| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.916083916083916, | |
| "eval_steps": 9, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.027972027972027972, | |
| "grad_norm": 5.75, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8618, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.027972027972027972, | |
| "eval_loss": 1.856866478919983, | |
| "eval_runtime": 4.5304, | |
| "eval_samples_per_second": 3.532, | |
| "eval_steps_per_second": 1.766, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.055944055944055944, | |
| "grad_norm": 5.5, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8343, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08391608391608392, | |
| "grad_norm": 5.75, | |
| "learning_rate": 6e-05, | |
| "loss": 1.87, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11188811188811189, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6976, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.13986013986013987, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0745, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16783216783216784, | |
| "grad_norm": 2.875, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5387, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1958041958041958, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 0.00014, | |
| "loss": 0.3055, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.22377622377622378, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0592, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0185, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "eval_loss": 0.0595741793513298, | |
| "eval_runtime": 4.6321, | |
| "eval_samples_per_second": 3.454, | |
| "eval_steps_per_second": 1.727, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.27972027972027974, | |
| "grad_norm": 1.65625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1101, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 6.375, | |
| "learning_rate": 0.00019997080140801932, | |
| "loss": 0.1327, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3356643356643357, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 0.00019988322268323268, | |
| "loss": 0.0082, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.00019973731496914914, | |
| "loss": 0.0049, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3916083916083916, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 0.00019953316347176488, | |
| "loss": 0.0024, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.4195804195804196, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 0.065, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.44755244755244755, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.0001989506399451051, | |
| "loss": 0.0029, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.4755244755244755, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 0.0001985726080931651, | |
| "loss": 0.0053, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 0.00019813701261394136, | |
| "loss": 0.0056, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "eval_loss": 0.02016551047563553, | |
| "eval_runtime": 4.6628, | |
| "eval_samples_per_second": 3.431, | |
| "eval_steps_per_second": 1.716, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5314685314685315, | |
| "grad_norm": 0.0791015625, | |
| "learning_rate": 0.00019764410788292722, | |
| "loss": 0.003, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.5594405594405595, | |
| "grad_norm": 0.032470703125, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 0.0011, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5874125874125874, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00019648755533435518, | |
| "loss": 0.0445, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 0.00019582458291091663, | |
| "loss": 0.0708, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6433566433566433, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0114, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6713286713286714, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0001943311813257743, | |
| "loss": 0.0243, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6993006993006993, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 0.0655, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.00019261746489577765, | |
| "loss": 0.0756, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "grad_norm": 0.0201416015625, | |
| "learning_rate": 0.00019167921953165825, | |
| "loss": 0.0008, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "eval_loss": 0.0004657781682908535, | |
| "eval_runtime": 4.6398, | |
| "eval_samples_per_second": 3.448, | |
| "eval_steps_per_second": 1.724, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7832167832167832, | |
| "grad_norm": 0.0186767578125, | |
| "learning_rate": 0.00019068743608505455, | |
| "loss": 0.0008, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8111888111888111, | |
| "grad_norm": 0.02392578125, | |
| "learning_rate": 0.00018964269372957038, | |
| "loss": 0.0008, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.8391608391608392, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 0.0046, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8671328671328671, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 0.0001873968032626518, | |
| "loss": 0.0049, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.8951048951048951, | |
| "grad_norm": 0.1220703125, | |
| "learning_rate": 0.00018619696668800492, | |
| "loss": 0.0047, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.0240478515625, | |
| "learning_rate": 0.0001849467935121521, | |
| "loss": 0.0012, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.951048951048951, | |
| "grad_norm": 0.032958984375, | |
| "learning_rate": 0.00018364701380102266, | |
| "loss": 0.0014, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.9790209790209791, | |
| "grad_norm": 0.0247802734375, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 0.001, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.006993006993007, | |
| "grad_norm": 0.01458740234375, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.0006, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.006993006993007, | |
| "eval_loss": 0.00015493688988499343, | |
| "eval_runtime": 4.6565, | |
| "eval_samples_per_second": 3.436, | |
| "eval_steps_per_second": 1.718, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.034965034965035, | |
| "grad_norm": 0.00191497802734375, | |
| "learning_rate": 0.00017945776797137543, | |
| "loss": 0.0001, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.062937062937063, | |
| "grad_norm": 0.00726318359375, | |
| "learning_rate": 0.00017796743540632223, | |
| "loss": 0.0003, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.002288818359375, | |
| "learning_rate": 0.00017643157205458483, | |
| "loss": 0.0001, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.118881118881119, | |
| "grad_norm": 0.0029144287109375, | |
| "learning_rate": 0.00017485107481711012, | |
| "loss": 0.0001, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1468531468531469, | |
| "grad_norm": 0.0009918212890625, | |
| "learning_rate": 0.00017322686665977737, | |
| "loss": 0.0001, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 1.1748251748251748, | |
| "grad_norm": 0.000743865966796875, | |
| "learning_rate": 0.00017155989607441213, | |
| "loss": 0.0, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 1.2027972027972027, | |
| "grad_norm": 0.001007080078125, | |
| "learning_rate": 0.00016985113652489374, | |
| "loss": 0.0001, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 0.002471923828125, | |
| "learning_rate": 0.00016810158587867973, | |
| "loss": 0.0001, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 1.2587412587412588, | |
| "grad_norm": 0.00103759765625, | |
| "learning_rate": 0.00016631226582407952, | |
| "loss": 0.0001, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.2587412587412588, | |
| "eval_loss": 2.9456097763613798e-05, | |
| "eval_runtime": 4.6586, | |
| "eval_samples_per_second": 3.434, | |
| "eval_steps_per_second": 1.717, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.2867132867132867, | |
| "grad_norm": 0.000637054443359375, | |
| "learning_rate": 0.00016448422127361706, | |
| "loss": 0.0, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.3146853146853146, | |
| "grad_norm": 0.000431060791015625, | |
| "learning_rate": 0.00016261851975383137, | |
| "loss": 0.0, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.3426573426573427, | |
| "grad_norm": 0.0020904541015625, | |
| "learning_rate": 0.00016071625078187114, | |
| "loss": 0.0001, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 1.3706293706293706, | |
| "grad_norm": 0.000568389892578125, | |
| "learning_rate": 0.00015877852522924732, | |
| "loss": 0.0, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 1.3986013986013985, | |
| "grad_norm": 0.00135040283203125, | |
| "learning_rate": 0.00015680647467311557, | |
| "loss": 0.0, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4265734265734267, | |
| "grad_norm": 0.000621795654296875, | |
| "learning_rate": 0.00015480125073546704, | |
| "loss": 0.0, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.7578125, | |
| "learning_rate": 0.0001527640244106133, | |
| "loss": 0.0953, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 1.4825174825174825, | |
| "grad_norm": 0.005279541015625, | |
| "learning_rate": 0.00015069598538135906, | |
| "loss": 0.0002, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 1.5104895104895104, | |
| "grad_norm": 0.0101318359375, | |
| "learning_rate": 0.0001485983413242606, | |
| "loss": 0.0004, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5104895104895104, | |
| "eval_loss": 0.0003907401696778834, | |
| "eval_runtime": 4.6412, | |
| "eval_samples_per_second": 3.447, | |
| "eval_steps_per_second": 1.724, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 0.007568359375, | |
| "learning_rate": 0.00014647231720437686, | |
| "loss": 0.0003, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.5664335664335665, | |
| "grad_norm": 0.0167236328125, | |
| "learning_rate": 0.00014431915455992414, | |
| "loss": 0.0007, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 1.5944055944055944, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 0.00014214011077725292, | |
| "loss": 0.0033, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 1.6223776223776225, | |
| "grad_norm": 0.046142578125, | |
| "learning_rate": 0.00013993645835656953, | |
| "loss": 0.0022, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 1.6503496503496504, | |
| "grad_norm": 0.32421875, | |
| "learning_rate": 0.00013770948416883205, | |
| "loss": 0.016, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 1.6783216783216783, | |
| "grad_norm": 0.12060546875, | |
| "learning_rate": 0.00013546048870425356, | |
| "loss": 0.0058, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7062937062937062, | |
| "grad_norm": 0.0341796875, | |
| "learning_rate": 0.00013319078531285285, | |
| "loss": 0.0019, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 1.7342657342657342, | |
| "grad_norm": 0.0166015625, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.001, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.762237762237762, | |
| "grad_norm": 0.01190185546875, | |
| "learning_rate": 0.00012859456783986893, | |
| "loss": 0.0007, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.762237762237762, | |
| "eval_loss": 0.00017181402654387057, | |
| "eval_runtime": 4.6193, | |
| "eval_samples_per_second": 3.464, | |
| "eval_steps_per_second": 1.732, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.7902097902097902, | |
| "grad_norm": 0.005401611328125, | |
| "learning_rate": 0.0001262707378198587, | |
| "loss": 0.0003, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.00157928466796875, | |
| "learning_rate": 0.0001239315664287558, | |
| "loss": 0.0001, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 0.0022125244140625, | |
| "learning_rate": 0.00012157841967678063, | |
| "loss": 0.0002, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.8741258741258742, | |
| "grad_norm": 0.00225830078125, | |
| "learning_rate": 0.00011921267173537086, | |
| "loss": 0.0001, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 1.902097902097902, | |
| "grad_norm": 0.00179290771484375, | |
| "learning_rate": 0.00011683570413470383, | |
| "loss": 0.0001, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 1.93006993006993, | |
| "grad_norm": 0.00188446044921875, | |
| "learning_rate": 0.00011444890495692213, | |
| "loss": 0.0001, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.958041958041958, | |
| "grad_norm": 0.0012054443359375, | |
| "learning_rate": 0.0001120536680255323, | |
| "loss": 0.0001, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.986013986013986, | |
| "grad_norm": 0.0030364990234375, | |
| "learning_rate": 0.00010965139209145152, | |
| "loss": 0.0002, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 2.013986013986014, | |
| "grad_norm": 0.0012359619140625, | |
| "learning_rate": 0.00010724348001617625, | |
| "loss": 0.0001, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.013986013986014, | |
| "eval_loss": 9.037779818754643e-05, | |
| "eval_runtime": 4.5862, | |
| "eval_samples_per_second": 3.489, | |
| "eval_steps_per_second": 1.744, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 2.041958041958042, | |
| "grad_norm": 0.0011444091796875, | |
| "learning_rate": 0.00010483133795255071, | |
| "loss": 0.0001, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 2.06993006993007, | |
| "grad_norm": 0.00121307373046875, | |
| "learning_rate": 0.00010241637452361323, | |
| "loss": 0.0001, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 2.097902097902098, | |
| "grad_norm": 0.002105712890625, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0001, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.125874125874126, | |
| "grad_norm": 0.002227783203125, | |
| "learning_rate": 9.75836254763868e-05, | |
| "loss": 0.0001, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 0.318359375, | |
| "learning_rate": 9.516866204744931e-05, | |
| "loss": 0.0985, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.0015869140625, | |
| "learning_rate": 9.275651998382377e-05, | |
| "loss": 0.0001, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 2.20979020979021, | |
| "grad_norm": 0.00093841552734375, | |
| "learning_rate": 9.034860790854849e-05, | |
| "loss": 0.0001, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 2.237762237762238, | |
| "grad_norm": 0.00142669677734375, | |
| "learning_rate": 8.79463319744677e-05, | |
| "loss": 0.0001, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.265734265734266, | |
| "grad_norm": 0.00189208984375, | |
| "learning_rate": 8.55510950430779e-05, | |
| "loss": 0.0001, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.265734265734266, | |
| "eval_loss": 0.00017878212383948267, | |
| "eval_runtime": 4.6091, | |
| "eval_samples_per_second": 3.471, | |
| "eval_steps_per_second": 1.736, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 2.2937062937062938, | |
| "grad_norm": 0.003875732421875, | |
| "learning_rate": 8.316429586529615e-05, | |
| "loss": 0.0002, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 2.3216783216783217, | |
| "grad_norm": 0.00482177734375, | |
| "learning_rate": 8.078732826462915e-05, | |
| "loss": 0.0003, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 2.3496503496503496, | |
| "grad_norm": 0.00628662109375, | |
| "learning_rate": 7.84215803232194e-05, | |
| "loss": 0.0004, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 2.3776223776223775, | |
| "grad_norm": 0.006072998046875, | |
| "learning_rate": 7.606843357124426e-05, | |
| "loss": 0.0003, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.4055944055944054, | |
| "grad_norm": 0.01068115234375, | |
| "learning_rate": 7.372926218014131e-05, | |
| "loss": 0.0005, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 2.4335664335664333, | |
| "grad_norm": 0.008056640625, | |
| "learning_rate": 7.14054321601311e-05, | |
| "loss": 0.0005, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 0.006927490234375, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.0004, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 2.4895104895104896, | |
| "grad_norm": 0.01263427734375, | |
| "learning_rate": 6.680921468714719e-05, | |
| "loss": 0.0007, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 2.5174825174825175, | |
| "grad_norm": 0.01202392578125, | |
| "learning_rate": 6.453951129574644e-05, | |
| "loss": 0.0006, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5174825174825175, | |
| "eval_loss": 0.00044604684808291495, | |
| "eval_runtime": 4.5475, | |
| "eval_samples_per_second": 3.518, | |
| "eval_steps_per_second": 1.759, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.01318359375, | |
| "learning_rate": 6.229051583116796e-05, | |
| "loss": 0.0007, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 2.5734265734265733, | |
| "grad_norm": 0.01214599609375, | |
| "learning_rate": 6.006354164343046e-05, | |
| "loss": 0.0007, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 2.6013986013986012, | |
| "grad_norm": 0.00799560546875, | |
| "learning_rate": 5.785988922274711e-05, | |
| "loss": 0.0006, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 2.629370629370629, | |
| "grad_norm": 0.0123291015625, | |
| "learning_rate": 5.568084544007588e-05, | |
| "loss": 0.0007, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.6573426573426575, | |
| "grad_norm": 0.011962890625, | |
| "learning_rate": 5.3527682795623146e-05, | |
| "loss": 0.0007, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.6853146853146854, | |
| "grad_norm": 0.007293701171875, | |
| "learning_rate": 5.14016586757394e-05, | |
| "loss": 0.0004, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 2.7132867132867133, | |
| "grad_norm": 0.01287841796875, | |
| "learning_rate": 4.9304014618640995e-05, | |
| "loss": 0.0007, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 2.7412587412587412, | |
| "grad_norm": 0.08154296875, | |
| "learning_rate": 4.723597558938672e-05, | |
| "loss": 0.0011, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 0.0096435546875, | |
| "learning_rate": 4.519874926453302e-05, | |
| "loss": 0.0006, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "eval_loss": 0.00037250021705403924, | |
| "eval_runtime": 4.6476, | |
| "eval_samples_per_second": 3.443, | |
| "eval_steps_per_second": 1.721, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 2.797202797202797, | |
| "grad_norm": 0.00970458984375, | |
| "learning_rate": 4.3193525326884435e-05, | |
| "loss": 0.0005, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.825174825174825, | |
| "grad_norm": 0.007781982421875, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.0004, | |
| "step": 101 | |
| }, | |
| { | |
| "epoch": 2.8531468531468533, | |
| "grad_norm": 0.0108642578125, | |
| "learning_rate": 3.9283749218128885e-05, | |
| "loss": 0.0006, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 2.8811188811188813, | |
| "grad_norm": 0.00811767578125, | |
| "learning_rate": 3.738148024616863e-05, | |
| "loss": 0.0005, | |
| "step": 103 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.007080078125, | |
| "learning_rate": 3.5515778726382966e-05, | |
| "loss": 0.0004, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 2.937062937062937, | |
| "grad_norm": 0.0140380859375, | |
| "learning_rate": 3.36877341759205e-05, | |
| "loss": 0.0008, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.965034965034965, | |
| "grad_norm": 0.01104736328125, | |
| "learning_rate": 3.1898414121320276e-05, | |
| "loss": 0.0007, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 2.993006993006993, | |
| "grad_norm": 0.00726318359375, | |
| "learning_rate": 3.0148863475106314e-05, | |
| "loss": 0.0004, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 3.020979020979021, | |
| "grad_norm": 0.007659912109375, | |
| "learning_rate": 2.84401039255879e-05, | |
| "loss": 0.0005, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 3.020979020979021, | |
| "eval_loss": 0.00027107546338811517, | |
| "eval_runtime": 4.6476, | |
| "eval_samples_per_second": 3.443, | |
| "eval_steps_per_second": 1.721, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 3.0489510489510487, | |
| "grad_norm": 0.006683349609375, | |
| "learning_rate": 2.677313334022268e-05, | |
| "loss": 0.0004, | |
| "step": 109 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 0.0067138671875, | |
| "learning_rate": 2.514892518288988e-05, | |
| "loss": 0.0004, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.104895104895105, | |
| "grad_norm": 0.004974365234375, | |
| "learning_rate": 2.356842794541516e-05, | |
| "loss": 0.0003, | |
| "step": 111 | |
| }, | |
| { | |
| "epoch": 3.132867132867133, | |
| "grad_norm": 0.004119873046875, | |
| "learning_rate": 2.2032564593677774e-05, | |
| "loss": 0.0003, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 3.160839160839161, | |
| "grad_norm": 0.005035400390625, | |
| "learning_rate": 2.0542232028624586e-05, | |
| "loss": 0.0003, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 3.1888111888111887, | |
| "grad_norm": 0.00799560546875, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0005, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 3.2167832167832167, | |
| "grad_norm": 0.00634765625, | |
| "learning_rate": 1.7701613410634365e-05, | |
| "loss": 0.0003, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.2447552447552446, | |
| "grad_norm": 0.00811767578125, | |
| "learning_rate": 1.6352986198977325e-05, | |
| "loss": 0.0004, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.006256103515625, | |
| "learning_rate": 1.5053206487847914e-05, | |
| "loss": 0.0003, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "eval_loss": 0.00024080852745100856, | |
| "eval_runtime": 4.6559, | |
| "eval_samples_per_second": 3.436, | |
| "eval_steps_per_second": 1.718, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 3.300699300699301, | |
| "grad_norm": 0.00762939453125, | |
| "learning_rate": 1.3803033311995072e-05, | |
| "loss": 0.0004, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 3.3286713286713288, | |
| "grad_norm": 0.00933837890625, | |
| "learning_rate": 1.260319673734821e-05, | |
| "loss": 0.0007, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 3.3566433566433567, | |
| "grad_norm": 0.003875732421875, | |
| "learning_rate": 1.1454397434679021e-05, | |
| "loss": 0.0003, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 0.0057373046875, | |
| "learning_rate": 1.0357306270429624e-05, | |
| "loss": 0.0003, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 3.4125874125874125, | |
| "grad_norm": 0.004913330078125, | |
| "learning_rate": 9.31256391494546e-06, | |
| "loss": 0.0003, | |
| "step": 122 | |
| }, | |
| { | |
| "epoch": 3.4405594405594404, | |
| "grad_norm": 0.00592041015625, | |
| "learning_rate": 8.32078046834176e-06, | |
| "loss": 0.0003, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 3.4685314685314683, | |
| "grad_norm": 0.0035247802734375, | |
| "learning_rate": 7.382535104222366e-06, | |
| "loss": 0.0002, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 3.4965034965034967, | |
| "grad_norm": 0.00482177734375, | |
| "learning_rate": 6.498375731458528e-06, | |
| "loss": 0.0002, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.5244755244755246, | |
| "grad_norm": 0.006683349609375, | |
| "learning_rate": 5.668818674225685e-06, | |
| "loss": 0.0004, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.5244755244755246, | |
| "eval_loss": 0.00024340736854355782, | |
| "eval_runtime": 4.6706, | |
| "eval_samples_per_second": 3.426, | |
| "eval_steps_per_second": 1.713, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 3.5524475524475525, | |
| "grad_norm": 0.007049560546875, | |
| "learning_rate": 4.8943483704846475e-06, | |
| "loss": 0.0004, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 3.5804195804195804, | |
| "grad_norm": 0.005950927734375, | |
| "learning_rate": 4.175417089083378e-06, | |
| "loss": 0.0003, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 3.6083916083916083, | |
| "grad_norm": 0.004486083984375, | |
| "learning_rate": 3.512444665644865e-06, | |
| "loss": 0.0003, | |
| "step": 129 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.02294921875, | |
| "learning_rate": 2.905818257394799e-06, | |
| "loss": 0.0006, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.664335664335664, | |
| "grad_norm": 0.006805419921875, | |
| "learning_rate": 2.3558921170727888e-06, | |
| "loss": 0.0004, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 0.006622314453125, | |
| "learning_rate": 1.8629873860586566e-06, | |
| "loss": 0.0004, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 3.7202797202797204, | |
| "grad_norm": 0.220703125, | |
| "learning_rate": 1.4273919068349184e-06, | |
| "loss": 0.0492, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 3.7482517482517483, | |
| "grad_norm": 0.0050048828125, | |
| "learning_rate": 1.0493600548948878e-06, | |
| "loss": 0.0003, | |
| "step": 134 | |
| }, | |
| { | |
| "epoch": 3.7762237762237763, | |
| "grad_norm": 0.00958251953125, | |
| "learning_rate": 7.291125901946027e-07, | |
| "loss": 0.0006, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.7762237762237763, | |
| "eval_loss": 0.0002461140393279493, | |
| "eval_runtime": 4.6249, | |
| "eval_samples_per_second": 3.46, | |
| "eval_steps_per_second": 1.73, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.804195804195804, | |
| "grad_norm": 0.004302978515625, | |
| "learning_rate": 4.668365282351372e-07, | |
| "loss": 0.0002, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 3.832167832167832, | |
| "grad_norm": 0.0074462890625, | |
| "learning_rate": 2.6268503085089547e-07, | |
| "loss": 0.0004, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 3.86013986013986, | |
| "grad_norm": 0.0048828125, | |
| "learning_rate": 1.1677731676733584e-07, | |
| "loss": 0.0003, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 3.8881118881118883, | |
| "grad_norm": 0.0048828125, | |
| "learning_rate": 2.9198591980705848e-08, | |
| "loss": 0.0003, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 3.916083916083916, | |
| "grad_norm": 0.0057373046875, | |
| "learning_rate": 0.0, | |
| "loss": 0.0003, | |
| "step": 140 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 35, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.083234187669668e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |