293, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.005737428282146473, | |
| "grad_norm": 0.20247291028499603, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.722, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.00590617617279784, | |
| "grad_norm": 0.2722700834274292, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.8538, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.006074924063449207, | |
| "grad_norm": 0.23295146226882935, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.7712, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006074924063449207, | |
| "eval_loss": 0.7734177708625793, | |
| "eval_runtime": 672.6791, | |
| "eval_samples_per_second": 7.418, | |
| "eval_steps_per_second": 0.928, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.006243671954100574, | |
| "grad_norm": 0.22734515368938446, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.6952, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.00641241984475194, | |
| "grad_norm": 0.23274950683116913, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6745, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0065811677354033074, | |
| "grad_norm": 0.2169954478740692, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.8216, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.006749915626054675, | |
| "grad_norm": 0.24286970496177673, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.7567, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.006918663516706041, | |
| "grad_norm": 0.21593761444091797, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.7137, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.007087411407357408, | |
| "grad_norm": 0.23738643527030945, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.7267, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.007256159298008775, | |
| "grad_norm": 0.23834103345870972, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.8647, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.007424907188660142, | |
| "grad_norm": 0.19728927314281464, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.6978, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.007593655079311508, | |
| "grad_norm": 0.25146934390068054, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.8439, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.007593655079311508, | |
| "eval_loss": 0.7586240172386169, | |
| "eval_runtime": 672.7804, | |
| "eval_samples_per_second": 7.417, | |
| "eval_steps_per_second": 0.927, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0077624029699628755, | |
| "grad_norm": 0.2514830529689789, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.7707, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.007931150860614243, | |
| "grad_norm": 0.21285653114318848, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.6869, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.00809989875126561, | |
| "grad_norm": 0.2110971063375473, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.8896, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.008268646641916975, | |
| "grad_norm": 0.22739623486995697, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.7584, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.008437394532568343, | |
| "grad_norm": 0.27134278416633606, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.8999, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.00860614242321971, | |
| "grad_norm": 0.22978180646896362, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.838, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.008774890313871077, | |
| "grad_norm": 0.2255326360464096, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.8183, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.008943638204522443, | |
| "grad_norm": 0.23889334499835968, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.7689, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.00911238609517381, | |
| "grad_norm": 0.23411183059215546, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.749, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.00911238609517381, | |
| "eval_loss": 0.7487614154815674, | |
| "eval_runtime": 672.4672, | |
| "eval_samples_per_second": 7.42, | |
| "eval_steps_per_second": 0.928, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.009281133985825177, | |
| "grad_norm": 0.21135364472866058, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6483, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.009449881876476544, | |
| "grad_norm": 0.22199489176273346, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.7055, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.009618629767127912, | |
| "grad_norm": 0.26034969091415405, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.8685, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.009787377657779277, | |
| "grad_norm": 0.2556268870830536, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.8371, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.009956125548430644, | |
| "grad_norm": 0.23750770092010498, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.8321, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.010124873439082012, | |
| "grad_norm": 0.25214332342147827, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.7922, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.010293621329733379, | |
| "grad_norm": 0.2333674132823944, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.7721, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.010462369220384746, | |
| "grad_norm": 0.2577410936355591, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.8861, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.010631117111036112, | |
| "grad_norm": 0.24156250059604645, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.7453, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.010631117111036112, | |
| "eval_loss": 0.742745041847229, | |
| "eval_runtime": 672.4867, | |
| "eval_samples_per_second": 7.42, | |
| "eval_steps_per_second": 0.928, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.010799865001687479, | |
| "grad_norm": 0.2269875854253769, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.6588, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.010968612892338846, | |
| "grad_norm": 0.2180250734090805, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.7415, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.011137360782990213, | |
| "grad_norm": 0.2491016685962677, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.8364, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.011306108673641579, | |
| "grad_norm": 0.23741914331912994, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.8251, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.011474856564292946, | |
| "grad_norm": 0.2304244041442871, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.792, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.011643604454944313, | |
| "grad_norm": 0.2569412589073181, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.6614, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.01181235234559568, | |
| "grad_norm": 0.20252695679664612, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.6474, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.011981100236247048, | |
| "grad_norm": 0.2276439368724823, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.7243, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.012149848126898413, | |
| "grad_norm": 0.28231948614120483, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.7426, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.012149848126898413, | |
| "eval_loss": 0.7391621470451355, | |
| "eval_runtime": 672.7088, | |
| "eval_samples_per_second": 7.418, | |
| "eval_steps_per_second": 0.928, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.01231859601754978, | |
| "grad_norm": 0.24270500242710114, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.6727, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.012487343908201148, | |
| "grad_norm": 0.24265378713607788, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.6515, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.012656091798852515, | |
| "grad_norm": 0.23576249182224274, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.8245, | |
| "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": 6.181229992869888e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |