| { | |
| "best_metric": 0.011823242530226707, | |
| "best_model_checkpoint": "/home/paperspace/Data/models/relianceV2/llm3br256/checkpoint-100", | |
| "epoch": 4.921212121212121, | |
| "eval_steps": 5, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.048484848484848485, | |
| "grad_norm": 0.13376639783382416, | |
| "learning_rate": 1e-05, | |
| "loss": 0.063, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.09696969696969697, | |
| "grad_norm": 0.12203846871852875, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0607, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.14545454545454545, | |
| "grad_norm": 0.13988564908504486, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0615, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.19393939393939394, | |
| "grad_norm": 0.10985743254423141, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0504, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "grad_norm": 0.07405027747154236, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0395, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.24242424242424243, | |
| "eval_loss": 0.04326783865690231, | |
| "eval_runtime": 11.3564, | |
| "eval_samples_per_second": 4.403, | |
| "eval_steps_per_second": 1.145, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2909090909090909, | |
| "grad_norm": 0.09203900396823883, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0419, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3393939393939394, | |
| "grad_norm": 0.07265866547822952, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0381, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3878787878787879, | |
| "grad_norm": 0.04461047425866127, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0341, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.43636363636363634, | |
| "grad_norm": 0.04746266454458237, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0345, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "grad_norm": 0.040007542818784714, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0324, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.48484848484848486, | |
| "eval_loss": 0.03000650927424431, | |
| "eval_runtime": 10.3489, | |
| "eval_samples_per_second": 4.831, | |
| "eval_steps_per_second": 1.256, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5333333333333333, | |
| "grad_norm": 0.03799016401171684, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.0331, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.5818181818181818, | |
| "grad_norm": 0.03196106478571892, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.029, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.6303030303030303, | |
| "grad_norm": 0.027229588478803635, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.0257, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.6787878787878788, | |
| "grad_norm": 0.02768268622457981, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.0215, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.032983552664518356, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.024, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "eval_loss": 0.024439169093966484, | |
| "eval_runtime": 10.352, | |
| "eval_samples_per_second": 4.83, | |
| "eval_steps_per_second": 1.256, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7757575757575758, | |
| "grad_norm": 0.03327898308634758, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0321, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.8242424242424242, | |
| "grad_norm": 0.02506704442203045, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.0281, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.8727272727272727, | |
| "grad_norm": 0.023538140580058098, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.0202, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.9212121212121213, | |
| "grad_norm": 0.02581767924129963, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.0206, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "grad_norm": 0.021269701421260834, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0189, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9696969696969697, | |
| "eval_loss": 0.0212231557816267, | |
| "eval_runtime": 10.4147, | |
| "eval_samples_per_second": 4.801, | |
| "eval_steps_per_second": 1.248, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0363636363636364, | |
| "grad_norm": 0.037210095673799515, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.0317, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.084848484848485, | |
| "grad_norm": 0.02000698819756508, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.019, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.1333333333333333, | |
| "grad_norm": 0.020116321742534637, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.019, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.1818181818181819, | |
| "grad_norm": 0.022886989638209343, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0181, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.2303030303030302, | |
| "grad_norm": 0.023031054064631462, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.0171, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2303030303030302, | |
| "eval_loss": 0.019027845934033394, | |
| "eval_runtime": 10.3362, | |
| "eval_samples_per_second": 4.837, | |
| "eval_steps_per_second": 1.258, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2787878787878788, | |
| "grad_norm": 0.01873927377164364, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0182, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.3272727272727272, | |
| "grad_norm": 0.018605511635541916, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.0163, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.3757575757575757, | |
| "grad_norm": 0.023363754153251648, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0177, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.4242424242424243, | |
| "grad_norm": 0.017733635380864143, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.014, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.4727272727272727, | |
| "grad_norm": 0.0225802194327116, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0146, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4727272727272727, | |
| "eval_loss": 0.01733844168484211, | |
| "eval_runtime": 10.3386, | |
| "eval_samples_per_second": 4.836, | |
| "eval_steps_per_second": 1.257, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.5212121212121212, | |
| "grad_norm": 0.018234577029943466, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.017, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.5696969696969698, | |
| "grad_norm": 0.024176109582185745, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0155, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.6181818181818182, | |
| "grad_norm": 0.023358961567282677, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.0157, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 0.025254715234041214, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0167, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.7151515151515153, | |
| "grad_norm": 0.021989861503243446, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.0144, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.7151515151515153, | |
| "eval_loss": 0.016134196892380714, | |
| "eval_runtime": 10.3976, | |
| "eval_samples_per_second": 4.809, | |
| "eval_steps_per_second": 1.25, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.7636363636363637, | |
| "grad_norm": 0.018357617780566216, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0162, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.812121212121212, | |
| "grad_norm": 0.021665558218955994, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.0194, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.8606060606060606, | |
| "grad_norm": 0.020184990018606186, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.0154, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.9090909090909092, | |
| "grad_norm": 0.016307702288031578, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.0134, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.9575757575757575, | |
| "grad_norm": 0.01873622089624405, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.0104, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.9575757575757575, | |
| "eval_loss": 0.015456685796380043, | |
| "eval_runtime": 10.3404, | |
| "eval_samples_per_second": 4.835, | |
| "eval_steps_per_second": 1.257, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0242424242424244, | |
| "grad_norm": 0.03928119316697121, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.0277, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.0727272727272728, | |
| "grad_norm": 0.01585806906223297, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.012, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.121212121212121, | |
| "grad_norm": 0.022943127900362015, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.0147, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.16969696969697, | |
| "grad_norm": 0.022961463779211044, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.014, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.2181818181818183, | |
| "grad_norm": 0.017532989382743835, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.0143, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.2181818181818183, | |
| "eval_loss": 0.01516613457351923, | |
| "eval_runtime": 10.3353, | |
| "eval_samples_per_second": 4.838, | |
| "eval_steps_per_second": 1.258, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.2666666666666666, | |
| "grad_norm": 0.026267599314451218, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.0126, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.315151515151515, | |
| "grad_norm": 0.024150898680090904, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.0118, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.3636363636363638, | |
| "grad_norm": 0.019730910658836365, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.0121, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.412121212121212, | |
| "grad_norm": 0.02562836930155754, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.0096, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.4606060606060605, | |
| "grad_norm": 0.01951598934829235, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.0117, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.4606060606060605, | |
| "eval_loss": 0.014072530902922153, | |
| "eval_runtime": 10.3868, | |
| "eval_samples_per_second": 4.814, | |
| "eval_steps_per_second": 1.252, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.509090909090909, | |
| "grad_norm": 0.02022629976272583, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.0139, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.5575757575757576, | |
| "grad_norm": 0.01750958524644375, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.0099, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.606060606060606, | |
| "grad_norm": 0.0212919432669878, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.013, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.6545454545454543, | |
| "grad_norm": 0.016065998002886772, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.0093, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.703030303030303, | |
| "grad_norm": 0.018324071541428566, | |
| "learning_rate": 5e-05, | |
| "loss": 0.015, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.703030303030303, | |
| "eval_loss": 0.013632840476930141, | |
| "eval_runtime": 10.3292, | |
| "eval_samples_per_second": 4.841, | |
| "eval_steps_per_second": 1.259, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.7515151515151515, | |
| "grad_norm": 0.02191258780658245, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.0109, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 0.01748000457882881, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.0104, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 2.8484848484848486, | |
| "grad_norm": 0.01487227063626051, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.0097, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.896969696969697, | |
| "grad_norm": 0.01992475800216198, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.0108, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 2.9454545454545453, | |
| "grad_norm": 0.01630261354148388, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.0092, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.9454545454545453, | |
| "eval_loss": 0.013053460977971554, | |
| "eval_runtime": 10.3658, | |
| "eval_samples_per_second": 4.824, | |
| "eval_steps_per_second": 1.254, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.012121212121212, | |
| "grad_norm": 0.03835081681609154, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.0172, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.0606060606060606, | |
| "grad_norm": 0.017339400947093964, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.008, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.109090909090909, | |
| "grad_norm": 0.01646747626364231, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.0121, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 3.1575757575757577, | |
| "grad_norm": 0.01738414168357849, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.008, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.206060606060606, | |
| "grad_norm": 0.022966962307691574, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.008, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.206060606060606, | |
| "eval_loss": 0.012713730335235596, | |
| "eval_runtime": 10.3449, | |
| "eval_samples_per_second": 4.833, | |
| "eval_steps_per_second": 1.257, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.2545454545454544, | |
| "grad_norm": 0.01976313628256321, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.0096, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 3.303030303030303, | |
| "grad_norm": 0.015671931207180023, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.0071, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 3.3515151515151516, | |
| "grad_norm": 0.02184975892305374, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.0092, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.021975932642817497, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.009, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.4484848484848483, | |
| "grad_norm": 0.02743859961628914, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.0109, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.4484848484848483, | |
| "eval_loss": 0.012468294240534306, | |
| "eval_runtime": 10.3407, | |
| "eval_samples_per_second": 4.835, | |
| "eval_steps_per_second": 1.257, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.496969696969697, | |
| "grad_norm": 0.02249360829591751, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.0123, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 3.5454545454545454, | |
| "grad_norm": 0.019340503960847855, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.0072, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.5939393939393938, | |
| "grad_norm": 0.017731212079524994, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.009, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 3.6424242424242426, | |
| "grad_norm": 0.017661720514297485, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.0078, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.690909090909091, | |
| "grad_norm": 0.01668357476592064, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.0085, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.690909090909091, | |
| "eval_loss": 0.012197648175060749, | |
| "eval_runtime": 10.3397, | |
| "eval_samples_per_second": 4.836, | |
| "eval_steps_per_second": 1.257, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.7393939393939393, | |
| "grad_norm": 0.019251663237810135, | |
| "learning_rate": 1.6543469682057106e-05, | |
| "loss": 0.0089, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 3.787878787878788, | |
| "grad_norm": 0.018358884379267693, | |
| "learning_rate": 1.526708147705013e-05, | |
| "loss": 0.0127, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 3.8363636363636364, | |
| "grad_norm": 0.018658000975847244, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.0085, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.8848484848484848, | |
| "grad_norm": 0.01768352836370468, | |
| "learning_rate": 1.2842758726130283e-05, | |
| "loss": 0.0079, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "grad_norm": 0.017675764858722687, | |
| "learning_rate": 1.1697777844051105e-05, | |
| "loss": 0.0089, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "eval_loss": 0.011956109665334225, | |
| "eval_runtime": 10.3374, | |
| "eval_samples_per_second": 4.837, | |
| "eval_steps_per_second": 1.258, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.981818181818182, | |
| "grad_norm": 0.04371407628059387, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.013, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 4.048484848484849, | |
| "grad_norm": 0.01942707970738411, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.0092, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.096969696969697, | |
| "grad_norm": 0.014996436424553394, | |
| "learning_rate": 8.548121372247918e-06, | |
| "loss": 0.0064, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 4.1454545454545455, | |
| "grad_norm": 0.01869269274175167, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.0086, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 4.193939393939394, | |
| "grad_norm": 0.017618168145418167, | |
| "learning_rate": 6.698729810778065e-06, | |
| "loss": 0.0074, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.193939393939394, | |
| "eval_loss": 0.011830544099211693, | |
| "eval_runtime": 10.3339, | |
| "eval_samples_per_second": 4.838, | |
| "eval_steps_per_second": 1.258, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.242424242424242, | |
| "grad_norm": 0.016551963984966278, | |
| "learning_rate": 5.852620357053651e-06, | |
| "loss": 0.009, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 4.290909090909091, | |
| "grad_norm": 0.01904563419520855, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.0074, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 4.33939393939394, | |
| "grad_norm": 0.02098240703344345, | |
| "learning_rate": 4.322727117869951e-06, | |
| "loss": 0.0094, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 4.387878787878788, | |
| "grad_norm": 0.015075408853590488, | |
| "learning_rate": 3.6408072716606346e-06, | |
| "loss": 0.0084, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 4.4363636363636365, | |
| "grad_norm": 0.016622379422187805, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.0074, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.4363636363636365, | |
| "eval_loss": 0.011814383789896965, | |
| "eval_runtime": 10.356, | |
| "eval_samples_per_second": 4.828, | |
| "eval_steps_per_second": 1.255, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.484848484848484, | |
| "grad_norm": 0.017493437975645065, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.0087, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.533333333333333, | |
| "grad_norm": 0.01581619866192341, | |
| "learning_rate": 1.9369152030840556e-06, | |
| "loss": 0.0073, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.581818181818182, | |
| "grad_norm": 0.020205620676279068, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.0075, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 4.63030303030303, | |
| "grad_norm": 0.016090713441371918, | |
| "learning_rate": 1.0926199633097157e-06, | |
| "loss": 0.0071, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 4.678787878787879, | |
| "grad_norm": 0.014329814352095127, | |
| "learning_rate": 7.596123493895991e-07, | |
| "loss": 0.0066, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.678787878787879, | |
| "eval_loss": 0.011838787235319614, | |
| "eval_runtime": 10.3353, | |
| "eval_samples_per_second": 4.838, | |
| "eval_steps_per_second": 1.258, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.015449507161974907, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.0063, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 4.775757575757575, | |
| "grad_norm": 0.016315339133143425, | |
| "learning_rate": 2.7390523158633554e-07, | |
| "loss": 0.0076, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 4.824242424242424, | |
| "grad_norm": 0.018415048718452454, | |
| "learning_rate": 1.2179748700879012e-07, | |
| "loss": 0.0103, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 4.872727272727273, | |
| "grad_norm": 0.014639717526733875, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.0057, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 4.921212121212121, | |
| "grad_norm": 0.018388710916042328, | |
| "learning_rate": 0.0, | |
| "loss": 0.0065, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.921212121212121, | |
| "eval_loss": 0.011823242530226707, | |
| "eval_runtime": 10.3329, | |
| "eval_samples_per_second": 4.839, | |
| "eval_steps_per_second": 1.258, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.921212121212121, | |
| "step": 100, | |
| "total_flos": 2.442632501894185e+17, | |
| "train_loss": 0.016519791511818767, | |
| "train_runtime": 2009.2463, | |
| "train_samples_per_second": 1.637, | |
| "train_steps_per_second": 0.05 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.442632501894185e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |