| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6224066390041494, | |
| "eval_steps": 9, | |
| "global_step": 75, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.008298755186721992, | |
| "grad_norm": 0.5584869980812073, | |
| "learning_rate": 1e-05, | |
| "loss": 0.4386, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.008298755186721992, | |
| "eval_loss": 0.525488555431366, | |
| "eval_runtime": 21.6583, | |
| "eval_samples_per_second": 4.71, | |
| "eval_steps_per_second": 0.6, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.016597510373443983, | |
| "grad_norm": 0.574292778968811, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4623, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.024896265560165973, | |
| "grad_norm": 0.5636454224586487, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9001, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.03319502074688797, | |
| "grad_norm": 0.464995801448822, | |
| "learning_rate": 4e-05, | |
| "loss": 0.3894, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.04149377593360996, | |
| "grad_norm": 0.48257189989089966, | |
| "learning_rate": 5e-05, | |
| "loss": 0.3823, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.04979253112033195, | |
| "grad_norm": 0.43475595116615295, | |
| "learning_rate": 6e-05, | |
| "loss": 0.3153, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.058091286307053944, | |
| "grad_norm": 0.5338840484619141, | |
| "learning_rate": 7e-05, | |
| "loss": 0.374, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.06639004149377593, | |
| "grad_norm": 0.4116763472557068, | |
| "learning_rate": 8e-05, | |
| "loss": 0.3165, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.07468879668049792, | |
| "grad_norm": 0.44430240988731384, | |
| "learning_rate": 9e-05, | |
| "loss": 0.3308, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.07468879668049792, | |
| "eval_loss": 0.36139607429504395, | |
| "eval_runtime": 20.832, | |
| "eval_samples_per_second": 4.896, | |
| "eval_steps_per_second": 0.624, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.08298755186721991, | |
| "grad_norm": 0.43339839577674866, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2451, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0912863070539419, | |
| "grad_norm": 0.41347187757492065, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.4802, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0995850622406639, | |
| "grad_norm": 0.39249882102012634, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.2027, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1078838174273859, | |
| "grad_norm": 0.3496515452861786, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.1702, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.11618257261410789, | |
| "grad_norm": 0.3440171480178833, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.1495, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.12448132780082988, | |
| "grad_norm": 0.40277042984962463, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.3027, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.13278008298755187, | |
| "grad_norm": 0.45786339044570923, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.156, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.14107883817427386, | |
| "grad_norm": 0.3107019066810608, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.1383, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.14937759336099585, | |
| "grad_norm": 0.2789028286933899, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.1316, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.14937759336099585, | |
| "eval_loss": 0.2620299756526947, | |
| "eval_runtime": 20.8181, | |
| "eval_samples_per_second": 4.9, | |
| "eval_steps_per_second": 0.624, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.15767634854771784, | |
| "grad_norm": 0.3484548032283783, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.152, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.16597510373443983, | |
| "grad_norm": 0.3380146324634552, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1643, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17427385892116182, | |
| "grad_norm": 0.47338756918907166, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.1325, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.1825726141078838, | |
| "grad_norm": 0.2783995568752289, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.1535, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.1908713692946058, | |
| "grad_norm": 0.4390009641647339, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.1582, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.1991701244813278, | |
| "grad_norm": 0.2531968951225281, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.129, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.2074688796680498, | |
| "grad_norm": 0.21148426830768585, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1381, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2157676348547718, | |
| "grad_norm": 0.38490021228790283, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.1646, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.22406639004149378, | |
| "grad_norm": 0.3353732228279114, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.1546, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.22406639004149378, | |
| "eval_loss": 0.25112995505332947, | |
| "eval_runtime": 20.8311, | |
| "eval_samples_per_second": 4.897, | |
| "eval_steps_per_second": 0.624, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.23236514522821577, | |
| "grad_norm": 0.20061597228050232, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1076, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.24066390041493776, | |
| "grad_norm": 0.33759647607803345, | |
| "learning_rate": 8.940053768033609e-05, | |
| "loss": 0.135, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.24896265560165975, | |
| "grad_norm": 0.44867655634880066, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.1674, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2572614107883817, | |
| "grad_norm": 0.22644516825675964, | |
| "learning_rate": 8.715724127386972e-05, | |
| "loss": 0.1157, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.26556016597510373, | |
| "grad_norm": 0.2607629597187042, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.1263, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.27385892116182575, | |
| "grad_norm": 0.37812289595603943, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.1232, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.2821576763485477, | |
| "grad_norm": 0.1757567971944809, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.1075, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.29045643153526973, | |
| "grad_norm": 0.29007238149642944, | |
| "learning_rate": 8.213938048432697e-05, | |
| "loss": 0.1213, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.2987551867219917, | |
| "grad_norm": 0.2690522372722626, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.126, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2987551867219917, | |
| "eval_loss": 0.24171073734760284, | |
| "eval_runtime": 20.8568, | |
| "eval_samples_per_second": 4.89, | |
| "eval_steps_per_second": 0.623, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.3070539419087137, | |
| "grad_norm": 0.1884515881538391, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.1081, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.3153526970954357, | |
| "grad_norm": 0.15631933510303497, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.1287, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.3236514522821577, | |
| "grad_norm": 0.37540963292121887, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.1166, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.33195020746887965, | |
| "grad_norm": 0.27853959798812866, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.1122, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.34024896265560167, | |
| "grad_norm": 0.3291212320327759, | |
| "learning_rate": 7.347357813929454e-05, | |
| "loss": 0.1254, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.34854771784232363, | |
| "grad_norm": 0.3194391131401062, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.1155, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.35684647302904565, | |
| "grad_norm": 0.43088459968566895, | |
| "learning_rate": 7.033683215379002e-05, | |
| "loss": 0.5447, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.3651452282157676, | |
| "grad_norm": 0.1883140206336975, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.1902, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.37344398340248963, | |
| "grad_norm": 0.21305282413959503, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.1173, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.37344398340248963, | |
| "eval_loss": 0.23829738795757294, | |
| "eval_runtime": 20.8696, | |
| "eval_samples_per_second": 4.887, | |
| "eval_steps_per_second": 0.623, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.3817427385892116, | |
| "grad_norm": 0.40837693214416504, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.1158, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.3900414937759336, | |
| "grad_norm": 0.22184498608112335, | |
| "learning_rate": 6.378186779084995e-05, | |
| "loss": 0.1132, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.3983402489626556, | |
| "grad_norm": 0.28481265902519226, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.1149, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.4066390041493776, | |
| "grad_norm": 0.3043873608112335, | |
| "learning_rate": 6.0395584540887963e-05, | |
| "loss": 0.12, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.4149377593360996, | |
| "grad_norm": 0.28662800788879395, | |
| "learning_rate": 5.868240888334653e-05, | |
| "loss": 0.1488, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.42323651452282157, | |
| "grad_norm": 0.5023450255393982, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.6348, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.4315352697095436, | |
| "grad_norm": 0.23465168476104736, | |
| "learning_rate": 5.522642316338268e-05, | |
| "loss": 0.1158, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 0.43983402489626555, | |
| "grad_norm": 0.34985145926475525, | |
| "learning_rate": 5.348782368720626e-05, | |
| "loss": 0.1243, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 0.44813278008298757, | |
| "grad_norm": 0.1788734644651413, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.165, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.44813278008298757, | |
| "eval_loss": 0.22840574383735657, | |
| "eval_runtime": 20.8342, | |
| "eval_samples_per_second": 4.896, | |
| "eval_steps_per_second": 0.624, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.45643153526970953, | |
| "grad_norm": 0.24725890159606934, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1135, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.46473029045643155, | |
| "grad_norm": 0.20031580328941345, | |
| "learning_rate": 4.825502516487497e-05, | |
| "loss": 0.1122, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.4730290456431535, | |
| "grad_norm": 0.2718360126018524, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.1847, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.48132780082987553, | |
| "grad_norm": 0.30479809641838074, | |
| "learning_rate": 4.477357683661734e-05, | |
| "loss": 0.1487, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 0.4896265560165975, | |
| "grad_norm": 0.31583982706069946, | |
| "learning_rate": 4.3041344951996746e-05, | |
| "loss": 0.6241, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 0.4979253112033195, | |
| "grad_norm": 0.2300901710987091, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.1186, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5062240663900415, | |
| "grad_norm": 0.24245917797088623, | |
| "learning_rate": 3.960441545911204e-05, | |
| "loss": 0.1174, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 0.5145228215767634, | |
| "grad_norm": 0.30403587222099304, | |
| "learning_rate": 3.790390522001662e-05, | |
| "loss": 0.1879, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 0.5228215767634855, | |
| "grad_norm": 0.22756731510162354, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.1182, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5228215767634855, | |
| "eval_loss": 0.22431056201457977, | |
| "eval_runtime": 20.8796, | |
| "eval_samples_per_second": 4.885, | |
| "eval_steps_per_second": 0.623, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.5311203319502075, | |
| "grad_norm": 0.2197120040655136, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.1106, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.5394190871369294, | |
| "grad_norm": 0.2344292849302292, | |
| "learning_rate": 3.289899283371657e-05, | |
| "loss": 0.0985, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.5477178423236515, | |
| "grad_norm": 0.22797583043575287, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.1015, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.5560165975103735, | |
| "grad_norm": 0.2751520872116089, | |
| "learning_rate": 2.9663167846209998e-05, | |
| "loss": 0.2086, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 0.5643153526970954, | |
| "grad_norm": 0.2529605031013489, | |
| "learning_rate": 2.8081442660546125e-05, | |
| "loss": 0.1148, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 0.5726141078838174, | |
| "grad_norm": 0.2754276990890503, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.1237, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.5809128630705395, | |
| "grad_norm": 0.263017863035202, | |
| "learning_rate": 2.500000000000001e-05, | |
| "loss": 0.3886, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.5892116182572614, | |
| "grad_norm": 0.23464225232601166, | |
| "learning_rate": 2.350403678833976e-05, | |
| "loss": 0.1013, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 0.5975103734439834, | |
| "grad_norm": 0.33201128244400024, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.1089, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.5975103734439834, | |
| "eval_loss": 0.22274896502494812, | |
| "eval_runtime": 20.8128, | |
| "eval_samples_per_second": 4.901, | |
| "eval_steps_per_second": 0.625, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.6058091286307054, | |
| "grad_norm": 0.27148982882499695, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.3075, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 0.6141078838174274, | |
| "grad_norm": 0.21130910515785217, | |
| "learning_rate": 1.9216926233717085e-05, | |
| "loss": 0.1422, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 0.6224066390041494, | |
| "grad_norm": 0.18204365670681, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.1057, | |
| "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": 9.42196850121769e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |