ali-elganzory/Qwen2.5-1.5B-SFT-Tulu3-decontaminated
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
7a8ebedfa194be15067a2b20f96a40e1bc1b23e4
Qwen2.5-1.5B-SFT-Tulu3-deco…/trainer_state.json
ModelHub XC 7a8ebedfa1 初始化项目，由ModelHub XC社区提供模型 
Model: ali-elganzory/Qwen2.5-1.5B-SFT-Tulu3-decontaminated
Source: Original Platform
2026-08-21 02:14:18 +08:00

14677 lines
414 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 14634,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 1.133642578125,
"epoch": 0.0013667737306088978,
"grad_norm": 3.4280447300326284,
"learning_rate": 1.0227272727272728e-07,
"loss": 1.2075,
"mean_token_accuracy": 0.7250663354992867,
"num_tokens": 894751.0,
"step": 10
},
{
"entropy": 1.105810546875,
"epoch": 0.0027335474612177955,
"grad_norm": 3.618605021752981,
"learning_rate": 2.1590909090909094e-07,
"loss": 1.1497,
"mean_token_accuracy": 0.7343597397208214,
"num_tokens": 1750308.0,
"step": 20
},
{
"entropy": 1.102099609375,
"epoch": 0.004100321191826693,
"grad_norm": 3.279639558021943,
"learning_rate": 3.2954545454545455e-07,
"loss": 1.1919,
"mean_token_accuracy": 0.7304547049105168,
"num_tokens": 2644413.0,
"step": 30
},
{
"entropy": 1.14296875,
"epoch": 0.005467094922435591,
"grad_norm": 3.1091859166894102,
"learning_rate": 4.431818181818182e-07,
"loss": 1.203,
"mean_token_accuracy": 0.7255290724337101,
"num_tokens": 3517264.0,
"step": 40
},
{
"entropy": 1.079052734375,
"epoch": 0.006833868653044488,
"grad_norm": 3.068930414193493,
"learning_rate": 5.568181818181818e-07,
"loss": 1.1212,
"mean_token_accuracy": 0.7394589193165302,
"num_tokens": 4410342.0,
"step": 50
},
{
"entropy": 1.1357421875,
"epoch": 0.008200642383653386,
"grad_norm": 2.0308832715086114,
"learning_rate": 6.704545454545456e-07,
"loss": 1.1509,
"mean_token_accuracy": 0.7299226187169552,
"num_tokens": 5335473.0,
"step": 60
},
{
"entropy": 1.182080078125,
"epoch": 0.009567416114262284,
"grad_norm": 2.300558552510473,
"learning_rate": 7.840909090909092e-07,
"loss": 1.199,
"mean_token_accuracy": 0.724257380515337,
"num_tokens": 6259118.0,
"step": 70
},
{
"entropy": 1.21337890625,
"epoch": 0.010934189844871182,
"grad_norm": 1.6465800515432338,
"learning_rate": 8.977272727272728e-07,
"loss": 1.1907,
"mean_token_accuracy": 0.7225596114993096,
"num_tokens": 7173208.0,
"step": 80
},
{
"entropy": 1.136572265625,
"epoch": 0.01230096357548008,
"grad_norm": 0.9418042958975457,
"learning_rate": 1.0113636363636365e-06,
"loss": 1.1056,
"mean_token_accuracy": 0.7350179180502892,
"num_tokens": 8004347.0,
"step": 90
},
{
"entropy": 1.099169921875,
"epoch": 0.013667737306088976,
"grad_norm": 0.732422404299948,
"learning_rate": 1.125e-06,
"loss": 1.0822,
"mean_token_accuracy": 0.7443214669823647,
"num_tokens": 8901160.0,
"step": 100
},
{
"entropy": 1.074267578125,
"epoch": 0.015034511036697874,
"grad_norm": 0.7009746602713247,
"learning_rate": 1.2386363636363638e-06,
"loss": 1.0556,
"mean_token_accuracy": 0.744538938254118,
"num_tokens": 9797773.0,
"step": 110
},
{
"entropy": 1.081201171875,
"epoch": 0.016401284767306772,
"grad_norm": 0.7524401106687102,
"learning_rate": 1.3522727272727273e-06,
"loss": 1.0776,
"mean_token_accuracy": 0.7413589105010032,
"num_tokens": 10670680.0,
"step": 120
},
{
"entropy": 1.07763671875,
"epoch": 0.01776805849791567,
"grad_norm": 0.7013391023866667,
"learning_rate": 1.465909090909091e-06,
"loss": 1.0527,
"mean_token_accuracy": 0.7439882136881352,
"num_tokens": 11561633.0,
"step": 130
},
{
"entropy": 1.085546875,
"epoch": 0.019134832228524568,
"grad_norm": 0.6823191333216855,
"learning_rate": 1.5795454545454547e-06,
"loss": 1.0786,
"mean_token_accuracy": 0.7401677779853344,
"num_tokens": 12473967.0,
"step": 140
},
{
"entropy": 1.090234375,
"epoch": 0.020501605959133466,
"grad_norm": 0.7158529039643341,
"learning_rate": 1.6931818181818182e-06,
"loss": 1.0743,
"mean_token_accuracy": 0.7422112219035626,
"num_tokens": 13386639.0,
"step": 150
},
{
"entropy": 1.073095703125,
"epoch": 0.021868379689742364,
"grad_norm": 0.7553636454160204,
"learning_rate": 1.8068181818181822e-06,
"loss": 1.0743,
"mean_token_accuracy": 0.7395212046802044,
"num_tokens": 14284815.0,
"step": 160
},
{
"entropy": 1.079833984375,
"epoch": 0.023235153420351262,
"grad_norm": 0.7827628195344934,
"learning_rate": 1.9204545454545457e-06,
"loss": 1.0692,
"mean_token_accuracy": 0.7428514145314693,
"num_tokens": 15109817.0,
"step": 170
},
{
"entropy": 1.02900390625,
"epoch": 0.02460192715096016,
"grad_norm": 0.6353356332992186,
"learning_rate": 2.034090909090909e-06,
"loss": 1.0027,
"mean_token_accuracy": 0.7537418395280838,
"num_tokens": 15983614.0,
"step": 180
},
{
"entropy": 1.06396484375,
"epoch": 0.025968700881569055,
"grad_norm": 0.6693921277678265,
"learning_rate": 2.147727272727273e-06,
"loss": 1.0561,
"mean_token_accuracy": 0.742389404028654,
"num_tokens": 16896096.0,
"step": 190
},
{
"entropy": 1.0849609375,
"epoch": 0.027335474612177953,
"grad_norm": 0.7278967595469739,
"learning_rate": 2.2613636363636366e-06,
"loss": 1.075,
"mean_token_accuracy": 0.7423841014504433,
"num_tokens": 17736266.0,
"step": 200
},
{
"entropy": 1.072021484375,
"epoch": 0.02870224834278685,
"grad_norm": 0.683426513050328,
"learning_rate": 2.375e-06,
"loss": 1.0626,
"mean_token_accuracy": 0.7420915260910987,
"num_tokens": 18638867.0,
"step": 210
},
{
"entropy": 1.093212890625,
"epoch": 0.03006902207339575,
"grad_norm": 0.7561285039409108,
"learning_rate": 2.488636363636364e-06,
"loss": 1.0672,
"mean_token_accuracy": 0.7425928525626659,
"num_tokens": 19534482.0,
"step": 220
},
{
"entropy": 1.01953125,
"epoch": 0.03143579580400465,
"grad_norm": 0.7205161200379568,
"learning_rate": 2.6022727272727276e-06,
"loss": 1.0077,
"mean_token_accuracy": 0.7540885612368584,
"num_tokens": 20445734.0,
"step": 230
},
{
"entropy": 1.055126953125,
"epoch": 0.032802569534613545,
"grad_norm": 0.678425215385315,
"learning_rate": 2.715909090909091e-06,
"loss": 1.035,
"mean_token_accuracy": 0.7469099894165993,
"num_tokens": 21340398.0,
"step": 240
},
{
"entropy": 1.04482421875,
"epoch": 0.03416934326522244,
"grad_norm": 0.70398390251816,
"learning_rate": 2.829545454545455e-06,
"loss": 1.0177,
"mean_token_accuracy": 0.7522810809314251,
"num_tokens": 22231277.0,
"step": 250
},
{
"entropy": 1.0083984375,
"epoch": 0.03553611699583134,
"grad_norm": 0.6622783470902116,
"learning_rate": 2.9431818181818185e-06,
"loss": 0.9976,
"mean_token_accuracy": 0.7571641318500042,
"num_tokens": 23128440.0,
"step": 260
},
{
"entropy": 1.0162109375,
"epoch": 0.036902890726440235,
"grad_norm": 0.6301165491426336,
"learning_rate": 3.056818181818182e-06,
"loss": 1.001,
"mean_token_accuracy": 0.7560463979840278,
"num_tokens": 24023611.0,
"step": 270
},
{
"entropy": 1.013232421875,
"epoch": 0.038269664457049136,
"grad_norm": 0.6423948758646932,
"learning_rate": 3.1704545454545456e-06,
"loss": 0.9941,
"mean_token_accuracy": 0.7537274546921253,
"num_tokens": 24931447.0,
"step": 280
},
{
"entropy": 1.0454345703125,
"epoch": 0.03963643818765803,
"grad_norm": 0.737561945502292,
"learning_rate": 3.2840909090909095e-06,
"loss": 1.026,
"mean_token_accuracy": 0.7486627273261547,
"num_tokens": 25822637.0,
"step": 290
},
{
"entropy": 1.0248046875,
"epoch": 0.04100321191826693,
"grad_norm": 0.6919526496124788,
"learning_rate": 3.397727272727273e-06,
"loss": 1.0143,
"mean_token_accuracy": 0.7536938957870006,
"num_tokens": 26705243.0,
"step": 300
},
{
"entropy": 1.055517578125,
"epoch": 0.04236998564887583,
"grad_norm": 0.7325215814036441,
"learning_rate": 3.5113636363636365e-06,
"loss": 1.0367,
"mean_token_accuracy": 0.7486682303249836,
"num_tokens": 27620336.0,
"step": 310
},
{
"entropy": 1.04072265625,
"epoch": 0.04373675937948473,
"grad_norm": 0.7265573617355544,
"learning_rate": 3.625e-06,
"loss": 1.0062,
"mean_token_accuracy": 0.7492509938776493,
"num_tokens": 28487030.0,
"step": 320
},
{
"entropy": 1.096337890625,
"epoch": 0.04510353311009362,
"grad_norm": 0.6922672171619181,
"learning_rate": 3.7386363636363635e-06,
"loss": 1.0873,
"mean_token_accuracy": 0.7375812739133835,
"num_tokens": 29362700.0,
"step": 330
},
{
"entropy": 0.9984375,
"epoch": 0.046470306840702524,
"grad_norm": 0.6938911522862108,
"learning_rate": 3.852272727272728e-06,
"loss": 0.9692,
"mean_token_accuracy": 0.758883186429739,
"num_tokens": 30216448.0,
"step": 340
},
{
"entropy": 0.98544921875,
"epoch": 0.04783708057131142,
"grad_norm": 0.6823106957043815,
"learning_rate": 3.965909090909091e-06,
"loss": 0.9506,
"mean_token_accuracy": 0.7612798042595387,
"num_tokens": 31124761.0,
"step": 350
},
{
"entropy": 1.050927734375,
"epoch": 0.04920385430192032,
"grad_norm": 0.5969178231377715,
"learning_rate": 4.079545454545455e-06,
"loss": 1.0253,
"mean_token_accuracy": 0.7488497033715248,
"num_tokens": 32021822.0,
"step": 360
},
{
"entropy": 1.049560546875,
"epoch": 0.050570628032529215,
"grad_norm": 0.7147920647409154,
"learning_rate": 4.193181818181819e-06,
"loss": 1.0362,
"mean_token_accuracy": 0.7463022604584694,
"num_tokens": 32911869.0,
"step": 370
},
{
"entropy": 1.047265625,
"epoch": 0.05193740176313811,
"grad_norm": 0.6884093050287405,
"learning_rate": 4.306818181818182e-06,
"loss": 1.0534,
"mean_token_accuracy": 0.7449572272598743,
"num_tokens": 33780250.0,
"step": 380
},
{
"entropy": 1.001416015625,
"epoch": 0.05330417549374701,
"grad_norm": 0.5995953083203266,
"learning_rate": 4.420454545454546e-06,
"loss": 1.0093,
"mean_token_accuracy": 0.7547092474997044,
"num_tokens": 34694542.0,
"step": 390
},
{
"entropy": 0.991552734375,
"epoch": 0.054670949224355905,
"grad_norm": 0.6372896340773203,
"learning_rate": 4.53409090909091e-06,
"loss": 0.9619,
"mean_token_accuracy": 0.7613359495997429,
"num_tokens": 35546961.0,
"step": 400
},
{
"entropy": 1.06650390625,
"epoch": 0.05603772295496481,
"grad_norm": 0.7391451599408864,
"learning_rate": 4.647727272727273e-06,
"loss": 1.0589,
"mean_token_accuracy": 0.7414283603429794,
"num_tokens": 36410375.0,
"step": 410
},
{
"entropy": 0.99140625,
"epoch": 0.0574044966855737,
"grad_norm": 0.6749365506563596,
"learning_rate": 4.761363636363637e-06,
"loss": 1.0052,
"mean_token_accuracy": 0.7557721860706806,
"num_tokens": 37311509.0,
"step": 420
},
{
"entropy": 0.987939453125,
"epoch": 0.0587712704161826,
"grad_norm": 0.66603376862157,
"learning_rate": 4.875e-06,
"loss": 0.9935,
"mean_token_accuracy": 0.7584005773067475,
"num_tokens": 38219048.0,
"step": 430
},
{
"entropy": 1.027978515625,
"epoch": 0.0601380441467915,
"grad_norm": 0.622056898105799,
"learning_rate": 4.988636363636364e-06,
"loss": 1.0181,
"mean_token_accuracy": 0.7506954029202462,
"num_tokens": 39127651.0,
"step": 440
},
{
"entropy": 1.035791015625,
"epoch": 0.0615048178774004,
"grad_norm": 0.7266670384580952,
"learning_rate": 4.996829646329435e-06,
"loss": 1.0317,
"mean_token_accuracy": 0.7445542335510253,
"num_tokens": 40015164.0,
"step": 450
},
{
"entropy": 1.067333984375,
"epoch": 0.0628715916080093,
"grad_norm": 0.6507103933714492,
"learning_rate": 4.993307031139919e-06,
"loss": 1.0765,
"mean_token_accuracy": 0.7405419059097766,
"num_tokens": 40925068.0,
"step": 460
},
{
"entropy": 1.03994140625,
"epoch": 0.0642383653386182,
"grad_norm": 0.61347766185431,
"learning_rate": 4.989784415950402e-06,
"loss": 1.0417,
"mean_token_accuracy": 0.7464969284832478,
"num_tokens": 41825875.0,
"step": 470
},
{
"entropy": 1.066455078125,
"epoch": 0.06560513906922709,
"grad_norm": 0.6650563232196003,
"learning_rate": 4.986261800760885e-06,
"loss": 1.0773,
"mean_token_accuracy": 0.737893583625555,
"num_tokens": 42703080.0,
"step": 480
},
{
"entropy": 1.0138671875,
"epoch": 0.06697191279983598,
"grad_norm": 0.6458096740287724,
"learning_rate": 4.9827391855713685e-06,
"loss": 1.0114,
"mean_token_accuracy": 0.7538170523941516,
"num_tokens": 43616920.0,
"step": 490
},
{
"entropy": 1.016259765625,
"epoch": 0.06833868653044488,
"grad_norm": 0.6950132567295455,
"learning_rate": 4.979216570381852e-06,
"loss": 1.0137,
"mean_token_accuracy": 0.7486440531909466,
"num_tokens": 44506039.0,
"step": 500
},
{
"entropy": 1.019189453125,
"epoch": 0.06970546026105379,
"grad_norm": 0.6356971178171562,
"learning_rate": 4.975693955192335e-06,
"loss": 1.0297,
"mean_token_accuracy": 0.7461608953773975,
"num_tokens": 45382350.0,
"step": 510
},
{
"entropy": 1.0373046875,
"epoch": 0.07107223399166268,
"grad_norm": 0.7111605823383721,
"learning_rate": 4.972171340002819e-06,
"loss": 1.0404,
"mean_token_accuracy": 0.7474403403699398,
"num_tokens": 46304368.0,
"step": 520
},
{
"entropy": 1.036865234375,
"epoch": 0.07243900772227158,
"grad_norm": 0.6512464219026485,
"learning_rate": 4.968648724813302e-06,
"loss": 1.031,
"mean_token_accuracy": 0.7461777545511723,
"num_tokens": 47189558.0,
"step": 530
},
{
"entropy": 1.0109130859375,
"epoch": 0.07380578145288047,
"grad_norm": 0.6668342294433599,
"learning_rate": 4.965126109623785e-06,
"loss": 1.0212,
"mean_token_accuracy": 0.7501662321388721,
"num_tokens": 48062199.0,
"step": 540
},
{
"entropy": 1.04189453125,
"epoch": 0.07517255518348938,
"grad_norm": 0.6545223473453724,
"learning_rate": 4.961603494434268e-06,
"loss": 1.0589,
"mean_token_accuracy": 0.7449680734425783,
"num_tokens": 48939443.0,
"step": 550
},
{
"entropy": 0.9900390625,
"epoch": 0.07653932891409827,
"grad_norm": 0.6829121933634417,
"learning_rate": 4.958080879244752e-06,
"loss": 0.9957,
"mean_token_accuracy": 0.7546392634510994,
"num_tokens": 49849066.0,
"step": 560
},
{
"entropy": 1.001611328125,
"epoch": 0.07790610264470717,
"grad_norm": 0.7605167675069774,
"learning_rate": 4.954558264055234e-06,
"loss": 0.9916,
"mean_token_accuracy": 0.7543714001774788,
"num_tokens": 50698168.0,
"step": 570
},
{
"entropy": 0.976806640625,
"epoch": 0.07927287637531606,
"grad_norm": 0.7264787971452071,
"learning_rate": 4.951035648865719e-06,
"loss": 0.966,
"mean_token_accuracy": 0.7579257391393185,
"num_tokens": 51559112.0,
"step": 580
},
{
"entropy": 0.995849609375,
"epoch": 0.08063965010592497,
"grad_norm": 0.6659640932613746,
"learning_rate": 4.9475130336762015e-06,
"loss": 1.009,
"mean_token_accuracy": 0.7532717436552048,
"num_tokens": 52433072.0,
"step": 590
},
{
"entropy": 1.038037109375,
"epoch": 0.08200642383653386,
"grad_norm": 0.6853337527954826,
"learning_rate": 4.943990418486685e-06,
"loss": 1.0505,
"mean_token_accuracy": 0.7430377215147018,
"num_tokens": 53357598.0,
"step": 600
},
{
"entropy": 1.01689453125,
"epoch": 0.08337319756714276,
"grad_norm": 0.661700690310068,
"learning_rate": 4.9404678032971685e-06,
"loss": 1.0063,
"mean_token_accuracy": 0.7542144954204559,
"num_tokens": 54234701.0,
"step": 610
},
{
"entropy": 0.995556640625,
"epoch": 0.08473997129775165,
"grad_norm": 0.6658696294129598,
"learning_rate": 4.936945188107651e-06,
"loss": 0.9788,
"mean_token_accuracy": 0.7603907085955143,
"num_tokens": 55143442.0,
"step": 620
},
{
"entropy": 0.94990234375,
"epoch": 0.08610674502836055,
"grad_norm": 0.6633196896539785,
"learning_rate": 4.933422572918135e-06,
"loss": 0.9259,
"mean_token_accuracy": 0.7676366038620472,
"num_tokens": 56014638.0,
"step": 630
},
{
"entropy": 0.9905517578125,
"epoch": 0.08747351875896946,
"grad_norm": 0.5928308113986804,
"learning_rate": 4.929899957728618e-06,
"loss": 0.9766,
"mean_token_accuracy": 0.7583583705127239,
"num_tokens": 56919559.0,
"step": 640
},
{
"entropy": 1.011767578125,
"epoch": 0.08884029248957835,
"grad_norm": 0.6992985119498211,
"learning_rate": 4.926377342539102e-06,
"loss": 0.9866,
"mean_token_accuracy": 0.7537693306803703,
"num_tokens": 57796634.0,
"step": 650
},
{
"entropy": 0.990576171875,
"epoch": 0.09020706622018725,
"grad_norm": 0.84677226436278,
"learning_rate": 4.922854727349585e-06,
"loss": 0.9809,
"mean_token_accuracy": 0.7564093492925167,
"num_tokens": 58681705.0,
"step": 660
},
{
"entropy": 0.999560546875,
"epoch": 0.09157383995079614,
"grad_norm": 0.6310808892953602,
"learning_rate": 4.919332112160068e-06,
"loss": 1.002,
"mean_token_accuracy": 0.7512515589594841,
"num_tokens": 59598109.0,
"step": 670
},
{
"entropy": 1.0138671875,
"epoch": 0.09294061368140505,
"grad_norm": 0.6650078251961363,
"learning_rate": 4.915809496970551e-06,
"loss": 1.0022,
"mean_token_accuracy": 0.7517239972949028,
"num_tokens": 60505590.0,
"step": 680
},
{
"entropy": 1.073583984375,
"epoch": 0.09430738741201394,
"grad_norm": 0.6318918262316185,
"learning_rate": 4.912286881781035e-06,
"loss": 1.1004,
"mean_token_accuracy": 0.7370693854987621,
"num_tokens": 61382600.0,
"step": 690
},
{
"entropy": 0.993701171875,
"epoch": 0.09567416114262284,
"grad_norm": 0.665954383221238,
"learning_rate": 4.908764266591518e-06,
"loss": 0.9817,
"mean_token_accuracy": 0.7537538498640061,
"num_tokens": 62258121.0,
"step": 700
},
{
"entropy": 1.00068359375,
"epoch": 0.09704093487323173,
"grad_norm": 0.6464780711551599,
"learning_rate": 4.9052416514020015e-06,
"loss": 0.9956,
"mean_token_accuracy": 0.7542749583721161,
"num_tokens": 63128686.0,
"step": 710
},
{
"entropy": 0.9648193359375,
"epoch": 0.09840770860384064,
"grad_norm": 0.6420618142645999,
"learning_rate": 4.901719036212484e-06,
"loss": 0.9662,
"mean_token_accuracy": 0.758437330275774,
"num_tokens": 64056881.0,
"step": 720
},
{
"entropy": 0.9724609375,
"epoch": 0.09977448233444954,
"grad_norm": 0.6404679000876007,
"learning_rate": 4.898196421022968e-06,
"loss": 0.9698,
"mean_token_accuracy": 0.7595405861735344,
"num_tokens": 64941116.0,
"step": 730
},
{
"entropy": 0.985400390625,
"epoch": 0.10114125606505843,
"grad_norm": 0.7066876483927433,
"learning_rate": 4.894673805833451e-06,
"loss": 0.9506,
"mean_token_accuracy": 0.7600325047969818,
"num_tokens": 65760286.0,
"step": 740
},
{
"entropy": 1.011474609375,
"epoch": 0.10250802979566732,
"grad_norm": 0.6408141335348894,
"learning_rate": 4.891151190643935e-06,
"loss": 0.9954,
"mean_token_accuracy": 0.7563378445804119,
"num_tokens": 66641601.0,
"step": 750
},
{
"entropy": 1.003662109375,
"epoch": 0.10387480352627622,
"grad_norm": 0.6290647649349624,
"learning_rate": 4.8876285754544175e-06,
"loss": 0.9926,
"mean_token_accuracy": 0.7534885846078396,
"num_tokens": 67556757.0,
"step": 760
},
{
"entropy": 1.040576171875,
"epoch": 0.10524157725688513,
"grad_norm": 0.71773040238029,
"learning_rate": 4.884105960264901e-06,
"loss": 1.0423,
"mean_token_accuracy": 0.7478190787136555,
"num_tokens": 68415204.0,
"step": 770
},
{
"entropy": 0.986279296875,
"epoch": 0.10660835098749402,
"grad_norm": 0.5956745887521789,
"learning_rate": 4.880583345075385e-06,
"loss": 0.9783,
"mean_token_accuracy": 0.7549893386662007,
"num_tokens": 69305673.0,
"step": 780
},
{
"entropy": 0.986474609375,
"epoch": 0.10797512471810292,
"grad_norm": 0.6181589135335007,
"learning_rate": 4.877060729885867e-06,
"loss": 0.9648,
"mean_token_accuracy": 0.7588886693120003,
"num_tokens": 70187437.0,
"step": 790
},
{
"entropy": 0.98427734375,
"epoch": 0.10934189844871181,
"grad_norm": 0.6818484296225196,
"learning_rate": 4.873538114696351e-06,
"loss": 0.9686,
"mean_token_accuracy": 0.7572630323469639,
"num_tokens": 71066843.0,
"step": 800
},
{
"entropy": 0.9986328125,
"epoch": 0.11070867217932072,
"grad_norm": 0.6717035575916456,
"learning_rate": 4.870015499506834e-06,
"loss": 0.9862,
"mean_token_accuracy": 0.7578598625957966,
"num_tokens": 71952220.0,
"step": 810
},
{
"entropy": 0.975830078125,
"epoch": 0.11207544590992961,
"grad_norm": 0.5767568485399869,
"learning_rate": 4.866492884317318e-06,
"loss": 0.9768,
"mean_token_accuracy": 0.7558931395411491,
"num_tokens": 72848678.0,
"step": 820
},
{
"entropy": 0.966455078125,
"epoch": 0.11344221964053851,
"grad_norm": 0.6923334648291468,
"learning_rate": 4.862970269127801e-06,
"loss": 0.9524,
"mean_token_accuracy": 0.7647132948040962,
"num_tokens": 73722522.0,
"step": 830
},
{
"entropy": 1.027587890625,
"epoch": 0.1148089933711474,
"grad_norm": 0.6238126661350346,
"learning_rate": 4.859447653938284e-06,
"loss": 0.9941,
"mean_token_accuracy": 0.7510923549532891,
"num_tokens": 74585045.0,
"step": 840
},
{
"entropy": 1.000439453125,
"epoch": 0.11617576710175631,
"grad_norm": 0.6272664632512938,
"learning_rate": 4.855925038748768e-06,
"loss": 1.0064,
"mean_token_accuracy": 0.7547958835959434,
"num_tokens": 75542861.0,
"step": 850
},
{
"entropy": 0.995751953125,
"epoch": 0.1175425408323652,
"grad_norm": 0.674856233875799,
"learning_rate": 4.852402423559251e-06,
"loss": 1.0037,
"mean_token_accuracy": 0.7548098973929882,
"num_tokens": 76472020.0,
"step": 860
},
{
"entropy": 1.029541015625,
"epoch": 0.1189093145629741,
"grad_norm": 0.628460515228581,
"learning_rate": 4.848879808369734e-06,
"loss": 1.0282,
"mean_token_accuracy": 0.7497413389384746,
"num_tokens": 77350339.0,
"step": 870
},
{
"entropy": 0.981689453125,
"epoch": 0.120276088293583,
"grad_norm": 0.6143284931439743,
"learning_rate": 4.8453571931802175e-06,
"loss": 0.9524,
"mean_token_accuracy": 0.7593360356986523,
"num_tokens": 78257082.0,
"step": 880
},
{
"entropy": 0.94912109375,
"epoch": 0.12164286202419189,
"grad_norm": 0.6561308825731347,
"learning_rate": 4.8418345779907e-06,
"loss": 0.9387,
"mean_token_accuracy": 0.7655358709394932,
"num_tokens": 79122216.0,
"step": 890
},
{
"entropy": 1.01923828125,
"epoch": 0.1230096357548008,
"grad_norm": 0.6595750925430152,
"learning_rate": 4.838311962801184e-06,
"loss": 1.026,
"mean_token_accuracy": 0.7501997977495194,
"num_tokens": 80032714.0,
"step": 900
},
{
"entropy": 0.996826171875,
"epoch": 0.12437640948540969,
"grad_norm": 0.6726760872748547,
"learning_rate": 4.834789347611667e-06,
"loss": 0.9805,
"mean_token_accuracy": 0.756825202703476,
"num_tokens": 80873820.0,
"step": 910
},
{
"entropy": 1.00732421875,
"epoch": 0.1257431832160186,
"grad_norm": 0.6885710708021324,
"learning_rate": 4.831266732422151e-06,
"loss": 0.9896,
"mean_token_accuracy": 0.7540356196463108,
"num_tokens": 81719615.0,
"step": 920
},
{
"entropy": 0.957470703125,
"epoch": 0.1271099569466275,
"grad_norm": 0.6547966490336906,
"learning_rate": 4.827744117232634e-06,
"loss": 0.9176,
"mean_token_accuracy": 0.7669717475771904,
"num_tokens": 82585299.0,
"step": 930
},
{
"entropy": 0.979541015625,
"epoch": 0.1284767306772364,
"grad_norm": 0.7042728625707763,
"learning_rate": 4.824221502043117e-06,
"loss": 0.9515,
"mean_token_accuracy": 0.7593542210757732,
"num_tokens": 83450499.0,
"step": 940
},
{
"entropy": 1.010107421875,
"epoch": 0.12984350440784528,
"grad_norm": 0.618233444510619,
"learning_rate": 4.820698886853601e-06,
"loss": 1.008,
"mean_token_accuracy": 0.7496823944151402,
"num_tokens": 84384201.0,
"step": 950
},
{
"entropy": 1.015625,
"epoch": 0.13121027813845418,
"grad_norm": 0.690455417601858,
"learning_rate": 4.817176271664084e-06,
"loss": 1.0063,
"mean_token_accuracy": 0.7510360911488533,
"num_tokens": 85282609.0,
"step": 960
},
{
"entropy": 0.955126953125,
"epoch": 0.13257705186906307,
"grad_norm": 0.6245802196145479,
"learning_rate": 4.813653656474567e-06,
"loss": 0.9369,
"mean_token_accuracy": 0.7640732683241367,
"num_tokens": 86130389.0,
"step": 970
},
{
"entropy": 1.0162109375,
"epoch": 0.13394382559967197,
"grad_norm": 0.732702775300104,
"learning_rate": 4.8101310412850505e-06,
"loss": 1.0263,
"mean_token_accuracy": 0.7495774812996387,
"num_tokens": 87013196.0,
"step": 980
},
{
"entropy": 0.991064453125,
"epoch": 0.13531059933028086,
"grad_norm": 0.9505768339434075,
"learning_rate": 4.806608426095534e-06,
"loss": 0.9812,
"mean_token_accuracy": 0.7597615152597428,
"num_tokens": 87940418.0,
"step": 990
},
{
"entropy": 0.959716796875,
"epoch": 0.13667737306088976,
"grad_norm": 0.5685320882987133,
"learning_rate": 4.803085810906017e-06,
"loss": 0.9429,
"mean_token_accuracy": 0.7627806380391121,
"num_tokens": 88835060.0,
"step": 1000
},
{
"entropy": 1.020703125,
"epoch": 0.13804414679149868,
"grad_norm": 1.0180117973938425,
"learning_rate": 4.7995631957165e-06,
"loss": 1.0416,
"mean_token_accuracy": 0.7433593206107616,
"num_tokens": 89750236.0,
"step": 1010
},
{
"entropy": 0.93583984375,
"epoch": 0.13941092052210757,
"grad_norm": 0.7853204723408518,
"learning_rate": 4.796040580526984e-06,
"loss": 0.9285,
"mean_token_accuracy": 0.7681406021118165,
"num_tokens": 90644061.0,
"step": 1020
},
{
"entropy": 0.972119140625,
"epoch": 0.14077769425271647,
"grad_norm": 0.7336607815657837,
"learning_rate": 4.792517965337467e-06,
"loss": 0.979,
"mean_token_accuracy": 0.7578564211726189,
"num_tokens": 91490004.0,
"step": 1030
},
{
"entropy": 0.959375,
"epoch": 0.14214446798332536,
"grad_norm": 0.6687744844547743,
"learning_rate": 4.78899535014795e-06,
"loss": 0.9783,
"mean_token_accuracy": 0.7606046959757805,
"num_tokens": 92377049.0,
"step": 1040
},
{
"entropy": 0.96796875,
"epoch": 0.14351124171393426,
"grad_norm": 0.618494208680913,
"learning_rate": 4.785472734958434e-06,
"loss": 0.9651,
"mean_token_accuracy": 0.760289304703474,
"num_tokens": 93288203.0,
"step": 1050
},
{
"entropy": 0.976708984375,
"epoch": 0.14487801544454315,
"grad_norm": 0.6673814579299123,
"learning_rate": 4.781950119768916e-06,
"loss": 0.9901,
"mean_token_accuracy": 0.7575847610831261,
"num_tokens": 94185325.0,
"step": 1060
},
{
"entropy": 0.934716796875,
"epoch": 0.14624478917515205,
"grad_norm": 0.6353615194450685,
"learning_rate": 4.778427504579401e-06,
"loss": 0.9077,
"mean_token_accuracy": 0.7675634264945984,
"num_tokens": 95085041.0,
"step": 1070
},
{
"entropy": 0.972802734375,
"epoch": 0.14761156290576094,
"grad_norm": 0.5724282817478571,
"learning_rate": 4.774904889389883e-06,
"loss": 0.9409,
"mean_token_accuracy": 0.7639177329838276,
"num_tokens": 96001882.0,
"step": 1080
},
{
"entropy": 0.975634765625,
"epoch": 0.14897833663636986,
"grad_norm": 0.6964750584426768,
"learning_rate": 4.771382274200367e-06,
"loss": 0.9523,
"mean_token_accuracy": 0.7624265760183334,
"num_tokens": 96888127.0,
"step": 1090
},
{
"entropy": 1.020458984375,
"epoch": 0.15034511036697876,
"grad_norm": 0.6199520559909174,
"learning_rate": 4.76785965901085e-06,
"loss": 1.0077,
"mean_token_accuracy": 0.7498815268278122,
"num_tokens": 97815019.0,
"step": 1100
},
{
"entropy": 1.03134765625,
"epoch": 0.15171188409758765,
"grad_norm": 0.6661214418460119,
"learning_rate": 4.764337043821333e-06,
"loss": 1.0483,
"mean_token_accuracy": 0.7452669091522693,
"num_tokens": 98683384.0,
"step": 1110
},
{
"entropy": 1.004736328125,
"epoch": 0.15307865782819655,
"grad_norm": 0.6570525636640497,
"learning_rate": 4.760814428631817e-06,
"loss": 1.0052,
"mean_token_accuracy": 0.7516034580767155,
"num_tokens": 99564393.0,
"step": 1120
},
{
"entropy": 0.9775390625,
"epoch": 0.15444543155880544,
"grad_norm": 0.6402502450823856,
"learning_rate": 4.7572918134423e-06,
"loss": 0.9819,
"mean_token_accuracy": 0.7525982886552811,
"num_tokens": 100408247.0,
"step": 1130
},
{
"entropy": 0.964501953125,
"epoch": 0.15581220528941434,
"grad_norm": 0.7643425468582651,
"learning_rate": 4.753769198252783e-06,
"loss": 0.9392,
"mean_token_accuracy": 0.7658330470323562,
"num_tokens": 101297373.0,
"step": 1140
},
{
"entropy": 1.023388671875,
"epoch": 0.15717897902002323,
"grad_norm": 0.6602800093717967,
"learning_rate": 4.7502465830632665e-06,
"loss": 1.0183,
"mean_token_accuracy": 0.7481300219893455,
"num_tokens": 102230124.0,
"step": 1150
},
{
"entropy": 0.938916015625,
"epoch": 0.15854575275063212,
"grad_norm": 0.6060882059623902,
"learning_rate": 4.74672396787375e-06,
"loss": 0.9289,
"mean_token_accuracy": 0.764824515581131,
"num_tokens": 103126908.0,
"step": 1160
},
{
"entropy": 0.963525390625,
"epoch": 0.15991252648124102,
"grad_norm": 0.5590459590657602,
"learning_rate": 4.743201352684233e-06,
"loss": 0.966,
"mean_token_accuracy": 0.7611152499914169,
"num_tokens": 104024292.0,
"step": 1170
},
{
"entropy": 0.97568359375,
"epoch": 0.16127930021184994,
"grad_norm": 0.6480206843984407,
"learning_rate": 4.739678737494716e-06,
"loss": 0.9488,
"mean_token_accuracy": 0.7638736627995968,
"num_tokens": 104898403.0,
"step": 1180
},
{
"entropy": 0.9816650390625,
"epoch": 0.16264607394245884,
"grad_norm": 0.9276836764714617,
"learning_rate": 4.7361561223052e-06,
"loss": 0.9802,
"mean_token_accuracy": 0.7555055499076844,
"num_tokens": 105782333.0,
"step": 1190
},
{
"entropy": 0.97451171875,
"epoch": 0.16401284767306773,
"grad_norm": 0.7045006783902796,
"learning_rate": 4.7326335071156834e-06,
"loss": 0.9589,
"mean_token_accuracy": 0.762077247351408,
"num_tokens": 106647573.0,
"step": 1200
},
{
"entropy": 0.9876953125,
"epoch": 0.16537962140367662,
"grad_norm": 0.6190870463248735,
"learning_rate": 4.729110891926166e-06,
"loss": 0.9946,
"mean_token_accuracy": 0.7539896950125694,
"num_tokens": 107566523.0,
"step": 1210
},
{
"entropy": 0.953125,
"epoch": 0.16674639513428552,
"grad_norm": 0.6961527709218684,
"learning_rate": 4.72558827673665e-06,
"loss": 0.9596,
"mean_token_accuracy": 0.7632116481661797,
"num_tokens": 108414707.0,
"step": 1220
},
{
"entropy": 0.972607421875,
"epoch": 0.1681131688648944,
"grad_norm": 0.6230457434183949,
"learning_rate": 4.722065661547132e-06,
"loss": 0.9696,
"mean_token_accuracy": 0.7617248825728893,
"num_tokens": 109308818.0,
"step": 1230
},
{
"entropy": 0.984423828125,
"epoch": 0.1694799425955033,
"grad_norm": 0.6763317210667288,
"learning_rate": 4.718543046357617e-06,
"loss": 0.9944,
"mean_token_accuracy": 0.7549416437745095,
"num_tokens": 110236510.0,
"step": 1240
},
{
"entropy": 0.95888671875,
"epoch": 0.1708467163261122,
"grad_norm": 0.6658500810590964,
"learning_rate": 4.7150204311680995e-06,
"loss": 0.9518,
"mean_token_accuracy": 0.7638459712266922,
"num_tokens": 111071984.0,
"step": 1250
},
{
"entropy": 1.010595703125,
"epoch": 0.1722134900567211,
"grad_norm": 0.660741478365423,
"learning_rate": 4.711497815978583e-06,
"loss": 0.9983,
"mean_token_accuracy": 0.7543517753481865,
"num_tokens": 111952526.0,
"step": 1260
},
{
"entropy": 0.999072265625,
"epoch": 0.17358026378733002,
"grad_norm": 0.5971144367497986,
"learning_rate": 4.707975200789066e-06,
"loss": 1.0188,
"mean_token_accuracy": 0.7507157512009144,
"num_tokens": 112881665.0,
"step": 1270
},
{
"entropy": 0.99150390625,
"epoch": 0.1749470375179389,
"grad_norm": 0.6586367888955206,
"learning_rate": 4.704452585599549e-06,
"loss": 0.9922,
"mean_token_accuracy": 0.7534755580127239,
"num_tokens": 113764242.0,
"step": 1280
},
{
"entropy": 0.96513671875,
"epoch": 0.1763138112485478,
"grad_norm": 0.6008020440848786,
"learning_rate": 4.700929970410033e-06,
"loss": 0.9601,
"mean_token_accuracy": 0.7618438422679901,
"num_tokens": 114647527.0,
"step": 1290
},
{
"entropy": 0.985693359375,
"epoch": 0.1776805849791567,
"grad_norm": 0.6271424889133779,
"learning_rate": 4.697407355220516e-06,
"loss": 0.938,
"mean_token_accuracy": 0.7637601956725121,
"num_tokens": 115562594.0,
"step": 1300
},
{
"entropy": 0.98525390625,
"epoch": 0.1790473587097656,
"grad_norm": 0.6771921139391213,
"learning_rate": 4.693884740030999e-06,
"loss": 0.96,
"mean_token_accuracy": 0.7552240870893001,
"num_tokens": 116394915.0,
"step": 1310
},
{
"entropy": 1.0002197265625,
"epoch": 0.1804141324403745,
"grad_norm": 0.6266281947997581,
"learning_rate": 4.690362124841483e-06,
"loss": 0.995,
"mean_token_accuracy": 0.7544123701751232,
"num_tokens": 117275706.0,
"step": 1320
},
{
"entropy": 0.99951171875,
"epoch": 0.1817809061709834,
"grad_norm": 0.6257670200082966,
"learning_rate": 4.686839509651966e-06,
"loss": 0.9907,
"mean_token_accuracy": 0.7558987960219383,
"num_tokens": 118163200.0,
"step": 1330
},
{
"entropy": 0.971923828125,
"epoch": 0.18314767990159228,
"grad_norm": 0.6083066826555856,
"learning_rate": 4.683316894462449e-06,
"loss": 0.9682,
"mean_token_accuracy": 0.7564503282308579,
"num_tokens": 119069034.0,
"step": 1340
},
{
"entropy": 0.9571533203125,
"epoch": 0.18451445363220118,
"grad_norm": 0.8042758897838691,
"learning_rate": 4.679794279272933e-06,
"loss": 0.9762,
"mean_token_accuracy": 0.7602034375071526,
"num_tokens": 119965397.0,
"step": 1350
},
{
"entropy": 0.95419921875,
"epoch": 0.1858812273628101,
"grad_norm": 0.7277472255742403,
"learning_rate": 4.676271664083416e-06,
"loss": 0.9589,
"mean_token_accuracy": 0.7628471940755844,
"num_tokens": 120876126.0,
"step": 1360
},
{
"entropy": 1.029345703125,
"epoch": 0.187248001093419,
"grad_norm": 0.8974241746181876,
"learning_rate": 4.6727490488938995e-06,
"loss": 1.0283,
"mean_token_accuracy": 0.7472648732364178,
"num_tokens": 121712882.0,
"step": 1370
},
{
"entropy": 0.984521484375,
"epoch": 0.1886147748240279,
"grad_norm": 0.5889712615756614,
"learning_rate": 4.669226433704382e-06,
"loss": 0.9551,
"mean_token_accuracy": 0.7596720844507218,
"num_tokens": 122597154.0,
"step": 1380
},
{
"entropy": 0.98408203125,
"epoch": 0.18998154855463678,
"grad_norm": 0.6539326844170138,
"learning_rate": 4.665703818514866e-06,
"loss": 0.9554,
"mean_token_accuracy": 0.7635039560496807,
"num_tokens": 123457814.0,
"step": 1390
},
{
"entropy": 0.985595703125,
"epoch": 0.19134832228524568,
"grad_norm": 0.589518542615011,
"learning_rate": 4.6621812033253484e-06,
"loss": 0.954,
"mean_token_accuracy": 0.7633778557181359,
"num_tokens": 124340451.0,
"step": 1400
},
{
"entropy": 0.903515625,
"epoch": 0.19271509601585457,
"grad_norm": 0.6389580498988359,
"learning_rate": 4.658658588135833e-06,
"loss": 0.865,
"mean_token_accuracy": 0.7780114367604256,
"num_tokens": 125222917.0,
"step": 1410
},
{
"entropy": 0.944970703125,
"epoch": 0.19408186974646346,
"grad_norm": 0.6885671995614167,
"learning_rate": 4.6551359729463155e-06,
"loss": 0.9204,
"mean_token_accuracy": 0.765720783919096,
"num_tokens": 126167349.0,
"step": 1420
},
{
"entropy": 0.993505859375,
"epoch": 0.19544864347707236,
"grad_norm": 0.6004334904951046,
"learning_rate": 4.651613357756799e-06,
"loss": 1.0156,
"mean_token_accuracy": 0.7497326716780662,
"num_tokens": 127053792.0,
"step": 1430
},
{
"entropy": 0.99619140625,
"epoch": 0.19681541720768128,
"grad_norm": 0.6912742358969891,
"learning_rate": 4.648090742567283e-06,
"loss": 1.0084,
"mean_token_accuracy": 0.7558760106563568,
"num_tokens": 127948703.0,
"step": 1440
},
{
"entropy": 0.965576171875,
"epoch": 0.19818219093829018,
"grad_norm": 0.6519328998921796,
"learning_rate": 4.644568127377765e-06,
"loss": 0.9694,
"mean_token_accuracy": 0.7593872673809529,
"num_tokens": 128861108.0,
"step": 1450
},
{
"entropy": 0.981005859375,
"epoch": 0.19954896466889907,
"grad_norm": 0.6340028521070149,
"learning_rate": 4.641045512188249e-06,
"loss": 0.9808,
"mean_token_accuracy": 0.7579961940646172,
"num_tokens": 129749270.0,
"step": 1460
},
{
"entropy": 0.943701171875,
"epoch": 0.20091573839950796,
"grad_norm": 0.6699430801818396,
"learning_rate": 4.6375228969987324e-06,
"loss": 0.9364,
"mean_token_accuracy": 0.7645595327019692,
"num_tokens": 130655000.0,
"step": 1470
},
{
"entropy": 0.96357421875,
"epoch": 0.20228251213011686,
"grad_norm": 0.6393383845561634,
"learning_rate": 4.634000281809216e-06,
"loss": 0.9721,
"mean_token_accuracy": 0.7585249237716198,
"num_tokens": 131564630.0,
"step": 1480
},
{
"entropy": 0.960498046875,
"epoch": 0.20364928586072575,
"grad_norm": 0.5971975625079909,
"learning_rate": 4.630477666619699e-06,
"loss": 0.9482,
"mean_token_accuracy": 0.762010820955038,
"num_tokens": 132430033.0,
"step": 1490
},
{
"entropy": 0.9328125,
"epoch": 0.20501605959133465,
"grad_norm": 0.5850164022579414,
"learning_rate": 4.626955051430182e-06,
"loss": 0.9216,
"mean_token_accuracy": 0.768496111780405,
"num_tokens": 133344093.0,
"step": 1500
},
{
"entropy": 0.99072265625,
"epoch": 0.20638283332194354,
"grad_norm": 0.6880075846597385,
"learning_rate": 4.623432436240665e-06,
"loss": 0.9808,
"mean_token_accuracy": 0.7553143031895161,
"num_tokens": 134170710.0,
"step": 1510
},
{
"entropy": 0.957470703125,
"epoch": 0.20774960705255244,
"grad_norm": 0.718955140068962,
"learning_rate": 4.619909821051149e-06,
"loss": 0.9344,
"mean_token_accuracy": 0.7634217724204063,
"num_tokens": 135002300.0,
"step": 1520
},
{
"entropy": 0.974755859375,
"epoch": 0.20911638078316136,
"grad_norm": 0.6555207337584779,
"learning_rate": 4.616387205861632e-06,
"loss": 0.9721,
"mean_token_accuracy": 0.7600927069783211,
"num_tokens": 135877256.0,
"step": 1530
},
{
"entropy": 0.967333984375,
"epoch": 0.21048315451377025,
"grad_norm": 0.6383233609552382,
"learning_rate": 4.6128645906721156e-06,
"loss": 0.9393,
"mean_token_accuracy": 0.7657891027629375,
"num_tokens": 136719178.0,
"step": 1540
},
{
"entropy": 1.010888671875,
"epoch": 0.21184992824437915,
"grad_norm": 0.9458297376783754,
"learning_rate": 4.609341975482598e-06,
"loss": 0.9998,
"mean_token_accuracy": 0.7548832461237908,
"num_tokens": 137587067.0,
"step": 1550
},
{
"entropy": 0.968017578125,
"epoch": 0.21321670197498804,
"grad_norm": 0.6417368393807718,
"learning_rate": 4.605819360293082e-06,
"loss": 0.9524,
"mean_token_accuracy": 0.7617362692952157,
"num_tokens": 138510147.0,
"step": 1560
},
{
"entropy": 0.9158203125,
"epoch": 0.21458347570559694,
"grad_norm": 0.5682414329562661,
"learning_rate": 4.602296745103565e-06,
"loss": 0.9198,
"mean_token_accuracy": 0.7677236050367355,
"num_tokens": 139391196.0,
"step": 1570
},
{
"entropy": 0.9833984375,
"epoch": 0.21595024943620583,
"grad_norm": 0.6723561880329522,
"learning_rate": 4.598774129914049e-06,
"loss": 1.009,
"mean_token_accuracy": 0.7541672974824906,
"num_tokens": 140283552.0,
"step": 1580
},
{
"entropy": 0.9877685546875,
"epoch": 0.21731702316681473,
"grad_norm": 0.7299747533710157,
"learning_rate": 4.595251514724532e-06,
"loss": 0.9816,
"mean_token_accuracy": 0.7523842424154281,
"num_tokens": 141150053.0,
"step": 1590
},
{
"entropy": 0.9478515625,
"epoch": 0.21868379689742362,
"grad_norm": 0.5884053009829695,
"learning_rate": 4.591728899535015e-06,
"loss": 0.9407,
"mean_token_accuracy": 0.7611678294837475,
"num_tokens": 142039072.0,
"step": 1600
},
{
"entropy": 0.943896484375,
"epoch": 0.22005057062803252,
"grad_norm": 0.663629087315831,
"learning_rate": 4.588206284345499e-06,
"loss": 0.9297,
"mean_token_accuracy": 0.7660427555441857,
"num_tokens": 142931990.0,
"step": 1610
},
{
"entropy": 1.008837890625,
"epoch": 0.22141734435864144,
"grad_norm": 0.7151625802922789,
"learning_rate": 4.584683669155981e-06,
"loss": 1.0288,
"mean_token_accuracy": 0.7493112236261368,
"num_tokens": 143831487.0,
"step": 1620
},
{
"entropy": 0.974853515625,
"epoch": 0.22278411808925033,
"grad_norm": 0.6388870902364759,
"learning_rate": 4.581161053966465e-06,
"loss": 0.9835,
"mean_token_accuracy": 0.7585226200520993,
"num_tokens": 144782154.0,
"step": 1630
},
{
"entropy": 1.004638671875,
"epoch": 0.22415089181985923,
"grad_norm": 0.7384278702450001,
"learning_rate": 4.5776384387769485e-06,
"loss": 1.016,
"mean_token_accuracy": 0.7529004834592342,
"num_tokens": 145684601.0,
"step": 1640
},
{
"entropy": 0.961474609375,
"epoch": 0.22551766555046812,
"grad_norm": 0.6539601659594253,
"learning_rate": 4.574115823587432e-06,
"loss": 0.9485,
"mean_token_accuracy": 0.7636470764875412,
"num_tokens": 146574905.0,
"step": 1650
},
{
"entropy": 0.953857421875,
"epoch": 0.22688443928107702,
"grad_norm": 0.6233078747521409,
"learning_rate": 4.570593208397915e-06,
"loss": 0.964,
"mean_token_accuracy": 0.7634499639272689,
"num_tokens": 147465145.0,
"step": 1660
},
{
"entropy": 0.9599609375,
"epoch": 0.2282512130116859,
"grad_norm": 0.5986683154095792,
"learning_rate": 4.567070593208398e-06,
"loss": 0.9571,
"mean_token_accuracy": 0.7602071113884449,
"num_tokens": 148329950.0,
"step": 1670
},
{
"entropy": 0.968359375,
"epoch": 0.2296179867422948,
"grad_norm": 0.6923835544626978,
"learning_rate": 4.563547978018882e-06,
"loss": 0.9589,
"mean_token_accuracy": 0.7585745595395565,
"num_tokens": 149228860.0,
"step": 1680
},
{
"entropy": 0.942138671875,
"epoch": 0.2309847604729037,
"grad_norm": 0.7113994814547963,
"learning_rate": 4.560025362829365e-06,
"loss": 0.9508,
"mean_token_accuracy": 0.7625337257981301,
"num_tokens": 150100793.0,
"step": 1690
},
{
"entropy": 0.986767578125,
"epoch": 0.23235153420351262,
"grad_norm": 0.6034396346575002,
"learning_rate": 4.556502747639848e-06,
"loss": 0.9664,
"mean_token_accuracy": 0.7602703474462033,
"num_tokens": 150995094.0,
"step": 1700
},
{
"entropy": 0.9890625,
"epoch": 0.23371830793412152,
"grad_norm": 0.6152795958791075,
"learning_rate": 4.552980132450332e-06,
"loss": 0.995,
"mean_token_accuracy": 0.7526928067207337,
"num_tokens": 151921452.0,
"step": 1710
},
{
"entropy": 0.956787109375,
"epoch": 0.2350850816647304,
"grad_norm": 0.5810736616838182,
"learning_rate": 4.549457517260814e-06,
"loss": 0.9621,
"mean_token_accuracy": 0.7604227617383004,
"num_tokens": 152829538.0,
"step": 1720
},
{
"entropy": 0.929296875,
"epoch": 0.2364518553953393,
"grad_norm": 0.6527614212817,
"learning_rate": 4.545934902071298e-06,
"loss": 0.9486,
"mean_token_accuracy": 0.7641879566013813,
"num_tokens": 153770075.0,
"step": 1730
},
{
"entropy": 0.98994140625,
"epoch": 0.2378186291259482,
"grad_norm": 0.7163433001387484,
"learning_rate": 4.5424122868817814e-06,
"loss": 0.9994,
"mean_token_accuracy": 0.7524978123605252,
"num_tokens": 154631539.0,
"step": 1740
},
{
"entropy": 0.9587890625,
"epoch": 0.2391854028565571,
"grad_norm": 0.6320826100719923,
"learning_rate": 4.538889671692265e-06,
"loss": 0.9298,
"mean_token_accuracy": 0.7669215567409993,
"num_tokens": 155495251.0,
"step": 1750
},
{
"entropy": 0.982861328125,
"epoch": 0.240552176587166,
"grad_norm": 0.674307410186055,
"learning_rate": 4.535367056502748e-06,
"loss": 0.9606,
"mean_token_accuracy": 0.7626395493745803,
"num_tokens": 156404778.0,
"step": 1760
},
{
"entropy": 0.97705078125,
"epoch": 0.24191895031777488,
"grad_norm": 0.5752522748003577,
"learning_rate": 4.531844441313231e-06,
"loss": 0.962,
"mean_token_accuracy": 0.7577393293380738,
"num_tokens": 157289886.0,
"step": 1770
},
{
"entropy": 0.930615234375,
"epoch": 0.24328572404838378,
"grad_norm": 0.5877156453150303,
"learning_rate": 4.528321826123715e-06,
"loss": 0.9362,
"mean_token_accuracy": 0.7666725993156434,
"num_tokens": 158159238.0,
"step": 1780
},
{
"entropy": 0.95732421875,
"epoch": 0.2446524977789927,
"grad_norm": 0.5463174967353166,
"learning_rate": 4.524799210934198e-06,
"loss": 0.9458,
"mean_token_accuracy": 0.76511884406209,
"num_tokens": 159054886.0,
"step": 1790
},
{
"entropy": 0.970947265625,
"epoch": 0.2460192715096016,
"grad_norm": 0.588369968024304,
"learning_rate": 4.521276595744681e-06,
"loss": 0.9499,
"mean_token_accuracy": 0.7620976120233536,
"num_tokens": 159916702.0,
"step": 1800
},
{
"entropy": 0.9655029296875,
"epoch": 0.2473860452402105,
"grad_norm": 0.6601482021793649,
"learning_rate": 4.5177539805551646e-06,
"loss": 0.9869,
"mean_token_accuracy": 0.7575971446931362,
"num_tokens": 160800051.0,
"step": 1810
},
{
"entropy": 0.953466796875,
"epoch": 0.24875281897081938,
"grad_norm": 0.6749722038888123,
"learning_rate": 4.514231365365648e-06,
"loss": 0.9437,
"mean_token_accuracy": 0.7643050238490104,
"num_tokens": 161649504.0,
"step": 1820
},
{
"entropy": 0.954150390625,
"epoch": 0.2501195927014283,
"grad_norm": 0.6356999718249062,
"learning_rate": 4.510708750176131e-06,
"loss": 0.9294,
"mean_token_accuracy": 0.7622107230126858,
"num_tokens": 162521710.0,
"step": 1830
},
{
"entropy": 0.944140625,
"epoch": 0.2514863664320372,
"grad_norm": 0.6104358786719225,
"learning_rate": 4.507186134986614e-06,
"loss": 0.9344,
"mean_token_accuracy": 0.7650333680212498,
"num_tokens": 163456795.0,
"step": 1840
},
{
"entropy": 0.944580078125,
"epoch": 0.2528531401626461,
"grad_norm": 0.619524392600988,
"learning_rate": 4.503663519797098e-06,
"loss": 0.9538,
"mean_token_accuracy": 0.7618495248258114,
"num_tokens": 164354274.0,
"step": 1850
},
{
"entropy": 0.9646484375,
"epoch": 0.254219913893255,
"grad_norm": 0.6650991067078843,
"learning_rate": 4.5001409046075814e-06,
"loss": 0.9517,
"mean_token_accuracy": 0.7627912171185016,
"num_tokens": 165231568.0,
"step": 1860
},
{
"entropy": 0.9513671875,
"epoch": 0.2555866876238639,
"grad_norm": 0.5873146872316886,
"learning_rate": 4.496618289418064e-06,
"loss": 0.9256,
"mean_token_accuracy": 0.7664681270718574,
"num_tokens": 166149414.0,
"step": 1870
},
{
"entropy": 0.967822265625,
"epoch": 0.2569534613544728,
"grad_norm": 0.6636062753625496,
"learning_rate": 4.493095674228548e-06,
"loss": 0.9444,
"mean_token_accuracy": 0.7636000894010067,
"num_tokens": 167006969.0,
"step": 1880
},
{
"entropy": 0.973583984375,
"epoch": 0.2583202350850817,
"grad_norm": 0.8033388227084874,
"learning_rate": 4.48957305903903e-06,
"loss": 0.9913,
"mean_token_accuracy": 0.7548758521676063,
"num_tokens": 167901440.0,
"step": 1890
},
{
"entropy": 0.9537109375,
"epoch": 0.25968700881569057,
"grad_norm": 0.6716540114868533,
"learning_rate": 4.486050443849515e-06,
"loss": 0.9682,
"mean_token_accuracy": 0.7617958843708038,
"num_tokens": 168783114.0,
"step": 1900
},
{
"entropy": 0.955224609375,
"epoch": 0.26105378254629946,
"grad_norm": 0.6114881471710274,
"learning_rate": 4.4825278286599975e-06,
"loss": 0.9497,
"mean_token_accuracy": 0.7634148001670837,
"num_tokens": 169685621.0,
"step": 1910
},
{
"entropy": 0.92001953125,
"epoch": 0.26242055627690836,
"grad_norm": 0.708019223748798,
"learning_rate": 4.479005213470481e-06,
"loss": 0.8983,
"mean_token_accuracy": 0.7730237439274787,
"num_tokens": 170573171.0,
"step": 1920
},
{
"entropy": 0.9775390625,
"epoch": 0.26378733000751725,
"grad_norm": 0.6591096203286871,
"learning_rate": 4.475482598280964e-06,
"loss": 0.9603,
"mean_token_accuracy": 0.7563680961728096,
"num_tokens": 171454806.0,
"step": 1930
},
{
"entropy": 0.925732421875,
"epoch": 0.26515410373812615,
"grad_norm": 0.6382427334754156,
"learning_rate": 4.471959983091447e-06,
"loss": 0.9234,
"mean_token_accuracy": 0.7690858602523803,
"num_tokens": 172346713.0,
"step": 1940
},
{
"entropy": 0.989111328125,
"epoch": 0.26652087746873504,
"grad_norm": 0.6338496940048642,
"learning_rate": 4.468437367901931e-06,
"loss": 0.972,
"mean_token_accuracy": 0.7603953361511231,
"num_tokens": 173233962.0,
"step": 1950
},
{
"entropy": 0.967822265625,
"epoch": 0.26788765119934393,
"grad_norm": 0.6703113542132306,
"learning_rate": 4.464914752712414e-06,
"loss": 0.9599,
"mean_token_accuracy": 0.7602911792695523,
"num_tokens": 174110792.0,
"step": 1960
},
{
"entropy": 0.9300537109375,
"epoch": 0.26925442492995283,
"grad_norm": 0.6288395930693714,
"learning_rate": 4.461392137522897e-06,
"loss": 0.9033,
"mean_token_accuracy": 0.7716706447303295,
"num_tokens": 175005864.0,
"step": 1970
},
{
"entropy": 0.928173828125,
"epoch": 0.2706211986605617,
"grad_norm": 0.5949692220332424,
"learning_rate": 4.457869522333381e-06,
"loss": 0.9024,
"mean_token_accuracy": 0.7726658187806607,
"num_tokens": 175870222.0,
"step": 1980
},
{
"entropy": 0.946044921875,
"epoch": 0.2719879723911706,
"grad_norm": 0.6735970434836677,
"learning_rate": 4.454346907143864e-06,
"loss": 0.9223,
"mean_token_accuracy": 0.7697566829621791,
"num_tokens": 176750406.0,
"step": 1990
},
{
"entropy": 0.980712890625,
"epoch": 0.2733547461217795,
"grad_norm": 0.6784837357836714,
"learning_rate": 4.450824291954347e-06,
"loss": 0.9677,
"mean_token_accuracy": 0.7596066750586032,
"num_tokens": 177646304.0,
"step": 2000
},
{
"entropy": 0.916015625,
"epoch": 0.27472151985238846,
"grad_norm": 0.5341074336417482,
"learning_rate": 4.44730167676483e-06,
"loss": 0.9133,
"mean_token_accuracy": 0.7714160323143006,
"num_tokens": 178544048.0,
"step": 2010
},
{
"entropy": 0.992626953125,
"epoch": 0.27608829358299736,
"grad_norm": 0.6343479541516627,
"learning_rate": 4.443779061575314e-06,
"loss": 0.9708,
"mean_token_accuracy": 0.7601838581264019,
"num_tokens": 179424878.0,
"step": 2020
},
{
"entropy": 0.9024169921875,
"epoch": 0.27745506731360625,
"grad_norm": 0.7261156018940499,
"learning_rate": 4.4402564463857975e-06,
"loss": 0.9017,
"mean_token_accuracy": 0.7708473280072212,
"num_tokens": 180290131.0,
"step": 2030
},
{
"entropy": 0.9783203125,
"epoch": 0.27882184104421515,
"grad_norm": 0.6095237768280096,
"learning_rate": 4.43673383119628e-06,
"loss": 0.9626,
"mean_token_accuracy": 0.7618141964077949,
"num_tokens": 181151230.0,
"step": 2040
},
{
"entropy": 0.9681640625,
"epoch": 0.28018861477482404,
"grad_norm": 0.6272270146702937,
"learning_rate": 4.433211216006764e-06,
"loss": 0.9466,
"mean_token_accuracy": 0.7618514016270638,
"num_tokens": 181997158.0,
"step": 2050
},
{
"entropy": 0.934228515625,
"epoch": 0.28155538850543294,
"grad_norm": 0.6019808247230355,
"learning_rate": 4.4296886008172465e-06,
"loss": 0.9366,
"mean_token_accuracy": 0.767015390843153,
"num_tokens": 182869862.0,
"step": 2060
},
{
"entropy": 0.926171875,
"epoch": 0.28292216223604183,
"grad_norm": 0.6433616451710493,
"learning_rate": 4.426165985627731e-06,
"loss": 0.9426,
"mean_token_accuracy": 0.765649500489235,
"num_tokens": 183793317.0,
"step": 2070
},
{
"entropy": 0.93056640625,
"epoch": 0.2842889359666507,
"grad_norm": 0.6227343822072434,
"learning_rate": 4.4226433704382136e-06,
"loss": 0.9139,
"mean_token_accuracy": 0.769086018204689,
"num_tokens": 184654232.0,
"step": 2080
},
{
"entropy": 0.93681640625,
"epoch": 0.2856557096972596,
"grad_norm": 0.6211745228393116,
"learning_rate": 4.419120755248697e-06,
"loss": 0.936,
"mean_token_accuracy": 0.7654936440289021,
"num_tokens": 185544158.0,
"step": 2090
},
{
"entropy": 1.007275390625,
"epoch": 0.2870224834278685,
"grad_norm": 0.5891567176731857,
"learning_rate": 4.41559814005918e-06,
"loss": 1.0139,
"mean_token_accuracy": 0.7500253945589066,
"num_tokens": 186444139.0,
"step": 2100
},
{
"entropy": 0.9215576171875,
"epoch": 0.2883892571584774,
"grad_norm": 0.644228770239019,
"learning_rate": 4.412075524869663e-06,
"loss": 0.9218,
"mean_token_accuracy": 0.769534221291542,
"num_tokens": 187281267.0,
"step": 2110
},
{
"entropy": 0.9626953125,
"epoch": 0.2897560308890863,
"grad_norm": 0.6353122380700604,
"learning_rate": 4.408552909680147e-06,
"loss": 0.949,
"mean_token_accuracy": 0.7632474407553673,
"num_tokens": 188147941.0,
"step": 2120
},
{
"entropy": 0.938916015625,
"epoch": 0.2911228046196952,
"grad_norm": 0.6202614074460888,
"learning_rate": 4.4050302944906304e-06,
"loss": 0.9472,
"mean_token_accuracy": 0.7642685241997242,
"num_tokens": 189047615.0,
"step": 2130
},
{
"entropy": 0.94091796875,
"epoch": 0.2924895783503041,
"grad_norm": 0.6518359505270835,
"learning_rate": 4.401507679301113e-06,
"loss": 0.9259,
"mean_token_accuracy": 0.7653455749154091,
"num_tokens": 189900274.0,
"step": 2140
},
{
"entropy": 1.00888671875,
"epoch": 0.293856352080913,
"grad_norm": 0.9539597227789041,
"learning_rate": 4.397985064111597e-06,
"loss": 1.0317,
"mean_token_accuracy": 0.7492226399481297,
"num_tokens": 190817864.0,
"step": 2150
},
{
"entropy": 0.94873046875,
"epoch": 0.2952231258115219,
"grad_norm": 0.5986691914153677,
"learning_rate": 4.39446244892208e-06,
"loss": 0.942,
"mean_token_accuracy": 0.7657283268868923,
"num_tokens": 191696200.0,
"step": 2160
},
{
"entropy": 0.9833984375,
"epoch": 0.2965898995421308,
"grad_norm": 0.7251659477673614,
"learning_rate": 4.390939833732563e-06,
"loss": 0.9475,
"mean_token_accuracy": 0.7626770831644535,
"num_tokens": 192504212.0,
"step": 2170
},
{
"entropy": 0.9486328125,
"epoch": 0.2979566732727397,
"grad_norm": 0.6793567483070109,
"learning_rate": 4.387417218543047e-06,
"loss": 0.9476,
"mean_token_accuracy": 0.7627994798123836,
"num_tokens": 193391915.0,
"step": 2180
},
{
"entropy": 0.920556640625,
"epoch": 0.2993234470033486,
"grad_norm": 0.6889372804742054,
"learning_rate": 4.38389460335353e-06,
"loss": 0.8904,
"mean_token_accuracy": 0.771683468669653,
"num_tokens": 194274927.0,
"step": 2190
},
{
"entropy": 0.9565673828125,
"epoch": 0.3006902207339575,
"grad_norm": 0.5651206538693608,
"learning_rate": 4.380371988164014e-06,
"loss": 0.9762,
"mean_token_accuracy": 0.756953302025795,
"num_tokens": 195216521.0,
"step": 2200
},
{
"entropy": 0.929833984375,
"epoch": 0.3020569944645664,
"grad_norm": 0.5755578778106992,
"learning_rate": 4.376849372974496e-06,
"loss": 0.943,
"mean_token_accuracy": 0.766729049384594,
"num_tokens": 196130208.0,
"step": 2210
},
{
"entropy": 0.9283203125,
"epoch": 0.3034237681951753,
"grad_norm": 0.5986357403959879,
"learning_rate": 4.37332675778498e-06,
"loss": 0.9096,
"mean_token_accuracy": 0.7672458060085774,
"num_tokens": 197028743.0,
"step": 2220
},
{
"entropy": 0.937158203125,
"epoch": 0.3047905419257842,
"grad_norm": 0.593467634960188,
"learning_rate": 4.369804142595463e-06,
"loss": 0.9475,
"mean_token_accuracy": 0.7653128646314145,
"num_tokens": 197898713.0,
"step": 2230
},
{
"entropy": 0.933251953125,
"epoch": 0.3061573156563931,
"grad_norm": 0.6662179824756784,
"learning_rate": 4.366281527405947e-06,
"loss": 0.951,
"mean_token_accuracy": 0.7607046350836754,
"num_tokens": 198815509.0,
"step": 2240
},
{
"entropy": 0.908984375,
"epoch": 0.307524089387002,
"grad_norm": 0.6344402549099978,
"learning_rate": 4.36275891221643e-06,
"loss": 0.8962,
"mean_token_accuracy": 0.7711473032832146,
"num_tokens": 199704367.0,
"step": 2250
},
{
"entropy": 0.8944091796875,
"epoch": 0.3088908631176109,
"grad_norm": 0.6728568875014157,
"learning_rate": 4.359236297026913e-06,
"loss": 0.8903,
"mean_token_accuracy": 0.7735740505158901,
"num_tokens": 200578924.0,
"step": 2260
},
{
"entropy": 0.9099609375,
"epoch": 0.3102576368482198,
"grad_norm": 0.711684787900784,
"learning_rate": 4.355713681837396e-06,
"loss": 0.8903,
"mean_token_accuracy": 0.7755497299134732,
"num_tokens": 201456944.0,
"step": 2270
},
{
"entropy": 0.917431640625,
"epoch": 0.31162441057882867,
"grad_norm": 0.6447354520730727,
"learning_rate": 4.352191066647879e-06,
"loss": 0.9283,
"mean_token_accuracy": 0.7677726872265339,
"num_tokens": 202323380.0,
"step": 2280
},
{
"entropy": 0.949755859375,
"epoch": 0.31299118430943756,
"grad_norm": 0.6573256117414116,
"learning_rate": 4.348668451458363e-06,
"loss": 0.9189,
"mean_token_accuracy": 0.7692284889519214,
"num_tokens": 203221545.0,
"step": 2290
},
{
"entropy": 0.9712890625,
"epoch": 0.31435795804004646,
"grad_norm": 0.6375554528805764,
"learning_rate": 4.3451458362688465e-06,
"loss": 0.9674,
"mean_token_accuracy": 0.7581154376268386,
"num_tokens": 204099161.0,
"step": 2300
},
{
"entropy": 0.95185546875,
"epoch": 0.31572473177065535,
"grad_norm": 0.6084126972785896,
"learning_rate": 4.34162322107933e-06,
"loss": 0.9329,
"mean_token_accuracy": 0.7653752952814102,
"num_tokens": 204961852.0,
"step": 2310
},
{
"entropy": 0.97763671875,
"epoch": 0.31709150550126425,
"grad_norm": 0.660140541485539,
"learning_rate": 4.338100605889813e-06,
"loss": 0.9425,
"mean_token_accuracy": 0.7628352038562298,
"num_tokens": 205811402.0,
"step": 2320
},
{
"entropy": 0.909228515625,
"epoch": 0.31845827923187314,
"grad_norm": 0.6692889330630295,
"learning_rate": 4.334577990700296e-06,
"loss": 0.8914,
"mean_token_accuracy": 0.7735836558043957,
"num_tokens": 206683323.0,
"step": 2330
},
{
"entropy": 0.962744140625,
"epoch": 0.31982505296248204,
"grad_norm": 0.638638626799313,
"learning_rate": 4.33105537551078e-06,
"loss": 0.9568,
"mean_token_accuracy": 0.760423143953085,
"num_tokens": 207548993.0,
"step": 2340
},
{
"entropy": 0.9115234375,
"epoch": 0.32119182669309093,
"grad_norm": 0.608819227362402,
"learning_rate": 4.327532760321263e-06,
"loss": 0.8849,
"mean_token_accuracy": 0.7738455638289452,
"num_tokens": 208393459.0,
"step": 2350
},
{
"entropy": 0.945654296875,
"epoch": 0.3225586004236999,
"grad_norm": 0.696075365495792,
"learning_rate": 4.324010145131746e-06,
"loss": 0.9414,
"mean_token_accuracy": 0.7634647272527217,
"num_tokens": 209266863.0,
"step": 2360
},
{
"entropy": 0.9439453125,
"epoch": 0.3239253741543088,
"grad_norm": 0.5692744767606693,
"learning_rate": 4.32048752994223e-06,
"loss": 0.953,
"mean_token_accuracy": 0.7632196754217148,
"num_tokens": 210197901.0,
"step": 2370
},
{
"entropy": 0.95810546875,
"epoch": 0.32529214788491767,
"grad_norm": 0.5917734996309398,
"learning_rate": 4.316964914752712e-06,
"loss": 0.9597,
"mean_token_accuracy": 0.7604008458554745,
"num_tokens": 211082714.0,
"step": 2380
},
{
"entropy": 0.962646484375,
"epoch": 0.32665892161552657,
"grad_norm": 0.6971296296253341,
"learning_rate": 4.313442299563196e-06,
"loss": 0.9635,
"mean_token_accuracy": 0.7615106001496315,
"num_tokens": 211920987.0,
"step": 2390
},
{
"entropy": 0.934521484375,
"epoch": 0.32802569534613546,
"grad_norm": 0.6881725770793218,
"learning_rate": 4.3099196843736794e-06,
"loss": 0.9305,
"mean_token_accuracy": 0.7667819730937481,
"num_tokens": 212778254.0,
"step": 2400
},
{
"entropy": 0.95,
"epoch": 0.32939246907674435,
"grad_norm": 0.6365948138305505,
"learning_rate": 4.306397069184163e-06,
"loss": 0.9312,
"mean_token_accuracy": 0.7617771111428737,
"num_tokens": 213665677.0,
"step": 2410
},
{
"entropy": 0.96171875,
"epoch": 0.33075924280735325,
"grad_norm": 0.715390563128823,
"learning_rate": 4.302874453994646e-06,
"loss": 0.9295,
"mean_token_accuracy": 0.7646388247609138,
"num_tokens": 214532016.0,
"step": 2420
},
{
"entropy": 0.9062744140625,
"epoch": 0.33212601653796214,
"grad_norm": 0.5972672736541248,
"learning_rate": 4.299351838805129e-06,
"loss": 0.8921,
"mean_token_accuracy": 0.7734146490693092,
"num_tokens": 215416984.0,
"step": 2430
},
{
"entropy": 0.989111328125,
"epoch": 0.33349279026857104,
"grad_norm": 0.5898684170377935,
"learning_rate": 4.295829223615613e-06,
"loss": 0.9649,
"mean_token_accuracy": 0.7597092792391777,
"num_tokens": 216317310.0,
"step": 2440
},
{
"entropy": 0.95419921875,
"epoch": 0.33485956399917993,
"grad_norm": 0.6491478159186219,
"learning_rate": 4.292306608426096e-06,
"loss": 0.9191,
"mean_token_accuracy": 0.7672670923173428,
"num_tokens": 217188245.0,
"step": 2450
},
{
"entropy": 0.902294921875,
"epoch": 0.3362263377297888,
"grad_norm": 0.5743270054238631,
"learning_rate": 4.288783993236579e-06,
"loss": 0.8933,
"mean_token_accuracy": 0.7729593485593795,
"num_tokens": 218084772.0,
"step": 2460
},
{
"entropy": 0.948095703125,
"epoch": 0.3375931114603977,
"grad_norm": 0.6499332169949742,
"learning_rate": 4.285261378047063e-06,
"loss": 0.9476,
"mean_token_accuracy": 0.7612086497247219,
"num_tokens": 218985412.0,
"step": 2470
},
{
"entropy": 0.96474609375,
"epoch": 0.3389598851910066,
"grad_norm": 0.6034303761956543,
"learning_rate": 4.281738762857546e-06,
"loss": 0.9385,
"mean_token_accuracy": 0.7637693896889687,
"num_tokens": 219878663.0,
"step": 2480
},
{
"entropy": 0.987890625,
"epoch": 0.3403266589216155,
"grad_norm": 0.651978764130767,
"learning_rate": 4.278216147668029e-06,
"loss": 0.9937,
"mean_token_accuracy": 0.753931286931038,
"num_tokens": 220775855.0,
"step": 2490
},
{
"entropy": 0.971630859375,
"epoch": 0.3416934326522244,
"grad_norm": 0.6522268646011458,
"learning_rate": 4.274693532478512e-06,
"loss": 0.9628,
"mean_token_accuracy": 0.75891033411026,
"num_tokens": 221663073.0,
"step": 2500
},
{
"entropy": 0.9159912109375,
"epoch": 0.3430602063828333,
"grad_norm": 0.5692529870401812,
"learning_rate": 4.271170917288996e-06,
"loss": 0.8953,
"mean_token_accuracy": 0.7721257641911506,
"num_tokens": 222544981.0,
"step": 2510
},
{
"entropy": 0.9342529296875,
"epoch": 0.3444269801134422,
"grad_norm": 0.6318340752950149,
"learning_rate": 4.2676483020994795e-06,
"loss": 0.9143,
"mean_token_accuracy": 0.7681442283093929,
"num_tokens": 223440024.0,
"step": 2520
},
{
"entropy": 0.96083984375,
"epoch": 0.34579375384405114,
"grad_norm": 0.5893731274934172,
"learning_rate": 4.264125686909962e-06,
"loss": 0.9672,
"mean_token_accuracy": 0.7625994205474853,
"num_tokens": 224363546.0,
"step": 2530
},
{
"entropy": 0.968994140625,
"epoch": 0.34716052757466004,
"grad_norm": 0.64610729949561,
"learning_rate": 4.260603071720446e-06,
"loss": 0.9544,
"mean_token_accuracy": 0.7634921133518219,
"num_tokens": 225225542.0,
"step": 2540
},
{
"entropy": 0.97392578125,
"epoch": 0.34852730130526893,
"grad_norm": 0.653304315818638,
"learning_rate": 4.257080456530928e-06,
"loss": 0.97,
"mean_token_accuracy": 0.7569543965160846,
"num_tokens": 226103209.0,
"step": 2550
},
{
"entropy": 0.917236328125,
"epoch": 0.3498940750358778,
"grad_norm": 0.6968122120218563,
"learning_rate": 4.253557841341413e-06,
"loss": 0.8944,
"mean_token_accuracy": 0.7724640846252442,
"num_tokens": 226998581.0,
"step": 2560
},
{
"entropy": 0.927294921875,
"epoch": 0.3512608487664867,
"grad_norm": 0.6781977538997959,
"learning_rate": 4.2500352261518955e-06,
"loss": 0.931,
"mean_token_accuracy": 0.7673283696174622,
"num_tokens": 227888802.0,
"step": 2570
},
{
"entropy": 0.8976318359375,
"epoch": 0.3526276224970956,
"grad_norm": 0.5982421098892895,
"learning_rate": 4.246512610962379e-06,
"loss": 0.8788,
"mean_token_accuracy": 0.7766573868691922,
"num_tokens": 228776580.0,
"step": 2580
},
{
"entropy": 0.956591796875,
"epoch": 0.3539943962277045,
"grad_norm": 0.6742282489969494,
"learning_rate": 4.242989995772862e-06,
"loss": 0.9516,
"mean_token_accuracy": 0.7631676279008388,
"num_tokens": 229624559.0,
"step": 2590
},
{
"entropy": 0.96103515625,
"epoch": 0.3553611699583134,
"grad_norm": 0.5789812298845777,
"learning_rate": 4.239467380583345e-06,
"loss": 0.9511,
"mean_token_accuracy": 0.7626226238906384,
"num_tokens": 230559792.0,
"step": 2600
},
{
"entropy": 0.932177734375,
"epoch": 0.3567279436889223,
"grad_norm": 0.6149617810914817,
"learning_rate": 4.235944765393829e-06,
"loss": 0.9042,
"mean_token_accuracy": 0.7703681223094463,
"num_tokens": 231437280.0,
"step": 2610
},
{
"entropy": 0.957177734375,
"epoch": 0.3580947174195312,
"grad_norm": 0.675637006580255,
"learning_rate": 4.232422150204312e-06,
"loss": 0.9534,
"mean_token_accuracy": 0.7619762703776359,
"num_tokens": 232292698.0,
"step": 2620
},
{
"entropy": 0.985205078125,
"epoch": 0.3594614911501401,
"grad_norm": 0.6262400648683922,
"learning_rate": 4.228899535014795e-06,
"loss": 0.9839,
"mean_token_accuracy": 0.7548864528536796,
"num_tokens": 233148286.0,
"step": 2630
},
{
"entropy": 0.9681640625,
"epoch": 0.360828264880749,
"grad_norm": 0.6645669749639399,
"learning_rate": 4.225376919825279e-06,
"loss": 0.9616,
"mean_token_accuracy": 0.7586772352457046,
"num_tokens": 234027940.0,
"step": 2640
},
{
"entropy": 0.9270751953125,
"epoch": 0.3621950386113579,
"grad_norm": 0.6039660355798939,
"learning_rate": 4.221854304635762e-06,
"loss": 0.9314,
"mean_token_accuracy": 0.7637597799301148,
"num_tokens": 234909727.0,
"step": 2650
},
{
"entropy": 0.99853515625,
"epoch": 0.3635618123419668,
"grad_norm": 0.6324720460448177,
"learning_rate": 4.218331689446245e-06,
"loss": 0.9884,
"mean_token_accuracy": 0.7578380532562733,
"num_tokens": 235799025.0,
"step": 2660
},
{
"entropy": 0.91650390625,
"epoch": 0.36492858607257567,
"grad_norm": 0.6272162900736367,
"learning_rate": 4.2148090742567284e-06,
"loss": 0.8822,
"mean_token_accuracy": 0.7754081815481186,
"num_tokens": 236681681.0,
"step": 2670
},
{
"entropy": 0.933447265625,
"epoch": 0.36629535980318456,
"grad_norm": 0.5770988868210146,
"learning_rate": 4.211286459067212e-06,
"loss": 0.9597,
"mean_token_accuracy": 0.7614934556186199,
"num_tokens": 237614496.0,
"step": 2680
},
{
"entropy": 0.908837890625,
"epoch": 0.36766213353379346,
"grad_norm": 0.5670471500989122,
"learning_rate": 4.2077638438776955e-06,
"loss": 0.8745,
"mean_token_accuracy": 0.7757954649627209,
"num_tokens": 238469453.0,
"step": 2690
},
{
"entropy": 0.94296875,
"epoch": 0.36902890726440235,
"grad_norm": 0.591844192260298,
"learning_rate": 4.204241228688178e-06,
"loss": 0.9191,
"mean_token_accuracy": 0.7704470001161099,
"num_tokens": 239444806.0,
"step": 2700
},
{
"entropy": 0.93203125,
"epoch": 0.3703956809950113,
"grad_norm": 0.6216847438157016,
"learning_rate": 4.200718613498662e-06,
"loss": 0.9145,
"mean_token_accuracy": 0.7694410622119904,
"num_tokens": 240347408.0,
"step": 2710
},
{
"entropy": 0.937841796875,
"epoch": 0.3717624547256202,
"grad_norm": 0.6515343516015123,
"learning_rate": 4.1971959983091445e-06,
"loss": 0.9119,
"mean_token_accuracy": 0.768502464145422,
"num_tokens": 241208094.0,
"step": 2720
},
{
"entropy": 0.907568359375,
"epoch": 0.3731292284562291,
"grad_norm": 0.604257517346589,
"learning_rate": 4.193673383119629e-06,
"loss": 0.8838,
"mean_token_accuracy": 0.7726956076920033,
"num_tokens": 242123416.0,
"step": 2730
},
{
"entropy": 0.96474609375,
"epoch": 0.374496002186838,
"grad_norm": 0.6200029347103642,
"learning_rate": 4.190150767930112e-06,
"loss": 0.9944,
"mean_token_accuracy": 0.7549713999032974,
"num_tokens": 243074693.0,
"step": 2740
},
{
"entropy": 0.971875,
"epoch": 0.3758627759174469,
"grad_norm": 0.6644203254047566,
"learning_rate": 4.186628152740595e-06,
"loss": 0.9406,
"mean_token_accuracy": 0.765952930599451,
"num_tokens": 243981066.0,
"step": 2750
},
{
"entropy": 0.938623046875,
"epoch": 0.3772295496480558,
"grad_norm": 0.6270559817509609,
"learning_rate": 4.183105537551078e-06,
"loss": 0.9132,
"mean_token_accuracy": 0.766964165866375,
"num_tokens": 244865595.0,
"step": 2760
},
{
"entropy": 0.97841796875,
"epoch": 0.37859632337866467,
"grad_norm": 0.6576906750058016,
"learning_rate": 4.179582922361561e-06,
"loss": 0.9822,
"mean_token_accuracy": 0.7570278130471706,
"num_tokens": 245726577.0,
"step": 2770
},
{
"entropy": 0.95576171875,
"epoch": 0.37996309710927356,
"grad_norm": 0.629951018374748,
"learning_rate": 4.176060307172045e-06,
"loss": 0.9552,
"mean_token_accuracy": 0.7589517496526241,
"num_tokens": 246639968.0,
"step": 2780
},
{
"entropy": 0.95888671875,
"epoch": 0.38132987083988246,
"grad_norm": 0.629159073964531,
"learning_rate": 4.1725376919825285e-06,
"loss": 0.9293,
"mean_token_accuracy": 0.76610107421875,
"num_tokens": 247494604.0,
"step": 2790
},
{
"entropy": 0.948095703125,
"epoch": 0.38269664457049135,
"grad_norm": 0.6364658767762714,
"learning_rate": 4.169015076793011e-06,
"loss": 0.9359,
"mean_token_accuracy": 0.7646566644310951,
"num_tokens": 248346580.0,
"step": 2800
},
{
"entropy": 0.9078125,
"epoch": 0.38406341830110025,
"grad_norm": 0.6687404673934824,
"learning_rate": 4.165492461603495e-06,
"loss": 0.9087,
"mean_token_accuracy": 0.7701289564371109,
"num_tokens": 249243321.0,
"step": 2810
},
{
"entropy": 0.92724609375,
"epoch": 0.38543019203170914,
"grad_norm": 0.6511792767816049,
"learning_rate": 4.161969846413978e-06,
"loss": 0.9186,
"mean_token_accuracy": 0.7685821935534477,
"num_tokens": 250100768.0,
"step": 2820
},
{
"entropy": 0.883935546875,
"epoch": 0.38679696576231803,
"grad_norm": 0.5901118321494334,
"learning_rate": 4.158447231224461e-06,
"loss": 0.8764,
"mean_token_accuracy": 0.7772910922765732,
"num_tokens": 251046334.0,
"step": 2830
},
{
"entropy": 0.941357421875,
"epoch": 0.38816373949292693,
"grad_norm": 0.5673234995641772,
"learning_rate": 4.1549246160349445e-06,
"loss": 0.9307,
"mean_token_accuracy": 0.7642626143991947,
"num_tokens": 251967634.0,
"step": 2840
},
{
"entropy": 0.92421875,
"epoch": 0.3895305132235358,
"grad_norm": 0.61705169376736,
"learning_rate": 4.151402000845428e-06,
"loss": 0.9309,
"mean_token_accuracy": 0.7648583747446537,
"num_tokens": 252873963.0,
"step": 2850
},
{
"entropy": 0.90546875,
"epoch": 0.3908972869541447,
"grad_norm": 0.5780984617576146,
"learning_rate": 4.147879385655912e-06,
"loss": 0.8923,
"mean_token_accuracy": 0.7733296319842339,
"num_tokens": 253752744.0,
"step": 2860
},
{
"entropy": 0.9291015625,
"epoch": 0.3922640606847536,
"grad_norm": 0.7392146589614133,
"learning_rate": 4.144356770466394e-06,
"loss": 0.9266,
"mean_token_accuracy": 0.7651013866066932,
"num_tokens": 254605874.0,
"step": 2870
},
{
"entropy": 1.001025390625,
"epoch": 0.39363083441536256,
"grad_norm": 0.6353917858047504,
"learning_rate": 4.140834155276878e-06,
"loss": 0.9804,
"mean_token_accuracy": 0.75335423797369,
"num_tokens": 255509596.0,
"step": 2880
},
{
"entropy": 0.948974609375,
"epoch": 0.39499760814597146,
"grad_norm": 0.6125466560295126,
"learning_rate": 4.137311540087361e-06,
"loss": 0.9404,
"mean_token_accuracy": 0.7652094975113869,
"num_tokens": 256444844.0,
"step": 2890
},
{
"entropy": 0.951416015625,
"epoch": 0.39636438187658035,
"grad_norm": 0.6721320561842112,
"learning_rate": 4.133788924897845e-06,
"loss": 0.9439,
"mean_token_accuracy": 0.7617141731083393,
"num_tokens": 257337194.0,
"step": 2900
},
{
"entropy": 0.895849609375,
"epoch": 0.39773115560718925,
"grad_norm": 0.5832193202133057,
"learning_rate": 4.130266309708328e-06,
"loss": 0.8768,
"mean_token_accuracy": 0.777929250895977,
"num_tokens": 258224564.0,
"step": 2910
},
{
"entropy": 0.961767578125,
"epoch": 0.39909792933779814,
"grad_norm": 0.657611207731794,
"learning_rate": 4.126743694518811e-06,
"loss": 0.9557,
"mean_token_accuracy": 0.7607496783137322,
"num_tokens": 259129739.0,
"step": 2920
},
{
"entropy": 0.952978515625,
"epoch": 0.40046470306840704,
"grad_norm": 0.6653381557647762,
"learning_rate": 4.123221079329294e-06,
"loss": 0.9176,
"mean_token_accuracy": 0.768390067666769,
"num_tokens": 259960983.0,
"step": 2930
},
{
"entropy": 0.972314453125,
"epoch": 0.40183147679901593,
"grad_norm": 0.6357799561583999,
"learning_rate": 4.1196984641397774e-06,
"loss": 0.9606,
"mean_token_accuracy": 0.75957006290555,
"num_tokens": 260822212.0,
"step": 2940
},
{
"entropy": 0.925927734375,
"epoch": 0.4031982505296248,
"grad_norm": 0.6197020936552756,
"learning_rate": 4.116175848950261e-06,
"loss": 0.8836,
"mean_token_accuracy": 0.7761113122105598,
"num_tokens": 261702691.0,
"step": 2950
},
{
"entropy": 0.88828125,
"epoch": 0.4045650242602337,
"grad_norm": 0.7701966542676406,
"learning_rate": 4.1126532337607445e-06,
"loss": 0.8645,
"mean_token_accuracy": 0.7774319730699062,
"num_tokens": 262516705.0,
"step": 2960
},
{
"entropy": 0.9758544921875,
"epoch": 0.4059317979908426,
"grad_norm": 0.7538027060619457,
"learning_rate": 4.109130618571227e-06,
"loss": 0.9831,
"mean_token_accuracy": 0.7561600483953953,
"num_tokens": 263406987.0,
"step": 2970
},
{
"entropy": 0.955029296875,
"epoch": 0.4072985717214515,
"grad_norm": 0.6884281376075932,
"learning_rate": 4.105608003381711e-06,
"loss": 0.9342,
"mean_token_accuracy": 0.7619048729538918,
"num_tokens": 264280538.0,
"step": 2980
},
{
"entropy": 0.908935546875,
"epoch": 0.4086653454520604,
"grad_norm": 0.6493785667832263,
"learning_rate": 4.102085388192194e-06,
"loss": 0.8847,
"mean_token_accuracy": 0.7729057133197784,
"num_tokens": 265155880.0,
"step": 2990
},
{
"entropy": 0.97236328125,
"epoch": 0.4100321191826693,
"grad_norm": 0.7077006706524858,
"learning_rate": 4.098562773002678e-06,
"loss": 0.9773,
"mean_token_accuracy": 0.7554806016385556,
"num_tokens": 266013552.0,
"step": 3000
},
{
"entropy": 0.9083984375,
"epoch": 0.4113988929132782,
"grad_norm": 0.592540567446806,
"learning_rate": 4.095040157813161e-06,
"loss": 0.9052,
"mean_token_accuracy": 0.7711334623396396,
"num_tokens": 266944824.0,
"step": 3010
},
{
"entropy": 0.947998046875,
"epoch": 0.4127656666438871,
"grad_norm": 0.6638717714219059,
"learning_rate": 4.091517542623644e-06,
"loss": 0.9457,
"mean_token_accuracy": 0.7644581459462643,
"num_tokens": 267814564.0,
"step": 3020
},
{
"entropy": 0.991748046875,
"epoch": 0.414132440374496,
"grad_norm": 0.6692940367420767,
"learning_rate": 4.087994927434128e-06,
"loss": 0.9974,
"mean_token_accuracy": 0.7544350862503052,
"num_tokens": 268692641.0,
"step": 3030
},
{
"entropy": 0.97197265625,
"epoch": 0.4154992141051049,
"grad_norm": 0.6318191748098028,
"learning_rate": 4.08447231224461e-06,
"loss": 0.9599,
"mean_token_accuracy": 0.7599162928760052,
"num_tokens": 269542807.0,
"step": 3040
},
{
"entropy": 0.9041015625,
"epoch": 0.4168659878357138,
"grad_norm": 1.0098964279779679,
"learning_rate": 4.080949697055094e-06,
"loss": 0.8723,
"mean_token_accuracy": 0.7763818196952343,
"num_tokens": 270383078.0,
"step": 3050
},
{
"entropy": 0.914990234375,
"epoch": 0.4182327615663227,
"grad_norm": 0.6591935080322798,
"learning_rate": 4.0774270818655775e-06,
"loss": 0.9065,
"mean_token_accuracy": 0.7690801069140434,
"num_tokens": 271284999.0,
"step": 3060
},
{
"entropy": 0.972021484375,
"epoch": 0.4195995352969316,
"grad_norm": 0.6623455605054593,
"learning_rate": 4.073904466676061e-06,
"loss": 0.9719,
"mean_token_accuracy": 0.756262069195509,
"num_tokens": 272148180.0,
"step": 3070
},
{
"entropy": 0.933349609375,
"epoch": 0.4209663090275405,
"grad_norm": 0.7155422834935311,
"learning_rate": 4.070381851486544e-06,
"loss": 0.9434,
"mean_token_accuracy": 0.7653803542256356,
"num_tokens": 273083027.0,
"step": 3080
},
{
"entropy": 0.926513671875,
"epoch": 0.4223330827581494,
"grad_norm": 0.6234773948130141,
"learning_rate": 4.066859236297027e-06,
"loss": 0.9271,
"mean_token_accuracy": 0.7681013584136963,
"num_tokens": 273982751.0,
"step": 3090
},
{
"entropy": 0.92900390625,
"epoch": 0.4236998564887583,
"grad_norm": 0.6595140214939107,
"learning_rate": 4.06333662110751e-06,
"loss": 0.925,
"mean_token_accuracy": 0.7678802318871021,
"num_tokens": 274840518.0,
"step": 3100
},
{
"entropy": 0.9912109375,
"epoch": 0.4250666302193672,
"grad_norm": 0.6266763475050634,
"learning_rate": 4.059814005917994e-06,
"loss": 0.9925,
"mean_token_accuracy": 0.7562396325170994,
"num_tokens": 275767278.0,
"step": 3110
},
{
"entropy": 0.946728515625,
"epoch": 0.4264334039499761,
"grad_norm": 0.6739894499189977,
"learning_rate": 4.056291390728477e-06,
"loss": 0.9222,
"mean_token_accuracy": 0.7680151842534542,
"num_tokens": 276644137.0,
"step": 3120
},
{
"entropy": 0.99033203125,
"epoch": 0.427800177680585,
"grad_norm": 0.7663357417701647,
"learning_rate": 4.052768775538961e-06,
"loss": 0.9863,
"mean_token_accuracy": 0.7567301906645298,
"num_tokens": 277549426.0,
"step": 3130
},
{
"entropy": 0.949560546875,
"epoch": 0.4291669514111939,
"grad_norm": 0.6935413982591924,
"learning_rate": 4.049246160349444e-06,
"loss": 0.9378,
"mean_token_accuracy": 0.764452163875103,
"num_tokens": 278466337.0,
"step": 3140
},
{
"entropy": 0.914111328125,
"epoch": 0.43053372514180277,
"grad_norm": 0.5703061763861456,
"learning_rate": 4.045723545159927e-06,
"loss": 0.895,
"mean_token_accuracy": 0.7725748740136623,
"num_tokens": 279324947.0,
"step": 3150
},
{
"entropy": 0.89765625,
"epoch": 0.43190049887241166,
"grad_norm": 0.584823512470594,
"learning_rate": 4.04220092997041e-06,
"loss": 0.8859,
"mean_token_accuracy": 0.7756792724132537,
"num_tokens": 280225654.0,
"step": 3160
},
{
"entropy": 0.91796875,
"epoch": 0.43326727260302056,
"grad_norm": 0.6040007433481087,
"learning_rate": 4.038678314780894e-06,
"loss": 0.9159,
"mean_token_accuracy": 0.7672983542084694,
"num_tokens": 281153133.0,
"step": 3170
},
{
"entropy": 0.961572265625,
"epoch": 0.43463404633362945,
"grad_norm": 0.6061804375603362,
"learning_rate": 4.0351556995913775e-06,
"loss": 0.9477,
"mean_token_accuracy": 0.7634034037590027,
"num_tokens": 282083125.0,
"step": 3180
},
{
"entropy": 0.9171875,
"epoch": 0.43600082006423835,
"grad_norm": 0.5966891828215897,
"learning_rate": 4.03163308440186e-06,
"loss": 0.9225,
"mean_token_accuracy": 0.7688467107713223,
"num_tokens": 282960536.0,
"step": 3190
},
{
"entropy": 0.9580078125,
"epoch": 0.43736759379484724,
"grad_norm": 0.6312202554970291,
"learning_rate": 4.028110469212344e-06,
"loss": 0.9392,
"mean_token_accuracy": 0.7657306291162967,
"num_tokens": 283800850.0,
"step": 3200
},
{
"entropy": 0.935791015625,
"epoch": 0.43873436752545614,
"grad_norm": 0.5893495010621845,
"learning_rate": 4.0245878540228264e-06,
"loss": 0.928,
"mean_token_accuracy": 0.7662612333893776,
"num_tokens": 284680091.0,
"step": 3210
},
{
"entropy": 0.965380859375,
"epoch": 0.44010114125606503,
"grad_norm": 0.6854442743670638,
"learning_rate": 4.02106523883331e-06,
"loss": 0.9576,
"mean_token_accuracy": 0.7616727530956269,
"num_tokens": 285538822.0,
"step": 3220
},
{
"entropy": 0.919091796875,
"epoch": 0.441467914986674,
"grad_norm": 0.6411033865114989,
"learning_rate": 4.0175426236437935e-06,
"loss": 0.9156,
"mean_token_accuracy": 0.7714936822652817,
"num_tokens": 286390073.0,
"step": 3230
},
{
"entropy": 0.9267822265625,
"epoch": 0.4428346887172829,
"grad_norm": 0.6150558944986921,
"learning_rate": 4.014020008454277e-06,
"loss": 0.9042,
"mean_token_accuracy": 0.7717894598841667,
"num_tokens": 287286685.0,
"step": 3240
},
{
"entropy": 0.921875,
"epoch": 0.44420146244789177,
"grad_norm": 0.5748828274832535,
"learning_rate": 4.01049739326476e-06,
"loss": 0.902,
"mean_token_accuracy": 0.7682955160737037,
"num_tokens": 288121053.0,
"step": 3250
},
{
"entropy": 0.927685546875,
"epoch": 0.44556823617850067,
"grad_norm": 0.6211022093497333,
"learning_rate": 4.006974778075243e-06,
"loss": 0.9146,
"mean_token_accuracy": 0.7702526494860649,
"num_tokens": 289016233.0,
"step": 3260
},
{
"entropy": 0.93466796875,
"epoch": 0.44693500990910956,
"grad_norm": 0.5831420057398989,
"learning_rate": 4.003452162885727e-06,
"loss": 0.9552,
"mean_token_accuracy": 0.7615006528794765,
"num_tokens": 289937154.0,
"step": 3270
},
{
"entropy": 0.893994140625,
"epoch": 0.44830178363971845,
"grad_norm": 0.7073051005502261,
"learning_rate": 3.99992954769621e-06,
"loss": 0.8949,
"mean_token_accuracy": 0.77388516664505,
"num_tokens": 290823440.0,
"step": 3280
},
{
"entropy": 0.914697265625,
"epoch": 0.44966855737032735,
"grad_norm": 0.6124447473403681,
"learning_rate": 3.996406932506693e-06,
"loss": 0.9058,
"mean_token_accuracy": 0.7719897344708443,
"num_tokens": 291744078.0,
"step": 3290
},
{
"entropy": 0.911376953125,
"epoch": 0.45103533110093624,
"grad_norm": 0.6681487654668247,
"learning_rate": 3.992884317317177e-06,
"loss": 0.9149,
"mean_token_accuracy": 0.771390688419342,
"num_tokens": 292683000.0,
"step": 3300
},
{
"entropy": 0.969189453125,
"epoch": 0.45240210483154514,
"grad_norm": 0.67043619548697,
"learning_rate": 3.98936170212766e-06,
"loss": 0.9474,
"mean_token_accuracy": 0.7647609241306782,
"num_tokens": 293591908.0,
"step": 3310
},
{
"entropy": 0.901220703125,
"epoch": 0.45376887856215403,
"grad_norm": 0.5947339592913287,
"learning_rate": 3.985839086938143e-06,
"loss": 0.8891,
"mean_token_accuracy": 0.772900614887476,
"num_tokens": 294468616.0,
"step": 3320
},
{
"entropy": 0.933056640625,
"epoch": 0.4551356522927629,
"grad_norm": 0.6355995487223831,
"learning_rate": 3.9823164717486265e-06,
"loss": 0.9376,
"mean_token_accuracy": 0.7642320804297924,
"num_tokens": 295394199.0,
"step": 3330
},
{
"entropy": 0.942333984375,
"epoch": 0.4565024260233718,
"grad_norm": 0.6215092866922419,
"learning_rate": 3.97879385655911e-06,
"loss": 0.9313,
"mean_token_accuracy": 0.7665700323879718,
"num_tokens": 296273215.0,
"step": 3340
},
{
"entropy": 0.94169921875,
"epoch": 0.4578691997539807,
"grad_norm": 0.6855196019686018,
"learning_rate": 3.9752712413695936e-06,
"loss": 0.9464,
"mean_token_accuracy": 0.7635574318468571,
"num_tokens": 297148572.0,
"step": 3350
},
{
"entropy": 1.007763671875,
"epoch": 0.4592359734845896,
"grad_norm": 0.6104964412644851,
"learning_rate": 3.971748626180076e-06,
"loss": 0.9956,
"mean_token_accuracy": 0.7542635776102543,
"num_tokens": 298039662.0,
"step": 3360
},
{
"entropy": 0.9405517578125,
"epoch": 0.4606027472151985,
"grad_norm": 0.621003168411592,
"learning_rate": 3.96822601099056e-06,
"loss": 0.9451,
"mean_token_accuracy": 0.7632085181772709,
"num_tokens": 298956252.0,
"step": 3370
},
{
"entropy": 0.923681640625,
"epoch": 0.4619695209458074,
"grad_norm": 0.89800344791506,
"learning_rate": 3.9647033958010425e-06,
"loss": 0.9331,
"mean_token_accuracy": 0.7664727859199048,
"num_tokens": 299823280.0,
"step": 3380
},
{
"entropy": 0.92607421875,
"epoch": 0.4633362946764163,
"grad_norm": 0.680622657927905,
"learning_rate": 3.961180780611527e-06,
"loss": 0.9353,
"mean_token_accuracy": 0.7659028135240078,
"num_tokens": 300704268.0,
"step": 3390
},
{
"entropy": 0.961865234375,
"epoch": 0.46470306840702524,
"grad_norm": 0.566404033073394,
"learning_rate": 3.95765816542201e-06,
"loss": 0.9476,
"mean_token_accuracy": 0.7629028640687465,
"num_tokens": 301638457.0,
"step": 3400
},
{
"entropy": 0.933154296875,
"epoch": 0.46606984213763414,
"grad_norm": 0.684905998138667,
"learning_rate": 3.954135550232493e-06,
"loss": 0.8795,
"mean_token_accuracy": 0.7734334848821163,
"num_tokens": 302495464.0,
"step": 3410
},
{
"entropy": 0.959521484375,
"epoch": 0.46743661586824303,
"grad_norm": 0.6240396289490382,
"learning_rate": 3.950612935042976e-06,
"loss": 0.9429,
"mean_token_accuracy": 0.7622483640909195,
"num_tokens": 303376908.0,
"step": 3420
},
{
"entropy": 0.908447265625,
"epoch": 0.4688033895988519,
"grad_norm": 0.6339265931155392,
"learning_rate": 3.947090319853459e-06,
"loss": 0.8747,
"mean_token_accuracy": 0.7746687076985836,
"num_tokens": 304255583.0,
"step": 3430
},
{
"entropy": 0.980322265625,
"epoch": 0.4701701633294608,
"grad_norm": 0.670148399571312,
"learning_rate": 3.943567704663943e-06,
"loss": 0.9683,
"mean_token_accuracy": 0.7581922888755799,
"num_tokens": 305117375.0,
"step": 3440
},
{
"entropy": 0.933349609375,
"epoch": 0.4715369370600697,
"grad_norm": 0.6860375690363979,
"learning_rate": 3.9400450894744265e-06,
"loss": 0.9195,
"mean_token_accuracy": 0.7671180970966816,
"num_tokens": 305964787.0,
"step": 3450
},
{
"entropy": 0.95810546875,
"epoch": 0.4729037107906786,
"grad_norm": 0.6118059553246599,
"learning_rate": 3.936522474284909e-06,
"loss": 0.9822,
"mean_token_accuracy": 0.7619119726121426,
"num_tokens": 306874371.0,
"step": 3460
},
{
"entropy": 0.92919921875,
"epoch": 0.4742704845212875,
"grad_norm": 0.736419686876982,
"learning_rate": 3.932999859095393e-06,
"loss": 0.9372,
"mean_token_accuracy": 0.7662431091070175,
"num_tokens": 307828178.0,
"step": 3470
},
{
"entropy": 0.919482421875,
"epoch": 0.4756372582518964,
"grad_norm": 0.6466942202353921,
"learning_rate": 3.929477243905876e-06,
"loss": 0.9005,
"mean_token_accuracy": 0.772038335353136,
"num_tokens": 308715305.0,
"step": 3480
},
{
"entropy": 0.961962890625,
"epoch": 0.4770040319825053,
"grad_norm": 0.7625891012572045,
"learning_rate": 3.925954628716359e-06,
"loss": 0.9462,
"mean_token_accuracy": 0.7647375896573066,
"num_tokens": 309596173.0,
"step": 3490
},
{
"entropy": 0.963818359375,
"epoch": 0.4783708057131142,
"grad_norm": 0.6515214505790619,
"learning_rate": 3.9224320135268425e-06,
"loss": 0.9379,
"mean_token_accuracy": 0.7619348064064979,
"num_tokens": 310464726.0,
"step": 3500
},
{
"entropy": 0.936572265625,
"epoch": 0.4797375794437231,
"grad_norm": 0.6712621237889889,
"learning_rate": 3.918909398337326e-06,
"loss": 0.9275,
"mean_token_accuracy": 0.7664429843425751,
"num_tokens": 311349143.0,
"step": 3510
},
{
"entropy": 0.93505859375,
"epoch": 0.481104353174332,
"grad_norm": 0.7309786882984751,
"learning_rate": 3.91538678314781e-06,
"loss": 0.908,
"mean_token_accuracy": 0.7704219534993172,
"num_tokens": 312240199.0,
"step": 3520
},
{
"entropy": 0.942041015625,
"epoch": 0.4824711269049409,
"grad_norm": 0.6407581938399605,
"learning_rate": 3.911864167958292e-06,
"loss": 0.9314,
"mean_token_accuracy": 0.7651642322540283,
"num_tokens": 313135937.0,
"step": 3530
},
{
"entropy": 0.93662109375,
"epoch": 0.48383790063554977,
"grad_norm": 0.6307022023108446,
"learning_rate": 3.908341552768776e-06,
"loss": 0.9165,
"mean_token_accuracy": 0.7664061166346073,
"num_tokens": 314018252.0,
"step": 3540
},
{
"entropy": 0.96005859375,
"epoch": 0.48520467436615866,
"grad_norm": 31.70864904605551,
"learning_rate": 3.9048189375792586e-06,
"loss": 0.9303,
"mean_token_accuracy": 0.7643258988857269,
"num_tokens": 314871616.0,
"step": 3550
},
{
"entropy": 0.959326171875,
"epoch": 0.48657144809676756,
"grad_norm": 0.6427744986093868,
"learning_rate": 3.901296322389743e-06,
"loss": 0.957,
"mean_token_accuracy": 0.7620796255767346,
"num_tokens": 315729836.0,
"step": 3560
},
{
"entropy": 0.955126953125,
"epoch": 0.48793822182737645,
"grad_norm": 0.6415277387620777,
"learning_rate": 3.897773707200226e-06,
"loss": 0.9373,
"mean_token_accuracy": 0.7609451964497567,
"num_tokens": 316602383.0,
"step": 3570
},
{
"entropy": 0.958349609375,
"epoch": 0.4893049955579854,
"grad_norm": 0.6799372853632959,
"learning_rate": 3.894251092010709e-06,
"loss": 0.9464,
"mean_token_accuracy": 0.7582819640636445,
"num_tokens": 317513326.0,
"step": 3580
},
{
"entropy": 0.899609375,
"epoch": 0.4906717692885943,
"grad_norm": 0.6091434047274472,
"learning_rate": 3.890728476821192e-06,
"loss": 0.8784,
"mean_token_accuracy": 0.7746723227202892,
"num_tokens": 318374110.0,
"step": 3590
},
{
"entropy": 0.894921875,
"epoch": 0.4920385430192032,
"grad_norm": 0.6734637751067465,
"learning_rate": 3.8872058616316755e-06,
"loss": 0.8527,
"mean_token_accuracy": 0.780231100320816,
"num_tokens": 319229245.0,
"step": 3600
},
{
"entropy": 0.924072265625,
"epoch": 0.4934053167498121,
"grad_norm": 0.637243344629856,
"learning_rate": 3.883683246442159e-06,
"loss": 0.9115,
"mean_token_accuracy": 0.7666994594037533,
"num_tokens": 320108614.0,
"step": 3610
},
{
"entropy": 0.9392822265625,
"epoch": 0.494772090480421,
"grad_norm": 0.6526394668290849,
"learning_rate": 3.8801606312526426e-06,
"loss": 0.9308,
"mean_token_accuracy": 0.7652523182332516,
"num_tokens": 321041845.0,
"step": 3620
},
{
"entropy": 0.879150390625,
"epoch": 0.4961388642110299,
"grad_norm": 0.6094219461149726,
"learning_rate": 3.876638016063125e-06,
"loss": 0.8368,
"mean_token_accuracy": 0.7845753863453865,
"num_tokens": 321942482.0,
"step": 3630
},
{
"entropy": 0.916796875,
"epoch": 0.49750563794163877,
"grad_norm": 0.6383933244034597,
"learning_rate": 3.873115400873609e-06,
"loss": 0.8899,
"mean_token_accuracy": 0.7745092153549195,
"num_tokens": 322786504.0,
"step": 3640
},
{
"entropy": 0.93232421875,
"epoch": 0.49887241167224766,
"grad_norm": 0.6359312854952249,
"learning_rate": 3.869592785684092e-06,
"loss": 0.9344,
"mean_token_accuracy": 0.7665737830102444,
"num_tokens": 323673486.0,
"step": 3650
},
{
"entropy": 0.9482421875,
"epoch": 0.5002391854028566,
"grad_norm": 0.5518157873928607,
"learning_rate": 3.866070170494575e-06,
"loss": 0.9274,
"mean_token_accuracy": 0.7674518674612045,
"num_tokens": 324539268.0,
"step": 3660
},
{
"entropy": 0.95625,
"epoch": 0.5016059591334655,
"grad_norm": 0.6951710602569751,
"learning_rate": 3.862547555305059e-06,
"loss": 0.9602,
"mean_token_accuracy": 0.7630286522209644,
"num_tokens": 325467313.0,
"step": 3670
},
{
"entropy": 0.95048828125,
"epoch": 0.5029727328640744,
"grad_norm": 0.708985472965003,
"learning_rate": 3.859024940115542e-06,
"loss": 0.9409,
"mean_token_accuracy": 0.765538503229618,
"num_tokens": 326360349.0,
"step": 3680
},
{
"entropy": 0.906689453125,
"epoch": 0.5043395065946833,
"grad_norm": 0.5954023421674405,
"learning_rate": 3.855502324926026e-06,
"loss": 0.8859,
"mean_token_accuracy": 0.7762729018926621,
"num_tokens": 327295014.0,
"step": 3690
},
{
"entropy": 0.91943359375,
"epoch": 0.5057062803252922,
"grad_norm": 0.6409225488523193,
"learning_rate": 3.851979709736508e-06,
"loss": 0.9259,
"mean_token_accuracy": 0.7683279164135456,
"num_tokens": 328166096.0,
"step": 3700
},
{
"entropy": 0.947900390625,
"epoch": 0.5070730540559011,
"grad_norm": 0.6255131803297045,
"learning_rate": 3.848457094546992e-06,
"loss": 0.9244,
"mean_token_accuracy": 0.7661083228886127,
"num_tokens": 329044476.0,
"step": 3710
},
{
"entropy": 0.9201171875,
"epoch": 0.50843982778651,
"grad_norm": 0.5741042827719574,
"learning_rate": 3.8449344793574755e-06,
"loss": 0.9011,
"mean_token_accuracy": 0.7717679902911186,
"num_tokens": 329954219.0,
"step": 3720
},
{
"entropy": 0.936083984375,
"epoch": 0.5098066015171189,
"grad_norm": 0.5744736010974905,
"learning_rate": 3.841411864167959e-06,
"loss": 0.9281,
"mean_token_accuracy": 0.7668012723326683,
"num_tokens": 330824306.0,
"step": 3730
},
{
"entropy": 0.95576171875,
"epoch": 0.5111733752477278,
"grad_norm": 0.6632032704169473,
"learning_rate": 3.837889248978442e-06,
"loss": 0.9596,
"mean_token_accuracy": 0.7598929472267628,
"num_tokens": 331697014.0,
"step": 3740
},
{
"entropy": 0.92333984375,
"epoch": 0.5125401489783367,
"grad_norm": 0.6571917121171212,
"learning_rate": 3.834366633788925e-06,
"loss": 0.9171,
"mean_token_accuracy": 0.7697374254465104,
"num_tokens": 332591752.0,
"step": 3750
},
{
"entropy": 0.938525390625,
"epoch": 0.5139069227089456,
"grad_norm": 0.7123039651645402,
"learning_rate": 3.830844018599408e-06,
"loss": 0.9242,
"mean_token_accuracy": 0.7646553538739681,
"num_tokens": 333428436.0,
"step": 3760
},
{
"entropy": 0.9556640625,
"epoch": 0.5152736964395545,
"grad_norm": 0.6065846947101122,
"learning_rate": 3.8273214034098915e-06,
"loss": 0.9343,
"mean_token_accuracy": 0.7637577272951603,
"num_tokens": 334312680.0,
"step": 3770
},
{
"entropy": 0.96875,
"epoch": 0.5166404701701633,
"grad_norm": 0.6396007886075189,
"learning_rate": 3.823798788220375e-06,
"loss": 0.9562,
"mean_token_accuracy": 0.7598132811486721,
"num_tokens": 335170349.0,
"step": 3780
},
{
"entropy": 0.903515625,
"epoch": 0.5180072439007722,
"grad_norm": 0.6730476330149454,
"learning_rate": 3.820276173030859e-06,
"loss": 0.9077,
"mean_token_accuracy": 0.7709536470472813,
"num_tokens": 336022826.0,
"step": 3790
},
{
"entropy": 0.91708984375,
"epoch": 0.5193740176313811,
"grad_norm": 1.0797856779645083,
"learning_rate": 3.816753557841341e-06,
"loss": 0.9175,
"mean_token_accuracy": 0.7672334156930447,
"num_tokens": 336888400.0,
"step": 3800
},
{
"entropy": 0.93994140625,
"epoch": 0.52074079136199,
"grad_norm": 0.6265567248535575,
"learning_rate": 3.813230942651825e-06,
"loss": 0.9219,
"mean_token_accuracy": 0.7664801649749279,
"num_tokens": 337726810.0,
"step": 3810
},
{
"entropy": 0.92109375,
"epoch": 0.5221075650925989,
"grad_norm": 0.6017537156121543,
"learning_rate": 3.809708327462308e-06,
"loss": 0.9085,
"mean_token_accuracy": 0.7713609606027603,
"num_tokens": 338587129.0,
"step": 3820
},
{
"entropy": 0.94150390625,
"epoch": 0.5234743388232078,
"grad_norm": 0.6090078848921925,
"learning_rate": 3.806185712272792e-06,
"loss": 0.9454,
"mean_token_accuracy": 0.7637524455785751,
"num_tokens": 339495137.0,
"step": 3830
},
{
"entropy": 0.99638671875,
"epoch": 0.5248411125538167,
"grad_norm": 0.6394959572948704,
"learning_rate": 3.802663097083275e-06,
"loss": 1.0165,
"mean_token_accuracy": 0.7499555930495262,
"num_tokens": 340326128.0,
"step": 3840
},
{
"entropy": 0.9384765625,
"epoch": 0.5262078862844256,
"grad_norm": 0.6424615838251556,
"learning_rate": 3.799140481893758e-06,
"loss": 0.9146,
"mean_token_accuracy": 0.7683775432407856,
"num_tokens": 341183998.0,
"step": 3850
},
{
"entropy": 0.91318359375,
"epoch": 0.5275746600150345,
"grad_norm": 0.6192876218278857,
"learning_rate": 3.7956178667042413e-06,
"loss": 0.8863,
"mean_token_accuracy": 0.7737067475914955,
"num_tokens": 342030730.0,
"step": 3860
},
{
"entropy": 0.89482421875,
"epoch": 0.5289414337456434,
"grad_norm": 0.6103823626970045,
"learning_rate": 3.792095251514725e-06,
"loss": 0.8868,
"mean_token_accuracy": 0.7719895906746388,
"num_tokens": 342928649.0,
"step": 3870
},
{
"entropy": 0.91640625,
"epoch": 0.5303082074762523,
"grad_norm": 0.6202437584901601,
"learning_rate": 3.788572636325208e-06,
"loss": 0.9051,
"mean_token_accuracy": 0.7725112460553646,
"num_tokens": 343852240.0,
"step": 3880
},
{
"entropy": 0.9005126953125,
"epoch": 0.5316749812068612,
"grad_norm": 0.5937632555452932,
"learning_rate": 3.7850500211356916e-06,
"loss": 0.8775,
"mean_token_accuracy": 0.776013159006834,
"num_tokens": 344750196.0,
"step": 3890
},
{
"entropy": 0.94921875,
"epoch": 0.5330417549374701,
"grad_norm": 0.6163163642918142,
"learning_rate": 3.7815274059461747e-06,
"loss": 0.9455,
"mean_token_accuracy": 0.7627267226576805,
"num_tokens": 345621623.0,
"step": 3900
},
{
"entropy": 0.924169921875,
"epoch": 0.534408528668079,
"grad_norm": 0.586624186788516,
"learning_rate": 3.7780047907566582e-06,
"loss": 0.9266,
"mean_token_accuracy": 0.7681899711489677,
"num_tokens": 346526744.0,
"step": 3910
},
{
"entropy": 0.9158447265625,
"epoch": 0.5357753023986879,
"grad_norm": 0.6104433763014383,
"learning_rate": 3.7744821755671413e-06,
"loss": 0.8953,
"mean_token_accuracy": 0.7751196049153805,
"num_tokens": 347436449.0,
"step": 3920
},
{
"entropy": 0.94462890625,
"epoch": 0.5371420761292968,
"grad_norm": 0.6251300424577241,
"learning_rate": 3.7709595603776245e-06,
"loss": 0.9616,
"mean_token_accuracy": 0.764511426538229,
"num_tokens": 348331917.0,
"step": 3930
},
{
"entropy": 0.959033203125,
"epoch": 0.5385088498599057,
"grad_norm": 0.7410197584618738,
"learning_rate": 3.767436945188108e-06,
"loss": 0.9504,
"mean_token_accuracy": 0.7622664347290993,
"num_tokens": 349199022.0,
"step": 3940
},
{
"entropy": 0.8919189453125,
"epoch": 0.5398756235905146,
"grad_norm": 0.6297088911838186,
"learning_rate": 3.7639143299985916e-06,
"loss": 0.8912,
"mean_token_accuracy": 0.7730370059609413,
"num_tokens": 350085246.0,
"step": 3950
},
{
"entropy": 0.945361328125,
"epoch": 0.5412423973211234,
"grad_norm": 0.6052192594879535,
"learning_rate": 3.7603917148090747e-06,
"loss": 0.9328,
"mean_token_accuracy": 0.7670720778405666,
"num_tokens": 350974056.0,
"step": 3960
},
{
"entropy": 0.92294921875,
"epoch": 0.5426091710517323,
"grad_norm": 0.6431427067784747,
"learning_rate": 3.756869099619558e-06,
"loss": 0.9042,
"mean_token_accuracy": 0.7732612743973732,
"num_tokens": 351864137.0,
"step": 3970
},
{
"entropy": 0.911474609375,
"epoch": 0.5439759447823412,
"grad_norm": 0.6121995923760664,
"learning_rate": 3.753346484430041e-06,
"loss": 0.9003,
"mean_token_accuracy": 0.7718327678740025,
"num_tokens": 352741822.0,
"step": 3980
},
{
"entropy": 0.91669921875,
"epoch": 0.5453427185129501,
"grad_norm": 0.644071175841464,
"learning_rate": 3.749823869240524e-06,
"loss": 0.8921,
"mean_token_accuracy": 0.7738240845501423,
"num_tokens": 353612932.0,
"step": 3990
},
{
"entropy": 0.897021484375,
"epoch": 0.546709492243559,
"grad_norm": 0.6581030559062012,
"learning_rate": 3.746301254051008e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7728954032063484,
"num_tokens": 354489469.0,
"step": 4000
},
{
"entropy": 0.905029296875,
"epoch": 0.548076265974168,
"grad_norm": 0.6049559001459911,
"learning_rate": 3.742778638861491e-06,
"loss": 0.8812,
"mean_token_accuracy": 0.7739376217126847,
"num_tokens": 355362081.0,
"step": 4010
},
{
"entropy": 0.93583984375,
"epoch": 0.5494430397047769,
"grad_norm": 0.6168104242155126,
"learning_rate": 3.7392560236719743e-06,
"loss": 0.9237,
"mean_token_accuracy": 0.7661426708102226,
"num_tokens": 356206559.0,
"step": 4020
},
{
"entropy": 0.88359375,
"epoch": 0.5508098134353858,
"grad_norm": 0.6204539663443699,
"learning_rate": 3.7357334084824574e-06,
"loss": 0.8659,
"mean_token_accuracy": 0.778882448375225,
"num_tokens": 357084489.0,
"step": 4030
},
{
"entropy": 0.908642578125,
"epoch": 0.5521765871659947,
"grad_norm": 0.6631942913325324,
"learning_rate": 3.732210793292941e-06,
"loss": 0.8847,
"mean_token_accuracy": 0.772475290298462,
"num_tokens": 357975361.0,
"step": 4040
},
{
"entropy": 0.937646484375,
"epoch": 0.5535433608966036,
"grad_norm": 0.6548668300219025,
"learning_rate": 3.7286881781034245e-06,
"loss": 0.9325,
"mean_token_accuracy": 0.7647103421390057,
"num_tokens": 358878614.0,
"step": 4050
},
{
"entropy": 0.909375,
"epoch": 0.5549101346272125,
"grad_norm": 0.590937427979614,
"learning_rate": 3.7251655629139076e-06,
"loss": 0.9124,
"mean_token_accuracy": 0.7692991971969605,
"num_tokens": 359801961.0,
"step": 4060
},
{
"entropy": 0.955126953125,
"epoch": 0.5562769083578214,
"grad_norm": 0.6053080704364314,
"learning_rate": 3.7216429477243907e-06,
"loss": 0.9514,
"mean_token_accuracy": 0.7586136005818844,
"num_tokens": 360684656.0,
"step": 4070
},
{
"entropy": 0.91943359375,
"epoch": 0.5576436820884303,
"grad_norm": 0.5951973660440234,
"learning_rate": 3.7181203325348743e-06,
"loss": 0.9119,
"mean_token_accuracy": 0.7701109416782856,
"num_tokens": 361604630.0,
"step": 4080
},
{
"entropy": 0.9647705078125,
"epoch": 0.5590104558190392,
"grad_norm": 0.688985301600651,
"learning_rate": 3.7145977173453574e-06,
"loss": 0.9572,
"mean_token_accuracy": 0.7595103152096272,
"num_tokens": 362453556.0,
"step": 4090
},
{
"entropy": 0.931884765625,
"epoch": 0.5603772295496481,
"grad_norm": 0.6287442931955897,
"learning_rate": 3.7110751021558405e-06,
"loss": 0.921,
"mean_token_accuracy": 0.7663753658533097,
"num_tokens": 363318401.0,
"step": 4100
},
{
"entropy": 0.903515625,
"epoch": 0.561744003280257,
"grad_norm": 0.6417338626167656,
"learning_rate": 3.7075524869663245e-06,
"loss": 0.8781,
"mean_token_accuracy": 0.7748869620263577,
"num_tokens": 364242506.0,
"step": 4110
},
{
"entropy": 0.9137939453125,
"epoch": 0.5631107770108659,
"grad_norm": 0.6059655472130787,
"learning_rate": 3.7040298717768076e-06,
"loss": 0.8961,
"mean_token_accuracy": 0.7746951416134834,
"num_tokens": 365155052.0,
"step": 4120
},
{
"entropy": 0.958544921875,
"epoch": 0.5644775507414748,
"grad_norm": 0.6125381193137558,
"learning_rate": 3.7005072565872908e-06,
"loss": 0.936,
"mean_token_accuracy": 0.7637355111539363,
"num_tokens": 366080792.0,
"step": 4130
},
{
"entropy": 0.93642578125,
"epoch": 0.5658443244720837,
"grad_norm": 0.6374473268765628,
"learning_rate": 3.696984641397774e-06,
"loss": 0.9403,
"mean_token_accuracy": 0.7638835191726685,
"num_tokens": 366962396.0,
"step": 4140
},
{
"entropy": 0.889697265625,
"epoch": 0.5672110982026926,
"grad_norm": 0.6484555907978654,
"learning_rate": 3.693462026208257e-06,
"loss": 0.8852,
"mean_token_accuracy": 0.7751211635768414,
"num_tokens": 367850915.0,
"step": 4150
},
{
"entropy": 0.93349609375,
"epoch": 0.5685778719333014,
"grad_norm": 0.5983566458396903,
"learning_rate": 3.689939411018741e-06,
"loss": 0.936,
"mean_token_accuracy": 0.7685567669570446,
"num_tokens": 368779867.0,
"step": 4160
},
{
"entropy": 0.882373046875,
"epoch": 0.5699446456639103,
"grad_norm": 0.6406053296389623,
"learning_rate": 3.686416795829224e-06,
"loss": 0.853,
"mean_token_accuracy": 0.7803356438875199,
"num_tokens": 369657800.0,
"step": 4170
},
{
"entropy": 0.934228515625,
"epoch": 0.5713114193945192,
"grad_norm": 0.6456649609485552,
"learning_rate": 3.6828941806397072e-06,
"loss": 0.9165,
"mean_token_accuracy": 0.771071308106184,
"num_tokens": 370511372.0,
"step": 4180
},
{
"entropy": 0.931982421875,
"epoch": 0.5726781931251281,
"grad_norm": 0.6302187780657302,
"learning_rate": 3.6793715654501903e-06,
"loss": 0.9093,
"mean_token_accuracy": 0.7697063222527504,
"num_tokens": 371416237.0,
"step": 4190
},
{
"entropy": 0.937841796875,
"epoch": 0.574044966855737,
"grad_norm": 0.7239343622929296,
"learning_rate": 3.6758489502606735e-06,
"loss": 0.9206,
"mean_token_accuracy": 0.7694468468427658,
"num_tokens": 372347297.0,
"step": 4200
},
{
"entropy": 0.942529296875,
"epoch": 0.5754117405863459,
"grad_norm": 0.7642466390812581,
"learning_rate": 3.672326335071157e-06,
"loss": 0.9146,
"mean_token_accuracy": 0.7691900305449962,
"num_tokens": 373210792.0,
"step": 4210
},
{
"entropy": 0.907763671875,
"epoch": 0.5767785143169548,
"grad_norm": 0.5855190817661039,
"learning_rate": 3.6688037198816406e-06,
"loss": 0.917,
"mean_token_accuracy": 0.7715086363255977,
"num_tokens": 374082669.0,
"step": 4220
},
{
"entropy": 0.9204833984375,
"epoch": 0.5781452880475637,
"grad_norm": 0.7002648373633297,
"learning_rate": 3.6652811046921237e-06,
"loss": 0.9144,
"mean_token_accuracy": 0.7700973838567734,
"num_tokens": 374934918.0,
"step": 4230
},
{
"entropy": 0.9138671875,
"epoch": 0.5795120617781726,
"grad_norm": 0.6121448997653711,
"learning_rate": 3.6617584895026072e-06,
"loss": 0.8955,
"mean_token_accuracy": 0.7730843558907509,
"num_tokens": 375792139.0,
"step": 4240
},
{
"entropy": 0.9775390625,
"epoch": 0.5808788355087815,
"grad_norm": 0.632599358328295,
"learning_rate": 3.6582358743130904e-06,
"loss": 0.9494,
"mean_token_accuracy": 0.7581640660762787,
"num_tokens": 376683914.0,
"step": 4250
},
{
"entropy": 0.919384765625,
"epoch": 0.5822456092393904,
"grad_norm": 0.6467662438699117,
"learning_rate": 3.6547132591235735e-06,
"loss": 0.9093,
"mean_token_accuracy": 0.769696743786335,
"num_tokens": 377543065.0,
"step": 4260
},
{
"entropy": 0.9513427734375,
"epoch": 0.5836123829699993,
"grad_norm": 0.7146317136012624,
"learning_rate": 3.651190643934057e-06,
"loss": 0.9426,
"mean_token_accuracy": 0.7643402151763439,
"num_tokens": 378369281.0,
"step": 4270
},
{
"entropy": 0.936669921875,
"epoch": 0.5849791567006082,
"grad_norm": 0.6237250508106679,
"learning_rate": 3.6476680287445406e-06,
"loss": 0.923,
"mean_token_accuracy": 0.7670198053121566,
"num_tokens": 379279878.0,
"step": 4280
},
{
"entropy": 0.9014404296875,
"epoch": 0.5863459304312171,
"grad_norm": 0.6145650210630451,
"learning_rate": 3.6441454135550237e-06,
"loss": 0.8865,
"mean_token_accuracy": 0.7741400316357613,
"num_tokens": 380191500.0,
"step": 4290
},
{
"entropy": 0.93740234375,
"epoch": 0.587712704161826,
"grad_norm": 0.6445120009421643,
"learning_rate": 3.640622798365507e-06,
"loss": 0.92,
"mean_token_accuracy": 0.7707940600812435,
"num_tokens": 381087074.0,
"step": 4300
},
{
"entropy": 0.96025390625,
"epoch": 0.5890794778924349,
"grad_norm": 0.6055128733719711,
"learning_rate": 3.63710018317599e-06,
"loss": 0.9426,
"mean_token_accuracy": 0.7615505717694759,
"num_tokens": 381979733.0,
"step": 4310
},
{
"entropy": 0.980712890625,
"epoch": 0.5904462516230438,
"grad_norm": 0.731513605132379,
"learning_rate": 3.633577567986473e-06,
"loss": 0.9713,
"mean_token_accuracy": 0.7588167160749435,
"num_tokens": 382847154.0,
"step": 4320
},
{
"entropy": 0.9400146484375,
"epoch": 0.5918130253536527,
"grad_norm": 0.6577601757657432,
"learning_rate": 3.630054952796957e-06,
"loss": 0.9161,
"mean_token_accuracy": 0.7675760932266712,
"num_tokens": 383707807.0,
"step": 4330
},
{
"entropy": 0.9642578125,
"epoch": 0.5931797990842615,
"grad_norm": 0.6542456874994412,
"learning_rate": 3.62653233760744e-06,
"loss": 0.9507,
"mean_token_accuracy": 0.7603037193417549,
"num_tokens": 384590386.0,
"step": 4340
},
{
"entropy": 0.939404296875,
"epoch": 0.5945465728148704,
"grad_norm": 0.8201014951954242,
"learning_rate": 3.6230097224179233e-06,
"loss": 0.9012,
"mean_token_accuracy": 0.7700928598642349,
"num_tokens": 385369969.0,
"step": 4350
},
{
"entropy": 0.916259765625,
"epoch": 0.5959133465454794,
"grad_norm": 0.7420875649673185,
"learning_rate": 3.6194871072284064e-06,
"loss": 0.9085,
"mean_token_accuracy": 0.7704723380506039,
"num_tokens": 386273775.0,
"step": 4360
},
{
"entropy": 0.9509765625,
"epoch": 0.5972801202760883,
"grad_norm": 0.6390786709377615,
"learning_rate": 3.61596449203889e-06,
"loss": 0.9276,
"mean_token_accuracy": 0.7650049902498722,
"num_tokens": 387155485.0,
"step": 4370
},
{
"entropy": 0.872216796875,
"epoch": 0.5986468940066972,
"grad_norm": 0.571554658526024,
"learning_rate": 3.6124418768493735e-06,
"loss": 0.85,
"mean_token_accuracy": 0.7815342307090759,
"num_tokens": 388029225.0,
"step": 4380
},
{
"entropy": 0.93876953125,
"epoch": 0.6000136677373061,
"grad_norm": 0.5904056365393243,
"learning_rate": 3.6089192616598566e-06,
"loss": 0.9385,
"mean_token_accuracy": 0.7656551599502563,
"num_tokens": 388973428.0,
"step": 4390
},
{
"entropy": 0.95107421875,
"epoch": 0.601380441467915,
"grad_norm": 0.6790607135868203,
"learning_rate": 3.6053966464703398e-06,
"loss": 0.9523,
"mean_token_accuracy": 0.7645857438445092,
"num_tokens": 389826917.0,
"step": 4400
},
{
"entropy": 0.96416015625,
"epoch": 0.6027472151985239,
"grad_norm": 0.5934280164167508,
"learning_rate": 3.6018740312808233e-06,
"loss": 0.9777,
"mean_token_accuracy": 0.758295975625515,
"num_tokens": 390768332.0,
"step": 4410
},
{
"entropy": 0.912744140625,
"epoch": 0.6041139889291328,
"grad_norm": 0.6664349945803706,
"learning_rate": 3.5983514160913064e-06,
"loss": 0.9113,
"mean_token_accuracy": 0.7669511348009109,
"num_tokens": 391673598.0,
"step": 4420
},
{
"entropy": 0.9395751953125,
"epoch": 0.6054807626597417,
"grad_norm": 0.6516230487628024,
"learning_rate": 3.5948288009017895e-06,
"loss": 0.9197,
"mean_token_accuracy": 0.7666677862405777,
"num_tokens": 392570813.0,
"step": 4430
},
{
"entropy": 0.946337890625,
"epoch": 0.6068475363903506,
"grad_norm": 0.6138786583071177,
"learning_rate": 3.591306185712273e-06,
"loss": 0.9302,
"mean_token_accuracy": 0.7651981383562088,
"num_tokens": 393449571.0,
"step": 4440
},
{
"entropy": 0.962646484375,
"epoch": 0.6082143101209595,
"grad_norm": 0.6017659070918718,
"learning_rate": 3.5877835705227566e-06,
"loss": 0.9367,
"mean_token_accuracy": 0.7651735477149486,
"num_tokens": 394310855.0,
"step": 4450
},
{
"entropy": 0.9087890625,
"epoch": 0.6095810838515684,
"grad_norm": 0.7943275920586389,
"learning_rate": 3.5842609553332398e-06,
"loss": 0.8718,
"mean_token_accuracy": 0.7770928591489792,
"num_tokens": 395201869.0,
"step": 4460
},
{
"entropy": 0.929443359375,
"epoch": 0.6109478575821773,
"grad_norm": 0.7051415752694872,
"learning_rate": 3.580738340143723e-06,
"loss": 0.9221,
"mean_token_accuracy": 0.7682494297623634,
"num_tokens": 396084305.0,
"step": 4470
},
{
"entropy": 0.929345703125,
"epoch": 0.6123146313127862,
"grad_norm": 0.6372744497515757,
"learning_rate": 3.577215724954206e-06,
"loss": 0.9216,
"mean_token_accuracy": 0.7697482287883759,
"num_tokens": 396999249.0,
"step": 4480
},
{
"entropy": 0.9570556640625,
"epoch": 0.6136814050433951,
"grad_norm": 0.5562812833148097,
"learning_rate": 3.57369310976469e-06,
"loss": 0.9442,
"mean_token_accuracy": 0.7618988133966923,
"num_tokens": 397841168.0,
"step": 4490
},
{
"entropy": 0.944482421875,
"epoch": 0.615048178774004,
"grad_norm": 0.6221986969471974,
"learning_rate": 3.570170494575173e-06,
"loss": 0.9277,
"mean_token_accuracy": 0.7679998643696309,
"num_tokens": 398754063.0,
"step": 4500
},
{
"entropy": 0.97587890625,
"epoch": 0.6164149525046129,
"grad_norm": 0.6401544035204912,
"learning_rate": 3.5666478793856562e-06,
"loss": 0.9858,
"mean_token_accuracy": 0.7576203636825085,
"num_tokens": 399621156.0,
"step": 4510
},
{
"entropy": 0.922509765625,
"epoch": 0.6177817262352218,
"grad_norm": 0.7175995194390957,
"learning_rate": 3.5631252641961394e-06,
"loss": 0.904,
"mean_token_accuracy": 0.77231165766716,
"num_tokens": 400472687.0,
"step": 4520
},
{
"entropy": 0.8770263671875,
"epoch": 0.6191484999658307,
"grad_norm": 0.6470345277201379,
"learning_rate": 3.5596026490066225e-06,
"loss": 0.8747,
"mean_token_accuracy": 0.7789335303008557,
"num_tokens": 401390570.0,
"step": 4530
},
{
"entropy": 0.9154052734375,
"epoch": 0.6205152736964396,
"grad_norm": 0.8826321458988151,
"learning_rate": 3.556080033817106e-06,
"loss": 0.9004,
"mean_token_accuracy": 0.7733095631003379,
"num_tokens": 402248186.0,
"step": 4540
},
{
"entropy": 0.983544921875,
"epoch": 0.6218820474270484,
"grad_norm": 0.6247401491263271,
"learning_rate": 3.5525574186275896e-06,
"loss": 0.9446,
"mean_token_accuracy": 0.7598119750618935,
"num_tokens": 403093559.0,
"step": 4550
},
{
"entropy": 0.974267578125,
"epoch": 0.6232488211576573,
"grad_norm": 0.641969521838033,
"learning_rate": 3.5490348034380727e-06,
"loss": 0.9705,
"mean_token_accuracy": 0.7606454133987427,
"num_tokens": 403992480.0,
"step": 4560
},
{
"entropy": 0.933154296875,
"epoch": 0.6246155948882662,
"grad_norm": 0.5750375140820787,
"learning_rate": 3.545512188248556e-06,
"loss": 0.9262,
"mean_token_accuracy": 0.7658180207014084,
"num_tokens": 404853974.0,
"step": 4570
},
{
"entropy": 0.90947265625,
"epoch": 0.6259823686188751,
"grad_norm": 0.6144136366218186,
"learning_rate": 3.5419895730590394e-06,
"loss": 0.8942,
"mean_token_accuracy": 0.7719666063785553,
"num_tokens": 405698785.0,
"step": 4580
},
{
"entropy": 0.9625,
"epoch": 0.627349142349484,
"grad_norm": 0.6849466567862318,
"learning_rate": 3.5384669578695225e-06,
"loss": 0.9646,
"mean_token_accuracy": 0.7593869663774967,
"num_tokens": 406574395.0,
"step": 4590
},
{
"entropy": 0.951171875,
"epoch": 0.6287159160800929,
"grad_norm": 0.6453728746931692,
"learning_rate": 3.534944342680006e-06,
"loss": 0.947,
"mean_token_accuracy": 0.7622104734182358,
"num_tokens": 407487257.0,
"step": 4600
},
{
"entropy": 0.907373046875,
"epoch": 0.6300826898107018,
"grad_norm": 0.6231684158585713,
"learning_rate": 3.5314217274904896e-06,
"loss": 0.903,
"mean_token_accuracy": 0.7712078802287579,
"num_tokens": 408376884.0,
"step": 4610
},
{
"entropy": 0.932421875,
"epoch": 0.6314494635413107,
"grad_norm": 0.5911209511245458,
"learning_rate": 3.5278991123009727e-06,
"loss": 0.9189,
"mean_token_accuracy": 0.7705645427107811,
"num_tokens": 409246879.0,
"step": 4620
},
{
"entropy": 0.9537353515625,
"epoch": 0.6328162372719196,
"grad_norm": 0.6876158357798131,
"learning_rate": 3.524376497111456e-06,
"loss": 0.9392,
"mean_token_accuracy": 0.7647826768457889,
"num_tokens": 410153996.0,
"step": 4630
},
{
"entropy": 0.908056640625,
"epoch": 0.6341830110025285,
"grad_norm": 0.7436240532991785,
"learning_rate": 3.520853881921939e-06,
"loss": 0.9012,
"mean_token_accuracy": 0.7710896670818329,
"num_tokens": 411026191.0,
"step": 4640
},
{
"entropy": 0.9185546875,
"epoch": 0.6355497847331374,
"grad_norm": 0.6326884958491744,
"learning_rate": 3.517331266732422e-06,
"loss": 0.9232,
"mean_token_accuracy": 0.7688418984413147,
"num_tokens": 411945110.0,
"step": 4650
},
{
"entropy": 0.981640625,
"epoch": 0.6369165584637463,
"grad_norm": 0.6006053822464673,
"learning_rate": 3.513808651542906e-06,
"loss": 0.9678,
"mean_token_accuracy": 0.7614172980189323,
"num_tokens": 412810803.0,
"step": 4660
},
{
"entropy": 0.934033203125,
"epoch": 0.6382833321943552,
"grad_norm": 0.6676977961605921,
"learning_rate": 3.510286036353389e-06,
"loss": 0.9029,
"mean_token_accuracy": 0.769856134057045,
"num_tokens": 413638372.0,
"step": 4670
},
{
"entropy": 0.92490234375,
"epoch": 0.6396501059249641,
"grad_norm": 0.5682213809667973,
"learning_rate": 3.5067634211638723e-06,
"loss": 0.9188,
"mean_token_accuracy": 0.7681633070111274,
"num_tokens": 414557899.0,
"step": 4680
},
{
"entropy": 0.92451171875,
"epoch": 0.641016879655573,
"grad_norm": 0.6222691334458051,
"learning_rate": 3.5032408059743554e-06,
"loss": 0.9207,
"mean_token_accuracy": 0.7701552018523217,
"num_tokens": 415455781.0,
"step": 4690
},
{
"entropy": 0.92451171875,
"epoch": 0.6423836533861819,
"grad_norm": 0.6244836601706479,
"learning_rate": 3.499718190784839e-06,
"loss": 0.9108,
"mean_token_accuracy": 0.7698039717972278,
"num_tokens": 416295102.0,
"step": 4700
},
{
"entropy": 0.9351318359375,
"epoch": 0.6437504271167909,
"grad_norm": 0.617020086729061,
"learning_rate": 3.4961955755953225e-06,
"loss": 0.9186,
"mean_token_accuracy": 0.7684897430241108,
"num_tokens": 417180585.0,
"step": 4710
},
{
"entropy": 0.96259765625,
"epoch": 0.6451172008473998,
"grad_norm": 0.6016620180554751,
"learning_rate": 3.4926729604058056e-06,
"loss": 0.9556,
"mean_token_accuracy": 0.7623788021504879,
"num_tokens": 418052757.0,
"step": 4720
},
{
"entropy": 0.91796875,
"epoch": 0.6464839745780087,
"grad_norm": 0.6560885442103777,
"learning_rate": 3.4891503452162888e-06,
"loss": 0.8944,
"mean_token_accuracy": 0.7716099195182323,
"num_tokens": 418895912.0,
"step": 4730
},
{
"entropy": 0.9077880859375,
"epoch": 0.6478507483086176,
"grad_norm": 0.5817300208373742,
"learning_rate": 3.4856277300267723e-06,
"loss": 0.8889,
"mean_token_accuracy": 0.7736640281975269,
"num_tokens": 419758665.0,
"step": 4740
},
{
"entropy": 0.9245849609375,
"epoch": 0.6492175220392264,
"grad_norm": 0.6226657406813666,
"learning_rate": 3.4821051148372554e-06,
"loss": 0.9123,
"mean_token_accuracy": 0.7672120496630669,
"num_tokens": 420619426.0,
"step": 4750
},
{
"entropy": 0.86650390625,
"epoch": 0.6505842957698353,
"grad_norm": 0.7076498471129132,
"learning_rate": 3.4785824996477386e-06,
"loss": 0.8338,
"mean_token_accuracy": 0.7833277694880962,
"num_tokens": 421466711.0,
"step": 4760
},
{
"entropy": 0.8564208984375,
"epoch": 0.6519510695004442,
"grad_norm": 0.596414389998348,
"learning_rate": 3.475059884458222e-06,
"loss": 0.8445,
"mean_token_accuracy": 0.7853332005441189,
"num_tokens": 422336352.0,
"step": 4770
},
{
"entropy": 0.92802734375,
"epoch": 0.6533178432310531,
"grad_norm": 0.6309409526172187,
"learning_rate": 3.4715372692687057e-06,
"loss": 0.894,
"mean_token_accuracy": 0.7731868393719197,
"num_tokens": 423184508.0,
"step": 4780
},
{
"entropy": 0.964404296875,
"epoch": 0.654684616961662,
"grad_norm": 0.6397130754014757,
"learning_rate": 3.4680146540791888e-06,
"loss": 0.945,
"mean_token_accuracy": 0.7647523067891597,
"num_tokens": 424081913.0,
"step": 4790
},
{
"entropy": 0.9284912109375,
"epoch": 0.6560513906922709,
"grad_norm": 0.636043433714453,
"learning_rate": 3.464492038889672e-06,
"loss": 0.9159,
"mean_token_accuracy": 0.7683800116181374,
"num_tokens": 424917247.0,
"step": 4800
},
{
"entropy": 0.933544921875,
"epoch": 0.6574181644228798,
"grad_norm": 0.6334294442893509,
"learning_rate": 3.460969423700155e-06,
"loss": 0.9449,
"mean_token_accuracy": 0.7627432972192765,
"num_tokens": 425842562.0,
"step": 4810
},
{
"entropy": 0.928125,
"epoch": 0.6587849381534887,
"grad_norm": 0.6150486911690238,
"learning_rate": 3.457446808510639e-06,
"loss": 0.9253,
"mean_token_accuracy": 0.7672046981751919,
"num_tokens": 426738885.0,
"step": 4820
},
{
"entropy": 0.8941162109375,
"epoch": 0.6601517118840976,
"grad_norm": 0.6214652902954265,
"learning_rate": 3.453924193321122e-06,
"loss": 0.8957,
"mean_token_accuracy": 0.7736395344138145,
"num_tokens": 427666165.0,
"step": 4830
},
{
"entropy": 0.91396484375,
"epoch": 0.6615184856147065,
"grad_norm": 0.6310678889629469,
"learning_rate": 3.4504015781316052e-06,
"loss": 0.8715,
"mean_token_accuracy": 0.7758303605020046,
"num_tokens": 428584533.0,
"step": 4840
},
{
"entropy": 0.948193359375,
"epoch": 0.6628852593453154,
"grad_norm": 0.6149038169440357,
"learning_rate": 3.4468789629420884e-06,
"loss": 0.9542,
"mean_token_accuracy": 0.7620439969003201,
"num_tokens": 429522884.0,
"step": 4850
},
{
"entropy": 0.899267578125,
"epoch": 0.6642520330759243,
"grad_norm": 0.6351817742975294,
"learning_rate": 3.4433563477525715e-06,
"loss": 0.8806,
"mean_token_accuracy": 0.7778452999889851,
"num_tokens": 430404622.0,
"step": 4860
},
{
"entropy": 0.91259765625,
"epoch": 0.6656188068065332,
"grad_norm": 0.6255235007303283,
"learning_rate": 3.439833732563055e-06,
"loss": 0.9038,
"mean_token_accuracy": 0.7711722984910011,
"num_tokens": 431275413.0,
"step": 4870
},
{
"entropy": 0.904736328125,
"epoch": 0.6669855805371421,
"grad_norm": 0.6602140456709596,
"learning_rate": 3.4363111173735386e-06,
"loss": 0.9037,
"mean_token_accuracy": 0.770679236203432,
"num_tokens": 432204493.0,
"step": 4880
},
{
"entropy": 0.94052734375,
"epoch": 0.668352354267751,
"grad_norm": 0.6132888678834465,
"learning_rate": 3.4327885021840217e-06,
"loss": 0.9245,
"mean_token_accuracy": 0.7682620540261269,
"num_tokens": 433107951.0,
"step": 4890
},
{
"entropy": 0.923974609375,
"epoch": 0.6697191279983599,
"grad_norm": 0.6126912018620074,
"learning_rate": 3.429265886994505e-06,
"loss": 0.9365,
"mean_token_accuracy": 0.7668405272066593,
"num_tokens": 434029585.0,
"step": 4900
},
{
"entropy": 0.90166015625,
"epoch": 0.6710859017289688,
"grad_norm": 0.7960183386445704,
"learning_rate": 3.4257432718049884e-06,
"loss": 0.8944,
"mean_token_accuracy": 0.7735520914196968,
"num_tokens": 434937916.0,
"step": 4910
},
{
"entropy": 0.9020263671875,
"epoch": 0.6724526754595777,
"grad_norm": 0.6267920602929132,
"learning_rate": 3.4222206566154715e-06,
"loss": 0.9287,
"mean_token_accuracy": 0.7712463982403278,
"num_tokens": 435852833.0,
"step": 4920
},
{
"entropy": 0.9603759765625,
"epoch": 0.6738194491901865,
"grad_norm": 0.6000055615799351,
"learning_rate": 3.418698041425955e-06,
"loss": 0.9588,
"mean_token_accuracy": 0.758978983014822,
"num_tokens": 436724128.0,
"step": 4930
},
{
"entropy": 0.9007568359375,
"epoch": 0.6751862229207954,
"grad_norm": 0.6203071772230346,
"learning_rate": 3.415175426236438e-06,
"loss": 0.8997,
"mean_token_accuracy": 0.7726327396929265,
"num_tokens": 437633875.0,
"step": 4940
},
{
"entropy": 0.865380859375,
"epoch": 0.6765529966514043,
"grad_norm": 0.5622788475518702,
"learning_rate": 3.4116528110469217e-06,
"loss": 0.8465,
"mean_token_accuracy": 0.7832522973418236,
"num_tokens": 438515714.0,
"step": 4950
},
{
"entropy": 0.921533203125,
"epoch": 0.6779197703820132,
"grad_norm": 0.6889218042602961,
"learning_rate": 3.408130195857405e-06,
"loss": 0.8906,
"mean_token_accuracy": 0.7733630277216434,
"num_tokens": 439364688.0,
"step": 4960
},
{
"entropy": 0.9194580078125,
"epoch": 0.6792865441126221,
"grad_norm": 0.9285207404492086,
"learning_rate": 3.404607580667888e-06,
"loss": 0.9049,
"mean_token_accuracy": 0.7701285541057586,
"num_tokens": 440259473.0,
"step": 4970
},
{
"entropy": 0.9029052734375,
"epoch": 0.680653317843231,
"grad_norm": 0.602228481166046,
"learning_rate": 3.401084965478371e-06,
"loss": 0.8821,
"mean_token_accuracy": 0.7772107750177384,
"num_tokens": 441127912.0,
"step": 4980
},
{
"entropy": 0.919189453125,
"epoch": 0.6820200915738399,
"grad_norm": 0.6101139879780795,
"learning_rate": 3.397562350288855e-06,
"loss": 0.9178,
"mean_token_accuracy": 0.7715644158422947,
"num_tokens": 442024079.0,
"step": 4990
},
{
"entropy": 0.9144775390625,
"epoch": 0.6833868653044488,
"grad_norm": 0.6172715809051087,
"learning_rate": 3.394039735099338e-06,
"loss": 0.9041,
"mean_token_accuracy": 0.7720178864896298,
"num_tokens": 442910647.0,
"step": 5000
},
{
"entropy": 0.8880859375,
"epoch": 0.6847536390350577,
"grad_norm": 0.7236777081565706,
"learning_rate": 3.3905171199098213e-06,
"loss": 0.8649,
"mean_token_accuracy": 0.7775885783135891,
"num_tokens": 443789732.0,
"step": 5010
},
{
"entropy": 0.9138427734375,
"epoch": 0.6861204127656666,
"grad_norm": 0.5854886312824747,
"learning_rate": 3.3869945047203044e-06,
"loss": 0.9115,
"mean_token_accuracy": 0.7703619621694088,
"num_tokens": 444692779.0,
"step": 5020
},
{
"entropy": 0.905078125,
"epoch": 0.6874871864962755,
"grad_norm": 0.6405683169204947,
"learning_rate": 3.3834718895307876e-06,
"loss": 0.8974,
"mean_token_accuracy": 0.769259586930275,
"num_tokens": 445556425.0,
"step": 5030
},
{
"entropy": 0.894580078125,
"epoch": 0.6888539602268844,
"grad_norm": 0.6087744394614885,
"learning_rate": 3.3799492743412715e-06,
"loss": 0.8938,
"mean_token_accuracy": 0.7755403906106949,
"num_tokens": 446440835.0,
"step": 5040
},
{
"entropy": 0.909130859375,
"epoch": 0.6902207339574933,
"grad_norm": 0.6136132811468907,
"learning_rate": 3.3764266591517547e-06,
"loss": 0.888,
"mean_token_accuracy": 0.7737716972827912,
"num_tokens": 447344281.0,
"step": 5050
},
{
"entropy": 0.9265380859375,
"epoch": 0.6915875076881023,
"grad_norm": 0.5936812505883273,
"learning_rate": 3.3729040439622378e-06,
"loss": 0.8928,
"mean_token_accuracy": 0.7716215014457702,
"num_tokens": 448244245.0,
"step": 5060
},
{
"entropy": 0.946630859375,
"epoch": 0.6929542814187112,
"grad_norm": 0.7149170327035338,
"learning_rate": 3.3693814287727213e-06,
"loss": 0.9202,
"mean_token_accuracy": 0.7656993992626667,
"num_tokens": 449100309.0,
"step": 5070
},
{
"entropy": 0.947900390625,
"epoch": 0.6943210551493201,
"grad_norm": 0.6948574418307258,
"learning_rate": 3.3658588135832044e-06,
"loss": 0.949,
"mean_token_accuracy": 0.76318179666996,
"num_tokens": 450042185.0,
"step": 5080
},
{
"entropy": 0.9011962890625,
"epoch": 0.695687828879929,
"grad_norm": 0.7260871143904716,
"learning_rate": 3.3623361983936876e-06,
"loss": 0.9039,
"mean_token_accuracy": 0.7730422839522362,
"num_tokens": 450988603.0,
"step": 5090
},
{
"entropy": 0.912109375,
"epoch": 0.6970546026105379,
"grad_norm": 0.640528096751195,
"learning_rate": 3.358813583204171e-06,
"loss": 0.9047,
"mean_token_accuracy": 0.7735993631184102,
"num_tokens": 451928945.0,
"step": 5100
},
{
"entropy": 0.971533203125,
"epoch": 0.6984213763411468,
"grad_norm": 0.6675129995911498,
"learning_rate": 3.3552909680146547e-06,
"loss": 0.9597,
"mean_token_accuracy": 0.7589392080903054,
"num_tokens": 452812071.0,
"step": 5110
},
{
"entropy": 0.896875,
"epoch": 0.6997881500717557,
"grad_norm": 0.6123352500179308,
"learning_rate": 3.3517683528251378e-06,
"loss": 0.8897,
"mean_token_accuracy": 0.7752141162753106,
"num_tokens": 453702565.0,
"step": 5120
},
{
"entropy": 0.8986328125,
"epoch": 0.7011549238023645,
"grad_norm": 0.6441339546415186,
"learning_rate": 3.348245737635621e-06,
"loss": 0.8815,
"mean_token_accuracy": 0.7749374061822891,
"num_tokens": 454608169.0,
"step": 5130
},
{
"entropy": 0.9224609375,
"epoch": 0.7025216975329734,
"grad_norm": 0.6712542152245352,
"learning_rate": 3.344723122446104e-06,
"loss": 0.8995,
"mean_token_accuracy": 0.7703351631760598,
"num_tokens": 455455664.0,
"step": 5140
},
{
"entropy": 0.9348388671875,
"epoch": 0.7038884712635823,
"grad_norm": 0.6348152065201295,
"learning_rate": 3.341200507256587e-06,
"loss": 0.9346,
"mean_token_accuracy": 0.7669262498617172,
"num_tokens": 456331907.0,
"step": 5150
},
{
"entropy": 0.9497802734375,
"epoch": 0.7052552449941912,
"grad_norm": 0.69486021692108,
"learning_rate": 3.337677892067071e-06,
"loss": 0.9411,
"mean_token_accuracy": 0.7648323938250542,
"num_tokens": 457208256.0,
"step": 5160
},
{
"entropy": 0.938427734375,
"epoch": 0.7066220187248001,
"grad_norm": 0.7190756630593559,
"learning_rate": 3.3341552768775543e-06,
"loss": 0.9305,
"mean_token_accuracy": 0.765125772356987,
"num_tokens": 458092829.0,
"step": 5170
},
{
"entropy": 0.937890625,
"epoch": 0.707988792455409,
"grad_norm": 0.6627548730358419,
"learning_rate": 3.3306326616880374e-06,
"loss": 0.9151,
"mean_token_accuracy": 0.7694108329713345,
"num_tokens": 458977684.0,
"step": 5180
},
{
"entropy": 0.926416015625,
"epoch": 0.7093555661860179,
"grad_norm": 0.5913253147720179,
"learning_rate": 3.3271100464985205e-06,
"loss": 0.8981,
"mean_token_accuracy": 0.7706363625824452,
"num_tokens": 459877261.0,
"step": 5190
},
{
"entropy": 0.91318359375,
"epoch": 0.7107223399166268,
"grad_norm": 0.604955891525787,
"learning_rate": 3.323587431309004e-06,
"loss": 0.8954,
"mean_token_accuracy": 0.7733483828604222,
"num_tokens": 460803559.0,
"step": 5200
},
{
"entropy": 0.94111328125,
"epoch": 0.7120891136472357,
"grad_norm": 0.6474338510217049,
"learning_rate": 3.3200648161194876e-06,
"loss": 0.9216,
"mean_token_accuracy": 0.7640735365450382,
"num_tokens": 461698177.0,
"step": 5210
},
{
"entropy": 0.905615234375,
"epoch": 0.7134558873778446,
"grad_norm": 0.6435502307830989,
"learning_rate": 3.3165422009299707e-06,
"loss": 0.9173,
"mean_token_accuracy": 0.7682828955352307,
"num_tokens": 462589164.0,
"step": 5220
},
{
"entropy": 0.9395751953125,
"epoch": 0.7148226611084535,
"grad_norm": 0.6377391522567587,
"learning_rate": 3.313019585740454e-06,
"loss": 0.9256,
"mean_token_accuracy": 0.7666286997497082,
"num_tokens": 463468365.0,
"step": 5230
},
{
"entropy": 0.951416015625,
"epoch": 0.7161894348390624,
"grad_norm": 0.6450366522718158,
"learning_rate": 3.3094969705509374e-06,
"loss": 0.954,
"mean_token_accuracy": 0.7618856251239776,
"num_tokens": 464350433.0,
"step": 5240
},
{
"entropy": 0.905517578125,
"epoch": 0.7175562085696713,
"grad_norm": 0.6480316904597011,
"learning_rate": 3.3059743553614205e-06,
"loss": 0.9033,
"mean_token_accuracy": 0.7736267954111099,
"num_tokens": 465221702.0,
"step": 5250
},
{
"entropy": 0.920849609375,
"epoch": 0.7189229823002802,
"grad_norm": 0.6487299296573664,
"learning_rate": 3.3024517401719036e-06,
"loss": 0.8969,
"mean_token_accuracy": 0.7695774272084236,
"num_tokens": 466090098.0,
"step": 5260
},
{
"entropy": 0.9374267578125,
"epoch": 0.7202897560308891,
"grad_norm": 0.7305533664796238,
"learning_rate": 3.298929124982387e-06,
"loss": 0.9507,
"mean_token_accuracy": 0.7678333856165409,
"num_tokens": 466990039.0,
"step": 5270
},
{
"entropy": 0.9267578125,
"epoch": 0.721656529761498,
"grad_norm": 0.6130817763195063,
"learning_rate": 3.2954065097928707e-06,
"loss": 0.9318,
"mean_token_accuracy": 0.768411710113287,
"num_tokens": 467896781.0,
"step": 5280
},
{
"entropy": 0.934033203125,
"epoch": 0.7230233034921069,
"grad_norm": 0.6316805580140246,
"learning_rate": 3.291883894603354e-06,
"loss": 0.9152,
"mean_token_accuracy": 0.7700305067002773,
"num_tokens": 468784684.0,
"step": 5290
},
{
"entropy": 0.9068359375,
"epoch": 0.7243900772227158,
"grad_norm": 0.670818631655748,
"learning_rate": 3.288361279413837e-06,
"loss": 0.8807,
"mean_token_accuracy": 0.7730885863304138,
"num_tokens": 469661887.0,
"step": 5300
},
{
"entropy": 0.8990234375,
"epoch": 0.7257568509533247,
"grad_norm": 0.5937829809449578,
"learning_rate": 3.28483866422432e-06,
"loss": 0.8898,
"mean_token_accuracy": 0.7759884350001812,
"num_tokens": 470540659.0,
"step": 5310
},
{
"entropy": 0.955322265625,
"epoch": 0.7271236246839335,
"grad_norm": 0.6301702585419089,
"learning_rate": 3.281316049034804e-06,
"loss": 0.9547,
"mean_token_accuracy": 0.760207387059927,
"num_tokens": 471483298.0,
"step": 5320
},
{
"entropy": 0.896875,
"epoch": 0.7284903984145424,
"grad_norm": 0.6713711906681539,
"learning_rate": 3.277793433845287e-06,
"loss": 0.8697,
"mean_token_accuracy": 0.7736578293144702,
"num_tokens": 472374357.0,
"step": 5330
},
{
"entropy": 0.90322265625,
"epoch": 0.7298571721451513,
"grad_norm": 0.5921988500212492,
"learning_rate": 3.2742708186557703e-06,
"loss": 0.8805,
"mean_token_accuracy": 0.7786927118897438,
"num_tokens": 473290428.0,
"step": 5340
},
{
"entropy": 0.943359375,
"epoch": 0.7312239458757602,
"grad_norm": 0.6721248726043073,
"learning_rate": 3.2707482034662534e-06,
"loss": 0.91,
"mean_token_accuracy": 0.7688648387789726,
"num_tokens": 474178439.0,
"step": 5350
},
{
"entropy": 0.92421875,
"epoch": 0.7325907196063691,
"grad_norm": 0.6521707229579653,
"learning_rate": 3.2672255882767366e-06,
"loss": 0.9157,
"mean_token_accuracy": 0.7687506228685379,
"num_tokens": 475073389.0,
"step": 5360
},
{
"entropy": 0.886962890625,
"epoch": 0.733957493336978,
"grad_norm": 0.5732833339507185,
"learning_rate": 3.26370297308722e-06,
"loss": 0.9047,
"mean_token_accuracy": 0.7718880631029605,
"num_tokens": 475982346.0,
"step": 5370
},
{
"entropy": 0.901708984375,
"epoch": 0.7353242670675869,
"grad_norm": 0.6245743844722369,
"learning_rate": 3.2601803578977037e-06,
"loss": 0.8984,
"mean_token_accuracy": 0.7739488154649734,
"num_tokens": 476863931.0,
"step": 5380
},
{
"entropy": 0.8754150390625,
"epoch": 0.7366910407981958,
"grad_norm": 0.5767592626885647,
"learning_rate": 3.2566577427081868e-06,
"loss": 0.88,
"mean_token_accuracy": 0.7769563727080822,
"num_tokens": 477781550.0,
"step": 5390
},
{
"entropy": 0.9031005859375,
"epoch": 0.7380578145288047,
"grad_norm": 0.6136869244402546,
"learning_rate": 3.25313512751867e-06,
"loss": 0.8988,
"mean_token_accuracy": 0.7742211513221264,
"num_tokens": 478654068.0,
"step": 5400
},
{
"entropy": 0.940966796875,
"epoch": 0.7394245882594137,
"grad_norm": 0.6129866487421717,
"learning_rate": 3.2496125123291535e-06,
"loss": 0.9682,
"mean_token_accuracy": 0.759343209117651,
"num_tokens": 479541351.0,
"step": 5410
},
{
"entropy": 0.9099609375,
"epoch": 0.7407913619900226,
"grad_norm": 0.6277970540457029,
"learning_rate": 3.2460898971396366e-06,
"loss": 0.8751,
"mean_token_accuracy": 0.7782999649643898,
"num_tokens": 480430820.0,
"step": 5420
},
{
"entropy": 0.92177734375,
"epoch": 0.7421581357206315,
"grad_norm": 0.6335707524185418,
"learning_rate": 3.24256728195012e-06,
"loss": 0.9207,
"mean_token_accuracy": 0.7687142990529537,
"num_tokens": 481306833.0,
"step": 5430
},
{
"entropy": 0.909814453125,
"epoch": 0.7435249094512404,
"grad_norm": 0.5876247108324927,
"learning_rate": 3.2390446667606037e-06,
"loss": 0.9285,
"mean_token_accuracy": 0.769102867692709,
"num_tokens": 482218500.0,
"step": 5440
},
{
"entropy": 0.940185546875,
"epoch": 0.7448916831818493,
"grad_norm": 0.6256856144809173,
"learning_rate": 3.235522051571087e-06,
"loss": 0.9176,
"mean_token_accuracy": 0.7678864672780037,
"num_tokens": 483099935.0,
"step": 5450
},
{
"entropy": 0.932373046875,
"epoch": 0.7462584569124582,
"grad_norm": 0.5503971164742403,
"learning_rate": 3.23199943638157e-06,
"loss": 0.9188,
"mean_token_accuracy": 0.7700562283396721,
"num_tokens": 483948988.0,
"step": 5460
},
{
"entropy": 0.93388671875,
"epoch": 0.7476252306430671,
"grad_norm": 0.7018108412424625,
"learning_rate": 3.228476821192053e-06,
"loss": 0.9347,
"mean_token_accuracy": 0.7669144205749034,
"num_tokens": 484844939.0,
"step": 5470
},
{
"entropy": 0.929541015625,
"epoch": 0.748992004373676,
"grad_norm": 0.6687946388175806,
"learning_rate": 3.224954206002536e-06,
"loss": 0.9418,
"mean_token_accuracy": 0.7643803462386132,
"num_tokens": 485763455.0,
"step": 5480
},
{
"entropy": 0.8823974609375,
"epoch": 0.7503587781042849,
"grad_norm": 0.6073711424162153,
"learning_rate": 3.22143159081302e-06,
"loss": 0.8626,
"mean_token_accuracy": 0.779666593670845,
"num_tokens": 486656249.0,
"step": 5490
},
{
"entropy": 0.9467041015625,
"epoch": 0.7517255518348938,
"grad_norm": 0.6756473811870625,
"learning_rate": 3.2179089756235033e-06,
"loss": 0.9148,
"mean_token_accuracy": 0.7655043423175811,
"num_tokens": 487540315.0,
"step": 5500
},
{
"entropy": 0.947119140625,
"epoch": 0.7530923255655027,
"grad_norm": 0.6293397442101021,
"learning_rate": 3.2143863604339864e-06,
"loss": 0.9286,
"mean_token_accuracy": 0.7674849264323711,
"num_tokens": 488445699.0,
"step": 5510
},
{
"entropy": 0.9134521484375,
"epoch": 0.7544590992961115,
"grad_norm": 0.6350135592892628,
"learning_rate": 3.2108637452444695e-06,
"loss": 0.9263,
"mean_token_accuracy": 0.7695359341800213,
"num_tokens": 489326155.0,
"step": 5520
},
{
"entropy": 0.9557373046875,
"epoch": 0.7558258730267204,
"grad_norm": 0.6083936385993641,
"learning_rate": 3.2073411300549526e-06,
"loss": 0.9381,
"mean_token_accuracy": 0.7635311223566532,
"num_tokens": 490191644.0,
"step": 5530
},
{
"entropy": 0.8634033203125,
"epoch": 0.7571926467573293,
"grad_norm": 0.6020311727387271,
"learning_rate": 3.2038185148654366e-06,
"loss": 0.8482,
"mean_token_accuracy": 0.7835816882550717,
"num_tokens": 491068717.0,
"step": 5540
},
{
"entropy": 0.9541015625,
"epoch": 0.7585594204879382,
"grad_norm": 0.6278092552051491,
"learning_rate": 3.2002958996759197e-06,
"loss": 0.9443,
"mean_token_accuracy": 0.7656134389340877,
"num_tokens": 491916966.0,
"step": 5550
},
{
"entropy": 0.95908203125,
"epoch": 0.7599261942185471,
"grad_norm": 0.6152640216740689,
"learning_rate": 3.196773284486403e-06,
"loss": 0.9483,
"mean_token_accuracy": 0.7621932193636894,
"num_tokens": 492799007.0,
"step": 5560
},
{
"entropy": 0.9646484375,
"epoch": 0.761292967949156,
"grad_norm": 0.6571522640212953,
"learning_rate": 3.1932506692968864e-06,
"loss": 0.9387,
"mean_token_accuracy": 0.7641196273267269,
"num_tokens": 493694735.0,
"step": 5570
},
{
"entropy": 0.918505859375,
"epoch": 0.7626597416797649,
"grad_norm": 0.6226034302536241,
"learning_rate": 3.1897280541073695e-06,
"loss": 0.8838,
"mean_token_accuracy": 0.7745287790894508,
"num_tokens": 494599096.0,
"step": 5580
},
{
"entropy": 0.9209228515625,
"epoch": 0.7640265154103738,
"grad_norm": 0.5900552726283053,
"learning_rate": 3.1862054389178526e-06,
"loss": 0.9216,
"mean_token_accuracy": 0.7691805146634578,
"num_tokens": 495533785.0,
"step": 5590
},
{
"entropy": 0.917578125,
"epoch": 0.7653932891409827,
"grad_norm": 0.6278267720849634,
"learning_rate": 3.182682823728336e-06,
"loss": 0.9112,
"mean_token_accuracy": 0.768238503485918,
"num_tokens": 496427337.0,
"step": 5600
},
{
"entropy": 0.89423828125,
"epoch": 0.7667600628715916,
"grad_norm": 0.5948259012581495,
"learning_rate": 3.1791602085388197e-06,
"loss": 0.8779,
"mean_token_accuracy": 0.7764112465083599,
"num_tokens": 497328578.0,
"step": 5610
},
{
"entropy": 0.962890625,
"epoch": 0.7681268366022005,
"grad_norm": 0.6241585549744524,
"learning_rate": 3.175637593349303e-06,
"loss": 0.9703,
"mean_token_accuracy": 0.759580048173666,
"num_tokens": 498197795.0,
"step": 5620
},
{
"entropy": 0.9478515625,
"epoch": 0.7694936103328094,
"grad_norm": 0.5996502232608913,
"learning_rate": 3.172114978159786e-06,
"loss": 0.9467,
"mean_token_accuracy": 0.7634208180010319,
"num_tokens": 499104957.0,
"step": 5630
},
{
"entropy": 0.919873046875,
"epoch": 0.7708603840634183,
"grad_norm": 0.5831183314186228,
"learning_rate": 3.168592362970269e-06,
"loss": 0.9209,
"mean_token_accuracy": 0.7711405895650387,
"num_tokens": 499955528.0,
"step": 5640
},
{
"entropy": 0.891162109375,
"epoch": 0.7722271577940272,
"grad_norm": 0.6373233467228837,
"learning_rate": 3.165069747780753e-06,
"loss": 0.8934,
"mean_token_accuracy": 0.7747475370764733,
"num_tokens": 500817635.0,
"step": 5650
},
{
"entropy": 0.933056640625,
"epoch": 0.7735939315246361,
"grad_norm": 0.6094478326720644,
"learning_rate": 3.161547132591236e-06,
"loss": 0.9337,
"mean_token_accuracy": 0.7668229512870312,
"num_tokens": 501718725.0,
"step": 5660
},
{
"entropy": 0.944580078125,
"epoch": 0.774960705255245,
"grad_norm": 0.5898256132178022,
"learning_rate": 3.1580245174017193e-06,
"loss": 0.9591,
"mean_token_accuracy": 0.7626009777188301,
"num_tokens": 502611865.0,
"step": 5670
},
{
"entropy": 0.93720703125,
"epoch": 0.7763274789858539,
"grad_norm": 0.6391232109635382,
"learning_rate": 3.1545019022122025e-06,
"loss": 0.921,
"mean_token_accuracy": 0.7685982584953308,
"num_tokens": 503500310.0,
"step": 5680
},
{
"entropy": 0.9709716796875,
"epoch": 0.7776942527164628,
"grad_norm": 0.5736536756217407,
"learning_rate": 3.1509792870226856e-06,
"loss": 0.9657,
"mean_token_accuracy": 0.7598110198974609,
"num_tokens": 504395432.0,
"step": 5690
},
{
"entropy": 0.91689453125,
"epoch": 0.7790610264470716,
"grad_norm": 0.7102601512195296,
"learning_rate": 3.147456671833169e-06,
"loss": 0.9064,
"mean_token_accuracy": 0.770981277525425,
"num_tokens": 505275664.0,
"step": 5700
},
{
"entropy": 0.905419921875,
"epoch": 0.7804278001776805,
"grad_norm": 0.7046606494742553,
"learning_rate": 3.1439340566436527e-06,
"loss": 0.9032,
"mean_token_accuracy": 0.7704074315726757,
"num_tokens": 506168945.0,
"step": 5710
},
{
"entropy": 0.9344482421875,
"epoch": 0.7817945739082894,
"grad_norm": 0.6064509922762849,
"learning_rate": 3.140411441454136e-06,
"loss": 0.8998,
"mean_token_accuracy": 0.7711231894791126,
"num_tokens": 507052908.0,
"step": 5720
},
{
"entropy": 0.928173828125,
"epoch": 0.7831613476388983,
"grad_norm": 0.6013562096588372,
"learning_rate": 3.136888826264619e-06,
"loss": 0.9298,
"mean_token_accuracy": 0.7652062311768532,
"num_tokens": 507938577.0,
"step": 5730
},
{
"entropy": 0.8828125,
"epoch": 0.7845281213695072,
"grad_norm": 0.6774270591441419,
"learning_rate": 3.1333662110751025e-06,
"loss": 0.8858,
"mean_token_accuracy": 0.7751082263886928,
"num_tokens": 508807062.0,
"step": 5740
},
{
"entropy": 0.888671875,
"epoch": 0.7858948951001162,
"grad_norm": 0.6455058768335712,
"learning_rate": 3.1298435958855856e-06,
"loss": 0.8473,
"mean_token_accuracy": 0.7814380072057248,
"num_tokens": 509646843.0,
"step": 5750
},
{
"entropy": 0.9285400390625,
"epoch": 0.7872616688307251,
"grad_norm": 0.618875911433539,
"learning_rate": 3.126320980696069e-06,
"loss": 0.902,
"mean_token_accuracy": 0.7702745974063874,
"num_tokens": 510507535.0,
"step": 5760
},
{
"entropy": 0.931005859375,
"epoch": 0.788628442561334,
"grad_norm": 0.6274143257224095,
"learning_rate": 3.1227983655065523e-06,
"loss": 0.9234,
"mean_token_accuracy": 0.7666326262056827,
"num_tokens": 511376176.0,
"step": 5770
},
{
"entropy": 0.8716796875,
"epoch": 0.7899952162919429,
"grad_norm": 0.5984195210740386,
"learning_rate": 3.119275750317036e-06,
"loss": 0.8677,
"mean_token_accuracy": 0.7767310164868831,
"num_tokens": 512257868.0,
"step": 5780
},
{
"entropy": 0.955419921875,
"epoch": 0.7913619900225518,
"grad_norm": 0.8551750124006953,
"learning_rate": 3.115753135127519e-06,
"loss": 0.9522,
"mean_token_accuracy": 0.7621162720024586,
"num_tokens": 513149614.0,
"step": 5790
},
{
"entropy": 0.924755859375,
"epoch": 0.7927287637531607,
"grad_norm": 0.6363800115227809,
"learning_rate": 3.112230519938002e-06,
"loss": 0.9163,
"mean_token_accuracy": 0.7687442176043987,
"num_tokens": 514060257.0,
"step": 5800
},
{
"entropy": 0.923388671875,
"epoch": 0.7940955374837696,
"grad_norm": 0.6965254875331773,
"learning_rate": 3.108707904748485e-06,
"loss": 0.8921,
"mean_token_accuracy": 0.7736927963793278,
"num_tokens": 514887972.0,
"step": 5810
},
{
"entropy": 0.92607421875,
"epoch": 0.7954623112143785,
"grad_norm": 0.5974540194038886,
"learning_rate": 3.105185289558969e-06,
"loss": 0.9173,
"mean_token_accuracy": 0.7719533018767833,
"num_tokens": 515812256.0,
"step": 5820
},
{
"entropy": 0.9245361328125,
"epoch": 0.7968290849449874,
"grad_norm": 0.6016138842594493,
"learning_rate": 3.1016626743694523e-06,
"loss": 0.904,
"mean_token_accuracy": 0.7710686899721623,
"num_tokens": 516697618.0,
"step": 5830
},
{
"entropy": 0.87529296875,
"epoch": 0.7981958586755963,
"grad_norm": 0.5742678173931933,
"learning_rate": 3.0981400591799354e-06,
"loss": 0.8644,
"mean_token_accuracy": 0.7790065869688988,
"num_tokens": 517646655.0,
"step": 5840
},
{
"entropy": 0.918505859375,
"epoch": 0.7995626324062052,
"grad_norm": 0.5824562922891173,
"learning_rate": 3.0946174439904185e-06,
"loss": 0.8867,
"mean_token_accuracy": 0.7751597911119461,
"num_tokens": 518550530.0,
"step": 5850
},
{
"entropy": 0.9150146484375,
"epoch": 0.8009294061368141,
"grad_norm": 0.614798575400779,
"learning_rate": 3.0910948288009016e-06,
"loss": 0.8797,
"mean_token_accuracy": 0.7719022639095783,
"num_tokens": 519408238.0,
"step": 5860
},
{
"entropy": 0.916064453125,
"epoch": 0.802296179867423,
"grad_norm": 0.6585036081849968,
"learning_rate": 3.0875722136113856e-06,
"loss": 0.9123,
"mean_token_accuracy": 0.7694119311869144,
"num_tokens": 520283433.0,
"step": 5870
},
{
"entropy": 0.948681640625,
"epoch": 0.8036629535980319,
"grad_norm": 0.6472911806550323,
"learning_rate": 3.0840495984218687e-06,
"loss": 0.9296,
"mean_token_accuracy": 0.7658033281564712,
"num_tokens": 521184354.0,
"step": 5880
},
{
"entropy": 0.94599609375,
"epoch": 0.8050297273286408,
"grad_norm": 0.5968939216336171,
"learning_rate": 3.080526983232352e-06,
"loss": 0.9459,
"mean_token_accuracy": 0.7636468335986137,
"num_tokens": 522052078.0,
"step": 5890
},
{
"entropy": 0.91806640625,
"epoch": 0.8063965010592496,
"grad_norm": 0.8828755227859019,
"learning_rate": 3.0770043680428354e-06,
"loss": 0.9063,
"mean_token_accuracy": 0.7679909251630306,
"num_tokens": 522948969.0,
"step": 5900
},
{
"entropy": 0.89541015625,
"epoch": 0.8077632747898585,
"grad_norm": 0.5810388416458373,
"learning_rate": 3.0734817528533185e-06,
"loss": 0.8666,
"mean_token_accuracy": 0.7798352010548115,
"num_tokens": 523817470.0,
"step": 5910
},
{
"entropy": 0.959423828125,
"epoch": 0.8091300485204674,
"grad_norm": 0.6262942222539903,
"learning_rate": 3.0699591376638017e-06,
"loss": 0.9465,
"mean_token_accuracy": 0.7625944443047047,
"num_tokens": 524695050.0,
"step": 5920
},
{
"entropy": 0.90556640625,
"epoch": 0.8104968222510763,
"grad_norm": 0.6345048465408092,
"learning_rate": 3.066436522474285e-06,
"loss": 0.8773,
"mean_token_accuracy": 0.7758624352514744,
"num_tokens": 525574755.0,
"step": 5930
},
{
"entropy": 0.873486328125,
"epoch": 0.8118635959816852,
"grad_norm": 0.6048328389970125,
"learning_rate": 3.0629139072847688e-06,
"loss": 0.86,
"mean_token_accuracy": 0.7805297449231148,
"num_tokens": 526509631.0,
"step": 5940
},
{
"entropy": 0.939208984375,
"epoch": 0.8132303697122941,
"grad_norm": 0.656559413545462,
"learning_rate": 3.059391292095252e-06,
"loss": 0.9258,
"mean_token_accuracy": 0.7641810491681099,
"num_tokens": 527382381.0,
"step": 5950
},
{
"entropy": 0.916015625,
"epoch": 0.814597143442903,
"grad_norm": 0.6216726881303944,
"learning_rate": 3.055868676905735e-06,
"loss": 0.9109,
"mean_token_accuracy": 0.7715377248823643,
"num_tokens": 528322905.0,
"step": 5960
},
{
"entropy": 0.9287109375,
"epoch": 0.8159639171735119,
"grad_norm": 0.634878948983,
"learning_rate": 3.052346061716218e-06,
"loss": 0.9295,
"mean_token_accuracy": 0.7675123102962971,
"num_tokens": 529151743.0,
"step": 5970
},
{
"entropy": 0.907666015625,
"epoch": 0.8173306909041208,
"grad_norm": 0.660200243676581,
"learning_rate": 3.048823446526702e-06,
"loss": 0.9122,
"mean_token_accuracy": 0.7696065753698349,
"num_tokens": 530042986.0,
"step": 5980
},
{
"entropy": 0.9489990234375,
"epoch": 0.8186974646347297,
"grad_norm": 0.652797876409655,
"learning_rate": 3.0453008313371852e-06,
"loss": 0.9313,
"mean_token_accuracy": 0.765207851678133,
"num_tokens": 530980320.0,
"step": 5990
},
{
"entropy": 0.919189453125,
"epoch": 0.8200642383653386,
"grad_norm": 0.6220957333030199,
"learning_rate": 3.0417782161476683e-06,
"loss": 0.9078,
"mean_token_accuracy": 0.7709616832435131,
"num_tokens": 531854285.0,
"step": 6000
},
{
"entropy": 0.971044921875,
"epoch": 0.8214310120959475,
"grad_norm": 0.6626224344804188,
"learning_rate": 3.0382556009581515e-06,
"loss": 0.954,
"mean_token_accuracy": 0.7644939221441746,
"num_tokens": 532722975.0,
"step": 6010
},
{
"entropy": 0.903271484375,
"epoch": 0.8227977858265564,
"grad_norm": 0.6061249827974882,
"learning_rate": 3.0347329857686346e-06,
"loss": 0.8771,
"mean_token_accuracy": 0.7723933540284633,
"num_tokens": 533582773.0,
"step": 6020
},
{
"entropy": 0.9044921875,
"epoch": 0.8241645595571653,
"grad_norm": 0.6337756534731916,
"learning_rate": 3.031210370579118e-06,
"loss": 0.8895,
"mean_token_accuracy": 0.7772169165313244,
"num_tokens": 534476179.0,
"step": 6030
},
{
"entropy": 0.914404296875,
"epoch": 0.8255313332877742,
"grad_norm": 0.5546062632650284,
"learning_rate": 3.0276877553896017e-06,
"loss": 0.8834,
"mean_token_accuracy": 0.7725088521838188,
"num_tokens": 535353086.0,
"step": 6040
},
{
"entropy": 0.92509765625,
"epoch": 0.8268981070183831,
"grad_norm": 0.6892767074175299,
"learning_rate": 3.024165140200085e-06,
"loss": 0.904,
"mean_token_accuracy": 0.7712945282459259,
"num_tokens": 536231879.0,
"step": 6050
},
{
"entropy": 0.924658203125,
"epoch": 0.828264880748992,
"grad_norm": 0.6411968152643918,
"learning_rate": 3.020642525010568e-06,
"loss": 0.917,
"mean_token_accuracy": 0.770191915333271,
"num_tokens": 537112252.0,
"step": 6060
},
{
"entropy": 0.89501953125,
"epoch": 0.8296316544796009,
"grad_norm": 0.6371907975606428,
"learning_rate": 3.0171199098210515e-06,
"loss": 0.8731,
"mean_token_accuracy": 0.7768353655934334,
"num_tokens": 537972593.0,
"step": 6070
},
{
"entropy": 0.942626953125,
"epoch": 0.8309984282102097,
"grad_norm": 0.6272734434785319,
"learning_rate": 3.0135972946315346e-06,
"loss": 0.9077,
"mean_token_accuracy": 0.7715313971042633,
"num_tokens": 538871586.0,
"step": 6080
},
{
"entropy": 0.928564453125,
"epoch": 0.8323652019408186,
"grad_norm": 0.6165708475231523,
"learning_rate": 3.010074679442018e-06,
"loss": 0.9184,
"mean_token_accuracy": 0.7693241387605667,
"num_tokens": 539732436.0,
"step": 6090
},
{
"entropy": 0.926513671875,
"epoch": 0.8337319756714276,
"grad_norm": 0.799453440992133,
"learning_rate": 3.0065520642525013e-06,
"loss": 0.9336,
"mean_token_accuracy": 0.7663158543407917,
"num_tokens": 540600028.0,
"step": 6100
},
{
"entropy": 0.960693359375,
"epoch": 0.8350987494020365,
"grad_norm": 0.6610891124646735,
"learning_rate": 3.003029449062985e-06,
"loss": 0.9252,
"mean_token_accuracy": 0.766525349766016,
"num_tokens": 541502670.0,
"step": 6110
},
{
"entropy": 0.9478515625,
"epoch": 0.8364655231326454,
"grad_norm": 0.688276177902839,
"learning_rate": 2.999506833873468e-06,
"loss": 0.9256,
"mean_token_accuracy": 0.7657857276499271,
"num_tokens": 542340098.0,
"step": 6120
},
{
"entropy": 0.936767578125,
"epoch": 0.8378322968632543,
"grad_norm": 0.63648294063846,
"learning_rate": 2.995984218683951e-06,
"loss": 0.9241,
"mean_token_accuracy": 0.763423866033554,
"num_tokens": 543218409.0,
"step": 6130
},
{
"entropy": 0.90771484375,
"epoch": 0.8391990705938632,
"grad_norm": 0.5708443171939778,
"learning_rate": 2.992461603494434e-06,
"loss": 0.8979,
"mean_token_accuracy": 0.7715304464101791,
"num_tokens": 544126052.0,
"step": 6140
},
{
"entropy": 0.92548828125,
"epoch": 0.8405658443244721,
"grad_norm": 0.6622472074763663,
"learning_rate": 2.988938988304918e-06,
"loss": 0.9159,
"mean_token_accuracy": 0.7686504371464252,
"num_tokens": 545029025.0,
"step": 6150
},
{
"entropy": 0.904150390625,
"epoch": 0.841932618055081,
"grad_norm": 0.5669670677615852,
"learning_rate": 2.9854163731154013e-06,
"loss": 0.8931,
"mean_token_accuracy": 0.774728886038065,
"num_tokens": 545905476.0,
"step": 6160
},
{
"entropy": 0.928271484375,
"epoch": 0.8432993917856899,
"grad_norm": 0.5833554169513141,
"learning_rate": 2.9818937579258844e-06,
"loss": 0.9339,
"mean_token_accuracy": 0.7652238883078099,
"num_tokens": 546749480.0,
"step": 6170
},
{
"entropy": 0.8931640625,
"epoch": 0.8446661655162988,
"grad_norm": 0.6369100875234818,
"learning_rate": 2.9783711427363675e-06,
"loss": 0.8813,
"mean_token_accuracy": 0.7770978838205338,
"num_tokens": 547635672.0,
"step": 6180
},
{
"entropy": 0.94091796875,
"epoch": 0.8460329392469077,
"grad_norm": 0.6841071068131978,
"learning_rate": 2.9748485275468507e-06,
"loss": 0.9125,
"mean_token_accuracy": 0.7708933457732201,
"num_tokens": 548511563.0,
"step": 6190
},
{
"entropy": 0.911083984375,
"epoch": 0.8473997129775166,
"grad_norm": 0.6033642450867948,
"learning_rate": 2.9713259123573346e-06,
"loss": 0.9059,
"mean_token_accuracy": 0.7719939380884171,
"num_tokens": 549358293.0,
"step": 6200
},
{
"entropy": 0.914697265625,
"epoch": 0.8487664867081255,
"grad_norm": 0.604820626331906,
"learning_rate": 2.9678032971678177e-06,
"loss": 0.9066,
"mean_token_accuracy": 0.7700458236038685,
"num_tokens": 550199472.0,
"step": 6210
},
{
"entropy": 0.9310546875,
"epoch": 0.8501332604387344,
"grad_norm": 0.6319763082131945,
"learning_rate": 2.964280681978301e-06,
"loss": 0.927,
"mean_token_accuracy": 0.7660935342311859,
"num_tokens": 551109412.0,
"step": 6220
},
{
"entropy": 0.8806640625,
"epoch": 0.8515000341693433,
"grad_norm": 0.7150246677406059,
"learning_rate": 2.960758066788784e-06,
"loss": 0.8642,
"mean_token_accuracy": 0.7775402359664441,
"num_tokens": 551947950.0,
"step": 6230
},
{
"entropy": 0.946484375,
"epoch": 0.8528668078999522,
"grad_norm": 0.622864861790224,
"learning_rate": 2.9572354515992675e-06,
"loss": 0.9501,
"mean_token_accuracy": 0.7584501430392265,
"num_tokens": 552800468.0,
"step": 6240
},
{
"entropy": 0.865576171875,
"epoch": 0.8542335816305611,
"grad_norm": 0.6184363502339332,
"learning_rate": 2.9537128364097507e-06,
"loss": 0.8391,
"mean_token_accuracy": 0.7862398467957974,
"num_tokens": 553683283.0,
"step": 6250
},
{
"entropy": 0.9090087890625,
"epoch": 0.85560035536117,
"grad_norm": 0.6104968211671159,
"learning_rate": 2.9501902212202342e-06,
"loss": 0.8971,
"mean_token_accuracy": 0.774077731370926,
"num_tokens": 554603674.0,
"step": 6260
},
{
"entropy": 0.93515625,
"epoch": 0.8569671290917789,
"grad_norm": 0.6021245588754303,
"learning_rate": 2.9466676060307178e-06,
"loss": 0.9186,
"mean_token_accuracy": 0.7673637121915817,
"num_tokens": 555498911.0,
"step": 6270
},
{
"entropy": 0.936669921875,
"epoch": 0.8583339028223878,
"grad_norm": 0.6379675875417462,
"learning_rate": 2.943144990841201e-06,
"loss": 0.9215,
"mean_token_accuracy": 0.7691512852907181,
"num_tokens": 556398231.0,
"step": 6280
},
{
"entropy": 0.915478515625,
"epoch": 0.8597006765529966,
"grad_norm": 0.6361788568086163,
"learning_rate": 2.939622375651684e-06,
"loss": 0.8821,
"mean_token_accuracy": 0.7756883852183819,
"num_tokens": 557289325.0,
"step": 6290
},
{
"entropy": 0.941943359375,
"epoch": 0.8610674502836055,
"grad_norm": 0.7563610734373586,
"learning_rate": 2.936099760462167e-06,
"loss": 0.9203,
"mean_token_accuracy": 0.7674103945493698,
"num_tokens": 558215409.0,
"step": 6300
},
{
"entropy": 0.894580078125,
"epoch": 0.8624342240142144,
"grad_norm": 0.6335202870125012,
"learning_rate": 2.932577145272651e-06,
"loss": 0.8731,
"mean_token_accuracy": 0.7783588618040085,
"num_tokens": 559078072.0,
"step": 6310
},
{
"entropy": 0.9382080078125,
"epoch": 0.8638009977448233,
"grad_norm": 0.5847335961711413,
"learning_rate": 2.9290545300831342e-06,
"loss": 0.9319,
"mean_token_accuracy": 0.766331996023655,
"num_tokens": 559985199.0,
"step": 6320
},
{
"entropy": 0.940576171875,
"epoch": 0.8651677714754322,
"grad_norm": 0.6310490560235693,
"learning_rate": 2.9255319148936174e-06,
"loss": 0.9299,
"mean_token_accuracy": 0.7678322218358516,
"num_tokens": 560873774.0,
"step": 6330
},
{
"entropy": 0.892724609375,
"epoch": 0.8665345452060411,
"grad_norm": 0.6303542339777937,
"learning_rate": 2.9220092997041005e-06,
"loss": 0.8536,
"mean_token_accuracy": 0.7811123430728912,
"num_tokens": 561773112.0,
"step": 6340
},
{
"entropy": 0.929931640625,
"epoch": 0.86790131893665,
"grad_norm": 0.5930221808682801,
"learning_rate": 2.9184866845145836e-06,
"loss": 0.9108,
"mean_token_accuracy": 0.7697659648954869,
"num_tokens": 562659591.0,
"step": 6350
},
{
"entropy": 0.9357421875,
"epoch": 0.8692680926672589,
"grad_norm": 0.6746973549845877,
"learning_rate": 2.9149640693250667e-06,
"loss": 0.9111,
"mean_token_accuracy": 0.7693645000457764,
"num_tokens": 563573747.0,
"step": 6360
},
{
"entropy": 0.914404296875,
"epoch": 0.8706348663978678,
"grad_norm": 0.6609440703580748,
"learning_rate": 2.9114414541355507e-06,
"loss": 0.9016,
"mean_token_accuracy": 0.7723480820655823,
"num_tokens": 564444810.0,
"step": 6370
},
{
"entropy": 0.928515625,
"epoch": 0.8720016401284767,
"grad_norm": 0.596460473472294,
"learning_rate": 2.907918838946034e-06,
"loss": 0.9201,
"mean_token_accuracy": 0.7672720484435558,
"num_tokens": 565334063.0,
"step": 6380
},
{
"entropy": 0.9003173828125,
"epoch": 0.8733684138590856,
"grad_norm": 0.6639089326449532,
"learning_rate": 2.904396223756517e-06,
"loss": 0.8997,
"mean_token_accuracy": 0.7712231576442719,
"num_tokens": 566217603.0,
"step": 6390
},
{
"entropy": 0.9229248046875,
"epoch": 0.8747351875896945,
"grad_norm": 0.6504951000786876,
"learning_rate": 2.9008736085670005e-06,
"loss": 0.9131,
"mean_token_accuracy": 0.7696467898786068,
"num_tokens": 567103733.0,
"step": 6400
},
{
"entropy": 0.8953125,
"epoch": 0.8761019613203034,
"grad_norm": 0.6914762259187871,
"learning_rate": 2.8973509933774836e-06,
"loss": 0.8611,
"mean_token_accuracy": 0.7776408359408379,
"num_tokens": 567958707.0,
"step": 6410
},
{
"entropy": 0.9339599609375,
"epoch": 0.8774687350509123,
"grad_norm": 0.5956720301526611,
"learning_rate": 2.893828378187967e-06,
"loss": 0.9438,
"mean_token_accuracy": 0.7653645791113377,
"num_tokens": 568857102.0,
"step": 6420
},
{
"entropy": 0.865771484375,
"epoch": 0.8788355087815212,
"grad_norm": 0.6524575897886308,
"learning_rate": 2.8903057629984503e-06,
"loss": 0.8661,
"mean_token_accuracy": 0.7783777870237827,
"num_tokens": 569724676.0,
"step": 6430
},
{
"entropy": 0.89609375,
"epoch": 0.8802022825121301,
"grad_norm": 0.6398375708814946,
"learning_rate": 2.886783147808934e-06,
"loss": 0.88,
"mean_token_accuracy": 0.7751635760068893,
"num_tokens": 570626696.0,
"step": 6440
},
{
"entropy": 0.968603515625,
"epoch": 0.8815690562427391,
"grad_norm": 0.6212038714529206,
"learning_rate": 2.883260532619417e-06,
"loss": 0.9421,
"mean_token_accuracy": 0.7609601475298404,
"num_tokens": 571490412.0,
"step": 6450
},
{
"entropy": 0.918798828125,
"epoch": 0.882935829973348,
"grad_norm": 0.7026829184441367,
"learning_rate": 2.8797379174299e-06,
"loss": 0.9037,
"mean_token_accuracy": 0.7686360657215119,
"num_tokens": 572332785.0,
"step": 6460
},
{
"entropy": 0.882763671875,
"epoch": 0.8843026037039569,
"grad_norm": 0.606246779982647,
"learning_rate": 2.876215302240383e-06,
"loss": 0.8492,
"mean_token_accuracy": 0.7833980552852153,
"num_tokens": 573204987.0,
"step": 6470
},
{
"entropy": 0.953662109375,
"epoch": 0.8856693774345658,
"grad_norm": 0.6559776587897801,
"learning_rate": 2.872692687050867e-06,
"loss": 0.9487,
"mean_token_accuracy": 0.7612179234623909,
"num_tokens": 574117752.0,
"step": 6480
},
{
"entropy": 0.933984375,
"epoch": 0.8870361511651746,
"grad_norm": 0.6037516125198049,
"learning_rate": 2.8691700718613503e-06,
"loss": 0.9285,
"mean_token_accuracy": 0.7662662513554096,
"num_tokens": 575003850.0,
"step": 6490
},
{
"entropy": 0.9005126953125,
"epoch": 0.8884029248957835,
"grad_norm": 0.6428731437902545,
"learning_rate": 2.8656474566718334e-06,
"loss": 0.9073,
"mean_token_accuracy": 0.769654193520546,
"num_tokens": 575921233.0,
"step": 6500
},
{
"entropy": 0.89140625,
"epoch": 0.8897696986263924,
"grad_norm": 0.571238360444478,
"learning_rate": 2.8621248414823165e-06,
"loss": 0.8844,
"mean_token_accuracy": 0.773760923743248,
"num_tokens": 576814818.0,
"step": 6510
},
{
"entropy": 0.936572265625,
"epoch": 0.8911364723570013,
"grad_norm": 0.7141844979182281,
"learning_rate": 2.8586022262927997e-06,
"loss": 0.9004,
"mean_token_accuracy": 0.7692235223948956,
"num_tokens": 577627640.0,
"step": 6520
},
{
"entropy": 0.94833984375,
"epoch": 0.8925032460876102,
"grad_norm": 0.6147243454849097,
"learning_rate": 2.8550796111032836e-06,
"loss": 0.9448,
"mean_token_accuracy": 0.7636757589876652,
"num_tokens": 578521956.0,
"step": 6530
},
{
"entropy": 0.915576171875,
"epoch": 0.8938700198182191,
"grad_norm": 0.6318104173969177,
"learning_rate": 2.8515569959137668e-06,
"loss": 0.9156,
"mean_token_accuracy": 0.7713035903871059,
"num_tokens": 579395782.0,
"step": 6540
},
{
"entropy": 0.925927734375,
"epoch": 0.895236793548828,
"grad_norm": 0.6276000848861687,
"learning_rate": 2.84803438072425e-06,
"loss": 0.902,
"mean_token_accuracy": 0.7721777103841305,
"num_tokens": 580272157.0,
"step": 6550
},
{
"entropy": 0.92705078125,
"epoch": 0.8966035672794369,
"grad_norm": 0.619119669825671,
"learning_rate": 2.844511765534733e-06,
"loss": 0.897,
"mean_token_accuracy": 0.7735455922782422,
"num_tokens": 581146537.0,
"step": 6560
},
{
"entropy": 0.8859619140625,
"epoch": 0.8979703410100458,
"grad_norm": 0.5543858284536771,
"learning_rate": 2.8409891503452166e-06,
"loss": 0.8673,
"mean_token_accuracy": 0.7771421484649181,
"num_tokens": 582048135.0,
"step": 6570
},
{
"entropy": 0.922607421875,
"epoch": 0.8993371147406547,
"grad_norm": 0.6584548041615412,
"learning_rate": 2.8374665351556997e-06,
"loss": 0.9322,
"mean_token_accuracy": 0.7660396069288253,
"num_tokens": 582904287.0,
"step": 6580
},
{
"entropy": 0.92919921875,
"epoch": 0.9007038884712636,
"grad_norm": 0.6073771673217173,
"learning_rate": 2.8339439199661832e-06,
"loss": 0.9081,
"mean_token_accuracy": 0.7705400481820106,
"num_tokens": 583826125.0,
"step": 6590
},
{
"entropy": 0.93798828125,
"epoch": 0.9020706622018725,
"grad_norm": 0.6435343551890607,
"learning_rate": 2.8304213047766663e-06,
"loss": 0.9202,
"mean_token_accuracy": 0.7686318263411522,
"num_tokens": 584674987.0,
"step": 6600
},
{
"entropy": 0.88916015625,
"epoch": 0.9034374359324814,
"grad_norm": 0.5993644783258539,
"learning_rate": 2.82689868958715e-06,
"loss": 0.8653,
"mean_token_accuracy": 0.7800382681190967,
"num_tokens": 585567786.0,
"step": 6610
},
{
"entropy": 0.924365234375,
"epoch": 0.9048042096630903,
"grad_norm": 0.6594929595714428,
"learning_rate": 2.823376074397633e-06,
"loss": 0.9064,
"mean_token_accuracy": 0.7689058221876621,
"num_tokens": 586446088.0,
"step": 6620
},
{
"entropy": 0.9078857421875,
"epoch": 0.9061709833936992,
"grad_norm": 0.6511223727875353,
"learning_rate": 2.819853459208116e-06,
"loss": 0.8939,
"mean_token_accuracy": 0.7732744947075844,
"num_tokens": 587291210.0,
"step": 6630
},
{
"entropy": 0.959423828125,
"epoch": 0.9075377571243081,
"grad_norm": 0.7469773831840323,
"learning_rate": 2.8163308440186e-06,
"loss": 0.9682,
"mean_token_accuracy": 0.7597765840590001,
"num_tokens": 588175587.0,
"step": 6640
},
{
"entropy": 0.923583984375,
"epoch": 0.908904530854917,
"grad_norm": 0.6754403186270822,
"learning_rate": 2.8128082288290832e-06,
"loss": 0.9102,
"mean_token_accuracy": 0.769044502824545,
"num_tokens": 589051406.0,
"step": 6650
},
{
"entropy": 0.9318359375,
"epoch": 0.9102713045855259,
"grad_norm": 0.6500948842908778,
"learning_rate": 2.8092856136395664e-06,
"loss": 0.912,
"mean_token_accuracy": 0.7692054845392704,
"num_tokens": 589942110.0,
"step": 6660
},
{
"entropy": 0.936474609375,
"epoch": 0.9116380783161347,
"grad_norm": 0.6210777958867869,
"learning_rate": 2.8057629984500495e-06,
"loss": 0.9286,
"mean_token_accuracy": 0.7643832109868527,
"num_tokens": 590877907.0,
"step": 6670
},
{
"entropy": 0.904150390625,
"epoch": 0.9130048520467436,
"grad_norm": 0.5992541996959254,
"learning_rate": 2.8022403832605326e-06,
"loss": 0.8884,
"mean_token_accuracy": 0.7755085773766041,
"num_tokens": 591773456.0,
"step": 6680
},
{
"entropy": 0.903955078125,
"epoch": 0.9143716257773525,
"grad_norm": 0.6069444202227541,
"learning_rate": 2.7987177680710157e-06,
"loss": 0.9136,
"mean_token_accuracy": 0.77174222022295,
"num_tokens": 592686986.0,
"step": 6690
},
{
"entropy": 0.9357421875,
"epoch": 0.9157383995079614,
"grad_norm": 0.6522015589542162,
"learning_rate": 2.7951951528814997e-06,
"loss": 0.9273,
"mean_token_accuracy": 0.7669572554528713,
"num_tokens": 593589156.0,
"step": 6700
},
{
"entropy": 0.919921875,
"epoch": 0.9171051732385703,
"grad_norm": 0.6926043979926507,
"learning_rate": 2.791672537691983e-06,
"loss": 0.9392,
"mean_token_accuracy": 0.7646652020514011,
"num_tokens": 594522006.0,
"step": 6710
},
{
"entropy": 0.8939208984375,
"epoch": 0.9184719469691792,
"grad_norm": 0.6306731156965873,
"learning_rate": 2.788149922502466e-06,
"loss": 0.8918,
"mean_token_accuracy": 0.772183109074831,
"num_tokens": 595400912.0,
"step": 6720
},
{
"entropy": 0.899560546875,
"epoch": 0.9198387206997881,
"grad_norm": 0.7638965212362246,
"learning_rate": 2.784627307312949e-06,
"loss": 0.8891,
"mean_token_accuracy": 0.7752847649157047,
"num_tokens": 596309300.0,
"step": 6730
},
{
"entropy": 0.891015625,
"epoch": 0.921205494430397,
"grad_norm": 0.6203359094785733,
"learning_rate": 2.7811046921234326e-06,
"loss": 0.853,
"mean_token_accuracy": 0.7779567196965218,
"num_tokens": 597210330.0,
"step": 6740
},
{
"entropy": 0.94130859375,
"epoch": 0.9225722681610059,
"grad_norm": 0.6223047894876488,
"learning_rate": 2.777582076933916e-06,
"loss": 0.9372,
"mean_token_accuracy": 0.76533083319664,
"num_tokens": 598081617.0,
"step": 6750
},
{
"entropy": 0.91171875,
"epoch": 0.9239390418916148,
"grad_norm": 0.5527296440525852,
"learning_rate": 2.7740594617443993e-06,
"loss": 0.9074,
"mean_token_accuracy": 0.7691813029348851,
"num_tokens": 598983716.0,
"step": 6760
},
{
"entropy": 0.887451171875,
"epoch": 0.9253058156222237,
"grad_norm": 0.6654086904966099,
"learning_rate": 2.770536846554883e-06,
"loss": 0.8762,
"mean_token_accuracy": 0.7779003098607064,
"num_tokens": 599918368.0,
"step": 6770
},
{
"entropy": 0.9203125,
"epoch": 0.9266725893528326,
"grad_norm": 0.6216708512694121,
"learning_rate": 2.767014231365366e-06,
"loss": 0.9305,
"mean_token_accuracy": 0.7681852832436562,
"num_tokens": 600794870.0,
"step": 6780
},
{
"entropy": 0.92099609375,
"epoch": 0.9280393630834415,
"grad_norm": 0.7013082219213335,
"learning_rate": 2.763491616175849e-06,
"loss": 0.8943,
"mean_token_accuracy": 0.7717525355517865,
"num_tokens": 601726884.0,
"step": 6790
},
{
"entropy": 0.896044921875,
"epoch": 0.9294061368140505,
"grad_norm": 0.6773330255327682,
"learning_rate": 2.759969000986332e-06,
"loss": 0.8899,
"mean_token_accuracy": 0.7728159792721272,
"num_tokens": 602627133.0,
"step": 6800
},
{
"entropy": 0.8740966796875,
"epoch": 0.9307729105446594,
"grad_norm": 0.6034330147489998,
"learning_rate": 2.756446385796816e-06,
"loss": 0.8416,
"mean_token_accuracy": 0.7820419616997242,
"num_tokens": 603501852.0,
"step": 6810
},
{
"entropy": 0.9007568359375,
"epoch": 0.9321396842752683,
"grad_norm": 0.5543588001007758,
"learning_rate": 2.7529237706072993e-06,
"loss": 0.8842,
"mean_token_accuracy": 0.7732481978833675,
"num_tokens": 604404913.0,
"step": 6820
},
{
"entropy": 0.9181884765625,
"epoch": 0.9335064580058772,
"grad_norm": 0.5984241160480006,
"learning_rate": 2.7494011554177824e-06,
"loss": 0.8896,
"mean_token_accuracy": 0.7724772319197655,
"num_tokens": 605285242.0,
"step": 6830
},
{
"entropy": 0.9405517578125,
"epoch": 0.9348732317364861,
"grad_norm": 0.647255156086798,
"learning_rate": 2.7458785402282656e-06,
"loss": 0.9384,
"mean_token_accuracy": 0.7642705902457237,
"num_tokens": 606154610.0,
"step": 6840
},
{
"entropy": 0.9399658203125,
"epoch": 0.936240005467095,
"grad_norm": 0.6385797832223212,
"learning_rate": 2.7423559250387487e-06,
"loss": 0.9323,
"mean_token_accuracy": 0.7659270398318767,
"num_tokens": 607046840.0,
"step": 6850
},
{
"entropy": 0.9072998046875,
"epoch": 0.9376067791977039,
"grad_norm": 0.6762822389377449,
"learning_rate": 2.7388333098492326e-06,
"loss": 0.8793,
"mean_token_accuracy": 0.7733124576508998,
"num_tokens": 607882778.0,
"step": 6860
},
{
"entropy": 0.893701171875,
"epoch": 0.9389735529283127,
"grad_norm": 0.6407203232010085,
"learning_rate": 2.7353106946597158e-06,
"loss": 0.9012,
"mean_token_accuracy": 0.7707997992634773,
"num_tokens": 608824241.0,
"step": 6870
},
{
"entropy": 0.923828125,
"epoch": 0.9403403266589216,
"grad_norm": 0.6945025490145846,
"learning_rate": 2.731788079470199e-06,
"loss": 0.9291,
"mean_token_accuracy": 0.7666601352393627,
"num_tokens": 609676651.0,
"step": 6880
},
{
"entropy": 0.919970703125,
"epoch": 0.9417071003895305,
"grad_norm": 0.6057133608032963,
"learning_rate": 2.728265464280682e-06,
"loss": 0.9155,
"mean_token_accuracy": 0.7688558541238308,
"num_tokens": 610543423.0,
"step": 6890
},
{
"entropy": 0.9157470703125,
"epoch": 0.9430738741201394,
"grad_norm": 0.6113087732335587,
"learning_rate": 2.7247428490911656e-06,
"loss": 0.9231,
"mean_token_accuracy": 0.7689621895551682,
"num_tokens": 611399783.0,
"step": 6900
},
{
"entropy": 0.9144287109375,
"epoch": 0.9444406478507483,
"grad_norm": 0.7091224796293008,
"learning_rate": 2.7212202339016487e-06,
"loss": 0.8951,
"mean_token_accuracy": 0.7737138338387013,
"num_tokens": 612298608.0,
"step": 6910
},
{
"entropy": 0.902294921875,
"epoch": 0.9458074215813572,
"grad_norm": 0.6692044239869139,
"learning_rate": 2.7176976187121322e-06,
"loss": 0.9093,
"mean_token_accuracy": 0.7720749072730542,
"num_tokens": 613248091.0,
"step": 6920
},
{
"entropy": 0.91767578125,
"epoch": 0.9471741953119661,
"grad_norm": 0.6754101904614696,
"learning_rate": 2.7141750035226154e-06,
"loss": 0.9113,
"mean_token_accuracy": 0.7679714664816857,
"num_tokens": 614177532.0,
"step": 6930
},
{
"entropy": 0.92890625,
"epoch": 0.948540969042575,
"grad_norm": 0.6257878077443771,
"learning_rate": 2.710652388333099e-06,
"loss": 0.9187,
"mean_token_accuracy": 0.7703791208565235,
"num_tokens": 615083286.0,
"step": 6940
},
{
"entropy": 0.94853515625,
"epoch": 0.9499077427731839,
"grad_norm": 0.5727037626991626,
"learning_rate": 2.707129773143582e-06,
"loss": 0.9219,
"mean_token_accuracy": 0.768681874871254,
"num_tokens": 616002316.0,
"step": 6950
},
{
"entropy": 0.8883544921875,
"epoch": 0.9512745165037928,
"grad_norm": 0.6317298387176081,
"learning_rate": 2.703607157954065e-06,
"loss": 0.8666,
"mean_token_accuracy": 0.7766501165926456,
"num_tokens": 616898102.0,
"step": 6960
},
{
"entropy": 0.937060546875,
"epoch": 0.9526412902344017,
"grad_norm": 0.6424781507380951,
"learning_rate": 2.7000845427645483e-06,
"loss": 0.9133,
"mean_token_accuracy": 0.7682504989206791,
"num_tokens": 617749598.0,
"step": 6970
},
{
"entropy": 0.883349609375,
"epoch": 0.9540080639650106,
"grad_norm": 0.6359985282765259,
"learning_rate": 2.6965619275750322e-06,
"loss": 0.8703,
"mean_token_accuracy": 0.7791894145309926,
"num_tokens": 618614766.0,
"step": 6980
},
{
"entropy": 0.9607666015625,
"epoch": 0.9553748376956195,
"grad_norm": 0.6741941925868783,
"learning_rate": 2.6930393123855154e-06,
"loss": 0.9703,
"mean_token_accuracy": 0.761593596637249,
"num_tokens": 619527250.0,
"step": 6990
},
{
"entropy": 0.95849609375,
"epoch": 0.9567416114262284,
"grad_norm": 0.6406496747431639,
"learning_rate": 2.6895166971959985e-06,
"loss": 0.9746,
"mean_token_accuracy": 0.7588690511882306,
"num_tokens": 620389290.0,
"step": 7000
},
{
"entropy": 0.8803955078125,
"epoch": 0.9581083851568373,
"grad_norm": 0.6832638914878877,
"learning_rate": 2.6859940820064816e-06,
"loss": 0.8624,
"mean_token_accuracy": 0.7803837396204472,
"num_tokens": 621260557.0,
"step": 7010
},
{
"entropy": 0.944677734375,
"epoch": 0.9594751588874462,
"grad_norm": 0.72924826225456,
"learning_rate": 2.6824714668169647e-06,
"loss": 0.9272,
"mean_token_accuracy": 0.7654004283249378,
"num_tokens": 622132062.0,
"step": 7020
},
{
"entropy": 0.92763671875,
"epoch": 0.9608419326180551,
"grad_norm": 0.6238006996635325,
"learning_rate": 2.6789488516274487e-06,
"loss": 0.9273,
"mean_token_accuracy": 0.768135180324316,
"num_tokens": 623083505.0,
"step": 7030
},
{
"entropy": 0.91640625,
"epoch": 0.962208706348664,
"grad_norm": 0.9550322558661815,
"learning_rate": 2.675426236437932e-06,
"loss": 0.9124,
"mean_token_accuracy": 0.7712678030133248,
"num_tokens": 624013009.0,
"step": 7040
},
{
"entropy": 0.893798828125,
"epoch": 0.9635754800792729,
"grad_norm": 0.5926062805900209,
"learning_rate": 2.671903621248415e-06,
"loss": 0.8965,
"mean_token_accuracy": 0.7711364291608334,
"num_tokens": 624866319.0,
"step": 7050
},
{
"entropy": 0.908251953125,
"epoch": 0.9649422538098817,
"grad_norm": 0.632081281679128,
"learning_rate": 2.668381006058898e-06,
"loss": 0.9114,
"mean_token_accuracy": 0.7693822115659714,
"num_tokens": 625727325.0,
"step": 7060
},
{
"entropy": 0.939794921875,
"epoch": 0.9663090275404906,
"grad_norm": 0.681585632823408,
"learning_rate": 2.6648583908693816e-06,
"loss": 0.921,
"mean_token_accuracy": 0.7665778495371341,
"num_tokens": 626605530.0,
"step": 7070
},
{
"entropy": 0.967236328125,
"epoch": 0.9676758012710995,
"grad_norm": 0.6876795894160002,
"learning_rate": 2.6613357756798648e-06,
"loss": 0.9767,
"mean_token_accuracy": 0.7585014723241329,
"num_tokens": 627487707.0,
"step": 7080
},
{
"entropy": 0.92841796875,
"epoch": 0.9690425750017084,
"grad_norm": 0.6441368601382563,
"learning_rate": 2.6578131604903483e-06,
"loss": 0.9026,
"mean_token_accuracy": 0.7725687108933925,
"num_tokens": 628336825.0,
"step": 7090
},
{
"entropy": 0.9029296875,
"epoch": 0.9704093487323173,
"grad_norm": 0.627882759241602,
"learning_rate": 2.6542905453008314e-06,
"loss": 0.8962,
"mean_token_accuracy": 0.7715781547129155,
"num_tokens": 629226993.0,
"step": 7100
},
{
"entropy": 0.924853515625,
"epoch": 0.9717761224629262,
"grad_norm": 0.5944612334051913,
"learning_rate": 2.650767930111315e-06,
"loss": 0.9042,
"mean_token_accuracy": 0.771028795838356,
"num_tokens": 630127171.0,
"step": 7110
},
{
"entropy": 0.9123046875,
"epoch": 0.9731428961935351,
"grad_norm": 0.6422182321876576,
"learning_rate": 2.647245314921798e-06,
"loss": 0.9035,
"mean_token_accuracy": 0.7726425133645535,
"num_tokens": 631049821.0,
"step": 7120
},
{
"entropy": 0.90615234375,
"epoch": 0.974509669924144,
"grad_norm": 0.5723535664754966,
"learning_rate": 2.6437226997322812e-06,
"loss": 0.8877,
"mean_token_accuracy": 0.7747730650007725,
"num_tokens": 631978925.0,
"step": 7130
},
{
"entropy": 0.93076171875,
"epoch": 0.9758764436547529,
"grad_norm": 0.6260511883189127,
"learning_rate": 2.640200084542765e-06,
"loss": 0.9033,
"mean_token_accuracy": 0.7707104437053204,
"num_tokens": 632874005.0,
"step": 7140
},
{
"entropy": 0.921875,
"epoch": 0.9772432173853619,
"grad_norm": 0.6985366339867247,
"learning_rate": 2.6366774693532483e-06,
"loss": 0.9047,
"mean_token_accuracy": 0.7708546347916126,
"num_tokens": 633703539.0,
"step": 7150
},
{
"entropy": 0.94951171875,
"epoch": 0.9786099911159708,
"grad_norm": 0.8057752636226275,
"learning_rate": 2.6331548541637314e-06,
"loss": 0.9368,
"mean_token_accuracy": 0.7662607923150062,
"num_tokens": 634587023.0,
"step": 7160
},
{
"entropy": 0.902392578125,
"epoch": 0.9799767648465797,
"grad_norm": 0.5669847209461535,
"learning_rate": 2.6296322389742146e-06,
"loss": 0.8852,
"mean_token_accuracy": 0.7758343160152436,
"num_tokens": 635488376.0,
"step": 7170
},
{
"entropy": 0.888134765625,
"epoch": 0.9813435385771886,
"grad_norm": 0.599414633863949,
"learning_rate": 2.6261096237846977e-06,
"loss": 0.8698,
"mean_token_accuracy": 0.7761620223522187,
"num_tokens": 636356843.0,
"step": 7180
},
{
"entropy": 0.904638671875,
"epoch": 0.9827103123077975,
"grad_norm": 0.6197137230135106,
"learning_rate": 2.622587008595181e-06,
"loss": 0.8842,
"mean_token_accuracy": 0.7751274138689042,
"num_tokens": 637214936.0,
"step": 7190
},
{
"entropy": 0.9365234375,
"epoch": 0.9840770860384064,
"grad_norm": 0.6375355087311735,
"learning_rate": 2.6190643934056648e-06,
"loss": 0.917,
"mean_token_accuracy": 0.765006122738123,
"num_tokens": 638111430.0,
"step": 7200
},
{
"entropy": 0.939697265625,
"epoch": 0.9854438597690153,
"grad_norm": 0.5941620167201024,
"learning_rate": 2.615541778216148e-06,
"loss": 0.9439,
"mean_token_accuracy": 0.7640200741589069,
"num_tokens": 638975613.0,
"step": 7210
},
{
"entropy": 0.9154052734375,
"epoch": 0.9868106334996242,
"grad_norm": 0.6213429390351977,
"learning_rate": 2.612019163026631e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7735110871493817,
"num_tokens": 639844309.0,
"step": 7220
},
{
"entropy": 0.944873046875,
"epoch": 0.9881774072302331,
"grad_norm": 0.6858834078258915,
"learning_rate": 2.6084965478371146e-06,
"loss": 0.96,
"mean_token_accuracy": 0.7622631765902043,
"num_tokens": 640732612.0,
"step": 7230
},
{
"entropy": 0.875048828125,
"epoch": 0.989544180960842,
"grad_norm": 0.5768188273897995,
"learning_rate": 2.6049739326475977e-06,
"loss": 0.8797,
"mean_token_accuracy": 0.7790592141449452,
"num_tokens": 641622965.0,
"step": 7240
},
{
"entropy": 0.9289794921875,
"epoch": 0.9909109546914509,
"grad_norm": 0.6631489932612652,
"learning_rate": 2.6014513174580812e-06,
"loss": 0.9122,
"mean_token_accuracy": 0.770894569158554,
"num_tokens": 642493831.0,
"step": 7250
},
{
"entropy": 0.91201171875,
"epoch": 0.9922777284220597,
"grad_norm": 0.6140616415669532,
"learning_rate": 2.5979287022685644e-06,
"loss": 0.9051,
"mean_token_accuracy": 0.7721701040863991,
"num_tokens": 643358509.0,
"step": 7260
},
{
"entropy": 0.871435546875,
"epoch": 0.9936445021526686,
"grad_norm": 0.6783538189571667,
"learning_rate": 2.594406087079048e-06,
"loss": 0.848,
"mean_token_accuracy": 0.782578169554472,
"num_tokens": 644234904.0,
"step": 7270
},
{
"entropy": 0.94873046875,
"epoch": 0.9950112758832775,
"grad_norm": 0.6339996611049753,
"learning_rate": 2.590883471889531e-06,
"loss": 0.923,
"mean_token_accuracy": 0.7670978769659996,
"num_tokens": 645107938.0,
"step": 7280
},
{
"entropy": 0.911669921875,
"epoch": 0.9963780496138864,
"grad_norm": 0.6425222091057121,
"learning_rate": 2.587360856700014e-06,
"loss": 0.8983,
"mean_token_accuracy": 0.7717308752238751,
"num_tokens": 645973503.0,
"step": 7290
},
{
"entropy": 0.92802734375,
"epoch": 0.9977448233444953,
"grad_norm": 0.6435093309484586,
"learning_rate": 2.5838382415104973e-06,
"loss": 0.9091,
"mean_token_accuracy": 0.7697232447564601,
"num_tokens": 646849773.0,
"step": 7300
},
{
"entropy": 0.94306640625,
"epoch": 0.9991115970751042,
"grad_norm": 0.5986002130496275,
"learning_rate": 2.5803156263209813e-06,
"loss": 0.969,
"mean_token_accuracy": 0.7594001807272435,
"num_tokens": 647714153.0,
"step": 7310
},
{
"entropy": 0.9505037006578947,
"epoch": 1.0004100321191827,
"grad_norm": 0.6757621290373451,
"learning_rate": 2.5767930111314644e-06,
"loss": 0.9456,
"mean_token_accuracy": 0.7635381731547808,
"num_tokens": 648591763.0,
"step": 7320
},
{
"entropy": 0.92568359375,
"epoch": 1.0017768058497916,
"grad_norm": 0.63779891092924,
"learning_rate": 2.5732703959419475e-06,
"loss": 0.9017,
"mean_token_accuracy": 0.7718878209590911,
"num_tokens": 649446852.0,
"step": 7330
},
{
"entropy": 0.89638671875,
"epoch": 1.0031435795804005,
"grad_norm": 0.5417885524015591,
"learning_rate": 2.5697477807524306e-06,
"loss": 0.8891,
"mean_token_accuracy": 0.7746891051530838,
"num_tokens": 650395487.0,
"step": 7340
},
{
"entropy": 0.85625,
"epoch": 1.0045103533110094,
"grad_norm": 0.6581871245197791,
"learning_rate": 2.5662251655629138e-06,
"loss": 0.8346,
"mean_token_accuracy": 0.7831880636513233,
"num_tokens": 651283020.0,
"step": 7350
},
{
"entropy": 0.918115234375,
"epoch": 1.0058771270416182,
"grad_norm": 1.2124482020591747,
"learning_rate": 2.5627025503733977e-06,
"loss": 0.9117,
"mean_token_accuracy": 0.7679999142885208,
"num_tokens": 652168918.0,
"step": 7360
},
{
"entropy": 0.920068359375,
"epoch": 1.0072439007722271,
"grad_norm": 0.6527083173875885,
"learning_rate": 2.559179935183881e-06,
"loss": 0.8788,
"mean_token_accuracy": 0.7759767524898052,
"num_tokens": 653038499.0,
"step": 7370
},
{
"entropy": 0.9114501953125,
"epoch": 1.008610674502836,
"grad_norm": 0.7095859229441521,
"learning_rate": 2.555657319994364e-06,
"loss": 0.9076,
"mean_token_accuracy": 0.7719812601804733,
"num_tokens": 653925430.0,
"step": 7380
},
{
"entropy": 0.892529296875,
"epoch": 1.009977448233445,
"grad_norm": 0.5952874349514018,
"learning_rate": 2.552134704804847e-06,
"loss": 0.8771,
"mean_token_accuracy": 0.7753948047757149,
"num_tokens": 654833925.0,
"step": 7390
},
{
"entropy": 0.9262939453125,
"epoch": 1.0113442219640538,
"grad_norm": 0.6560899535376913,
"learning_rate": 2.5486120896153306e-06,
"loss": 0.9447,
"mean_token_accuracy": 0.7659755513072014,
"num_tokens": 655696986.0,
"step": 7400
},
{
"entropy": 0.890283203125,
"epoch": 1.0127109956946627,
"grad_norm": 0.677296356802976,
"learning_rate": 2.5450894744258138e-06,
"loss": 0.8673,
"mean_token_accuracy": 0.7791869185864926,
"num_tokens": 656585655.0,
"step": 7410
},
{
"entropy": 0.8530029296875,
"epoch": 1.0140777694252716,
"grad_norm": 0.6677179163813526,
"learning_rate": 2.5415668592362973e-06,
"loss": 0.818,
"mean_token_accuracy": 0.7881573684513569,
"num_tokens": 657442572.0,
"step": 7420
},
{
"entropy": 0.880712890625,
"epoch": 1.0154445431558805,
"grad_norm": 0.6107589972577132,
"learning_rate": 2.5380442440467804e-06,
"loss": 0.8832,
"mean_token_accuracy": 0.7787267602980137,
"num_tokens": 658344385.0,
"step": 7430
},
{
"entropy": 0.939306640625,
"epoch": 1.0168113168864894,
"grad_norm": 0.716977888445125,
"learning_rate": 2.534521628857264e-06,
"loss": 0.9353,
"mean_token_accuracy": 0.7647909060120582,
"num_tokens": 659214294.0,
"step": 7440
},
{
"entropy": 0.883642578125,
"epoch": 1.0181780906170983,
"grad_norm": 0.5751448507148568,
"learning_rate": 2.530999013667747e-06,
"loss": 0.8736,
"mean_token_accuracy": 0.7778988890349865,
"num_tokens": 660072515.0,
"step": 7450
},
{
"entropy": 0.9642822265625,
"epoch": 1.0195448643477072,
"grad_norm": 0.7585078127135412,
"learning_rate": 2.5274763984782302e-06,
"loss": 0.9498,
"mean_token_accuracy": 0.7625981524586678,
"num_tokens": 660959998.0,
"step": 7460
},
{
"entropy": 0.8624267578125,
"epoch": 1.020911638078316,
"grad_norm": 0.6384203772709837,
"learning_rate": 2.5239537832887138e-06,
"loss": 0.8442,
"mean_token_accuracy": 0.7848257854580879,
"num_tokens": 661875700.0,
"step": 7470
},
{
"entropy": 0.909716796875,
"epoch": 1.022278411808925,
"grad_norm": 0.6585286407145617,
"learning_rate": 2.5204311680991973e-06,
"loss": 0.8991,
"mean_token_accuracy": 0.771257969737053,
"num_tokens": 662774852.0,
"step": 7480
},
{
"entropy": 0.89619140625,
"epoch": 1.0236451855395339,
"grad_norm": 0.5685885939497795,
"learning_rate": 2.5169085529096804e-06,
"loss": 0.8524,
"mean_token_accuracy": 0.7827370822429657,
"num_tokens": 663594608.0,
"step": 7490
},
{
"entropy": 0.911083984375,
"epoch": 1.0250119592701428,
"grad_norm": 0.6855472086416394,
"learning_rate": 2.5133859377201636e-06,
"loss": 0.9208,
"mean_token_accuracy": 0.7682478643953801,
"num_tokens": 664461143.0,
"step": 7500
},
{
"entropy": 0.928759765625,
"epoch": 1.0263787330007517,
"grad_norm": 0.6267705335033117,
"learning_rate": 2.5098633225306467e-06,
"loss": 0.9378,
"mean_token_accuracy": 0.7648865349590779,
"num_tokens": 665355418.0,
"step": 7510
},
{
"entropy": 0.8812255859375,
"epoch": 1.0277455067313606,
"grad_norm": 0.6137919473088351,
"learning_rate": 2.50634070734113e-06,
"loss": 0.8855,
"mean_token_accuracy": 0.7753292836248875,
"num_tokens": 666259994.0,
"step": 7520
},
{
"entropy": 0.887255859375,
"epoch": 1.0291122804619695,
"grad_norm": 0.5913802729636496,
"learning_rate": 2.502818092151614e-06,
"loss": 0.8709,
"mean_token_accuracy": 0.7792847007513046,
"num_tokens": 667126480.0,
"step": 7530
},
{
"entropy": 0.878515625,
"epoch": 1.0304790541925783,
"grad_norm": 0.688488695724841,
"learning_rate": 2.499295476962097e-06,
"loss": 0.8462,
"mean_token_accuracy": 0.7805165648460388,
"num_tokens": 668015004.0,
"step": 7540
},
{
"entropy": 0.8981201171875,
"epoch": 1.0318458279231872,
"grad_norm": 0.598514661074622,
"learning_rate": 2.49577286177258e-06,
"loss": 0.8819,
"mean_token_accuracy": 0.7780529037117958,
"num_tokens": 668927065.0,
"step": 7550
},
{
"entropy": 0.88818359375,
"epoch": 1.0332126016537961,
"grad_norm": 0.6084087208801823,
"learning_rate": 2.492250246583063e-06,
"loss": 0.8847,
"mean_token_accuracy": 0.7767330013215542,
"num_tokens": 669799875.0,
"step": 7560
},
{
"entropy": 0.90478515625,
"epoch": 1.034579375384405,
"grad_norm": 0.6122234375258254,
"learning_rate": 2.4887276313935467e-06,
"loss": 0.8715,
"mean_token_accuracy": 0.7771121956408024,
"num_tokens": 670673474.0,
"step": 7570
},
{
"entropy": 0.885302734375,
"epoch": 1.035946149115014,
"grad_norm": 0.5815606375753577,
"learning_rate": 2.48520501620403e-06,
"loss": 0.8614,
"mean_token_accuracy": 0.7785942025482655,
"num_tokens": 671584047.0,
"step": 7580
},
{
"entropy": 0.88125,
"epoch": 1.0373129228456228,
"grad_norm": 0.5876914159346844,
"learning_rate": 2.4816824010145134e-06,
"loss": 0.8736,
"mean_token_accuracy": 0.7782637283205986,
"num_tokens": 672451040.0,
"step": 7590
},
{
"entropy": 0.9045654296875,
"epoch": 1.0386796965762317,
"grad_norm": 0.6726365049667543,
"learning_rate": 2.478159785824997e-06,
"loss": 0.8888,
"mean_token_accuracy": 0.7738964095711708,
"num_tokens": 673394919.0,
"step": 7600
},
{
"entropy": 0.89033203125,
"epoch": 1.0400464703068406,
"grad_norm": 0.6492852496299003,
"learning_rate": 2.47463717063548e-06,
"loss": 0.9107,
"mean_token_accuracy": 0.7721129879355431,
"num_tokens": 674297118.0,
"step": 7610
},
{
"entropy": 0.9233642578125,
"epoch": 1.0414132440374495,
"grad_norm": 0.6407352630062247,
"learning_rate": 2.4711145554459636e-06,
"loss": 0.9121,
"mean_token_accuracy": 0.7682387575507164,
"num_tokens": 675208936.0,
"step": 7620
},
{
"entropy": 0.913330078125,
"epoch": 1.0427800177680584,
"grad_norm": 0.6235843005603872,
"learning_rate": 2.4675919402564467e-06,
"loss": 0.914,
"mean_token_accuracy": 0.769565113633871,
"num_tokens": 676084432.0,
"step": 7630
},
{
"entropy": 0.9145263671875,
"epoch": 1.0441467914986673,
"grad_norm": 0.7032979530070943,
"learning_rate": 2.46406932506693e-06,
"loss": 0.9143,
"mean_token_accuracy": 0.7696544714272022,
"num_tokens": 677002994.0,
"step": 7640
},
{
"entropy": 0.8744873046875,
"epoch": 1.0455135652292762,
"grad_norm": 0.6805735231655696,
"learning_rate": 2.4605467098774134e-06,
"loss": 0.8714,
"mean_token_accuracy": 0.7791687659919262,
"num_tokens": 677933461.0,
"step": 7650
},
{
"entropy": 0.87919921875,
"epoch": 1.0468803389598853,
"grad_norm": 0.5922712443962107,
"learning_rate": 2.4570240946878965e-06,
"loss": 0.8757,
"mean_token_accuracy": 0.7762679263949395,
"num_tokens": 678803569.0,
"step": 7660
},
{
"entropy": 0.901904296875,
"epoch": 1.0482471126904942,
"grad_norm": 0.6499869931627323,
"learning_rate": 2.4535014794983796e-06,
"loss": 0.9021,
"mean_token_accuracy": 0.7746988721191883,
"num_tokens": 679680533.0,
"step": 7670
},
{
"entropy": 0.86240234375,
"epoch": 1.049613886421103,
"grad_norm": 0.6784166183252662,
"learning_rate": 2.449978864308863e-06,
"loss": 0.8514,
"mean_token_accuracy": 0.7823827266693115,
"num_tokens": 680523782.0,
"step": 7680
},
{
"entropy": 0.90068359375,
"epoch": 1.050980660151712,
"grad_norm": 0.6437144702026948,
"learning_rate": 2.4464562491193463e-06,
"loss": 0.8718,
"mean_token_accuracy": 0.7766845606267452,
"num_tokens": 681377055.0,
"step": 7690
},
{
"entropy": 0.90322265625,
"epoch": 1.0523474338823209,
"grad_norm": 0.6741831221453607,
"learning_rate": 2.4429336339298294e-06,
"loss": 0.9176,
"mean_token_accuracy": 0.7719850815832615,
"num_tokens": 682286031.0,
"step": 7700
},
{
"entropy": 0.8538818359375,
"epoch": 1.0537142076129298,
"grad_norm": 0.6353092238038135,
"learning_rate": 2.439411018740313e-06,
"loss": 0.8375,
"mean_token_accuracy": 0.7829391933977604,
"num_tokens": 683137335.0,
"step": 7710
},
{
"entropy": 0.8732177734375,
"epoch": 1.0550809813435387,
"grad_norm": 0.589673886100118,
"learning_rate": 2.435888403550796e-06,
"loss": 0.8444,
"mean_token_accuracy": 0.7833970680832862,
"num_tokens": 683993688.0,
"step": 7720
},
{
"entropy": 0.9086669921875,
"epoch": 1.0564477550741476,
"grad_norm": 0.6270307743737362,
"learning_rate": 2.4323657883612797e-06,
"loss": 0.9051,
"mean_token_accuracy": 0.7719081409275532,
"num_tokens": 684887017.0,
"step": 7730
},
{
"entropy": 0.918359375,
"epoch": 1.0578145288047565,
"grad_norm": 0.6249840130433536,
"learning_rate": 2.4288431731717628e-06,
"loss": 0.8969,
"mean_token_accuracy": 0.7729722797870636,
"num_tokens": 685740410.0,
"step": 7740
},
{
"entropy": 0.898779296875,
"epoch": 1.0591813025353654,
"grad_norm": 0.7275456211965842,
"learning_rate": 2.4253205579822463e-06,
"loss": 0.8771,
"mean_token_accuracy": 0.7759522691369056,
"num_tokens": 686594655.0,
"step": 7750
},
{
"entropy": 0.86611328125,
"epoch": 1.0605480762659742,
"grad_norm": 0.6289326658370727,
"learning_rate": 2.4217979427927294e-06,
"loss": 0.859,
"mean_token_accuracy": 0.7790747597813606,
"num_tokens": 687452138.0,
"step": 7760
},
{
"entropy": 0.857958984375,
"epoch": 1.0619148499965831,
"grad_norm": 0.6077636612571147,
"learning_rate": 2.418275327603213e-06,
"loss": 0.8587,
"mean_token_accuracy": 0.7801697321236134,
"num_tokens": 688323704.0,
"step": 7770
},
{
"entropy": 0.89140625,
"epoch": 1.063281623727192,
"grad_norm": 0.6852248517178815,
"learning_rate": 2.414752712413696e-06,
"loss": 0.881,
"mean_token_accuracy": 0.7739756129682064,
"num_tokens": 689208797.0,
"step": 7780
},
{
"entropy": 0.858544921875,
"epoch": 1.064648397457801,
"grad_norm": 0.6175620695398291,
"learning_rate": 2.4112300972241797e-06,
"loss": 0.825,
"mean_token_accuracy": 0.7875398695468903,
"num_tokens": 690082488.0,
"step": 7790
},
{
"entropy": 0.8882080078125,
"epoch": 1.0660151711884098,
"grad_norm": 0.6484123547902582,
"learning_rate": 2.4077074820346628e-06,
"loss": 0.8974,
"mean_token_accuracy": 0.7734080836176872,
"num_tokens": 690974253.0,
"step": 7800
},
{
"entropy": 0.9244140625,
"epoch": 1.0673819449190187,
"grad_norm": 0.6697028375425489,
"learning_rate": 2.404184866845146e-06,
"loss": 0.9201,
"mean_token_accuracy": 0.7668568588793278,
"num_tokens": 691884193.0,
"step": 7810
},
{
"entropy": 0.90634765625,
"epoch": 1.0687487186496276,
"grad_norm": 0.6441179801331183,
"learning_rate": 2.4006622516556295e-06,
"loss": 0.8915,
"mean_token_accuracy": 0.7735776349902153,
"num_tokens": 692774904.0,
"step": 7820
},
{
"entropy": 0.9092041015625,
"epoch": 1.0701154923802365,
"grad_norm": 0.6633320997111479,
"learning_rate": 2.3971396364661126e-06,
"loss": 0.9025,
"mean_token_accuracy": 0.7721963256597519,
"num_tokens": 693642265.0,
"step": 7830
},
{
"entropy": 0.904638671875,
"epoch": 1.0714822661108454,
"grad_norm": 0.6446558325133522,
"learning_rate": 2.393617021276596e-06,
"loss": 0.9132,
"mean_token_accuracy": 0.7709735915064811,
"num_tokens": 694520615.0,
"step": 7840
},
{
"entropy": 0.8833984375,
"epoch": 1.0728490398414543,
"grad_norm": 0.5975542350036984,
"learning_rate": 2.3900944060870793e-06,
"loss": 0.8727,
"mean_token_accuracy": 0.7791016139090061,
"num_tokens": 695379137.0,
"step": 7850
},
{
"entropy": 0.916015625,
"epoch": 1.0742158135720632,
"grad_norm": 0.6585920856521388,
"learning_rate": 2.3865717908975624e-06,
"loss": 0.8919,
"mean_token_accuracy": 0.7716780744493008,
"num_tokens": 696271994.0,
"step": 7860
},
{
"entropy": 0.90224609375,
"epoch": 1.075582587302672,
"grad_norm": 0.6597594567842892,
"learning_rate": 2.383049175708046e-06,
"loss": 0.8809,
"mean_token_accuracy": 0.7755672670900822,
"num_tokens": 697156934.0,
"step": 7870
},
{
"entropy": 0.874609375,
"epoch": 1.076949361033281,
"grad_norm": 0.7043204945536683,
"learning_rate": 2.379526560518529e-06,
"loss": 0.8611,
"mean_token_accuracy": 0.7808612525463104,
"num_tokens": 698027349.0,
"step": 7880
},
{
"entropy": 0.9260986328125,
"epoch": 1.0783161347638899,
"grad_norm": 0.6629613529320012,
"learning_rate": 2.376003945329012e-06,
"loss": 0.9119,
"mean_token_accuracy": 0.7672599844634533,
"num_tokens": 698930447.0,
"step": 7890
},
{
"entropy": 0.839208984375,
"epoch": 1.0796829084944988,
"grad_norm": 0.6552377623926978,
"learning_rate": 2.3724813301394957e-06,
"loss": 0.814,
"mean_token_accuracy": 0.788493923842907,
"num_tokens": 699791840.0,
"step": 7900
},
{
"entropy": 0.851708984375,
"epoch": 1.0810496822251077,
"grad_norm": 0.6382855525238956,
"learning_rate": 2.368958714949979e-06,
"loss": 0.8296,
"mean_token_accuracy": 0.7869559310376644,
"num_tokens": 700643155.0,
"step": 7910
},
{
"entropy": 0.8937255859375,
"epoch": 1.0824164559557166,
"grad_norm": 0.6193284929888466,
"learning_rate": 2.3654360997604624e-06,
"loss": 0.8725,
"mean_token_accuracy": 0.7736146353185177,
"num_tokens": 701513474.0,
"step": 7920
},
{
"entropy": 0.885107421875,
"epoch": 1.0837832296863255,
"grad_norm": 0.616164560693422,
"learning_rate": 2.3619134845709455e-06,
"loss": 0.8954,
"mean_token_accuracy": 0.7768044374883175,
"num_tokens": 702423369.0,
"step": 7930
},
{
"entropy": 0.915283203125,
"epoch": 1.0851500034169344,
"grad_norm": 0.5789316732983976,
"learning_rate": 2.358390869381429e-06,
"loss": 0.8784,
"mean_token_accuracy": 0.7759137384593486,
"num_tokens": 703297260.0,
"step": 7940
},
{
"entropy": 0.898828125,
"epoch": 1.0865167771475432,
"grad_norm": 1.416621264009509,
"learning_rate": 2.354868254191912e-06,
"loss": 0.9044,
"mean_token_accuracy": 0.7732366874814034,
"num_tokens": 704211864.0,
"step": 7950
},
{
"entropy": 0.9076904296875,
"epoch": 1.0878835508781521,
"grad_norm": 0.6317656432052376,
"learning_rate": 2.3513456390023957e-06,
"loss": 0.882,
"mean_token_accuracy": 0.7760845065116883,
"num_tokens": 705113336.0,
"step": 7960
},
{
"entropy": 0.921533203125,
"epoch": 1.089250324608761,
"grad_norm": 0.6290978459490035,
"learning_rate": 2.347823023812879e-06,
"loss": 0.9258,
"mean_token_accuracy": 0.7670378096401691,
"num_tokens": 705967220.0,
"step": 7970
},
{
"entropy": 0.92568359375,
"epoch": 1.09061709833937,
"grad_norm": 0.6199878822603744,
"learning_rate": 2.3443004086233624e-06,
"loss": 0.9046,
"mean_token_accuracy": 0.7698243573307991,
"num_tokens": 706841729.0,
"step": 7980
},
{
"entropy": 0.875048828125,
"epoch": 1.0919838720699788,
"grad_norm": 0.6652397558572815,
"learning_rate": 2.3407777934338455e-06,
"loss": 0.8659,
"mean_token_accuracy": 0.7810400113463402,
"num_tokens": 707742723.0,
"step": 7990
},
{
"entropy": 0.901025390625,
"epoch": 1.0933506458005877,
"grad_norm": 0.6422706073721475,
"learning_rate": 2.3372551782443286e-06,
"loss": 0.8566,
"mean_token_accuracy": 0.7799086675047875,
"num_tokens": 708621892.0,
"step": 8000
},
{
"entropy": 0.904931640625,
"epoch": 1.0947174195311966,
"grad_norm": 0.6104146184984741,
"learning_rate": 2.333732563054812e-06,
"loss": 0.8913,
"mean_token_accuracy": 0.775169862061739,
"num_tokens": 709529009.0,
"step": 8010
},
{
"entropy": 0.8920166015625,
"epoch": 1.0960841932618055,
"grad_norm": 0.6786116087205967,
"learning_rate": 2.3302099478652953e-06,
"loss": 0.8813,
"mean_token_accuracy": 0.7767438299953937,
"num_tokens": 710431822.0,
"step": 8020
},
{
"entropy": 0.910009765625,
"epoch": 1.0974509669924144,
"grad_norm": 0.6553361133909408,
"learning_rate": 2.3266873326757784e-06,
"loss": 0.8911,
"mean_token_accuracy": 0.7773710764944554,
"num_tokens": 711282131.0,
"step": 8030
},
{
"entropy": 0.966943359375,
"epoch": 1.0988177407230233,
"grad_norm": 0.6817539386253159,
"learning_rate": 2.323164717486262e-06,
"loss": 0.961,
"mean_token_accuracy": 0.7609073407948017,
"num_tokens": 712177392.0,
"step": 8040
},
{
"entropy": 0.8399169921875,
"epoch": 1.1001845144536322,
"grad_norm": 0.5855197563149663,
"learning_rate": 2.319642102296745e-06,
"loss": 0.8238,
"mean_token_accuracy": 0.7886503674089909,
"num_tokens": 713063521.0,
"step": 8050
},
{
"entropy": 0.8802734375,
"epoch": 1.101551288184241,
"grad_norm": 0.6447882212394849,
"learning_rate": 2.3161194871072287e-06,
"loss": 0.8787,
"mean_token_accuracy": 0.7753881633281707,
"num_tokens": 713932139.0,
"step": 8060
},
{
"entropy": 0.862939453125,
"epoch": 1.10291806191485,
"grad_norm": 0.6196129530607546,
"learning_rate": 2.3125968719177118e-06,
"loss": 0.8502,
"mean_token_accuracy": 0.781496525555849,
"num_tokens": 714800743.0,
"step": 8070
},
{
"entropy": 0.8967041015625,
"epoch": 1.1042848356454589,
"grad_norm": 0.6787912427681907,
"learning_rate": 2.3090742567281953e-06,
"loss": 0.9058,
"mean_token_accuracy": 0.7705842301249504,
"num_tokens": 715716767.0,
"step": 8080
},
{
"entropy": 0.914453125,
"epoch": 1.1056516093760678,
"grad_norm": 0.5837538061908562,
"learning_rate": 2.3055516415386785e-06,
"loss": 0.8719,
"mean_token_accuracy": 0.7783690758049489,
"num_tokens": 716586164.0,
"step": 8090
},
{
"entropy": 0.891943359375,
"epoch": 1.1070183831066767,
"grad_norm": 0.774945104731988,
"learning_rate": 2.302029026349162e-06,
"loss": 0.8967,
"mean_token_accuracy": 0.7749120138585568,
"num_tokens": 717454801.0,
"step": 8100
},
{
"entropy": 0.902783203125,
"epoch": 1.1083851568372856,
"grad_norm": 0.7291037553860992,
"learning_rate": 2.298506411159645e-06,
"loss": 0.8998,
"mean_token_accuracy": 0.7726473718881607,
"num_tokens": 718358774.0,
"step": 8110
},
{
"entropy": 0.906982421875,
"epoch": 1.1097519305678945,
"grad_norm": 0.6234459954042425,
"learning_rate": 2.2949837959701287e-06,
"loss": 0.8957,
"mean_token_accuracy": 0.7709045022726059,
"num_tokens": 719306865.0,
"step": 8120
},
{
"entropy": 0.89716796875,
"epoch": 1.1111187042985033,
"grad_norm": 0.6431255152104038,
"learning_rate": 2.291461180780612e-06,
"loss": 0.8862,
"mean_token_accuracy": 0.7743942230939865,
"num_tokens": 720218145.0,
"step": 8130
},
{
"entropy": 0.88505859375,
"epoch": 1.1124854780291122,
"grad_norm": 0.6482420353358689,
"learning_rate": 2.287938565591095e-06,
"loss": 0.8781,
"mean_token_accuracy": 0.7768988996744156,
"num_tokens": 721112536.0,
"step": 8140
},
{
"entropy": 0.856298828125,
"epoch": 1.1138522517597211,
"grad_norm": 0.5776730618026176,
"learning_rate": 2.2844159504015785e-06,
"loss": 0.8396,
"mean_token_accuracy": 0.7839043006300926,
"num_tokens": 722020751.0,
"step": 8150
},
{
"entropy": 0.8985595703125,
"epoch": 1.11521902549033,
"grad_norm": 0.6272387362429889,
"learning_rate": 2.2808933352120616e-06,
"loss": 0.8767,
"mean_token_accuracy": 0.7769060365855693,
"num_tokens": 722858329.0,
"step": 8160
},
{
"entropy": 0.8931640625,
"epoch": 1.116585799220939,
"grad_norm": 0.6124084520919972,
"learning_rate": 2.277370720022545e-06,
"loss": 0.8786,
"mean_token_accuracy": 0.774334105104208,
"num_tokens": 723733723.0,
"step": 8170
},
{
"entropy": 0.9068359375,
"epoch": 1.1179525729515478,
"grad_norm": 0.673327198826502,
"learning_rate": 2.2738481048330283e-06,
"loss": 0.88,
"mean_token_accuracy": 0.7736827217042446,
"num_tokens": 724643810.0,
"step": 8180
},
{
"entropy": 0.8830078125,
"epoch": 1.1193193466821567,
"grad_norm": 0.6457582205720874,
"learning_rate": 2.2703254896435114e-06,
"loss": 0.8758,
"mean_token_accuracy": 0.7769770972430706,
"num_tokens": 725562682.0,
"step": 8190
},
{
"entropy": 0.901806640625,
"epoch": 1.1206861204127656,
"grad_norm": 0.6119981021596242,
"learning_rate": 2.266802874453995e-06,
"loss": 0.9203,
"mean_token_accuracy": 0.7692897401750087,
"num_tokens": 726462837.0,
"step": 8200
},
{
"entropy": 0.8982421875,
"epoch": 1.1220528941433745,
"grad_norm": 0.6975611279508085,
"learning_rate": 2.263280259264478e-06,
"loss": 0.8885,
"mean_token_accuracy": 0.7743327416479587,
"num_tokens": 727331679.0,
"step": 8210
},
{
"entropy": 0.855224609375,
"epoch": 1.1234196678739834,
"grad_norm": 0.6392929906023048,
"learning_rate": 2.259757644074961e-06,
"loss": 0.813,
"mean_token_accuracy": 0.7876626826822758,
"num_tokens": 728213250.0,
"step": 8220
},
{
"entropy": 0.92099609375,
"epoch": 1.1247864416045923,
"grad_norm": 0.6744703336002006,
"learning_rate": 2.2562350288854447e-06,
"loss": 0.9002,
"mean_token_accuracy": 0.7720523461699486,
"num_tokens": 729111281.0,
"step": 8230
},
{
"entropy": 0.868359375,
"epoch": 1.1261532153352012,
"grad_norm": 0.6601734967724888,
"learning_rate": 2.252712413695928e-06,
"loss": 0.8408,
"mean_token_accuracy": 0.7828718081116677,
"num_tokens": 729959862.0,
"step": 8240
},
{
"entropy": 0.9015625,
"epoch": 1.12751998906581,
"grad_norm": 0.6379127866596914,
"learning_rate": 2.2491897985064114e-06,
"loss": 0.8976,
"mean_token_accuracy": 0.7719614543020725,
"num_tokens": 730854597.0,
"step": 8250
},
{
"entropy": 0.904443359375,
"epoch": 1.128886762796419,
"grad_norm": 0.7054775816033368,
"learning_rate": 2.2456671833168945e-06,
"loss": 0.8732,
"mean_token_accuracy": 0.7783226862549781,
"num_tokens": 731702559.0,
"step": 8260
},
{
"entropy": 0.8921142578125,
"epoch": 1.1302535365270279,
"grad_norm": 0.6675166136081409,
"learning_rate": 2.242144568127378e-06,
"loss": 0.8841,
"mean_token_accuracy": 0.7759886823594571,
"num_tokens": 732555774.0,
"step": 8270
},
{
"entropy": 0.915283203125,
"epoch": 1.1316203102576368,
"grad_norm": 0.5984195773118474,
"learning_rate": 2.238621952937861e-06,
"loss": 0.9175,
"mean_token_accuracy": 0.7696954436600209,
"num_tokens": 733472536.0,
"step": 8280
},
{
"entropy": 0.920703125,
"epoch": 1.1329870839882457,
"grad_norm": 0.6824772830027784,
"learning_rate": 2.2350993377483447e-06,
"loss": 0.9149,
"mean_token_accuracy": 0.7669693656265736,
"num_tokens": 734343325.0,
"step": 8290
},
{
"entropy": 0.8926513671875,
"epoch": 1.1343538577188546,
"grad_norm": 0.6387264177099558,
"learning_rate": 2.231576722558828e-06,
"loss": 0.8651,
"mean_token_accuracy": 0.777832293510437,
"num_tokens": 735209167.0,
"step": 8300
},
{
"entropy": 0.8662353515625,
"epoch": 1.1357206314494634,
"grad_norm": 0.6613039418992037,
"learning_rate": 2.2280541073693114e-06,
"loss": 0.8524,
"mean_token_accuracy": 0.7827970243990421,
"num_tokens": 736102499.0,
"step": 8310
},
{
"entropy": 0.9318359375,
"epoch": 1.1370874051800723,
"grad_norm": 0.6739714435340602,
"learning_rate": 2.2245314921797945e-06,
"loss": 0.9175,
"mean_token_accuracy": 0.7691331170499325,
"num_tokens": 736993752.0,
"step": 8320
},
{
"entropy": 0.903515625,
"epoch": 1.1384541789106812,
"grad_norm": 0.6224969722538038,
"learning_rate": 2.2210088769902777e-06,
"loss": 0.8963,
"mean_token_accuracy": 0.7728459484875202,
"num_tokens": 737905549.0,
"step": 8330
},
{
"entropy": 0.836083984375,
"epoch": 1.1398209526412901,
"grad_norm": 0.6686508892509263,
"learning_rate": 2.217486261800761e-06,
"loss": 0.8115,
"mean_token_accuracy": 0.7883894488215446,
"num_tokens": 738743099.0,
"step": 8340
},
{
"entropy": 0.8956298828125,
"epoch": 1.141187726371899,
"grad_norm": 0.6289430268275589,
"learning_rate": 2.2139636466112443e-06,
"loss": 0.885,
"mean_token_accuracy": 0.7765215002000332,
"num_tokens": 739666732.0,
"step": 8350
},
{
"entropy": 0.8606201171875,
"epoch": 1.142554500102508,
"grad_norm": 0.6642512089201957,
"learning_rate": 2.2104410314217275e-06,
"loss": 0.8486,
"mean_token_accuracy": 0.7807374276220799,
"num_tokens": 740545036.0,
"step": 8360
},
{
"entropy": 0.9107666015625,
"epoch": 1.1439212738331168,
"grad_norm": 0.6840047057410128,
"learning_rate": 2.206918416232211e-06,
"loss": 0.9062,
"mean_token_accuracy": 0.7726617507636547,
"num_tokens": 741452530.0,
"step": 8370
},
{
"entropy": 0.9131591796875,
"epoch": 1.1452880475637257,
"grad_norm": 0.5915713801932108,
"learning_rate": 2.203395801042694e-06,
"loss": 0.9142,
"mean_token_accuracy": 0.7661910966038704,
"num_tokens": 742353736.0,
"step": 8380
},
{
"entropy": 0.893310546875,
"epoch": 1.1466548212943346,
"grad_norm": 0.7091780608862464,
"learning_rate": 2.1998731858531777e-06,
"loss": 0.9003,
"mean_token_accuracy": 0.7742526255548,
"num_tokens": 743291796.0,
"step": 8390
},
{
"entropy": 0.896875,
"epoch": 1.1480215950249435,
"grad_norm": 0.6994489284531417,
"learning_rate": 2.196350570663661e-06,
"loss": 0.8957,
"mean_token_accuracy": 0.7758879758417606,
"num_tokens": 744191576.0,
"step": 8400
},
{
"entropy": 0.908056640625,
"epoch": 1.1493883687555526,
"grad_norm": 0.6697906754252302,
"learning_rate": 2.192827955474144e-06,
"loss": 0.88,
"mean_token_accuracy": 0.7756712570786476,
"num_tokens": 745053472.0,
"step": 8410
},
{
"entropy": 0.879833984375,
"epoch": 1.1507551424861615,
"grad_norm": 0.5840120312430608,
"learning_rate": 2.1893053402846275e-06,
"loss": 0.847,
"mean_token_accuracy": 0.7822894498705864,
"num_tokens": 745945754.0,
"step": 8420
},
{
"entropy": 0.9228515625,
"epoch": 1.1521219162167704,
"grad_norm": 0.6249995493645794,
"learning_rate": 2.1857827250951106e-06,
"loss": 0.909,
"mean_token_accuracy": 0.7719705693423748,
"num_tokens": 746861785.0,
"step": 8430
},
{
"entropy": 0.933154296875,
"epoch": 1.1534886899473793,
"grad_norm": 0.655511624580773,
"learning_rate": 2.182260109905594e-06,
"loss": 0.9219,
"mean_token_accuracy": 0.7690572202205658,
"num_tokens": 747742441.0,
"step": 8440
},
{
"entropy": 0.878955078125,
"epoch": 1.1548554636779882,
"grad_norm": 0.6438495618402751,
"learning_rate": 2.1787374947160777e-06,
"loss": 0.8546,
"mean_token_accuracy": 0.780893737822771,
"num_tokens": 748609606.0,
"step": 8450
},
{
"entropy": 0.864794921875,
"epoch": 1.156222237408597,
"grad_norm": 0.5990932978131096,
"learning_rate": 2.175214879526561e-06,
"loss": 0.8702,
"mean_token_accuracy": 0.7788433104753494,
"num_tokens": 749527900.0,
"step": 8460
},
{
"entropy": 0.916162109375,
"epoch": 1.157589011139206,
"grad_norm": 0.6248547754741081,
"learning_rate": 2.171692264337044e-06,
"loss": 0.898,
"mean_token_accuracy": 0.7696771867573261,
"num_tokens": 750396477.0,
"step": 8470
},
{
"entropy": 0.886279296875,
"epoch": 1.1589557848698149,
"grad_norm": 0.6515982610637115,
"learning_rate": 2.1681696491475275e-06,
"loss": 0.8634,
"mean_token_accuracy": 0.7764886133372784,
"num_tokens": 751256327.0,
"step": 8480
},
{
"entropy": 0.912060546875,
"epoch": 1.1603225586004238,
"grad_norm": 0.6137383539399678,
"learning_rate": 2.1646470339580106e-06,
"loss": 0.8822,
"mean_token_accuracy": 0.7747414082288742,
"num_tokens": 752202319.0,
"step": 8490
},
{
"entropy": 0.8611572265625,
"epoch": 1.1616893323310327,
"grad_norm": 0.6369354889578388,
"learning_rate": 2.1611244187684937e-06,
"loss": 0.8409,
"mean_token_accuracy": 0.7860894352197647,
"num_tokens": 753076870.0,
"step": 8500
},
{
"entropy": 0.894384765625,
"epoch": 1.1630561060616416,
"grad_norm": 0.6073302399054902,
"learning_rate": 2.1576018035789773e-06,
"loss": 0.8771,
"mean_token_accuracy": 0.7755731858313084,
"num_tokens": 753992426.0,
"step": 8510
},
{
"entropy": 0.8780029296875,
"epoch": 1.1644228797922505,
"grad_norm": 0.6605533013264712,
"learning_rate": 2.1540791883894604e-06,
"loss": 0.8506,
"mean_token_accuracy": 0.779113968461752,
"num_tokens": 754824603.0,
"step": 8520
},
{
"entropy": 0.90634765625,
"epoch": 1.1657896535228593,
"grad_norm": 0.6365414602042601,
"learning_rate": 2.150556573199944e-06,
"loss": 0.904,
"mean_token_accuracy": 0.7746730744838715,
"num_tokens": 755695653.0,
"step": 8530
},
{
"entropy": 0.88076171875,
"epoch": 1.1671564272534682,
"grad_norm": 0.6093977599317174,
"learning_rate": 2.147033958010427e-06,
"loss": 0.8744,
"mean_token_accuracy": 0.7766316190361977,
"num_tokens": 756583392.0,
"step": 8540
},
{
"entropy": 0.9140625,
"epoch": 1.1685232009840771,
"grad_norm": 0.7188150851167744,
"learning_rate": 2.14351134282091e-06,
"loss": 0.8985,
"mean_token_accuracy": 0.771044161170721,
"num_tokens": 757484772.0,
"step": 8550
},
{
"entropy": 0.8487548828125,
"epoch": 1.169889974714686,
"grad_norm": 0.6524639281729816,
"learning_rate": 2.1399887276313937e-06,
"loss": 0.8279,
"mean_token_accuracy": 0.7840420648455619,
"num_tokens": 758398852.0,
"step": 8560
},
{
"entropy": 0.9099609375,
"epoch": 1.171256748445295,
"grad_norm": 0.5954534104964879,
"learning_rate": 2.136466112441877e-06,
"loss": 0.9146,
"mean_token_accuracy": 0.7700278349220753,
"num_tokens": 759304325.0,
"step": 8570
},
{
"entropy": 0.87880859375,
"epoch": 1.1726235221759038,
"grad_norm": 0.6636239811933494,
"learning_rate": 2.1329434972523604e-06,
"loss": 0.8684,
"mean_token_accuracy": 0.7789595000445843,
"num_tokens": 760182897.0,
"step": 8580
},
{
"entropy": 0.904443359375,
"epoch": 1.1739902959065127,
"grad_norm": 0.6292147513528402,
"learning_rate": 2.1294208820628435e-06,
"loss": 0.873,
"mean_token_accuracy": 0.7757314458489418,
"num_tokens": 761046394.0,
"step": 8590
},
{
"entropy": 0.91669921875,
"epoch": 1.1753570696371216,
"grad_norm": 0.6803356786315845,
"learning_rate": 2.125898266873327e-06,
"loss": 0.9209,
"mean_token_accuracy": 0.7665342092514038,
"num_tokens": 761983231.0,
"step": 8600
},
{
"entropy": 0.90322265625,
"epoch": 1.1767238433677305,
"grad_norm": 0.614154063541608,
"learning_rate": 2.12237565168381e-06,
"loss": 0.8931,
"mean_token_accuracy": 0.7717494130134582,
"num_tokens": 762882518.0,
"step": 8610
},
{
"entropy": 0.895556640625,
"epoch": 1.1780906170983394,
"grad_norm": 0.6631372098320691,
"learning_rate": 2.1188530364942938e-06,
"loss": 0.8915,
"mean_token_accuracy": 0.7736919656395912,
"num_tokens": 763770291.0,
"step": 8620
},
{
"entropy": 0.846728515625,
"epoch": 1.1794573908289483,
"grad_norm": 0.5742270093405668,
"learning_rate": 2.115330421304777e-06,
"loss": 0.8235,
"mean_token_accuracy": 0.7864968098700047,
"num_tokens": 764654419.0,
"step": 8630
},
{
"entropy": 0.8765625,
"epoch": 1.1808241645595572,
"grad_norm": 0.6027435634334174,
"learning_rate": 2.1118078061152604e-06,
"loss": 0.8761,
"mean_token_accuracy": 0.7798977442085743,
"num_tokens": 765587578.0,
"step": 8640
},
{
"entropy": 0.863330078125,
"epoch": 1.182190938290166,
"grad_norm": 0.6907382322401672,
"learning_rate": 2.1082851909257435e-06,
"loss": 0.8512,
"mean_token_accuracy": 0.7832790940999985,
"num_tokens": 766486320.0,
"step": 8650
},
{
"entropy": 0.897900390625,
"epoch": 1.183557712020775,
"grad_norm": 0.6254518153234523,
"learning_rate": 2.1047625757362267e-06,
"loss": 0.8697,
"mean_token_accuracy": 0.7778702549636364,
"num_tokens": 767398093.0,
"step": 8660
},
{
"entropy": 0.8896484375,
"epoch": 1.1849244857513839,
"grad_norm": 0.6513017354484407,
"learning_rate": 2.1012399605467102e-06,
"loss": 0.8861,
"mean_token_accuracy": 0.7761388055980205,
"num_tokens": 768249145.0,
"step": 8670
},
{
"entropy": 0.898681640625,
"epoch": 1.1862912594819928,
"grad_norm": 0.6137282253529995,
"learning_rate": 2.0977173453571933e-06,
"loss": 0.8959,
"mean_token_accuracy": 0.7752281829714776,
"num_tokens": 769095450.0,
"step": 8680
},
{
"entropy": 0.90478515625,
"epoch": 1.1876580332126017,
"grad_norm": 0.6859600011733008,
"learning_rate": 2.0941947301676765e-06,
"loss": 0.9208,
"mean_token_accuracy": 0.7701627373695373,
"num_tokens": 769933084.0,
"step": 8690
},
{
"entropy": 0.8863525390625,
"epoch": 1.1890248069432106,
"grad_norm": 0.654791125416128,
"learning_rate": 2.09067211497816e-06,
"loss": 0.855,
"mean_token_accuracy": 0.7772803239524364,
"num_tokens": 770828053.0,
"step": 8700
},
{
"entropy": 0.910009765625,
"epoch": 1.1903915806738194,
"grad_norm": 0.6456966098462408,
"learning_rate": 2.087149499788643e-06,
"loss": 0.8962,
"mean_token_accuracy": 0.7714630216360092,
"num_tokens": 771702147.0,
"step": 8710
},
{
"entropy": 0.9068359375,
"epoch": 1.1917583544044283,
"grad_norm": 0.7460549715688074,
"learning_rate": 2.0836268845991263e-06,
"loss": 0.9,
"mean_token_accuracy": 0.7729074157774448,
"num_tokens": 772586595.0,
"step": 8720
},
{
"entropy": 0.90693359375,
"epoch": 1.1931251281350372,
"grad_norm": 0.6565883984219623,
"learning_rate": 2.08010426940961e-06,
"loss": 0.8972,
"mean_token_accuracy": 0.7751238532364368,
"num_tokens": 773492144.0,
"step": 8730
},
{
"entropy": 0.907958984375,
"epoch": 1.1944919018656461,
"grad_norm": 0.6893792675527011,
"learning_rate": 2.076581654220093e-06,
"loss": 0.9083,
"mean_token_accuracy": 0.7721623227000236,
"num_tokens": 774384486.0,
"step": 8740
},
{
"entropy": 0.85498046875,
"epoch": 1.195858675596255,
"grad_norm": 0.8136742077154877,
"learning_rate": 2.0730590390305765e-06,
"loss": 0.8484,
"mean_token_accuracy": 0.7826823115348815,
"num_tokens": 775257458.0,
"step": 8750
},
{
"entropy": 0.8934814453125,
"epoch": 1.197225449326864,
"grad_norm": 0.688227161984929,
"learning_rate": 2.0695364238410596e-06,
"loss": 0.8819,
"mean_token_accuracy": 0.7727421529591083,
"num_tokens": 776188791.0,
"step": 8760
},
{
"entropy": 0.9015625,
"epoch": 1.1985922230574728,
"grad_norm": 0.5982211537313886,
"learning_rate": 2.066013808651543e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7736355490982533,
"num_tokens": 777067934.0,
"step": 8770
},
{
"entropy": 0.902001953125,
"epoch": 1.1999589967880817,
"grad_norm": 0.7200257769283286,
"learning_rate": 2.0624911934620263e-06,
"loss": 0.9073,
"mean_token_accuracy": 0.7703949965536594,
"num_tokens": 777918469.0,
"step": 8780
},
{
"entropy": 0.911572265625,
"epoch": 1.2013257705186906,
"grad_norm": 0.6158555304478843,
"learning_rate": 2.05896857827251e-06,
"loss": 0.9132,
"mean_token_accuracy": 0.770005152374506,
"num_tokens": 778840876.0,
"step": 8790
},
{
"entropy": 0.8742431640625,
"epoch": 1.2026925442492995,
"grad_norm": 0.640854377692924,
"learning_rate": 2.055445963082993e-06,
"loss": 0.839,
"mean_token_accuracy": 0.7845941469073295,
"num_tokens": 779722325.0,
"step": 8800
},
{
"entropy": 0.85908203125,
"epoch": 1.2040593179799084,
"grad_norm": 0.630968482597706,
"learning_rate": 2.0519233478934765e-06,
"loss": 0.8754,
"mean_token_accuracy": 0.7794738031923771,
"num_tokens": 780628547.0,
"step": 8810
},
{
"entropy": 0.9075927734375,
"epoch": 1.2054260917105173,
"grad_norm": 0.6368965057758514,
"learning_rate": 2.0484007327039596e-06,
"loss": 0.8853,
"mean_token_accuracy": 0.7754197388887405,
"num_tokens": 781503881.0,
"step": 8820
},
{
"entropy": 0.908203125,
"epoch": 1.2067928654411262,
"grad_norm": 0.6495116100490108,
"learning_rate": 2.0448781175144427e-06,
"loss": 0.8973,
"mean_token_accuracy": 0.7720845669507981,
"num_tokens": 782375550.0,
"step": 8830
},
{
"entropy": 0.873193359375,
"epoch": 1.208159639171735,
"grad_norm": 0.7421467296296794,
"learning_rate": 2.0413555023249263e-06,
"loss": 0.8537,
"mean_token_accuracy": 0.7824203111231327,
"num_tokens": 783264796.0,
"step": 8840
},
{
"entropy": 0.888671875,
"epoch": 1.209526412902344,
"grad_norm": 0.6522868360294027,
"learning_rate": 2.0378328871354094e-06,
"loss": 0.8728,
"mean_token_accuracy": 0.7773667141795159,
"num_tokens": 784182648.0,
"step": 8850
},
{
"entropy": 0.8849365234375,
"epoch": 1.2108931866329529,
"grad_norm": 0.6048037289511006,
"learning_rate": 2.034310271945893e-06,
"loss": 0.8751,
"mean_token_accuracy": 0.7790608935058116,
"num_tokens": 785110665.0,
"step": 8860
},
{
"entropy": 0.88427734375,
"epoch": 1.2122599603635618,
"grad_norm": 0.6777402433075365,
"learning_rate": 2.030787656756376e-06,
"loss": 0.8734,
"mean_token_accuracy": 0.7758014447987079,
"num_tokens": 785985140.0,
"step": 8870
},
{
"entropy": 0.87236328125,
"epoch": 1.2136267340941707,
"grad_norm": 0.589395842680932,
"learning_rate": 2.027265041566859e-06,
"loss": 0.878,
"mean_token_accuracy": 0.7801454551517963,
"num_tokens": 786868691.0,
"step": 8880
},
{
"entropy": 0.880322265625,
"epoch": 1.2149935078247796,
"grad_norm": 0.6782478000094176,
"learning_rate": 2.0237424263773427e-06,
"loss": 0.8468,
"mean_token_accuracy": 0.7818550907075406,
"num_tokens": 787720812.0,
"step": 8890
},
{
"entropy": 0.864453125,
"epoch": 1.2163602815553884,
"grad_norm": 0.5742068729487767,
"learning_rate": 2.020219811187826e-06,
"loss": 0.839,
"mean_token_accuracy": 0.7836993567645549,
"num_tokens": 788632265.0,
"step": 8900
},
{
"entropy": 0.9001953125,
"epoch": 1.2177270552859973,
"grad_norm": 0.5990955439449626,
"learning_rate": 2.0166971959983094e-06,
"loss": 0.8927,
"mean_token_accuracy": 0.7737382337450981,
"num_tokens": 789527672.0,
"step": 8910
},
{
"entropy": 0.87333984375,
"epoch": 1.2190938290166062,
"grad_norm": 0.5761093129133186,
"learning_rate": 2.0131745808087925e-06,
"loss": 0.8519,
"mean_token_accuracy": 0.7829400211572647,
"num_tokens": 790430285.0,
"step": 8920
},
{
"entropy": 0.874658203125,
"epoch": 1.2204606027472151,
"grad_norm": 0.6562270589366388,
"learning_rate": 2.009651965619276e-06,
"loss": 0.8626,
"mean_token_accuracy": 0.779813601821661,
"num_tokens": 791320202.0,
"step": 8930
},
{
"entropy": 0.9099609375,
"epoch": 1.221827376477824,
"grad_norm": 0.6188422269322296,
"learning_rate": 2.0061293504297592e-06,
"loss": 0.9117,
"mean_token_accuracy": 0.771654486656189,
"num_tokens": 792228711.0,
"step": 8940
},
{
"entropy": 0.8564697265625,
"epoch": 1.223194150208433,
"grad_norm": 0.6077032382660479,
"learning_rate": 2.0026067352402428e-06,
"loss": 0.8234,
"mean_token_accuracy": 0.7869372144341469,
"num_tokens": 793111439.0,
"step": 8950
},
{
"entropy": 0.88056640625,
"epoch": 1.2245609239390418,
"grad_norm": 0.5699496805836997,
"learning_rate": 1.999084120050726e-06,
"loss": 0.8977,
"mean_token_accuracy": 0.7743222512304783,
"num_tokens": 794001756.0,
"step": 8960
},
{
"entropy": 0.871044921875,
"epoch": 1.2259276976696507,
"grad_norm": 0.5905840818518614,
"learning_rate": 1.9955615048612094e-06,
"loss": 0.8512,
"mean_token_accuracy": 0.7808790810406208,
"num_tokens": 794884233.0,
"step": 8970
},
{
"entropy": 0.872607421875,
"epoch": 1.2272944714002596,
"grad_norm": 0.6656990368421166,
"learning_rate": 1.9920388896716926e-06,
"loss": 0.8717,
"mean_token_accuracy": 0.7782989010214806,
"num_tokens": 795772500.0,
"step": 8980
},
{
"entropy": 0.893115234375,
"epoch": 1.2286612451308685,
"grad_norm": 0.5868094458168432,
"learning_rate": 1.9885162744821757e-06,
"loss": 0.8621,
"mean_token_accuracy": 0.7825874857604503,
"num_tokens": 796668379.0,
"step": 8990
},
{
"entropy": 0.8959228515625,
"epoch": 1.2300280188614774,
"grad_norm": 0.8281890288175602,
"learning_rate": 1.9849936592926592e-06,
"loss": 0.8454,
"mean_token_accuracy": 0.7815780282020569,
"num_tokens": 797490645.0,
"step": 9000
},
{
"entropy": 0.90263671875,
"epoch": 1.2313947925920863,
"grad_norm": 0.6139551908186021,
"learning_rate": 1.9814710441031424e-06,
"loss": 0.859,
"mean_token_accuracy": 0.777613727003336,
"num_tokens": 798370052.0,
"step": 9010
},
{
"entropy": 0.8748779296875,
"epoch": 1.2327615663226954,
"grad_norm": 0.6565176085245547,
"learning_rate": 1.9779484289136255e-06,
"loss": 0.8564,
"mean_token_accuracy": 0.7794780880212784,
"num_tokens": 799229778.0,
"step": 9020
},
{
"entropy": 0.9048828125,
"epoch": 1.2341283400533043,
"grad_norm": 0.624390599257827,
"learning_rate": 1.974425813724109e-06,
"loss": 0.8885,
"mean_token_accuracy": 0.7745428755879402,
"num_tokens": 800125708.0,
"step": 9030
},
{
"entropy": 0.8734130859375,
"epoch": 1.2354951137839132,
"grad_norm": 0.5840701027358182,
"learning_rate": 1.970903198534592e-06,
"loss": 0.8379,
"mean_token_accuracy": 0.7848003648221493,
"num_tokens": 801028037.0,
"step": 9040
},
{
"entropy": 0.8798828125,
"epoch": 1.236861887514522,
"grad_norm": 0.6971237083402785,
"learning_rate": 1.9673805833450753e-06,
"loss": 0.8816,
"mean_token_accuracy": 0.7738574564456939,
"num_tokens": 801926801.0,
"step": 9050
},
{
"entropy": 0.921435546875,
"epoch": 1.238228661245131,
"grad_norm": 0.5328979691697455,
"learning_rate": 1.963857968155559e-06,
"loss": 0.9131,
"mean_token_accuracy": 0.7696879908442498,
"num_tokens": 802832899.0,
"step": 9060
},
{
"entropy": 0.89794921875,
"epoch": 1.2395954349757399,
"grad_norm": 0.7079385779985577,
"learning_rate": 1.960335352966042e-06,
"loss": 0.866,
"mean_token_accuracy": 0.7784696780145168,
"num_tokens": 803717792.0,
"step": 9070
},
{
"entropy": 0.8943359375,
"epoch": 1.2409622087063488,
"grad_norm": 0.6153689896419035,
"learning_rate": 1.9568127377765255e-06,
"loss": 0.887,
"mean_token_accuracy": 0.770531153678894,
"num_tokens": 804630519.0,
"step": 9080
},
{
"entropy": 0.866015625,
"epoch": 1.2423289824369577,
"grad_norm": 0.5957281486905083,
"learning_rate": 1.9532901225870086e-06,
"loss": 0.8466,
"mean_token_accuracy": 0.7828671276569367,
"num_tokens": 805565081.0,
"step": 9090
},
{
"entropy": 0.8927734375,
"epoch": 1.2436957561675666,
"grad_norm": 0.6559395639274657,
"learning_rate": 1.949767507397492e-06,
"loss": 0.8751,
"mean_token_accuracy": 0.7766449302434921,
"num_tokens": 806425066.0,
"step": 9100
},
{
"entropy": 0.9033203125,
"epoch": 1.2450625298981755,
"grad_norm": 0.6189874930106518,
"learning_rate": 1.9462448922079753e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7733691856265068,
"num_tokens": 807299718.0,
"step": 9110
},
{
"entropy": 0.893359375,
"epoch": 1.2464293036287843,
"grad_norm": 0.6971128617403005,
"learning_rate": 1.942722277018459e-06,
"loss": 0.8749,
"mean_token_accuracy": 0.7768423147499561,
"num_tokens": 808167887.0,
"step": 9120
},
{
"entropy": 0.92197265625,
"epoch": 1.2477960773593932,
"grad_norm": 0.6840582308820735,
"learning_rate": 1.939199661828942e-06,
"loss": 0.9236,
"mean_token_accuracy": 0.7686198830604554,
"num_tokens": 809042851.0,
"step": 9130
},
{
"entropy": 0.862109375,
"epoch": 1.2491628510900021,
"grad_norm": 0.647500289998666,
"learning_rate": 1.9356770466394255e-06,
"loss": 0.8442,
"mean_token_accuracy": 0.783176413923502,
"num_tokens": 809871246.0,
"step": 9140
},
{
"entropy": 0.856689453125,
"epoch": 1.250529624820611,
"grad_norm": 0.6098673911092052,
"learning_rate": 1.9321544314499086e-06,
"loss": 0.8454,
"mean_token_accuracy": 0.7823197074234486,
"num_tokens": 810754067.0,
"step": 9150
},
{
"entropy": 0.8813720703125,
"epoch": 1.25189639855122,
"grad_norm": 0.7057101859280207,
"learning_rate": 1.9286318162603917e-06,
"loss": 0.8789,
"mean_token_accuracy": 0.7762944452464581,
"num_tokens": 811652395.0,
"step": 9160
},
{
"entropy": 0.8490966796875,
"epoch": 1.2532631722818288,
"grad_norm": 0.6098215127135332,
"learning_rate": 1.9251092010708753e-06,
"loss": 0.8408,
"mean_token_accuracy": 0.786857920140028,
"num_tokens": 812537595.0,
"step": 9170
},
{
"entropy": 0.8939697265625,
"epoch": 1.2546299460124377,
"grad_norm": 0.5644040585480365,
"learning_rate": 1.9215865858813584e-06,
"loss": 0.8959,
"mean_token_accuracy": 0.7753555335104465,
"num_tokens": 813418052.0,
"step": 9180
},
{
"entropy": 0.9033203125,
"epoch": 1.2559967197430466,
"grad_norm": 0.5645520426434595,
"learning_rate": 1.918063970691842e-06,
"loss": 0.894,
"mean_token_accuracy": 0.7721271120011807,
"num_tokens": 814316613.0,
"step": 9190
},
{
"entropy": 0.91259765625,
"epoch": 1.2573634934736555,
"grad_norm": 0.6428827588278848,
"learning_rate": 1.914541355502325e-06,
"loss": 0.8805,
"mean_token_accuracy": 0.7716483265161515,
"num_tokens": 815160785.0,
"step": 9200
},
{
"entropy": 0.891162109375,
"epoch": 1.2587302672042644,
"grad_norm": 0.5884243058283878,
"learning_rate": 1.9110187403128082e-06,
"loss": 0.86,
"mean_token_accuracy": 0.778124725073576,
"num_tokens": 816036366.0,
"step": 9210
},
{
"entropy": 0.8760498046875,
"epoch": 1.2600970409348733,
"grad_norm": 0.6023933238281334,
"learning_rate": 1.9074961251232918e-06,
"loss": 0.865,
"mean_token_accuracy": 0.7798771880567074,
"num_tokens": 816965958.0,
"step": 9220
},
{
"entropy": 0.864990234375,
"epoch": 1.2614638146654822,
"grad_norm": 0.6375320986819911,
"learning_rate": 1.903973509933775e-06,
"loss": 0.8517,
"mean_token_accuracy": 0.7786759153008461,
"num_tokens": 817830932.0,
"step": 9230
},
{
"entropy": 0.92138671875,
"epoch": 1.262830588396091,
"grad_norm": 0.6605296208134678,
"learning_rate": 1.9004508947442582e-06,
"loss": 0.8916,
"mean_token_accuracy": 0.7725426994264126,
"num_tokens": 818690604.0,
"step": 9240
},
{
"entropy": 0.869580078125,
"epoch": 1.2641973621267,
"grad_norm": 0.7135518661358268,
"learning_rate": 1.8969282795547418e-06,
"loss": 0.8485,
"mean_token_accuracy": 0.7848115012049675,
"num_tokens": 819552184.0,
"step": 9250
},
{
"entropy": 0.898974609375,
"epoch": 1.2655641358573089,
"grad_norm": 0.6141851457081157,
"learning_rate": 1.8934056643652249e-06,
"loss": 0.8808,
"mean_token_accuracy": 0.7768262773752213,
"num_tokens": 820423497.0,
"step": 9260
},
{
"entropy": 0.8758544921875,
"epoch": 1.2669309095879178,
"grad_norm": 0.6100746889651343,
"learning_rate": 1.889883049175708e-06,
"loss": 0.8626,
"mean_token_accuracy": 0.7804457984864712,
"num_tokens": 821295538.0,
"step": 9270
},
{
"entropy": 0.836474609375,
"epoch": 1.2682976833185267,
"grad_norm": 0.6367051766375864,
"learning_rate": 1.8863604339861916e-06,
"loss": 0.822,
"mean_token_accuracy": 0.7859038561582565,
"num_tokens": 822180580.0,
"step": 9280
},
{
"entropy": 0.9169921875,
"epoch": 1.2696644570491356,
"grad_norm": 0.6688614608504452,
"learning_rate": 1.8828378187966747e-06,
"loss": 0.8857,
"mean_token_accuracy": 0.7752997860312462,
"num_tokens": 823090612.0,
"step": 9290
},
{
"entropy": 0.851904296875,
"epoch": 1.2710312307797444,
"grad_norm": 0.6215099346536026,
"learning_rate": 1.8793152036071582e-06,
"loss": 0.8552,
"mean_token_accuracy": 0.7810220293700695,
"num_tokens": 823965696.0,
"step": 9300
},
{
"entropy": 0.881298828125,
"epoch": 1.2723980045103533,
"grad_norm": 0.6057109122735109,
"learning_rate": 1.8757925884176414e-06,
"loss": 0.8477,
"mean_token_accuracy": 0.7819186106324196,
"num_tokens": 824871297.0,
"step": 9310
},
{
"entropy": 0.9122802734375,
"epoch": 1.2737647782409622,
"grad_norm": 0.7124787774469434,
"learning_rate": 1.8722699732281247e-06,
"loss": 0.9195,
"mean_token_accuracy": 0.7679477736353875,
"num_tokens": 825811434.0,
"step": 9320
},
{
"entropy": 0.908544921875,
"epoch": 1.2751315519715711,
"grad_norm": 0.6389695828586279,
"learning_rate": 1.868747358038608e-06,
"loss": 0.8995,
"mean_token_accuracy": 0.7727701015770435,
"num_tokens": 826682534.0,
"step": 9330
},
{
"entropy": 0.8605224609375,
"epoch": 1.27649832570218,
"grad_norm": 0.6138212656426449,
"learning_rate": 1.8652247428490914e-06,
"loss": 0.8507,
"mean_token_accuracy": 0.7827878817915916,
"num_tokens": 827565061.0,
"step": 9340
},
{
"entropy": 0.90478515625,
"epoch": 1.277865099432789,
"grad_norm": 0.7196911088178871,
"learning_rate": 1.8617021276595745e-06,
"loss": 0.8871,
"mean_token_accuracy": 0.7757990032434463,
"num_tokens": 828433291.0,
"step": 9350
},
{
"entropy": 0.875390625,
"epoch": 1.2792318731633978,
"grad_norm": 0.6314955915309679,
"learning_rate": 1.858179512470058e-06,
"loss": 0.8694,
"mean_token_accuracy": 0.7793812312185764,
"num_tokens": 829312319.0,
"step": 9360
},
{
"entropy": 0.86806640625,
"epoch": 1.2805986468940067,
"grad_norm": 0.631889217637198,
"learning_rate": 1.8546568972805412e-06,
"loss": 0.8531,
"mean_token_accuracy": 0.7829393729567528,
"num_tokens": 830252194.0,
"step": 9370
},
{
"entropy": 0.877783203125,
"epoch": 1.2819654206246156,
"grad_norm": 0.6371175277589287,
"learning_rate": 1.8511342820910245e-06,
"loss": 0.8401,
"mean_token_accuracy": 0.782128469645977,
"num_tokens": 831128961.0,
"step": 9380
},
{
"entropy": 0.853662109375,
"epoch": 1.2833321943552245,
"grad_norm": 0.6484183975027233,
"learning_rate": 1.8476116669015078e-06,
"loss": 0.8368,
"mean_token_accuracy": 0.7860616594552994,
"num_tokens": 832029477.0,
"step": 9390
},
{
"entropy": 0.8645263671875,
"epoch": 1.2846989680858334,
"grad_norm": 0.5818140349203766,
"learning_rate": 1.8440890517119912e-06,
"loss": 0.841,
"mean_token_accuracy": 0.7826697573065757,
"num_tokens": 832931731.0,
"step": 9400
},
{
"entropy": 0.874560546875,
"epoch": 1.2860657418164423,
"grad_norm": 0.6779998958368199,
"learning_rate": 1.8405664365224743e-06,
"loss": 0.8472,
"mean_token_accuracy": 0.7794681914150715,
"num_tokens": 833785182.0,
"step": 9410
},
{
"entropy": 0.885302734375,
"epoch": 1.2874325155470512,
"grad_norm": 0.6344081930115768,
"learning_rate": 1.8370438213329578e-06,
"loss": 0.8748,
"mean_token_accuracy": 0.7804376646876335,
"num_tokens": 834717467.0,
"step": 9420
},
{
"entropy": 0.8952880859375,
"epoch": 1.28879928927766,
"grad_norm": 0.6424615420341114,
"learning_rate": 1.833521206143441e-06,
"loss": 0.9003,
"mean_token_accuracy": 0.7722059540450573,
"num_tokens": 835602297.0,
"step": 9430
},
{
"entropy": 0.924560546875,
"epoch": 1.290166063008269,
"grad_norm": 0.6249084907327024,
"learning_rate": 1.8299985909539245e-06,
"loss": 0.9172,
"mean_token_accuracy": 0.768189200758934,
"num_tokens": 836448213.0,
"step": 9440
},
{
"entropy": 0.887451171875,
"epoch": 1.2915328367388779,
"grad_norm": 0.616805457097103,
"learning_rate": 1.8264759757644076e-06,
"loss": 0.8797,
"mean_token_accuracy": 0.7752693988382816,
"num_tokens": 837327129.0,
"step": 9450
},
{
"entropy": 0.864453125,
"epoch": 1.2928996104694868,
"grad_norm": 0.7399071288933752,
"learning_rate": 1.822953360574891e-06,
"loss": 0.8381,
"mean_token_accuracy": 0.7844132207334041,
"num_tokens": 838172987.0,
"step": 9460
},
{
"entropy": 0.851123046875,
"epoch": 1.2942663842000957,
"grad_norm": 0.5312825764356638,
"learning_rate": 1.8194307453853743e-06,
"loss": 0.8218,
"mean_token_accuracy": 0.7862383343279362,
"num_tokens": 839057791.0,
"step": 9470
},
{
"entropy": 0.9296875,
"epoch": 1.2956331579307045,
"grad_norm": 0.6892654944952115,
"learning_rate": 1.8159081301958576e-06,
"loss": 0.9186,
"mean_token_accuracy": 0.7714765183627605,
"num_tokens": 839956502.0,
"step": 9480
},
{
"entropy": 0.865478515625,
"epoch": 1.2969999316613134,
"grad_norm": 0.5702746571113436,
"learning_rate": 1.8123855150063408e-06,
"loss": 0.8382,
"mean_token_accuracy": 0.7849998950958252,
"num_tokens": 840854870.0,
"step": 9490
},
{
"entropy": 0.8805419921875,
"epoch": 1.2983667053919223,
"grad_norm": 1.0300249368065109,
"learning_rate": 1.8088628998168243e-06,
"loss": 0.86,
"mean_token_accuracy": 0.7819328658282757,
"num_tokens": 841722038.0,
"step": 9500
},
{
"entropy": 0.853955078125,
"epoch": 1.2997334791225312,
"grad_norm": 0.6770133788062203,
"learning_rate": 1.8053402846273074e-06,
"loss": 0.8345,
"mean_token_accuracy": 0.782308854907751,
"num_tokens": 842603404.0,
"step": 9510
},
{
"entropy": 0.839453125,
"epoch": 1.3011002528531401,
"grad_norm": 0.5896461681044857,
"learning_rate": 1.8018176694377906e-06,
"loss": 0.8301,
"mean_token_accuracy": 0.7856783114373684,
"num_tokens": 843468734.0,
"step": 9520
},
{
"entropy": 0.8845703125,
"epoch": 1.302467026583749,
"grad_norm": 0.6348154594919553,
"learning_rate": 1.798295054248274e-06,
"loss": 0.8791,
"mean_token_accuracy": 0.7788964614272118,
"num_tokens": 844321422.0,
"step": 9530
},
{
"entropy": 0.891650390625,
"epoch": 1.303833800314358,
"grad_norm": 0.7360636572137819,
"learning_rate": 1.7947724390587572e-06,
"loss": 0.859,
"mean_token_accuracy": 0.7785930201411247,
"num_tokens": 845185266.0,
"step": 9540
},
{
"entropy": 0.91005859375,
"epoch": 1.3052005740449668,
"grad_norm": 0.6495713451558776,
"learning_rate": 1.7912498238692408e-06,
"loss": 0.8978,
"mean_token_accuracy": 0.7717495799064636,
"num_tokens": 846083764.0,
"step": 9550
},
{
"entropy": 0.883544921875,
"epoch": 1.3065673477755757,
"grad_norm": 0.6091780502473008,
"learning_rate": 1.7877272086797239e-06,
"loss": 0.8703,
"mean_token_accuracy": 0.7801335245370865,
"num_tokens": 846968868.0,
"step": 9560
},
{
"entropy": 0.90458984375,
"epoch": 1.3079341215061846,
"grad_norm": 0.6095584331095178,
"learning_rate": 1.7842045934902072e-06,
"loss": 0.8909,
"mean_token_accuracy": 0.7740511707961559,
"num_tokens": 847867293.0,
"step": 9570
},
{
"entropy": 0.889501953125,
"epoch": 1.3093008952367935,
"grad_norm": 0.5854581700520355,
"learning_rate": 1.7806819783006908e-06,
"loss": 0.879,
"mean_token_accuracy": 0.7753022626042366,
"num_tokens": 848738392.0,
"step": 9580
},
{
"entropy": 0.883447265625,
"epoch": 1.3106676689674024,
"grad_norm": 0.6078723612062978,
"learning_rate": 1.777159363111174e-06,
"loss": 0.8798,
"mean_token_accuracy": 0.7795827366411686,
"num_tokens": 849620845.0,
"step": 9590
},
{
"entropy": 0.87578125,
"epoch": 1.3120344426980113,
"grad_norm": 0.5490950263735851,
"learning_rate": 1.773636747921657e-06,
"loss": 0.8547,
"mean_token_accuracy": 0.7777528196573258,
"num_tokens": 850463193.0,
"step": 9600
},
{
"entropy": 0.897216796875,
"epoch": 1.3134012164286202,
"grad_norm": 0.6861598947625639,
"learning_rate": 1.7701141327321406e-06,
"loss": 0.8566,
"mean_token_accuracy": 0.7774150021374225,
"num_tokens": 851372856.0,
"step": 9610
},
{
"entropy": 0.8720703125,
"epoch": 1.314767990159229,
"grad_norm": 0.5643992584208034,
"learning_rate": 1.7665915175426237e-06,
"loss": 0.8912,
"mean_token_accuracy": 0.7747424736618995,
"num_tokens": 852308597.0,
"step": 9620
},
{
"entropy": 0.863818359375,
"epoch": 1.316134763889838,
"grad_norm": 0.6512834755103465,
"learning_rate": 1.763068902353107e-06,
"loss": 0.854,
"mean_token_accuracy": 0.781435540318489,
"num_tokens": 853172611.0,
"step": 9630
},
{
"entropy": 0.915576171875,
"epoch": 1.3175015376204469,
"grad_norm": 0.6308999277520303,
"learning_rate": 1.7595462871635904e-06,
"loss": 0.8963,
"mean_token_accuracy": 0.7709020853042603,
"num_tokens": 854025251.0,
"step": 9640
},
{
"entropy": 0.8502685546875,
"epoch": 1.3188683113510558,
"grad_norm": 0.6095570383080909,
"learning_rate": 1.7560236719740737e-06,
"loss": 0.8229,
"mean_token_accuracy": 0.7872369073331356,
"num_tokens": 854918609.0,
"step": 9650
},
{
"entropy": 0.8765869140625,
"epoch": 1.3202350850816646,
"grad_norm": 0.563889655768779,
"learning_rate": 1.752501056784557e-06,
"loss": 0.8606,
"mean_token_accuracy": 0.7802076056599617,
"num_tokens": 855840317.0,
"step": 9660
},
{
"entropy": 0.8755126953125,
"epoch": 1.3216018588122735,
"grad_norm": 0.5887846475364755,
"learning_rate": 1.7489784415950404e-06,
"loss": 0.8492,
"mean_token_accuracy": 0.7815208174288273,
"num_tokens": 856694202.0,
"step": 9670
},
{
"entropy": 0.879052734375,
"epoch": 1.3229686325428824,
"grad_norm": 0.6060414812226678,
"learning_rate": 1.7454558264055235e-06,
"loss": 0.8398,
"mean_token_accuracy": 0.7804498597979546,
"num_tokens": 857560944.0,
"step": 9680
},
{
"entropy": 0.8783935546875,
"epoch": 1.3243354062734913,
"grad_norm": 0.6082013966500308,
"learning_rate": 1.741933211216007e-06,
"loss": 0.8581,
"mean_token_accuracy": 0.7828697450459003,
"num_tokens": 858425303.0,
"step": 9690
},
{
"entropy": 0.8900390625,
"epoch": 1.3257021800041002,
"grad_norm": 0.6222214256370517,
"learning_rate": 1.7384105960264902e-06,
"loss": 0.8653,
"mean_token_accuracy": 0.7791015192866325,
"num_tokens": 859322782.0,
"step": 9700
},
{
"entropy": 0.93037109375,
"epoch": 1.3270689537347091,
"grad_norm": 0.6705518684954962,
"learning_rate": 1.7348879808369735e-06,
"loss": 0.931,
"mean_token_accuracy": 0.7628317460417747,
"num_tokens": 860228542.0,
"step": 9710
},
{
"entropy": 0.87802734375,
"epoch": 1.328435727465318,
"grad_norm": 0.6477233601525666,
"learning_rate": 1.7313653656474568e-06,
"loss": 0.892,
"mean_token_accuracy": 0.7760573118925095,
"num_tokens": 861128996.0,
"step": 9720
},
{
"entropy": 0.908447265625,
"epoch": 1.329802501195927,
"grad_norm": 0.6592103924326732,
"learning_rate": 1.7278427504579402e-06,
"loss": 0.9043,
"mean_token_accuracy": 0.7738235987722873,
"num_tokens": 862045862.0,
"step": 9730
},
{
"entropy": 0.907080078125,
"epoch": 1.3311692749265358,
"grad_norm": 0.6418888780213199,
"learning_rate": 1.7243201352684233e-06,
"loss": 0.9102,
"mean_token_accuracy": 0.7748648524284363,
"num_tokens": 862980918.0,
"step": 9740
},
{
"entropy": 0.909716796875,
"epoch": 1.3325360486571447,
"grad_norm": 0.6471016999333733,
"learning_rate": 1.7207975200789068e-06,
"loss": 0.9169,
"mean_token_accuracy": 0.7703908666968345,
"num_tokens": 863864536.0,
"step": 9750
},
{
"entropy": 0.8905029296875,
"epoch": 1.3339028223877536,
"grad_norm": 0.6839383441967917,
"learning_rate": 1.71727490488939e-06,
"loss": 0.875,
"mean_token_accuracy": 0.7771003901958465,
"num_tokens": 864692208.0,
"step": 9760
},
{
"entropy": 0.873779296875,
"epoch": 1.3352695961183625,
"grad_norm": 0.7054787517657191,
"learning_rate": 1.7137522896998735e-06,
"loss": 0.8798,
"mean_token_accuracy": 0.7795041009783745,
"num_tokens": 865583812.0,
"step": 9770
},
{
"entropy": 0.8489013671875,
"epoch": 1.3366363698489714,
"grad_norm": 0.6361760424302663,
"learning_rate": 1.7102296745103566e-06,
"loss": 0.8301,
"mean_token_accuracy": 0.7883041873574257,
"num_tokens": 866442714.0,
"step": 9780
},
{
"entropy": 0.8918212890625,
"epoch": 1.3380031435795803,
"grad_norm": 0.6461352896973152,
"learning_rate": 1.7067070593208398e-06,
"loss": 0.8723,
"mean_token_accuracy": 0.7755745522677898,
"num_tokens": 867348693.0,
"step": 9790
},
{
"entropy": 0.87216796875,
"epoch": 1.3393699173101892,
"grad_norm": 0.6717532685991074,
"learning_rate": 1.7031844441313233e-06,
"loss": 0.8703,
"mean_token_accuracy": 0.7792750619351864,
"num_tokens": 868245125.0,
"step": 9800
},
{
"entropy": 0.878076171875,
"epoch": 1.340736691040798,
"grad_norm": 0.6452127699607376,
"learning_rate": 1.6996618289418064e-06,
"loss": 0.8639,
"mean_token_accuracy": 0.7815754629671574,
"num_tokens": 869147634.0,
"step": 9810
},
{
"entropy": 0.8826171875,
"epoch": 1.342103464771407,
"grad_norm": 0.6354065236695657,
"learning_rate": 1.6961392137522898e-06,
"loss": 0.8615,
"mean_token_accuracy": 0.7777621813118458,
"num_tokens": 869984569.0,
"step": 9820
},
{
"entropy": 0.857275390625,
"epoch": 1.3434702385020159,
"grad_norm": 0.5901859925338285,
"learning_rate": 1.6926165985627733e-06,
"loss": 0.8269,
"mean_token_accuracy": 0.7852850370109081,
"num_tokens": 870822937.0,
"step": 9830
},
{
"entropy": 0.9034423828125,
"epoch": 1.3448370122326248,
"grad_norm": 0.6260094248860993,
"learning_rate": 1.6890939833732564e-06,
"loss": 0.9195,
"mean_token_accuracy": 0.7682308733463288,
"num_tokens": 871733262.0,
"step": 9840
},
{
"entropy": 0.911474609375,
"epoch": 1.3462037859632339,
"grad_norm": 0.6539887655154177,
"learning_rate": 1.6855713681837396e-06,
"loss": 0.8986,
"mean_token_accuracy": 0.7738306798040867,
"num_tokens": 872630340.0,
"step": 9850
},
{
"entropy": 0.86953125,
"epoch": 1.3475705596938428,
"grad_norm": 0.7374917955828285,
"learning_rate": 1.6820487529942231e-06,
"loss": 0.8356,
"mean_token_accuracy": 0.7835083834826946,
"num_tokens": 873508298.0,
"step": 9860
},
{
"entropy": 0.8703125,
"epoch": 1.3489373334244517,
"grad_norm": 0.6740098136991473,
"learning_rate": 1.6785261378047062e-06,
"loss": 0.8363,
"mean_token_accuracy": 0.782510943710804,
"num_tokens": 874391916.0,
"step": 9870
},
{
"entropy": 0.89775390625,
"epoch": 1.3503041071550606,
"grad_norm": 0.6570113443267154,
"learning_rate": 1.6750035226151898e-06,
"loss": 0.8678,
"mean_token_accuracy": 0.7777207411825657,
"num_tokens": 875231876.0,
"step": 9880
},
{
"entropy": 0.89716796875,
"epoch": 1.3516708808856694,
"grad_norm": 0.6757201803251682,
"learning_rate": 1.671480907425673e-06,
"loss": 0.8653,
"mean_token_accuracy": 0.7785965092480183,
"num_tokens": 876127902.0,
"step": 9890
},
{
"entropy": 0.878515625,
"epoch": 1.3530376546162783,
"grad_norm": 0.6091121358918695,
"learning_rate": 1.6679582922361562e-06,
"loss": 0.8834,
"mean_token_accuracy": 0.7751582078635693,
"num_tokens": 876985950.0,
"step": 9900
},
{
"entropy": 0.87685546875,
"epoch": 1.3544044283468872,
"grad_norm": 0.6657647247450642,
"learning_rate": 1.6644356770466396e-06,
"loss": 0.864,
"mean_token_accuracy": 0.7795547924935817,
"num_tokens": 877880193.0,
"step": 9910
},
{
"entropy": 0.8850341796875,
"epoch": 1.3557712020774961,
"grad_norm": 0.6887182888081465,
"learning_rate": 1.660913061857123e-06,
"loss": 0.8656,
"mean_token_accuracy": 0.7794572599232197,
"num_tokens": 878770858.0,
"step": 9920
},
{
"entropy": 0.8510498046875,
"epoch": 1.357137975808105,
"grad_norm": 0.6525214585913073,
"learning_rate": 1.657390446667606e-06,
"loss": 0.8216,
"mean_token_accuracy": 0.7857621617615222,
"num_tokens": 879667998.0,
"step": 9930
},
{
"entropy": 0.876171875,
"epoch": 1.358504749538714,
"grad_norm": 0.6086931930355584,
"learning_rate": 1.6538678314780896e-06,
"loss": 0.87,
"mean_token_accuracy": 0.7794688597321511,
"num_tokens": 880544633.0,
"step": 9940
},
{
"entropy": 0.9483154296875,
"epoch": 1.3598715232693228,
"grad_norm": 0.692765053822939,
"learning_rate": 1.6503452162885727e-06,
"loss": 0.9266,
"mean_token_accuracy": 0.7674866408109665,
"num_tokens": 881402825.0,
"step": 9950
},
{
"entropy": 0.8789794921875,
"epoch": 1.3612382969999317,
"grad_norm": 0.7422057745919242,
"learning_rate": 1.646822601099056e-06,
"loss": 0.857,
"mean_token_accuracy": 0.7777922756969928,
"num_tokens": 882291149.0,
"step": 9960
},
{
"entropy": 0.8505859375,
"epoch": 1.3626050707305406,
"grad_norm": 0.6273805051076747,
"learning_rate": 1.6432999859095394e-06,
"loss": 0.8514,
"mean_token_accuracy": 0.7848576217889786,
"num_tokens": 883176457.0,
"step": 9970
},
{
"entropy": 0.919580078125,
"epoch": 1.3639718444611495,
"grad_norm": 0.7151584575337099,
"learning_rate": 1.6397773707200227e-06,
"loss": 0.9108,
"mean_token_accuracy": 0.7687877170741558,
"num_tokens": 884063256.0,
"step": 9980
},
{
"entropy": 0.8807861328125,
"epoch": 1.3653386181917584,
"grad_norm": 0.5749706292830427,
"learning_rate": 1.636254755530506e-06,
"loss": 0.8666,
"mean_token_accuracy": 0.7784491300582885,
"num_tokens": 884906668.0,
"step": 9990
},
{
"entropy": 0.83046875,
"epoch": 1.3667053919223673,
"grad_norm": 0.6154759198104087,
"learning_rate": 1.6327321403409894e-06,
"loss": 0.8039,
"mean_token_accuracy": 0.7919099286198616,
"num_tokens": 885782940.0,
"step": 10000
},
{
"entropy": 0.90302734375,
"epoch": 1.3680721656529762,
"grad_norm": 0.7063533296312744,
"learning_rate": 1.6292095251514725e-06,
"loss": 0.8924,
"mean_token_accuracy": 0.7761632561683655,
"num_tokens": 886655374.0,
"step": 10010
},
{
"entropy": 0.903857421875,
"epoch": 1.369438939383585,
"grad_norm": 0.7028149697267406,
"learning_rate": 1.625686909961956e-06,
"loss": 0.8806,
"mean_token_accuracy": 0.7759507477283478,
"num_tokens": 887551775.0,
"step": 10020
},
{
"entropy": 0.896728515625,
"epoch": 1.370805713114194,
"grad_norm": 0.6733098164820794,
"learning_rate": 1.6221642947724392e-06,
"loss": 0.8815,
"mean_token_accuracy": 0.7768429256975651,
"num_tokens": 888388302.0,
"step": 10030
},
{
"entropy": 0.9132568359375,
"epoch": 1.3721724868448029,
"grad_norm": 0.6596183561871105,
"learning_rate": 1.6186416795829223e-06,
"loss": 0.897,
"mean_token_accuracy": 0.7705864317715168,
"num_tokens": 889290281.0,
"step": 10040
},
{
"entropy": 0.90625,
"epoch": 1.3735392605754118,
"grad_norm": 0.6716996176831272,
"learning_rate": 1.6151190643934058e-06,
"loss": 0.8832,
"mean_token_accuracy": 0.7751258246600627,
"num_tokens": 890192489.0,
"step": 10050
},
{
"entropy": 0.859326171875,
"epoch": 1.3749060343060207,
"grad_norm": 0.6273119623608236,
"learning_rate": 1.6115964492038892e-06,
"loss": 0.8416,
"mean_token_accuracy": 0.783117139339447,
"num_tokens": 891071923.0,
"step": 10060
},
{
"entropy": 0.872314453125,
"epoch": 1.3762728080366295,
"grad_norm": 0.6363798937428152,
"learning_rate": 1.6080738340143723e-06,
"loss": 0.8806,
"mean_token_accuracy": 0.7787937097251415,
"num_tokens": 891978878.0,
"step": 10070
},
{
"entropy": 0.867236328125,
"epoch": 1.3776395817672384,
"grad_norm": 0.6292932863792872,
"learning_rate": 1.6045512188248559e-06,
"loss": 0.8392,
"mean_token_accuracy": 0.7845931209623813,
"num_tokens": 892814800.0,
"step": 10080
},
{
"entropy": 0.888720703125,
"epoch": 1.3790063554978473,
"grad_norm": 0.6243261675059499,
"learning_rate": 1.601028603635339e-06,
"loss": 0.8779,
"mean_token_accuracy": 0.7772016569972038,
"num_tokens": 893678292.0,
"step": 10090
},
{
"entropy": 0.870361328125,
"epoch": 1.3803731292284562,
"grad_norm": 0.7325674908673301,
"learning_rate": 1.5975059884458225e-06,
"loss": 0.8765,
"mean_token_accuracy": 0.7799178250133991,
"num_tokens": 894600271.0,
"step": 10100
},
{
"entropy": 0.875341796875,
"epoch": 1.3817399029590651,
"grad_norm": 0.6525512986841929,
"learning_rate": 1.5939833732563056e-06,
"loss": 0.8521,
"mean_token_accuracy": 0.7825881227850914,
"num_tokens": 895489860.0,
"step": 10110
},
{
"entropy": 0.8810302734375,
"epoch": 1.383106676689674,
"grad_norm": 0.5742253137386478,
"learning_rate": 1.5904607580667888e-06,
"loss": 0.8736,
"mean_token_accuracy": 0.777743661403656,
"num_tokens": 896400113.0,
"step": 10120
},
{
"entropy": 0.883154296875,
"epoch": 1.384473450420283,
"grad_norm": 0.704910349455101,
"learning_rate": 1.5869381428772723e-06,
"loss": 0.8781,
"mean_token_accuracy": 0.7772124655544758,
"num_tokens": 897273945.0,
"step": 10130
},
{
"entropy": 0.8938720703125,
"epoch": 1.3858402241508918,
"grad_norm": 0.6416323543849882,
"learning_rate": 1.5834155276877554e-06,
"loss": 0.8717,
"mean_token_accuracy": 0.7765508249402047,
"num_tokens": 898141031.0,
"step": 10140
},
{
"entropy": 0.9155517578125,
"epoch": 1.3872069978815007,
"grad_norm": 0.6134043375192502,
"learning_rate": 1.5798929124982388e-06,
"loss": 0.8777,
"mean_token_accuracy": 0.7755227439105511,
"num_tokens": 898982893.0,
"step": 10150
},
{
"entropy": 0.8649658203125,
"epoch": 1.3885737716121096,
"grad_norm": 0.6676787586915455,
"learning_rate": 1.5763702973087221e-06,
"loss": 0.8929,
"mean_token_accuracy": 0.7762259654700756,
"num_tokens": 899851800.0,
"step": 10160
},
{
"entropy": 0.8535400390625,
"epoch": 1.3899405453427185,
"grad_norm": 0.9476778743584067,
"learning_rate": 1.5728476821192054e-06,
"loss": 0.8151,
"mean_token_accuracy": 0.7903167277574539,
"num_tokens": 900727352.0,
"step": 10170
},
{
"entropy": 0.9126953125,
"epoch": 1.3913073190733274,
"grad_norm": 0.9072720016895268,
"learning_rate": 1.5693250669296886e-06,
"loss": 0.9072,
"mean_token_accuracy": 0.7711271405220032,
"num_tokens": 901637094.0,
"step": 10180
},
{
"entropy": 0.8537841796875,
"epoch": 1.3926740928039363,
"grad_norm": 0.6436526380126867,
"learning_rate": 1.5658024517401721e-06,
"loss": 0.852,
"mean_token_accuracy": 0.7825956776738167,
"num_tokens": 902522639.0,
"step": 10190
},
{
"entropy": 0.873876953125,
"epoch": 1.3940408665345452,
"grad_norm": 0.5644343537699938,
"learning_rate": 1.5622798365506552e-06,
"loss": 0.8522,
"mean_token_accuracy": 0.7837581798434258,
"num_tokens": 903429509.0,
"step": 10200
},
{
"entropy": 0.884375,
"epoch": 1.395407640265154,
"grad_norm": 0.8517659680980705,
"learning_rate": 1.5587572213611388e-06,
"loss": 0.8694,
"mean_token_accuracy": 0.7783410370349884,
"num_tokens": 904317564.0,
"step": 10210
},
{
"entropy": 0.9185546875,
"epoch": 1.396774413995763,
"grad_norm": 0.8948795534080601,
"learning_rate": 1.555234606171622e-06,
"loss": 0.8888,
"mean_token_accuracy": 0.7726988896727562,
"num_tokens": 905224463.0,
"step": 10220
},
{
"entropy": 0.8722900390625,
"epoch": 1.3981411877263719,
"grad_norm": 0.5884951549855936,
"learning_rate": 1.5517119909821052e-06,
"loss": 0.8747,
"mean_token_accuracy": 0.7771315529942513,
"num_tokens": 906142295.0,
"step": 10230
},
{
"entropy": 0.9030517578125,
"epoch": 1.3995079614569808,
"grad_norm": 0.6074314827508206,
"learning_rate": 1.5481893757925886e-06,
"loss": 0.8939,
"mean_token_accuracy": 0.7728781782090663,
"num_tokens": 907054448.0,
"step": 10240
},
{
"entropy": 0.917529296875,
"epoch": 1.4008747351875896,
"grad_norm": 0.730108547543514,
"learning_rate": 1.544666760603072e-06,
"loss": 0.8826,
"mean_token_accuracy": 0.7757147312164306,
"num_tokens": 907895116.0,
"step": 10250
},
{
"entropy": 0.917333984375,
"epoch": 1.4022415089181985,
"grad_norm": 0.720367552781828,
"learning_rate": 1.541144145413555e-06,
"loss": 0.9268,
"mean_token_accuracy": 0.7698544099926948,
"num_tokens": 908802189.0,
"step": 10260
},
{
"entropy": 0.8682373046875,
"epoch": 1.4036082826488074,
"grad_norm": 0.623493597801881,
"learning_rate": 1.5376215302240386e-06,
"loss": 0.8654,
"mean_token_accuracy": 0.7825506269931793,
"num_tokens": 909755000.0,
"step": 10270
},
{
"entropy": 0.884375,
"epoch": 1.4049750563794163,
"grad_norm": 0.5803718650417758,
"learning_rate": 1.5340989150345217e-06,
"loss": 0.8979,
"mean_token_accuracy": 0.7755442500114441,
"num_tokens": 910650883.0,
"step": 10280
},
{
"entropy": 0.8247802734375,
"epoch": 1.4063418301100252,
"grad_norm": 1.0064260372985268,
"learning_rate": 1.5305762998450048e-06,
"loss": 0.8026,
"mean_token_accuracy": 0.7929730333387852,
"num_tokens": 911523735.0,
"step": 10290
},
{
"entropy": 0.9270751953125,
"epoch": 1.4077086038406341,
"grad_norm": 0.7242303014745257,
"learning_rate": 1.5270536846554884e-06,
"loss": 0.9046,
"mean_token_accuracy": 0.7728070452809334,
"num_tokens": 912401560.0,
"step": 10300
},
{
"entropy": 0.890380859375,
"epoch": 1.409075377571243,
"grad_norm": 0.6854699289735641,
"learning_rate": 1.5235310694659717e-06,
"loss": 0.8931,
"mean_token_accuracy": 0.773138527572155,
"num_tokens": 913302312.0,
"step": 10310
},
{
"entropy": 0.8533203125,
"epoch": 1.410442151301852,
"grad_norm": 0.675488568157104,
"learning_rate": 1.5200084542764548e-06,
"loss": 0.8222,
"mean_token_accuracy": 0.7845391899347305,
"num_tokens": 914208968.0,
"step": 10320
},
{
"entropy": 0.8943359375,
"epoch": 1.4118089250324608,
"grad_norm": 0.660116094376874,
"learning_rate": 1.5164858390869384e-06,
"loss": 0.8884,
"mean_token_accuracy": 0.7740811176598072,
"num_tokens": 915092764.0,
"step": 10330
},
{
"entropy": 0.9355712890625,
"epoch": 1.4131756987630697,
"grad_norm": 0.6907591149896903,
"learning_rate": 1.5129632238974215e-06,
"loss": 0.9505,
"mean_token_accuracy": 0.763513357937336,
"num_tokens": 915947203.0,
"step": 10340
},
{
"entropy": 0.887158203125,
"epoch": 1.4145424724936786,
"grad_norm": 0.6023092298375576,
"learning_rate": 1.509440608707905e-06,
"loss": 0.8862,
"mean_token_accuracy": 0.7741846047341824,
"num_tokens": 916862972.0,
"step": 10350
},
{
"entropy": 0.89755859375,
"epoch": 1.4159092462242875,
"grad_norm": 0.6483384102849886,
"learning_rate": 1.5059179935183882e-06,
"loss": 0.895,
"mean_token_accuracy": 0.772338493168354,
"num_tokens": 917739748.0,
"step": 10360
},
{
"entropy": 0.886328125,
"epoch": 1.4172760199548966,
"grad_norm": 0.6395234818891629,
"learning_rate": 1.5023953783288713e-06,
"loss": 0.8911,
"mean_token_accuracy": 0.7750992380082607,
"num_tokens": 918638904.0,
"step": 10370
},
{
"entropy": 0.88623046875,
"epoch": 1.4186427936855055,
"grad_norm": 0.6894896525817777,
"learning_rate": 1.4988727631393549e-06,
"loss": 0.8903,
"mean_token_accuracy": 0.7719627745449543,
"num_tokens": 919508022.0,
"step": 10380
},
{
"entropy": 0.9142578125,
"epoch": 1.4200095674161144,
"grad_norm": 0.7029931922877469,
"learning_rate": 1.495350147949838e-06,
"loss": 0.8876,
"mean_token_accuracy": 0.7747014030814171,
"num_tokens": 920421270.0,
"step": 10390
},
{
"entropy": 0.87265625,
"epoch": 1.4213763411467233,
"grad_norm": 0.6422393612621936,
"learning_rate": 1.4918275327603213e-06,
"loss": 0.8377,
"mean_token_accuracy": 0.7859460040926933,
"num_tokens": 921268963.0,
"step": 10400
},
{
"entropy": 0.861083984375,
"epoch": 1.4227431148773322,
"grad_norm": 0.7616006750424962,
"learning_rate": 1.4883049175708047e-06,
"loss": 0.8397,
"mean_token_accuracy": 0.7830014176666736,
"num_tokens": 922110156.0,
"step": 10410
},
{
"entropy": 0.894580078125,
"epoch": 1.424109888607941,
"grad_norm": 0.6198597548728054,
"learning_rate": 1.484782302381288e-06,
"loss": 0.8632,
"mean_token_accuracy": 0.7750765137374401,
"num_tokens": 923000326.0,
"step": 10420
},
{
"entropy": 0.893115234375,
"epoch": 1.42547666233855,
"grad_norm": 0.6865121492143131,
"learning_rate": 1.4812596871917711e-06,
"loss": 0.8887,
"mean_token_accuracy": 0.7749325193464756,
"num_tokens": 923866880.0,
"step": 10430
},
{
"entropy": 0.869921875,
"epoch": 1.4268434360691589,
"grad_norm": 0.6639333709431997,
"learning_rate": 1.4777370720022547e-06,
"loss": 0.8428,
"mean_token_accuracy": 0.7822246931493282,
"num_tokens": 924736744.0,
"step": 10440
},
{
"entropy": 0.875439453125,
"epoch": 1.4282102097997678,
"grad_norm": 0.6610778452246109,
"learning_rate": 1.4742144568127378e-06,
"loss": 0.8714,
"mean_token_accuracy": 0.7761015571653843,
"num_tokens": 925648312.0,
"step": 10450
},
{
"entropy": 0.855322265625,
"epoch": 1.4295769835303767,
"grad_norm": 0.6336505788617716,
"learning_rate": 1.4706918416232213e-06,
"loss": 0.8281,
"mean_token_accuracy": 0.7860072173178196,
"num_tokens": 926529038.0,
"step": 10460
},
{
"entropy": 0.896728515625,
"epoch": 1.4309437572609855,
"grad_norm": 0.5932409213380131,
"learning_rate": 1.4671692264337045e-06,
"loss": 0.8616,
"mean_token_accuracy": 0.7776559375226497,
"num_tokens": 927441768.0,
"step": 10470
},
{
"entropy": 0.882177734375,
"epoch": 1.4323105309915944,
"grad_norm": 0.5531434093600732,
"learning_rate": 1.4636466112441878e-06,
"loss": 0.868,
"mean_token_accuracy": 0.7777209177613258,
"num_tokens": 928358695.0,
"step": 10480
},
{
"entropy": 0.865673828125,
"epoch": 1.4336773047222033,
"grad_norm": 0.6090126340361574,
"learning_rate": 1.4601239960546711e-06,
"loss": 0.8416,
"mean_token_accuracy": 0.7817985102534294,
"num_tokens": 929240770.0,
"step": 10490
},
{
"entropy": 0.891748046875,
"epoch": 1.4350440784528122,
"grad_norm": 0.7232837754605569,
"learning_rate": 1.4566013808651545e-06,
"loss": 0.8687,
"mean_token_accuracy": 0.7776722058653831,
"num_tokens": 930170684.0,
"step": 10500
},
{
"entropy": 0.84697265625,
"epoch": 1.4364108521834211,
"grad_norm": 0.6079199705486035,
"learning_rate": 1.4530787656756376e-06,
"loss": 0.8377,
"mean_token_accuracy": 0.786101184040308,
"num_tokens": 931088028.0,
"step": 10510
},
{
"entropy": 0.8712890625,
"epoch": 1.43777762591403,
"grad_norm": 0.6720386321676393,
"learning_rate": 1.4495561504861211e-06,
"loss": 0.8533,
"mean_token_accuracy": 0.7805227816104889,
"num_tokens": 932014724.0,
"step": 10520
},
{
"entropy": 0.89619140625,
"epoch": 1.439144399644639,
"grad_norm": 0.5816684233181901,
"learning_rate": 1.4460335352966043e-06,
"loss": 0.8697,
"mean_token_accuracy": 0.779954056441784,
"num_tokens": 932888906.0,
"step": 10530
},
{
"entropy": 0.940673828125,
"epoch": 1.4405111733752478,
"grad_norm": 0.7763028312526392,
"learning_rate": 1.4425109201070876e-06,
"loss": 0.9374,
"mean_token_accuracy": 0.7647092245519161,
"num_tokens": 933851911.0,
"step": 10540
},
{
"entropy": 0.8780029296875,
"epoch": 1.4418779471058567,
"grad_norm": 1.1031609539649712,
"learning_rate": 1.438988304917571e-06,
"loss": 0.8598,
"mean_token_accuracy": 0.7780480168759822,
"num_tokens": 934766976.0,
"step": 10550
},
{
"entropy": 0.9062255859375,
"epoch": 1.4432447208364656,
"grad_norm": 0.6050536387235281,
"learning_rate": 1.4354656897280543e-06,
"loss": 0.8922,
"mean_token_accuracy": 0.7736318409442902,
"num_tokens": 935658392.0,
"step": 10560
},
{
"entropy": 0.88798828125,
"epoch": 1.4446114945670745,
"grad_norm": 0.636219520084406,
"learning_rate": 1.4319430745385376e-06,
"loss": 0.8669,
"mean_token_accuracy": 0.7772695124149323,
"num_tokens": 936532519.0,
"step": 10570
},
{
"entropy": 0.8797119140625,
"epoch": 1.4459782682976834,
"grad_norm": 0.6082025038780904,
"learning_rate": 1.428420459349021e-06,
"loss": 0.859,
"mean_token_accuracy": 0.7829842545092106,
"num_tokens": 937413962.0,
"step": 10580
},
{
"entropy": 0.846484375,
"epoch": 1.4473450420282923,
"grad_norm": 0.6622165791136518,
"learning_rate": 1.424897844159504e-06,
"loss": 0.8318,
"mean_token_accuracy": 0.785702820122242,
"num_tokens": 938298992.0,
"step": 10590
},
{
"entropy": 0.876953125,
"epoch": 1.4487118157589012,
"grad_norm": 0.7477253238388694,
"learning_rate": 1.4213752289699876e-06,
"loss": 0.8629,
"mean_token_accuracy": 0.778553469479084,
"num_tokens": 939178986.0,
"step": 10600
},
{
"entropy": 0.8494140625,
"epoch": 1.45007858948951,
"grad_norm": 0.5505871481843785,
"learning_rate": 1.4178526137804707e-06,
"loss": 0.8394,
"mean_token_accuracy": 0.7824489019811154,
"num_tokens": 940050901.0,
"step": 10610
},
{
"entropy": 0.874169921875,
"epoch": 1.451445363220119,
"grad_norm": 0.635743240259711,
"learning_rate": 1.4143299985909538e-06,
"loss": 0.8344,
"mean_token_accuracy": 0.7854037299752236,
"num_tokens": 940916956.0,
"step": 10620
},
{
"entropy": 0.8732421875,
"epoch": 1.4528121369507279,
"grad_norm": 0.6782285973770746,
"learning_rate": 1.4108073834014374e-06,
"loss": 0.8464,
"mean_token_accuracy": 0.780683160573244,
"num_tokens": 941829889.0,
"step": 10630
},
{
"entropy": 0.8555419921875,
"epoch": 1.4541789106813368,
"grad_norm": 0.6373884573026494,
"learning_rate": 1.4072847682119205e-06,
"loss": 0.8481,
"mean_token_accuracy": 0.7843662969768047,
"num_tokens": 942736711.0,
"step": 10640
},
{
"entropy": 0.8633056640625,
"epoch": 1.4555456844119457,
"grad_norm": 0.6189036098917922,
"learning_rate": 1.4037621530224039e-06,
"loss": 0.8529,
"mean_token_accuracy": 0.7822520762681962,
"num_tokens": 943671086.0,
"step": 10650
},
{
"entropy": 0.8798095703125,
"epoch": 1.4569124581425545,
"grad_norm": 0.6117400637832722,
"learning_rate": 1.4002395378328872e-06,
"loss": 0.8571,
"mean_token_accuracy": 0.7801527351140976,
"num_tokens": 944598233.0,
"step": 10660
},
{
"entropy": 0.870068359375,
"epoch": 1.4582792318731634,
"grad_norm": 0.6474514220086224,
"learning_rate": 1.3967169226433705e-06,
"loss": 0.8361,
"mean_token_accuracy": 0.7852485448122024,
"num_tokens": 945480002.0,
"step": 10670
},
{
"entropy": 0.8858154296875,
"epoch": 1.4596460056037723,
"grad_norm": 0.6286751767952814,
"learning_rate": 1.393194307453854e-06,
"loss": 0.8809,
"mean_token_accuracy": 0.7755683712661267,
"num_tokens": 946388987.0,
"step": 10680
},
{
"entropy": 0.9116455078125,
"epoch": 1.4610127793343812,
"grad_norm": 0.6678815666429074,
"learning_rate": 1.3896716922643372e-06,
"loss": 0.8887,
"mean_token_accuracy": 0.7714920826256275,
"num_tokens": 947253180.0,
"step": 10690
},
{
"entropy": 0.864306640625,
"epoch": 1.4623795530649901,
"grad_norm": 0.5634596999864351,
"learning_rate": 1.3861490770748203e-06,
"loss": 0.852,
"mean_token_accuracy": 0.7819933436810971,
"num_tokens": 948160131.0,
"step": 10700
},
{
"entropy": 0.89716796875,
"epoch": 1.463746326795599,
"grad_norm": 0.6916602396001389,
"learning_rate": 1.3826264618853039e-06,
"loss": 0.8674,
"mean_token_accuracy": 0.7761421859264374,
"num_tokens": 949000608.0,
"step": 10710
},
{
"entropy": 0.9085693359375,
"epoch": 1.465113100526208,
"grad_norm": 0.5963874678553922,
"learning_rate": 1.379103846695787e-06,
"loss": 0.9003,
"mean_token_accuracy": 0.7730742216110229,
"num_tokens": 949951490.0,
"step": 10720
},
{
"entropy": 0.90380859375,
"epoch": 1.4664798742568168,
"grad_norm": 0.6412130486176985,
"learning_rate": 1.3755812315062703e-06,
"loss": 0.883,
"mean_token_accuracy": 0.777616772800684,
"num_tokens": 950835351.0,
"step": 10730
},
{
"entropy": 0.85693359375,
"epoch": 1.4678466479874257,
"grad_norm": 0.5557819526205052,
"learning_rate": 1.3720586163167537e-06,
"loss": 0.8474,
"mean_token_accuracy": 0.7831106811761857,
"num_tokens": 951720959.0,
"step": 10740
},
{
"entropy": 0.91376953125,
"epoch": 1.4692134217180346,
"grad_norm": 0.7036249737696106,
"learning_rate": 1.368536001127237e-06,
"loss": 0.9037,
"mean_token_accuracy": 0.7733076870441437,
"num_tokens": 952584602.0,
"step": 10750
},
{
"entropy": 0.8474609375,
"epoch": 1.4705801954486435,
"grad_norm": 0.6081170224976907,
"learning_rate": 1.3650133859377201e-06,
"loss": 0.8291,
"mean_token_accuracy": 0.7847199484705925,
"num_tokens": 953459805.0,
"step": 10760
},
{
"entropy": 0.89208984375,
"epoch": 1.4719469691792524,
"grad_norm": 0.6108201955535805,
"learning_rate": 1.3614907707482037e-06,
"loss": 0.8756,
"mean_token_accuracy": 0.7748410783708095,
"num_tokens": 954324502.0,
"step": 10770
},
{
"entropy": 0.94208984375,
"epoch": 1.4733137429098613,
"grad_norm": 0.6223855423881038,
"learning_rate": 1.3579681555586868e-06,
"loss": 0.9071,
"mean_token_accuracy": 0.7695388652384281,
"num_tokens": 955233860.0,
"step": 10780
},
{
"entropy": 0.901416015625,
"epoch": 1.4746805166404702,
"grad_norm": 0.5408366312693538,
"learning_rate": 1.3544455403691703e-06,
"loss": 0.8724,
"mean_token_accuracy": 0.7771421194076538,
"num_tokens": 956148137.0,
"step": 10790
},
{
"entropy": 0.86923828125,
"epoch": 1.476047290371079,
"grad_norm": 0.6389966767930418,
"learning_rate": 1.3509229251796535e-06,
"loss": 0.8403,
"mean_token_accuracy": 0.782866370677948,
"num_tokens": 956998510.0,
"step": 10800
},
{
"entropy": 0.885498046875,
"epoch": 1.477414064101688,
"grad_norm": 0.6238512973970612,
"learning_rate": 1.3474003099901368e-06,
"loss": 0.8771,
"mean_token_accuracy": 0.7754251681268215,
"num_tokens": 957859645.0,
"step": 10810
},
{
"entropy": 0.855224609375,
"epoch": 1.4787808378322969,
"grad_norm": 0.582197961588558,
"learning_rate": 1.3438776948006201e-06,
"loss": 0.8223,
"mean_token_accuracy": 0.7858412027359009,
"num_tokens": 958729483.0,
"step": 10820
},
{
"entropy": 0.85185546875,
"epoch": 1.4801476115629058,
"grad_norm": 0.6026286471734382,
"learning_rate": 1.3403550796111035e-06,
"loss": 0.8315,
"mean_token_accuracy": 0.786886865645647,
"num_tokens": 959617772.0,
"step": 10830
},
{
"entropy": 0.866796875,
"epoch": 1.4815143852935146,
"grad_norm": 0.7042155476737803,
"learning_rate": 1.3368324644215866e-06,
"loss": 0.8512,
"mean_token_accuracy": 0.780373913794756,
"num_tokens": 960496565.0,
"step": 10840
},
{
"entropy": 0.8495849609375,
"epoch": 1.4828811590241235,
"grad_norm": 0.5742955518646963,
"learning_rate": 1.3333098492320701e-06,
"loss": 0.8122,
"mean_token_accuracy": 0.7885249681770802,
"num_tokens": 961390449.0,
"step": 10850
},
{
"entropy": 0.913720703125,
"epoch": 1.4842479327547324,
"grad_norm": 0.7150157716866442,
"learning_rate": 1.3297872340425533e-06,
"loss": 0.9007,
"mean_token_accuracy": 0.7699067160487175,
"num_tokens": 962243695.0,
"step": 10860
},
{
"entropy": 0.90830078125,
"epoch": 1.4856147064853413,
"grad_norm": 0.6518728186176995,
"learning_rate": 1.3262646188530364e-06,
"loss": 0.921,
"mean_token_accuracy": 0.7720843970775604,
"num_tokens": 963171977.0,
"step": 10870
},
{
"entropy": 0.863720703125,
"epoch": 1.4869814802159502,
"grad_norm": 0.5737093703016354,
"learning_rate": 1.32274200366352e-06,
"loss": 0.8453,
"mean_token_accuracy": 0.7855264782905579,
"num_tokens": 964055991.0,
"step": 10880
},
{
"entropy": 0.8646240234375,
"epoch": 1.4883482539465591,
"grad_norm": 0.5901970752103564,
"learning_rate": 1.319219388474003e-06,
"loss": 0.841,
"mean_token_accuracy": 0.7840908542275429,
"num_tokens": 964895585.0,
"step": 10890
},
{
"entropy": 0.896875,
"epoch": 1.489715027677168,
"grad_norm": 0.6245991354502892,
"learning_rate": 1.3156967732844866e-06,
"loss": 0.887,
"mean_token_accuracy": 0.775787065923214,
"num_tokens": 965796547.0,
"step": 10900
},
{
"entropy": 0.8514892578125,
"epoch": 1.491081801407777,
"grad_norm": 0.6435324104594672,
"learning_rate": 1.31217415809497e-06,
"loss": 0.8129,
"mean_token_accuracy": 0.7891143605113029,
"num_tokens": 966706376.0,
"step": 10910
},
{
"entropy": 0.928564453125,
"epoch": 1.4924485751383858,
"grad_norm": 0.6511673865030082,
"learning_rate": 1.308651542905453e-06,
"loss": 0.9483,
"mean_token_accuracy": 0.7670082464814186,
"num_tokens": 967641987.0,
"step": 10920
},
{
"entropy": 0.8625732421875,
"epoch": 1.4938153488689947,
"grad_norm": 0.6729170468308742,
"learning_rate": 1.3051289277159366e-06,
"loss": 0.853,
"mean_token_accuracy": 0.7813981883227825,
"num_tokens": 968513222.0,
"step": 10930
},
{
"entropy": 0.886962890625,
"epoch": 1.4951821225996036,
"grad_norm": 0.6371234933037652,
"learning_rate": 1.3016063125264197e-06,
"loss": 0.8773,
"mean_token_accuracy": 0.7788613051176071,
"num_tokens": 969379276.0,
"step": 10940
},
{
"entropy": 0.89921875,
"epoch": 1.4965488963302125,
"grad_norm": 0.6932387750342471,
"learning_rate": 1.2980836973369029e-06,
"loss": 0.8801,
"mean_token_accuracy": 0.77412104383111,
"num_tokens": 970233883.0,
"step": 10950
},
{
"entropy": 0.8658935546875,
"epoch": 1.4979156700608214,
"grad_norm": 0.6671726231646383,
"learning_rate": 1.2945610821473864e-06,
"loss": 0.857,
"mean_token_accuracy": 0.7810898371040821,
"num_tokens": 971089571.0,
"step": 10960
},
{
"entropy": 0.87587890625,
"epoch": 1.4992824437914303,
"grad_norm": 0.6755824731804873,
"learning_rate": 1.2910384669578695e-06,
"loss": 0.8248,
"mean_token_accuracy": 0.789128589630127,
"num_tokens": 971969845.0,
"step": 10970
},
{
"entropy": 0.865185546875,
"epoch": 1.5006492175220392,
"grad_norm": 0.6779982294835583,
"learning_rate": 1.2875158517683529e-06,
"loss": 0.8457,
"mean_token_accuracy": 0.7839606180787086,
"num_tokens": 972830400.0,
"step": 10980
},
{
"entropy": 0.90205078125,
"epoch": 1.502015991252648,
"grad_norm": 0.6403232533125717,
"learning_rate": 1.2839932365788362e-06,
"loss": 0.896,
"mean_token_accuracy": 0.7714507639408111,
"num_tokens": 973711071.0,
"step": 10990
},
{
"entropy": 0.8745361328125,
"epoch": 1.503382764983257,
"grad_norm": 0.5779912407257277,
"learning_rate": 1.2804706213893195e-06,
"loss": 0.8758,
"mean_token_accuracy": 0.7775009199976921,
"num_tokens": 974615703.0,
"step": 11000
},
{
"entropy": 0.9189208984375,
"epoch": 1.5047495387138659,
"grad_norm": 0.6400847166203043,
"learning_rate": 1.2769480061998029e-06,
"loss": 0.9074,
"mean_token_accuracy": 0.7721088133752346,
"num_tokens": 975506409.0,
"step": 11010
},
{
"entropy": 0.8544677734375,
"epoch": 1.5061163124444747,
"grad_norm": 0.6803292134390948,
"learning_rate": 1.2734253910102862e-06,
"loss": 0.8311,
"mean_token_accuracy": 0.7881742708384991,
"num_tokens": 976439918.0,
"step": 11020
},
{
"entropy": 0.8749267578125,
"epoch": 1.5074830861750836,
"grad_norm": 0.6353222559776464,
"learning_rate": 1.2699027758207693e-06,
"loss": 0.8673,
"mean_token_accuracy": 0.7807885713875293,
"num_tokens": 977319557.0,
"step": 11030
},
{
"entropy": 0.9337890625,
"epoch": 1.5088498599056925,
"grad_norm": 0.647953558914027,
"learning_rate": 1.2663801606312529e-06,
"loss": 0.9224,
"mean_token_accuracy": 0.7667840108275413,
"num_tokens": 978259078.0,
"step": 11040
},
{
"entropy": 0.8861328125,
"epoch": 1.5102166336363014,
"grad_norm": 0.6759503569151725,
"learning_rate": 1.262857545441736e-06,
"loss": 0.8281,
"mean_token_accuracy": 0.7868098236620427,
"num_tokens": 979110803.0,
"step": 11050
},
{
"entropy": 0.8874755859375,
"epoch": 1.5115834073669103,
"grad_norm": 0.6723906177131002,
"learning_rate": 1.2593349302522193e-06,
"loss": 0.8585,
"mean_token_accuracy": 0.7794128999114036,
"num_tokens": 979995472.0,
"step": 11060
},
{
"entropy": 0.891552734375,
"epoch": 1.5129501810975192,
"grad_norm": 0.6904660296822535,
"learning_rate": 1.2558123150627027e-06,
"loss": 0.8826,
"mean_token_accuracy": 0.7776944287121296,
"num_tokens": 980896411.0,
"step": 11070
},
{
"entropy": 0.90205078125,
"epoch": 1.5143169548281281,
"grad_norm": 0.7176628481522106,
"learning_rate": 1.252289699873186e-06,
"loss": 0.9166,
"mean_token_accuracy": 0.7698030136525631,
"num_tokens": 981782545.0,
"step": 11080
},
{
"entropy": 0.889599609375,
"epoch": 1.515683728558737,
"grad_norm": 0.6944735991026267,
"learning_rate": 1.2487670846836693e-06,
"loss": 0.8857,
"mean_token_accuracy": 0.7738656453788281,
"num_tokens": 982689516.0,
"step": 11090
},
{
"entropy": 0.937744140625,
"epoch": 1.517050502289346,
"grad_norm": 0.7448223578762644,
"learning_rate": 1.2452444694941527e-06,
"loss": 0.949,
"mean_token_accuracy": 0.7632369257509708,
"num_tokens": 983535293.0,
"step": 11100
},
{
"entropy": 0.9007568359375,
"epoch": 1.5184172760199548,
"grad_norm": 0.5663924267985717,
"learning_rate": 1.2417218543046358e-06,
"loss": 0.8699,
"mean_token_accuracy": 0.7765246257185936,
"num_tokens": 984435788.0,
"step": 11110
},
{
"entropy": 0.8890625,
"epoch": 1.5197840497505637,
"grad_norm": 0.6549532580044107,
"learning_rate": 1.2381992391151191e-06,
"loss": 0.8692,
"mean_token_accuracy": 0.7790604948997497,
"num_tokens": 985303807.0,
"step": 11120
},
{
"entropy": 0.892578125,
"epoch": 1.5211508234811726,
"grad_norm": 0.6585395360065014,
"learning_rate": 1.2346766239256025e-06,
"loss": 0.899,
"mean_token_accuracy": 0.7728531070053577,
"num_tokens": 986218767.0,
"step": 11130
},
{
"entropy": 0.8734375,
"epoch": 1.5225175972117815,
"grad_norm": 0.6086054413818591,
"learning_rate": 1.2311540087360858e-06,
"loss": 0.8554,
"mean_token_accuracy": 0.781261432170868,
"num_tokens": 987143778.0,
"step": 11140
},
{
"entropy": 0.868408203125,
"epoch": 1.5238843709423904,
"grad_norm": 0.6739251680577202,
"learning_rate": 1.2276313935465691e-06,
"loss": 0.8368,
"mean_token_accuracy": 0.7812109686434269,
"num_tokens": 988031175.0,
"step": 11150
},
{
"entropy": 0.8728271484375,
"epoch": 1.5252511446729993,
"grad_norm": 0.5913326797942106,
"learning_rate": 1.2241087783570525e-06,
"loss": 0.8448,
"mean_token_accuracy": 0.7846134670078755,
"num_tokens": 988923325.0,
"step": 11160
},
{
"entropy": 0.8970458984375,
"epoch": 1.5266179184036082,
"grad_norm": 0.610322202146902,
"learning_rate": 1.2205861631675358e-06,
"loss": 0.8946,
"mean_token_accuracy": 0.7736282549798489,
"num_tokens": 989811919.0,
"step": 11170
},
{
"entropy": 0.903759765625,
"epoch": 1.527984692134217,
"grad_norm": 0.7112549821555393,
"learning_rate": 1.217063547978019e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7737945802509785,
"num_tokens": 990679973.0,
"step": 11180
},
{
"entropy": 0.8777587890625,
"epoch": 1.529351465864826,
"grad_norm": 0.597340457473233,
"learning_rate": 1.2135409327885023e-06,
"loss": 0.865,
"mean_token_accuracy": 0.7786882184445858,
"num_tokens": 991562442.0,
"step": 11190
},
{
"entropy": 0.859814453125,
"epoch": 1.5307182395954348,
"grad_norm": 0.5922455909152896,
"learning_rate": 1.2100183175989856e-06,
"loss": 0.8419,
"mean_token_accuracy": 0.782082162052393,
"num_tokens": 992429060.0,
"step": 11200
},
{
"entropy": 0.845556640625,
"epoch": 1.5320850133260437,
"grad_norm": 0.8617921421854747,
"learning_rate": 1.206495702409469e-06,
"loss": 0.8319,
"mean_token_accuracy": 0.7861634030938148,
"num_tokens": 993299078.0,
"step": 11210
},
{
"entropy": 0.8489990234375,
"epoch": 1.5334517870566526,
"grad_norm": 0.6816374939952295,
"learning_rate": 1.202973087219952e-06,
"loss": 0.8275,
"mean_token_accuracy": 0.7856946356594563,
"num_tokens": 994183754.0,
"step": 11220
},
{
"entropy": 0.87255859375,
"epoch": 1.5348185607872615,
"grad_norm": 0.5951491836366938,
"learning_rate": 1.1994504720304354e-06,
"loss": 0.8701,
"mean_token_accuracy": 0.7798071287572383,
"num_tokens": 995090732.0,
"step": 11230
},
{
"entropy": 0.868359375,
"epoch": 1.5361853345178704,
"grad_norm": 0.7262020498632568,
"learning_rate": 1.1959278568409187e-06,
"loss": 0.8345,
"mean_token_accuracy": 0.7825100533664227,
"num_tokens": 996004572.0,
"step": 11240
},
{
"entropy": 0.868994140625,
"epoch": 1.5375521082484793,
"grad_norm": 0.6871092898560237,
"learning_rate": 1.192405241651402e-06,
"loss": 0.8476,
"mean_token_accuracy": 0.7814276769757271,
"num_tokens": 996890955.0,
"step": 11250
},
{
"entropy": 0.873828125,
"epoch": 1.5389188819790882,
"grad_norm": 0.6129891215584968,
"learning_rate": 1.1888826264618854e-06,
"loss": 0.8657,
"mean_token_accuracy": 0.7777640290558339,
"num_tokens": 997769284.0,
"step": 11260
},
{
"entropy": 0.876171875,
"epoch": 1.540285655709697,
"grad_norm": 0.6251094156960004,
"learning_rate": 1.1853600112723687e-06,
"loss": 0.8501,
"mean_token_accuracy": 0.7825486585497856,
"num_tokens": 998677050.0,
"step": 11270
},
{
"entropy": 0.8638671875,
"epoch": 1.541652429440306,
"grad_norm": 0.6699665709992728,
"learning_rate": 1.181837396082852e-06,
"loss": 0.8419,
"mean_token_accuracy": 0.7841992788016796,
"num_tokens": 999575420.0,
"step": 11280
},
{
"entropy": 0.8545654296875,
"epoch": 1.543019203170915,
"grad_norm": 0.5967691613766036,
"learning_rate": 1.1783147808933352e-06,
"loss": 0.8495,
"mean_token_accuracy": 0.7835985727608203,
"num_tokens": 1000467865.0,
"step": 11290
},
{
"entropy": 0.944677734375,
"epoch": 1.5443859769015238,
"grad_norm": 0.6534988044006271,
"learning_rate": 1.1747921657038185e-06,
"loss": 0.9476,
"mean_token_accuracy": 0.7651697643101215,
"num_tokens": 1001343857.0,
"step": 11300
},
{
"entropy": 0.8619140625,
"epoch": 1.5457527506321327,
"grad_norm": 0.573152478548352,
"learning_rate": 1.1712695505143019e-06,
"loss": 0.8301,
"mean_token_accuracy": 0.7838553510606289,
"num_tokens": 1002206031.0,
"step": 11310
},
{
"entropy": 0.8789794921875,
"epoch": 1.5471195243627416,
"grad_norm": 0.6357242268736055,
"learning_rate": 1.1677469353247852e-06,
"loss": 0.8519,
"mean_token_accuracy": 0.7807807184755802,
"num_tokens": 1003046032.0,
"step": 11320
},
{
"entropy": 0.9027587890625,
"epoch": 1.5484862980933507,
"grad_norm": 0.6169287502112748,
"learning_rate": 1.1642243201352685e-06,
"loss": 0.8957,
"mean_token_accuracy": 0.771472591906786,
"num_tokens": 1003968005.0,
"step": 11330
},
{
"entropy": 0.874755859375,
"epoch": 1.5498530718239596,
"grad_norm": 0.6446646635528585,
"learning_rate": 1.1607017049457519e-06,
"loss": 0.8493,
"mean_token_accuracy": 0.7805164344608784,
"num_tokens": 1004874914.0,
"step": 11340
},
{
"entropy": 0.908544921875,
"epoch": 1.5512198455545685,
"grad_norm": 0.6386352356976869,
"learning_rate": 1.1571790897562352e-06,
"loss": 0.9016,
"mean_token_accuracy": 0.7721534669399261,
"num_tokens": 1005735910.0,
"step": 11350
},
{
"entropy": 0.848828125,
"epoch": 1.5525866192851774,
"grad_norm": 0.6410988136866715,
"learning_rate": 1.1536564745667183e-06,
"loss": 0.8364,
"mean_token_accuracy": 0.7829449519515037,
"num_tokens": 1006600794.0,
"step": 11360
},
{
"entropy": 0.8791015625,
"epoch": 1.5539533930157863,
"grad_norm": 0.6927974061448415,
"learning_rate": 1.1501338593772017e-06,
"loss": 0.8412,
"mean_token_accuracy": 0.7829448252916336,
"num_tokens": 1007467493.0,
"step": 11370
},
{
"entropy": 0.889501953125,
"epoch": 1.5553201667463952,
"grad_norm": 0.7119710851322546,
"learning_rate": 1.146611244187685e-06,
"loss": 0.8775,
"mean_token_accuracy": 0.7782140858471394,
"num_tokens": 1008379489.0,
"step": 11380
},
{
"entropy": 0.900146484375,
"epoch": 1.556686940477004,
"grad_norm": 0.636655678888353,
"learning_rate": 1.1430886289981683e-06,
"loss": 0.9012,
"mean_token_accuracy": 0.7706985451281071,
"num_tokens": 1009259295.0,
"step": 11390
},
{
"entropy": 0.86005859375,
"epoch": 1.558053714207613,
"grad_norm": 0.6326456636648298,
"learning_rate": 1.1395660138086517e-06,
"loss": 0.8185,
"mean_token_accuracy": 0.7858917094767094,
"num_tokens": 1010088513.0,
"step": 11400
},
{
"entropy": 0.833544921875,
"epoch": 1.5594204879382219,
"grad_norm": 0.6456844694804781,
"learning_rate": 1.136043398619135e-06,
"loss": 0.8307,
"mean_token_accuracy": 0.7873340487480164,
"num_tokens": 1010964392.0,
"step": 11410
},
{
"entropy": 0.92666015625,
"epoch": 1.5607872616688307,
"grad_norm": 0.69164787960966,
"learning_rate": 1.1325207834296184e-06,
"loss": 0.9188,
"mean_token_accuracy": 0.7667729690670967,
"num_tokens": 1011863256.0,
"step": 11420
},
{
"entropy": 0.85166015625,
"epoch": 1.5621540353994396,
"grad_norm": 0.6147450924220452,
"learning_rate": 1.1289981682401017e-06,
"loss": 0.8236,
"mean_token_accuracy": 0.7869856923818588,
"num_tokens": 1012715405.0,
"step": 11430
},
{
"entropy": 0.881884765625,
"epoch": 1.5635208091300485,
"grad_norm": 0.5800619800486037,
"learning_rate": 1.1254755530505848e-06,
"loss": 0.8505,
"mean_token_accuracy": 0.7803284093737602,
"num_tokens": 1013602800.0,
"step": 11440
},
{
"entropy": 0.906787109375,
"epoch": 1.5648875828606574,
"grad_norm": 0.5736618250250684,
"learning_rate": 1.1219529378610681e-06,
"loss": 0.8867,
"mean_token_accuracy": 0.772984130680561,
"num_tokens": 1014477602.0,
"step": 11450
},
{
"entropy": 0.882177734375,
"epoch": 1.5662543565912663,
"grad_norm": 0.6369654079742901,
"learning_rate": 1.1184303226715515e-06,
"loss": 0.8699,
"mean_token_accuracy": 0.7788866408169269,
"num_tokens": 1015376894.0,
"step": 11460
},
{
"entropy": 0.9174560546875,
"epoch": 1.5676211303218752,
"grad_norm": 0.6518316345035471,
"learning_rate": 1.1149077074820346e-06,
"loss": 0.9045,
"mean_token_accuracy": 0.7698356963694095,
"num_tokens": 1016269694.0,
"step": 11470
},
{
"entropy": 0.924609375,
"epoch": 1.5689879040524841,
"grad_norm": 0.6733935027755287,
"learning_rate": 1.111385092292518e-06,
"loss": 0.9295,
"mean_token_accuracy": 0.7677959211170673,
"num_tokens": 1017136473.0,
"step": 11480
},
{
"entropy": 0.86640625,
"epoch": 1.570354677783093,
"grad_norm": 0.5875306166529677,
"learning_rate": 1.1078624771030013e-06,
"loss": 0.8632,
"mean_token_accuracy": 0.7803583785891532,
"num_tokens": 1018025066.0,
"step": 11490
},
{
"entropy": 0.8994140625,
"epoch": 1.571721451513702,
"grad_norm": 0.6086244564346265,
"learning_rate": 1.1043398619134846e-06,
"loss": 0.8917,
"mean_token_accuracy": 0.7740721426904201,
"num_tokens": 1018946433.0,
"step": 11500
},
{
"entropy": 0.902685546875,
"epoch": 1.5730882252443108,
"grad_norm": 0.6226353337970392,
"learning_rate": 1.100817246723968e-06,
"loss": 0.8847,
"mean_token_accuracy": 0.7744199648499489,
"num_tokens": 1019818403.0,
"step": 11510
},
{
"entropy": 0.8796875,
"epoch": 1.5744549989749197,
"grad_norm": 0.6728966966874322,
"learning_rate": 1.0972946315344513e-06,
"loss": 0.8968,
"mean_token_accuracy": 0.774585147947073,
"num_tokens": 1020734420.0,
"step": 11520
},
{
"entropy": 0.8718017578125,
"epoch": 1.5758217727055286,
"grad_norm": 0.6010966368301275,
"learning_rate": 1.0937720163449346e-06,
"loss": 0.8563,
"mean_token_accuracy": 0.7826667279005051,
"num_tokens": 1021610608.0,
"step": 11530
},
{
"entropy": 0.871435546875,
"epoch": 1.5771885464361375,
"grad_norm": 0.612790051016904,
"learning_rate": 1.0902494011554177e-06,
"loss": 0.8615,
"mean_token_accuracy": 0.7790874943137169,
"num_tokens": 1022507878.0,
"step": 11540
},
{
"entropy": 0.852099609375,
"epoch": 1.5785553201667464,
"grad_norm": 0.6474245213101412,
"learning_rate": 1.086726785965901e-06,
"loss": 0.8366,
"mean_token_accuracy": 0.7858131222426892,
"num_tokens": 1023437587.0,
"step": 11550
},
{
"entropy": 0.91689453125,
"epoch": 1.5799220938973553,
"grad_norm": 0.660923174345203,
"learning_rate": 1.0832041707763844e-06,
"loss": 0.9053,
"mean_token_accuracy": 0.7718389116227626,
"num_tokens": 1024328404.0,
"step": 11560
},
{
"entropy": 0.8814453125,
"epoch": 1.5812888676279642,
"grad_norm": 0.60720281378498,
"learning_rate": 1.0796815555868678e-06,
"loss": 0.8857,
"mean_token_accuracy": 0.7770010598003865,
"num_tokens": 1025278391.0,
"step": 11570
},
{
"entropy": 0.9017578125,
"epoch": 1.582655641358573,
"grad_norm": 0.6033541561465195,
"learning_rate": 1.076158940397351e-06,
"loss": 0.9165,
"mean_token_accuracy": 0.7693294674158097,
"num_tokens": 1026166802.0,
"step": 11580
},
{
"entropy": 0.888525390625,
"epoch": 1.584022415089182,
"grad_norm": 0.5996849074167735,
"learning_rate": 1.0726363252078344e-06,
"loss": 0.8822,
"mean_token_accuracy": 0.7769663088023663,
"num_tokens": 1027024240.0,
"step": 11590
},
{
"entropy": 0.8583740234375,
"epoch": 1.5853891888197909,
"grad_norm": 0.569594818436103,
"learning_rate": 1.0691137100183178e-06,
"loss": 0.8293,
"mean_token_accuracy": 0.7856789931654931,
"num_tokens": 1027904455.0,
"step": 11600
},
{
"entropy": 0.8603515625,
"epoch": 1.5867559625503997,
"grad_norm": 0.6576811777835115,
"learning_rate": 1.065591094828801e-06,
"loss": 0.8681,
"mean_token_accuracy": 0.7792346000671386,
"num_tokens": 1028833540.0,
"step": 11610
},
{
"entropy": 0.8716796875,
"epoch": 1.5881227362810086,
"grad_norm": 0.6736883278672954,
"learning_rate": 1.0620684796392842e-06,
"loss": 0.8405,
"mean_token_accuracy": 0.7819972522556782,
"num_tokens": 1029655720.0,
"step": 11620
},
{
"entropy": 0.8271240234375,
"epoch": 1.5894895100116175,
"grad_norm": 0.5786092562027514,
"learning_rate": 1.0585458644497676e-06,
"loss": 0.801,
"mean_token_accuracy": 0.792154909670353,
"num_tokens": 1030505941.0,
"step": 11630
},
{
"entropy": 0.915869140625,
"epoch": 1.5908562837422264,
"grad_norm": 0.6313629668073789,
"learning_rate": 1.0550232492602509e-06,
"loss": 0.8896,
"mean_token_accuracy": 0.7745629355311394,
"num_tokens": 1031370313.0,
"step": 11640
},
{
"entropy": 0.89140625,
"epoch": 1.5922230574728353,
"grad_norm": 0.5922256823036701,
"learning_rate": 1.0515006340707342e-06,
"loss": 0.8596,
"mean_token_accuracy": 0.7791000813245773,
"num_tokens": 1032243138.0,
"step": 11650
},
{
"entropy": 0.860791015625,
"epoch": 1.5935898312034442,
"grad_norm": 0.6567745930672148,
"learning_rate": 1.0479780188812176e-06,
"loss": 0.8341,
"mean_token_accuracy": 0.7848434187471867,
"num_tokens": 1033081238.0,
"step": 11660
},
{
"entropy": 0.870361328125,
"epoch": 1.5949566049340531,
"grad_norm": 0.6357472207048299,
"learning_rate": 1.0444554036917009e-06,
"loss": 0.8606,
"mean_token_accuracy": 0.7800371676683426,
"num_tokens": 1033975506.0,
"step": 11670
},
{
"entropy": 0.912939453125,
"epoch": 1.5963233786646622,
"grad_norm": 0.6303436051000985,
"learning_rate": 1.0409327885021842e-06,
"loss": 0.9181,
"mean_token_accuracy": 0.7674969054758549,
"num_tokens": 1034879253.0,
"step": 11680
},
{
"entropy": 0.851513671875,
"epoch": 1.5976901523952711,
"grad_norm": 0.6473305885664726,
"learning_rate": 1.0374101733126674e-06,
"loss": 0.8456,
"mean_token_accuracy": 0.7848185129463673,
"num_tokens": 1035764543.0,
"step": 11690
},
{
"entropy": 0.8580810546875,
"epoch": 1.59905692612588,
"grad_norm": 0.6994656256010295,
"learning_rate": 1.0338875581231507e-06,
"loss": 0.81,
"mean_token_accuracy": 0.7882751733064651,
"num_tokens": 1036597658.0,
"step": 11700
},
{
"entropy": 0.9119140625,
"epoch": 1.600423699856489,
"grad_norm": 0.6433741610042779,
"learning_rate": 1.030364942933634e-06,
"loss": 0.8938,
"mean_token_accuracy": 0.7727463014423848,
"num_tokens": 1037461178.0,
"step": 11710
},
{
"entropy": 0.87578125,
"epoch": 1.6017904735870978,
"grad_norm": 0.5449401696593634,
"learning_rate": 1.0268423277441174e-06,
"loss": 0.8952,
"mean_token_accuracy": 0.7723553106188774,
"num_tokens": 1038320543.0,
"step": 11720
},
{
"entropy": 0.873095703125,
"epoch": 1.6031572473177067,
"grad_norm": 0.6466679996801727,
"learning_rate": 1.0233197125546005e-06,
"loss": 0.8393,
"mean_token_accuracy": 0.7821230553090572,
"num_tokens": 1039158322.0,
"step": 11730
},
{
"entropy": 0.886181640625,
"epoch": 1.6045240210483156,
"grad_norm": 0.652136608517785,
"learning_rate": 1.0197970973650838e-06,
"loss": 0.8779,
"mean_token_accuracy": 0.776769807934761,
"num_tokens": 1040044664.0,
"step": 11740
},
{
"entropy": 0.8616943359375,
"epoch": 1.6058907947789245,
"grad_norm": 0.6920456343689589,
"learning_rate": 1.0162744821755674e-06,
"loss": 0.8557,
"mean_token_accuracy": 0.782530790567398,
"num_tokens": 1040933733.0,
"step": 11750
},
{
"entropy": 0.8561279296875,
"epoch": 1.6072575685095334,
"grad_norm": 0.6523697324377878,
"learning_rate": 1.0127518669860505e-06,
"loss": 0.8548,
"mean_token_accuracy": 0.7817596890032291,
"num_tokens": 1041821739.0,
"step": 11760
},
{
"entropy": 0.8979736328125,
"epoch": 1.6086243422401423,
"grad_norm": 0.713992435525201,
"learning_rate": 1.0092292517965338e-06,
"loss": 0.9102,
"mean_token_accuracy": 0.772839991748333,
"num_tokens": 1042734726.0,
"step": 11770
},
{
"entropy": 0.898486328125,
"epoch": 1.6099911159707512,
"grad_norm": 0.6522050546785709,
"learning_rate": 1.0057066366070172e-06,
"loss": 0.8823,
"mean_token_accuracy": 0.7755483098328113,
"num_tokens": 1043605674.0,
"step": 11780
},
{
"entropy": 0.854296875,
"epoch": 1.61135788970136,
"grad_norm": 0.8458991446652431,
"learning_rate": 1.0021840214175005e-06,
"loss": 0.8342,
"mean_token_accuracy": 0.7845298103988171,
"num_tokens": 1044489733.0,
"step": 11790
},
{
"entropy": 0.884130859375,
"epoch": 1.612724663431969,
"grad_norm": 0.7191892528411045,
"learning_rate": 9.986614062279836e-07,
"loss": 0.8746,
"mean_token_accuracy": 0.7798868030309677,
"num_tokens": 1045386114.0,
"step": 11800
},
{
"entropy": 0.848828125,
"epoch": 1.6140914371625779,
"grad_norm": 0.6141203262313625,
"learning_rate": 9.95138791038467e-07,
"loss": 0.8187,
"mean_token_accuracy": 0.7872417323291302,
"num_tokens": 1046281529.0,
"step": 11810
},
{
"entropy": 0.85576171875,
"epoch": 1.6154582108931868,
"grad_norm": 0.6085109834719004,
"learning_rate": 9.916161758489503e-07,
"loss": 0.8372,
"mean_token_accuracy": 0.7851675182580948,
"num_tokens": 1047172601.0,
"step": 11820
},
{
"entropy": 0.89560546875,
"epoch": 1.6168249846237956,
"grad_norm": 0.6183151842716972,
"learning_rate": 9.880935606594336e-07,
"loss": 0.8897,
"mean_token_accuracy": 0.7772857986390591,
"num_tokens": 1048111551.0,
"step": 11830
},
{
"entropy": 0.8646240234375,
"epoch": 1.6181917583544045,
"grad_norm": 0.5896541119269015,
"learning_rate": 9.84570945469917e-07,
"loss": 0.8616,
"mean_token_accuracy": 0.7821886517107487,
"num_tokens": 1048993658.0,
"step": 11840
},
{
"entropy": 0.9060546875,
"epoch": 1.6195585320850134,
"grad_norm": 0.6412904851242678,
"learning_rate": 9.810483302804003e-07,
"loss": 0.8826,
"mean_token_accuracy": 0.7733226165175437,
"num_tokens": 1049881721.0,
"step": 11850
},
{
"entropy": 0.8951171875,
"epoch": 1.6209253058156223,
"grad_norm": 0.6734985776361488,
"learning_rate": 9.775257150908836e-07,
"loss": 0.859,
"mean_token_accuracy": 0.777108097076416,
"num_tokens": 1050738642.0,
"step": 11860
},
{
"entropy": 0.854736328125,
"epoch": 1.6222920795462312,
"grad_norm": 0.6694491202047714,
"learning_rate": 9.740030999013668e-07,
"loss": 0.8607,
"mean_token_accuracy": 0.7841218486428261,
"num_tokens": 1051607144.0,
"step": 11870
},
{
"entropy": 0.8619384765625,
"epoch": 1.6236588532768401,
"grad_norm": 0.600869046100873,
"learning_rate": 9.7048048471185e-07,
"loss": 0.8502,
"mean_token_accuracy": 0.7803348951041699,
"num_tokens": 1052460566.0,
"step": 11880
},
{
"entropy": 0.890185546875,
"epoch": 1.625025627007449,
"grad_norm": 0.5405481033599301,
"learning_rate": 9.669578695223334e-07,
"loss": 0.8943,
"mean_token_accuracy": 0.773001367598772,
"num_tokens": 1053364568.0,
"step": 11890
},
{
"entropy": 0.8716796875,
"epoch": 1.626392400738058,
"grad_norm": 0.630049546171816,
"learning_rate": 9.634352543328168e-07,
"loss": 0.8556,
"mean_token_accuracy": 0.7795073382556439,
"num_tokens": 1054222654.0,
"step": 11900
},
{
"entropy": 0.8807861328125,
"epoch": 1.6277591744686668,
"grad_norm": 0.672812245586794,
"learning_rate": 9.599126391433e-07,
"loss": 0.8722,
"mean_token_accuracy": 0.7790414713323116,
"num_tokens": 1055112095.0,
"step": 11910
},
{
"entropy": 0.898876953125,
"epoch": 1.6291259481992757,
"grad_norm": 0.6576808694630778,
"learning_rate": 9.563900239537834e-07,
"loss": 0.8797,
"mean_token_accuracy": 0.7753811351954937,
"num_tokens": 1055990084.0,
"step": 11920
},
{
"entropy": 0.8605712890625,
"epoch": 1.6304927219298846,
"grad_norm": 0.6589641691540885,
"learning_rate": 9.528674087642667e-07,
"loss": 0.8544,
"mean_token_accuracy": 0.7834944441914559,
"num_tokens": 1056906052.0,
"step": 11930
},
{
"entropy": 0.841650390625,
"epoch": 1.6318594956604935,
"grad_norm": 0.6525554444051017,
"learning_rate": 9.4934479357475e-07,
"loss": 0.8172,
"mean_token_accuracy": 0.7892064653337002,
"num_tokens": 1057804689.0,
"step": 11940
},
{
"entropy": 0.922314453125,
"epoch": 1.6332262693911024,
"grad_norm": 0.5936187187533785,
"learning_rate": 9.458221783852332e-07,
"loss": 0.8972,
"mean_token_accuracy": 0.7710883520543576,
"num_tokens": 1058720254.0,
"step": 11950
},
{
"entropy": 0.875927734375,
"epoch": 1.6345930431217113,
"grad_norm": 0.6427208745003881,
"learning_rate": 9.422995631957166e-07,
"loss": 0.8504,
"mean_token_accuracy": 0.7829153761267662,
"num_tokens": 1059561064.0,
"step": 11960
},
{
"entropy": 0.88125,
"epoch": 1.6359598168523202,
"grad_norm": 0.6804565281102498,
"learning_rate": 9.387769480061999e-07,
"loss": 0.8793,
"mean_token_accuracy": 0.7756216943264007,
"num_tokens": 1060429575.0,
"step": 11970
},
{
"entropy": 0.8974609375,
"epoch": 1.637326590582929,
"grad_norm": 0.6320187727227752,
"learning_rate": 9.352543328166831e-07,
"loss": 0.862,
"mean_token_accuracy": 0.7772472649812698,
"num_tokens": 1061352006.0,
"step": 11980
},
{
"entropy": 0.882177734375,
"epoch": 1.638693364313538,
"grad_norm": 0.5751693648570979,
"learning_rate": 9.317317176271665e-07,
"loss": 0.8625,
"mean_token_accuracy": 0.7802124932408333,
"num_tokens": 1062196979.0,
"step": 11990
},
{
"entropy": 0.9099609375,
"epoch": 1.6400601380441469,
"grad_norm": 0.5444241601657916,
"learning_rate": 9.282091024376498e-07,
"loss": 0.9253,
"mean_token_accuracy": 0.7681181281805038,
"num_tokens": 1063136300.0,
"step": 12000
},
{
"entropy": 0.919482421875,
"epoch": 1.6414269117747557,
"grad_norm": 0.72139177907399,
"learning_rate": 9.246864872481331e-07,
"loss": 0.9094,
"mean_token_accuracy": 0.7699120812118053,
"num_tokens": 1063984611.0,
"step": 12010
},
{
"entropy": 0.8716064453125,
"epoch": 1.6427936855053646,
"grad_norm": 0.6051930966307631,
"learning_rate": 9.211638720586164e-07,
"loss": 0.847,
"mean_token_accuracy": 0.7827926233410836,
"num_tokens": 1064831535.0,
"step": 12020
},
{
"entropy": 0.876708984375,
"epoch": 1.6441604592359735,
"grad_norm": 0.6838887001745763,
"learning_rate": 9.176412568690997e-07,
"loss": 0.832,
"mean_token_accuracy": 0.7849361278116703,
"num_tokens": 1065725334.0,
"step": 12030
},
{
"entropy": 0.8777099609375,
"epoch": 1.6455272329665824,
"grad_norm": 0.6752762245632044,
"learning_rate": 9.14118641679583e-07,
"loss": 0.8866,
"mean_token_accuracy": 0.7705355636775494,
"num_tokens": 1066626508.0,
"step": 12040
},
{
"entropy": 0.874365234375,
"epoch": 1.6468940066971913,
"grad_norm": 0.6672950035253807,
"learning_rate": 9.105960264900663e-07,
"loss": 0.8655,
"mean_token_accuracy": 0.7789167106151581,
"num_tokens": 1067460843.0,
"step": 12050
},
{
"entropy": 0.858740234375,
"epoch": 1.6482607804278002,
"grad_norm": 0.6131594298930778,
"learning_rate": 9.070734113005496e-07,
"loss": 0.869,
"mean_token_accuracy": 0.7828976139426231,
"num_tokens": 1068399394.0,
"step": 12060
},
{
"entropy": 0.912939453125,
"epoch": 1.6496275541584091,
"grad_norm": 0.6464349326535827,
"learning_rate": 9.035507961110329e-07,
"loss": 0.9177,
"mean_token_accuracy": 0.7658012449741364,
"num_tokens": 1069269064.0,
"step": 12070
},
{
"entropy": 0.87421875,
"epoch": 1.650994327889018,
"grad_norm": 0.6576783163059311,
"learning_rate": 9.000281809215163e-07,
"loss": 0.8603,
"mean_token_accuracy": 0.7821370333433151,
"num_tokens": 1070151947.0,
"step": 12080
},
{
"entropy": 0.8677734375,
"epoch": 1.652361101619627,
"grad_norm": 0.6470125328046743,
"learning_rate": 8.965055657319995e-07,
"loss": 0.8227,
"mean_token_accuracy": 0.7890266194939614,
"num_tokens": 1071027874.0,
"step": 12090
},
{
"entropy": 0.8745849609375,
"epoch": 1.6537278753502358,
"grad_norm": 0.5852357542368867,
"learning_rate": 8.929829505424828e-07,
"loss": 0.8652,
"mean_token_accuracy": 0.7785955846309662,
"num_tokens": 1071914954.0,
"step": 12100
},
{
"entropy": 0.87216796875,
"epoch": 1.6550946490808447,
"grad_norm": 0.5943157835418952,
"learning_rate": 8.894603353529662e-07,
"loss": 0.8503,
"mean_token_accuracy": 0.7798246741294861,
"num_tokens": 1072794976.0,
"step": 12110
},
{
"entropy": 0.8575927734375,
"epoch": 1.6564614228114536,
"grad_norm": 0.5808386457001793,
"learning_rate": 8.859377201634495e-07,
"loss": 0.8357,
"mean_token_accuracy": 0.7852286137640476,
"num_tokens": 1073716171.0,
"step": 12120
},
{
"entropy": 0.92001953125,
"epoch": 1.6578281965420625,
"grad_norm": 0.5957710004607907,
"learning_rate": 8.824151049739326e-07,
"loss": 0.9147,
"mean_token_accuracy": 0.7672912120819092,
"num_tokens": 1074611877.0,
"step": 12130
},
{
"entropy": 0.8490234375,
"epoch": 1.6591949702726714,
"grad_norm": 0.6287461450778391,
"learning_rate": 8.788924897844161e-07,
"loss": 0.8346,
"mean_token_accuracy": 0.7865989118814468,
"num_tokens": 1075528862.0,
"step": 12140
},
{
"entropy": 0.859716796875,
"epoch": 1.6605617440032803,
"grad_norm": 0.6601491317942232,
"learning_rate": 8.753698745948994e-07,
"loss": 0.836,
"mean_token_accuracy": 0.7837707214057446,
"num_tokens": 1076437769.0,
"step": 12150
},
{
"entropy": 0.86025390625,
"epoch": 1.6619285177338892,
"grad_norm": 0.6712968272785057,
"learning_rate": 8.718472594053825e-07,
"loss": 0.8366,
"mean_token_accuracy": 0.7816597737371922,
"num_tokens": 1077282147.0,
"step": 12160
},
{
"entropy": 0.90771484375,
"epoch": 1.663295291464498,
"grad_norm": 0.6608155691442885,
"learning_rate": 8.683246442158659e-07,
"loss": 0.8552,
"mean_token_accuracy": 0.7790877275168896,
"num_tokens": 1078141716.0,
"step": 12170
},
{
"entropy": 0.8826416015625,
"epoch": 1.664662065195107,
"grad_norm": 0.6197918906597017,
"learning_rate": 8.648020290263492e-07,
"loss": 0.8328,
"mean_token_accuracy": 0.7836525142192841,
"num_tokens": 1078987019.0,
"step": 12180
},
{
"entropy": 0.866748046875,
"epoch": 1.6660288389257158,
"grad_norm": 0.6370517981936552,
"learning_rate": 8.612794138368325e-07,
"loss": 0.8594,
"mean_token_accuracy": 0.7808619849383831,
"num_tokens": 1079883292.0,
"step": 12190
},
{
"entropy": 0.888720703125,
"epoch": 1.6673956126563247,
"grad_norm": 0.6712891314967837,
"learning_rate": 8.577567986473158e-07,
"loss": 0.8765,
"mean_token_accuracy": 0.7772718377411365,
"num_tokens": 1080734797.0,
"step": 12200
},
{
"entropy": 0.86845703125,
"epoch": 1.6687623863869336,
"grad_norm": 0.6529782184769779,
"learning_rate": 8.542341834577991e-07,
"loss": 0.8437,
"mean_token_accuracy": 0.7833823926746846,
"num_tokens": 1081602784.0,
"step": 12210
},
{
"entropy": 0.869189453125,
"epoch": 1.6701291601175425,
"grad_norm": 0.6467798017572669,
"learning_rate": 8.507115682682824e-07,
"loss": 0.8404,
"mean_token_accuracy": 0.783482625335455,
"num_tokens": 1082468814.0,
"step": 12220
},
{
"entropy": 0.9113037109375,
"epoch": 1.6714959338481514,
"grad_norm": 0.694471723441864,
"learning_rate": 8.471889530787658e-07,
"loss": 0.9266,
"mean_token_accuracy": 0.7680496253073216,
"num_tokens": 1083378491.0,
"step": 12230
},
{
"entropy": 0.87509765625,
"epoch": 1.6728627075787603,
"grad_norm": 0.6241550418994637,
"learning_rate": 8.43666337889249e-07,
"loss": 0.8618,
"mean_token_accuracy": 0.7821087643504143,
"num_tokens": 1084256436.0,
"step": 12240
},
{
"entropy": 0.8412841796875,
"epoch": 1.6742294813093692,
"grad_norm": 0.5513133248402139,
"learning_rate": 8.401437226997323e-07,
"loss": 0.8452,
"mean_token_accuracy": 0.7869075298309326,
"num_tokens": 1085167492.0,
"step": 12250
},
{
"entropy": 0.875830078125,
"epoch": 1.675596255039978,
"grad_norm": 0.6599587437944319,
"learning_rate": 8.366211075102157e-07,
"loss": 0.8485,
"mean_token_accuracy": 0.7814508520066739,
"num_tokens": 1085994348.0,
"step": 12260
},
{
"entropy": 0.86416015625,
"epoch": 1.676963028770587,
"grad_norm": 0.6393215785253491,
"learning_rate": 8.330984923206989e-07,
"loss": 0.8356,
"mean_token_accuracy": 0.7825547091662883,
"num_tokens": 1086909354.0,
"step": 12270
},
{
"entropy": 0.893408203125,
"epoch": 1.678329802501196,
"grad_norm": 0.6377579353319165,
"learning_rate": 8.295758771311822e-07,
"loss": 0.9088,
"mean_token_accuracy": 0.7730342917144298,
"num_tokens": 1087731177.0,
"step": 12280
},
{
"entropy": 0.877490234375,
"epoch": 1.6796965762318048,
"grad_norm": 0.5655064403514389,
"learning_rate": 8.260532619416656e-07,
"loss": 0.8768,
"mean_token_accuracy": 0.7770364142954349,
"num_tokens": 1088587754.0,
"step": 12290
},
{
"entropy": 0.8691650390625,
"epoch": 1.6810633499624137,
"grad_norm": 0.5996043966857054,
"learning_rate": 8.225306467521489e-07,
"loss": 0.854,
"mean_token_accuracy": 0.7816752053797245,
"num_tokens": 1089479095.0,
"step": 12300
},
{
"entropy": 0.8730712890625,
"epoch": 1.6824301236930226,
"grad_norm": 0.6205422445354924,
"learning_rate": 8.190080315626321e-07,
"loss": 0.8665,
"mean_token_accuracy": 0.7807982131838799,
"num_tokens": 1090337044.0,
"step": 12310
},
{
"entropy": 0.878515625,
"epoch": 1.6837968974236315,
"grad_norm": 0.6179606813031737,
"learning_rate": 8.154854163731155e-07,
"loss": 0.8562,
"mean_token_accuracy": 0.7794875994324684,
"num_tokens": 1091244361.0,
"step": 12320
},
{
"entropy": 0.89482421875,
"epoch": 1.6851636711542404,
"grad_norm": 0.6839782740997792,
"learning_rate": 8.119628011835988e-07,
"loss": 0.8825,
"mean_token_accuracy": 0.7742877453565598,
"num_tokens": 1092118184.0,
"step": 12330
},
{
"entropy": 0.86845703125,
"epoch": 1.6865304448848493,
"grad_norm": 0.5972049785180878,
"learning_rate": 8.084401859940821e-07,
"loss": 0.8531,
"mean_token_accuracy": 0.7796293377876282,
"num_tokens": 1092980010.0,
"step": 12340
},
{
"entropy": 0.894384765625,
"epoch": 1.6878972186154582,
"grad_norm": 0.6378263425527652,
"learning_rate": 8.049175708045654e-07,
"loss": 0.9036,
"mean_token_accuracy": 0.7726514115929604,
"num_tokens": 1093871943.0,
"step": 12350
},
{
"entropy": 0.8744140625,
"epoch": 1.689263992346067,
"grad_norm": 0.5741415390415991,
"learning_rate": 8.013949556150487e-07,
"loss": 0.8312,
"mean_token_accuracy": 0.7839008167386055,
"num_tokens": 1094708515.0,
"step": 12360
},
{
"entropy": 0.8676513671875,
"epoch": 1.690630766076676,
"grad_norm": 0.6634693032033321,
"learning_rate": 7.97872340425532e-07,
"loss": 0.8481,
"mean_token_accuracy": 0.7818403616547585,
"num_tokens": 1095605283.0,
"step": 12370
},
{
"entropy": 0.89326171875,
"epoch": 1.6919975398072848,
"grad_norm": 0.7426725449031375,
"learning_rate": 7.943497252360153e-07,
"loss": 0.8844,
"mean_token_accuracy": 0.7738823764026165,
"num_tokens": 1096471521.0,
"step": 12380
},
{
"entropy": 0.914453125,
"epoch": 1.6933643135378937,
"grad_norm": 0.6475663963340884,
"learning_rate": 7.908271100464986e-07,
"loss": 0.9038,
"mean_token_accuracy": 0.7697284333407879,
"num_tokens": 1097380895.0,
"step": 12390
},
{
"entropy": 0.901708984375,
"epoch": 1.6947310872685026,
"grad_norm": 0.6933591691102208,
"learning_rate": 7.873044948569819e-07,
"loss": 0.9015,
"mean_token_accuracy": 0.775365948677063,
"num_tokens": 1098264769.0,
"step": 12400
},
{
"entropy": 0.89189453125,
"epoch": 1.6960978609991115,
"grad_norm": 0.6222753619371514,
"learning_rate": 7.837818796674653e-07,
"loss": 0.8637,
"mean_token_accuracy": 0.7783260501921176,
"num_tokens": 1099166534.0,
"step": 12410
},
{
"entropy": 0.87548828125,
"epoch": 1.6974646347297204,
"grad_norm": 0.7046997317140495,
"learning_rate": 7.802592644779484e-07,
"loss": 0.8591,
"mean_token_accuracy": 0.7785611361265182,
"num_tokens": 1100092895.0,
"step": 12420
},
{
"entropy": 0.89755859375,
"epoch": 1.6988314084603293,
"grad_norm": 0.7124594858544343,
"learning_rate": 7.767366492884317e-07,
"loss": 0.8966,
"mean_token_accuracy": 0.7729496449232102,
"num_tokens": 1100961965.0,
"step": 12430
},
{
"entropy": 0.872119140625,
"epoch": 1.7001981821909382,
"grad_norm": 0.662471431646776,
"learning_rate": 7.732140340989152e-07,
"loss": 0.8803,
"mean_token_accuracy": 0.7756528481841087,
"num_tokens": 1101887589.0,
"step": 12440
},
{
"entropy": 0.891845703125,
"epoch": 1.701564955921547,
"grad_norm": 0.597561053713222,
"learning_rate": 7.696914189093985e-07,
"loss": 0.8682,
"mean_token_accuracy": 0.7784192897379398,
"num_tokens": 1102768436.0,
"step": 12450
},
{
"entropy": 0.8700439453125,
"epoch": 1.702931729652156,
"grad_norm": 0.5624836330334663,
"learning_rate": 7.661688037198816e-07,
"loss": 0.8497,
"mean_token_accuracy": 0.78179931640625,
"num_tokens": 1103662460.0,
"step": 12460
},
{
"entropy": 0.8533935546875,
"epoch": 1.704298503382765,
"grad_norm": 0.8868289459628534,
"learning_rate": 7.62646188530365e-07,
"loss": 0.8372,
"mean_token_accuracy": 0.7851487189531327,
"num_tokens": 1104539373.0,
"step": 12470
},
{
"entropy": 0.838037109375,
"epoch": 1.7056652771133738,
"grad_norm": 0.6082339729241616,
"learning_rate": 7.591235733408483e-07,
"loss": 0.8247,
"mean_token_accuracy": 0.784604874253273,
"num_tokens": 1105434657.0,
"step": 12480
},
{
"entropy": 0.843115234375,
"epoch": 1.7070320508439827,
"grad_norm": 0.6685484825221335,
"learning_rate": 7.556009581513315e-07,
"loss": 0.8189,
"mean_token_accuracy": 0.7872738040983677,
"num_tokens": 1106311018.0,
"step": 12490
},
{
"entropy": 0.839794921875,
"epoch": 1.7083988245745916,
"grad_norm": 0.6385554220073073,
"learning_rate": 7.520783429618149e-07,
"loss": 0.827,
"mean_token_accuracy": 0.7882353827357292,
"num_tokens": 1107229488.0,
"step": 12500
},
{
"entropy": 0.84580078125,
"epoch": 1.7097655983052005,
"grad_norm": 0.6064936382573342,
"learning_rate": 7.485557277722982e-07,
"loss": 0.8451,
"mean_token_accuracy": 0.7848631627857685,
"num_tokens": 1108129586.0,
"step": 12510
},
{
"entropy": 0.871923828125,
"epoch": 1.7111323720358094,
"grad_norm": 0.5726556265564926,
"learning_rate": 7.450331125827815e-07,
"loss": 0.8598,
"mean_token_accuracy": 0.7793069384992123,
"num_tokens": 1109025467.0,
"step": 12520
},
{
"entropy": 0.878466796875,
"epoch": 1.7124991457664183,
"grad_norm": 0.651738139212967,
"learning_rate": 7.415104973932648e-07,
"loss": 0.8634,
"mean_token_accuracy": 0.7784644268453121,
"num_tokens": 1109928602.0,
"step": 12530
},
{
"entropy": 0.8448974609375,
"epoch": 1.7138659194970272,
"grad_norm": 0.6703659091059984,
"learning_rate": 7.379878822037481e-07,
"loss": 0.8238,
"mean_token_accuracy": 0.787165955454111,
"num_tokens": 1110759388.0,
"step": 12540
},
{
"entropy": 0.898046875,
"epoch": 1.715232693227636,
"grad_norm": 0.6489465435800134,
"learning_rate": 7.344652670142314e-07,
"loss": 0.8817,
"mean_token_accuracy": 0.7741785563528538,
"num_tokens": 1111652033.0,
"step": 12550
},
{
"entropy": 0.8843994140625,
"epoch": 1.716599466958245,
"grad_norm": 0.620978043042264,
"learning_rate": 7.309426518247147e-07,
"loss": 0.8615,
"mean_token_accuracy": 0.7773758247494698,
"num_tokens": 1112559338.0,
"step": 12560
},
{
"entropy": 0.894970703125,
"epoch": 1.7179662406888538,
"grad_norm": 0.5948733443442683,
"learning_rate": 7.27420036635198e-07,
"loss": 0.8736,
"mean_token_accuracy": 0.7786666020751,
"num_tokens": 1113422543.0,
"step": 12570
},
{
"entropy": 0.8997314453125,
"epoch": 1.7193330144194627,
"grad_norm": 0.6186192844767272,
"learning_rate": 7.238974214456813e-07,
"loss": 0.883,
"mean_token_accuracy": 0.7746313631534576,
"num_tokens": 1114320790.0,
"step": 12580
},
{
"entropy": 0.87744140625,
"epoch": 1.7206997881500716,
"grad_norm": 0.6153246382495627,
"learning_rate": 7.203748062561647e-07,
"loss": 0.8501,
"mean_token_accuracy": 0.7832189902663231,
"num_tokens": 1115279578.0,
"step": 12590
},
{
"entropy": 0.8755859375,
"epoch": 1.7220665618806805,
"grad_norm": 0.6677713336304814,
"learning_rate": 7.168521910666479e-07,
"loss": 0.861,
"mean_token_accuracy": 0.7813778266310691,
"num_tokens": 1116158991.0,
"step": 12600
},
{
"entropy": 0.88662109375,
"epoch": 1.7234333356112894,
"grad_norm": 0.6176807520672073,
"learning_rate": 7.133295758771312e-07,
"loss": 0.8838,
"mean_token_accuracy": 0.7760683447122574,
"num_tokens": 1117041491.0,
"step": 12610
},
{
"entropy": 0.866015625,
"epoch": 1.7248001093418983,
"grad_norm": 0.6486242915321803,
"learning_rate": 7.098069606876146e-07,
"loss": 0.8352,
"mean_token_accuracy": 0.7847947396337986,
"num_tokens": 1117918004.0,
"step": 12620
},
{
"entropy": 0.841650390625,
"epoch": 1.7261668830725072,
"grad_norm": 0.7063487973465534,
"learning_rate": 7.062843454980979e-07,
"loss": 0.8316,
"mean_token_accuracy": 0.7871500171720982,
"num_tokens": 1118785040.0,
"step": 12630
},
{
"entropy": 0.8598876953125,
"epoch": 1.727533656803116,
"grad_norm": 0.5805543435133351,
"learning_rate": 7.027617303085811e-07,
"loss": 0.8391,
"mean_token_accuracy": 0.7842237673699856,
"num_tokens": 1119691878.0,
"step": 12640
},
{
"entropy": 0.85048828125,
"epoch": 1.728900430533725,
"grad_norm": 0.6543026544346401,
"learning_rate": 6.992391151190645e-07,
"loss": 0.8517,
"mean_token_accuracy": 0.7842083856463432,
"num_tokens": 1120588565.0,
"step": 12650
},
{
"entropy": 0.9030029296875,
"epoch": 1.730267204264334,
"grad_norm": 0.7080863076266929,
"learning_rate": 6.957164999295478e-07,
"loss": 0.8886,
"mean_token_accuracy": 0.7737496547400952,
"num_tokens": 1121432713.0,
"step": 12660
},
{
"entropy": 0.8896484375,
"epoch": 1.7316339779949428,
"grad_norm": 0.6216982961118652,
"learning_rate": 6.921938847400309e-07,
"loss": 0.8489,
"mean_token_accuracy": 0.7786870442330838,
"num_tokens": 1122273894.0,
"step": 12670
},
{
"entropy": 0.86923828125,
"epoch": 1.7330007517255517,
"grad_norm": 0.6120767441227928,
"learning_rate": 6.886712695505144e-07,
"loss": 0.8532,
"mean_token_accuracy": 0.7817772842943669,
"num_tokens": 1123187375.0,
"step": 12680
},
{
"entropy": 0.905322265625,
"epoch": 1.7343675254561606,
"grad_norm": 0.6279302280162432,
"learning_rate": 6.851486543609977e-07,
"loss": 0.9254,
"mean_token_accuracy": 0.7689151346683503,
"num_tokens": 1124085975.0,
"step": 12690
},
{
"entropy": 0.91171875,
"epoch": 1.7357342991867695,
"grad_norm": 1.0164972746608998,
"learning_rate": 6.816260391714811e-07,
"loss": 0.8856,
"mean_token_accuracy": 0.7767547070980072,
"num_tokens": 1124957939.0,
"step": 12700
},
{
"entropy": 0.87783203125,
"epoch": 1.7371010729173784,
"grad_norm": 0.6752542824702497,
"learning_rate": 6.781034239819642e-07,
"loss": 0.8855,
"mean_token_accuracy": 0.7757141493260861,
"num_tokens": 1125845835.0,
"step": 12710
},
{
"entropy": 0.86025390625,
"epoch": 1.7384678466479875,
"grad_norm": 0.6066185510880894,
"learning_rate": 6.745808087924475e-07,
"loss": 0.8395,
"mean_token_accuracy": 0.7836493082344532,
"num_tokens": 1126726139.0,
"step": 12720
},
{
"entropy": 0.8966796875,
"epoch": 1.7398346203785964,
"grad_norm": 0.6444453997842056,
"learning_rate": 6.710581936029308e-07,
"loss": 0.8851,
"mean_token_accuracy": 0.7794846341013908,
"num_tokens": 1127591299.0,
"step": 12730
},
{
"entropy": 0.856787109375,
"epoch": 1.7412013941092053,
"grad_norm": 0.6136324728752277,
"learning_rate": 6.675355784134143e-07,
"loss": 0.8502,
"mean_token_accuracy": 0.7820504263043404,
"num_tokens": 1128505092.0,
"step": 12740
},
{
"entropy": 0.8770751953125,
"epoch": 1.7425681678398142,
"grad_norm": 0.626898470236672,
"learning_rate": 6.640129632238974e-07,
"loss": 0.8571,
"mean_token_accuracy": 0.7808243274688721,
"num_tokens": 1129395781.0,
"step": 12750
},
{
"entropy": 0.890771484375,
"epoch": 1.743934941570423,
"grad_norm": 0.6003078297628306,
"learning_rate": 6.604903480343807e-07,
"loss": 0.9027,
"mean_token_accuracy": 0.7707201808691024,
"num_tokens": 1130301530.0,
"step": 12760
},
{
"entropy": 0.8599609375,
"epoch": 1.745301715301032,
"grad_norm": 0.6900270891039557,
"learning_rate": 6.569677328448641e-07,
"loss": 0.848,
"mean_token_accuracy": 0.7834755666553974,
"num_tokens": 1131161646.0,
"step": 12770
},
{
"entropy": 0.8471923828125,
"epoch": 1.7466684890316408,
"grad_norm": 0.6229983310952886,
"learning_rate": 6.534451176553473e-07,
"loss": 0.8157,
"mean_token_accuracy": 0.7885446555912494,
"num_tokens": 1132051345.0,
"step": 12780
},
{
"entropy": 0.87158203125,
"epoch": 1.7480352627622497,
"grad_norm": 0.5753599122280351,
"learning_rate": 6.499225024658306e-07,
"loss": 0.8579,
"mean_token_accuracy": 0.7798839941620826,
"num_tokens": 1132962042.0,
"step": 12790
},
{
"entropy": 0.88193359375,
"epoch": 1.7494020364928586,
"grad_norm": 0.6157410937981205,
"learning_rate": 6.46399887276314e-07,
"loss": 0.8421,
"mean_token_accuracy": 0.7835871770977973,
"num_tokens": 1133802402.0,
"step": 12800
},
{
"entropy": 0.8994384765625,
"epoch": 1.7507688102234675,
"grad_norm": 0.7523232798002717,
"learning_rate": 6.428772720867973e-07,
"loss": 0.8854,
"mean_token_accuracy": 0.7747002005577087,
"num_tokens": 1134670099.0,
"step": 12810
},
{
"entropy": 0.8614013671875,
"epoch": 1.7521355839540764,
"grad_norm": 0.6405771896127452,
"learning_rate": 6.393546568972805e-07,
"loss": 0.8532,
"mean_token_accuracy": 0.7814978487789631,
"num_tokens": 1135525749.0,
"step": 12820
},
{
"entropy": 0.891455078125,
"epoch": 1.7535023576846853,
"grad_norm": 0.6518997756242859,
"learning_rate": 6.358320417077639e-07,
"loss": 0.8698,
"mean_token_accuracy": 0.7766164131462574,
"num_tokens": 1136407810.0,
"step": 12830
},
{
"entropy": 0.870654296875,
"epoch": 1.7548691314152942,
"grad_norm": 0.630743083470493,
"learning_rate": 6.323094265182472e-07,
"loss": 0.8564,
"mean_token_accuracy": 0.7804528072476387,
"num_tokens": 1137249574.0,
"step": 12840
},
{
"entropy": 0.843359375,
"epoch": 1.756235905145903,
"grad_norm": 0.5663431106600659,
"learning_rate": 6.287868113287306e-07,
"loss": 0.8146,
"mean_token_accuracy": 0.7875705815851688,
"num_tokens": 1138164235.0,
"step": 12850
},
{
"entropy": 0.86005859375,
"epoch": 1.757602678876512,
"grad_norm": 0.6008001719313958,
"learning_rate": 6.252641961392138e-07,
"loss": 0.8613,
"mean_token_accuracy": 0.7812321692705154,
"num_tokens": 1139067214.0,
"step": 12860
},
{
"entropy": 0.8353271484375,
"epoch": 1.758969452607121,
"grad_norm": 0.6195897186203309,
"learning_rate": 6.217415809496971e-07,
"loss": 0.8321,
"mean_token_accuracy": 0.7875038295984268,
"num_tokens": 1139953795.0,
"step": 12870
},
{
"entropy": 0.87939453125,
"epoch": 1.7603362263377298,
"grad_norm": 0.6343027368544522,
"learning_rate": 6.182189657601804e-07,
"loss": 0.8994,
"mean_token_accuracy": 0.7764708869159221,
"num_tokens": 1140862640.0,
"step": 12880
},
{
"entropy": 0.8677490234375,
"epoch": 1.7617030000683387,
"grad_norm": 0.6412298940023232,
"learning_rate": 6.146963505706637e-07,
"loss": 0.8451,
"mean_token_accuracy": 0.7848143734037876,
"num_tokens": 1141743254.0,
"step": 12890
},
{
"entropy": 0.8748046875,
"epoch": 1.7630697737989476,
"grad_norm": 0.5901632713300002,
"learning_rate": 6.11173735381147e-07,
"loss": 0.8683,
"mean_token_accuracy": 0.7782444164156914,
"num_tokens": 1142600515.0,
"step": 12900
},
{
"entropy": 0.90634765625,
"epoch": 1.7644365475295565,
"grad_norm": 0.5516697563912818,
"learning_rate": 6.076511201916304e-07,
"loss": 0.8924,
"mean_token_accuracy": 0.7770768634974956,
"num_tokens": 1143475412.0,
"step": 12910
},
{
"entropy": 0.87236328125,
"epoch": 1.7658033212601654,
"grad_norm": 0.7005388212807965,
"learning_rate": 6.041285050021136e-07,
"loss": 0.8568,
"mean_token_accuracy": 0.7795761942863464,
"num_tokens": 1144302196.0,
"step": 12920
},
{
"entropy": 0.9015869140625,
"epoch": 1.7671700949907743,
"grad_norm": 0.5638590768277398,
"learning_rate": 6.006058898125969e-07,
"loss": 0.8928,
"mean_token_accuracy": 0.7733306311070919,
"num_tokens": 1145232557.0,
"step": 12930
},
{
"entropy": 0.89228515625,
"epoch": 1.7685368687213832,
"grad_norm": 0.6440338837826973,
"learning_rate": 5.970832746230803e-07,
"loss": 0.897,
"mean_token_accuracy": 0.7761762537062168,
"num_tokens": 1146194324.0,
"step": 12940
},
{
"entropy": 0.880322265625,
"epoch": 1.769903642451992,
"grad_norm": 0.6487329485526836,
"learning_rate": 5.935606594335636e-07,
"loss": 0.8616,
"mean_token_accuracy": 0.7793267570436001,
"num_tokens": 1147126259.0,
"step": 12950
},
{
"entropy": 0.878564453125,
"epoch": 1.771270416182601,
"grad_norm": 0.6175105564152714,
"learning_rate": 5.900380442440468e-07,
"loss": 0.881,
"mean_token_accuracy": 0.7776012472808361,
"num_tokens": 1147998971.0,
"step": 12960
},
{
"entropy": 0.91513671875,
"epoch": 1.7726371899132098,
"grad_norm": 0.61928039260753,
"learning_rate": 5.8651542905453e-07,
"loss": 0.92,
"mean_token_accuracy": 0.766179695725441,
"num_tokens": 1148904341.0,
"step": 12970
},
{
"entropy": 0.864306640625,
"epoch": 1.7740039636438187,
"grad_norm": 0.657690916843788,
"learning_rate": 5.829928138650135e-07,
"loss": 0.837,
"mean_token_accuracy": 0.7867551848292351,
"num_tokens": 1149783289.0,
"step": 12980
},
{
"entropy": 0.8633544921875,
"epoch": 1.7753707373744276,
"grad_norm": 0.5746443060237572,
"learning_rate": 5.794701986754967e-07,
"loss": 0.8478,
"mean_token_accuracy": 0.7826368264853955,
"num_tokens": 1150669938.0,
"step": 12990
},
{
"entropy": 0.900244140625,
"epoch": 1.7767375111050365,
"grad_norm": 0.5996103518396152,
"learning_rate": 5.759475834859801e-07,
"loss": 0.8826,
"mean_token_accuracy": 0.7778916575014592,
"num_tokens": 1151559575.0,
"step": 13000
},
{
"entropy": 0.91806640625,
"epoch": 1.7781042848356454,
"grad_norm": 0.664560993971928,
"learning_rate": 5.724249682964633e-07,
"loss": 0.8981,
"mean_token_accuracy": 0.772744408249855,
"num_tokens": 1152414438.0,
"step": 13010
},
{
"entropy": 0.8822509765625,
"epoch": 1.7794710585662543,
"grad_norm": 0.6817025704277182,
"learning_rate": 5.689023531069466e-07,
"loss": 0.8718,
"mean_token_accuracy": 0.7773973047733307,
"num_tokens": 1153345079.0,
"step": 13020
},
{
"entropy": 0.87958984375,
"epoch": 1.7808378322968632,
"grad_norm": 0.6061692631954559,
"learning_rate": 5.6537973791743e-07,
"loss": 0.8799,
"mean_token_accuracy": 0.778936629742384,
"num_tokens": 1154226611.0,
"step": 13030
},
{
"entropy": 0.856396484375,
"epoch": 1.782204606027472,
"grad_norm": 0.6770291886788563,
"learning_rate": 5.618571227279133e-07,
"loss": 0.8417,
"mean_token_accuracy": 0.7843591861426831,
"num_tokens": 1155145591.0,
"step": 13040
},
{
"entropy": 0.83046875,
"epoch": 1.783571379758081,
"grad_norm": 0.6652301323062134,
"learning_rate": 5.583345075383965e-07,
"loss": 0.833,
"mean_token_accuracy": 0.7852985866367816,
"num_tokens": 1156065067.0,
"step": 13050
},
{
"entropy": 0.862841796875,
"epoch": 1.78493815348869,
"grad_norm": 0.7086383115548384,
"learning_rate": 5.548118923488799e-07,
"loss": 0.8229,
"mean_token_accuracy": 0.7877714663743973,
"num_tokens": 1156911394.0,
"step": 13060
},
{
"entropy": 0.87255859375,
"epoch": 1.786304927219299,
"grad_norm": 0.6220520983588382,
"learning_rate": 5.512892771593632e-07,
"loss": 0.8477,
"mean_token_accuracy": 0.780919574201107,
"num_tokens": 1157807466.0,
"step": 13070
},
{
"entropy": 0.92333984375,
"epoch": 1.787671700949908,
"grad_norm": 0.5757849003757352,
"learning_rate": 5.477666619698464e-07,
"loss": 0.9105,
"mean_token_accuracy": 0.7707650043070317,
"num_tokens": 1158708008.0,
"step": 13080
},
{
"entropy": 0.8928955078125,
"epoch": 1.7890384746805168,
"grad_norm": 0.6068999127898511,
"learning_rate": 5.442440467803298e-07,
"loss": 0.8654,
"mean_token_accuracy": 0.778403778374195,
"num_tokens": 1159583725.0,
"step": 13090
},
{
"entropy": 0.88828125,
"epoch": 1.7904052484111257,
"grad_norm": 0.752886745008753,
"learning_rate": 5.407214315908131e-07,
"loss": 0.8664,
"mean_token_accuracy": 0.7771011278033256,
"num_tokens": 1160491493.0,
"step": 13100
},
{
"entropy": 0.863818359375,
"epoch": 1.7917720221417346,
"grad_norm": 0.5931876250045046,
"learning_rate": 5.371988164012964e-07,
"loss": 0.8696,
"mean_token_accuracy": 0.7762885741889477,
"num_tokens": 1161409970.0,
"step": 13110
},
{
"entropy": 0.8439208984375,
"epoch": 1.7931387958723435,
"grad_norm": 0.5970590370168406,
"learning_rate": 5.336762012117797e-07,
"loss": 0.8242,
"mean_token_accuracy": 0.7867983095347881,
"num_tokens": 1162313347.0,
"step": 13120
},
{
"entropy": 0.888427734375,
"epoch": 1.7945055696029524,
"grad_norm": 0.6452449737993415,
"learning_rate": 5.30153586022263e-07,
"loss": 0.8855,
"mean_token_accuracy": 0.7758688434958458,
"num_tokens": 1163220881.0,
"step": 13130
},
{
"entropy": 0.877734375,
"epoch": 1.7958723433335613,
"grad_norm": 0.574020620771317,
"learning_rate": 5.266309708327462e-07,
"loss": 0.8522,
"mean_token_accuracy": 0.7797754444181919,
"num_tokens": 1164094809.0,
"step": 13140
},
{
"entropy": 0.85732421875,
"epoch": 1.7972391170641702,
"grad_norm": 0.6037787709609526,
"learning_rate": 5.231083556432296e-07,
"loss": 0.8384,
"mean_token_accuracy": 0.7874787151813507,
"num_tokens": 1164971338.0,
"step": 13150
},
{
"entropy": 0.9025390625,
"epoch": 1.798605890794779,
"grad_norm": 0.6325527736909728,
"learning_rate": 5.195857404537129e-07,
"loss": 0.9316,
"mean_token_accuracy": 0.7716842234134674,
"num_tokens": 1165850482.0,
"step": 13160
},
{
"entropy": 0.8919921875,
"epoch": 1.799972664525388,
"grad_norm": 0.6312894090327134,
"learning_rate": 5.160631252641961e-07,
"loss": 0.855,
"mean_token_accuracy": 0.7807838283479214,
"num_tokens": 1166706374.0,
"step": 13170
},
{
"entropy": 0.86376953125,
"epoch": 1.8013394382559968,
"grad_norm": 0.9174005622628827,
"learning_rate": 5.125405100746795e-07,
"loss": 0.833,
"mean_token_accuracy": 0.7853959977626801,
"num_tokens": 1167624146.0,
"step": 13180
},
{
"entropy": 0.881494140625,
"epoch": 1.8027062119866057,
"grad_norm": 0.644707800769561,
"learning_rate": 5.090178948851628e-07,
"loss": 0.8952,
"mean_token_accuracy": 0.7748947650194168,
"num_tokens": 1168541396.0,
"step": 13190
},
{
"entropy": 0.906103515625,
"epoch": 1.8040729857172146,
"grad_norm": 0.6537508456618294,
"learning_rate": 5.054952796956461e-07,
"loss": 0.8969,
"mean_token_accuracy": 0.7712951928377152,
"num_tokens": 1169402817.0,
"step": 13200
},
{
"entropy": 0.8541748046875,
"epoch": 1.8054397594478235,
"grad_norm": 0.64719012522013,
"learning_rate": 5.019726645061294e-07,
"loss": 0.837,
"mean_token_accuracy": 0.7865447029471397,
"num_tokens": 1170317297.0,
"step": 13210
},
{
"entropy": 0.89970703125,
"epoch": 1.8068065331784324,
"grad_norm": 0.6846857557005489,
"learning_rate": 4.984500493166127e-07,
"loss": 0.8985,
"mean_token_accuracy": 0.7755084104835988,
"num_tokens": 1171217665.0,
"step": 13220
},
{
"entropy": 0.834912109375,
"epoch": 1.8081733069090413,
"grad_norm": 0.6536632843586978,
"learning_rate": 4.94927434127096e-07,
"loss": 0.8085,
"mean_token_accuracy": 0.7920351237058639,
"num_tokens": 1172087543.0,
"step": 13230
},
{
"entropy": 0.8439453125,
"epoch": 1.8095400806396502,
"grad_norm": 0.702345923300384,
"learning_rate": 4.914048189375794e-07,
"loss": 0.8493,
"mean_token_accuracy": 0.7821595445275307,
"num_tokens": 1172967896.0,
"step": 13240
},
{
"entropy": 0.8876953125,
"epoch": 1.8109068543702591,
"grad_norm": 0.6685311962864883,
"learning_rate": 4.878822037480626e-07,
"loss": 0.8512,
"mean_token_accuracy": 0.7795375771820545,
"num_tokens": 1173765974.0,
"step": 13250
},
{
"entropy": 0.86455078125,
"epoch": 1.812273628100868,
"grad_norm": 0.5989072903124874,
"learning_rate": 4.843595885585459e-07,
"loss": 0.8642,
"mean_token_accuracy": 0.7790997169911862,
"num_tokens": 1174621145.0,
"step": 13260
},
{
"entropy": 0.8712890625,
"epoch": 1.813640401831477,
"grad_norm": 0.7548576168266844,
"learning_rate": 4.808369733690292e-07,
"loss": 0.879,
"mean_token_accuracy": 0.7781397521495819,
"num_tokens": 1175529749.0,
"step": 13270
},
{
"entropy": 0.9130859375,
"epoch": 1.8150071755620858,
"grad_norm": 0.6560977197960545,
"learning_rate": 4.773143581795125e-07,
"loss": 0.9176,
"mean_token_accuracy": 0.7702094197273255,
"num_tokens": 1176461343.0,
"step": 13280
},
{
"entropy": 0.823876953125,
"epoch": 1.8163739492926947,
"grad_norm": 0.6121989190525112,
"learning_rate": 4.7379174298999583e-07,
"loss": 0.7985,
"mean_token_accuracy": 0.7905144296586514,
"num_tokens": 1177334280.0,
"step": 13290
},
{
"entropy": 0.85302734375,
"epoch": 1.8177407230233036,
"grad_norm": 0.6409881334996248,
"learning_rate": 4.702691278004791e-07,
"loss": 0.8421,
"mean_token_accuracy": 0.7844751797616482,
"num_tokens": 1178258174.0,
"step": 13300
},
{
"entropy": 0.8578369140625,
"epoch": 1.8191074967539125,
"grad_norm": 0.7366951404956733,
"learning_rate": 4.6674651261096245e-07,
"loss": 0.856,
"mean_token_accuracy": 0.781606062501669,
"num_tokens": 1179152856.0,
"step": 13310
},
{
"entropy": 0.844970703125,
"epoch": 1.8204742704845214,
"grad_norm": 0.6443275955579284,
"learning_rate": 4.632238974214457e-07,
"loss": 0.8232,
"mean_token_accuracy": 0.7876503251492977,
"num_tokens": 1180048625.0,
"step": 13320
},
{
"entropy": 0.840283203125,
"epoch": 1.8218410442151303,
"grad_norm": 0.5737295420408239,
"learning_rate": 4.5970128223192907e-07,
"loss": 0.8217,
"mean_token_accuracy": 0.7876095399260521,
"num_tokens": 1180938231.0,
"step": 13330
},
{
"entropy": 0.83173828125,
"epoch": 1.8232078179457392,
"grad_norm": 0.586541446046978,
"learning_rate": 4.561786670424123e-07,
"loss": 0.8089,
"mean_token_accuracy": 0.7895360514521599,
"num_tokens": 1181813930.0,
"step": 13340
},
{
"entropy": 0.88125,
"epoch": 1.824574591676348,
"grad_norm": 0.6636349983378554,
"learning_rate": 4.5265605185289563e-07,
"loss": 0.8757,
"mean_token_accuracy": 0.7802544936537743,
"num_tokens": 1182727375.0,
"step": 13350
},
{
"entropy": 0.870556640625,
"epoch": 1.825941365406957,
"grad_norm": 0.5887210296102945,
"learning_rate": 4.491334366633789e-07,
"loss": 0.843,
"mean_token_accuracy": 0.7828044414520263,
"num_tokens": 1183623201.0,
"step": 13360
},
{
"entropy": 0.8943115234375,
"epoch": 1.8273081391375658,
"grad_norm": 0.6474623031117906,
"learning_rate": 4.456108214738622e-07,
"loss": 0.8754,
"mean_token_accuracy": 0.7779895782470703,
"num_tokens": 1184513685.0,
"step": 13370
},
{
"entropy": 0.861572265625,
"epoch": 1.8286749128681747,
"grad_norm": 0.6707157842141764,
"learning_rate": 4.4208820628434553e-07,
"loss": 0.8456,
"mean_token_accuracy": 0.7822755672037601,
"num_tokens": 1185400077.0,
"step": 13380
},
{
"entropy": 0.903857421875,
"epoch": 1.8300416865987836,
"grad_norm": 0.6284846910932648,
"learning_rate": 4.385655910948288e-07,
"loss": 0.8889,
"mean_token_accuracy": 0.772513072192669,
"num_tokens": 1186291424.0,
"step": 13390
},
{
"entropy": 0.890966796875,
"epoch": 1.8314084603293925,
"grad_norm": 0.6336387033238788,
"learning_rate": 4.3504297590531215e-07,
"loss": 0.902,
"mean_token_accuracy": 0.7734977349638938,
"num_tokens": 1187194562.0,
"step": 13400
},
{
"entropy": 0.845556640625,
"epoch": 1.8327752340600014,
"grad_norm": 0.6311395306992467,
"learning_rate": 4.3152036071579543e-07,
"loss": 0.86,
"mean_token_accuracy": 0.7851853474974633,
"num_tokens": 1188098751.0,
"step": 13410
},
{
"entropy": 0.844091796875,
"epoch": 1.8341420077906103,
"grad_norm": 0.6913883475308271,
"learning_rate": 4.2799774552627877e-07,
"loss": 0.848,
"mean_token_accuracy": 0.7827664114534855,
"num_tokens": 1189005385.0,
"step": 13420
},
{
"entropy": 0.84931640625,
"epoch": 1.8355087815212192,
"grad_norm": 0.7170700811478197,
"learning_rate": 4.2447513033676205e-07,
"loss": 0.848,
"mean_token_accuracy": 0.7819149784743786,
"num_tokens": 1189882489.0,
"step": 13430
},
{
"entropy": 0.869091796875,
"epoch": 1.836875555251828,
"grad_norm": 0.6842771926547755,
"learning_rate": 4.209525151472454e-07,
"loss": 0.8222,
"mean_token_accuracy": 0.7842157624661923,
"num_tokens": 1190750704.0,
"step": 13440
},
{
"entropy": 0.881640625,
"epoch": 1.838242328982437,
"grad_norm": 0.6842152033214174,
"learning_rate": 4.1742989995772867e-07,
"loss": 0.8678,
"mean_token_accuracy": 0.778414761275053,
"num_tokens": 1191660625.0,
"step": 13450
},
{
"entropy": 0.898193359375,
"epoch": 1.839609102713046,
"grad_norm": 0.6271669219059234,
"learning_rate": 4.13907284768212e-07,
"loss": 0.8976,
"mean_token_accuracy": 0.7754556991159915,
"num_tokens": 1192508000.0,
"step": 13460
},
{
"entropy": 0.90205078125,
"epoch": 1.8409758764436548,
"grad_norm": 0.6046698772148272,
"learning_rate": 4.1038466957869523e-07,
"loss": 0.8733,
"mean_token_accuracy": 0.7789119511842728,
"num_tokens": 1193362849.0,
"step": 13470
},
{
"entropy": 0.8619873046875,
"epoch": 1.8423426501742637,
"grad_norm": 0.6087451414696002,
"learning_rate": 4.068620543891785e-07,
"loss": 0.8418,
"mean_token_accuracy": 0.785490658134222,
"num_tokens": 1194316124.0,
"step": 13480
},
{
"entropy": 0.88134765625,
"epoch": 1.8437094239048726,
"grad_norm": 0.6156999182617078,
"learning_rate": 4.0333943919966185e-07,
"loss": 0.8714,
"mean_token_accuracy": 0.7789288111031055,
"num_tokens": 1195255727.0,
"step": 13490
},
{
"entropy": 0.8626953125,
"epoch": 1.8450761976354815,
"grad_norm": 0.6483412378751585,
"learning_rate": 3.9981682401014513e-07,
"loss": 0.8367,
"mean_token_accuracy": 0.7850859962403774,
"num_tokens": 1196182466.0,
"step": 13500
},
{
"entropy": 0.847314453125,
"epoch": 1.8464429713660904,
"grad_norm": 0.6136548170248578,
"learning_rate": 3.9629420882062847e-07,
"loss": 0.8329,
"mean_token_accuracy": 0.785405520349741,
"num_tokens": 1197059992.0,
"step": 13510
},
{
"entropy": 0.8910400390625,
"epoch": 1.8478097450966993,
"grad_norm": 0.6322998090764322,
"learning_rate": 3.9277159363111175e-07,
"loss": 0.8986,
"mean_token_accuracy": 0.7739111892879009,
"num_tokens": 1197948995.0,
"step": 13520
},
{
"entropy": 0.8666015625,
"epoch": 1.8491765188273082,
"grad_norm": 0.6342165954796632,
"learning_rate": 3.892489784415951e-07,
"loss": 0.8698,
"mean_token_accuracy": 0.7786829575896264,
"num_tokens": 1198824386.0,
"step": 13530
},
{
"entropy": 0.88544921875,
"epoch": 1.850543292557917,
"grad_norm": 0.6671267749048717,
"learning_rate": 3.8572636325207837e-07,
"loss": 0.8838,
"mean_token_accuracy": 0.7771362036466598,
"num_tokens": 1199644308.0,
"step": 13540
},
{
"entropy": 0.901318359375,
"epoch": 1.851910066288526,
"grad_norm": 0.6511734643198841,
"learning_rate": 3.822037480625617e-07,
"loss": 0.8791,
"mean_token_accuracy": 0.7758109159767628,
"num_tokens": 1200529892.0,
"step": 13550
},
{
"entropy": 0.83779296875,
"epoch": 1.8532768400191348,
"grad_norm": 0.578823941700517,
"learning_rate": 3.78681132873045e-07,
"loss": 0.8216,
"mean_token_accuracy": 0.7866666719317437,
"num_tokens": 1201485843.0,
"step": 13560
},
{
"entropy": 0.88662109375,
"epoch": 1.8546436137497437,
"grad_norm": 0.575199142284016,
"learning_rate": 3.7515851768352827e-07,
"loss": 0.8625,
"mean_token_accuracy": 0.7775881864130497,
"num_tokens": 1202365458.0,
"step": 13570
},
{
"entropy": 0.9112060546875,
"epoch": 1.8560103874803526,
"grad_norm": 0.7778792830205572,
"learning_rate": 3.716359024940116e-07,
"loss": 0.8942,
"mean_token_accuracy": 0.7732493795454503,
"num_tokens": 1203210624.0,
"step": 13580
},
{
"entropy": 0.8962646484375,
"epoch": 1.8573771612109615,
"grad_norm": 0.7231169080239824,
"learning_rate": 3.6811328730449484e-07,
"loss": 0.8777,
"mean_token_accuracy": 0.7759856030344963,
"num_tokens": 1204109870.0,
"step": 13590
},
{
"entropy": 0.88125,
"epoch": 1.8587439349415704,
"grad_norm": 0.6249295777343477,
"learning_rate": 3.645906721149782e-07,
"loss": 0.8531,
"mean_token_accuracy": 0.7808558225631714,
"num_tokens": 1204941911.0,
"step": 13600
},
{
"entropy": 0.8836669921875,
"epoch": 1.8601107086721793,
"grad_norm": 0.6041862348818317,
"learning_rate": 3.6106805692546145e-07,
"loss": 0.869,
"mean_token_accuracy": 0.7789572946727276,
"num_tokens": 1205867891.0,
"step": 13610
},
{
"entropy": 0.8845947265625,
"epoch": 1.8614774824027882,
"grad_norm": 0.6128517239039806,
"learning_rate": 3.575454417359448e-07,
"loss": 0.8674,
"mean_token_accuracy": 0.7783841900527477,
"num_tokens": 1206743074.0,
"step": 13620
},
{
"entropy": 0.864990234375,
"epoch": 1.862844256133397,
"grad_norm": 0.6202851732915954,
"learning_rate": 3.5402282654642807e-07,
"loss": 0.8343,
"mean_token_accuracy": 0.7838932871818542,
"num_tokens": 1207606486.0,
"step": 13630
},
{
"entropy": 0.869140625,
"epoch": 1.864211029864006,
"grad_norm": 0.6438782688248773,
"learning_rate": 3.505002113569114e-07,
"loss": 0.8806,
"mean_token_accuracy": 0.7764981500804424,
"num_tokens": 1208513818.0,
"step": 13640
},
{
"entropy": 0.88828125,
"epoch": 1.865577803594615,
"grad_norm": 0.6741497614997025,
"learning_rate": 3.469775961673947e-07,
"loss": 0.8668,
"mean_token_accuracy": 0.779580508172512,
"num_tokens": 1209364561.0,
"step": 13650
},
{
"entropy": 0.8387451171875,
"epoch": 1.8669445773252238,
"grad_norm": 0.5616816646635658,
"learning_rate": 3.43454980977878e-07,
"loss": 0.8112,
"mean_token_accuracy": 0.7932673089206219,
"num_tokens": 1210238798.0,
"step": 13660
},
{
"entropy": 0.890185546875,
"epoch": 1.8683113510558327,
"grad_norm": 0.674999835918783,
"learning_rate": 3.399323657883613e-07,
"loss": 0.8789,
"mean_token_accuracy": 0.7755057364702225,
"num_tokens": 1211114581.0,
"step": 13670
},
{
"entropy": 0.894970703125,
"epoch": 1.8696781247864416,
"grad_norm": 0.6312831338073221,
"learning_rate": 3.364097505988446e-07,
"loss": 0.8784,
"mean_token_accuracy": 0.7768735095858574,
"num_tokens": 1212024219.0,
"step": 13680
},
{
"entropy": 0.8678466796875,
"epoch": 1.8710448985170505,
"grad_norm": 0.5937214360239252,
"learning_rate": 3.328871354093279e-07,
"loss": 0.8522,
"mean_token_accuracy": 0.7819060996174813,
"num_tokens": 1212946824.0,
"step": 13690
},
{
"entropy": 0.888671875,
"epoch": 1.8724116722476594,
"grad_norm": 0.7754051630641834,
"learning_rate": 3.293645202198112e-07,
"loss": 0.8812,
"mean_token_accuracy": 0.7758187994360923,
"num_tokens": 1213839645.0,
"step": 13700
},
{
"entropy": 0.887890625,
"epoch": 1.8737784459782683,
"grad_norm": 0.6021181764057624,
"learning_rate": 3.2584190503029454e-07,
"loss": 0.8767,
"mean_token_accuracy": 0.7760289460420609,
"num_tokens": 1214692776.0,
"step": 13710
},
{
"entropy": 0.85830078125,
"epoch": 1.8751452197088772,
"grad_norm": 0.6173444889881202,
"learning_rate": 3.223192898407778e-07,
"loss": 0.8437,
"mean_token_accuracy": 0.7866519972681999,
"num_tokens": 1215588902.0,
"step": 13720
},
{
"entropy": 0.858544921875,
"epoch": 1.876511993439486,
"grad_norm": 1.1700983925818609,
"learning_rate": 3.1879667465126116e-07,
"loss": 0.8484,
"mean_token_accuracy": 0.7846133060753345,
"num_tokens": 1216493221.0,
"step": 13730
},
{
"entropy": 0.85419921875,
"epoch": 1.877878767170095,
"grad_norm": 0.5676962448530135,
"learning_rate": 3.152740594617444e-07,
"loss": 0.8465,
"mean_token_accuracy": 0.7853829421103,
"num_tokens": 1217440087.0,
"step": 13740
},
{
"entropy": 0.876123046875,
"epoch": 1.8792455409007038,
"grad_norm": 0.601441002653187,
"learning_rate": 3.117514442722277e-07,
"loss": 0.8674,
"mean_token_accuracy": 0.7770878776907921,
"num_tokens": 1218357098.0,
"step": 13750
},
{
"entropy": 0.8341552734375,
"epoch": 1.8806123146313127,
"grad_norm": 0.5717231004256154,
"learning_rate": 3.08228829082711e-07,
"loss": 0.7981,
"mean_token_accuracy": 0.7945769697427749,
"num_tokens": 1219287510.0,
"step": 13760
},
{
"entropy": 0.8765625,
"epoch": 1.8819790883619216,
"grad_norm": 0.6684715324717744,
"learning_rate": 3.0470621389319434e-07,
"loss": 0.8643,
"mean_token_accuracy": 0.7783019848167896,
"num_tokens": 1220145054.0,
"step": 13770
},
{
"entropy": 0.8861328125,
"epoch": 1.8833458620925305,
"grad_norm": 0.5644858412208444,
"learning_rate": 3.0118359870367763e-07,
"loss": 0.8509,
"mean_token_accuracy": 0.7840838946402073,
"num_tokens": 1220986637.0,
"step": 13780
},
{
"entropy": 0.85791015625,
"epoch": 1.8847126358231394,
"grad_norm": 0.6151869073387672,
"learning_rate": 2.9766098351416096e-07,
"loss": 0.8245,
"mean_token_accuracy": 0.7859044969081879,
"num_tokens": 1221882549.0,
"step": 13790
},
{
"entropy": 0.887939453125,
"epoch": 1.8860794095537483,
"grad_norm": 0.5827998680518758,
"learning_rate": 2.9413836832464424e-07,
"loss": 0.8763,
"mean_token_accuracy": 0.7785092808306218,
"num_tokens": 1222799616.0,
"step": 13800
},
{
"entropy": 0.90791015625,
"epoch": 1.8874461832843572,
"grad_norm": 0.6111397387300581,
"learning_rate": 2.906157531351276e-07,
"loss": 0.9156,
"mean_token_accuracy": 0.7725523419678211,
"num_tokens": 1223692723.0,
"step": 13810
},
{
"entropy": 0.8697265625,
"epoch": 1.888812957014966,
"grad_norm": 0.6666113860270796,
"learning_rate": 2.8709313794561086e-07,
"loss": 0.8581,
"mean_token_accuracy": 0.77978570535779,
"num_tokens": 1224602296.0,
"step": 13820
},
{
"entropy": 0.8912109375,
"epoch": 1.890179730745575,
"grad_norm": 0.6928650422925297,
"learning_rate": 2.8357052275609415e-07,
"loss": 0.8821,
"mean_token_accuracy": 0.7744676426053048,
"num_tokens": 1225444602.0,
"step": 13830
},
{
"entropy": 0.91083984375,
"epoch": 1.8915465044761839,
"grad_norm": 0.6613783822436018,
"learning_rate": 2.8004790756657743e-07,
"loss": 0.8984,
"mean_token_accuracy": 0.775581170618534,
"num_tokens": 1226326865.0,
"step": 13840
},
{
"entropy": 0.865966796875,
"epoch": 1.8929132782067928,
"grad_norm": 0.5537369441189711,
"learning_rate": 2.7652529237706076e-07,
"loss": 0.8361,
"mean_token_accuracy": 0.780582831799984,
"num_tokens": 1227216431.0,
"step": 13850
},
{
"entropy": 0.84921875,
"epoch": 1.8942800519374017,
"grad_norm": 0.6130260410736645,
"learning_rate": 2.7300267718754405e-07,
"loss": 0.8243,
"mean_token_accuracy": 0.7882960870862007,
"num_tokens": 1228073184.0,
"step": 13860
},
{
"entropy": 0.812255859375,
"epoch": 1.8956468256680106,
"grad_norm": 0.5972060380671202,
"learning_rate": 2.694800619980274e-07,
"loss": 0.7844,
"mean_token_accuracy": 0.7968228377401829,
"num_tokens": 1228961273.0,
"step": 13870
},
{
"entropy": 0.862109375,
"epoch": 1.8970135993986195,
"grad_norm": 0.7092554096581005,
"learning_rate": 2.6595744680851066e-07,
"loss": 0.865,
"mean_token_accuracy": 0.7789915844798088,
"num_tokens": 1229885119.0,
"step": 13880
},
{
"entropy": 0.906201171875,
"epoch": 1.8983803731292284,
"grad_norm": 0.6040524892516147,
"learning_rate": 2.6243483161899395e-07,
"loss": 0.8928,
"mean_token_accuracy": 0.7752930887043477,
"num_tokens": 1230756421.0,
"step": 13890
},
{
"entropy": 0.8772705078125,
"epoch": 1.8997471468598373,
"grad_norm": 0.6204203728215769,
"learning_rate": 2.589122164294773e-07,
"loss": 0.8702,
"mean_token_accuracy": 0.7776052705943585,
"num_tokens": 1231686900.0,
"step": 13900
},
{
"entropy": 0.8646240234375,
"epoch": 1.9011139205904461,
"grad_norm": 0.646778051818254,
"learning_rate": 2.5538960123996056e-07,
"loss": 0.8693,
"mean_token_accuracy": 0.7815115749835968,
"num_tokens": 1232568957.0,
"step": 13910
},
{
"entropy": 0.89462890625,
"epoch": 1.902480694321055,
"grad_norm": 0.6803021554382997,
"learning_rate": 2.518669860504439e-07,
"loss": 0.8627,
"mean_token_accuracy": 0.7783646024763584,
"num_tokens": 1233407782.0,
"step": 13920
},
{
"entropy": 0.879541015625,
"epoch": 1.903847468051664,
"grad_norm": 0.651214189736717,
"learning_rate": 2.483443708609272e-07,
"loss": 0.8636,
"mean_token_accuracy": 0.7799850106239319,
"num_tokens": 1234277152.0,
"step": 13930
},
{
"entropy": 0.844677734375,
"epoch": 1.9052142417822728,
"grad_norm": 0.775199062426369,
"learning_rate": 2.4482175567141046e-07,
"loss": 0.849,
"mean_token_accuracy": 0.7839196600019932,
"num_tokens": 1235185409.0,
"step": 13940
},
{
"entropy": 0.86875,
"epoch": 1.9065810155128817,
"grad_norm": 0.6237268503502167,
"learning_rate": 2.4129914048189375e-07,
"loss": 0.8708,
"mean_token_accuracy": 0.7789345040917397,
"num_tokens": 1236117547.0,
"step": 13950
},
{
"entropy": 0.89130859375,
"epoch": 1.9079477892434906,
"grad_norm": 0.6004899089843776,
"learning_rate": 2.3777652529237708e-07,
"loss": 0.8767,
"mean_token_accuracy": 0.779845853894949,
"num_tokens": 1237019384.0,
"step": 13960
},
{
"entropy": 0.9169921875,
"epoch": 1.9093145629740995,
"grad_norm": 0.6550984848772777,
"learning_rate": 2.342539101028604e-07,
"loss": 0.9191,
"mean_token_accuracy": 0.769962540268898,
"num_tokens": 1237946995.0,
"step": 13970
},
{
"entropy": 0.922216796875,
"epoch": 1.9106813367047084,
"grad_norm": 0.6771974792441479,
"learning_rate": 2.307312949133437e-07,
"loss": 0.9128,
"mean_token_accuracy": 0.7719332106411457,
"num_tokens": 1238820204.0,
"step": 13980
},
{
"entropy": 0.8789306640625,
"epoch": 1.9120481104353173,
"grad_norm": 0.6021162719829456,
"learning_rate": 2.2720867972382698e-07,
"loss": 0.8681,
"mean_token_accuracy": 0.7801465056836605,
"num_tokens": 1239710908.0,
"step": 13990
},
{
"entropy": 0.87880859375,
"epoch": 1.9134148841659262,
"grad_norm": 0.5863192430585343,
"learning_rate": 2.236860645343103e-07,
"loss": 0.871,
"mean_token_accuracy": 0.7748402863740921,
"num_tokens": 1240579007.0,
"step": 14000
},
{
"entropy": 0.85517578125,
"epoch": 1.914781657896535,
"grad_norm": 0.7166902257737592,
"learning_rate": 2.201634493447936e-07,
"loss": 0.8173,
"mean_token_accuracy": 0.7842565469443799,
"num_tokens": 1241438818.0,
"step": 14010
},
{
"entropy": 0.875048828125,
"epoch": 1.916148431627144,
"grad_norm": 0.6358208583500532,
"learning_rate": 2.166408341552769e-07,
"loss": 0.8708,
"mean_token_accuracy": 0.7793443039059639,
"num_tokens": 1242302339.0,
"step": 14020
},
{
"entropy": 0.855078125,
"epoch": 1.9175152053577529,
"grad_norm": 1.034436998907917,
"learning_rate": 2.131182189657602e-07,
"loss": 0.8331,
"mean_token_accuracy": 0.7850712344050408,
"num_tokens": 1243178877.0,
"step": 14030
},
{
"entropy": 0.866748046875,
"epoch": 1.9188819790883618,
"grad_norm": 0.659997946393946,
"learning_rate": 2.095956037762435e-07,
"loss": 0.8574,
"mean_token_accuracy": 0.7807396940886975,
"num_tokens": 1244068968.0,
"step": 14040
},
{
"entropy": 0.878271484375,
"epoch": 1.9202487528189707,
"grad_norm": 0.6021674466267385,
"learning_rate": 2.0607298858672678e-07,
"loss": 0.8718,
"mean_token_accuracy": 0.7802112899720669,
"num_tokens": 1244958623.0,
"step": 14050
},
{
"entropy": 0.862548828125,
"epoch": 1.9216155265495796,
"grad_norm": 0.65321045313265,
"learning_rate": 2.025503733972101e-07,
"loss": 0.8696,
"mean_token_accuracy": 0.7806286722421646,
"num_tokens": 1245856273.0,
"step": 14060
},
{
"entropy": 0.907958984375,
"epoch": 1.9229823002801885,
"grad_norm": 0.5584449393352221,
"learning_rate": 1.990277582076934e-07,
"loss": 0.9044,
"mean_token_accuracy": 0.7738801248371601,
"num_tokens": 1246770150.0,
"step": 14070
},
{
"entropy": 0.8788818359375,
"epoch": 1.9243490740107974,
"grad_norm": 0.6124849988633743,
"learning_rate": 1.955051430181767e-07,
"loss": 0.8455,
"mean_token_accuracy": 0.7799909695982933,
"num_tokens": 1247607077.0,
"step": 14080
},
{
"entropy": 0.8666259765625,
"epoch": 1.9257158477414063,
"grad_norm": 0.6603681779975706,
"learning_rate": 1.9198252782866002e-07,
"loss": 0.8423,
"mean_token_accuracy": 0.7829136103391647,
"num_tokens": 1248487610.0,
"step": 14090
},
{
"entropy": 0.8647216796875,
"epoch": 1.9270826214720151,
"grad_norm": 0.5977846674118141,
"learning_rate": 1.8845991263914333e-07,
"loss": 0.8556,
"mean_token_accuracy": 0.7800504051148891,
"num_tokens": 1249385378.0,
"step": 14100
},
{
"entropy": 0.895263671875,
"epoch": 1.928449395202624,
"grad_norm": 0.7052202636851796,
"learning_rate": 1.8493729744962664e-07,
"loss": 0.8723,
"mean_token_accuracy": 0.7776886217296124,
"num_tokens": 1250196035.0,
"step": 14110
},
{
"entropy": 0.8943115234375,
"epoch": 1.9298161689332332,
"grad_norm": 0.5956876430150408,
"learning_rate": 1.8141468226010995e-07,
"loss": 0.9141,
"mean_token_accuracy": 0.769616425782442,
"num_tokens": 1251106482.0,
"step": 14120
},
{
"entropy": 0.853759765625,
"epoch": 1.931182942663842,
"grad_norm": 0.6664445760013396,
"learning_rate": 1.778920670705932e-07,
"loss": 0.828,
"mean_token_accuracy": 0.7859554663300514,
"num_tokens": 1251995675.0,
"step": 14130
},
{
"entropy": 0.854736328125,
"epoch": 1.932549716394451,
"grad_norm": 0.5793333618455437,
"learning_rate": 1.743694518810765e-07,
"loss": 0.8268,
"mean_token_accuracy": 0.786456075310707,
"num_tokens": 1252902201.0,
"step": 14140
},
{
"entropy": 0.8594482421875,
"epoch": 1.9339164901250598,
"grad_norm": 0.615322992280405,
"learning_rate": 1.7084683669155982e-07,
"loss": 0.8282,
"mean_token_accuracy": 0.7843310475349426,
"num_tokens": 1253798317.0,
"step": 14150
},
{
"entropy": 0.9089111328125,
"epoch": 1.9352832638556687,
"grad_norm": 0.6451830580608401,
"learning_rate": 1.6732422150204313e-07,
"loss": 0.9015,
"mean_token_accuracy": 0.772354094684124,
"num_tokens": 1254659906.0,
"step": 14160
},
{
"entropy": 0.841259765625,
"epoch": 1.9366500375862776,
"grad_norm": 0.5821533460125914,
"learning_rate": 1.6380160631252644e-07,
"loss": 0.8273,
"mean_token_accuracy": 0.7886179476976395,
"num_tokens": 1255597488.0,
"step": 14170
},
{
"entropy": 0.86806640625,
"epoch": 1.9380168113168865,
"grad_norm": 0.5806479362232033,
"learning_rate": 1.6027899112300975e-07,
"loss": 0.8607,
"mean_token_accuracy": 0.7796605959534645,
"num_tokens": 1256472898.0,
"step": 14180
},
{
"entropy": 0.847705078125,
"epoch": 1.9393835850474954,
"grad_norm": 0.5858058609441618,
"learning_rate": 1.5675637593349303e-07,
"loss": 0.7976,
"mean_token_accuracy": 0.7936233215034008,
"num_tokens": 1257391119.0,
"step": 14190
},
{
"entropy": 0.86083984375,
"epoch": 1.9407503587781043,
"grad_norm": 0.63625515345928,
"learning_rate": 1.5323376074397634e-07,
"loss": 0.848,
"mean_token_accuracy": 0.779336966574192,
"num_tokens": 1258295967.0,
"step": 14200
},
{
"entropy": 0.941845703125,
"epoch": 1.9421171325087132,
"grad_norm": 0.7016896554344426,
"learning_rate": 1.4971114555445965e-07,
"loss": 0.9537,
"mean_token_accuracy": 0.7636564113199711,
"num_tokens": 1259195309.0,
"step": 14210
},
{
"entropy": 0.8882080078125,
"epoch": 1.943483906239322,
"grad_norm": 0.6861409592117784,
"learning_rate": 1.4618853036494293e-07,
"loss": 0.9069,
"mean_token_accuracy": 0.7723307706415653,
"num_tokens": 1260067768.0,
"step": 14220
},
{
"entropy": 0.8856201171875,
"epoch": 1.944850679969931,
"grad_norm": 0.6164110621378277,
"learning_rate": 1.4266591517542624e-07,
"loss": 0.8732,
"mean_token_accuracy": 0.7777280576527119,
"num_tokens": 1260932295.0,
"step": 14230
},
{
"entropy": 0.8828857421875,
"epoch": 1.94621745370054,
"grad_norm": 0.5801836621530847,
"learning_rate": 1.3914329998590955e-07,
"loss": 0.84,
"mean_token_accuracy": 0.7838250868022442,
"num_tokens": 1261753749.0,
"step": 14240
},
{
"entropy": 0.878515625,
"epoch": 1.9475842274311488,
"grad_norm": 0.603414800295112,
"learning_rate": 1.3562068479639286e-07,
"loss": 0.8483,
"mean_token_accuracy": 0.7812985017895698,
"num_tokens": 1262672908.0,
"step": 14250
},
{
"entropy": 0.846142578125,
"epoch": 1.9489510011617577,
"grad_norm": 0.6435928990713268,
"learning_rate": 1.3209806960687614e-07,
"loss": 0.8173,
"mean_token_accuracy": 0.7875230647623539,
"num_tokens": 1263528802.0,
"step": 14260
},
{
"entropy": 0.8662841796875,
"epoch": 1.9503177748923666,
"grad_norm": 0.6678552807533968,
"learning_rate": 1.2857545441735945e-07,
"loss": 0.8541,
"mean_token_accuracy": 0.7831369064748287,
"num_tokens": 1264405077.0,
"step": 14270
},
{
"entropy": 0.8980224609375,
"epoch": 1.9516845486229755,
"grad_norm": 0.6677463938392563,
"learning_rate": 1.2505283922784276e-07,
"loss": 0.8822,
"mean_token_accuracy": 0.7769231639802456,
"num_tokens": 1265320781.0,
"step": 14280
},
{
"entropy": 0.864599609375,
"epoch": 1.9530513223535844,
"grad_norm": 0.6673364125952983,
"learning_rate": 1.2153022403832607e-07,
"loss": 0.8479,
"mean_token_accuracy": 0.78209308385849,
"num_tokens": 1266222145.0,
"step": 14290
},
{
"entropy": 0.868017578125,
"epoch": 1.9544180960841933,
"grad_norm": 0.635626344094964,
"learning_rate": 1.1800760884880937e-07,
"loss": 0.8698,
"mean_token_accuracy": 0.7792714685201645,
"num_tokens": 1267147735.0,
"step": 14300
},
{
"entropy": 0.88720703125,
"epoch": 1.9557848698148022,
"grad_norm": 0.6063108982404437,
"learning_rate": 1.1448499365929266e-07,
"loss": 0.875,
"mean_token_accuracy": 0.7782995782792568,
"num_tokens": 1268033827.0,
"step": 14310
},
{
"entropy": 0.8851806640625,
"epoch": 1.957151643545411,
"grad_norm": 0.6297334757849625,
"learning_rate": 1.1096237846977597e-07,
"loss": 0.8597,
"mean_token_accuracy": 0.7800509043037891,
"num_tokens": 1268910863.0,
"step": 14320
},
{
"entropy": 0.902197265625,
"epoch": 1.95851841727602,
"grad_norm": 0.6462971084308888,
"learning_rate": 1.0743976328025928e-07,
"loss": 0.8719,
"mean_token_accuracy": 0.7753344900906086,
"num_tokens": 1269814412.0,
"step": 14330
},
{
"entropy": 0.844580078125,
"epoch": 1.9598851910066288,
"grad_norm": 0.6426034070666903,
"learning_rate": 1.0391714809074258e-07,
"loss": 0.8366,
"mean_token_accuracy": 0.7836333081126213,
"num_tokens": 1270716800.0,
"step": 14340
},
{
"entropy": 0.86630859375,
"epoch": 1.9612519647372377,
"grad_norm": 0.623673488584865,
"learning_rate": 1.0039453290122588e-07,
"loss": 0.8665,
"mean_token_accuracy": 0.77717809304595,
"num_tokens": 1271593925.0,
"step": 14350
},
{
"entropy": 0.855419921875,
"epoch": 1.9626187384678466,
"grad_norm": 0.6468222216278748,
"learning_rate": 9.687191771170918e-08,
"loss": 0.8352,
"mean_token_accuracy": 0.7845522791147232,
"num_tokens": 1272462660.0,
"step": 14360
},
{
"entropy": 0.897998046875,
"epoch": 1.9639855121984555,
"grad_norm": 0.6681862574563373,
"learning_rate": 9.334930252219248e-08,
"loss": 0.8979,
"mean_token_accuracy": 0.7740066908299923,
"num_tokens": 1273382850.0,
"step": 14370
},
{
"entropy": 0.867333984375,
"epoch": 1.9653522859290644,
"grad_norm": 0.6242326563396422,
"learning_rate": 8.982668733267578e-08,
"loss": 0.8438,
"mean_token_accuracy": 0.7860009931027889,
"num_tokens": 1274290572.0,
"step": 14380
},
{
"entropy": 0.8512939453125,
"epoch": 1.9667190596596733,
"grad_norm": 0.5814155480895156,
"learning_rate": 8.630407214315909e-08,
"loss": 0.8194,
"mean_token_accuracy": 0.7882739640772343,
"num_tokens": 1275174274.0,
"step": 14390
},
{
"entropy": 0.902880859375,
"epoch": 1.9680858333902822,
"grad_norm": 0.6256145287987471,
"learning_rate": 8.27814569536424e-08,
"loss": 0.8676,
"mean_token_accuracy": 0.778792018443346,
"num_tokens": 1276074274.0,
"step": 14400
},
{
"entropy": 0.905078125,
"epoch": 1.969452607120891,
"grad_norm": 0.7265842882842429,
"learning_rate": 7.925884176412568e-08,
"loss": 0.9054,
"mean_token_accuracy": 0.7750126637518406,
"num_tokens": 1276932931.0,
"step": 14410
},
{
"entropy": 0.9159912109375,
"epoch": 1.9708193808515,
"grad_norm": 0.6341446970878963,
"learning_rate": 7.573622657460899e-08,
"loss": 0.9054,
"mean_token_accuracy": 0.7741668440401555,
"num_tokens": 1277813012.0,
"step": 14420
},
{
"entropy": 0.847998046875,
"epoch": 1.9721861545821089,
"grad_norm": 0.6022736893706243,
"learning_rate": 7.22136113850923e-08,
"loss": 0.8152,
"mean_token_accuracy": 0.7887401349842549,
"num_tokens": 1278702374.0,
"step": 14430
},
{
"entropy": 0.849072265625,
"epoch": 1.9735529283127178,
"grad_norm": 0.6020872269695193,
"learning_rate": 6.869099619557561e-08,
"loss": 0.8153,
"mean_token_accuracy": 0.7894223801791668,
"num_tokens": 1279586770.0,
"step": 14440
},
{
"entropy": 0.8651123046875,
"epoch": 1.9749197020433267,
"grad_norm": 0.7155021618143219,
"learning_rate": 6.51683810060589e-08,
"loss": 0.8675,
"mean_token_accuracy": 0.7815263785421849,
"num_tokens": 1280510572.0,
"step": 14450
},
{
"entropy": 0.886279296875,
"epoch": 1.9762864757739358,
"grad_norm": 0.7029734527388513,
"learning_rate": 6.164576581654221e-08,
"loss": 0.8733,
"mean_token_accuracy": 0.7772665232419967,
"num_tokens": 1281364572.0,
"step": 14460
},
{
"entropy": 0.869287109375,
"epoch": 1.9776532495045447,
"grad_norm": 0.641438883307566,
"learning_rate": 5.8123150627025515e-08,
"loss": 0.8347,
"mean_token_accuracy": 0.7828988552093505,
"num_tokens": 1282224524.0,
"step": 14470
},
{
"entropy": 0.86640625,
"epoch": 1.9790200232351536,
"grad_norm": 0.6232797356498359,
"learning_rate": 5.460053543750881e-08,
"loss": 0.8451,
"mean_token_accuracy": 0.783694152534008,
"num_tokens": 1283046842.0,
"step": 14480
},
{
"entropy": 0.886328125,
"epoch": 1.9803867969657625,
"grad_norm": 0.6547873274695531,
"learning_rate": 5.107792024799211e-08,
"loss": 0.9047,
"mean_token_accuracy": 0.7732194416224957,
"num_tokens": 1283926789.0,
"step": 14490
},
{
"entropy": 0.925390625,
"epoch": 1.9817535706963714,
"grad_norm": 0.6658809441330159,
"learning_rate": 4.7555305058475415e-08,
"loss": 0.8977,
"mean_token_accuracy": 0.7715370461344719,
"num_tokens": 1284778482.0,
"step": 14500
},
{
"entropy": 0.882177734375,
"epoch": 1.9831203444269803,
"grad_norm": 0.6442397812458255,
"learning_rate": 4.403268986895872e-08,
"loss": 0.891,
"mean_token_accuracy": 0.7781330294907093,
"num_tokens": 1285717361.0,
"step": 14510
},
{
"entropy": 0.88759765625,
"epoch": 1.9844871181575892,
"grad_norm": 0.6435395199414549,
"learning_rate": 4.0510074679442026e-08,
"loss": 0.8894,
"mean_token_accuracy": 0.7747194200754166,
"num_tokens": 1286605150.0,
"step": 14520
},
{
"entropy": 0.9047607421875,
"epoch": 1.985853891888198,
"grad_norm": 0.6278560895846574,
"learning_rate": 3.698745948992532e-08,
"loss": 0.8546,
"mean_token_accuracy": 0.7796093784272671,
"num_tokens": 1287429714.0,
"step": 14530
},
{
"entropy": 0.8635498046875,
"epoch": 1.987220665618807,
"grad_norm": 0.633473529808102,
"learning_rate": 3.3464844300408624e-08,
"loss": 0.8385,
"mean_token_accuracy": 0.7821372672915459,
"num_tokens": 1288274856.0,
"step": 14540
},
{
"entropy": 0.8583740234375,
"epoch": 1.9885874393494158,
"grad_norm": 0.6268955879754733,
"learning_rate": 2.994222911089193e-08,
"loss": 0.8409,
"mean_token_accuracy": 0.783450861275196,
"num_tokens": 1289184678.0,
"step": 14550
},
{
"entropy": 0.8943359375,
"epoch": 1.9899542130800247,
"grad_norm": 0.632037899112269,
"learning_rate": 2.6419613921375232e-08,
"loss": 0.8881,
"mean_token_accuracy": 0.7759060755372047,
"num_tokens": 1290104163.0,
"step": 14560
},
{
"entropy": 0.848291015625,
"epoch": 1.9913209868106336,
"grad_norm": 0.6457538674864827,
"learning_rate": 2.289699873185853e-08,
"loss": 0.8436,
"mean_token_accuracy": 0.782917107641697,
"num_tokens": 1290984420.0,
"step": 14570
},
{
"entropy": 0.9302734375,
"epoch": 1.9926877605412425,
"grad_norm": 0.6041236591901165,
"learning_rate": 1.9374383542341837e-08,
"loss": 0.9214,
"mean_token_accuracy": 0.7684244722127914,
"num_tokens": 1291827527.0,
"step": 14580
},
{
"entropy": 0.8504638671875,
"epoch": 1.9940545342718514,
"grad_norm": 0.6022852228737959,
"learning_rate": 1.585176835282514e-08,
"loss": 0.8211,
"mean_token_accuracy": 0.7858721174299717,
"num_tokens": 1292724554.0,
"step": 14590
},
{
"entropy": 0.92724609375,
"epoch": 1.9954213080024603,
"grad_norm": 0.6472079976801512,
"learning_rate": 1.2329153163308442e-08,
"loss": 0.9175,
"mean_token_accuracy": 0.768941281735897,
"num_tokens": 1293618300.0,
"step": 14600
},
{
"entropy": 0.8853515625,
"epoch": 1.9967880817330692,
"grad_norm": 0.6595030593134943,
"learning_rate": 8.806537973791744e-09,
"loss": 0.8734,
"mean_token_accuracy": 0.7770158238708973,
"num_tokens": 1294534796.0,
"step": 14610
},
{
"entropy": 0.856298828125,
"epoch": 1.998154855463678,
"grad_norm": 0.5963360498985301,
"learning_rate": 5.2839227842750465e-09,
"loss": 0.8535,
"mean_token_accuracy": 0.7829092271625996,
"num_tokens": 1295448543.0,
"step": 14620
},
{
"entropy": 0.8841796875,
"epoch": 1.999521629194287,
"grad_norm": 0.664000367394873,
"learning_rate": 1.7613075947583486e-09,
"loss": 0.864,
"mean_token_accuracy": 0.7774464026093483,
"num_tokens": 1296340286.0,
"step": 14630
},
{
"entropy": 0.8878348214285714,
"epoch": 2.0,
"mean_token_accuracy": 0.7744237844433103,
"num_tokens": 1296663702.0,
"step": 14634,
"total_flos": 2402749059235840.0,
"train_loss": 0.903982289814633,
"train_runtime": 66066.3042,
"train_samples_per_second": 28.351,
"train_steps_per_second": 0.222
}
],
"logging_steps": 10,
"max_steps": 14634,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 200,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2402749059235840.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink