| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9902912621359223, | |
| "eval_steps": 500, | |
| "global_step": 51, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.019417475728155338, | |
| "grad_norm": 6.841878890991211, | |
| "learning_rate": 2e-05, | |
| "loss": 4.5864, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.038834951456310676, | |
| "grad_norm": 6.640402793884277, | |
| "learning_rate": 4e-05, | |
| "loss": 4.6138, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.05825242718446602, | |
| "grad_norm": 6.41209077835083, | |
| "learning_rate": 6e-05, | |
| "loss": 4.6064, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.07766990291262135, | |
| "grad_norm": 6.082383632659912, | |
| "learning_rate": 8e-05, | |
| "loss": 4.5534, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0970873786407767, | |
| "grad_norm": 4.00201416015625, | |
| "learning_rate": 0.0001, | |
| "loss": 4.3245, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.11650485436893204, | |
| "grad_norm": 3.1707470417022705, | |
| "learning_rate": 0.00012, | |
| "loss": 4.0971, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.13592233009708737, | |
| "grad_norm": 3.160670518875122, | |
| "learning_rate": 0.00014, | |
| "loss": 3.789, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1553398058252427, | |
| "grad_norm": 2.994746208190918, | |
| "learning_rate": 0.00016, | |
| "loss": 3.3244, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.17475728155339806, | |
| "grad_norm": 2.6664786338806152, | |
| "learning_rate": 0.00018, | |
| "loss": 2.8561, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1941747572815534, | |
| "grad_norm": 2.029910087585449, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3926, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.21359223300970873, | |
| "grad_norm": 1.3144137859344482, | |
| "learning_rate": 0.00019998688836656323, | |
| "loss": 2.0283, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.23300970873786409, | |
| "grad_norm": 0.8263471722602844, | |
| "learning_rate": 0.00019994755690455152, | |
| "loss": 1.8683, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2524271844660194, | |
| "grad_norm": 0.7193414568901062, | |
| "learning_rate": 0.0001998820159279591, | |
| "loss": 1.732, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.27184466019417475, | |
| "grad_norm": 0.8350910544395447, | |
| "learning_rate": 0.00019979028262377118, | |
| "loss": 1.6208, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.2912621359223301, | |
| "grad_norm": 0.6700139045715332, | |
| "learning_rate": 0.00019967238104745696, | |
| "loss": 1.493, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3106796116504854, | |
| "grad_norm": 1.1213518381118774, | |
| "learning_rate": 0.0001995283421166614, | |
| "loss": 1.4857, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3300970873786408, | |
| "grad_norm": 0.7505896687507629, | |
| "learning_rate": 0.00019935820360309777, | |
| "loss": 1.4643, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.34951456310679613, | |
| "grad_norm": 0.6984190940856934, | |
| "learning_rate": 0.00019916201012264254, | |
| "loss": 1.4227, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.36893203883495146, | |
| "grad_norm": 0.46819525957107544, | |
| "learning_rate": 0.00019893981312363562, | |
| "loss": 1.3889, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.3883495145631068, | |
| "grad_norm": 0.5214670300483704, | |
| "learning_rate": 0.00019869167087338907, | |
| "loss": 1.3764, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4077669902912621, | |
| "grad_norm": 0.43329402804374695, | |
| "learning_rate": 0.00019841764844290744, | |
| "loss": 1.351, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.42718446601941745, | |
| "grad_norm": 0.6535053253173828, | |
| "learning_rate": 0.0001981178176898239, | |
| "loss": 1.3576, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.44660194174757284, | |
| "grad_norm": 0.6363189816474915, | |
| "learning_rate": 0.00019779225723955707, | |
| "loss": 1.3159, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.46601941747572817, | |
| "grad_norm": 0.42140620946884155, | |
| "learning_rate": 0.00019744105246469263, | |
| "loss": 1.3072, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.4854368932038835, | |
| "grad_norm": 0.6757440567016602, | |
| "learning_rate": 0.00019706429546259593, | |
| "loss": 1.3261, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5048543689320388, | |
| "grad_norm": 0.4116632044315338, | |
| "learning_rate": 0.00019666208503126112, | |
| "loss": 1.317, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.5242718446601942, | |
| "grad_norm": 0.5359898209571838, | |
| "learning_rate": 0.00019623452664340306, | |
| "loss": 1.2932, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5436893203883495, | |
| "grad_norm": 0.4596949517726898, | |
| "learning_rate": 0.00019578173241879872, | |
| "loss": 1.2601, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.5631067961165048, | |
| "grad_norm": 0.6857442855834961, | |
| "learning_rate": 0.0001953038210948861, | |
| "loss": 1.2734, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.5825242718446602, | |
| "grad_norm": 0.4453957676887512, | |
| "learning_rate": 0.00019480091799562704, | |
| "loss": 1.2521, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6019417475728155, | |
| "grad_norm": 0.6118847131729126, | |
| "learning_rate": 0.00019427315499864344, | |
| "loss": 1.2252, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.6213592233009708, | |
| "grad_norm": 0.4894339442253113, | |
| "learning_rate": 0.00019372067050063438, | |
| "loss": 1.2369, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.6407766990291263, | |
| "grad_norm": 0.5901785492897034, | |
| "learning_rate": 0.00019314360938108425, | |
| "loss": 1.2507, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.6601941747572816, | |
| "grad_norm": 0.4147897958755493, | |
| "learning_rate": 0.00019254212296427044, | |
| "loss": 1.2454, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.6796116504854369, | |
| "grad_norm": 0.2836126685142517, | |
| "learning_rate": 0.00019191636897958122, | |
| "loss": 1.2515, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.6990291262135923, | |
| "grad_norm": 0.4523944556713104, | |
| "learning_rate": 0.00019126651152015403, | |
| "loss": 1.2217, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7184466019417476, | |
| "grad_norm": 0.4187013804912567, | |
| "learning_rate": 0.0001905927209998447, | |
| "loss": 1.2382, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.7378640776699029, | |
| "grad_norm": 0.516381561756134, | |
| "learning_rate": 0.00018989517410853955, | |
| "loss": 1.2214, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.7572815533980582, | |
| "grad_norm": 0.5928078889846802, | |
| "learning_rate": 0.00018917405376582145, | |
| "loss": 1.2407, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.7766990291262136, | |
| "grad_norm": 0.4062119722366333, | |
| "learning_rate": 0.00018842954907300236, | |
| "loss": 1.1958, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.7961165048543689, | |
| "grad_norm": 0.5966278910636902, | |
| "learning_rate": 0.0001876618552635348, | |
| "loss": 1.2043, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.8155339805825242, | |
| "grad_norm": 0.47196483612060547, | |
| "learning_rate": 0.00018687117365181512, | |
| "loss": 1.2321, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.8349514563106796, | |
| "grad_norm": 0.6465126872062683, | |
| "learning_rate": 0.00018605771158039253, | |
| "loss": 1.1996, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.8543689320388349, | |
| "grad_norm": 0.4234643578529358, | |
| "learning_rate": 0.00018522168236559695, | |
| "loss": 1.198, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.8737864077669902, | |
| "grad_norm": 0.3910824656486511, | |
| "learning_rate": 0.00018436330524160047, | |
| "loss": 1.1895, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8932038834951457, | |
| "grad_norm": 0.5380635261535645, | |
| "learning_rate": 0.00018348280530292713, | |
| "loss": 1.1693, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.912621359223301, | |
| "grad_norm": 0.31823453307151794, | |
| "learning_rate": 0.00018258041344542566, | |
| "loss": 1.159, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.9320388349514563, | |
| "grad_norm": 0.5679419636726379, | |
| "learning_rate": 0.0001816563663057211, | |
| "loss": 1.1931, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.9514563106796117, | |
| "grad_norm": 0.3980861008167267, | |
| "learning_rate": 0.00018071090619916093, | |
| "loss": 1.1826, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.970873786407767, | |
| "grad_norm": 0.3688671588897705, | |
| "learning_rate": 0.00017974428105627208, | |
| "loss": 1.176, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.9902912621359223, | |
| "grad_norm": 0.4137047827243805, | |
| "learning_rate": 0.00017875674435774547, | |
| "loss": 1.142, | |
| "step": 51 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 204, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 51, | |
| "total_flos": 1.0626058999037952e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |