ali-elganzory/Qwen3-1.7B-Base-SFT-Tulu3-decontaminated-masked
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
Qwen3-1.7B-Base-SFT-Tulu3-d…/trainer_state.json
ModelHub XC 56f6c8ebe9 初始化项目，由ModelHub XC社区提供模型 
Model: ali-elganzory/Qwen3-1.7B-Base-SFT-Tulu3-decontaminated-masked
Source: Original Platform
2026-07-19 01:21:11 +08:00

14677 lines
410 KiB
JSON
Raw Permalink 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": 1.108203125,
"epoch": 0.0013667737306088978,
"grad_norm": 2.4833950867336028,
"learning_rate": 1.0227272727272728e-07,
"loss": 0.8499,
"mean_token_accuracy": 0.779207244515419,
"num_tokens": 894751.0,
"step": 10
},
{
"entropy": 1.0833984375,
"epoch": 0.0027335474612177955,
"grad_norm": 2.448528272173453,
"learning_rate": 2.1590909090909094e-07,
"loss": 0.7969,
"mean_token_accuracy": 0.7947208464145661,
"num_tokens": 1750308.0,
"step": 20
},
{
"entropy": 1.0720703125,
"epoch": 0.004100321191826693,
"grad_norm": 2.51110046423478,
"learning_rate": 3.2954545454545455e-07,
"loss": 0.8115,
"mean_token_accuracy": 0.7888881891965867,
"num_tokens": 2644413.0,
"step": 30
},
{
"entropy": 1.1150390625,
"epoch": 0.005467094922435591,
"grad_norm": 1.8783594186439598,
"learning_rate": 4.431818181818182e-07,
"loss": 0.8353,
"mean_token_accuracy": 0.7858321815729141,
"num_tokens": 3517264.0,
"step": 40
},
{
"entropy": 1.0501953125,
"epoch": 0.006833868653044488,
"grad_norm": 1.9275433760801968,
"learning_rate": 5.568181818181818e-07,
"loss": 0.7861,
"mean_token_accuracy": 0.7940415740013123,
"num_tokens": 4410342.0,
"step": 50
},
{
"entropy": 1.0953125,
"epoch": 0.008200642383653386,
"grad_norm": 1.1587303733397005,
"learning_rate": 6.704545454545456e-07,
"loss": 0.7992,
"mean_token_accuracy": 0.7910040110349655,
"num_tokens": 5335473.0,
"step": 60
},
{
"entropy": 1.1373046875,
"epoch": 0.009567416114262284,
"grad_norm": 1.1148868704213344,
"learning_rate": 7.840909090909092e-07,
"loss": 0.8212,
"mean_token_accuracy": 0.7874140411615371,
"num_tokens": 6259118.0,
"step": 70
},
{
"entropy": 1.1607421875,
"epoch": 0.010934189844871182,
"grad_norm": 1.0651745453375532,
"learning_rate": 8.977272727272728e-07,
"loss": 0.8471,
"mean_token_accuracy": 0.7830597758293152,
"num_tokens": 7173208.0,
"step": 80
},
{
"entropy": 1.1025390625,
"epoch": 0.01230096357548008,
"grad_norm": 0.8393764941455704,
"learning_rate": 1.0113636363636365e-06,
"loss": 0.76,
"mean_token_accuracy": 0.7968711405992508,
"num_tokens": 8004347.0,
"step": 90
},
{
"entropy": 1.0767578125,
"epoch": 0.013667737306088976,
"grad_norm": 0.9099768218292918,
"learning_rate": 1.125e-06,
"loss": 0.7766,
"mean_token_accuracy": 0.7981488019227981,
"num_tokens": 8901160.0,
"step": 100
},
{
"entropy": 1.0689453125,
"epoch": 0.015034511036697874,
"grad_norm": 0.844960436483788,
"learning_rate": 1.2386363636363638e-06,
"loss": 0.7467,
"mean_token_accuracy": 0.8018648207187653,
"num_tokens": 9797773.0,
"step": 110
},
{
"entropy": 1.06796875,
"epoch": 0.016401284767306772,
"grad_norm": 0.8881330042023365,
"learning_rate": 1.3522727272727273e-06,
"loss": 0.7572,
"mean_token_accuracy": 0.799167063832283,
"num_tokens": 10670680.0,
"step": 120
},
{
"entropy": 1.0638671875,
"epoch": 0.01776805849791567,
"grad_norm": 0.8256390271448409,
"learning_rate": 1.465909090909091e-06,
"loss": 0.7494,
"mean_token_accuracy": 0.8005818784236908,
"num_tokens": 11561633.0,
"step": 130
},
{
"entropy": 1.0771484375,
"epoch": 0.019134832228524568,
"grad_norm": 0.7945873851619824,
"learning_rate": 1.5795454545454547e-06,
"loss": 0.734,
"mean_token_accuracy": 0.8044161438941956,
"num_tokens": 12473967.0,
"step": 140
},
{
"entropy": 1.0841796875,
"epoch": 0.020501605959133466,
"grad_norm": 0.8362941444082493,
"learning_rate": 1.6931818181818182e-06,
"loss": 0.7621,
"mean_token_accuracy": 0.8023771911859512,
"num_tokens": 13386639.0,
"step": 150
},
{
"entropy": 1.0671875,
"epoch": 0.021868379689742364,
"grad_norm": 0.9116663074119534,
"learning_rate": 1.8068181818181822e-06,
"loss": 0.7801,
"mean_token_accuracy": 0.7942297756671906,
"num_tokens": 14284815.0,
"step": 160
},
{
"entropy": 1.08671875,
"epoch": 0.023235153420351262,
"grad_norm": 0.8470619179866946,
"learning_rate": 1.9204545454545457e-06,
"loss": 0.769,
"mean_token_accuracy": 0.7988474547863007,
"num_tokens": 15109817.0,
"step": 170
},
{
"entropy": 1.02890625,
"epoch": 0.02460192715096016,
"grad_norm": 0.7479213914334186,
"learning_rate": 2.034090909090909e-06,
"loss": 0.7363,
"mean_token_accuracy": 0.8044725477695465,
"num_tokens": 15983614.0,
"step": 180
},
{
"entropy": 1.0505859375,
"epoch": 0.025968700881569055,
"grad_norm": 0.7745011877291036,
"learning_rate": 2.147727272727273e-06,
"loss": 0.7773,
"mean_token_accuracy": 0.7959740787744523,
"num_tokens": 16896096.0,
"step": 190
},
{
"entropy": 1.0890625,
"epoch": 0.027335474612177953,
"grad_norm": 0.8417562616712069,
"learning_rate": 2.2613636363636366e-06,
"loss": 0.781,
"mean_token_accuracy": 0.7974732279777527,
"num_tokens": 17736266.0,
"step": 200
},
{
"entropy": 1.066796875,
"epoch": 0.02870224834278685,
"grad_norm": 0.8674020706105897,
"learning_rate": 2.375e-06,
"loss": 0.7668,
"mean_token_accuracy": 0.7961135774850845,
"num_tokens": 18638867.0,
"step": 210
},
{
"entropy": 1.08984375,
"epoch": 0.03006902207339575,
"grad_norm": 0.9901938336942466,
"learning_rate": 2.488636363636364e-06,
"loss": 0.7593,
"mean_token_accuracy": 0.7995853871107101,
"num_tokens": 19534482.0,
"step": 220
},
{
"entropy": 1.025390625,
"epoch": 0.03143579580400465,
"grad_norm": 0.8650790589128151,
"learning_rate": 2.6022727272727276e-06,
"loss": 0.7035,
"mean_token_accuracy": 0.8100258558988571,
"num_tokens": 20445734.0,
"step": 230
},
{
"entropy": 1.05859375,
"epoch": 0.032802569534613545,
"grad_norm": 0.8165384462431529,
"learning_rate": 2.715909090909091e-06,
"loss": 0.7424,
"mean_token_accuracy": 0.8022427886724472,
"num_tokens": 21340398.0,
"step": 240
},
{
"entropy": 1.0466796875,
"epoch": 0.03416934326522244,
"grad_norm": 0.8624240732408183,
"learning_rate": 2.829545454545455e-06,
"loss": 0.7333,
"mean_token_accuracy": 0.8060361295938492,
"num_tokens": 22231277.0,
"step": 250
},
{
"entropy": 1.0068359375,
"epoch": 0.03553611699583134,
"grad_norm": 0.7709505884837659,
"learning_rate": 2.9431818181818185e-06,
"loss": 0.7063,
"mean_token_accuracy": 0.8124743491411209,
"num_tokens": 23128440.0,
"step": 260
},
{
"entropy": 1.023046875,
"epoch": 0.036902890726440235,
"grad_norm": 0.6784493682659523,
"learning_rate": 3.056818181818182e-06,
"loss": 0.7,
"mean_token_accuracy": 0.8119614213705063,
"num_tokens": 24023611.0,
"step": 270
},
{
"entropy": 1.01875,
"epoch": 0.038269664457049136,
"grad_norm": 0.7574711864822365,
"learning_rate": 3.1704545454545456e-06,
"loss": 0.73,
"mean_token_accuracy": 0.8068324208259583,
"num_tokens": 24931447.0,
"step": 280
},
{
"entropy": 1.047265625,
"epoch": 0.03963643818765803,
"grad_norm": 0.8544490431521309,
"learning_rate": 3.2840909090909095e-06,
"loss": 0.7476,
"mean_token_accuracy": 0.8026481896638871,
"num_tokens": 25822637.0,
"step": 290
},
{
"entropy": 1.0224609375,
"epoch": 0.04100321191826693,
"grad_norm": 0.8576997368885168,
"learning_rate": 3.397727272727273e-06,
"loss": 0.7295,
"mean_token_accuracy": 0.8045373916625976,
"num_tokens": 26705243.0,
"step": 300
},
{
"entropy": 1.0544921875,
"epoch": 0.04236998564887583,
"grad_norm": 0.863965066703299,
"learning_rate": 3.5113636363636365e-06,
"loss": 0.7551,
"mean_token_accuracy": 0.8028387516736984,
"num_tokens": 27620336.0,
"step": 310
},
{
"entropy": 1.0390625,
"epoch": 0.04373675937948473,
"grad_norm": 0.8029676374122844,
"learning_rate": 3.625e-06,
"loss": 0.7319,
"mean_token_accuracy": 0.8025805532932282,
"num_tokens": 28487030.0,
"step": 320
},
{
"entropy": 1.098828125,
"epoch": 0.04510353311009362,
"grad_norm": 0.835802364306144,
"learning_rate": 3.7386363636363635e-06,
"loss": 0.8044,
"mean_token_accuracy": 0.793497633934021,
"num_tokens": 29362700.0,
"step": 330
},
{
"entropy": 1.001953125,
"epoch": 0.046470306840702524,
"grad_norm": 0.8416376219152584,
"learning_rate": 3.852272727272728e-06,
"loss": 0.7007,
"mean_token_accuracy": 0.8117711633443833,
"num_tokens": 30216448.0,
"step": 340
},
{
"entropy": 0.9962890625,
"epoch": 0.04783708057131142,
"grad_norm": 0.834018280872051,
"learning_rate": 3.965909090909091e-06,
"loss": 0.6953,
"mean_token_accuracy": 0.8136526644229889,
"num_tokens": 31124761.0,
"step": 350
},
{
"entropy": 1.0564453125,
"epoch": 0.04920385430192032,
"grad_norm": 0.7639003461158121,
"learning_rate": 4.079545454545455e-06,
"loss": 0.7678,
"mean_token_accuracy": 0.7981337636709214,
"num_tokens": 32021822.0,
"step": 360
},
{
"entropy": 1.060546875,
"epoch": 0.050570628032529215,
"grad_norm": 0.8741751280247381,
"learning_rate": 4.193181818181819e-06,
"loss": 0.7434,
"mean_token_accuracy": 0.8021952897310257,
"num_tokens": 32911869.0,
"step": 370
},
{
"entropy": 1.0662109375,
"epoch": 0.05193740176313811,
"grad_norm": 0.8972234566098974,
"learning_rate": 4.306818181818182e-06,
"loss": 0.7755,
"mean_token_accuracy": 0.7974812805652618,
"num_tokens": 33780250.0,
"step": 380
},
{
"entropy": 1.0123046875,
"epoch": 0.05330417549374701,
"grad_norm": 0.7845921398865803,
"learning_rate": 4.420454545454546e-06,
"loss": 0.7208,
"mean_token_accuracy": 0.807163092494011,
"num_tokens": 34694542.0,
"step": 390
},
{
"entropy": 1.0111328125,
"epoch": 0.054670949224355905,
"grad_norm": 0.7392176868674488,
"learning_rate": 4.53409090909091e-06,
"loss": 0.7079,
"mean_token_accuracy": 0.8117583870887757,
"num_tokens": 35546961.0,
"step": 400
},
{
"entropy": 1.0837890625,
"epoch": 0.05603772295496481,
"grad_norm": 0.9325305843495422,
"learning_rate": 4.647727272727273e-06,
"loss": 0.7566,
"mean_token_accuracy": 0.797896096110344,
"num_tokens": 36410375.0,
"step": 410
},
{
"entropy": 1.01640625,
"epoch": 0.0574044966855737,
"grad_norm": 0.8602457834872663,
"learning_rate": 4.761363636363637e-06,
"loss": 0.7247,
"mean_token_accuracy": 0.8072879076004028,
"num_tokens": 37311509.0,
"step": 420
},
{
"entropy": 1.003515625,
"epoch": 0.0587712704161826,
"grad_norm": 0.8200131160840981,
"learning_rate": 4.875e-06,
"loss": 0.6909,
"mean_token_accuracy": 0.815858119726181,
"num_tokens": 38219048.0,
"step": 430
},
{
"entropy": 1.0421875,
"epoch": 0.0601380441467915,
"grad_norm": 0.7768523588453945,
"learning_rate": 4.988636363636364e-06,
"loss": 0.7874,
"mean_token_accuracy": 0.7942088812589645,
"num_tokens": 39127651.0,
"step": 440
},
{
"entropy": 1.0515625,
"epoch": 0.0615048178774004,
"grad_norm": 0.7769441004008638,
"learning_rate": 4.996829646329435e-06,
"loss": 0.7629,
"mean_token_accuracy": 0.7960357218980789,
"num_tokens": 40015164.0,
"step": 450
},
{
"entropy": 1.086328125,
"epoch": 0.0628715916080093,
"grad_norm": 0.8385953222739866,
"learning_rate": 4.993307031139919e-06,
"loss": 0.7986,
"mean_token_accuracy": 0.7921811133623123,
"num_tokens": 40925068.0,
"step": 460
},
{
"entropy": 1.06484375,
"epoch": 0.0642383653386182,
"grad_norm": 0.7811730793312988,
"learning_rate": 4.989784415950402e-06,
"loss": 0.7949,
"mean_token_accuracy": 0.791978332400322,
"num_tokens": 41825875.0,
"step": 470
},
{
"entropy": 1.090234375,
"epoch": 0.06560513906922709,
"grad_norm": 0.8951092433887612,
"learning_rate": 4.986261800760885e-06,
"loss": 0.7873,
"mean_token_accuracy": 0.7919754981994629,
"num_tokens": 42703080.0,
"step": 480
},
{
"entropy": 1.03359375,
"epoch": 0.06697191279983598,
"grad_norm": 0.7925777019123804,
"learning_rate": 4.9827391855713685e-06,
"loss": 0.7498,
"mean_token_accuracy": 0.8040930300951004,
"num_tokens": 43616920.0,
"step": 490
},
{
"entropy": 1.040625,
"epoch": 0.06833868653044488,
"grad_norm": 0.8833245141950362,
"learning_rate": 4.979216570381852e-06,
"loss": 0.7039,
"mean_token_accuracy": 0.8094258815050125,
"num_tokens": 44506039.0,
"step": 500
},
{
"entropy": 1.0416015625,
"epoch": 0.06970546026105379,
"grad_norm": 0.8217713843409041,
"learning_rate": 4.975693955192335e-06,
"loss": 0.7534,
"mean_token_accuracy": 0.7984866559505462,
"num_tokens": 45382350.0,
"step": 510
},
{
"entropy": 1.0666015625,
"epoch": 0.07107223399166268,
"grad_norm": 0.8968708017861274,
"learning_rate": 4.972171340002819e-06,
"loss": 0.761,
"mean_token_accuracy": 0.8001853406429291,
"num_tokens": 46304368.0,
"step": 520
},
{
"entropy": 1.0662109375,
"epoch": 0.07243900772227158,
"grad_norm": 0.8422021141099825,
"learning_rate": 4.968648724813302e-06,
"loss": 0.7527,
"mean_token_accuracy": 0.7999732434749603,
"num_tokens": 47189558.0,
"step": 530
},
{
"entropy": 1.0427734375,
"epoch": 0.07380578145288047,
"grad_norm": 0.8625146537561087,
"learning_rate": 4.965126109623785e-06,
"loss": 0.7557,
"mean_token_accuracy": 0.8013763636350631,
"num_tokens": 48062199.0,
"step": 540
},
{
"entropy": 1.0671875,
"epoch": 0.07517255518348938,
"grad_norm": 0.8564612749461629,
"learning_rate": 4.961603494434268e-06,
"loss": 0.7959,
"mean_token_accuracy": 0.7936537563800812,
"num_tokens": 48939443.0,
"step": 550
},
{
"entropy": 1.0197265625,
"epoch": 0.07653932891409827,
"grad_norm": 0.8600929235512221,
"learning_rate": 4.958080879244752e-06,
"loss": 0.7022,
"mean_token_accuracy": 0.8107238352298737,
"num_tokens": 49849066.0,
"step": 560
},
{
"entropy": 1.02578125,
"epoch": 0.07790610264470717,
"grad_norm": 0.8839106554748335,
"learning_rate": 4.954558264055234e-06,
"loss": 0.7374,
"mean_token_accuracy": 0.8023515343666077,
"num_tokens": 50698168.0,
"step": 570
},
{
"entropy": 1.00390625,
"epoch": 0.07927287637531606,
"grad_norm": 0.9814255234984038,
"learning_rate": 4.951035648865719e-06,
"loss": 0.7255,
"mean_token_accuracy": 0.8053876549005509,
"num_tokens": 51559112.0,
"step": 580
},
{
"entropy": 1.0193359375,
"epoch": 0.08063965010592497,
"grad_norm": 0.8576811300000117,
"learning_rate": 4.9475130336762015e-06,
"loss": 0.7153,
"mean_token_accuracy": 0.8070975095033646,
"num_tokens": 52433072.0,
"step": 590
},
{
"entropy": 1.062890625,
"epoch": 0.08200642383653386,
"grad_norm": 0.8635361750833235,
"learning_rate": 4.943990418486685e-06,
"loss": 0.7598,
"mean_token_accuracy": 0.7992813557386398,
"num_tokens": 53357598.0,
"step": 600
},
{
"entropy": 1.0427734375,
"epoch": 0.08337319756714276,
"grad_norm": 0.8117536189959161,
"learning_rate": 4.9404678032971685e-06,
"loss": 0.7614,
"mean_token_accuracy": 0.8018678963184357,
"num_tokens": 54234701.0,
"step": 610
},
{
"entropy": 1.0205078125,
"epoch": 0.08473997129775165,
"grad_norm": 0.8141101926036644,
"learning_rate": 4.936945188107651e-06,
"loss": 0.7553,
"mean_token_accuracy": 0.8012787312269211,
"num_tokens": 55143442.0,
"step": 620
},
{
"entropy": 0.979296875,
"epoch": 0.08610674502836055,
"grad_norm": 0.8486927204368574,
"learning_rate": 4.933422572918135e-06,
"loss": 0.702,
"mean_token_accuracy": 0.8136590659618378,
"num_tokens": 56014638.0,
"step": 630
},
{
"entropy": 1.0220703125,
"epoch": 0.08747351875896946,
"grad_norm": 0.761297096251087,
"learning_rate": 4.929899957728618e-06,
"loss": 0.726,
"mean_token_accuracy": 0.8083883941173553,
"num_tokens": 56919559.0,
"step": 640
},
{
"entropy": 1.0396484375,
"epoch": 0.08884029248957835,
"grad_norm": 0.9647705739576775,
"learning_rate": 4.926377342539102e-06,
"loss": 0.7215,
"mean_token_accuracy": 0.8077186554670334,
"num_tokens": 57796634.0,
"step": 650
},
{
"entropy": 1.0224609375,
"epoch": 0.09020706622018725,
"grad_norm": 0.8174345084167174,
"learning_rate": 4.922854727349585e-06,
"loss": 0.7231,
"mean_token_accuracy": 0.8060762941837311,
"num_tokens": 58681705.0,
"step": 660
},
{
"entropy": 1.0298828125,
"epoch": 0.09157383995079614,
"grad_norm": 0.7720523708847083,
"learning_rate": 4.919332112160068e-06,
"loss": 0.7424,
"mean_token_accuracy": 0.8008512645959854,
"num_tokens": 59598109.0,
"step": 670
},
{
"entropy": 1.0396484375,
"epoch": 0.09294061368140505,
"grad_norm": 0.7783246652259422,
"learning_rate": 4.915809496970551e-06,
"loss": 0.7295,
"mean_token_accuracy": 0.805041617155075,
"num_tokens": 60505590.0,
"step": 680
},
{
"entropy": 1.09375,
"epoch": 0.09430738741201394,
"grad_norm": 0.8463351763288609,
"learning_rate": 4.912286881781035e-06,
"loss": 0.8337,
"mean_token_accuracy": 0.7833318382501602,
"num_tokens": 61382600.0,
"step": 690
},
{
"entropy": 1.0154296875,
"epoch": 0.09567416114262284,
"grad_norm": 0.8670151286489363,
"learning_rate": 4.908764266591518e-06,
"loss": 0.7036,
"mean_token_accuracy": 0.8109525233507157,
"num_tokens": 62258121.0,
"step": 700
},
{
"entropy": 1.0380859375,
"epoch": 0.09704093487323173,
"grad_norm": 0.8358537867057667,
"learning_rate": 4.9052416514020015e-06,
"loss": 0.7141,
"mean_token_accuracy": 0.8098570466041565,
"num_tokens": 63128686.0,
"step": 710
},
{
"entropy": 0.9935546875,
"epoch": 0.09840770860384064,
"grad_norm": 0.83395497431297,
"learning_rate": 4.901719036212484e-06,
"loss": 0.711,
"mean_token_accuracy": 0.8081421583890915,
"num_tokens": 64056881.0,
"step": 720
},
{
"entropy": 1.016015625,
"epoch": 0.09977448233444954,
"grad_norm": 0.8650983165060004,
"learning_rate": 4.898196421022968e-06,
"loss": 0.7141,
"mean_token_accuracy": 0.8080666363239288,
"num_tokens": 64941116.0,
"step": 730
},
{
"entropy": 1.0212890625,
"epoch": 0.10114125606505843,
"grad_norm": 0.9155051312806181,
"learning_rate": 4.894673805833451e-06,
"loss": 0.6733,
"mean_token_accuracy": 0.8170630067586899,
"num_tokens": 65760286.0,
"step": 740
},
{
"entropy": 1.040234375,
"epoch": 0.10250802979566732,
"grad_norm": 0.8162557386789376,
"learning_rate": 4.891151190643935e-06,
"loss": 0.7478,
"mean_token_accuracy": 0.802932733297348,
"num_tokens": 66641601.0,
"step": 750
},
{
"entropy": 1.034375,
"epoch": 0.10387480352627622,
"grad_norm": 0.8096287142272944,
"learning_rate": 4.8876285754544175e-06,
"loss": 0.7322,
"mean_token_accuracy": 0.8050035834312439,
"num_tokens": 67556757.0,
"step": 760
},
{
"entropy": 1.0716796875,
"epoch": 0.10524157725688513,
"grad_norm": 0.8869071697854729,
"learning_rate": 4.884105960264901e-06,
"loss": 0.776,
"mean_token_accuracy": 0.7978862911462784,
"num_tokens": 68415204.0,
"step": 770
},
{
"entropy": 1.0146484375,
"epoch": 0.10660835098749402,
"grad_norm": 0.7867866216232045,
"learning_rate": 4.880583345075385e-06,
"loss": 0.7282,
"mean_token_accuracy": 0.8062320828437806,
"num_tokens": 69305673.0,
"step": 780
},
{
"entropy": 1.0185546875,
"epoch": 0.10797512471810292,
"grad_norm": 0.7345963059557391,
"learning_rate": 4.877060729885867e-06,
"loss": 0.6866,
"mean_token_accuracy": 0.8148729801177979,
"num_tokens": 70187437.0,
"step": 790
},
{
"entropy": 1.0189453125,
"epoch": 0.10934189844871181,
"grad_norm": 0.870276029234116,
"learning_rate": 4.873538114696351e-06,
"loss": 0.7133,
"mean_token_accuracy": 0.8078462302684783,
"num_tokens": 71066843.0,
"step": 800
},
{
"entropy": 1.03671875,
"epoch": 0.11070867217932072,
"grad_norm": 0.8434562301088327,
"learning_rate": 4.870015499506834e-06,
"loss": 0.705,
"mean_token_accuracy": 0.812209153175354,
"num_tokens": 71952220.0,
"step": 810
},
{
"entropy": 1.0044921875,
"epoch": 0.11207544590992961,
"grad_norm": 0.7877139853870151,
"learning_rate": 4.866492884317318e-06,
"loss": 0.7125,
"mean_token_accuracy": 0.8070896804332733,
"num_tokens": 72848678.0,
"step": 820
},
{
"entropy": 1.004296875,
"epoch": 0.11344221964053851,
"grad_norm": 0.8737894700241836,
"learning_rate": 4.862970269127801e-06,
"loss": 0.7108,
"mean_token_accuracy": 0.811753922700882,
"num_tokens": 73722522.0,
"step": 830
},
{
"entropy": 1.0583984375,
"epoch": 0.1148089933711474,
"grad_norm": 0.8259957286948921,
"learning_rate": 4.859447653938284e-06,
"loss": 0.746,
"mean_token_accuracy": 0.8021477669477463,
"num_tokens": 74585045.0,
"step": 840
},
{
"entropy": 1.0263671875,
"epoch": 0.11617576710175631,
"grad_norm": 0.7321908204374379,
"learning_rate": 4.855925038748768e-06,
"loss": 0.7886,
"mean_token_accuracy": 0.7949862241744995,
"num_tokens": 75542861.0,
"step": 850
},
{
"entropy": 1.02578125,
"epoch": 0.1175425408323652,
"grad_norm": 0.823762249315054,
"learning_rate": 4.852402423559251e-06,
"loss": 0.7275,
"mean_token_accuracy": 0.8068595975637436,
"num_tokens": 76472020.0,
"step": 860
},
{
"entropy": 1.0638671875,
"epoch": 0.1189093145629741,
"grad_norm": 0.7766392902179192,
"learning_rate": 4.848879808369734e-06,
"loss": 0.7875,
"mean_token_accuracy": 0.7940213292837143,
"num_tokens": 77350339.0,
"step": 870
},
{
"entropy": 1.0166015625,
"epoch": 0.120276088293583,
"grad_norm": 0.782692046509434,
"learning_rate": 4.8453571931802175e-06,
"loss": 0.6838,
"mean_token_accuracy": 0.8149712204933166,
"num_tokens": 78257082.0,
"step": 880
},
{
"entropy": 0.9779296875,
"epoch": 0.12164286202419189,
"grad_norm": 0.7350024550775773,
"learning_rate": 4.8418345779907e-06,
"loss": 0.7466,
"mean_token_accuracy": 0.8014688998460769,
"num_tokens": 79122216.0,
"step": 890
},
{
"entropy": 1.058203125,
"epoch": 0.1230096357548008,
"grad_norm": 0.8153880103850201,
"learning_rate": 4.838311962801184e-06,
"loss": 0.7518,
"mean_token_accuracy": 0.7998087197542191,
"num_tokens": 80032714.0,
"step": 900
},
{
"entropy": 1.0396484375,
"epoch": 0.12437640948540969,
"grad_norm": 0.8630079974024805,
"learning_rate": 4.834789347611667e-06,
"loss": 0.7446,
"mean_token_accuracy": 0.8022423714399338,
"num_tokens": 80873820.0,
"step": 910
},
{
"entropy": 1.0591796875,
"epoch": 0.1257431832160186,
"grad_norm": 0.8822906034875729,
"learning_rate": 4.831266732422151e-06,
"loss": 0.7431,
"mean_token_accuracy": 0.8036547392606735,
"num_tokens": 81719615.0,
"step": 920
},
{
"entropy": 0.991015625,
"epoch": 0.1271099569466275,
"grad_norm": 0.8339258299590762,
"learning_rate": 4.827744117232634e-06,
"loss": 0.6704,
"mean_token_accuracy": 0.8177006393671036,
"num_tokens": 82585299.0,
"step": 930
},
{
"entropy": 1.0119140625,
"epoch": 0.1284767306772364,
"grad_norm": 0.8728370222103438,
"learning_rate": 4.824221502043117e-06,
"loss": 0.6822,
"mean_token_accuracy": 0.8155334293842316,
"num_tokens": 83450499.0,
"step": 940
},
{
"entropy": 1.042578125,
"epoch": 0.12984350440784528,
"grad_norm": 0.7487341100504479,
"learning_rate": 4.820698886853601e-06,
"loss": 0.7575,
"mean_token_accuracy": 0.7984952032566071,
"num_tokens": 84384201.0,
"step": 950
},
{
"entropy": 1.0587890625,
"epoch": 0.13121027813845418,
"grad_norm": 0.7897783937884317,
"learning_rate": 4.817176271664084e-06,
"loss": 0.7483,
"mean_token_accuracy": 0.8013297080993652,
"num_tokens": 85282609.0,
"step": 960
},
{
"entropy": 0.9958984375,
"epoch": 0.13257705186906307,
"grad_norm": 0.8390491860242915,
"learning_rate": 4.813653656474567e-06,
"loss": 0.6979,
"mean_token_accuracy": 0.8123379617929458,
"num_tokens": 86130389.0,
"step": 970
},
{
"entropy": 1.0474609375,
"epoch": 0.13394382559967197,
"grad_norm": 0.916864548701589,
"learning_rate": 4.8101310412850505e-06,
"loss": 0.7352,
"mean_token_accuracy": 0.8037164121866226,
"num_tokens": 87013196.0,
"step": 980
},
{
"entropy": 1.0185546875,
"epoch": 0.13531059933028086,
"grad_norm": 0.7864178830314299,
"learning_rate": 4.806608426095534e-06,
"loss": 0.714,
"mean_token_accuracy": 0.8094704717397689,
"num_tokens": 87940418.0,
"step": 990
},
{
"entropy": 0.9974609375,
"epoch": 0.13667737306088976,
"grad_norm": 0.7074294037882203,
"learning_rate": 4.803085810906017e-06,
"loss": 0.6975,
"mean_token_accuracy": 0.8115524739027024,
"num_tokens": 88835060.0,
"step": 1000
},
{
"entropy": 1.051953125,
"epoch": 0.13804414679149868,
"grad_norm": 0.7451670008654181,
"learning_rate": 4.7995631957165e-06,
"loss": 0.7527,
"mean_token_accuracy": 0.799260500073433,
"num_tokens": 89750236.0,
"step": 1010
},
{
"entropy": 0.975,
"epoch": 0.13941092052210757,
"grad_norm": 0.837410804357508,
"learning_rate": 4.796040580526984e-06,
"loss": 0.6769,
"mean_token_accuracy": 0.8167847871780396,
"num_tokens": 90644061.0,
"step": 1020
},
{
"entropy": 0.9970703125,
"epoch": 0.14077769425271647,
"grad_norm": 0.928946548182078,
"learning_rate": 4.792517965337467e-06,
"loss": 0.7144,
"mean_token_accuracy": 0.8086564093828201,
"num_tokens": 91490004.0,
"step": 1030
},
{
"entropy": 0.98359375,
"epoch": 0.14214446798332536,
"grad_norm": 0.8480110598502361,
"learning_rate": 4.78899535014795e-06,
"loss": 0.6744,
"mean_token_accuracy": 0.8183129161596299,
"num_tokens": 92377049.0,
"step": 1040
},
{
"entropy": 1.004296875,
"epoch": 0.14351124171393426,
"grad_norm": 0.7670430306067388,
"learning_rate": 4.785472734958434e-06,
"loss": 0.6994,
"mean_token_accuracy": 0.8122775435447693,
"num_tokens": 93288203.0,
"step": 1050
},
{
"entropy": 1.01640625,
"epoch": 0.14487801544454315,
"grad_norm": 0.7847429228649154,
"learning_rate": 4.781950119768916e-06,
"loss": 0.6988,
"mean_token_accuracy": 0.811975285410881,
"num_tokens": 94185325.0,
"step": 1060
},
{
"entropy": 0.979296875,
"epoch": 0.14624478917515205,
"grad_norm": 0.7521938219628429,
"learning_rate": 4.778427504579401e-06,
"loss": 0.6772,
"mean_token_accuracy": 0.815192025899887,
"num_tokens": 95085041.0,
"step": 1070
},
{
"entropy": 1.0005859375,
"epoch": 0.14761156290576094,
"grad_norm": 0.6764513015245464,
"learning_rate": 4.774904889389883e-06,
"loss": 0.7026,
"mean_token_accuracy": 0.8113694369792939,
"num_tokens": 96001882.0,
"step": 1080
},
{
"entropy": 1.01484375,
"epoch": 0.14897833663636986,
"grad_norm": 0.8723329887811756,
"learning_rate": 4.771382274200367e-06,
"loss": 0.7281,
"mean_token_accuracy": 0.8048442512750625,
"num_tokens": 96888127.0,
"step": 1090
},
{
"entropy": 1.0431640625,
"epoch": 0.15034511036697876,
"grad_norm": 0.7524074610959879,
"learning_rate": 4.76785965901085e-06,
"loss": 0.7786,
"mean_token_accuracy": 0.7962271630764007,
"num_tokens": 97815019.0,
"step": 1100
},
{
"entropy": 1.06953125,
"epoch": 0.15171188409758765,
"grad_norm": 0.8085016681122469,
"learning_rate": 4.764337043821333e-06,
"loss": 0.7988,
"mean_token_accuracy": 0.7902297347784042,
"num_tokens": 98683384.0,
"step": 1110
},
{
"entropy": 1.0294921875,
"epoch": 0.15307865782819655,
"grad_norm": 0.8313708780113097,
"learning_rate": 4.760814428631817e-06,
"loss": 0.734,
"mean_token_accuracy": 0.8045684903860092,
"num_tokens": 99564393.0,
"step": 1120
},
{
"entropy": 1.01484375,
"epoch": 0.15444543155880544,
"grad_norm": 0.8016234368718247,
"learning_rate": 4.7572918134423e-06,
"loss": 0.737,
"mean_token_accuracy": 0.8029660999774932,
"num_tokens": 100408247.0,
"step": 1130
},
{
"entropy": 0.99375,
"epoch": 0.15581220528941434,
"grad_norm": 0.7833278821575662,
"learning_rate": 4.753769198252783e-06,
"loss": 0.6942,
"mean_token_accuracy": 0.8125752955675125,
"num_tokens": 101297373.0,
"step": 1140
},
{
"entropy": 1.06328125,
"epoch": 0.15717897902002323,
"grad_norm": 0.8916382333941759,
"learning_rate": 4.7502465830632665e-06,
"loss": 0.7458,
"mean_token_accuracy": 0.8012790769338608,
"num_tokens": 102230124.0,
"step": 1150
},
{
"entropy": 0.9689453125,
"epoch": 0.15854575275063212,
"grad_norm": 0.7845876687017688,
"learning_rate": 4.74672396787375e-06,
"loss": 0.6952,
"mean_token_accuracy": 0.8131747364997863,
"num_tokens": 103126908.0,
"step": 1160
},
{
"entropy": 0.991796875,
"epoch": 0.15991252648124102,
"grad_norm": 0.6771244347120736,
"learning_rate": 4.743201352684233e-06,
"loss": 0.7051,
"mean_token_accuracy": 0.8113730013370514,
"num_tokens": 104024292.0,
"step": 1170
},
{
"entropy": 1.015234375,
"epoch": 0.16127930021184994,
"grad_norm": 0.8966323491513247,
"learning_rate": 4.739678737494716e-06,
"loss": 0.7244,
"mean_token_accuracy": 0.8083579182624817,
"num_tokens": 104898403.0,
"step": 1180
},
{
"entropy": 1.01171875,
"epoch": 0.16264607394245884,
"grad_norm": 0.7495637743411313,
"learning_rate": 4.7361561223052e-06,
"loss": 0.7262,
"mean_token_accuracy": 0.8062534421682358,
"num_tokens": 105782333.0,
"step": 1190
},
{
"entropy": 1.0017578125,
"epoch": 0.16401284767306773,
"grad_norm": 0.8542020320553048,
"learning_rate": 4.7326335071156834e-06,
"loss": 0.714,
"mean_token_accuracy": 0.8098825544118882,
"num_tokens": 106647573.0,
"step": 1200
},
{
"entropy": 1.0080078125,
"epoch": 0.16537962140367662,
"grad_norm": 0.7462671295917825,
"learning_rate": 4.729110891926166e-06,
"loss": 0.7325,
"mean_token_accuracy": 0.8040844023227691,
"num_tokens": 107566523.0,
"step": 1210
},
{
"entropy": 0.973046875,
"epoch": 0.16674639513428552,
"grad_norm": 0.9598011894069124,
"learning_rate": 4.72558827673665e-06,
"loss": 0.7074,
"mean_token_accuracy": 0.8103570759296417,
"num_tokens": 108414707.0,
"step": 1220
},
{
"entropy": 1.009375,
"epoch": 0.1681131688648944,
"grad_norm": 0.7683247344191266,
"learning_rate": 4.722065661547132e-06,
"loss": 0.7347,
"mean_token_accuracy": 0.8043894797563553,
"num_tokens": 109308818.0,
"step": 1230
},
{
"entropy": 1.008984375,
"epoch": 0.1694799425955033,
"grad_norm": 0.8707582006394764,
"learning_rate": 4.718543046357617e-06,
"loss": 0.7516,
"mean_token_accuracy": 0.8016739398241043,
"num_tokens": 110236510.0,
"step": 1240
},
{
"entropy": 0.974609375,
"epoch": 0.1708467163261122,
"grad_norm": 0.8750532030957177,
"learning_rate": 4.7150204311680995e-06,
"loss": 0.7006,
"mean_token_accuracy": 0.8115582317113876,
"num_tokens": 111071984.0,
"step": 1250
},
{
"entropy": 1.0400390625,
"epoch": 0.1722134900567211,
"grad_norm": 0.8694884789831856,
"learning_rate": 4.711497815978583e-06,
"loss": 0.7332,
"mean_token_accuracy": 0.8029281735420227,
"num_tokens": 111952526.0,
"step": 1260
},
{
"entropy": 1.02265625,
"epoch": 0.17358026378733002,
"grad_norm": 0.7235842229439367,
"learning_rate": 4.707975200789066e-06,
"loss": 0.7437,
"mean_token_accuracy": 0.8007719039916992,
"num_tokens": 112881665.0,
"step": 1270
},
{
"entropy": 1.0224609375,
"epoch": 0.1749470375179389,
"grad_norm": 0.842777076547381,
"learning_rate": 4.704452585599549e-06,
"loss": 0.7465,
"mean_token_accuracy": 0.8018406927585602,
"num_tokens": 113764242.0,
"step": 1280
},
{
"entropy": 1.0052734375,
"epoch": 0.1763138112485478,
"grad_norm": 0.756488203648542,
"learning_rate": 4.700929970410033e-06,
"loss": 0.711,
"mean_token_accuracy": 0.8092076689004898,
"num_tokens": 114647527.0,
"step": 1290
},
{
"entropy": 1.0154296875,
"epoch": 0.1776805849791567,
"grad_norm": 0.8028285830845061,
"learning_rate": 4.697407355220516e-06,
"loss": 0.73,
"mean_token_accuracy": 0.8059550434350967,
"num_tokens": 115562594.0,
"step": 1300
},
{
"entropy": 1.01875,
"epoch": 0.1790473587097656,
"grad_norm": 0.8000567126091751,
"learning_rate": 4.693884740030999e-06,
"loss": 0.7022,
"mean_token_accuracy": 0.8077812612056732,
"num_tokens": 116394915.0,
"step": 1310
},
{
"entropy": 1.03671875,
"epoch": 0.1804141324403745,
"grad_norm": 0.7642913728405574,
"learning_rate": 4.690362124841483e-06,
"loss": 0.7601,
"mean_token_accuracy": 0.7979720711708069,
"num_tokens": 117275706.0,
"step": 1320
},
{
"entropy": 1.03203125,
"epoch": 0.1817809061709834,
"grad_norm": 0.7421657118070474,
"learning_rate": 4.686839509651966e-06,
"loss": 0.7042,
"mean_token_accuracy": 0.8107543766498566,
"num_tokens": 118163200.0,
"step": 1330
},
{
"entropy": 1.007421875,
"epoch": 0.18314767990159228,
"grad_norm": 0.8444710636832825,
"learning_rate": 4.683316894462449e-06,
"loss": 0.7195,
"mean_token_accuracy": 0.8064839601516723,
"num_tokens": 119069034.0,
"step": 1340
},
{
"entropy": 0.9865234375,
"epoch": 0.18451445363220118,
"grad_norm": 0.8296751771679198,
"learning_rate": 4.679794279272933e-06,
"loss": 0.7358,
"mean_token_accuracy": 0.8030135661363602,
"num_tokens": 119965397.0,
"step": 1350
},
{
"entropy": 0.9943359375,
"epoch": 0.1858812273628101,
"grad_norm": 0.8472645763030341,
"learning_rate": 4.676271664083416e-06,
"loss": 0.7004,
"mean_token_accuracy": 0.8123631566762924,
"num_tokens": 120876126.0,
"step": 1360
},
{
"entropy": 1.0619140625,
"epoch": 0.187248001093419,
"grad_norm": 0.9158934142107483,
"learning_rate": 4.6727490488938995e-06,
"loss": 0.753,
"mean_token_accuracy": 0.801296254992485,
"num_tokens": 121712882.0,
"step": 1370
},
{
"entropy": 1.0125,
"epoch": 0.1886147748240279,
"grad_norm": 0.7480199962551197,
"learning_rate": 4.669226433704382e-06,
"loss": 0.7278,
"mean_token_accuracy": 0.8054364323616028,
"num_tokens": 122597154.0,
"step": 1380
},
{
"entropy": 1.021875,
"epoch": 0.18998154855463678,
"grad_norm": 0.825605760157343,
"learning_rate": 4.665703818514866e-06,
"loss": 0.7322,
"mean_token_accuracy": 0.8073795437812805,
"num_tokens": 123457814.0,
"step": 1390
},
{
"entropy": 1.026171875,
"epoch": 0.19134832228524568,
"grad_norm": 0.7269329092479063,
"learning_rate": 4.6621812033253484e-06,
"loss": 0.7048,
"mean_token_accuracy": 0.810889008641243,
"num_tokens": 124340451.0,
"step": 1400
},
{
"entropy": 0.94296875,
"epoch": 0.19271509601585457,
"grad_norm": 0.7635348281400209,
"learning_rate": 4.658658588135833e-06,
"loss": 0.648,
"mean_token_accuracy": 0.823466083407402,
"num_tokens": 125222917.0,
"step": 1410
},
{
"entropy": 0.9765625,
"epoch": 0.19408186974646346,
"grad_norm": 0.8931183599528667,
"learning_rate": 4.6551359729463155e-06,
"loss": 0.6972,
"mean_token_accuracy": 0.8123978108167649,
"num_tokens": 126167349.0,
"step": 1420
},
{
"entropy": 1.0275390625,
"epoch": 0.19544864347707236,
"grad_norm": 0.7834535507743707,
"learning_rate": 4.651613357756799e-06,
"loss": 0.7736,
"mean_token_accuracy": 0.7938032239675522,
"num_tokens": 127053792.0,
"step": 1430
},
{
"entropy": 1.0330078125,
"epoch": 0.19681541720768128,
"grad_norm": 0.792923725081107,
"learning_rate": 4.648090742567283e-06,
"loss": 0.7274,
"mean_token_accuracy": 0.8064889699220658,
"num_tokens": 127948703.0,
"step": 1440
},
{
"entropy": 1.0001953125,
"epoch": 0.19818219093829018,
"grad_norm": 0.8475826100161549,
"learning_rate": 4.644568127377765e-06,
"loss": 0.7125,
"mean_token_accuracy": 0.8083245992660523,
"num_tokens": 128861108.0,
"step": 1450
},
{
"entropy": 1.012890625,
"epoch": 0.19954896466889907,
"grad_norm": 0.8721043131649978,
"learning_rate": 4.641045512188249e-06,
"loss": 0.7544,
"mean_token_accuracy": 0.7999853760004043,
"num_tokens": 129749270.0,
"step": 1460
},
{
"entropy": 0.979296875,
"epoch": 0.20091573839950796,
"grad_norm": 0.7635743259149621,
"learning_rate": 4.6375228969987324e-06,
"loss": 0.7024,
"mean_token_accuracy": 0.811814370751381,
"num_tokens": 130655000.0,
"step": 1470
},
{
"entropy": 0.9955078125,
"epoch": 0.20228251213011686,
"grad_norm": 0.7674265965994246,
"learning_rate": 4.634000281809216e-06,
"loss": 0.7102,
"mean_token_accuracy": 0.8094262987375259,
"num_tokens": 131564630.0,
"step": 1480
},
{
"entropy": 0.995703125,
"epoch": 0.20364928586072575,
"grad_norm": 0.6924611121293675,
"learning_rate": 4.630477666619699e-06,
"loss": 0.7249,
"mean_token_accuracy": 0.8066471993923188,
"num_tokens": 132430033.0,
"step": 1490
},
{
"entropy": 0.96328125,
"epoch": 0.20501605959133465,
"grad_norm": 0.7961177821925648,
"learning_rate": 4.626955051430182e-06,
"loss": 0.6861,
"mean_token_accuracy": 0.8156867176294327,
"num_tokens": 133344093.0,
"step": 1500
},
{
"entropy": 1.032421875,
"epoch": 0.20638283332194354,
"grad_norm": 0.8339905127416445,
"learning_rate": 4.623432436240665e-06,
"loss": 0.7552,
"mean_token_accuracy": 0.7987716943025589,
"num_tokens": 134170710.0,
"step": 1510
},
{
"entropy": 0.9951171875,
"epoch": 0.20774960705255244,
"grad_norm": 0.8283930297814062,
"learning_rate": 4.619909821051149e-06,
"loss": 0.6984,
"mean_token_accuracy": 0.809878408908844,
"num_tokens": 135002300.0,
"step": 1520
},
{
"entropy": 1.0109375,
"epoch": 0.20911638078316136,
"grad_norm": 0.8766220659853394,
"learning_rate": 4.616387205861632e-06,
"loss": 0.7276,
"mean_token_accuracy": 0.805756589770317,
"num_tokens": 135877256.0,
"step": 1530
},
{
"entropy": 1.0060546875,
"epoch": 0.21048315451377025,
"grad_norm": 0.8321106282190728,
"learning_rate": 4.6128645906721156e-06,
"loss": 0.7311,
"mean_token_accuracy": 0.8064899802207947,
"num_tokens": 136719178.0,
"step": 1540
},
{
"entropy": 1.0564453125,
"epoch": 0.21184992824437915,
"grad_norm": 0.8129077073869128,
"learning_rate": 4.609341975482598e-06,
"loss": 0.7449,
"mean_token_accuracy": 0.8040707319974899,
"num_tokens": 137587067.0,
"step": 1550
},
{
"entropy": 0.99765625,
"epoch": 0.21321670197498804,
"grad_norm": 0.847220874180088,
"learning_rate": 4.605819360293082e-06,
"loss": 0.6782,
"mean_token_accuracy": 0.8150618463754654,
"num_tokens": 138510147.0,
"step": 1560
},
{
"entropy": 0.951953125,
"epoch": 0.21458347570559694,
"grad_norm": 0.7558106877783107,
"learning_rate": 4.602296745103565e-06,
"loss": 0.6512,
"mean_token_accuracy": 0.8207267612218857,
"num_tokens": 139391196.0,
"step": 1570
},
{
"entropy": 1.0076171875,
"epoch": 0.21595024943620583,
"grad_norm": 0.8176221638516442,
"learning_rate": 4.598774129914049e-06,
"loss": 0.77,
"mean_token_accuracy": 0.7984687000513077,
"num_tokens": 140283552.0,
"step": 1580
},
{
"entropy": 1.01953125,
"epoch": 0.21731702316681473,
"grad_norm": 0.9109111807472255,
"learning_rate": 4.595251514724532e-06,
"loss": 0.7441,
"mean_token_accuracy": 0.7998974859714508,
"num_tokens": 141150053.0,
"step": 1590
},
{
"entropy": 0.9740234375,
"epoch": 0.21868379689742362,
"grad_norm": 0.7481221402950338,
"learning_rate": 4.591728899535015e-06,
"loss": 0.698,
"mean_token_accuracy": 0.8094686448574067,
"num_tokens": 142039072.0,
"step": 1600
},
{
"entropy": 0.975390625,
"epoch": 0.22005057062803252,
"grad_norm": 0.8162086762228699,
"learning_rate": 4.588206284345499e-06,
"loss": 0.6629,
"mean_token_accuracy": 0.8204103320837021,
"num_tokens": 142931990.0,
"step": 1610
},
{
"entropy": 1.0525390625,
"epoch": 0.22141734435864144,
"grad_norm": 0.9415754290957723,
"learning_rate": 4.584683669155981e-06,
"loss": 0.7817,
"mean_token_accuracy": 0.7959058314561844,
"num_tokens": 143831487.0,
"step": 1620
},
{
"entropy": 1.01015625,
"epoch": 0.22278411808925033,
"grad_norm": 0.7532127210398842,
"learning_rate": 4.581161053966465e-06,
"loss": 0.723,
"mean_token_accuracy": 0.8083644956350327,
"num_tokens": 144782154.0,
"step": 1630
},
{
"entropy": 1.039453125,
"epoch": 0.22415089181985923,
"grad_norm": 0.9291991678203696,
"learning_rate": 4.5776384387769485e-06,
"loss": 0.7331,
"mean_token_accuracy": 0.804180845618248,
"num_tokens": 145684601.0,
"step": 1640
},
{
"entropy": 0.9919921875,
"epoch": 0.22551766555046812,
"grad_norm": 0.8860896080117581,
"learning_rate": 4.574115823587432e-06,
"loss": 0.704,
"mean_token_accuracy": 0.8120960474014283,
"num_tokens": 146574905.0,
"step": 1650
},
{
"entropy": 0.9787109375,
"epoch": 0.22688443928107702,
"grad_norm": 0.7471612820524387,
"learning_rate": 4.570593208397915e-06,
"loss": 0.7015,
"mean_token_accuracy": 0.8134039908647537,
"num_tokens": 147465145.0,
"step": 1660
},
{
"entropy": 0.9919921875,
"epoch": 0.2282512130116859,
"grad_norm": 0.7059414958953119,
"learning_rate": 4.567070593208398e-06,
"loss": 0.7362,
"mean_token_accuracy": 0.8049653559923172,
"num_tokens": 148329950.0,
"step": 1670
},
{
"entropy": 0.9970703125,
"epoch": 0.2296179867422948,
"grad_norm": 0.8804459122468066,
"learning_rate": 4.563547978018882e-06,
"loss": 0.7188,
"mean_token_accuracy": 0.8054957121610642,
"num_tokens": 149228860.0,
"step": 1680
},
{
"entropy": 0.9666015625,
"epoch": 0.2309847604729037,
"grad_norm": 0.8002198537260644,
"learning_rate": 4.560025362829365e-06,
"loss": 0.7111,
"mean_token_accuracy": 0.8086749643087388,
"num_tokens": 150100793.0,
"step": 1690
},
{
"entropy": 1.012109375,
"epoch": 0.23235153420351262,
"grad_norm": 0.7292242259202798,
"learning_rate": 4.556502747639848e-06,
"loss": 0.7278,
"mean_token_accuracy": 0.8041989475488662,
"num_tokens": 150995094.0,
"step": 1700
},
{
"entropy": 1.015625,
"epoch": 0.23371830793412152,
"grad_norm": 0.7332247286971317,
"learning_rate": 4.552980132450332e-06,
"loss": 0.746,
"mean_token_accuracy": 0.8005778104066849,
"num_tokens": 151921452.0,
"step": 1710
},
{
"entropy": 0.9806640625,
"epoch": 0.2350850816647304,
"grad_norm": 0.687928089254978,
"learning_rate": 4.549457517260814e-06,
"loss": 0.725,
"mean_token_accuracy": 0.8087247878313064,
"num_tokens": 152829538.0,
"step": 1720
},
{
"entropy": 0.9595703125,
"epoch": 0.2364518553953393,
"grad_norm": 0.8247617411155244,
"learning_rate": 4.545934902071298e-06,
"loss": 0.6888,
"mean_token_accuracy": 0.8110770285129547,
"num_tokens": 153770075.0,
"step": 1730
},
{
"entropy": 1.022265625,
"epoch": 0.2378186291259482,
"grad_norm": 0.8750687720719393,
"learning_rate": 4.5424122868817814e-06,
"loss": 0.7186,
"mean_token_accuracy": 0.80419262945652,
"num_tokens": 154631539.0,
"step": 1740
},
{
"entropy": 0.980078125,
"epoch": 0.2391854028565571,
"grad_norm": 0.756663636288404,
"learning_rate": 4.538889671692265e-06,
"loss": 0.6916,
"mean_token_accuracy": 0.8136232405900955,
"num_tokens": 155495251.0,
"step": 1750
},
{
"entropy": 1.0119140625,
"epoch": 0.240552176587166,
"grad_norm": 0.8096119032834547,
"learning_rate": 4.535367056502748e-06,
"loss": 0.7068,
"mean_token_accuracy": 0.8078958243131638,
"num_tokens": 156404778.0,
"step": 1760
},
{
"entropy": 1.0177734375,
"epoch": 0.24191895031777488,
"grad_norm": 0.7317193109965263,
"learning_rate": 4.531844441313231e-06,
"loss": 0.732,
"mean_token_accuracy": 0.8033632397651672,
"num_tokens": 157289886.0,
"step": 1770
},
{
"entropy": 0.9609375,
"epoch": 0.24328572404838378,
"grad_norm": 0.736891483747059,
"learning_rate": 4.528321826123715e-06,
"loss": 0.6813,
"mean_token_accuracy": 0.8158172100782395,
"num_tokens": 158159238.0,
"step": 1780
},
{
"entropy": 1.009765625,
"epoch": 0.2446524977789927,
"grad_norm": 0.7204321824859703,
"learning_rate": 4.524799210934198e-06,
"loss": 0.7048,
"mean_token_accuracy": 0.8121180057525634,
"num_tokens": 159054886.0,
"step": 1790
},
{
"entropy": 1.0060546875,
"epoch": 0.2460192715096016,
"grad_norm": 0.7149468251168896,
"learning_rate": 4.521276595744681e-06,
"loss": 0.6679,
"mean_token_accuracy": 0.8205239117145539,
"num_tokens": 159916702.0,
"step": 1800
},
{
"entropy": 0.9998046875,
"epoch": 0.2473860452402105,
"grad_norm": 0.8442014196503996,
"learning_rate": 4.5177539805551646e-06,
"loss": 0.7433,
"mean_token_accuracy": 0.8023593068122864,
"num_tokens": 160800051.0,
"step": 1810
},
{
"entropy": 0.9833984375,
"epoch": 0.24875281897081938,
"grad_norm": 0.8159661938327386,
"learning_rate": 4.514231365365648e-06,
"loss": 0.6677,
"mean_token_accuracy": 0.8191746413707733,
"num_tokens": 161649504.0,
"step": 1820
},
{
"entropy": 0.9830078125,
"epoch": 0.2501195927014283,
"grad_norm": 0.7508219603630336,
"learning_rate": 4.510708750176131e-06,
"loss": 0.6946,
"mean_token_accuracy": 0.81060032248497,
"num_tokens": 162521710.0,
"step": 1830
},
{
"entropy": 0.9869140625,
"epoch": 0.2514863664320372,
"grad_norm": 0.7730010913545102,
"learning_rate": 4.507186134986614e-06,
"loss": 0.7031,
"mean_token_accuracy": 0.8124055057764054,
"num_tokens": 163456795.0,
"step": 1840
},
{
"entropy": 0.9806640625,
"epoch": 0.2528531401626461,
"grad_norm": 0.6958553702964516,
"learning_rate": 4.503663519797098e-06,
"loss": 0.7068,
"mean_token_accuracy": 0.8096126854419708,
"num_tokens": 164354274.0,
"step": 1850
},
{
"entropy": 0.9927734375,
"epoch": 0.254219913893255,
"grad_norm": 0.8448851293412462,
"learning_rate": 4.5001409046075814e-06,
"loss": 0.7018,
"mean_token_accuracy": 0.8108880281448364,
"num_tokens": 165231568.0,
"step": 1860
},
{
"entropy": 0.9828125,
"epoch": 0.2555866876238639,
"grad_norm": 0.7371766618136034,
"learning_rate": 4.496618289418064e-06,
"loss": 0.6914,
"mean_token_accuracy": 0.8130333214998245,
"num_tokens": 166149414.0,
"step": 1870
},
{
"entropy": 0.99375,
"epoch": 0.2569534613544728,
"grad_norm": 0.8456402417119336,
"learning_rate": 4.493095674228548e-06,
"loss": 0.6964,
"mean_token_accuracy": 0.8135116070508956,
"num_tokens": 167006969.0,
"step": 1880
},
{
"entropy": 1.0048828125,
"epoch": 0.2583202350850817,
"grad_norm": 0.8310391775901148,
"learning_rate": 4.48957305903903e-06,
"loss": 0.7316,
"mean_token_accuracy": 0.8028609335422516,
"num_tokens": 167901440.0,
"step": 1890
},
{
"entropy": 0.9845703125,
"epoch": 0.25968700881569057,
"grad_norm": 0.8516327638833464,
"learning_rate": 4.486050443849515e-06,
"loss": 0.6746,
"mean_token_accuracy": 0.818128016591072,
"num_tokens": 168783114.0,
"step": 1900
},
{
"entropy": 0.995703125,
"epoch": 0.26105378254629946,
"grad_norm": 0.7377855307296071,
"learning_rate": 4.4825278286599975e-06,
"loss": 0.6959,
"mean_token_accuracy": 0.811351990699768,
"num_tokens": 169685621.0,
"step": 1910
},
{
"entropy": 0.9578125,
"epoch": 0.26242055627690836,
"grad_norm": 0.9050761446451233,
"learning_rate": 4.479005213470481e-06,
"loss": 0.6891,
"mean_token_accuracy": 0.8134806841611862,
"num_tokens": 170573171.0,
"step": 1920
},
{
"entropy": 1.00234375,
"epoch": 0.26378733000751725,
"grad_norm": 0.8845761670947171,
"learning_rate": 4.475482598280964e-06,
"loss": 0.7598,
"mean_token_accuracy": 0.7947903752326966,
"num_tokens": 171454806.0,
"step": 1930
},
{
"entropy": 0.970703125,
"epoch": 0.26515410373812615,
"grad_norm": 0.7878844431938701,
"learning_rate": 4.471959983091447e-06,
"loss": 0.7025,
"mean_token_accuracy": 0.8096184253692627,
"num_tokens": 172346713.0,
"step": 1940
},
{
"entropy": 1.0125,
"epoch": 0.26652087746873504,
"grad_norm": 0.7853323089591413,
"learning_rate": 4.468437367901931e-06,
"loss": 0.6854,
"mean_token_accuracy": 0.8137507259845733,
"num_tokens": 173233962.0,
"step": 1950
},
{
"entropy": 1.01484375,
"epoch": 0.26788765119934393,
"grad_norm": 0.8166934619465657,
"learning_rate": 4.464914752712414e-06,
"loss": 0.7466,
"mean_token_accuracy": 0.8011311709880828,
"num_tokens": 174110792.0,
"step": 1960
},
{
"entropy": 0.9673828125,
"epoch": 0.26925442492995283,
"grad_norm": 0.7599205154062314,
"learning_rate": 4.461392137522897e-06,
"loss": 0.6945,
"mean_token_accuracy": 0.8110936760902405,
"num_tokens": 175005864.0,
"step": 1970
},
{
"entropy": 0.9701171875,
"epoch": 0.2706211986605617,
"grad_norm": 0.7604279915797904,
"learning_rate": 4.457869522333381e-06,
"loss": 0.6918,
"mean_token_accuracy": 0.8137934476137161,
"num_tokens": 175870222.0,
"step": 1980
},
{
"entropy": 0.991015625,
"epoch": 0.2719879723911706,
"grad_norm": 0.7812829801624434,
"learning_rate": 4.454346907143864e-06,
"loss": 0.69,
"mean_token_accuracy": 0.8140656471252441,
"num_tokens": 176750406.0,
"step": 1990
},
{
"entropy": 1.0287109375,
"epoch": 0.2733547461217795,
"grad_norm": 0.8131696981995804,
"learning_rate": 4.450824291954347e-06,
"loss": 0.7049,
"mean_token_accuracy": 0.8098706394433975,
"num_tokens": 177646304.0,
"step": 2000
},
{
"entropy": 0.95859375,
"epoch": 0.27472151985238846,
"grad_norm": 0.6889718612224057,
"learning_rate": 4.44730167676483e-06,
"loss": 0.6687,
"mean_token_accuracy": 0.8189610451459884,
"num_tokens": 178544048.0,
"step": 2010
},
{
"entropy": 1.0265625,
"epoch": 0.27608829358299736,
"grad_norm": 0.7736286797675228,
"learning_rate": 4.443779061575314e-06,
"loss": 0.7506,
"mean_token_accuracy": 0.8027170360088348,
"num_tokens": 179424878.0,
"step": 2020
},
{
"entropy": 0.9451171875,
"epoch": 0.27745506731360625,
"grad_norm": 0.8609642988664381,
"learning_rate": 4.4402564463857975e-06,
"loss": 0.6705,
"mean_token_accuracy": 0.8172019362449646,
"num_tokens": 180290131.0,
"step": 2030
},
{
"entropy": 1.0146484375,
"epoch": 0.27882184104421515,
"grad_norm": 0.7323221383340515,
"learning_rate": 4.43673383119628e-06,
"loss": 0.705,
"mean_token_accuracy": 0.8105100184679032,
"num_tokens": 181151230.0,
"step": 2040
},
{
"entropy": 0.984765625,
"epoch": 0.28018861477482404,
"grad_norm": 0.781060914423855,
"learning_rate": 4.433211216006764e-06,
"loss": 0.6893,
"mean_token_accuracy": 0.8120083957910538,
"num_tokens": 181997158.0,
"step": 2050
},
{
"entropy": 0.9693359375,
"epoch": 0.28155538850543294,
"grad_norm": 0.7290147987290493,
"learning_rate": 4.4296886008172465e-06,
"loss": 0.7035,
"mean_token_accuracy": 0.8097601443529129,
"num_tokens": 182869862.0,
"step": 2060
},
{
"entropy": 0.9619140625,
"epoch": 0.28292216223604183,
"grad_norm": 0.7714751602716319,
"learning_rate": 4.426165985627731e-06,
"loss": 0.6761,
"mean_token_accuracy": 0.8158272117376327,
"num_tokens": 183793317.0,
"step": 2070
},
{
"entropy": 0.955078125,
"epoch": 0.2842889359666507,
"grad_norm": 0.7816720782909408,
"learning_rate": 4.4226433704382136e-06,
"loss": 0.6904,
"mean_token_accuracy": 0.8155831098556519,
"num_tokens": 184654232.0,
"step": 2080
},
{
"entropy": 0.965625,
"epoch": 0.2856557096972596,
"grad_norm": 0.7560046469077146,
"learning_rate": 4.419120755248697e-06,
"loss": 0.676,
"mean_token_accuracy": 0.817731311917305,
"num_tokens": 185544158.0,
"step": 2090
},
{
"entropy": 1.045703125,
"epoch": 0.2870224834278685,
"grad_norm": 0.835230784702763,
"learning_rate": 4.41559814005918e-06,
"loss": 0.763,
"mean_token_accuracy": 0.7979078769683838,
"num_tokens": 186444139.0,
"step": 2100
},
{
"entropy": 0.9646484375,
"epoch": 0.2883892571584774,
"grad_norm": 0.8181221652523085,
"learning_rate": 4.412075524869663e-06,
"loss": 0.7016,
"mean_token_accuracy": 0.8126494795083999,
"num_tokens": 187281267.0,
"step": 2110
},
{
"entropy": 1.0123046875,
"epoch": 0.2897560308890863,
"grad_norm": 0.8280280751764884,
"learning_rate": 4.408552909680147e-06,
"loss": 0.7298,
"mean_token_accuracy": 0.8047745078802109,
"num_tokens": 188147941.0,
"step": 2120
},
{
"entropy": 0.9775390625,
"epoch": 0.2911228046196952,
"grad_norm": 0.7273997926519497,
"learning_rate": 4.4050302944906304e-06,
"loss": 0.7134,
"mean_token_accuracy": 0.807851466536522,
"num_tokens": 189047615.0,
"step": 2130
},
{
"entropy": 0.98046875,
"epoch": 0.2924895783503041,
"grad_norm": 0.8268163021354655,
"learning_rate": 4.401507679301113e-06,
"loss": 0.6803,
"mean_token_accuracy": 0.8158050328493118,
"num_tokens": 189900274.0,
"step": 2140
},
{
"entropy": 1.0568359375,
"epoch": 0.293856352080913,
"grad_norm": 0.9180346360363243,
"learning_rate": 4.397985064111597e-06,
"loss": 0.7875,
"mean_token_accuracy": 0.7931036829948426,
"num_tokens": 190817864.0,
"step": 2150
},
{
"entropy": 0.988671875,
"epoch": 0.2952231258115219,
"grad_norm": 0.7336535602841817,
"learning_rate": 4.39446244892208e-06,
"loss": 0.6935,
"mean_token_accuracy": 0.8136212021112442,
"num_tokens": 191696200.0,
"step": 2160
},
{
"entropy": 1.0166015625,
"epoch": 0.2965898995421308,
"grad_norm": 0.886861084278737,
"learning_rate": 4.390939833732563e-06,
"loss": 0.7097,
"mean_token_accuracy": 0.808835831284523,
"num_tokens": 192504212.0,
"step": 2170
},
{
"entropy": 0.9876953125,
"epoch": 0.2979566732727397,
"grad_norm": 0.8363225202689489,
"learning_rate": 4.387417218543047e-06,
"loss": 0.7053,
"mean_token_accuracy": 0.8108400017023086,
"num_tokens": 193391915.0,
"step": 2180
},
{
"entropy": 0.95859375,
"epoch": 0.2993234470033486,
"grad_norm": 0.8037657959715913,
"learning_rate": 4.38389460335353e-06,
"loss": 0.6543,
"mean_token_accuracy": 0.8221141487360001,
"num_tokens": 194274927.0,
"step": 2190
},
{
"entropy": 0.9845703125,
"epoch": 0.3006902207339575,
"grad_norm": 0.7011082784919492,
"learning_rate": 4.380371988164014e-06,
"loss": 0.7561,
"mean_token_accuracy": 0.7989371389150619,
"num_tokens": 195216521.0,
"step": 2200
},
{
"entropy": 0.9630859375,
"epoch": 0.3020569944645664,
"grad_norm": 0.7642499115371695,
"learning_rate": 4.376849372974496e-06,
"loss": 0.6951,
"mean_token_accuracy": 0.815863487124443,
"num_tokens": 196130208.0,
"step": 2210
},
{
"entropy": 0.9712890625,
"epoch": 0.3034237681951753,
"grad_norm": 0.745489572351828,
"learning_rate": 4.37332675778498e-06,
"loss": 0.7022,
"mean_token_accuracy": 0.8099752932786941,
"num_tokens": 197028743.0,
"step": 2220
},
{
"entropy": 0.980078125,
"epoch": 0.3047905419257842,
"grad_norm": 0.7510540551729908,
"learning_rate": 4.369804142595463e-06,
"loss": 0.7318,
"mean_token_accuracy": 0.8064709633588791,
"num_tokens": 197898713.0,
"step": 2230
},
{
"entropy": 0.962890625,
"epoch": 0.3061573156563931,
"grad_norm": 0.8020047593711526,
"learning_rate": 4.366281527405947e-06,
"loss": 0.711,
"mean_token_accuracy": 0.8066793024539948,
"num_tokens": 198815509.0,
"step": 2240
},
{
"entropy": 0.951171875,
"epoch": 0.307524089387002,
"grad_norm": 0.8069944158701267,
"learning_rate": 4.36275891221643e-06,
"loss": 0.6755,
"mean_token_accuracy": 0.8139858067035675,
"num_tokens": 199704367.0,
"step": 2250
},
{
"entropy": 0.93984375,
"epoch": 0.3088908631176109,
"grad_norm": 0.8177964372292853,
"learning_rate": 4.359236297026913e-06,
"loss": 0.6852,
"mean_token_accuracy": 0.8142912417650223,
"num_tokens": 200578924.0,
"step": 2260
},
{
"entropy": 0.946875,
"epoch": 0.3102576368482198,
"grad_norm": 0.8216380351462543,
"learning_rate": 4.355713681837396e-06,
"loss": 0.6582,
"mean_token_accuracy": 0.8201599180698395,
"num_tokens": 201456944.0,
"step": 2270
},
{
"entropy": 0.9552734375,
"epoch": 0.31162441057882867,
"grad_norm": 0.7884378296489564,
"learning_rate": 4.352191066647879e-06,
"loss": 0.6895,
"mean_token_accuracy": 0.8122224867343902,
"num_tokens": 202323380.0,
"step": 2280
},
{
"entropy": 0.994921875,
"epoch": 0.31299118430943756,
"grad_norm": 0.7892620779005551,
"learning_rate": 4.348668451458363e-06,
"loss": 0.7161,
"mean_token_accuracy": 0.8072948008775711,
"num_tokens": 203221545.0,
"step": 2290
},
{
"entropy": 1.0029296875,
"epoch": 0.31435795804004646,
"grad_norm": 0.7558858150508347,
"learning_rate": 4.3451458362688465e-06,
"loss": 0.7117,
"mean_token_accuracy": 0.8091292232275009,
"num_tokens": 204099161.0,
"step": 2300
},
{
"entropy": 1.003125,
"epoch": 0.31572473177065535,
"grad_norm": 0.7308076990056759,
"learning_rate": 4.34162322107933e-06,
"loss": 0.6999,
"mean_token_accuracy": 0.8117423176765441,
"num_tokens": 204961852.0,
"step": 2310
},
{
"entropy": 1.01953125,
"epoch": 0.31709150550126425,
"grad_norm": 0.785979279255561,
"learning_rate": 4.338100605889813e-06,
"loss": 0.6902,
"mean_token_accuracy": 0.8124176949262619,
"num_tokens": 205811402.0,
"step": 2320
},
{
"entropy": 0.9513671875,
"epoch": 0.31845827923187314,
"grad_norm": 0.803249349052139,
"learning_rate": 4.334577990700296e-06,
"loss": 0.6278,
"mean_token_accuracy": 0.8293322622776031,
"num_tokens": 206683323.0,
"step": 2330
},
{
"entropy": 1.00859375,
"epoch": 0.31982505296248204,
"grad_norm": 0.7844225703112512,
"learning_rate": 4.33105537551078e-06,
"loss": 0.7451,
"mean_token_accuracy": 0.8004480391740799,
"num_tokens": 207548993.0,
"step": 2340
},
{
"entropy": 0.9572265625,
"epoch": 0.32119182669309093,
"grad_norm": 0.8091187875476478,
"learning_rate": 4.327532760321263e-06,
"loss": 0.6571,
"mean_token_accuracy": 0.820944881439209,
"num_tokens": 208393459.0,
"step": 2350
},
{
"entropy": 0.9833984375,
"epoch": 0.3225586004236999,
"grad_norm": 0.8166771879268846,
"learning_rate": 4.324010145131746e-06,
"loss": 0.7292,
"mean_token_accuracy": 0.8047452986240387,
"num_tokens": 209266863.0,
"step": 2360
},
{
"entropy": 0.9810546875,
"epoch": 0.3239253741543088,
"grad_norm": 0.7322137311293395,
"learning_rate": 4.32048752994223e-06,
"loss": 0.7148,
"mean_token_accuracy": 0.8070944160223007,
"num_tokens": 210197901.0,
"step": 2370
},
{
"entropy": 1.000390625,
"epoch": 0.32529214788491767,
"grad_norm": 0.7553126971385115,
"learning_rate": 4.316964914752712e-06,
"loss": 0.7142,
"mean_token_accuracy": 0.8068336486816406,
"num_tokens": 211082714.0,
"step": 2380
},
{
"entropy": 0.9994140625,
"epoch": 0.32665892161552657,
"grad_norm": 0.8884126674901496,
"learning_rate": 4.313442299563196e-06,
"loss": 0.7746,
"mean_token_accuracy": 0.7983190208673477,
"num_tokens": 211920987.0,
"step": 2390
},
{
"entropy": 0.9744140625,
"epoch": 0.32802569534613546,
"grad_norm": 0.8707089116317029,
"learning_rate": 4.3099196843736794e-06,
"loss": 0.6934,
"mean_token_accuracy": 0.812601774930954,
"num_tokens": 212778254.0,
"step": 2400
},
{
"entropy": 0.998046875,
"epoch": 0.32939246907674435,
"grad_norm": 0.7664866423315292,
"learning_rate": 4.306397069184163e-06,
"loss": 0.7009,
"mean_token_accuracy": 0.8098969250917435,
"num_tokens": 213665677.0,
"step": 2410
},
{
"entropy": 1.0125,
"epoch": 0.33075924280735325,
"grad_norm": 0.8773471869082041,
"learning_rate": 4.302874453994646e-06,
"loss": 0.7173,
"mean_token_accuracy": 0.8066916465759277,
"num_tokens": 214532016.0,
"step": 2420
},
{
"entropy": 0.962109375,
"epoch": 0.33212601653796214,
"grad_norm": 0.7324837251756169,
"learning_rate": 4.299351838805129e-06,
"loss": 0.6603,
"mean_token_accuracy": 0.8183428555727005,
"num_tokens": 215416984.0,
"step": 2430
},
{
"entropy": 1.019140625,
"epoch": 0.33349279026857104,
"grad_norm": 0.6917979461083614,
"learning_rate": 4.295829223615613e-06,
"loss": 0.7063,
"mean_token_accuracy": 0.8099083662033081,
"num_tokens": 216317310.0,
"step": 2440
},
{
"entropy": 1.0029296875,
"epoch": 0.33485956399917993,
"grad_norm": 0.801216952535054,
"learning_rate": 4.292306608426096e-06,
"loss": 0.7051,
"mean_token_accuracy": 0.8117168486118317,
"num_tokens": 217188245.0,
"step": 2450
},
{
"entropy": 0.939453125,
"epoch": 0.3362263377297888,
"grad_norm": 0.7052507057120119,
"learning_rate": 4.288783993236579e-06,
"loss": 0.6797,
"mean_token_accuracy": 0.8175667226314545,
"num_tokens": 218084772.0,
"step": 2460
},
{
"entropy": 0.990625,
"epoch": 0.3375931114603977,
"grad_norm": 0.8356143964022139,
"learning_rate": 4.285261378047063e-06,
"loss": 0.6847,
"mean_token_accuracy": 0.8129307985305786,
"num_tokens": 218985412.0,
"step": 2470
},
{
"entropy": 1.00390625,
"epoch": 0.3389598851910066,
"grad_norm": 0.7304053917760452,
"learning_rate": 4.281738762857546e-06,
"loss": 0.7124,
"mean_token_accuracy": 0.8076407134532928,
"num_tokens": 219878663.0,
"step": 2480
},
{
"entropy": 1.0322265625,
"epoch": 0.3403266589216155,
"grad_norm": 0.8276748109369251,
"learning_rate": 4.278216147668029e-06,
"loss": 0.7046,
"mean_token_accuracy": 0.809708696603775,
"num_tokens": 220775855.0,
"step": 2490
},
{
"entropy": 0.9978515625,
"epoch": 0.3416934326522244,
"grad_norm": 0.8007716765163533,
"learning_rate": 4.274693532478512e-06,
"loss": 0.745,
"mean_token_accuracy": 0.7994920641183854,
"num_tokens": 221663073.0,
"step": 2500
},
{
"entropy": 0.9521484375,
"epoch": 0.3430602063828333,
"grad_norm": 0.6921537029108707,
"learning_rate": 4.271170917288996e-06,
"loss": 0.6863,
"mean_token_accuracy": 0.8150428891181946,
"num_tokens": 222544981.0,
"step": 2510
},
{
"entropy": 0.9712890625,
"epoch": 0.3444269801134422,
"grad_norm": 0.7862544922678399,
"learning_rate": 4.2676483020994795e-06,
"loss": 0.6716,
"mean_token_accuracy": 0.8171723186969757,
"num_tokens": 223440024.0,
"step": 2520
},
{
"entropy": 0.9923828125,
"epoch": 0.34579375384405114,
"grad_norm": 0.765878884777755,
"learning_rate": 4.264125686909962e-06,
"loss": 0.7316,
"mean_token_accuracy": 0.8050745308399201,
"num_tokens": 224363546.0,
"step": 2530
},
{
"entropy": 1.003125,
"epoch": 0.34716052757466004,
"grad_norm": 0.7731153031217866,
"learning_rate": 4.260603071720446e-06,
"loss": 0.7327,
"mean_token_accuracy": 0.8071545690298081,
"num_tokens": 225225542.0,
"step": 2540
},
{
"entropy": 1.0205078125,
"epoch": 0.34852730130526893,
"grad_norm": 0.8637954501797291,
"learning_rate": 4.257080456530928e-06,
"loss": 0.7476,
"mean_token_accuracy": 0.7991413682699203,
"num_tokens": 226103209.0,
"step": 2550
},
{
"entropy": 0.9673828125,
"epoch": 0.3498940750358778,
"grad_norm": 0.7907064840976512,
"learning_rate": 4.253557841341413e-06,
"loss": 0.6642,
"mean_token_accuracy": 0.8191326141357422,
"num_tokens": 226998581.0,
"step": 2560
},
{
"entropy": 0.976953125,
"epoch": 0.3512608487664867,
"grad_norm": 0.8184717769752532,
"learning_rate": 4.2500352261518955e-06,
"loss": 0.706,
"mean_token_accuracy": 0.8100301504135132,
"num_tokens": 227888802.0,
"step": 2570
},
{
"entropy": 0.9384765625,
"epoch": 0.3526276224970956,
"grad_norm": 0.7785325505411812,
"learning_rate": 4.246512610962379e-06,
"loss": 0.683,
"mean_token_accuracy": 0.8163754999637604,
"num_tokens": 228776580.0,
"step": 2580
},
{
"entropy": 1.0029296875,
"epoch": 0.3539943962277045,
"grad_norm": 0.8304272339137139,
"learning_rate": 4.242989995772862e-06,
"loss": 0.7185,
"mean_token_accuracy": 0.8077067404985427,
"num_tokens": 229624559.0,
"step": 2590
},
{
"entropy": 1.002734375,
"epoch": 0.3553611699583134,
"grad_norm": 0.6779037886749544,
"learning_rate": 4.239467380583345e-06,
"loss": 0.7055,
"mean_token_accuracy": 0.8099199414253235,
"num_tokens": 230559792.0,
"step": 2600
},
{
"entropy": 0.9763671875,
"epoch": 0.3567279436889223,
"grad_norm": 0.6906803362204267,
"learning_rate": 4.235944765393829e-06,
"loss": 0.6661,
"mean_token_accuracy": 0.819983783364296,
"num_tokens": 231437280.0,
"step": 2610
},
{
"entropy": 0.99296875,
"epoch": 0.3580947174195312,
"grad_norm": 0.9035824907549277,
"learning_rate": 4.232422150204312e-06,
"loss": 0.7283,
"mean_token_accuracy": 0.8047367334365845,
"num_tokens": 232292698.0,
"step": 2620
},
{
"entropy": 1.0294921875,
"epoch": 0.3594614911501401,
"grad_norm": 0.7744496946576592,
"learning_rate": 4.228899535014795e-06,
"loss": 0.7552,
"mean_token_accuracy": 0.7989694058895112,
"num_tokens": 233148286.0,
"step": 2630
},
{
"entropy": 1.0064453125,
"epoch": 0.360828264880749,
"grad_norm": 0.9173095592297467,
"learning_rate": 4.225376919825279e-06,
"loss": 0.7258,
"mean_token_accuracy": 0.8056702435016632,
"num_tokens": 234027940.0,
"step": 2640
},
{
"entropy": 0.9634765625,
"epoch": 0.3621950386113579,
"grad_norm": 0.7371413380970807,
"learning_rate": 4.221854304635762e-06,
"loss": 0.6982,
"mean_token_accuracy": 0.8101388037204742,
"num_tokens": 234909727.0,
"step": 2650
},
{
"entropy": 1.046875,
"epoch": 0.3635618123419668,
"grad_norm": 0.7304181082590024,
"learning_rate": 4.218331689446245e-06,
"loss": 0.7666,
"mean_token_accuracy": 0.7969495594501496,
"num_tokens": 235799025.0,
"step": 2660
},
{
"entropy": 0.9736328125,
"epoch": 0.36492858607257567,
"grad_norm": 0.8518789203203548,
"learning_rate": 4.2148090742567284e-06,
"loss": 0.6646,
"mean_token_accuracy": 0.8204658299684524,
"num_tokens": 236681681.0,
"step": 2670
},
{
"entropy": 0.9708984375,
"epoch": 0.36629535980318456,
"grad_norm": 0.755689661257198,
"learning_rate": 4.211286459067212e-06,
"loss": 0.7329,
"mean_token_accuracy": 0.8037692099809647,
"num_tokens": 237614496.0,
"step": 2680
},
{
"entropy": 0.9583984375,
"epoch": 0.36766213353379346,
"grad_norm": 0.6804297935364744,
"learning_rate": 4.2077638438776955e-06,
"loss": 0.6544,
"mean_token_accuracy": 0.8217035442590713,
"num_tokens": 238469453.0,
"step": 2690
},
{
"entropy": 0.9783203125,
"epoch": 0.36902890726440235,
"grad_norm": 0.7184222037041044,
"learning_rate": 4.204241228688178e-06,
"loss": 0.6799,
"mean_token_accuracy": 0.8147988736629486,
"num_tokens": 239444806.0,
"step": 2700
},
{
"entropy": 0.9693359375,
"epoch": 0.3703956809950113,
"grad_norm": 0.7348063464836252,
"learning_rate": 4.200718613498662e-06,
"loss": 0.6603,
"mean_token_accuracy": 0.8200602173805237,
"num_tokens": 240347408.0,
"step": 2710
},
{
"entropy": 0.9755859375,
"epoch": 0.3717624547256202,
"grad_norm": 0.8602240389525622,
"learning_rate": 4.1971959983091445e-06,
"loss": 0.6744,
"mean_token_accuracy": 0.8159934461116791,
"num_tokens": 241208094.0,
"step": 2720
},
{
"entropy": 0.95234375,
"epoch": 0.3731292284562291,
"grad_norm": 0.6965065486614727,
"learning_rate": 4.193673383119629e-06,
"loss": 0.6587,
"mean_token_accuracy": 0.820703050494194,
"num_tokens": 242123416.0,
"step": 2730
},
{
"entropy": 1.0080078125,
"epoch": 0.374496002186838,
"grad_norm": 0.7946279212246405,
"learning_rate": 4.190150767930112e-06,
"loss": 0.7739,
"mean_token_accuracy": 0.7970874726772308,
"num_tokens": 243074693.0,
"step": 2740
},
{
"entropy": 1.0220703125,
"epoch": 0.3758627759174469,
"grad_norm": 0.7840817807307842,
"learning_rate": 4.186628152740595e-06,
"loss": 0.7217,
"mean_token_accuracy": 0.8064022898674011,
"num_tokens": 243981066.0,
"step": 2750
},
{
"entropy": 0.9830078125,
"epoch": 0.3772295496480558,
"grad_norm": 0.754668589083318,
"learning_rate": 4.183105537551078e-06,
"loss": 0.7004,
"mean_token_accuracy": 0.8104293555021286,
"num_tokens": 244865595.0,
"step": 2760
},
{
"entropy": 1.0068359375,
"epoch": 0.37859632337866467,
"grad_norm": 0.8722189595547772,
"learning_rate": 4.179582922361561e-06,
"loss": 0.7266,
"mean_token_accuracy": 0.8056835323572159,
"num_tokens": 245726577.0,
"step": 2770
},
{
"entropy": 0.9908203125,
"epoch": 0.37996309710927356,
"grad_norm": 0.8191892196368383,
"learning_rate": 4.176060307172045e-06,
"loss": 0.7024,
"mean_token_accuracy": 0.8074427217245101,
"num_tokens": 246639968.0,
"step": 2780
},
{
"entropy": 0.9916015625,
"epoch": 0.38132987083988246,
"grad_norm": 0.7224678464429879,
"learning_rate": 4.1725376919825285e-06,
"loss": 0.6744,
"mean_token_accuracy": 0.8165394067764282,
"num_tokens": 247494604.0,
"step": 2790
},
{
"entropy": 0.98203125,
"epoch": 0.38269664457049135,
"grad_norm": 0.7802962727235394,
"learning_rate": 4.169015076793011e-06,
"loss": 0.7103,
"mean_token_accuracy": 0.8092028260231018,
"num_tokens": 248346580.0,
"step": 2800
},
{
"entropy": 0.9431640625,
"epoch": 0.38406341830110025,
"grad_norm": 0.8552953903399854,
"learning_rate": 4.165492461603495e-06,
"loss": 0.6834,
"mean_token_accuracy": 0.8133321195840836,
"num_tokens": 249243321.0,
"step": 2810
},
{
"entropy": 0.9630859375,
"epoch": 0.38543019203170914,
"grad_norm": 0.799419353811644,
"learning_rate": 4.161969846413978e-06,
"loss": 0.6819,
"mean_token_accuracy": 0.815095180273056,
"num_tokens": 250100768.0,
"step": 2820
},
{
"entropy": 0.921875,
"epoch": 0.38679696576231803,
"grad_norm": 0.7149362566831643,
"learning_rate": 4.158447231224461e-06,
"loss": 0.6698,
"mean_token_accuracy": 0.817307648062706,
"num_tokens": 251046334.0,
"step": 2830
},
{
"entropy": 0.9822265625,
"epoch": 0.38816373949292693,
"grad_norm": 0.7488295461297557,
"learning_rate": 4.1549246160349445e-06,
"loss": 0.7061,
"mean_token_accuracy": 0.8096099197864532,
"num_tokens": 251967634.0,
"step": 2840
},
{
"entropy": 0.9609375,
"epoch": 0.3895305132235358,
"grad_norm": 0.7285672097234401,
"learning_rate": 4.151402000845428e-06,
"loss": 0.6805,
"mean_token_accuracy": 0.8134467422962188,
"num_tokens": 252873963.0,
"step": 2850
},
{
"entropy": 0.9490234375,
"epoch": 0.3908972869541447,
"grad_norm": 0.6967027212080041,
"learning_rate": 4.147879385655912e-06,
"loss": 0.6446,
"mean_token_accuracy": 0.8242575168609619,
"num_tokens": 253752744.0,
"step": 2860
},
{
"entropy": 0.9724609375,
"epoch": 0.3922640606847536,
"grad_norm": 0.8430452065072676,
"learning_rate": 4.144356770466394e-06,
"loss": 0.6944,
"mean_token_accuracy": 0.8123599410057067,
"num_tokens": 254605874.0,
"step": 2870
},
{
"entropy": 1.049609375,
"epoch": 0.39363083441536256,
"grad_norm": 0.8135044162408059,
"learning_rate": 4.140834155276878e-06,
"loss": 0.7616,
"mean_token_accuracy": 0.7965049803256988,
"num_tokens": 255509596.0,
"step": 2880
},
{
"entropy": 0.9931640625,
"epoch": 0.39499760814597146,
"grad_norm": 0.7921037200371817,
"learning_rate": 4.137311540087361e-06,
"loss": 0.693,
"mean_token_accuracy": 0.8147242426872253,
"num_tokens": 256444844.0,
"step": 2890
},
{
"entropy": 1.0029296875,
"epoch": 0.39636438187658035,
"grad_norm": 0.8840696362325742,
"learning_rate": 4.133788924897845e-06,
"loss": 0.7207,
"mean_token_accuracy": 0.8075321644544602,
"num_tokens": 257337194.0,
"step": 2900
},
{
"entropy": 0.947265625,
"epoch": 0.39773115560718925,
"grad_norm": 0.651354438698841,
"learning_rate": 4.130266309708328e-06,
"loss": 0.6388,
"mean_token_accuracy": 0.8261900246143341,
"num_tokens": 258224564.0,
"step": 2910
},
{
"entropy": 1.00546875,
"epoch": 0.39909792933779814,
"grad_norm": 0.8000528064937826,
"learning_rate": 4.126743694518811e-06,
"loss": 0.7376,
"mean_token_accuracy": 0.8036564260721206,
"num_tokens": 259129739.0,
"step": 2920
},
{
"entropy": 1.003125,
"epoch": 0.40046470306840704,
"grad_norm": 0.8216091857652157,
"learning_rate": 4.123221079329294e-06,
"loss": 0.6781,
"mean_token_accuracy": 0.8166547894477845,
"num_tokens": 259960983.0,
"step": 2930
},
{
"entropy": 1.01328125,
"epoch": 0.40183147679901593,
"grad_norm": 0.8013331664430924,
"learning_rate": 4.1196984641397774e-06,
"loss": 0.7226,
"mean_token_accuracy": 0.8058714538812637,
"num_tokens": 260822212.0,
"step": 2940
},
{
"entropy": 0.965234375,
"epoch": 0.4031982505296248,
"grad_norm": 0.7545400917457757,
"learning_rate": 4.116175848950261e-06,
"loss": 0.6391,
"mean_token_accuracy": 0.8253423810005188,
"num_tokens": 261702691.0,
"step": 2950
},
{
"entropy": 0.9482421875,
"epoch": 0.4045650242602337,
"grad_norm": 0.8212195751030741,
"learning_rate": 4.1126532337607445e-06,
"loss": 0.6479,
"mean_token_accuracy": 0.8216013163328171,
"num_tokens": 262516705.0,
"step": 2960
},
{
"entropy": 1.0181640625,
"epoch": 0.4059317979908426,
"grad_norm": 0.8648610330411163,
"learning_rate": 4.109130618571227e-06,
"loss": 0.7205,
"mean_token_accuracy": 0.806904774904251,
"num_tokens": 263406987.0,
"step": 2970
},
{
"entropy": 1.008984375,
"epoch": 0.4072985717214515,
"grad_norm": 0.789797535946681,
"learning_rate": 4.105608003381711e-06,
"loss": 0.7024,
"mean_token_accuracy": 0.810305905342102,
"num_tokens": 264280538.0,
"step": 2980
},
{
"entropy": 0.954296875,
"epoch": 0.4086653454520604,
"grad_norm": 0.8468026216894693,
"learning_rate": 4.102085388192194e-06,
"loss": 0.6756,
"mean_token_accuracy": 0.8172462195158005,
"num_tokens": 265155880.0,
"step": 2990
},
{
"entropy": 1.0015625,
"epoch": 0.4100321191826693,
"grad_norm": 0.8380305968344867,
"learning_rate": 4.098562773002678e-06,
"loss": 0.7274,
"mean_token_accuracy": 0.8045136123895645,
"num_tokens": 266013552.0,
"step": 3000
},
{
"entropy": 0.940234375,
"epoch": 0.4113988929132782,
"grad_norm": 0.7959776037191618,
"learning_rate": 4.095040157813161e-06,
"loss": 0.6685,
"mean_token_accuracy": 0.8175415098667145,
"num_tokens": 266944824.0,
"step": 3010
},
{
"entropy": 0.9953125,
"epoch": 0.4127656666438871,
"grad_norm": 0.830127370758811,
"learning_rate": 4.091517542623644e-06,
"loss": 0.7023,
"mean_token_accuracy": 0.8113681733608246,
"num_tokens": 267814564.0,
"step": 3020
},
{
"entropy": 1.0400390625,
"epoch": 0.414132440374496,
"grad_norm": 0.8488022091345874,
"learning_rate": 4.087994927434128e-06,
"loss": 0.7353,
"mean_token_accuracy": 0.8035644203424454,
"num_tokens": 268692641.0,
"step": 3030
},
{
"entropy": 1.0078125,
"epoch": 0.4154992141051049,
"grad_norm": 0.7455624970467237,
"learning_rate": 4.08447231224461e-06,
"loss": 0.7133,
"mean_token_accuracy": 0.807159823179245,
"num_tokens": 269542807.0,
"step": 3040
},
{
"entropy": 0.9572265625,
"epoch": 0.4168659878357138,
"grad_norm": 0.7659366575056075,
"learning_rate": 4.080949697055094e-06,
"loss": 0.6393,
"mean_token_accuracy": 0.8241441696882248,
"num_tokens": 270383078.0,
"step": 3050
},
{
"entropy": 0.97265625,
"epoch": 0.4182327615663227,
"grad_norm": 0.8552541334147095,
"learning_rate": 4.0774270818655775e-06,
"loss": 0.6886,
"mean_token_accuracy": 0.813379654288292,
"num_tokens": 271284999.0,
"step": 3060
},
{
"entropy": 1.0208984375,
"epoch": 0.4195995352969316,
"grad_norm": 0.8295168041862435,
"learning_rate": 4.073904466676061e-06,
"loss": 0.7413,
"mean_token_accuracy": 0.8024969160556793,
"num_tokens": 272148180.0,
"step": 3070
},
{
"entropy": 0.9833984375,
"epoch": 0.4209663090275405,
"grad_norm": 0.8597461154718612,
"learning_rate": 4.070381851486544e-06,
"loss": 0.695,
"mean_token_accuracy": 0.8136145919561386,
"num_tokens": 273083027.0,
"step": 3080
},
{
"entropy": 0.978515625,
"epoch": 0.4223330827581494,
"grad_norm": 0.7366571112347251,
"learning_rate": 4.066859236297027e-06,
"loss": 0.6614,
"mean_token_accuracy": 0.8197453916072845,
"num_tokens": 273982751.0,
"step": 3090
},
{
"entropy": 0.991015625,
"epoch": 0.4236998564887583,
"grad_norm": 0.8469965962543816,
"learning_rate": 4.06333662110751e-06,
"loss": 0.6872,
"mean_token_accuracy": 0.815274041891098,
"num_tokens": 274840518.0,
"step": 3100
},
{
"entropy": 1.040234375,
"epoch": 0.4250666302193672,
"grad_norm": 0.7905286273100829,
"learning_rate": 4.059814005917994e-06,
"loss": 0.7336,
"mean_token_accuracy": 0.8035754024982452,
"num_tokens": 275767278.0,
"step": 3110
},
{
"entropy": 1.00625,
"epoch": 0.4264334039499761,
"grad_norm": 0.8366727740327329,
"learning_rate": 4.056291390728477e-06,
"loss": 0.7059,
"mean_token_accuracy": 0.8102634966373443,
"num_tokens": 276644137.0,
"step": 3120
},
{
"entropy": 1.0458984375,
"epoch": 0.427800177680585,
"grad_norm": 1.0010530291772748,
"learning_rate": 4.052768775538961e-06,
"loss": 0.7276,
"mean_token_accuracy": 0.8049431145191193,
"num_tokens": 277549426.0,
"step": 3130
},
{
"entropy": 0.992578125,
"epoch": 0.4291669514111939,
"grad_norm": 0.906534275779086,
"learning_rate": 4.049246160349444e-06,
"loss": 0.733,
"mean_token_accuracy": 0.806847682595253,
"num_tokens": 278466337.0,
"step": 3140
},
{
"entropy": 0.9677734375,
"epoch": 0.43053372514180277,
"grad_norm": 0.776315257490987,
"learning_rate": 4.045723545159927e-06,
"loss": 0.6616,
"mean_token_accuracy": 0.8208817392587662,
"num_tokens": 279324947.0,
"step": 3150
},
{
"entropy": 0.94140625,
"epoch": 0.43190049887241166,
"grad_norm": 0.7662935500414488,
"learning_rate": 4.04220092997041e-06,
"loss": 0.6345,
"mean_token_accuracy": 0.8247306197881699,
"num_tokens": 280225654.0,
"step": 3160
},
{
"entropy": 0.964453125,
"epoch": 0.43326727260302056,
"grad_norm": 0.7231671477053645,
"learning_rate": 4.038678314780894e-06,
"loss": 0.6881,
"mean_token_accuracy": 0.815484020113945,
"num_tokens": 281153133.0,
"step": 3170
},
{
"entropy": 0.9953125,
"epoch": 0.43463404633362945,
"grad_norm": 0.7114495569086311,
"learning_rate": 4.0351556995913775e-06,
"loss": 0.7319,
"mean_token_accuracy": 0.8033326953649521,
"num_tokens": 282083125.0,
"step": 3180
},
{
"entropy": 0.9654296875,
"epoch": 0.43600082006423835,
"grad_norm": 0.6867597371336408,
"learning_rate": 4.03163308440186e-06,
"loss": 0.7054,
"mean_token_accuracy": 0.8108426719903946,
"num_tokens": 282960536.0,
"step": 3190
},
{
"entropy": 1.0119140625,
"epoch": 0.43736759379484724,
"grad_norm": 0.7876484404714228,
"learning_rate": 4.028110469212344e-06,
"loss": 0.7287,
"mean_token_accuracy": 0.8081396758556366,
"num_tokens": 283800850.0,
"step": 3200
},
{
"entropy": 0.9953125,
"epoch": 0.43873436752545614,
"grad_norm": 0.7782336499144333,
"learning_rate": 4.0245878540228264e-06,
"loss": 0.6967,
"mean_token_accuracy": 0.8096145689487457,
"num_tokens": 284680091.0,
"step": 3210
},
{
"entropy": 1.0125,
"epoch": 0.44010114125606503,
"grad_norm": 0.8131863333847568,
"learning_rate": 4.02106523883331e-06,
"loss": 0.7135,
"mean_token_accuracy": 0.8097018271684646,
"num_tokens": 285538822.0,
"step": 3220
},
{
"entropy": 0.9599609375,
"epoch": 0.441467914986674,
"grad_norm": 0.7816436317133533,
"learning_rate": 4.0175426236437935e-06,
"loss": 0.7009,
"mean_token_accuracy": 0.8143548309803009,
"num_tokens": 286390073.0,
"step": 3230
},
{
"entropy": 0.9822265625,
"epoch": 0.4428346887172829,
"grad_norm": 0.7926182318675519,
"learning_rate": 4.014020008454277e-06,
"loss": 0.6924,
"mean_token_accuracy": 0.8136519998311996,
"num_tokens": 287286685.0,
"step": 3240
},
{
"entropy": 0.96875,
"epoch": 0.44420146244789177,
"grad_norm": 0.6731418092345045,
"learning_rate": 4.01049739326476e-06,
"loss": 0.6907,
"mean_token_accuracy": 0.8100464314222335,
"num_tokens": 288121053.0,
"step": 3250
},
{
"entropy": 0.9869140625,
"epoch": 0.44556823617850067,
"grad_norm": 0.7274601263444969,
"learning_rate": 4.006974778075243e-06,
"loss": 0.6977,
"mean_token_accuracy": 0.812783470749855,
"num_tokens": 289016233.0,
"step": 3260
},
{
"entropy": 0.9828125,
"epoch": 0.44693500990910956,
"grad_norm": 0.6590714435988143,
"learning_rate": 4.003452162885727e-06,
"loss": 0.7191,
"mean_token_accuracy": 0.8063458651304245,
"num_tokens": 289937154.0,
"step": 3270
},
{
"entropy": 0.94140625,
"epoch": 0.44830178363971845,
"grad_norm": 0.8540856499675822,
"learning_rate": 3.99992954769621e-06,
"loss": 0.6696,
"mean_token_accuracy": 0.8183889657258987,
"num_tokens": 290823440.0,
"step": 3280
},
{
"entropy": 0.9630859375,
"epoch": 0.44966855737032735,
"grad_norm": 0.7681226550509954,
"learning_rate": 3.996406932506693e-06,
"loss": 0.699,
"mean_token_accuracy": 0.8107506036758423,
"num_tokens": 291744078.0,
"step": 3290
},
{
"entropy": 0.9587890625,
"epoch": 0.45103533110093624,
"grad_norm": 0.825865928363738,
"learning_rate": 3.992884317317177e-06,
"loss": 0.6949,
"mean_token_accuracy": 0.8128736197948456,
"num_tokens": 292683000.0,
"step": 3300
},
{
"entropy": 1.0419921875,
"epoch": 0.45240210483154514,
"grad_norm": 0.8106605528526303,
"learning_rate": 3.98936170212766e-06,
"loss": 0.7196,
"mean_token_accuracy": 0.8080102622509002,
"num_tokens": 293591908.0,
"step": 3310
},
{
"entropy": 0.9525390625,
"epoch": 0.45376887856215403,
"grad_norm": 0.7419304883387207,
"learning_rate": 3.985839086938143e-06,
"loss": 0.6629,
"mean_token_accuracy": 0.8202334433794022,
"num_tokens": 294468616.0,
"step": 3320
},
{
"entropy": 0.9923828125,
"epoch": 0.4551356522927629,
"grad_norm": 0.8271329313851044,
"learning_rate": 3.9823164717486265e-06,
"loss": 0.7423,
"mean_token_accuracy": 0.8021346062421799,
"num_tokens": 295394199.0,
"step": 3330
},
{
"entropy": 0.9875,
"epoch": 0.4565024260233718,
"grad_norm": 0.7559050986728351,
"learning_rate": 3.97879385655911e-06,
"loss": 0.6949,
"mean_token_accuracy": 0.8124124825000762,
"num_tokens": 296273215.0,
"step": 3340
},
{
"entropy": 0.9955078125,
"epoch": 0.4578691997539807,
"grad_norm": 0.810518238736224,
"learning_rate": 3.9752712413695936e-06,
"loss": 0.7115,
"mean_token_accuracy": 0.8074479401111603,
"num_tokens": 297148572.0,
"step": 3350
},
{
"entropy": 1.0697265625,
"epoch": 0.4592359734845896,
"grad_norm": 0.7411614479848697,
"learning_rate": 3.971748626180076e-06,
"loss": 0.7449,
"mean_token_accuracy": 0.8028772681951523,
"num_tokens": 298039662.0,
"step": 3360
},
{
"entropy": 0.990625,
"epoch": 0.4606027472151985,
"grad_norm": 0.7682807816985432,
"learning_rate": 3.96822601099056e-06,
"loss": 0.7102,
"mean_token_accuracy": 0.8087541908025742,
"num_tokens": 298956252.0,
"step": 3370
},
{
"entropy": 0.98515625,
"epoch": 0.4619695209458074,
"grad_norm": 0.805598811587094,
"learning_rate": 3.9647033958010425e-06,
"loss": 0.7457,
"mean_token_accuracy": 0.8021170705556869,
"num_tokens": 299823280.0,
"step": 3380
},
{
"entropy": 0.994140625,
"epoch": 0.4633362946764163,
"grad_norm": 0.8438720640673301,
"learning_rate": 3.961180780611527e-06,
"loss": 0.689,
"mean_token_accuracy": 0.8134701699018478,
"num_tokens": 300704268.0,
"step": 3390
},
{
"entropy": 1.018359375,
"epoch": 0.46470306840702524,
"grad_norm": 0.681927003766349,
"learning_rate": 3.95765816542201e-06,
"loss": 0.7459,
"mean_token_accuracy": 0.8031022727489472,
"num_tokens": 301638457.0,
"step": 3400
},
{
"entropy": 0.991796875,
"epoch": 0.46606984213763414,
"grad_norm": 0.7299359776574085,
"learning_rate": 3.954135550232493e-06,
"loss": 0.6702,
"mean_token_accuracy": 0.8159130454063416,
"num_tokens": 302495464.0,
"step": 3410
},
{
"entropy": 1.009765625,
"epoch": 0.46743661586824303,
"grad_norm": 0.770472971581311,
"learning_rate": 3.950612935042976e-06,
"loss": 0.729,
"mean_token_accuracy": 0.8045765280723571,
"num_tokens": 303376908.0,
"step": 3420
},
{
"entropy": 0.967578125,
"epoch": 0.4688033895988519,
"grad_norm": 0.7992851381908447,
"learning_rate": 3.947090319853459e-06,
"loss": 0.6948,
"mean_token_accuracy": 0.8128030180931092,
"num_tokens": 304255583.0,
"step": 3430
},
{
"entropy": 1.034765625,
"epoch": 0.4701701633294608,
"grad_norm": 0.7966461640227416,
"learning_rate": 3.943567704663943e-06,
"loss": 0.7344,
"mean_token_accuracy": 0.805388194322586,
"num_tokens": 305117375.0,
"step": 3440
},
{
"entropy": 0.9818359375,
"epoch": 0.4715369370600697,
"grad_norm": 0.8620989516697106,
"learning_rate": 3.9400450894744265e-06,
"loss": 0.7242,
"mean_token_accuracy": 0.8052465289831161,
"num_tokens": 305964787.0,
"step": 3450
},
{
"entropy": 1.023828125,
"epoch": 0.4729037107906786,
"grad_norm": 0.8989857405370105,
"learning_rate": 3.936522474284909e-06,
"loss": 0.726,
"mean_token_accuracy": 0.8049546211957932,
"num_tokens": 306874371.0,
"step": 3460
},
{
"entropy": 0.9921875,
"epoch": 0.4742704845212875,
"grad_norm": 0.8945862744228753,
"learning_rate": 3.932999859095393e-06,
"loss": 0.6971,
"mean_token_accuracy": 0.8127100199460984,
"num_tokens": 307828178.0,
"step": 3470
},
{
"entropy": 0.967578125,
"epoch": 0.4756372582518964,
"grad_norm": 0.8264513501577553,
"learning_rate": 3.929477243905876e-06,
"loss": 0.6767,
"mean_token_accuracy": 0.8166712671518326,
"num_tokens": 308715305.0,
"step": 3480
},
{
"entropy": 1.0216796875,
"epoch": 0.4770040319825053,
"grad_norm": 0.9547904713998239,
"learning_rate": 3.925954628716359e-06,
"loss": 0.7011,
"mean_token_accuracy": 0.8141042828559876,
"num_tokens": 309596173.0,
"step": 3490
},
{
"entropy": 1.024609375,
"epoch": 0.4783708057131142,
"grad_norm": 0.8559587155505296,
"learning_rate": 3.9224320135268425e-06,
"loss": 0.7013,
"mean_token_accuracy": 0.8088554441928864,
"num_tokens": 310464726.0,
"step": 3500
},
{
"entropy": 0.9998046875,
"epoch": 0.4797375794437231,
"grad_norm": 0.825966875143227,
"learning_rate": 3.918909398337326e-06,
"loss": 0.6959,
"mean_token_accuracy": 0.8139429271221161,
"num_tokens": 311349143.0,
"step": 3510
},
{
"entropy": 0.9982421875,
"epoch": 0.481104353174332,
"grad_norm": 0.785653079634919,
"learning_rate": 3.91538678314781e-06,
"loss": 0.7106,
"mean_token_accuracy": 0.8095379114151001,
"num_tokens": 312240199.0,
"step": 3520
},
{
"entropy": 0.99765625,
"epoch": 0.4824711269049409,
"grad_norm": 0.7830141193421727,
"learning_rate": 3.911864167958292e-06,
"loss": 0.6978,
"mean_token_accuracy": 0.8119673669338227,
"num_tokens": 313135937.0,
"step": 3530
},
{
"entropy": 0.98203125,
"epoch": 0.48383790063554977,
"grad_norm": 0.7909696592323343,
"learning_rate": 3.908341552768776e-06,
"loss": 0.7143,
"mean_token_accuracy": 0.8079461932182312,
"num_tokens": 314018252.0,
"step": 3540
},
{
"entropy": 1.0048828125,
"epoch": 0.48520467436615866,
"grad_norm": 0.7985093309744903,
"learning_rate": 3.9048189375792586e-06,
"loss": 0.7008,
"mean_token_accuracy": 0.8084679931402207,
"num_tokens": 314871616.0,
"step": 3550
},
{
"entropy": 1.01015625,
"epoch": 0.48657144809676756,
"grad_norm": 0.7603281131252385,
"learning_rate": 3.901296322389743e-06,
"loss": 0.7398,
"mean_token_accuracy": 0.8039510101079941,
"num_tokens": 315729836.0,
"step": 3560
},
{
"entropy": 1.01171875,
"epoch": 0.48793822182737645,
"grad_norm": 0.7731912394670131,
"learning_rate": 3.897773707200226e-06,
"loss": 0.7056,
"mean_token_accuracy": 0.8091530799865723,
"num_tokens": 316602383.0,
"step": 3570
},
{
"entropy": 0.998046875,
"epoch": 0.4893049955579854,
"grad_norm": 0.8365918066100918,
"learning_rate": 3.894251092010709e-06,
"loss": 0.6892,
"mean_token_accuracy": 0.8102873176336288,
"num_tokens": 317513326.0,
"step": 3580
},
{
"entropy": 0.9640625,
"epoch": 0.4906717692885943,
"grad_norm": 0.7621376940212303,
"learning_rate": 3.890728476821192e-06,
"loss": 0.6767,
"mean_token_accuracy": 0.8162740916013718,
"num_tokens": 318374110.0,
"step": 3590
},
{
"entropy": 0.9529296875,
"epoch": 0.4920385430192032,
"grad_norm": 0.7548418753796782,
"learning_rate": 3.8872058616316755e-06,
"loss": 0.6575,
"mean_token_accuracy": 0.8217393815517425,
"num_tokens": 319229245.0,
"step": 3600
},
{
"entropy": 0.9646484375,
"epoch": 0.4934053167498121,
"grad_norm": 0.7977671983852296,
"learning_rate": 3.883683246442159e-06,
"loss": 0.7114,
"mean_token_accuracy": 0.8085690617561341,
"num_tokens": 320108614.0,
"step": 3610
},
{
"entropy": 0.990625,
"epoch": 0.494772090480421,
"grad_norm": 0.7189894828629616,
"learning_rate": 3.8801606312526426e-06,
"loss": 0.6665,
"mean_token_accuracy": 0.8176935076713562,
"num_tokens": 321041845.0,
"step": 3620
},
{
"entropy": 0.9330078125,
"epoch": 0.4961388642110299,
"grad_norm": 0.8048752787882473,
"learning_rate": 3.876638016063125e-06,
"loss": 0.6211,
"mean_token_accuracy": 0.8292361527681351,
"num_tokens": 321942482.0,
"step": 3630
},
{
"entropy": 0.9779296875,
"epoch": 0.49750563794163877,
"grad_norm": 0.8398229395757382,
"learning_rate": 3.873115400873609e-06,
"loss": 0.6686,
"mean_token_accuracy": 0.8174511134624481,
"num_tokens": 322786504.0,
"step": 3640
},
{
"entropy": 0.9865234375,
"epoch": 0.49887241167224766,
"grad_norm": 0.7974771722869434,
"learning_rate": 3.869592785684092e-06,
"loss": 0.7168,
"mean_token_accuracy": 0.8074921637773513,
"num_tokens": 323673486.0,
"step": 3650
},
{
"entropy": 1.0111328125,
"epoch": 0.5002391854028566,
"grad_norm": 0.6859225409713021,
"learning_rate": 3.866070170494575e-06,
"loss": 0.6996,
"mean_token_accuracy": 0.8135553151369095,
"num_tokens": 324539268.0,
"step": 3660
},
{
"entropy": 1.0095703125,
"epoch": 0.5016059591334655,
"grad_norm": 0.878800978192418,
"learning_rate": 3.862547555305059e-06,
"loss": 0.72,
"mean_token_accuracy": 0.8064991861581803,
"num_tokens": 325467313.0,
"step": 3670
},
{
"entropy": 1.0044921875,
"epoch": 0.5029727328640744,
"grad_norm": 0.8831838896405875,
"learning_rate": 3.859024940115542e-06,
"loss": 0.6881,
"mean_token_accuracy": 0.8142493724822998,
"num_tokens": 326360349.0,
"step": 3680
},
{
"entropy": 0.9595703125,
"epoch": 0.5043395065946833,
"grad_norm": 0.6820775235014569,
"learning_rate": 3.855502324926026e-06,
"loss": 0.6537,
"mean_token_accuracy": 0.8228391766548157,
"num_tokens": 327295014.0,
"step": 3690
},
{
"entropy": 0.9708984375,
"epoch": 0.5057062803252922,
"grad_norm": 0.7863643029463756,
"learning_rate": 3.851979709736508e-06,
"loss": 0.6903,
"mean_token_accuracy": 0.8135940253734588,
"num_tokens": 328166096.0,
"step": 3700
},
{
"entropy": 1.0115234375,
"epoch": 0.5070730540559011,
"grad_norm": 0.7768721386952211,
"learning_rate": 3.848457094546992e-06,
"loss": 0.7126,
"mean_token_accuracy": 0.8092552363872528,
"num_tokens": 329044476.0,
"step": 3710
},
{
"entropy": 0.965625,
"epoch": 0.50843982778651,
"grad_norm": 0.7587031469264036,
"learning_rate": 3.8449344793574755e-06,
"loss": 0.7037,
"mean_token_accuracy": 0.811672705411911,
"num_tokens": 329954219.0,
"step": 3720
},
{
"entropy": 0.988671875,
"epoch": 0.5098066015171189,
"grad_norm": 0.6967271623958298,
"learning_rate": 3.841411864167959e-06,
"loss": 0.7323,
"mean_token_accuracy": 0.8047899842262268,
"num_tokens": 330824306.0,
"step": 3730
},
{
"entropy": 0.9986328125,
"epoch": 0.5111733752477278,
"grad_norm": 0.8503831614070991,
"learning_rate": 3.837889248978442e-06,
"loss": 0.7347,
"mean_token_accuracy": 0.8027462720870971,
"num_tokens": 331697014.0,
"step": 3740
},
{
"entropy": 0.9826171875,
"epoch": 0.5125401489783367,
"grad_norm": 0.8105546006699571,
"learning_rate": 3.834366633788925e-06,
"loss": 0.6662,
"mean_token_accuracy": 0.8206939905881881,
"num_tokens": 332591752.0,
"step": 3750
},
{
"entropy": 0.98828125,
"epoch": 0.5139069227089456,
"grad_norm": 0.823595551002211,
"learning_rate": 3.830844018599408e-06,
"loss": 0.6734,
"mean_token_accuracy": 0.8167521536350251,
"num_tokens": 333428436.0,
"step": 3760
},
{
"entropy": 1.0134765625,
"epoch": 0.5152736964395545,
"grad_norm": 0.8123564208482943,
"learning_rate": 3.8273214034098915e-06,
"loss": 0.711,
"mean_token_accuracy": 0.8110816299915313,
"num_tokens": 334312680.0,
"step": 3770
},
{
"entropy": 1.0201171875,
"epoch": 0.5166404701701633,
"grad_norm": 0.7485976808587699,
"learning_rate": 3.823798788220375e-06,
"loss": 0.7083,
"mean_token_accuracy": 0.8089155167341232,
"num_tokens": 335170349.0,
"step": 3780
},
{
"entropy": 0.9578125,
"epoch": 0.5180072439007722,
"grad_norm": 0.8519725396521569,
"learning_rate": 3.820276173030859e-06,
"loss": 0.6866,
"mean_token_accuracy": 0.8143474221229553,
"num_tokens": 336022826.0,
"step": 3790
},
{
"entropy": 0.96640625,
"epoch": 0.5193740176313811,
"grad_norm": 0.7908591166333419,
"learning_rate": 3.816753557841341e-06,
"loss": 0.6673,
"mean_token_accuracy": 0.8180392920970917,
"num_tokens": 336888400.0,
"step": 3800
},
{
"entropy": 0.98515625,
"epoch": 0.52074079136199,
"grad_norm": 0.7535535999979862,
"learning_rate": 3.813230942651825e-06,
"loss": 0.7158,
"mean_token_accuracy": 0.8083377659320832,
"num_tokens": 337726810.0,
"step": 3810
},
{
"entropy": 0.9943359375,
"epoch": 0.5221075650925989,
"grad_norm": 0.7884277527026199,
"learning_rate": 3.809708327462308e-06,
"loss": 0.6739,
"mean_token_accuracy": 0.8161290466785431,
"num_tokens": 338587129.0,
"step": 3820
},
{
"entropy": 0.9966796875,
"epoch": 0.5234743388232078,
"grad_norm": 0.7589094292290308,
"learning_rate": 3.806185712272792e-06,
"loss": 0.7135,
"mean_token_accuracy": 0.8094353407621384,
"num_tokens": 339495137.0,
"step": 3830
},
{
"entropy": 1.0580078125,
"epoch": 0.5248411125538167,
"grad_norm": 0.7809116323735399,
"learning_rate": 3.802663097083275e-06,
"loss": 0.7694,
"mean_token_accuracy": 0.7971067488193512,
"num_tokens": 340326128.0,
"step": 3840
},
{
"entropy": 0.9908203125,
"epoch": 0.5262078862844256,
"grad_norm": 0.7832026142032406,
"learning_rate": 3.799140481893758e-06,
"loss": 0.6907,
"mean_token_accuracy": 0.8109352797269821,
"num_tokens": 341183998.0,
"step": 3850
},
{
"entropy": 0.95390625,
"epoch": 0.5275746600150345,
"grad_norm": 0.72140778245727,
"learning_rate": 3.7956178667042413e-06,
"loss": 0.6723,
"mean_token_accuracy": 0.8183467477560044,
"num_tokens": 342030730.0,
"step": 3860
},
{
"entropy": 0.9435546875,
"epoch": 0.5289414337456434,
"grad_norm": 0.7758773982873882,
"learning_rate": 3.792095251514725e-06,
"loss": 0.6797,
"mean_token_accuracy": 0.8152454227209092,
"num_tokens": 342928649.0,
"step": 3870
},
{
"entropy": 0.96875,
"epoch": 0.5303082074762523,
"grad_norm": 0.7537239528631098,
"learning_rate": 3.788572636325208e-06,
"loss": 0.6925,
"mean_token_accuracy": 0.8148498982191086,
"num_tokens": 343852240.0,
"step": 3880
},
{
"entropy": 0.96328125,
"epoch": 0.5316749812068612,
"grad_norm": 0.7282066830169055,
"learning_rate": 3.7850500211356916e-06,
"loss": 0.6634,
"mean_token_accuracy": 0.8177864879369736,
"num_tokens": 344750196.0,
"step": 3890
},
{
"entropy": 0.993359375,
"epoch": 0.5330417549374701,
"grad_norm": 0.7836377257557595,
"learning_rate": 3.7815274059461747e-06,
"loss": 0.6962,
"mean_token_accuracy": 0.811986431479454,
"num_tokens": 345621623.0,
"step": 3900
},
{
"entropy": 0.9779296875,
"epoch": 0.534408528668079,
"grad_norm": 0.6932161295027599,
"learning_rate": 3.7780047907566582e-06,
"loss": 0.6487,
"mean_token_accuracy": 0.8219556152820587,
"num_tokens": 346526744.0,
"step": 3910
},
{
"entropy": 0.98203125,
"epoch": 0.5357753023986879,
"grad_norm": 0.7063941651920479,
"learning_rate": 3.7744821755671413e-06,
"loss": 0.6722,
"mean_token_accuracy": 0.8189321964979172,
"num_tokens": 347436449.0,
"step": 3920
},
{
"entropy": 0.9921875,
"epoch": 0.5371420761292968,
"grad_norm": 0.7499906267687931,
"learning_rate": 3.7709595603776245e-06,
"loss": 0.6927,
"mean_token_accuracy": 0.814979350566864,
"num_tokens": 348331917.0,
"step": 3930
},
{
"entropy": 1.01484375,
"epoch": 0.5385088498599057,
"grad_norm": 0.8579048569154432,
"learning_rate": 3.767436945188108e-06,
"loss": 0.7146,
"mean_token_accuracy": 0.8069836616516113,
"num_tokens": 349199022.0,
"step": 3940
},
{
"entropy": 0.95859375,
"epoch": 0.5398756235905146,
"grad_norm": 0.755390816477522,
"learning_rate": 3.7639143299985916e-06,
"loss": 0.6851,
"mean_token_accuracy": 0.8136724472045899,
"num_tokens": 350085246.0,
"step": 3950
},
{
"entropy": 1.0109375,
"epoch": 0.5412423973211234,
"grad_norm": 0.7173087328213487,
"learning_rate": 3.7603917148090747e-06,
"loss": 0.6869,
"mean_token_accuracy": 0.8145682960748672,
"num_tokens": 350974056.0,
"step": 3960
},
{
"entropy": 0.9765625,
"epoch": 0.5426091710517323,
"grad_norm": 0.7540226013303124,
"learning_rate": 3.756869099619558e-06,
"loss": 0.6648,
"mean_token_accuracy": 0.8185296386480332,
"num_tokens": 351864137.0,
"step": 3970
},
{
"entropy": 0.9697265625,
"epoch": 0.5439759447823412,
"grad_norm": 0.7596391519018107,
"learning_rate": 3.753346484430041e-06,
"loss": 0.6854,
"mean_token_accuracy": 0.814040195941925,
"num_tokens": 352741822.0,
"step": 3980
},
{
"entropy": 0.968359375,
"epoch": 0.5453427185129501,
"grad_norm": 0.7895807708054654,
"learning_rate": 3.749823869240524e-06,
"loss": 0.6643,
"mean_token_accuracy": 0.820447462797165,
"num_tokens": 353612932.0,
"step": 3990
},
{
"entropy": 0.9447265625,
"epoch": 0.546709492243559,
"grad_norm": 0.8384090779226839,
"learning_rate": 3.746301254051008e-06,
"loss": 0.6426,
"mean_token_accuracy": 0.8242950141429901,
"num_tokens": 354489469.0,
"step": 4000
},
{
"entropy": 0.95703125,
"epoch": 0.548076265974168,
"grad_norm": 0.7434710413259598,
"learning_rate": 3.742778638861491e-06,
"loss": 0.6803,
"mean_token_accuracy": 0.8143187820911407,
"num_tokens": 355362081.0,
"step": 4010
},
{
"entropy": 0.9814453125,
"epoch": 0.5494430397047769,
"grad_norm": 0.7604044573784576,
"learning_rate": 3.7392560236719743e-06,
"loss": 0.696,
"mean_token_accuracy": 0.8103440076112747,
"num_tokens": 356206559.0,
"step": 4020
},
{
"entropy": 0.931640625,
"epoch": 0.5508098134353858,
"grad_norm": 0.7239121066687831,
"learning_rate": 3.7357334084824574e-06,
"loss": 0.6406,
"mean_token_accuracy": 0.8241357058286667,
"num_tokens": 357084489.0,
"step": 4030
},
{
"entropy": 0.9587890625,
"epoch": 0.5521765871659947,
"grad_norm": 0.7996419875764378,
"learning_rate": 3.732210793292941e-06,
"loss": 0.6671,
"mean_token_accuracy": 0.8181318491697311,
"num_tokens": 357975361.0,
"step": 4040
},
{
"entropy": 1.0009765625,
"epoch": 0.5535433608966036,
"grad_norm": 0.8014506724781437,
"learning_rate": 3.7286881781034245e-06,
"loss": 0.6886,
"mean_token_accuracy": 0.8123732447624207,
"num_tokens": 358878614.0,
"step": 4050
},
{
"entropy": 0.9646484375,
"epoch": 0.5549101346272125,
"grad_norm": 0.7280835096680386,
"learning_rate": 3.7251655629139076e-06,
"loss": 0.6517,
"mean_token_accuracy": 0.8215162783861161,
"num_tokens": 359801961.0,
"step": 4060
},
{
"entropy": 1.0142578125,
"epoch": 0.5562769083578214,
"grad_norm": 0.7130757003859051,
"learning_rate": 3.7216429477243907e-06,
"loss": 0.7029,
"mean_token_accuracy": 0.8083680272102356,
"num_tokens": 360684656.0,
"step": 4070
},
{
"entropy": 0.9833984375,
"epoch": 0.5576436820884303,
"grad_norm": 0.7858005165002516,
"learning_rate": 3.7181203325348743e-06,
"loss": 0.6914,
"mean_token_accuracy": 0.8126088351011276,
"num_tokens": 361604630.0,
"step": 4080
},
{
"entropy": 1.0251953125,
"epoch": 0.5590104558190392,
"grad_norm": 0.8145060113671897,
"learning_rate": 3.7145977173453574e-06,
"loss": 0.7281,
"mean_token_accuracy": 0.8052356630563736,
"num_tokens": 362453556.0,
"step": 4090
},
{
"entropy": 0.978125,
"epoch": 0.5603772295496481,
"grad_norm": 0.7526561246451242,
"learning_rate": 3.7110751021558405e-06,
"loss": 0.6806,
"mean_token_accuracy": 0.8139560252428055,
"num_tokens": 363318401.0,
"step": 4100
},
{
"entropy": 0.9693359375,
"epoch": 0.561744003280257,
"grad_norm": 0.7440179752023197,
"learning_rate": 3.7075524869663245e-06,
"loss": 0.6659,
"mean_token_accuracy": 0.8187500089406967,
"num_tokens": 364242506.0,
"step": 4110
},
{
"entropy": 0.97890625,
"epoch": 0.5631107770108659,
"grad_norm": 0.7066267765574754,
"learning_rate": 3.7040298717768076e-06,
"loss": 0.7103,
"mean_token_accuracy": 0.8112464189529419,
"num_tokens": 365155052.0,
"step": 4120
},
{
"entropy": 1.0068359375,
"epoch": 0.5644775507414748,
"grad_norm": 0.7886614879259739,
"learning_rate": 3.7005072565872908e-06,
"loss": 0.7154,
"mean_token_accuracy": 0.8082158446311951,
"num_tokens": 366080792.0,
"step": 4130
},
{
"entropy": 0.9951171875,
"epoch": 0.5658443244720837,
"grad_norm": 0.7933640773298268,
"learning_rate": 3.696984641397774e-06,
"loss": 0.7214,
"mean_token_accuracy": 0.8051944375038147,
"num_tokens": 366962396.0,
"step": 4140
},
{
"entropy": 0.9369140625,
"epoch": 0.5672110982026926,
"grad_norm": 0.8557231778132872,
"learning_rate": 3.693462026208257e-06,
"loss": 0.6493,
"mean_token_accuracy": 0.8220056235790253,
"num_tokens": 367850915.0,
"step": 4150
},
{
"entropy": 0.983984375,
"epoch": 0.5685778719333014,
"grad_norm": 0.7207935374820416,
"learning_rate": 3.689939411018741e-06,
"loss": 0.6896,
"mean_token_accuracy": 0.8157393008470535,
"num_tokens": 368779867.0,
"step": 4160
},
{
"entropy": 0.944921875,
"epoch": 0.5699446456639103,
"grad_norm": 0.7671497743245514,
"learning_rate": 3.686416795829224e-06,
"loss": 0.628,
"mean_token_accuracy": 0.8276755094528199,
"num_tokens": 369657800.0,
"step": 4170
},
{
"entropy": 0.982421875,
"epoch": 0.5713114193945192,
"grad_norm": 0.7778467257215443,
"learning_rate": 3.6828941806397072e-06,
"loss": 0.6945,
"mean_token_accuracy": 0.8128043502569199,
"num_tokens": 370511372.0,
"step": 4180
},
{
"entropy": 0.9802734375,
"epoch": 0.5726781931251281,
"grad_norm": 0.7512217390124868,
"learning_rate": 3.6793715654501903e-06,
"loss": 0.7089,
"mean_token_accuracy": 0.808812353014946,
"num_tokens": 371416237.0,
"step": 4190
},
{
"entropy": 0.9916015625,
"epoch": 0.574044966855737,
"grad_norm": 0.8757153612273556,
"learning_rate": 3.6758489502606735e-06,
"loss": 0.7093,
"mean_token_accuracy": 0.8098421782255173,
"num_tokens": 372347297.0,
"step": 4200
},
{
"entropy": 0.9962890625,
"epoch": 0.5754117405863459,
"grad_norm": 0.8007065444272512,
"learning_rate": 3.672326335071157e-06,
"loss": 0.6897,
"mean_token_accuracy": 0.8128375589847565,
"num_tokens": 373210792.0,
"step": 4210
},
{
"entropy": 0.9603515625,
"epoch": 0.5767785143169548,
"grad_norm": 0.7872203323926492,
"learning_rate": 3.6688037198816406e-06,
"loss": 0.6737,
"mean_token_accuracy": 0.8175820916891098,
"num_tokens": 374082669.0,
"step": 4220
},
{
"entropy": 0.9703125,
"epoch": 0.5781452880475637,
"grad_norm": 0.8815229211134221,
"learning_rate": 3.6652811046921237e-06,
"loss": 0.6821,
"mean_token_accuracy": 0.8153961598873138,
"num_tokens": 374934918.0,
"step": 4230
},
{
"entropy": 0.973828125,
"epoch": 0.5795120617781726,
"grad_norm": 0.7631057864575165,
"learning_rate": 3.6617584895026072e-06,
"loss": 0.6629,
"mean_token_accuracy": 0.8192988693714142,
"num_tokens": 375792139.0,
"step": 4240
},
{
"entropy": 1.0234375,
"epoch": 0.5808788355087815,
"grad_norm": 0.8218713242008784,
"learning_rate": 3.6582358743130904e-06,
"loss": 0.7319,
"mean_token_accuracy": 0.800834345817566,
"num_tokens": 376683914.0,
"step": 4250
},
{
"entropy": 0.9744140625,
"epoch": 0.5822456092393904,
"grad_norm": 0.7956652762791976,
"learning_rate": 3.6547132591235735e-06,
"loss": 0.6985,
"mean_token_accuracy": 0.8100746244192123,
"num_tokens": 377543065.0,
"step": 4260
},
{
"entropy": 1.0111328125,
"epoch": 0.5836123829699993,
"grad_norm": 0.8595736721744676,
"learning_rate": 3.651190643934057e-06,
"loss": 0.7244,
"mean_token_accuracy": 0.806061190366745,
"num_tokens": 378369281.0,
"step": 4270
},
{
"entropy": 0.98046875,
"epoch": 0.5849791567006082,
"grad_norm": 0.7749562979154775,
"learning_rate": 3.6476680287445406e-06,
"loss": 0.6965,
"mean_token_accuracy": 0.8130518466234207,
"num_tokens": 379279878.0,
"step": 4280
},
{
"entropy": 0.9546875,
"epoch": 0.5863459304312171,
"grad_norm": 0.750535323374358,
"learning_rate": 3.6441454135550237e-06,
"loss": 0.6655,
"mean_token_accuracy": 0.8187259316444397,
"num_tokens": 380191500.0,
"step": 4290
},
{
"entropy": 1.00390625,
"epoch": 0.587712704161826,
"grad_norm": 0.7778947062528565,
"learning_rate": 3.640622798365507e-06,
"loss": 0.6951,
"mean_token_accuracy": 0.8135675817728043,
"num_tokens": 381087074.0,
"step": 4300
},
{
"entropy": 1.0275390625,
"epoch": 0.5890794778924349,
"grad_norm": 0.7040744285313633,
"learning_rate": 3.63710018317599e-06,
"loss": 0.7315,
"mean_token_accuracy": 0.8044957607984543,
"num_tokens": 381979733.0,
"step": 4310
},
{
"entropy": 1.04140625,
"epoch": 0.5904462516230438,
"grad_norm": 0.829058410428729,
"learning_rate": 3.633577567986473e-06,
"loss": 0.7141,
"mean_token_accuracy": 0.8078227430582047,
"num_tokens": 382847154.0,
"step": 4320
},
{
"entropy": 0.986328125,
"epoch": 0.5918130253536527,
"grad_norm": 0.7576175164505192,
"learning_rate": 3.630054952796957e-06,
"loss": 0.6654,
"mean_token_accuracy": 0.8169815123081208,
"num_tokens": 383707807.0,
"step": 4330
},
{
"entropy": 1.0224609375,
"epoch": 0.5931797990842615,
"grad_norm": 0.7499205351957751,
"learning_rate": 3.62653233760744e-06,
"loss": 0.7646,
"mean_token_accuracy": 0.7967443227767944,
"num_tokens": 384590386.0,
"step": 4340
},
{
"entropy": 0.998828125,
"epoch": 0.5945465728148704,
"grad_norm": 0.8478753252730006,
"learning_rate": 3.6230097224179233e-06,
"loss": 0.6713,
"mean_token_accuracy": 0.8190259963274003,
"num_tokens": 385369969.0,
"step": 4350
},
{
"entropy": 0.9791015625,
"epoch": 0.5959133465454794,
"grad_norm": 0.8449120019787558,
"learning_rate": 3.6194871072284064e-06,
"loss": 0.6944,
"mean_token_accuracy": 0.8128062814474106,
"num_tokens": 386273775.0,
"step": 4360
},
{
"entropy": 1.0123046875,
"epoch": 0.5972801202760883,
"grad_norm": 0.8239699550375721,
"learning_rate": 3.61596449203889e-06,
"loss": 0.683,
"mean_token_accuracy": 0.8138798087835312,
"num_tokens": 387155485.0,
"step": 4370
},
{
"entropy": 0.9349609375,
"epoch": 0.5986468940066972,
"grad_norm": 0.6852018650098914,
"learning_rate": 3.6124418768493735e-06,
"loss": 0.6573,
"mean_token_accuracy": 0.82138112783432,
"num_tokens": 388029225.0,
"step": 4380
},
{
"entropy": 0.9806640625,
"epoch": 0.6000136677373061,
"grad_norm": 0.7011348690560667,
"learning_rate": 3.6089192616598566e-06,
"loss": 0.7023,
"mean_token_accuracy": 0.811756843328476,
"num_tokens": 388973428.0,
"step": 4390
},
{
"entropy": 1.0033203125,
"epoch": 0.601380441467915,
"grad_norm": 0.8325375460184776,
"learning_rate": 3.6053966464703398e-06,
"loss": 0.7434,
"mean_token_accuracy": 0.8035388976335526,
"num_tokens": 389826917.0,
"step": 4400
},
{
"entropy": 1.0162109375,
"epoch": 0.6027472151985239,
"grad_norm": 0.7468714317875655,
"learning_rate": 3.6018740312808233e-06,
"loss": 0.7227,
"mean_token_accuracy": 0.8047880202531814,
"num_tokens": 390768332.0,
"step": 4410
},
{
"entropy": 0.9724609375,
"epoch": 0.6041139889291328,
"grad_norm": 0.8252190927175495,
"learning_rate": 3.5983514160913064e-06,
"loss": 0.6801,
"mean_token_accuracy": 0.813497906923294,
"num_tokens": 391673598.0,
"step": 4420
},
{
"entropy": 0.9982421875,
"epoch": 0.6054807626597417,
"grad_norm": 0.7147319709359161,
"learning_rate": 3.5948288009017895e-06,
"loss": 0.7018,
"mean_token_accuracy": 0.8130618840456009,
"num_tokens": 392570813.0,
"step": 4430
},
{
"entropy": 1.0060546875,
"epoch": 0.6068475363903506,
"grad_norm": 0.8010099783838767,
"learning_rate": 3.591306185712273e-06,
"loss": 0.7383,
"mean_token_accuracy": 0.8041794985532761,
"num_tokens": 393449571.0,
"step": 4440
},
{
"entropy": 1.034765625,
"epoch": 0.6082143101209595,
"grad_norm": 0.7578374921780676,
"learning_rate": 3.5877835705227566e-06,
"loss": 0.725,
"mean_token_accuracy": 0.8061411321163178,
"num_tokens": 394310855.0,
"step": 4450
},
{
"entropy": 0.9693359375,
"epoch": 0.6095810838515684,
"grad_norm": 1.071165642769431,
"learning_rate": 3.5842609553332398e-06,
"loss": 0.6856,
"mean_token_accuracy": 0.8157458066940307,
"num_tokens": 395201869.0,
"step": 4460
},
{
"entropy": 0.9814453125,
"epoch": 0.6109478575821773,
"grad_norm": 0.7921633829437134,
"learning_rate": 3.580738340143723e-06,
"loss": 0.6908,
"mean_token_accuracy": 0.8135299861431122,
"num_tokens": 396084305.0,
"step": 4470
},
{
"entropy": 0.9880859375,
"epoch": 0.6123146313127862,
"grad_norm": 0.7795139876059526,
"learning_rate": 3.577215724954206e-06,
"loss": 0.7016,
"mean_token_accuracy": 0.8112848818302154,
"num_tokens": 396999249.0,
"step": 4480
},
{
"entropy": 1.0203125,
"epoch": 0.6136814050433951,
"grad_norm": 0.6948212820119783,
"learning_rate": 3.57369310976469e-06,
"loss": 0.7303,
"mean_token_accuracy": 0.8034562349319458,
"num_tokens": 397841168.0,
"step": 4490
},
{
"entropy": 1.0150390625,
"epoch": 0.615048178774004,
"grad_norm": 0.7469473076744361,
"learning_rate": 3.570170494575173e-06,
"loss": 0.7139,
"mean_token_accuracy": 0.8089707493782043,
"num_tokens": 398754063.0,
"step": 4500
},
{
"entropy": 1.0435546875,
"epoch": 0.6164149525046129,
"grad_norm": 0.7656024175265354,
"learning_rate": 3.5666478793856562e-06,
"loss": 0.7783,
"mean_token_accuracy": 0.793326461315155,
"num_tokens": 399621156.0,
"step": 4510
},
{
"entropy": 0.9787109375,
"epoch": 0.6177817262352218,
"grad_norm": 0.8526249877362821,
"learning_rate": 3.5631252641961394e-06,
"loss": 0.6818,
"mean_token_accuracy": 0.8143033504486084,
"num_tokens": 400472687.0,
"step": 4520
},
{
"entropy": 0.933984375,
"epoch": 0.6191484999658307,
"grad_norm": 0.7479363741795042,
"learning_rate": 3.5596026490066225e-06,
"loss": 0.6504,
"mean_token_accuracy": 0.821707233786583,
"num_tokens": 401390570.0,
"step": 4530
},
{
"entropy": 0.978515625,
"epoch": 0.6205152736964396,
"grad_norm": 0.7876823375693502,
"learning_rate": 3.556080033817106e-06,
"loss": 0.6786,
"mean_token_accuracy": 0.8164149731397629,
"num_tokens": 402248186.0,
"step": 4540
},
{
"entropy": 1.0412109375,
"epoch": 0.6218820474270484,
"grad_norm": 0.7796170878595704,
"learning_rate": 3.5525574186275896e-06,
"loss": 0.7481,
"mean_token_accuracy": 0.8013476938009262,
"num_tokens": 403093559.0,
"step": 4550
},
{
"entropy": 1.0296875,
"epoch": 0.6232488211576573,
"grad_norm": 0.8458091683467284,
"learning_rate": 3.5490348034380727e-06,
"loss": 0.7248,
"mean_token_accuracy": 0.806901079416275,
"num_tokens": 403992480.0,
"step": 4560
},
{
"entropy": 0.9953125,
"epoch": 0.6246155948882662,
"grad_norm": 0.7473278833932094,
"learning_rate": 3.545512188248556e-06,
"loss": 0.6898,
"mean_token_accuracy": 0.8141829431056976,
"num_tokens": 404853974.0,
"step": 4570
},
{
"entropy": 0.9763671875,
"epoch": 0.6259823686188751,
"grad_norm": 0.7354880061355658,
"learning_rate": 3.5419895730590394e-06,
"loss": 0.6837,
"mean_token_accuracy": 0.8130509853363037,
"num_tokens": 405698785.0,
"step": 4580
},
{
"entropy": 1.0306640625,
"epoch": 0.627349142349484,
"grad_norm": 0.7636651540740428,
"learning_rate": 3.5384669578695225e-06,
"loss": 0.6988,
"mean_token_accuracy": 0.8105187326669693,
"num_tokens": 406574395.0,
"step": 4590
},
{
"entropy": 1.025390625,
"epoch": 0.6287159160800929,
"grad_norm": 0.7973293735772822,
"learning_rate": 3.534944342680006e-06,
"loss": 0.7543,
"mean_token_accuracy": 0.801137262582779,
"num_tokens": 407487257.0,
"step": 4600
},
{
"entropy": 0.9716796875,
"epoch": 0.6300826898107018,
"grad_norm": 0.7833131317449497,
"learning_rate": 3.5314217274904896e-06,
"loss": 0.7061,
"mean_token_accuracy": 0.8094787478446961,
"num_tokens": 408376884.0,
"step": 4610
},
{
"entropy": 1.0076171875,
"epoch": 0.6314494635413107,
"grad_norm": 0.71042897956098,
"learning_rate": 3.5278991123009727e-06,
"loss": 0.7078,
"mean_token_accuracy": 0.8103502720594407,
"num_tokens": 409246879.0,
"step": 4620
},
{
"entropy": 1.030859375,
"epoch": 0.6328162372719196,
"grad_norm": 0.7998210090002549,
"learning_rate": 3.524376497111456e-06,
"loss": 0.7065,
"mean_token_accuracy": 0.808617752790451,
"num_tokens": 410153996.0,
"step": 4630
},
{
"entropy": 0.967578125,
"epoch": 0.6341830110025285,
"grad_norm": 0.8245583157613016,
"learning_rate": 3.520853881921939e-06,
"loss": 0.6733,
"mean_token_accuracy": 0.8186338365077972,
"num_tokens": 411026191.0,
"step": 4640
},
{
"entropy": 0.966796875,
"epoch": 0.6355497847331374,
"grad_norm": 0.7748247823583794,
"learning_rate": 3.517331266732422e-06,
"loss": 0.7117,
"mean_token_accuracy": 0.8104182779788971,
"num_tokens": 411945110.0,
"step": 4650
},
{
"entropy": 1.0587890625,
"epoch": 0.6369165584637463,
"grad_norm": 0.8215991462918023,
"learning_rate": 3.513808651542906e-06,
"loss": 0.7986,
"mean_token_accuracy": 0.7936654210090637,
"num_tokens": 412810803.0,
"step": 4660
},
{
"entropy": 0.99609375,
"epoch": 0.6382833321943552,
"grad_norm": 0.8013440404409816,
"learning_rate": 3.510286036353389e-06,
"loss": 0.6629,
"mean_token_accuracy": 0.8191068142652511,
"num_tokens": 413638372.0,
"step": 4670
},
{
"entropy": 0.98125,
"epoch": 0.6396501059249641,
"grad_norm": 0.6886685227093562,
"learning_rate": 3.5067634211638723e-06,
"loss": 0.6677,
"mean_token_accuracy": 0.8166958808898925,
"num_tokens": 414557899.0,
"step": 4680
},
{
"entropy": 0.9884765625,
"epoch": 0.641016879655573,
"grad_norm": 0.754053253083038,
"learning_rate": 3.5032408059743554e-06,
"loss": 0.6856,
"mean_token_accuracy": 0.813271215558052,
"num_tokens": 415455781.0,
"step": 4690
},
{
"entropy": 0.9873046875,
"epoch": 0.6423836533861819,
"grad_norm": 0.7804370669218913,
"learning_rate": 3.499718190784839e-06,
"loss": 0.6943,
"mean_token_accuracy": 0.8123634487390519,
"num_tokens": 416295102.0,
"step": 4700
},
{
"entropy": 0.9912109375,
"epoch": 0.6437504271167909,
"grad_norm": 0.7921701259702444,
"learning_rate": 3.4961955755953225e-06,
"loss": 0.7203,
"mean_token_accuracy": 0.8074250906705857,
"num_tokens": 417180585.0,
"step": 4710
},
{
"entropy": 1.0283203125,
"epoch": 0.6451172008473998,
"grad_norm": 0.7357318184259359,
"learning_rate": 3.4926729604058056e-06,
"loss": 0.704,
"mean_token_accuracy": 0.8103884845972061,
"num_tokens": 418052757.0,
"step": 4720
},
{
"entropy": 0.983984375,
"epoch": 0.6464839745780087,
"grad_norm": 0.8794226513360007,
"learning_rate": 3.4891503452162888e-06,
"loss": 0.6875,
"mean_token_accuracy": 0.8144732654094696,
"num_tokens": 418895912.0,
"step": 4730
},
{
"entropy": 0.973046875,
"epoch": 0.6478507483086176,
"grad_norm": 0.706085953195538,
"learning_rate": 3.4856277300267723e-06,
"loss": 0.6581,
"mean_token_accuracy": 0.8197737693786621,
"num_tokens": 419758665.0,
"step": 4740
},
{
"entropy": 0.99140625,
"epoch": 0.6492175220392264,
"grad_norm": 0.8083089594936401,
"learning_rate": 3.4821051148372554e-06,
"loss": 0.7063,
"mean_token_accuracy": 0.8093706965446472,
"num_tokens": 420619426.0,
"step": 4750
},
{
"entropy": 0.9435546875,
"epoch": 0.6505842957698353,
"grad_norm": 0.8461727702730071,
"learning_rate": 3.4785824996477386e-06,
"loss": 0.6284,
"mean_token_accuracy": 0.8264329731464386,
"num_tokens": 421466711.0,
"step": 4760
},
{
"entropy": 0.9373046875,
"epoch": 0.6519510695004442,
"grad_norm": 0.8245601478932103,
"learning_rate": 3.475059884458222e-06,
"loss": 0.666,
"mean_token_accuracy": 0.8191474914550781,
"num_tokens": 422336352.0,
"step": 4770
},
{
"entropy": 0.98828125,
"epoch": 0.6533178432310531,
"grad_norm": 0.7752098737364052,
"learning_rate": 3.4715372692687057e-06,
"loss": 0.6719,
"mean_token_accuracy": 0.8171870946884155,
"num_tokens": 423184508.0,
"step": 4780
},
{
"entropy": 1.0462890625,
"epoch": 0.654684616961662,
"grad_norm": 0.7936171032161543,
"learning_rate": 3.4680146540791888e-06,
"loss": 0.7325,
"mean_token_accuracy": 0.806986403465271,
"num_tokens": 424081913.0,
"step": 4790
},
{
"entropy": 0.9939453125,
"epoch": 0.6560513906922709,
"grad_norm": 0.7302265068067899,
"learning_rate": 3.464492038889672e-06,
"loss": 0.7077,
"mean_token_accuracy": 0.8106041520833969,
"num_tokens": 424917247.0,
"step": 4800
},
{
"entropy": 0.984765625,
"epoch": 0.6574181644228798,
"grad_norm": 0.8248106428225372,
"learning_rate": 3.460969423700155e-06,
"loss": 0.7385,
"mean_token_accuracy": 0.8018399208784104,
"num_tokens": 425842562.0,
"step": 4810
},
{
"entropy": 0.9896484375,
"epoch": 0.6587849381534887,
"grad_norm": 0.7500695788518175,
"learning_rate": 3.457446808510639e-06,
"loss": 0.6999,
"mean_token_accuracy": 0.8119636684656143,
"num_tokens": 426738885.0,
"step": 4820
},
{
"entropy": 0.9623046875,
"epoch": 0.6601517118840976,
"grad_norm": 0.7828513213625707,
"learning_rate": 3.453924193321122e-06,
"loss": 0.6865,
"mean_token_accuracy": 0.8142214059829712,
"num_tokens": 427666165.0,
"step": 4830
},
{
"entropy": 0.96484375,
"epoch": 0.6615184856147065,
"grad_norm": 0.7958518917360529,
"learning_rate": 3.4504015781316052e-06,
"loss": 0.6628,
"mean_token_accuracy": 0.8215816587209701,
"num_tokens": 428584533.0,
"step": 4840
},
{
"entropy": 1.009375,
"epoch": 0.6628852593453154,
"grad_norm": 0.7881943365367393,
"learning_rate": 3.4468789629420884e-06,
"loss": 0.7066,
"mean_token_accuracy": 0.8093484163284301,
"num_tokens": 429522884.0,
"step": 4850
},
{
"entropy": 0.953125,
"epoch": 0.6642520330759243,
"grad_norm": 0.7703850008059547,
"learning_rate": 3.4433563477525715e-06,
"loss": 0.6802,
"mean_token_accuracy": 0.8150426268577575,
"num_tokens": 430404622.0,
"step": 4860
},
{
"entropy": 0.9748046875,
"epoch": 0.6656188068065332,
"grad_norm": 0.8057162220008176,
"learning_rate": 3.439833732563055e-06,
"loss": 0.6658,
"mean_token_accuracy": 0.8181778013706207,
"num_tokens": 431275413.0,
"step": 4870
},
{
"entropy": 0.9787109375,
"epoch": 0.6669855805371421,
"grad_norm": 0.8801839886077731,
"learning_rate": 3.4363111173735386e-06,
"loss": 0.6852,
"mean_token_accuracy": 0.8137222230434418,
"num_tokens": 432204493.0,
"step": 4880
},
{
"entropy": 0.9984375,
"epoch": 0.668352354267751,
"grad_norm": 0.8055210425732408,
"learning_rate": 3.4327885021840217e-06,
"loss": 0.7008,
"mean_token_accuracy": 0.8121373176574707,
"num_tokens": 433107951.0,
"step": 4890
},
{
"entropy": 0.9748046875,
"epoch": 0.6697191279983599,
"grad_norm": 0.735180172938854,
"learning_rate": 3.429265886994505e-06,
"loss": 0.7238,
"mean_token_accuracy": 0.8066814750432968,
"num_tokens": 434029585.0,
"step": 4900
},
{
"entropy": 0.9650390625,
"epoch": 0.6710859017289688,
"grad_norm": 0.7896278172005097,
"learning_rate": 3.4257432718049884e-06,
"loss": 0.6703,
"mean_token_accuracy": 0.817569237947464,
"num_tokens": 434937916.0,
"step": 4910
},
{
"entropy": 0.950390625,
"epoch": 0.6724526754595777,
"grad_norm": 0.7592762477124009,
"learning_rate": 3.4222206566154715e-06,
"loss": 0.7197,
"mean_token_accuracy": 0.8089982807636261,
"num_tokens": 435852833.0,
"step": 4920
},
{
"entropy": 1.0130859375,
"epoch": 0.6738194491901865,
"grad_norm": 0.7605195634694151,
"learning_rate": 3.418698041425955e-06,
"loss": 0.7361,
"mean_token_accuracy": 0.8032962709665299,
"num_tokens": 436724128.0,
"step": 4930
},
{
"entropy": 0.9529296875,
"epoch": 0.6751862229207954,
"grad_norm": 0.7853391469333263,
"learning_rate": 3.415175426236438e-06,
"loss": 0.664,
"mean_token_accuracy": 0.8194535046815872,
"num_tokens": 437633875.0,
"step": 4940
},
{
"entropy": 0.9220703125,
"epoch": 0.6765529966514043,
"grad_norm": 0.7010159288730016,
"learning_rate": 3.4116528110469217e-06,
"loss": 0.6266,
"mean_token_accuracy": 0.8283613711595536,
"num_tokens": 438515714.0,
"step": 4950
},
{
"entropy": 0.980078125,
"epoch": 0.6779197703820132,
"grad_norm": 0.8020582641256698,
"learning_rate": 3.408130195857405e-06,
"loss": 0.6553,
"mean_token_accuracy": 0.8216167658567428,
"num_tokens": 439364688.0,
"step": 4960
},
{
"entropy": 0.9974609375,
"epoch": 0.6792865441126221,
"grad_norm": 0.8537365128279352,
"learning_rate": 3.404607580667888e-06,
"loss": 0.6966,
"mean_token_accuracy": 0.8113814920186997,
"num_tokens": 440259473.0,
"step": 4970
},
{
"entropy": 0.965234375,
"epoch": 0.680653317843231,
"grad_norm": 0.7385278710879903,
"learning_rate": 3.401084965478371e-06,
"loss": 0.6436,
"mean_token_accuracy": 0.8243313044309616,
"num_tokens": 441127912.0,
"step": 4980
},
{
"entropy": 0.989453125,
"epoch": 0.6820200915738399,
"grad_norm": 0.7265718418236922,
"learning_rate": 3.397562350288855e-06,
"loss": 0.699,
"mean_token_accuracy": 0.811851117014885,
"num_tokens": 442024079.0,
"step": 4990
},
{
"entropy": 0.9775390625,
"epoch": 0.6833868653044488,
"grad_norm": 0.7649915055995465,
"learning_rate": 3.394039735099338e-06,
"loss": 0.6515,
"mean_token_accuracy": 0.8207689315080643,
"num_tokens": 442910647.0,
"step": 5000
},
{
"entropy": 0.95234375,
"epoch": 0.6847536390350577,
"grad_norm": 0.8275150058486563,
"learning_rate": 3.3905171199098213e-06,
"loss": 0.6375,
"mean_token_accuracy": 0.8233121156692504,
"num_tokens": 443789732.0,
"step": 5010
},
{
"entropy": 0.9654296875,
"epoch": 0.6861204127656666,
"grad_norm": 0.7030953536312434,
"learning_rate": 3.3869945047203044e-06,
"loss": 0.7016,
"mean_token_accuracy": 0.8120089203119278,
"num_tokens": 444692779.0,
"step": 5020
},
{
"entropy": 0.966015625,
"epoch": 0.6874871864962755,
"grad_norm": 0.786784971686847,
"learning_rate": 3.3834718895307876e-06,
"loss": 0.6579,
"mean_token_accuracy": 0.8200263470411301,
"num_tokens": 445556425.0,
"step": 5030
},
{
"entropy": 0.9494140625,
"epoch": 0.6888539602268844,
"grad_norm": 0.713823912661295,
"learning_rate": 3.3799492743412715e-06,
"loss": 0.6601,
"mean_token_accuracy": 0.8212069690227508,
"num_tokens": 446440835.0,
"step": 5040
},
{
"entropy": 0.965234375,
"epoch": 0.6902207339574933,
"grad_norm": 0.7145196729480466,
"learning_rate": 3.3764266591517547e-06,
"loss": 0.6666,
"mean_token_accuracy": 0.8177187502384186,
"num_tokens": 447344281.0,
"step": 5050
},
{
"entropy": 0.981640625,
"epoch": 0.6915875076881023,
"grad_norm": 0.7385335928313015,
"learning_rate": 3.3729040439622378e-06,
"loss": 0.6836,
"mean_token_accuracy": 0.8160740435123444,
"num_tokens": 448244245.0,
"step": 5060
},
{
"entropy": 1.0064453125,
"epoch": 0.6929542814187112,
"grad_norm": 0.8867927328307652,
"learning_rate": 3.3693814287727213e-06,
"loss": 0.7214,
"mean_token_accuracy": 0.8073490649461746,
"num_tokens": 449100309.0,
"step": 5070
},
{
"entropy": 1.0158203125,
"epoch": 0.6943210551493201,
"grad_norm": 0.8548441574524056,
"learning_rate": 3.3658588135832044e-06,
"loss": 0.7197,
"mean_token_accuracy": 0.8067093312740325,
"num_tokens": 450042185.0,
"step": 5080
},
{
"entropy": 0.95703125,
"epoch": 0.695687828879929,
"grad_norm": 0.6690838643513302,
"learning_rate": 3.3623361983936876e-06,
"loss": 0.6896,
"mean_token_accuracy": 0.8147846072912216,
"num_tokens": 450988603.0,
"step": 5090
},
{
"entropy": 0.9802734375,
"epoch": 0.6970546026105379,
"grad_norm": 0.7463463755099142,
"learning_rate": 3.358813583204171e-06,
"loss": 0.6721,
"mean_token_accuracy": 0.8182615458965301,
"num_tokens": 451928945.0,
"step": 5100
},
{
"entropy": 1.042578125,
"epoch": 0.6984213763411468,
"grad_norm": 0.8233570978033394,
"learning_rate": 3.3552909680146547e-06,
"loss": 0.7278,
"mean_token_accuracy": 0.803802227973938,
"num_tokens": 452812071.0,
"step": 5110
},
{
"entropy": 0.959375,
"epoch": 0.6997881500717557,
"grad_norm": 0.7458494422889655,
"learning_rate": 3.3517683528251378e-06,
"loss": 0.6666,
"mean_token_accuracy": 0.8179999560117721,
"num_tokens": 453702565.0,
"step": 5120
},
{
"entropy": 0.9703125,
"epoch": 0.7011549238023645,
"grad_norm": 0.7962870207491588,
"learning_rate": 3.348245737635621e-06,
"loss": 0.6609,
"mean_token_accuracy": 0.819678321480751,
"num_tokens": 454608169.0,
"step": 5130
},
{
"entropy": 0.989453125,
"epoch": 0.7025216975329734,
"grad_norm": 0.7366577117843924,
"learning_rate": 3.344723122446104e-06,
"loss": 0.6738,
"mean_token_accuracy": 0.8172920554876327,
"num_tokens": 455455664.0,
"step": 5140
},
{
"entropy": 0.996875,
"epoch": 0.7038884712635823,
"grad_norm": 0.8106944807698179,
"learning_rate": 3.341200507256587e-06,
"loss": 0.6978,
"mean_token_accuracy": 0.8135390132665634,
"num_tokens": 456331907.0,
"step": 5150
},
{
"entropy": 1.020703125,
"epoch": 0.7052552449941912,
"grad_norm": 0.839168444579637,
"learning_rate": 3.337677892067071e-06,
"loss": 0.6818,
"mean_token_accuracy": 0.815180429816246,
"num_tokens": 457208256.0,
"step": 5160
},
{
"entropy": 0.9990234375,
"epoch": 0.7066220187248001,
"grad_norm": 0.9289539777129413,
"learning_rate": 3.3341552768775543e-06,
"loss": 0.7332,
"mean_token_accuracy": 0.8035592555999755,
"num_tokens": 458092829.0,
"step": 5170
},
{
"entropy": 0.9990234375,
"epoch": 0.707988792455409,
"grad_norm": 0.7789936491377681,
"learning_rate": 3.3306326616880374e-06,
"loss": 0.6625,
"mean_token_accuracy": 0.8212026566267013,
"num_tokens": 458977684.0,
"step": 5180
},
{
"entropy": 1.008984375,
"epoch": 0.7093555661860179,
"grad_norm": 0.7500121200596604,
"learning_rate": 3.3271100464985205e-06,
"loss": 0.7027,
"mean_token_accuracy": 0.8096355408430099,
"num_tokens": 459877261.0,
"step": 5190
},
{
"entropy": 0.971875,
"epoch": 0.7107223399166268,
"grad_norm": 0.73694088038812,
"learning_rate": 3.323587431309004e-06,
"loss": 0.6724,
"mean_token_accuracy": 0.8191078066825866,
"num_tokens": 460803559.0,
"step": 5200
},
{
"entropy": 0.9990234375,
"epoch": 0.7120891136472357,
"grad_norm": 0.8006977201788347,
"learning_rate": 3.3200648161194876e-06,
"loss": 0.6847,
"mean_token_accuracy": 0.814735722541809,
"num_tokens": 461698177.0,
"step": 5210
},
{
"entropy": 0.9697265625,
"epoch": 0.7134558873778446,
"grad_norm": 0.8045650549766643,
"learning_rate": 3.3165422009299707e-06,
"loss": 0.6851,
"mean_token_accuracy": 0.8142364144325256,
"num_tokens": 462589164.0,
"step": 5220
},
{
"entropy": 1.0083984375,
"epoch": 0.7148226611084535,
"grad_norm": 0.7516171789491205,
"learning_rate": 3.313019585740454e-06,
"loss": 0.683,
"mean_token_accuracy": 0.8134177654981614,
"num_tokens": 463468365.0,
"step": 5230
},
{
"entropy": 1.0087890625,
"epoch": 0.7161894348390624,
"grad_norm": 0.7920893230502382,
"learning_rate": 3.3094969705509374e-06,
"loss": 0.7384,
"mean_token_accuracy": 0.8031978994607926,
"num_tokens": 464350433.0,
"step": 5240
},
{
"entropy": 0.975,
"epoch": 0.7175562085696713,
"grad_norm": 0.7586331506033891,
"learning_rate": 3.3059743553614205e-06,
"loss": 0.701,
"mean_token_accuracy": 0.8124682247638703,
"num_tokens": 465221702.0,
"step": 5250
},
{
"entropy": 0.9765625,
"epoch": 0.7189229823002802,
"grad_norm": 0.7861269959912984,
"learning_rate": 3.3024517401719036e-06,
"loss": 0.6805,
"mean_token_accuracy": 0.8138928323984146,
"num_tokens": 466090098.0,
"step": 5260
},
{
"entropy": 0.9955078125,
"epoch": 0.7202897560308891,
"grad_norm": 0.9782764585007452,
"learning_rate": 3.298929124982387e-06,
"loss": 0.7527,
"mean_token_accuracy": 0.8096537053585052,
"num_tokens": 466990039.0,
"step": 5270
},
{
"entropy": 1.0111328125,
"epoch": 0.721656529761498,
"grad_norm": 0.748989195129646,
"learning_rate": 3.2954065097928707e-06,
"loss": 0.7155,
"mean_token_accuracy": 0.8090497046709061,
"num_tokens": 467896781.0,
"step": 5280
},
{
"entropy": 0.9943359375,
"epoch": 0.7230233034921069,
"grad_norm": 0.7820790774126962,
"learning_rate": 3.291883894603354e-06,
"loss": 0.6991,
"mean_token_accuracy": 0.8125780820846558,
"num_tokens": 468784684.0,
"step": 5290
},
{
"entropy": 0.9626953125,
"epoch": 0.7243900772227158,
"grad_norm": 0.8212461849669104,
"learning_rate": 3.288361279413837e-06,
"loss": 0.6864,
"mean_token_accuracy": 0.8125094622373581,
"num_tokens": 469661887.0,
"step": 5300
},
{
"entropy": 0.964453125,
"epoch": 0.7257568509533247,
"grad_norm": 0.6939835170124516,
"learning_rate": 3.28483866422432e-06,
"loss": 0.6537,
"mean_token_accuracy": 0.8224819511175155,
"num_tokens": 470540659.0,
"step": 5310
},
{
"entropy": 1.0095703125,
"epoch": 0.7271236246839335,
"grad_norm": 0.7413581697632078,
"learning_rate": 3.281316049034804e-06,
"loss": 0.7302,
"mean_token_accuracy": 0.8021710962057114,
"num_tokens": 471483298.0,
"step": 5320
},
{
"entropy": 0.96328125,
"epoch": 0.7284903984145424,
"grad_norm": 0.7491249897929223,
"learning_rate": 3.277793433845287e-06,
"loss": 0.6515,
"mean_token_accuracy": 0.8194939762353897,
"num_tokens": 472374357.0,
"step": 5330
},
{
"entropy": 0.9853515625,
"epoch": 0.7298571721451513,
"grad_norm": 0.7831037770783656,
"learning_rate": 3.2742708186557703e-06,
"loss": 0.6663,
"mean_token_accuracy": 0.8200128763914108,
"num_tokens": 473290428.0,
"step": 5340
},
{
"entropy": 1.023046875,
"epoch": 0.7312239458757602,
"grad_norm": 0.7979761078658298,
"learning_rate": 3.2707482034662534e-06,
"loss": 0.6992,
"mean_token_accuracy": 0.8108828246593476,
"num_tokens": 474178439.0,
"step": 5350
},
{
"entropy": 0.98671875,
"epoch": 0.7325907196063691,
"grad_norm": 0.8053953799226069,
"learning_rate": 3.2672255882767366e-06,
"loss": 0.6942,
"mean_token_accuracy": 0.8123347967863083,
"num_tokens": 475073389.0,
"step": 5360
},
{
"entropy": 0.944921875,
"epoch": 0.733957493336978,
"grad_norm": 0.6903897271747469,
"learning_rate": 3.26370297308722e-06,
"loss": 0.6497,
"mean_token_accuracy": 0.8228194087743759,
"num_tokens": 475982346.0,
"step": 5370
},
{
"entropy": 0.9767578125,
"epoch": 0.7353242670675869,
"grad_norm": 0.740687113249482,
"learning_rate": 3.2601803578977037e-06,
"loss": 0.701,
"mean_token_accuracy": 0.8102813512086868,
"num_tokens": 476863931.0,
"step": 5380
},
{
"entropy": 0.9478515625,
"epoch": 0.7366910407981958,
"grad_norm": 0.7138499582876328,
"learning_rate": 3.2566577427081868e-06,
"loss": 0.6594,
"mean_token_accuracy": 0.8197508186101914,
"num_tokens": 477781550.0,
"step": 5390
},
{
"entropy": 0.9796875,
"epoch": 0.7380578145288047,
"grad_norm": 0.7678922415703211,
"learning_rate": 3.25313512751867e-06,
"loss": 0.6821,
"mean_token_accuracy": 0.8164928674697876,
"num_tokens": 478654068.0,
"step": 5400
},
{
"entropy": 1.0037109375,
"epoch": 0.7394245882594137,
"grad_norm": 0.6954115720647032,
"learning_rate": 3.2496125123291535e-06,
"loss": 0.7207,
"mean_token_accuracy": 0.8070971161127091,
"num_tokens": 479541351.0,
"step": 5410
},
{
"entropy": 0.9896484375,
"epoch": 0.7407913619900226,
"grad_norm": 0.7613021705954046,
"learning_rate": 3.2460898971396366e-06,
"loss": 0.665,
"mean_token_accuracy": 0.8195907413959503,
"num_tokens": 480430820.0,
"step": 5420
},
{
"entropy": 0.99140625,
"epoch": 0.7421581357206315,
"grad_norm": 0.7931564502746363,
"learning_rate": 3.24256728195012e-06,
"loss": 0.6956,
"mean_token_accuracy": 0.8144201606512069,
"num_tokens": 481306833.0,
"step": 5430
},
{
"entropy": 0.9638671875,
"epoch": 0.7435249094512404,
"grad_norm": 0.7661585541195064,
"learning_rate": 3.2390446667606037e-06,
"loss": 0.6765,
"mean_token_accuracy": 0.8177257180213928,
"num_tokens": 482218500.0,
"step": 5440
},
{
"entropy": 0.994921875,
"epoch": 0.7448916831818493,
"grad_norm": 0.7946780830496234,
"learning_rate": 3.235522051571087e-06,
"loss": 0.6759,
"mean_token_accuracy": 0.8169646680355072,
"num_tokens": 483099935.0,
"step": 5450
},
{
"entropy": 1.0009765625,
"epoch": 0.7462584569124582,
"grad_norm": 0.6644718748853345,
"learning_rate": 3.23199943638157e-06,
"loss": 0.6818,
"mean_token_accuracy": 0.8160134613513946,
"num_tokens": 483948988.0,
"step": 5460
},
{
"entropy": 1.006640625,
"epoch": 0.7476252306430671,
"grad_norm": 0.8832145947204654,
"learning_rate": 3.228476821192053e-06,
"loss": 0.6945,
"mean_token_accuracy": 0.8116898357868194,
"num_tokens": 484844939.0,
"step": 5470
},
{
"entropy": 0.9978515625,
"epoch": 0.748992004373676,
"grad_norm": 0.8228008751203593,
"learning_rate": 3.224954206002536e-06,
"loss": 0.7302,
"mean_token_accuracy": 0.8033615976572037,
"num_tokens": 485763455.0,
"step": 5480
},
{
"entropy": 0.9611328125,
"epoch": 0.7503587781042849,
"grad_norm": 0.7260958096819845,
"learning_rate": 3.22143159081302e-06,
"loss": 0.6267,
"mean_token_accuracy": 0.8273924142122269,
"num_tokens": 486656249.0,
"step": 5490
},
{
"entropy": 1.017578125,
"epoch": 0.7517255518348938,
"grad_norm": 0.8877073119414157,
"learning_rate": 3.2179089756235033e-06,
"loss": 0.6987,
"mean_token_accuracy": 0.8094022601842881,
"num_tokens": 487540315.0,
"step": 5500
},
{
"entropy": 1.0201171875,
"epoch": 0.7530923255655027,
"grad_norm": 0.7276912936949571,
"learning_rate": 3.2143863604339864e-06,
"loss": 0.7107,
"mean_token_accuracy": 0.8096262782812118,
"num_tokens": 488445699.0,
"step": 5510
},
{
"entropy": 0.985546875,
"epoch": 0.7544590992961115,
"grad_norm": 0.7165040973047131,
"learning_rate": 3.2108637452444695e-06,
"loss": 0.6956,
"mean_token_accuracy": 0.8135731875896454,
"num_tokens": 489326155.0,
"step": 5520
},
{
"entropy": 1.0296875,
"epoch": 0.7558258730267204,
"grad_norm": 0.7344883478027465,
"learning_rate": 3.2073411300549526e-06,
"loss": 0.7343,
"mean_token_accuracy": 0.8044039338827134,
"num_tokens": 490191644.0,
"step": 5530
},
{
"entropy": 0.948046875,
"epoch": 0.7571926467573293,
"grad_norm": 0.6879540546873437,
"learning_rate": 3.2038185148654366e-06,
"loss": 0.6287,
"mean_token_accuracy": 0.8271225363016128,
"num_tokens": 491068717.0,
"step": 5540
},
{
"entropy": 1.0328125,
"epoch": 0.7585594204879382,
"grad_norm": 0.7642456569172467,
"learning_rate": 3.2002958996759197e-06,
"loss": 0.7601,
"mean_token_accuracy": 0.8016656130552292,
"num_tokens": 491916966.0,
"step": 5550
},
{
"entropy": 1.037109375,
"epoch": 0.7599261942185471,
"grad_norm": 0.853382962780325,
"learning_rate": 3.196773284486403e-06,
"loss": 0.7253,
"mean_token_accuracy": 0.8037023663520813,
"num_tokens": 492799007.0,
"step": 5560
},
{
"entropy": 1.02890625,
"epoch": 0.761292967949156,
"grad_norm": 0.7908111237215389,
"learning_rate": 3.1932506692968864e-06,
"loss": 0.7313,
"mean_token_accuracy": 0.8061786532402039,
"num_tokens": 493694735.0,
"step": 5570
},
{
"entropy": 0.9759765625,
"epoch": 0.7626597416797649,
"grad_norm": 0.7807152730478378,
"learning_rate": 3.1897280541073695e-06,
"loss": 0.6635,
"mean_token_accuracy": 0.8187363833189011,
"num_tokens": 494599096.0,
"step": 5580
},
{
"entropy": 0.9912109375,
"epoch": 0.7640265154103738,
"grad_norm": 0.6871073801521973,
"learning_rate": 3.1862054389178526e-06,
"loss": 0.676,
"mean_token_accuracy": 0.8147340655326843,
"num_tokens": 495533785.0,
"step": 5590
},
{
"entropy": 0.9748046875,
"epoch": 0.7653932891409827,
"grad_norm": 0.7591202845444404,
"learning_rate": 3.182682823728336e-06,
"loss": 0.6855,
"mean_token_accuracy": 0.8144841700792312,
"num_tokens": 496427337.0,
"step": 5600
},
{
"entropy": 0.9611328125,
"epoch": 0.7667600628715916,
"grad_norm": 0.7444544330156877,
"learning_rate": 3.1791602085388197e-06,
"loss": 0.6747,
"mean_token_accuracy": 0.8156604588031768,
"num_tokens": 497328578.0,
"step": 5610
},
{
"entropy": 1.0294921875,
"epoch": 0.7681268366022005,
"grad_norm": 0.7415592958193216,
"learning_rate": 3.175637593349303e-06,
"loss": 0.7437,
"mean_token_accuracy": 0.8028131932020187,
"num_tokens": 498197795.0,
"step": 5620
},
{
"entropy": 1.0185546875,
"epoch": 0.7694936103328094,
"grad_norm": 0.7195317679961294,
"learning_rate": 3.172114978159786e-06,
"loss": 0.7131,
"mean_token_accuracy": 0.8090102344751358,
"num_tokens": 499104957.0,
"step": 5630
},
{
"entropy": 0.9986328125,
"epoch": 0.7708603840634183,
"grad_norm": 0.7472628305627229,
"learning_rate": 3.168592362970269e-06,
"loss": 0.6857,
"mean_token_accuracy": 0.8149809598922729,
"num_tokens": 499955528.0,
"step": 5640
},
{
"entropy": 0.96171875,
"epoch": 0.7722271577940272,
"grad_norm": 0.7818044673725272,
"learning_rate": 3.165069747780753e-06,
"loss": 0.6589,
"mean_token_accuracy": 0.8206013828516007,
"num_tokens": 500817635.0,
"step": 5650
},
{
"entropy": 1.0197265625,
"epoch": 0.7735939315246361,
"grad_norm": 0.7775916854966156,
"learning_rate": 3.161547132591236e-06,
"loss": 0.7097,
"mean_token_accuracy": 0.8084346503019333,
"num_tokens": 501718725.0,
"step": 5660
},
{
"entropy": 1.009375,
"epoch": 0.774960705255245,
"grad_norm": 0.6367508292477452,
"learning_rate": 3.1580245174017193e-06,
"loss": 0.7036,
"mean_token_accuracy": 0.809009513258934,
"num_tokens": 502611865.0,
"step": 5670
},
{
"entropy": 1.01171875,
"epoch": 0.7763274789858539,
"grad_norm": 0.7472303087261686,
"learning_rate": 3.1545019022122025e-06,
"loss": 0.6663,
"mean_token_accuracy": 0.8186325848102569,
"num_tokens": 503500310.0,
"step": 5680
},
{
"entropy": 1.0453125,
"epoch": 0.7776942527164628,
"grad_norm": 0.7333216438916991,
"learning_rate": 3.1509792870226856e-06,
"loss": 0.7223,
"mean_token_accuracy": 0.8051281929016113,
"num_tokens": 504395432.0,
"step": 5690
},
{
"entropy": 0.9794921875,
"epoch": 0.7790610264470716,
"grad_norm": 0.8408100317787087,
"learning_rate": 3.147456671833169e-06,
"loss": 0.6753,
"mean_token_accuracy": 0.816343829035759,
"num_tokens": 505275664.0,
"step": 5700
},
{
"entropy": 0.9787109375,
"epoch": 0.7804278001776805,
"grad_norm": 0.7670324351472672,
"learning_rate": 3.1439340566436527e-06,
"loss": 0.6903,
"mean_token_accuracy": 0.8152349263429641,
"num_tokens": 506168945.0,
"step": 5710
},
{
"entropy": 1.019140625,
"epoch": 0.7817945739082894,
"grad_norm": 0.7991261431628542,
"learning_rate": 3.140411441454136e-06,
"loss": 0.7055,
"mean_token_accuracy": 0.8099988222122192,
"num_tokens": 507052908.0,
"step": 5720
},
{
"entropy": 0.9775390625,
"epoch": 0.7831613476388983,
"grad_norm": 0.7739066478911489,
"learning_rate": 3.136888826264619e-06,
"loss": 0.7005,
"mean_token_accuracy": 0.8106383144855499,
"num_tokens": 507938577.0,
"step": 5730
},
{
"entropy": 0.9630859375,
"epoch": 0.7845281213695072,
"grad_norm": 0.8184687845345091,
"learning_rate": 3.1333662110751025e-06,
"loss": 0.6968,
"mean_token_accuracy": 0.8107469886541366,
"num_tokens": 508807062.0,
"step": 5740
},
{
"entropy": 0.9708984375,
"epoch": 0.7858948951001162,
"grad_norm": 0.7483277220839287,
"learning_rate": 3.1298435958855856e-06,
"loss": 0.6623,
"mean_token_accuracy": 0.8190160781145096,
"num_tokens": 509646843.0,
"step": 5750
},
{
"entropy": 0.995703125,
"epoch": 0.7872616688307251,
"grad_norm": 0.749231010492405,
"learning_rate": 3.126320980696069e-06,
"loss": 0.7195,
"mean_token_accuracy": 0.8057454884052276,
"num_tokens": 510507535.0,
"step": 5760
},
{
"entropy": 0.9958984375,
"epoch": 0.788628442561334,
"grad_norm": 0.7834758882979491,
"learning_rate": 3.1227983655065523e-06,
"loss": 0.6823,
"mean_token_accuracy": 0.8128932535648346,
"num_tokens": 511376176.0,
"step": 5770
},
{
"entropy": 0.946875,
"epoch": 0.7899952162919429,
"grad_norm": 0.7063425516036448,
"learning_rate": 3.119275750317036e-06,
"loss": 0.6632,
"mean_token_accuracy": 0.819043293595314,
"num_tokens": 512257868.0,
"step": 5780
},
{
"entropy": 1.024609375,
"epoch": 0.7913619900225518,
"grad_norm": 0.7437037268134126,
"learning_rate": 3.115753135127519e-06,
"loss": 0.7134,
"mean_token_accuracy": 0.8066636949777604,
"num_tokens": 513149614.0,
"step": 5790
},
{
"entropy": 0.9794921875,
"epoch": 0.7927287637531607,
"grad_norm": 0.7653433754638388,
"learning_rate": 3.112230519938002e-06,
"loss": 0.6884,
"mean_token_accuracy": 0.8147072225809098,
"num_tokens": 514060257.0,
"step": 5800
},
{
"entropy": 0.9927734375,
"epoch": 0.7940955374837696,
"grad_norm": 0.8671929906018734,
"learning_rate": 3.108707904748485e-06,
"loss": 0.6667,
"mean_token_accuracy": 0.8177355140447616,
"num_tokens": 514887972.0,
"step": 5810
},
{
"entropy": 1.0013671875,
"epoch": 0.7954623112143785,
"grad_norm": 0.6892537421692612,
"learning_rate": 3.105185289558969e-06,
"loss": 0.693,
"mean_token_accuracy": 0.8141306638717651,
"num_tokens": 515812256.0,
"step": 5820
},
{
"entropy": 0.9970703125,
"epoch": 0.7968290849449874,
"grad_norm": 0.7294834301533548,
"learning_rate": 3.1016626743694523e-06,
"loss": 0.6919,
"mean_token_accuracy": 0.8117187738418579,
"num_tokens": 516697618.0,
"step": 5830
},
{
"entropy": 0.93515625,
"epoch": 0.7981958586755963,
"grad_norm": 0.7284416036245421,
"learning_rate": 3.0981400591799354e-06,
"loss": 0.6542,
"mean_token_accuracy": 0.8195067435503006,
"num_tokens": 517646655.0,
"step": 5840
},
{
"entropy": 0.991796875,
"epoch": 0.7995626324062052,
"grad_norm": 0.6796177673569014,
"learning_rate": 3.0946174439904185e-06,
"loss": 0.6595,
"mean_token_accuracy": 0.8192854344844818,
"num_tokens": 518550530.0,
"step": 5850
},
{
"entropy": 0.9810546875,
"epoch": 0.8009294061368141,
"grad_norm": 0.7583420821091867,
"learning_rate": 3.0910948288009016e-06,
"loss": 0.6604,
"mean_token_accuracy": 0.818585854768753,
"num_tokens": 519408238.0,
"step": 5860
},
{
"entropy": 0.9759765625,
"epoch": 0.802296179867423,
"grad_norm": 0.8146297378586211,
"learning_rate": 3.0875722136113856e-06,
"loss": 0.7022,
"mean_token_accuracy": 0.8115995168685913,
"num_tokens": 520283433.0,
"step": 5870
},
{
"entropy": 1.0103515625,
"epoch": 0.8036629535980319,
"grad_norm": 0.7941075614415739,
"learning_rate": 3.0840495984218687e-06,
"loss": 0.7082,
"mean_token_accuracy": 0.8108259946107864,
"num_tokens": 521184354.0,
"step": 5880
},
{
"entropy": 1.0076171875,
"epoch": 0.8050297273286408,
"grad_norm": 0.6937162034842981,
"learning_rate": 3.080526983232352e-06,
"loss": 0.6932,
"mean_token_accuracy": 0.8115349233150482,
"num_tokens": 522052078.0,
"step": 5890
},
{
"entropy": 0.9837890625,
"epoch": 0.8063965010592496,
"grad_norm": 0.7677956604343856,
"learning_rate": 3.0770043680428354e-06,
"loss": 0.7048,
"mean_token_accuracy": 0.8099212169647216,
"num_tokens": 522948969.0,
"step": 5900
},
{
"entropy": 0.96796875,
"epoch": 0.8077632747898585,
"grad_norm": 0.7551092029334426,
"learning_rate": 3.0734817528533185e-06,
"loss": 0.6535,
"mean_token_accuracy": 0.8231275618076325,
"num_tokens": 523817470.0,
"step": 5910
},
{
"entropy": 1.024609375,
"epoch": 0.8091300485204674,
"grad_norm": 0.826773181239107,
"learning_rate": 3.0699591376638017e-06,
"loss": 0.7225,
"mean_token_accuracy": 0.8064969033002853,
"num_tokens": 524695050.0,
"step": 5920
},
{
"entropy": 0.9669921875,
"epoch": 0.8104968222510763,
"grad_norm": 0.7654854406250229,
"learning_rate": 3.066436522474285e-06,
"loss": 0.6882,
"mean_token_accuracy": 0.814867427945137,
"num_tokens": 525574755.0,
"step": 5930
},
{
"entropy": 0.946484375,
"epoch": 0.8118635959816852,
"grad_norm": 0.6927196476450459,
"learning_rate": 3.0629139072847688e-06,
"loss": 0.6477,
"mean_token_accuracy": 0.8211459249258042,
"num_tokens": 526509631.0,
"step": 5940
},
{
"entropy": 1.0052734375,
"epoch": 0.8132303697122941,
"grad_norm": 0.852936815120187,
"learning_rate": 3.059391292095252e-06,
"loss": 0.7169,
"mean_token_accuracy": 0.805453184247017,
"num_tokens": 527382381.0,
"step": 5950
},
{
"entropy": 0.987890625,
"epoch": 0.814597143442903,
"grad_norm": 0.7855040846683587,
"learning_rate": 3.055868676905735e-06,
"loss": 0.7085,
"mean_token_accuracy": 0.8099881798028946,
"num_tokens": 528322905.0,
"step": 5960
},
{
"entropy": 1.0052734375,
"epoch": 0.8159639171735119,
"grad_norm": 0.8302877000590402,
"learning_rate": 3.052346061716218e-06,
"loss": 0.7176,
"mean_token_accuracy": 0.8072595655918121,
"num_tokens": 529151743.0,
"step": 5970
},
{
"entropy": 0.97421875,
"epoch": 0.8173306909041208,
"grad_norm": 0.7524334737724573,
"learning_rate": 3.048823446526702e-06,
"loss": 0.6828,
"mean_token_accuracy": 0.8140360325574875,
"num_tokens": 530042986.0,
"step": 5980
},
{
"entropy": 1.0150390625,
"epoch": 0.8186974646347297,
"grad_norm": 0.7923235143157089,
"learning_rate": 3.0453008313371852e-06,
"loss": 0.7102,
"mean_token_accuracy": 0.8105103611946106,
"num_tokens": 530980320.0,
"step": 5990
},
{
"entropy": 0.98828125,
"epoch": 0.8200642383653386,
"grad_norm": 0.8121505727076137,
"learning_rate": 3.0417782161476683e-06,
"loss": 0.6958,
"mean_token_accuracy": 0.8118116587400437,
"num_tokens": 531854285.0,
"step": 6000
},
{
"entropy": 1.0484375,
"epoch": 0.8214310120959475,
"grad_norm": 0.8452899481813168,
"learning_rate": 3.0382556009581515e-06,
"loss": 0.7151,
"mean_token_accuracy": 0.8074131280183792,
"num_tokens": 532722975.0,
"step": 6010
},
{
"entropy": 0.961328125,
"epoch": 0.8227977858265564,
"grad_norm": 0.8112135214674147,
"learning_rate": 3.0347329857686346e-06,
"loss": 0.6459,
"mean_token_accuracy": 0.8213377326726914,
"num_tokens": 533582773.0,
"step": 6020
},
{
"entropy": 0.969921875,
"epoch": 0.8241645595571653,
"grad_norm": 0.8092670091509929,
"learning_rate": 3.031210370579118e-06,
"loss": 0.6424,
"mean_token_accuracy": 0.8251520484685898,
"num_tokens": 534476179.0,
"step": 6030
},
{
"entropy": 0.987890625,
"epoch": 0.8255313332877742,
"grad_norm": 0.73198142158311,
"learning_rate": 3.0276877553896017e-06,
"loss": 0.6819,
"mean_token_accuracy": 0.8112536936998367,
"num_tokens": 535353086.0,
"step": 6040
},
{
"entropy": 0.9947265625,
"epoch": 0.8268981070183831,
"grad_norm": 0.846654487710214,
"learning_rate": 3.024165140200085e-06,
"loss": 0.67,
"mean_token_accuracy": 0.8167933404445649,
"num_tokens": 536231879.0,
"step": 6050
},
{
"entropy": 0.992578125,
"epoch": 0.828264880748992,
"grad_norm": 0.7789799898596376,
"learning_rate": 3.020642525010568e-06,
"loss": 0.6937,
"mean_token_accuracy": 0.8138351976871491,
"num_tokens": 537112252.0,
"step": 6060
},
{
"entropy": 0.98125,
"epoch": 0.8296316544796009,
"grad_norm": 0.8117915631749149,
"learning_rate": 3.0171199098210515e-06,
"loss": 0.6699,
"mean_token_accuracy": 0.8175642877817154,
"num_tokens": 537972593.0,
"step": 6070
},
{
"entropy": 1.0044921875,
"epoch": 0.8309984282102097,
"grad_norm": 0.7443134017630383,
"learning_rate": 3.0135972946315346e-06,
"loss": 0.6999,
"mean_token_accuracy": 0.8132076978683471,
"num_tokens": 538871586.0,
"step": 6080
},
{
"entropy": 1.0015625,
"epoch": 0.8323652019408186,
"grad_norm": 0.7552305885624925,
"learning_rate": 3.010074679442018e-06,
"loss": 0.7122,
"mean_token_accuracy": 0.8088525474071503,
"num_tokens": 539732436.0,
"step": 6090
},
{
"entropy": 1.0015625,
"epoch": 0.8337319756714276,
"grad_norm": 0.8505526247160717,
"learning_rate": 3.0065520642525013e-06,
"loss": 0.7197,
"mean_token_accuracy": 0.809402683377266,
"num_tokens": 540600028.0,
"step": 6100
},
{
"entropy": 1.0345703125,
"epoch": 0.8350987494020365,
"grad_norm": 0.7921422715992686,
"learning_rate": 3.003029449062985e-06,
"loss": 0.7048,
"mean_token_accuracy": 0.810922110080719,
"num_tokens": 541502670.0,
"step": 6110
},
{
"entropy": 1.0060546875,
"epoch": 0.8364655231326454,
"grad_norm": 0.8260219190041819,
"learning_rate": 2.999506833873468e-06,
"loss": 0.7222,
"mean_token_accuracy": 0.8063399940729141,
"num_tokens": 542340098.0,
"step": 6120
},
{
"entropy": 1.020703125,
"epoch": 0.8378322968632543,
"grad_norm": 0.711922371836695,
"learning_rate": 2.995984218683951e-06,
"loss": 0.7382,
"mean_token_accuracy": 0.8011481434106826,
"num_tokens": 543218409.0,
"step": 6130
},
{
"entropy": 0.971484375,
"epoch": 0.8391990705938632,
"grad_norm": 0.7122760346664355,
"learning_rate": 2.992461603494434e-06,
"loss": 0.6611,
"mean_token_accuracy": 0.8175148397684098,
"num_tokens": 544126052.0,
"step": 6140
},
{
"entropy": 0.982421875,
"epoch": 0.8405658443244721,
"grad_norm": 0.8253283972865734,
"learning_rate": 2.988938988304918e-06,
"loss": 0.6984,
"mean_token_accuracy": 0.8114474058151245,
"num_tokens": 545029025.0,
"step": 6150
},
{
"entropy": 0.971484375,
"epoch": 0.841932618055081,
"grad_norm": 0.7507788832093284,
"learning_rate": 2.9854163731154013e-06,
"loss": 0.6696,
"mean_token_accuracy": 0.8191411525011063,
"num_tokens": 545905476.0,
"step": 6160
},
{
"entropy": 0.9888671875,
"epoch": 0.8432993917856899,
"grad_norm": 0.718187609498899,
"learning_rate": 2.9818937579258844e-06,
"loss": 0.6778,
"mean_token_accuracy": 0.8157603293657303,
"num_tokens": 546749480.0,
"step": 6170
},
{
"entropy": 0.977734375,
"epoch": 0.8446661655162988,
"grad_norm": 0.8045005349251302,
"learning_rate": 2.9783711427363675e-06,
"loss": 0.657,
"mean_token_accuracy": 0.8204722493886948,
"num_tokens": 547635672.0,
"step": 6180
},
{
"entropy": 1.00625,
"epoch": 0.8460329392469077,
"grad_norm": 0.8906623108959373,
"learning_rate": 2.9748485275468507e-06,
"loss": 0.6944,
"mean_token_accuracy": 0.8123522400856018,
"num_tokens": 548511563.0,
"step": 6190
},
{
"entropy": 0.983203125,
"epoch": 0.8473997129775166,
"grad_norm": 0.7440098942117939,
"learning_rate": 2.9713259123573346e-06,
"loss": 0.6959,
"mean_token_accuracy": 0.8109301745891571,
"num_tokens": 549358293.0,
"step": 6200
},
{
"entropy": 0.97578125,
"epoch": 0.8487664867081255,
"grad_norm": 0.7288422482725624,
"learning_rate": 2.9678032971678177e-06,
"loss": 0.7029,
"mean_token_accuracy": 0.8103498667478561,
"num_tokens": 550199472.0,
"step": 6210
},
{
"entropy": 0.996484375,
"epoch": 0.8501332604387344,
"grad_norm": 0.754012252593423,
"learning_rate": 2.964280681978301e-06,
"loss": 0.7444,
"mean_token_accuracy": 0.8021257877349853,
"num_tokens": 551109412.0,
"step": 6220
},
{
"entropy": 0.96015625,
"epoch": 0.8515000341693433,
"grad_norm": 0.8945893110777644,
"learning_rate": 2.960758066788784e-06,
"loss": 0.6746,
"mean_token_accuracy": 0.8158706903457642,
"num_tokens": 551947950.0,
"step": 6230
},
{
"entropy": 0.9978515625,
"epoch": 0.8528668078999522,
"grad_norm": 0.7671372543081093,
"learning_rate": 2.9572354515992675e-06,
"loss": 0.7086,
"mean_token_accuracy": 0.8058241993188858,
"num_tokens": 552800468.0,
"step": 6240
},
{
"entropy": 0.937109375,
"epoch": 0.8542335816305611,
"grad_norm": 0.7522979771717737,
"learning_rate": 2.9537128364097507e-06,
"loss": 0.643,
"mean_token_accuracy": 0.823775252699852,
"num_tokens": 553683283.0,
"step": 6250
},
{
"entropy": 0.97890625,
"epoch": 0.85560035536117,
"grad_norm": 0.6872994441874493,
"learning_rate": 2.9501902212202342e-06,
"loss": 0.6863,
"mean_token_accuracy": 0.8148276388645173,
"num_tokens": 554603674.0,
"step": 6260
},
{
"entropy": 1.0025390625,
"epoch": 0.8569671290917789,
"grad_norm": 0.7257641382803074,
"learning_rate": 2.9466676060307178e-06,
"loss": 0.7153,
"mean_token_accuracy": 0.8071107029914856,
"num_tokens": 555498911.0,
"step": 6270
},
{
"entropy": 1.0154296875,
"epoch": 0.8583339028223878,
"grad_norm": 0.7093668268912044,
"learning_rate": 2.943144990841201e-06,
"loss": 0.7056,
"mean_token_accuracy": 0.8111616283655166,
"num_tokens": 556398231.0,
"step": 6280
},
{
"entropy": 0.982421875,
"epoch": 0.8597006765529966,
"grad_norm": 0.810553802294539,
"learning_rate": 2.939622375651684e-06,
"loss": 0.6629,
"mean_token_accuracy": 0.8196200847625732,
"num_tokens": 557289325.0,
"step": 6290
},
{
"entropy": 1.0080078125,
"epoch": 0.8610674502836055,
"grad_norm": 0.8641637379392659,
"learning_rate": 2.936099760462167e-06,
"loss": 0.7332,
"mean_token_accuracy": 0.8044625729322433,
"num_tokens": 558215409.0,
"step": 6300
},
{
"entropy": 0.9740234375,
"epoch": 0.8624342240142144,
"grad_norm": 0.7569531500707845,
"learning_rate": 2.932577145272651e-06,
"loss": 0.6273,
"mean_token_accuracy": 0.8275295585393906,
"num_tokens": 559078072.0,
"step": 6310
},
{
"entropy": 1.0103515625,
"epoch": 0.8638009977448233,
"grad_norm": 0.7445434283336702,
"learning_rate": 2.9290545300831342e-06,
"loss": 0.7033,
"mean_token_accuracy": 0.8115330815315247,
"num_tokens": 559985199.0,
"step": 6320
},
{
"entropy": 1.01953125,
"epoch": 0.8651677714754322,
"grad_norm": 0.8058948656295389,
"learning_rate": 2.9255319148936174e-06,
"loss": 0.7134,
"mean_token_accuracy": 0.8087217301130295,
"num_tokens": 560873774.0,
"step": 6330
},
{
"entropy": 0.960546875,
"epoch": 0.8665345452060411,
"grad_norm": 0.7847022714597921,
"learning_rate": 2.9220092997041005e-06,
"loss": 0.6337,
"mean_token_accuracy": 0.8265541791915894,
"num_tokens": 561773112.0,
"step": 6340
},
{
"entropy": 1.0037109375,
"epoch": 0.86790131893665,
"grad_norm": 0.7464790040382401,
"learning_rate": 2.9184866845145836e-06,
"loss": 0.7023,
"mean_token_accuracy": 0.813403582572937,
"num_tokens": 562659591.0,
"step": 6350
},
{
"entropy": 1.0033203125,
"epoch": 0.8692680926672589,
"grad_norm": 0.7826730123348531,
"learning_rate": 2.9149640693250667e-06,
"loss": 0.7022,
"mean_token_accuracy": 0.8081123381853104,
"num_tokens": 563573747.0,
"step": 6360
},
{
"entropy": 0.9798828125,
"epoch": 0.8706348663978678,
"grad_norm": 0.800189734319328,
"learning_rate": 2.9114414541355507e-06,
"loss": 0.6972,
"mean_token_accuracy": 0.8115439116954803,
"num_tokens": 564444810.0,
"step": 6370
},
{
"entropy": 1.0048828125,
"epoch": 0.8720016401284767,
"grad_norm": 0.6984134483836663,
"learning_rate": 2.907918838946034e-06,
"loss": 0.6868,
"mean_token_accuracy": 0.8152987480163574,
"num_tokens": 565334063.0,
"step": 6380
},
{
"entropy": 0.976953125,
"epoch": 0.8733684138590856,
"grad_norm": 0.8223083691076293,
"learning_rate": 2.904396223756517e-06,
"loss": 0.6884,
"mean_token_accuracy": 0.8125899881124496,
"num_tokens": 566217603.0,
"step": 6390
},
{
"entropy": 1.0037109375,
"epoch": 0.8747351875896945,
"grad_norm": 0.8149920677395663,
"learning_rate": 2.9008736085670005e-06,
"loss": 0.7001,
"mean_token_accuracy": 0.8116345524787902,
"num_tokens": 567103733.0,
"step": 6400
},
{
"entropy": 0.9623046875,
"epoch": 0.8761019613203034,
"grad_norm": 0.8661334588392458,
"learning_rate": 2.8973509933774836e-06,
"loss": 0.6631,
"mean_token_accuracy": 0.8186804950237274,
"num_tokens": 567958707.0,
"step": 6410
},
{
"entropy": 0.990625,
"epoch": 0.8774687350509123,
"grad_norm": 0.729180911798163,
"learning_rate": 2.893828378187967e-06,
"loss": 0.7228,
"mean_token_accuracy": 0.8069371014833451,
"num_tokens": 568857102.0,
"step": 6420
},
{
"entropy": 0.9546875,
"epoch": 0.8788355087815212,
"grad_norm": 0.7666054588026547,
"learning_rate": 2.8903057629984503e-06,
"loss": 0.6615,
"mean_token_accuracy": 0.818892952799797,
"num_tokens": 569724676.0,
"step": 6430
},
{
"entropy": 0.970703125,
"epoch": 0.8802022825121301,
"grad_norm": 0.8257952113018665,
"learning_rate": 2.886783147808934e-06,
"loss": 0.6592,
"mean_token_accuracy": 0.8188104838132858,
"num_tokens": 570626696.0,
"step": 6440
},
{
"entropy": 1.033203125,
"epoch": 0.8815690562427391,
"grad_norm": 0.7588848969558735,
"learning_rate": 2.883260532619417e-06,
"loss": 0.7276,
"mean_token_accuracy": 0.8040929704904556,
"num_tokens": 571490412.0,
"step": 6450
},
{
"entropy": 0.9857421875,
"epoch": 0.882935829973348,
"grad_norm": 0.8697181661437575,
"learning_rate": 2.8797379174299e-06,
"loss": 0.6759,
"mean_token_accuracy": 0.8146070003509521,
"num_tokens": 572332785.0,
"step": 6460
},
{
"entropy": 0.9505859375,
"epoch": 0.8843026037039569,
"grad_norm": 0.7290433167873468,
"learning_rate": 2.876215302240383e-06,
"loss": 0.6372,
"mean_token_accuracy": 0.8263335436582565,
"num_tokens": 573204987.0,
"step": 6470
},
{
"entropy": 1.01796875,
"epoch": 0.8856693774345658,
"grad_norm": 0.8581411238850338,
"learning_rate": 2.872692687050867e-06,
"loss": 0.7241,
"mean_token_accuracy": 0.8070144712924957,
"num_tokens": 574117752.0,
"step": 6480
},
{
"entropy": 1.00078125,
"epoch": 0.8870361511651746,
"grad_norm": 0.719086862718769,
"learning_rate": 2.8691700718613503e-06,
"loss": 0.6799,
"mean_token_accuracy": 0.8134649664163589,
"num_tokens": 575003850.0,
"step": 6490
},
{
"entropy": 0.9716796875,
"epoch": 0.8884029248957835,
"grad_norm": 0.831102004365895,
"learning_rate": 2.8656474566718334e-06,
"loss": 0.6726,
"mean_token_accuracy": 0.815576845407486,
"num_tokens": 575921233.0,
"step": 6500
},
{
"entropy": 0.9546875,
"epoch": 0.8897696986263924,
"grad_norm": 0.7024923510928729,
"learning_rate": 2.8621248414823165e-06,
"loss": 0.6418,
"mean_token_accuracy": 0.8227149248123169,
"num_tokens": 576814818.0,
"step": 6510
},
{
"entropy": 1.018359375,
"epoch": 0.8911364723570013,
"grad_norm": 0.8668200886502808,
"learning_rate": 2.8586022262927997e-06,
"loss": 0.6919,
"mean_token_accuracy": 0.8118092447519303,
"num_tokens": 577627640.0,
"step": 6520
},
{
"entropy": 1.0392578125,
"epoch": 0.8925032460876102,
"grad_norm": 0.712166028181192,
"learning_rate": 2.8550796111032836e-06,
"loss": 0.736,
"mean_token_accuracy": 0.8025434374809265,
"num_tokens": 578521956.0,
"step": 6530
},
{
"entropy": 0.9845703125,
"epoch": 0.8938700198182191,
"grad_norm": 0.761516961803004,
"learning_rate": 2.8515569959137668e-06,
"loss": 0.6907,
"mean_token_accuracy": 0.8136275738477707,
"num_tokens": 579395782.0,
"step": 6540
},
{
"entropy": 0.993359375,
"epoch": 0.895236793548828,
"grad_norm": 0.7259370362996954,
"learning_rate": 2.84803438072425e-06,
"loss": 0.6766,
"mean_token_accuracy": 0.8175433605909348,
"num_tokens": 580272157.0,
"step": 6550
},
{
"entropy": 0.9966796875,
"epoch": 0.8966035672794369,
"grad_norm": 0.7814845957726428,
"learning_rate": 2.844511765534733e-06,
"loss": 0.6801,
"mean_token_accuracy": 0.816193026304245,
"num_tokens": 581146537.0,
"step": 6560
},
{
"entropy": 0.9513671875,
"epoch": 0.8979703410100458,
"grad_norm": 0.7120578845631831,
"learning_rate": 2.8409891503452166e-06,
"loss": 0.6796,
"mean_token_accuracy": 0.8156316012144089,
"num_tokens": 582048135.0,
"step": 6570
},
{
"entropy": 0.9962890625,
"epoch": 0.8993371147406547,
"grad_norm": 0.8155753907849608,
"learning_rate": 2.8374665351556997e-06,
"loss": 0.6993,
"mean_token_accuracy": 0.8115780144929886,
"num_tokens": 582904287.0,
"step": 6580
},
{
"entropy": 1.0025390625,
"epoch": 0.9007038884712636,
"grad_norm": 0.7446932110019452,
"learning_rate": 2.8339439199661832e-06,
"loss": 0.706,
"mean_token_accuracy": 0.8107634842395782,
"num_tokens": 583826125.0,
"step": 6590
},
{
"entropy": 1.0111328125,
"epoch": 0.9020706622018725,
"grad_norm": 0.7556388440613876,
"learning_rate": 2.8304213047766663e-06,
"loss": 0.72,
"mean_token_accuracy": 0.8054689705371857,
"num_tokens": 584674987.0,
"step": 6600
},
{
"entropy": 0.970703125,
"epoch": 0.9034374359324814,
"grad_norm": 0.7549025361100669,
"learning_rate": 2.82689868958715e-06,
"loss": 0.6694,
"mean_token_accuracy": 0.8177745014429092,
"num_tokens": 585567786.0,
"step": 6610
},
{
"entropy": 0.986328125,
"epoch": 0.9048042096630903,
"grad_norm": 0.8093381627428929,
"learning_rate": 2.823376074397633e-06,
"loss": 0.7187,
"mean_token_accuracy": 0.8091011345386505,
"num_tokens": 586446088.0,
"step": 6620
},
{
"entropy": 0.9908203125,
"epoch": 0.9061709833936992,
"grad_norm": 0.8265596484315829,
"learning_rate": 2.819853459208116e-06,
"loss": 0.6959,
"mean_token_accuracy": 0.8124132186174393,
"num_tokens": 587291210.0,
"step": 6630
},
{
"entropy": 1.026171875,
"epoch": 0.9075377571243081,
"grad_norm": 0.7553046008940949,
"learning_rate": 2.8163308440186e-06,
"loss": 0.7275,
"mean_token_accuracy": 0.805382776260376,
"num_tokens": 588175587.0,
"step": 6640
},
{
"entropy": 1.0009765625,
"epoch": 0.908904530854917,
"grad_norm": 1.0266596460256423,
"learning_rate": 2.8128082288290832e-06,
"loss": 0.6641,
"mean_token_accuracy": 0.8196311891078949,
"num_tokens": 589051406.0,
"step": 6650
},
{
"entropy": 1.01015625,
"epoch": 0.9102713045855259,
"grad_norm": 0.8272075147146867,
"learning_rate": 2.8092856136395664e-06,
"loss": 0.6882,
"mean_token_accuracy": 0.8117015928030014,
"num_tokens": 589942110.0,
"step": 6660
},
{
"entropy": 1.0080078125,
"epoch": 0.9116380783161347,
"grad_norm": 0.7922706995522857,
"learning_rate": 2.8057629984500495e-06,
"loss": 0.7049,
"mean_token_accuracy": 0.8096364349126816,
"num_tokens": 590877907.0,
"step": 6670
},
{
"entropy": 0.9861328125,
"epoch": 0.9130048520467436,
"grad_norm": 0.7226065264390196,
"learning_rate": 2.8022403832605326e-06,
"loss": 0.6924,
"mean_token_accuracy": 0.8133171498775482,
"num_tokens": 591773456.0,
"step": 6680
},
{
"entropy": 0.979296875,
"epoch": 0.9143716257773525,
"grad_norm": 0.7278628054688634,
"learning_rate": 2.7987177680710157e-06,
"loss": 0.655,
"mean_token_accuracy": 0.8209361851215362,
"num_tokens": 592686986.0,
"step": 6690
},
{
"entropy": 1.0037109375,
"epoch": 0.9157383995079614,
"grad_norm": 0.8691293830769832,
"learning_rate": 2.7951951528814997e-06,
"loss": 0.6949,
"mean_token_accuracy": 0.8114201694726944,
"num_tokens": 593589156.0,
"step": 6700
},
{
"entropy": 0.9845703125,
"epoch": 0.9171051732385703,
"grad_norm": 0.7579761575072722,
"learning_rate": 2.791672537691983e-06,
"loss": 0.7047,
"mean_token_accuracy": 0.8100018560886383,
"num_tokens": 594522006.0,
"step": 6710
},
{
"entropy": 0.956640625,
"epoch": 0.9184719469691792,
"grad_norm": 0.7888342305915186,
"learning_rate": 2.788149922502466e-06,
"loss": 0.6887,
"mean_token_accuracy": 0.8140259444713592,
"num_tokens": 595400912.0,
"step": 6720
},
{
"entropy": 0.9802734375,
"epoch": 0.9198387206997881,
"grad_norm": 0.9397363486862614,
"learning_rate": 2.784627307312949e-06,
"loss": 0.6791,
"mean_token_accuracy": 0.8163236826658249,
"num_tokens": 596309300.0,
"step": 6730
},
{
"entropy": 0.9650390625,
"epoch": 0.921205494430397,
"grad_norm": 0.748759443689676,
"learning_rate": 2.7811046921234326e-06,
"loss": 0.6434,
"mean_token_accuracy": 0.8228376299142838,
"num_tokens": 597210330.0,
"step": 6740
},
{
"entropy": 1.0125,
"epoch": 0.9225722681610059,
"grad_norm": 0.814889783807825,
"learning_rate": 2.777582076933916e-06,
"loss": 0.7073,
"mean_token_accuracy": 0.8105467259883881,
"num_tokens": 598081617.0,
"step": 6750
},
{
"entropy": 0.9857421875,
"epoch": 0.9239390418916148,
"grad_norm": 0.7092489000373994,
"learning_rate": 2.7740594617443993e-06,
"loss": 0.6906,
"mean_token_accuracy": 0.8125540643930436,
"num_tokens": 598983716.0,
"step": 6760
},
{
"entropy": 0.966015625,
"epoch": 0.9253058156222237,
"grad_norm": 0.7994826142852147,
"learning_rate": 2.770536846554883e-06,
"loss": 0.6639,
"mean_token_accuracy": 0.8187306612730026,
"num_tokens": 599918368.0,
"step": 6770
},
{
"entropy": 0.9958984375,
"epoch": 0.9266725893528326,
"grad_norm": 0.7657198836912561,
"learning_rate": 2.767014231365366e-06,
"loss": 0.7071,
"mean_token_accuracy": 0.8106543004512787,
"num_tokens": 600794870.0,
"step": 6780
},
{
"entropy": 1.003515625,
"epoch": 0.9280393630834415,
"grad_norm": 0.8014893201004748,
"learning_rate": 2.763491616175849e-06,
"loss": 0.7276,
"mean_token_accuracy": 0.8077147841453552,
"num_tokens": 601726884.0,
"step": 6790
},
{
"entropy": 0.9697265625,
"epoch": 0.9294061368140505,
"grad_norm": 0.8281105662469749,
"learning_rate": 2.759969000986332e-06,
"loss": 0.6933,
"mean_token_accuracy": 0.8125525563955307,
"num_tokens": 602627133.0,
"step": 6800
},
{
"entropy": 0.9517578125,
"epoch": 0.9307729105446594,
"grad_norm": 0.7706325110737335,
"learning_rate": 2.756446385796816e-06,
"loss": 0.6588,
"mean_token_accuracy": 0.8193507373332978,
"num_tokens": 603501852.0,
"step": 6810
},
{
"entropy": 0.969921875,
"epoch": 0.9321396842752683,
"grad_norm": 0.7614510645483153,
"learning_rate": 2.7529237706072993e-06,
"loss": 0.6538,
"mean_token_accuracy": 0.8214022547006607,
"num_tokens": 604404913.0,
"step": 6820
},
{
"entropy": 0.9857421875,
"epoch": 0.9335064580058772,
"grad_norm": 0.7313873025752333,
"learning_rate": 2.7494011554177824e-06,
"loss": 0.7008,
"mean_token_accuracy": 0.8113573282957077,
"num_tokens": 605285242.0,
"step": 6830
},
{
"entropy": 1.0203125,
"epoch": 0.9348732317364861,
"grad_norm": 0.8076554834963564,
"learning_rate": 2.7458785402282656e-06,
"loss": 0.7194,
"mean_token_accuracy": 0.8047234088182449,
"num_tokens": 606154610.0,
"step": 6840
},
{
"entropy": 1.01953125,
"epoch": 0.936240005467095,
"grad_norm": 0.7905098799520331,
"learning_rate": 2.7423559250387487e-06,
"loss": 0.7227,
"mean_token_accuracy": 0.8066396147012711,
"num_tokens": 607046840.0,
"step": 6850
},
{
"entropy": 0.9787109375,
"epoch": 0.9376067791977039,
"grad_norm": 0.7235728458057872,
"learning_rate": 2.7388333098492326e-06,
"loss": 0.6873,
"mean_token_accuracy": 0.814787495136261,
"num_tokens": 607882778.0,
"step": 6860
},
{
"entropy": 0.9630859375,
"epoch": 0.9389735529283127,
"grad_norm": 0.7780060349732635,
"learning_rate": 2.7353106946597158e-06,
"loss": 0.701,
"mean_token_accuracy": 0.8112201422452927,
"num_tokens": 608824241.0,
"step": 6870
},
{
"entropy": 1.0169921875,
"epoch": 0.9403403266589216,
"grad_norm": 0.8661180082322555,
"learning_rate": 2.731788079470199e-06,
"loss": 0.704,
"mean_token_accuracy": 0.8086902290582657,
"num_tokens": 609676651.0,
"step": 6880
},
{
"entropy": 1.002734375,
"epoch": 0.9417071003895305,
"grad_norm": 0.7599666236602641,
"learning_rate": 2.728265464280682e-06,
"loss": 0.7194,
"mean_token_accuracy": 0.8078528851270675,
"num_tokens": 610543423.0,
"step": 6890
},
{
"entropy": 0.994921875,
"epoch": 0.9430738741201394,
"grad_norm": 0.714232198300505,
"learning_rate": 2.7247428490911656e-06,
"loss": 0.676,
"mean_token_accuracy": 0.816844317317009,
"num_tokens": 611399783.0,
"step": 6900
},
{
"entropy": 1.0048828125,
"epoch": 0.9444406478507483,
"grad_norm": 0.8212573289384504,
"learning_rate": 2.7212202339016487e-06,
"loss": 0.7029,
"mean_token_accuracy": 0.8105417877435684,
"num_tokens": 612298608.0,
"step": 6910
},
{
"entropy": 0.9712890625,
"epoch": 0.9458074215813572,
"grad_norm": 0.7123591398465731,
"learning_rate": 2.7176976187121322e-06,
"loss": 0.7165,
"mean_token_accuracy": 0.8076278567314148,
"num_tokens": 613248091.0,
"step": 6920
},
{
"entropy": 0.99765625,
"epoch": 0.9471741953119661,
"grad_norm": 0.738729877354312,
"learning_rate": 2.7141750035226154e-06,
"loss": 0.6918,
"mean_token_accuracy": 0.8125291556119919,
"num_tokens": 614177532.0,
"step": 6930
},
{
"entropy": 1.001953125,
"epoch": 0.948540969042575,
"grad_norm": 0.8219282975174166,
"learning_rate": 2.710652388333099e-06,
"loss": 0.705,
"mean_token_accuracy": 0.8110216349363327,
"num_tokens": 615083286.0,
"step": 6940
},
{
"entropy": 1.0146484375,
"epoch": 0.9499077427731839,
"grad_norm": 0.7204507414878372,
"learning_rate": 2.707129773143582e-06,
"loss": 0.7523,
"mean_token_accuracy": 0.7980942964553833,
"num_tokens": 616002316.0,
"step": 6950
},
{
"entropy": 0.9609375,
"epoch": 0.9512745165037928,
"grad_norm": 0.7984899067458674,
"learning_rate": 2.703607157954065e-06,
"loss": 0.6763,
"mean_token_accuracy": 0.8143421471118927,
"num_tokens": 616898102.0,
"step": 6960
},
{
"entropy": 1.02421875,
"epoch": 0.9526412902344017,
"grad_norm": 0.7984612228783475,
"learning_rate": 2.7000845427645483e-06,
"loss": 0.7301,
"mean_token_accuracy": 0.8048859983682632,
"num_tokens": 617749598.0,
"step": 6970
},
{
"entropy": 0.957421875,
"epoch": 0.9540080639650106,
"grad_norm": 0.8206633389427582,
"learning_rate": 2.6965619275750322e-06,
"loss": 0.6607,
"mean_token_accuracy": 0.8191677689552307,
"num_tokens": 618614766.0,
"step": 6980
},
{
"entropy": 1.0078125,
"epoch": 0.9553748376956195,
"grad_norm": 0.8741868657250778,
"learning_rate": 2.6930393123855154e-06,
"loss": 0.7307,
"mean_token_accuracy": 0.8070410043001175,
"num_tokens": 619527250.0,
"step": 6990
},
{
"entropy": 1.0310546875,
"epoch": 0.9567416114262284,
"grad_norm": 0.7918825093643986,
"learning_rate": 2.6895166971959985e-06,
"loss": 0.7394,
"mean_token_accuracy": 0.8026743561029435,
"num_tokens": 620389290.0,
"step": 7000
},
{
"entropy": 0.9642578125,
"epoch": 0.9581083851568373,
"grad_norm": 0.7574361952729899,
"learning_rate": 2.6859940820064816e-06,
"loss": 0.6674,
"mean_token_accuracy": 0.820996567606926,
"num_tokens": 621260557.0,
"step": 7010
},
{
"entropy": 1.0318359375,
"epoch": 0.9594751588874462,
"grad_norm": 0.9064986963621744,
"learning_rate": 2.6824714668169647e-06,
"loss": 0.7003,
"mean_token_accuracy": 0.8106149971485138,
"num_tokens": 622132062.0,
"step": 7020
},
{
"entropy": 0.9900390625,
"epoch": 0.9608419326180551,
"grad_norm": 0.7376823028418074,
"learning_rate": 2.6789488516274487e-06,
"loss": 0.7368,
"mean_token_accuracy": 0.8055850207805634,
"num_tokens": 623083505.0,
"step": 7030
},
{
"entropy": 1.0015625,
"epoch": 0.962208706348664,
"grad_norm": 0.8296961905324908,
"learning_rate": 2.675426236437932e-06,
"loss": 0.6639,
"mean_token_accuracy": 0.8191018015146255,
"num_tokens": 624013009.0,
"step": 7040
},
{
"entropy": 0.9625,
"epoch": 0.9635754800792729,
"grad_norm": 0.7256359492165542,
"learning_rate": 2.671903621248415e-06,
"loss": 0.6908,
"mean_token_accuracy": 0.8108186930418014,
"num_tokens": 624866319.0,
"step": 7050
},
{
"entropy": 0.967578125,
"epoch": 0.9649422538098817,
"grad_norm": 0.8168319545057097,
"learning_rate": 2.668381006058898e-06,
"loss": 0.7009,
"mean_token_accuracy": 0.8095557421445847,
"num_tokens": 625727325.0,
"step": 7060
},
{
"entropy": 1.0154296875,
"epoch": 0.9663090275404906,
"grad_norm": 0.7566500126709597,
"learning_rate": 2.6648583908693816e-06,
"loss": 0.6821,
"mean_token_accuracy": 0.8141679257154465,
"num_tokens": 626605530.0,
"step": 7070
},
{
"entropy": 1.0283203125,
"epoch": 0.9676758012710995,
"grad_norm": 0.8604088608540846,
"learning_rate": 2.6613357756798648e-06,
"loss": 0.7528,
"mean_token_accuracy": 0.8017343699932098,
"num_tokens": 627487707.0,
"step": 7080
},
{
"entropy": 1.0095703125,
"epoch": 0.9690425750017084,
"grad_norm": 0.7522134770594657,
"learning_rate": 2.6578131604903483e-06,
"loss": 0.7073,
"mean_token_accuracy": 0.8102524966001511,
"num_tokens": 628336825.0,
"step": 7090
},
{
"entropy": 0.9716796875,
"epoch": 0.9704093487323173,
"grad_norm": 0.751015149354093,
"learning_rate": 2.6542905453008314e-06,
"loss": 0.6682,
"mean_token_accuracy": 0.819304171204567,
"num_tokens": 629226993.0,
"step": 7100
},
{
"entropy": 1.003515625,
"epoch": 0.9717761224629262,
"grad_norm": 0.7657010290714672,
"learning_rate": 2.650767930111315e-06,
"loss": 0.707,
"mean_token_accuracy": 0.8103487193584442,
"num_tokens": 630127171.0,
"step": 7110
},
{
"entropy": 0.995703125,
"epoch": 0.9731428961935351,
"grad_norm": 0.7097652866852381,
"learning_rate": 2.647245314921798e-06,
"loss": 0.6787,
"mean_token_accuracy": 0.8154985010623932,
"num_tokens": 631049821.0,
"step": 7120
},
{
"entropy": 0.9828125,
"epoch": 0.974509669924144,
"grad_norm": 0.6911164626511238,
"learning_rate": 2.6437226997322812e-06,
"loss": 0.682,
"mean_token_accuracy": 0.8170451998710633,
"num_tokens": 631978925.0,
"step": 7130
},
{
"entropy": 1.00234375,
"epoch": 0.9758764436547529,
"grad_norm": 0.7756043494116309,
"learning_rate": 2.640200084542765e-06,
"loss": 0.6691,
"mean_token_accuracy": 0.8160231649875641,
"num_tokens": 632874005.0,
"step": 7140
},
{
"entropy": 1.0021484375,
"epoch": 0.9772432173853619,
"grad_norm": 0.8647673919726134,
"learning_rate": 2.6366774693532483e-06,
"loss": 0.7073,
"mean_token_accuracy": 0.8092184156179428,
"num_tokens": 633703539.0,
"step": 7150
},
{
"entropy": 1.0287109375,
"epoch": 0.9786099911159708,
"grad_norm": 0.8374942475665127,
"learning_rate": 2.6331548541637314e-06,
"loss": 0.7223,
"mean_token_accuracy": 0.8053885847330093,
"num_tokens": 634587023.0,
"step": 7160
},
{
"entropy": 0.97265625,
"epoch": 0.9799767648465797,
"grad_norm": 0.686848269729656,
"learning_rate": 2.6296322389742146e-06,
"loss": 0.6678,
"mean_token_accuracy": 0.8188909679651261,
"num_tokens": 635488376.0,
"step": 7170
},
{
"entropy": 0.9537109375,
"epoch": 0.9813435385771886,
"grad_norm": 0.7828571786019828,
"learning_rate": 2.6261096237846977e-06,
"loss": 0.6718,
"mean_token_accuracy": 0.8154725968837738,
"num_tokens": 636356843.0,
"step": 7180
},
{
"entropy": 0.9841796875,
"epoch": 0.9827103123077975,
"grad_norm": 0.8054547118287042,
"learning_rate": 2.622587008595181e-06,
"loss": 0.7011,
"mean_token_accuracy": 0.8118202149868011,
"num_tokens": 637214936.0,
"step": 7190
},
{
"entropy": 1.003125,
"epoch": 0.9840770860384064,
"grad_norm": 0.8246621744928478,
"learning_rate": 2.6190643934056648e-06,
"loss": 0.7059,
"mean_token_accuracy": 0.8082175046205521,
"num_tokens": 638111430.0,
"step": 7200
},
{
"entropy": 1.0177734375,
"epoch": 0.9854438597690153,
"grad_norm": 0.7465948387613044,
"learning_rate": 2.615541778216148e-06,
"loss": 0.7272,
"mean_token_accuracy": 0.8070796489715576,
"num_tokens": 638975613.0,
"step": 7210
},
{
"entropy": 0.9869140625,
"epoch": 0.9868106334996242,
"grad_norm": 0.7810965948010813,
"learning_rate": 2.612019163026631e-06,
"loss": 0.674,
"mean_token_accuracy": 0.8157333403825759,
"num_tokens": 639844309.0,
"step": 7220
},
{
"entropy": 1.0263671875,
"epoch": 0.9881774072302331,
"grad_norm": 0.8325643735321173,
"learning_rate": 2.6084965478371146e-06,
"loss": 0.742,
"mean_token_accuracy": 0.8020849257707596,
"num_tokens": 640732612.0,
"step": 7230
},
{
"entropy": 0.94453125,
"epoch": 0.989544180960842,
"grad_norm": 0.741564549839553,
"learning_rate": 2.6049739326475977e-06,
"loss": 0.6749,
"mean_token_accuracy": 0.8179847300052643,
"num_tokens": 641622965.0,
"step": 7240
},
{
"entropy": 1.0212890625,
"epoch": 0.9909109546914509,
"grad_norm": 0.8745564378740422,
"learning_rate": 2.6014513174580812e-06,
"loss": 0.6805,
"mean_token_accuracy": 0.8149160206317901,
"num_tokens": 642493831.0,
"step": 7250
},
{
"entropy": 0.976953125,
"epoch": 0.9922777284220597,
"grad_norm": 0.7817413338408525,
"learning_rate": 2.5979287022685644e-06,
"loss": 0.6762,
"mean_token_accuracy": 0.8179818391799927,
"num_tokens": 643358509.0,
"step": 7260
},
{
"entropy": 0.9529296875,
"epoch": 0.9936445021526686,
"grad_norm": 0.7665345169811426,
"learning_rate": 2.594406087079048e-06,
"loss": 0.6508,
"mean_token_accuracy": 0.8217165142297744,
"num_tokens": 644234904.0,
"step": 7270
},
{
"entropy": 1.0224609375,
"epoch": 0.9950112758832775,
"grad_norm": 0.7453273730967398,
"learning_rate": 2.590883471889531e-06,
"loss": 0.732,
"mean_token_accuracy": 0.8033026039600373,
"num_tokens": 645107938.0,
"step": 7280
},
{
"entropy": 0.9837890625,
"epoch": 0.9963780496138864,
"grad_norm": 0.7277663036305626,
"learning_rate": 2.587360856700014e-06,
"loss": 0.682,
"mean_token_accuracy": 0.814745482802391,
"num_tokens": 645973503.0,
"step": 7290
},
{
"entropy": 0.9904296875,
"epoch": 0.9977448233444953,
"grad_norm": 0.7968977823903666,
"learning_rate": 2.5838382415104973e-06,
"loss": 0.6927,
"mean_token_accuracy": 0.8138006657361985,
"num_tokens": 646849773.0,
"step": 7300
},
{
"entropy": 1.0119140625,
"epoch": 0.9991115970751042,
"grad_norm": 0.7514092472427303,
"learning_rate": 2.5803156263209813e-06,
"loss": 0.7123,
"mean_token_accuracy": 0.8097608327865601,
"num_tokens": 647714153.0,
"step": 7310
},
{
"entropy": 1.0291940789473684,
"epoch": 1.0004100321191827,
"grad_norm": 0.8311405247902348,
"learning_rate": 2.5767930111314644e-06,
"loss": 0.7634,
"mean_token_accuracy": 0.8008053992923937,
"num_tokens": 648591763.0,
"step": 7320
},
{
"entropy": 1.0029296875,
"epoch": 1.0017768058497916,
"grad_norm": 0.7603682726265689,
"learning_rate": 2.5732703959419475e-06,
"loss": 0.6848,
"mean_token_accuracy": 0.8138793170452118,
"num_tokens": 649446852.0,
"step": 7330
},
{
"entropy": 0.9548828125,
"epoch": 1.0031435795804005,
"grad_norm": 0.6724456099545856,
"learning_rate": 2.5697477807524306e-06,
"loss": 0.6766,
"mean_token_accuracy": 0.8157681912183762,
"num_tokens": 650395487.0,
"step": 7340
},
{
"entropy": 0.93125,
"epoch": 1.0045103533110094,
"grad_norm": 0.7909402056273646,
"learning_rate": 2.5662251655629138e-06,
"loss": 0.6574,
"mean_token_accuracy": 0.820235139131546,
"num_tokens": 651283020.0,
"step": 7350
},
{
"entropy": 0.97734375,
"epoch": 1.0058771270416182,
"grad_norm": 0.8211858452514211,
"learning_rate": 2.5627025503733977e-06,
"loss": 0.6834,
"mean_token_accuracy": 0.8122252076864243,
"num_tokens": 652168918.0,
"step": 7360
},
{
"entropy": 0.9935546875,
"epoch": 1.0072439007722271,
"grad_norm": 0.8006045904850417,
"learning_rate": 2.559179935183881e-06,
"loss": 0.7164,
"mean_token_accuracy": 0.8089279919862747,
"num_tokens": 653038499.0,
"step": 7370
},
{
"entropy": 0.9767578125,
"epoch": 1.008610674502836,
"grad_norm": 0.8633169757376867,
"learning_rate": 2.555657319994364e-06,
"loss": 0.6828,
"mean_token_accuracy": 0.8141887754201889,
"num_tokens": 653925430.0,
"step": 7380
},
{
"entropy": 0.9603515625,
"epoch": 1.009977448233445,
"grad_norm": 0.7339645928247789,
"learning_rate": 2.552134704804847e-06,
"loss": 0.6737,
"mean_token_accuracy": 0.8170844435691833,
"num_tokens": 654833925.0,
"step": 7390
},
{
"entropy": 1.008984375,
"epoch": 1.0113442219640538,
"grad_norm": 0.8127316279243577,
"learning_rate": 2.5486120896153306e-06,
"loss": 0.7166,
"mean_token_accuracy": 0.8074491947889328,
"num_tokens": 655696986.0,
"step": 7400
},
{
"entropy": 0.9736328125,
"epoch": 1.0127109956946627,
"grad_norm": 0.8359915237671496,
"learning_rate": 2.5450894744258138e-06,
"loss": 0.6589,
"mean_token_accuracy": 0.8211406648159028,
"num_tokens": 656585655.0,
"step": 7410
},
{
"entropy": 0.935546875,
"epoch": 1.0140777694252716,
"grad_norm": 0.7979147377981276,
"learning_rate": 2.5415668592362973e-06,
"loss": 0.6323,
"mean_token_accuracy": 0.8269622176885605,
"num_tokens": 657442572.0,
"step": 7420
},
{
"entropy": 0.9435546875,
"epoch": 1.0154445431558805,
"grad_norm": 0.7426172465865568,
"learning_rate": 2.5380442440467804e-06,
"loss": 0.6464,
"mean_token_accuracy": 0.8234199106693267,
"num_tokens": 658344385.0,
"step": 7430
},
{
"entropy": 1.0134765625,
"epoch": 1.0168113168864894,
"grad_norm": 0.9217442894495533,
"learning_rate": 2.534521628857264e-06,
"loss": 0.7243,
"mean_token_accuracy": 0.8046896904706955,
"num_tokens": 659214294.0,
"step": 7440
},
{
"entropy": 0.9564453125,
"epoch": 1.0181780906170983,
"grad_norm": 0.6919460523437397,
"learning_rate": 2.530999013667747e-06,
"loss": 0.677,
"mean_token_accuracy": 0.8162459403276443,
"num_tokens": 660072515.0,
"step": 7450
},
{
"entropy": 1.044140625,
"epoch": 1.0195448643477072,
"grad_norm": 0.9468953745020394,
"learning_rate": 2.5274763984782302e-06,
"loss": 0.7371,
"mean_token_accuracy": 0.804921418428421,
"num_tokens": 660959998.0,
"step": 7460
},
{
"entropy": 0.934375,
"epoch": 1.020911638078316,
"grad_norm": 0.7967819404874233,
"learning_rate": 2.5239537832887138e-06,
"loss": 0.652,
"mean_token_accuracy": 0.822745481133461,
"num_tokens": 661875700.0,
"step": 7470
},
{
"entropy": 0.97734375,
"epoch": 1.022278411808925,
"grad_norm": 0.7719169554346371,
"learning_rate": 2.5204311680991973e-06,
"loss": 0.6603,
"mean_token_accuracy": 0.8189722031354905,
"num_tokens": 662774852.0,
"step": 7480
},
{
"entropy": 0.9712890625,
"epoch": 1.0236451855395339,
"grad_norm": 0.7343573321865391,
"learning_rate": 2.5169085529096804e-06,
"loss": 0.6336,
"mean_token_accuracy": 0.8257413148880005,
"num_tokens": 663594608.0,
"step": 7490
},
{
"entropy": 0.9814453125,
"epoch": 1.0250119592701428,
"grad_norm": 0.8734286970910649,
"learning_rate": 2.5133859377201636e-06,
"loss": 0.6992,
"mean_token_accuracy": 0.8103983730077744,
"num_tokens": 664461143.0,
"step": 7500
},
{
"entropy": 1.0029296875,
"epoch": 1.0263787330007517,
"grad_norm": 0.759928104123206,
"learning_rate": 2.5098633225306467e-06,
"loss": 0.6969,
"mean_token_accuracy": 0.8103911191225052,
"num_tokens": 665355418.0,
"step": 7510
},
{
"entropy": 0.947265625,
"epoch": 1.0277455067313606,
"grad_norm": 0.7738520174072514,
"learning_rate": 2.50634070734113e-06,
"loss": 0.6598,
"mean_token_accuracy": 0.8174798607826232,
"num_tokens": 666259994.0,
"step": 7520
},
{
"entropy": 0.9671875,
"epoch": 1.0291122804619695,
"grad_norm": 0.7169895958824931,
"learning_rate": 2.502818092151614e-06,
"loss": 0.685,
"mean_token_accuracy": 0.8163608372211456,
"num_tokens": 667126480.0,
"step": 7530
},
{
"entropy": 0.95234375,
"epoch": 1.0304790541925783,
"grad_norm": 0.8156663480109132,
"learning_rate": 2.499295476962097e-06,
"loss": 0.6733,
"mean_token_accuracy": 0.815731143951416,
"num_tokens": 668015004.0,
"step": 7540
},
{
"entropy": 0.9642578125,
"epoch": 1.0318458279231872,
"grad_norm": 0.7447537475668428,
"learning_rate": 2.49577286177258e-06,
"loss": 0.676,
"mean_token_accuracy": 0.8168157696723938,
"num_tokens": 668927065.0,
"step": 7550
},
{
"entropy": 0.9564453125,
"epoch": 1.0332126016537961,
"grad_norm": 0.7661475731025528,
"learning_rate": 2.492250246583063e-06,
"loss": 0.7135,
"mean_token_accuracy": 0.8099840223789215,
"num_tokens": 669799875.0,
"step": 7560
},
{
"entropy": 0.9736328125,
"epoch": 1.034579375384405,
"grad_norm": 0.801697911377627,
"learning_rate": 2.4887276313935467e-06,
"loss": 0.6364,
"mean_token_accuracy": 0.824469143152237,
"num_tokens": 670673474.0,
"step": 7570
},
{
"entropy": 0.95703125,
"epoch": 1.035946149115014,
"grad_norm": 0.7108425334735763,
"learning_rate": 2.48520501620403e-06,
"loss": 0.6551,
"mean_token_accuracy": 0.8217111229896545,
"num_tokens": 671584047.0,
"step": 7580
},
{
"entropy": 0.9349609375,
"epoch": 1.0373129228456228,
"grad_norm": 0.7638923542580708,
"learning_rate": 2.4816824010145134e-06,
"loss": 0.6264,
"mean_token_accuracy": 0.824051558971405,
"num_tokens": 672451040.0,
"step": 7590
},
{
"entropy": 0.966015625,
"epoch": 1.0386796965762317,
"grad_norm": 0.886946011893497,
"learning_rate": 2.478159785824997e-06,
"loss": 0.6881,
"mean_token_accuracy": 0.814638489484787,
"num_tokens": 673394919.0,
"step": 7600
},
{
"entropy": 0.9541015625,
"epoch": 1.0400464703068406,
"grad_norm": 0.7967557972448058,
"learning_rate": 2.47463717063548e-06,
"loss": 0.6533,
"mean_token_accuracy": 0.8219844698905945,
"num_tokens": 674297118.0,
"step": 7610
},
{
"entropy": 0.99921875,
"epoch": 1.0414132440374495,
"grad_norm": 0.7212351098951874,
"learning_rate": 2.4711145554459636e-06,
"loss": 0.7077,
"mean_token_accuracy": 0.8098427653312683,
"num_tokens": 675208936.0,
"step": 7620
},
{
"entropy": 0.9814453125,
"epoch": 1.0427800177680584,
"grad_norm": 0.7785067168182617,
"learning_rate": 2.4675919402564467e-06,
"loss": 0.6837,
"mean_token_accuracy": 0.8144280701875687,
"num_tokens": 676084432.0,
"step": 7630
},
{
"entropy": 0.98046875,
"epoch": 1.0441467914986673,
"grad_norm": 0.7314973814246908,
"learning_rate": 2.46406932506693e-06,
"loss": 0.6753,
"mean_token_accuracy": 0.8155555456876755,
"num_tokens": 677002994.0,
"step": 7640
},
{
"entropy": 0.96015625,
"epoch": 1.0455135652292762,
"grad_norm": 0.8288639025433413,
"learning_rate": 2.4605467098774134e-06,
"loss": 0.6945,
"mean_token_accuracy": 0.812854814529419,
"num_tokens": 677933461.0,
"step": 7650
},
{
"entropy": 0.9376953125,
"epoch": 1.0468803389598853,
"grad_norm": 0.711794251040153,
"learning_rate": 2.4570240946878965e-06,
"loss": 0.6875,
"mean_token_accuracy": 0.8137983679771423,
"num_tokens": 678803569.0,
"step": 7660
},
{
"entropy": 0.9578125,
"epoch": 1.0482471126904942,
"grad_norm": 0.7456744215159891,
"learning_rate": 2.4535014794983796e-06,
"loss": 0.69,
"mean_token_accuracy": 0.8136455297470093,
"num_tokens": 679680533.0,
"step": 7670
},
{
"entropy": 0.9408203125,
"epoch": 1.049613886421103,
"grad_norm": 0.7816445205079304,
"learning_rate": 2.449978864308863e-06,
"loss": 0.6443,
"mean_token_accuracy": 0.8234420239925384,
"num_tokens": 680523782.0,
"step": 7680
},
{
"entropy": 0.97265625,
"epoch": 1.050980660151712,
"grad_norm": 0.7798952208443849,
"learning_rate": 2.4464562491193463e-06,
"loss": 0.6741,
"mean_token_accuracy": 0.8169470697641372,
"num_tokens": 681377055.0,
"step": 7690
},
{
"entropy": 0.9814453125,
"epoch": 1.0523474338823209,
"grad_norm": 0.9090805136952367,
"learning_rate": 2.4429336339298294e-06,
"loss": 0.694,
"mean_token_accuracy": 0.8134478002786636,
"num_tokens": 682286031.0,
"step": 7700
},
{
"entropy": 0.9130859375,
"epoch": 1.0537142076129298,
"grad_norm": 0.7851247719339721,
"learning_rate": 2.439411018740313e-06,
"loss": 0.626,
"mean_token_accuracy": 0.8241081058979034,
"num_tokens": 683137335.0,
"step": 7710
},
{
"entropy": 0.9548828125,
"epoch": 1.0550809813435387,
"grad_norm": 0.6863060057158384,
"learning_rate": 2.435888403550796e-06,
"loss": 0.6239,
"mean_token_accuracy": 0.826748576760292,
"num_tokens": 683993688.0,
"step": 7720
},
{
"entropy": 0.97421875,
"epoch": 1.0564477550741476,
"grad_norm": 0.7830454395075448,
"learning_rate": 2.4323657883612797e-06,
"loss": 0.7148,
"mean_token_accuracy": 0.8086974918842316,
"num_tokens": 684887017.0,
"step": 7730
},
{
"entropy": 0.9875,
"epoch": 1.0578145288047565,
"grad_norm": 0.797837402470115,
"learning_rate": 2.4288431731717628e-06,
"loss": 0.676,
"mean_token_accuracy": 0.8142283320426941,
"num_tokens": 685740410.0,
"step": 7740
},
{
"entropy": 0.965234375,
"epoch": 1.0591813025353654,
"grad_norm": 0.8804483874922113,
"learning_rate": 2.4253205579822463e-06,
"loss": 0.6743,
"mean_token_accuracy": 0.8154237508773804,
"num_tokens": 686594655.0,
"step": 7750
},
{
"entropy": 0.928125,
"epoch": 1.0605480762659742,
"grad_norm": 0.6913285934940032,
"learning_rate": 2.4217979427927294e-06,
"loss": 0.6523,
"mean_token_accuracy": 0.8215391755104064,
"num_tokens": 687452138.0,
"step": 7760
},
{
"entropy": 0.9181640625,
"epoch": 1.0619148499965831,
"grad_norm": 0.7248651688659661,
"learning_rate": 2.418275327603213e-06,
"loss": 0.6812,
"mean_token_accuracy": 0.8171747356653214,
"num_tokens": 688323704.0,
"step": 7770
},
{
"entropy": 0.9546875,
"epoch": 1.063281623727192,
"grad_norm": 0.8828367787505916,
"learning_rate": 2.414752712413696e-06,
"loss": 0.6601,
"mean_token_accuracy": 0.8189290136098861,
"num_tokens": 689208797.0,
"step": 7780
},
{
"entropy": 0.9419921875,
"epoch": 1.064648397457801,
"grad_norm": 0.7638312055304237,
"learning_rate": 2.4112300972241797e-06,
"loss": 0.6333,
"mean_token_accuracy": 0.8253736853599548,
"num_tokens": 690082488.0,
"step": 7790
},
{
"entropy": 0.9564453125,
"epoch": 1.0660151711884098,
"grad_norm": 0.7269243641445289,
"learning_rate": 2.4077074820346628e-06,
"loss": 0.7055,
"mean_token_accuracy": 0.8109775811433793,
"num_tokens": 690974253.0,
"step": 7800
},
{
"entropy": 0.9865234375,
"epoch": 1.0673819449190187,
"grad_norm": 0.8116388964779555,
"learning_rate": 2.404184866845146e-06,
"loss": 0.686,
"mean_token_accuracy": 0.8127865374088288,
"num_tokens": 691884193.0,
"step": 7810
},
{
"entropy": 0.972265625,
"epoch": 1.0687487186496276,
"grad_norm": 0.7726694009206726,
"learning_rate": 2.4006622516556295e-06,
"loss": 0.6912,
"mean_token_accuracy": 0.8137483835220337,
"num_tokens": 692774904.0,
"step": 7820
},
{
"entropy": 0.9890625,
"epoch": 1.0701154923802365,
"grad_norm": 0.7838820576854507,
"learning_rate": 2.3971396364661126e-06,
"loss": 0.6955,
"mean_token_accuracy": 0.8127223193645478,
"num_tokens": 693642265.0,
"step": 7830
},
{
"entropy": 0.975,
"epoch": 1.0714822661108454,
"grad_norm": 0.7868694382646395,
"learning_rate": 2.393617021276596e-06,
"loss": 0.6708,
"mean_token_accuracy": 0.8181735187768936,
"num_tokens": 694520615.0,
"step": 7840
},
{
"entropy": 0.9669921875,
"epoch": 1.0728490398414543,
"grad_norm": 0.726104450725915,
"learning_rate": 2.3900944060870793e-06,
"loss": 0.6642,
"mean_token_accuracy": 0.8191784292459487,
"num_tokens": 695379137.0,
"step": 7850
},
{
"entropy": 0.9826171875,
"epoch": 1.0742158135720632,
"grad_norm": 0.733825060570749,
"learning_rate": 2.3865717908975624e-06,
"loss": 0.6715,
"mean_token_accuracy": 0.8154716521501542,
"num_tokens": 696271994.0,
"step": 7860
},
{
"entropy": 0.9814453125,
"epoch": 1.075582587302672,
"grad_norm": 0.8583524661670753,
"learning_rate": 2.383049175708046e-06,
"loss": 0.6665,
"mean_token_accuracy": 0.8184741348028183,
"num_tokens": 697156934.0,
"step": 7870
},
{
"entropy": 0.93671875,
"epoch": 1.076949361033281,
"grad_norm": 0.8881768640916475,
"learning_rate": 2.379526560518529e-06,
"loss": 0.683,
"mean_token_accuracy": 0.8144725084304809,
"num_tokens": 698027349.0,
"step": 7880
},
{
"entropy": 0.9998046875,
"epoch": 1.0783161347638899,
"grad_norm": 0.7818799843654242,
"learning_rate": 2.376003945329012e-06,
"loss": 0.7221,
"mean_token_accuracy": 0.8056115120649338,
"num_tokens": 698930447.0,
"step": 7890
},
{
"entropy": 0.9126953125,
"epoch": 1.0796829084944988,
"grad_norm": 0.8140013089602809,
"learning_rate": 2.3724813301394957e-06,
"loss": 0.5953,
"mean_token_accuracy": 0.8338286340236664,
"num_tokens": 699791840.0,
"step": 7900
},
{
"entropy": 0.9353515625,
"epoch": 1.0810496822251077,
"grad_norm": 0.719659936782654,
"learning_rate": 2.368958714949979e-06,
"loss": 0.6246,
"mean_token_accuracy": 0.8274411022663116,
"num_tokens": 700643155.0,
"step": 7910
},
{
"entropy": 0.9630859375,
"epoch": 1.0824164559557166,
"grad_norm": 0.7750958164554245,
"learning_rate": 2.3654360997604624e-06,
"loss": 0.6432,
"mean_token_accuracy": 0.8227149903774261,
"num_tokens": 701513474.0,
"step": 7920
},
{
"entropy": 0.9626953125,
"epoch": 1.0837832296863255,
"grad_norm": 0.7789064619460003,
"learning_rate": 2.3619134845709455e-06,
"loss": 0.6589,
"mean_token_accuracy": 0.8216227233409882,
"num_tokens": 702423369.0,
"step": 7930
},
{
"entropy": 0.990625,
"epoch": 1.0851500034169344,
"grad_norm": 0.7125812319603912,
"learning_rate": 2.358390869381429e-06,
"loss": 0.6619,
"mean_token_accuracy": 0.8194294333457947,
"num_tokens": 703297260.0,
"step": 7940
},
{
"entropy": 0.9765625,
"epoch": 1.0865167771475432,
"grad_norm": 0.7548163895773605,
"learning_rate": 2.354868254191912e-06,
"loss": 0.689,
"mean_token_accuracy": 0.812394094467163,
"num_tokens": 704211864.0,
"step": 7950
},
{
"entropy": 0.986328125,
"epoch": 1.0878835508781521,
"grad_norm": 0.8302223024275136,
"learning_rate": 2.3513456390023957e-06,
"loss": 0.6649,
"mean_token_accuracy": 0.8193449884653091,
"num_tokens": 705113336.0,
"step": 7960
},
{
"entropy": 0.9853515625,
"epoch": 1.089250324608761,
"grad_norm": 0.7919081515224167,
"learning_rate": 2.347823023812879e-06,
"loss": 0.6727,
"mean_token_accuracy": 0.8157610595226288,
"num_tokens": 705967220.0,
"step": 7970
},
{
"entropy": 1.008203125,
"epoch": 1.09061709833937,
"grad_norm": 0.7677055004290724,
"learning_rate": 2.3443004086233624e-06,
"loss": 0.6493,
"mean_token_accuracy": 0.8213225275278091,
"num_tokens": 706841729.0,
"step": 7980
},
{
"entropy": 0.9365234375,
"epoch": 1.0919838720699788,
"grad_norm": 0.8434223961184751,
"learning_rate": 2.3407777934338455e-06,
"loss": 0.6196,
"mean_token_accuracy": 0.8278831958770752,
"num_tokens": 707742723.0,
"step": 7990
},
{
"entropy": 0.973828125,
"epoch": 1.0933506458005877,
"grad_norm": 0.7797359259466415,
"learning_rate": 2.3372551782443286e-06,
"loss": 0.6596,
"mean_token_accuracy": 0.8211466372013092,
"num_tokens": 708621892.0,
"step": 8000
},
{
"entropy": 0.9703125,
"epoch": 1.0947174195311966,
"grad_norm": 0.7530030783226226,
"learning_rate": 2.333732563054812e-06,
"loss": 0.6731,
"mean_token_accuracy": 0.8156515240669251,
"num_tokens": 709529009.0,
"step": 8010
},
{
"entropy": 0.9697265625,
"epoch": 1.0960841932618055,
"grad_norm": 0.8437187364469441,
"learning_rate": 2.3302099478652953e-06,
"loss": 0.6645,
"mean_token_accuracy": 0.8200674712657928,
"num_tokens": 710431822.0,
"step": 8020
},
{
"entropy": 0.98515625,
"epoch": 1.0974509669924144,
"grad_norm": 0.867534955938422,
"learning_rate": 2.3266873326757784e-06,
"loss": 0.6798,
"mean_token_accuracy": 0.8175362914800643,
"num_tokens": 711282131.0,
"step": 8030
},
{
"entropy": 1.0353515625,
"epoch": 1.0988177407230233,
"grad_norm": 0.8315945621852996,
"learning_rate": 2.323164717486262e-06,
"loss": 0.7391,
"mean_token_accuracy": 0.8013202905654907,
"num_tokens": 712177392.0,
"step": 8040
},
{
"entropy": 0.9228515625,
"epoch": 1.1001845144536322,
"grad_norm": 0.7102760376936019,
"learning_rate": 2.319642102296745e-06,
"loss": 0.6532,
"mean_token_accuracy": 0.8230769604444503,
"num_tokens": 713063521.0,
"step": 8050
},
{
"entropy": 0.95859375,
"epoch": 1.101551288184241,
"grad_norm": 0.7668191729123122,
"learning_rate": 2.3161194871072287e-06,
"loss": 0.6634,
"mean_token_accuracy": 0.8183923840522767,
"num_tokens": 713932139.0,
"step": 8060
},
{
"entropy": 0.933984375,
"epoch": 1.10291806191485,
"grad_norm": 0.8031892306128253,
"learning_rate": 2.3125968719177118e-06,
"loss": 0.646,
"mean_token_accuracy": 0.8214219093322754,
"num_tokens": 714800743.0,
"step": 8070
},
{
"entropy": 0.98359375,
"epoch": 1.1042848356454589,
"grad_norm": 0.8352978483597865,
"learning_rate": 2.3090742567281953e-06,
"loss": 0.6873,
"mean_token_accuracy": 0.8135174512863159,
"num_tokens": 715716767.0,
"step": 8080
},
{
"entropy": 0.9818359375,
"epoch": 1.1056516093760678,
"grad_norm": 0.7460892680028064,
"learning_rate": 2.3055516415386785e-06,
"loss": 0.7056,
"mean_token_accuracy": 0.8113393634557724,
"num_tokens": 716586164.0,
"step": 8090
},
{
"entropy": 0.971484375,
"epoch": 1.1070183831066767,
"grad_norm": 0.9002301974043125,
"learning_rate": 2.302029026349162e-06,
"loss": 0.6789,
"mean_token_accuracy": 0.8176020890474319,
"num_tokens": 717454801.0,
"step": 8100
},
{
"entropy": 0.974609375,
"epoch": 1.1083851568372856,
"grad_norm": 0.9377311123286918,
"learning_rate": 2.298506411159645e-06,
"loss": 0.6757,
"mean_token_accuracy": 0.8172421872615814,
"num_tokens": 718358774.0,
"step": 8110
},
{
"entropy": 0.9658203125,
"epoch": 1.1097519305678945,
"grad_norm": 0.8434223263685562,
"learning_rate": 2.2949837959701287e-06,
"loss": 0.6726,
"mean_token_accuracy": 0.8152223974466324,
"num_tokens": 719306865.0,
"step": 8120
},
{
"entropy": 0.97265625,
"epoch": 1.1111187042985033,
"grad_norm": 0.7901516107591189,
"learning_rate": 2.291461180780612e-06,
"loss": 0.6763,
"mean_token_accuracy": 0.8157725840806961,
"num_tokens": 720218145.0,
"step": 8130
},
{
"entropy": 0.959765625,
"epoch": 1.1124854780291122,
"grad_norm": 0.7814739988194319,
"learning_rate": 2.287938565591095e-06,
"loss": 0.6238,
"mean_token_accuracy": 0.8263473153114319,
"num_tokens": 721112536.0,
"step": 8140
},
{
"entropy": 0.9328125,
"epoch": 1.1138522517597211,
"grad_norm": 0.6961197111801879,
"learning_rate": 2.2844159504015785e-06,
"loss": 0.6545,
"mean_token_accuracy": 0.8202046036720276,
"num_tokens": 722020751.0,
"step": 8150
},
{
"entropy": 0.968359375,
"epoch": 1.11521902549033,
"grad_norm": 0.7974970970446214,
"learning_rate": 2.2808933352120616e-06,
"loss": 0.6696,
"mean_token_accuracy": 0.8199956625699997,
"num_tokens": 722858329.0,
"step": 8160
},
{
"entropy": 0.959765625,
"epoch": 1.116585799220939,
"grad_norm": 0.7674857061023236,
"learning_rate": 2.277370720022545e-06,
"loss": 0.6604,
"mean_token_accuracy": 0.8169799625873566,
"num_tokens": 723733723.0,
"step": 8170
},
{
"entropy": 0.97265625,
"epoch": 1.1179525729515478,
"grad_norm": 0.7893937671186776,
"learning_rate": 2.2738481048330283e-06,
"loss": 0.6947,
"mean_token_accuracy": 0.8107841551303864,
"num_tokens": 724643810.0,
"step": 8180
},
{
"entropy": 0.9478515625,
"epoch": 1.1193193466821567,
"grad_norm": 0.8215111312569354,
"learning_rate": 2.2703254896435114e-06,
"loss": 0.6846,
"mean_token_accuracy": 0.8143371641635895,
"num_tokens": 725562682.0,
"step": 8190
},
{
"entropy": 0.9689453125,
"epoch": 1.1206861204127656,
"grad_norm": 0.7263203728989827,
"learning_rate": 2.266802874453995e-06,
"loss": 0.7011,
"mean_token_accuracy": 0.8116056501865387,
"num_tokens": 726462837.0,
"step": 8200
},
{
"entropy": 0.9736328125,
"epoch": 1.1220528941433745,
"grad_norm": 0.9120384495741448,
"learning_rate": 2.263280259264478e-06,
"loss": 0.677,
"mean_token_accuracy": 0.8142390966415405,
"num_tokens": 727331679.0,
"step": 8210
},
{
"entropy": 0.9244140625,
"epoch": 1.1234196678739834,
"grad_norm": 0.7220464056649812,
"learning_rate": 2.259757644074961e-06,
"loss": 0.5901,
"mean_token_accuracy": 0.8349061369895935,
"num_tokens": 728213250.0,
"step": 8220
},
{
"entropy": 0.978125,
"epoch": 1.1247864416045923,
"grad_norm": 0.7442961882724038,
"learning_rate": 2.2562350288854447e-06,
"loss": 0.685,
"mean_token_accuracy": 0.8158569544553756,
"num_tokens": 729111281.0,
"step": 8230
},
{
"entropy": 0.94921875,
"epoch": 1.1261532153352012,
"grad_norm": 0.8219899329308207,
"learning_rate": 2.252712413695928e-06,
"loss": 0.6524,
"mean_token_accuracy": 0.8203908503055573,
"num_tokens": 729959862.0,
"step": 8240
},
{
"entropy": 0.9734375,
"epoch": 1.12751998906581,
"grad_norm": 0.8302928982787369,
"learning_rate": 2.2491897985064114e-06,
"loss": 0.6986,
"mean_token_accuracy": 0.8095601618289947,
"num_tokens": 730854597.0,
"step": 8250
},
{
"entropy": 0.962109375,
"epoch": 1.128886762796419,
"grad_norm": 0.8235064167580243,
"learning_rate": 2.2456671833168945e-06,
"loss": 0.6514,
"mean_token_accuracy": 0.8211957395076752,
"num_tokens": 731702559.0,
"step": 8260
},
{
"entropy": 0.9509765625,
"epoch": 1.1302535365270279,
"grad_norm": 0.8787623516565186,
"learning_rate": 2.242144568127378e-06,
"loss": 0.7001,
"mean_token_accuracy": 0.811849445104599,
"num_tokens": 732555774.0,
"step": 8270
},
{
"entropy": 0.9904296875,
"epoch": 1.1316203102576368,
"grad_norm": 0.6762662127713559,
"learning_rate": 2.238621952937861e-06,
"loss": 0.7005,
"mean_token_accuracy": 0.8109658926725387,
"num_tokens": 733472536.0,
"step": 8280
},
{
"entropy": 0.9962890625,
"epoch": 1.1329870839882457,
"grad_norm": 0.8643300546683632,
"learning_rate": 2.2350993377483447e-06,
"loss": 0.7028,
"mean_token_accuracy": 0.8093641489744187,
"num_tokens": 734343325.0,
"step": 8290
},
{
"entropy": 0.965625,
"epoch": 1.1343538577188546,
"grad_norm": 0.7851848658020254,
"learning_rate": 2.231576722558828e-06,
"loss": 0.6811,
"mean_token_accuracy": 0.814346757531166,
"num_tokens": 735209167.0,
"step": 8300
},
{
"entropy": 0.9447265625,
"epoch": 1.1357206314494634,
"grad_norm": 0.7989145659431506,
"learning_rate": 2.2280541073693114e-06,
"loss": 0.6278,
"mean_token_accuracy": 0.8257664501667022,
"num_tokens": 736102499.0,
"step": 8310
},
{
"entropy": 1.0044921875,
"epoch": 1.1370874051800723,
"grad_norm": 0.8598750496257392,
"learning_rate": 2.2245314921797945e-06,
"loss": 0.7179,
"mean_token_accuracy": 0.8052316069602966,
"num_tokens": 736993752.0,
"step": 8320
},
{
"entropy": 0.96875,
"epoch": 1.1384541789106812,
"grad_norm": 0.7501312727896907,
"learning_rate": 2.2210088769902777e-06,
"loss": 0.6725,
"mean_token_accuracy": 0.8164238125085831,
"num_tokens": 737905549.0,
"step": 8330
},
{
"entropy": 0.8943359375,
"epoch": 1.1398209526412901,
"grad_norm": 0.7826777751637151,
"learning_rate": 2.217486261800761e-06,
"loss": 0.5843,
"mean_token_accuracy": 0.8360004186630249,
"num_tokens": 738743099.0,
"step": 8340
},
{
"entropy": 0.9814453125,
"epoch": 1.141187726371899,
"grad_norm": 0.797315621956627,
"learning_rate": 2.2139636466112443e-06,
"loss": 0.6832,
"mean_token_accuracy": 0.8157221019268036,
"num_tokens": 739666732.0,
"step": 8350
},
{
"entropy": 0.93046875,
"epoch": 1.142554500102508,
"grad_norm": 0.7873445827403641,
"learning_rate": 2.2104410314217275e-06,
"loss": 0.6321,
"mean_token_accuracy": 0.826965942978859,
"num_tokens": 740545036.0,
"step": 8360
},
{
"entropy": 0.9796875,
"epoch": 1.1439212738331168,
"grad_norm": 0.8465488039989977,
"learning_rate": 2.206918416232211e-06,
"loss": 0.684,
"mean_token_accuracy": 0.8142398655414581,
"num_tokens": 741452530.0,
"step": 8370
},
{
"entropy": 0.973046875,
"epoch": 1.1452880475637257,
"grad_norm": 0.758548796780733,
"learning_rate": 2.203395801042694e-06,
"loss": 0.7077,
"mean_token_accuracy": 0.8076848864555359,
"num_tokens": 742353736.0,
"step": 8380
},
{
"entropy": 0.9642578125,
"epoch": 1.1466548212943346,
"grad_norm": 0.8157975149634431,
"learning_rate": 2.1998731858531777e-06,
"loss": 0.6563,
"mean_token_accuracy": 0.8203108608722687,
"num_tokens": 743291796.0,
"step": 8390
},
{
"entropy": 0.9728515625,
"epoch": 1.1480215950249435,
"grad_norm": 0.8649538544933941,
"learning_rate": 2.196350570663661e-06,
"loss": 0.6302,
"mean_token_accuracy": 0.8263802021741867,
"num_tokens": 744191576.0,
"step": 8400
},
{
"entropy": 0.976171875,
"epoch": 1.1493883687555526,
"grad_norm": 0.720560414297124,
"learning_rate": 2.192827955474144e-06,
"loss": 0.6486,
"mean_token_accuracy": 0.8233198612928391,
"num_tokens": 745053472.0,
"step": 8410
},
{
"entropy": 0.9248046875,
"epoch": 1.1507551424861615,
"grad_norm": 0.7166047193389415,
"learning_rate": 2.1893053402846275e-06,
"loss": 0.6423,
"mean_token_accuracy": 0.8239462971687317,
"num_tokens": 745945754.0,
"step": 8420
},
{
"entropy": 1.0072265625,
"epoch": 1.1521219162167704,
"grad_norm": 0.8389937594756707,
"learning_rate": 2.1857827250951106e-06,
"loss": 0.7109,
"mean_token_accuracy": 0.8113535821437836,
"num_tokens": 746861785.0,
"step": 8430
},
{
"entropy": 1.0060546875,
"epoch": 1.1534886899473793,
"grad_norm": 0.8109093773938069,
"learning_rate": 2.182260109905594e-06,
"loss": 0.697,
"mean_token_accuracy": 0.8093420594930649,
"num_tokens": 747742441.0,
"step": 8440
},
{
"entropy": 0.9486328125,
"epoch": 1.1548554636779882,
"grad_norm": 0.774854243073793,
"learning_rate": 2.1787374947160777e-06,
"loss": 0.6583,
"mean_token_accuracy": 0.8213986486196518,
"num_tokens": 748609606.0,
"step": 8450
},
{
"entropy": 0.9224609375,
"epoch": 1.156222237408597,
"grad_norm": 0.672187150707633,
"learning_rate": 2.175214879526561e-06,
"loss": 0.6876,
"mean_token_accuracy": 0.8146886378526688,
"num_tokens": 749527900.0,
"step": 8460
},
{
"entropy": 0.98203125,
"epoch": 1.157589011139206,
"grad_norm": 0.8023796903946218,
"learning_rate": 2.171692264337044e-06,
"loss": 0.6717,
"mean_token_accuracy": 0.8158406257629395,
"num_tokens": 750396477.0,
"step": 8470
},
{
"entropy": 0.9611328125,
"epoch": 1.1589557848698149,
"grad_norm": 0.7807825641692069,
"learning_rate": 2.1681696491475275e-06,
"loss": 0.6666,
"mean_token_accuracy": 0.8186038047075271,
"num_tokens": 751256327.0,
"step": 8480
},
{
"entropy": 0.975390625,
"epoch": 1.1603225586004238,
"grad_norm": 0.7126333491419276,
"learning_rate": 2.1646470339580106e-06,
"loss": 0.6523,
"mean_token_accuracy": 0.821704438328743,
"num_tokens": 752202319.0,
"step": 8490
},
{
"entropy": 0.9486328125,
"epoch": 1.1616893323310327,
"grad_norm": 0.8285907582206887,
"learning_rate": 2.1611244187684937e-06,
"loss": 0.6424,
"mean_token_accuracy": 0.8250924706459045,
"num_tokens": 753076870.0,
"step": 8500
},
{
"entropy": 0.9572265625,
"epoch": 1.1630561060616416,
"grad_norm": 0.7069468664199636,
"learning_rate": 2.1576018035789773e-06,
"loss": 0.6725,
"mean_token_accuracy": 0.8177714496850967,
"num_tokens": 753992426.0,
"step": 8510
},
{
"entropy": 0.933203125,
"epoch": 1.1644228797922505,
"grad_norm": 0.8093126660723126,
"learning_rate": 2.1540791883894604e-06,
"loss": 0.6493,
"mean_token_accuracy": 0.8202078878879547,
"num_tokens": 754824603.0,
"step": 8520
},
{
"entropy": 0.978515625,
"epoch": 1.1657896535228593,
"grad_norm": 0.7621477238979495,
"learning_rate": 2.150556573199944e-06,
"loss": 0.6691,
"mean_token_accuracy": 0.8196769654750824,
"num_tokens": 755695653.0,
"step": 8530
},
{
"entropy": 0.938671875,
"epoch": 1.1671564272534682,
"grad_norm": 0.7237759962148493,
"learning_rate": 2.147033958010427e-06,
"loss": 0.6378,
"mean_token_accuracy": 0.8258242815732956,
"num_tokens": 756583392.0,
"step": 8540
},
{
"entropy": 0.97734375,
"epoch": 1.1685232009840771,
"grad_norm": 0.8371306413372314,
"learning_rate": 2.14351134282091e-06,
"loss": 0.6616,
"mean_token_accuracy": 0.8188671410083771,
"num_tokens": 757484772.0,
"step": 8550
},
{
"entropy": 0.9333984375,
"epoch": 1.169889974714686,
"grad_norm": 0.8214297492147601,
"learning_rate": 2.1399887276313937e-06,
"loss": 0.6076,
"mean_token_accuracy": 0.8313104271888733,
"num_tokens": 758398852.0,
"step": 8560
},
{
"entropy": 0.97265625,
"epoch": 1.171256748445295,
"grad_norm": 0.721318293856647,
"learning_rate": 2.136466112441877e-06,
"loss": 0.7,
"mean_token_accuracy": 0.8121876418590546,
"num_tokens": 759304325.0,
"step": 8570
},
{
"entropy": 0.95390625,
"epoch": 1.1726235221759038,
"grad_norm": 0.7858357817825518,
"learning_rate": 2.1329434972523604e-06,
"loss": 0.6412,
"mean_token_accuracy": 0.8223109185695648,
"num_tokens": 760182897.0,
"step": 8580
},
{
"entropy": 0.9763671875,
"epoch": 1.1739902959065127,
"grad_norm": 0.8022851620980066,
"learning_rate": 2.1294208820628435e-06,
"loss": 0.6619,
"mean_token_accuracy": 0.8180845558643342,
"num_tokens": 761046394.0,
"step": 8590
},
{
"entropy": 0.98984375,
"epoch": 1.1753570696371216,
"grad_norm": 0.8579508842606604,
"learning_rate": 2.125898266873327e-06,
"loss": 0.6846,
"mean_token_accuracy": 0.8130091339349746,
"num_tokens": 761983231.0,
"step": 8600
},
{
"entropy": 0.9908203125,
"epoch": 1.1767238433677305,
"grad_norm": 0.7179188017226305,
"learning_rate": 2.12237565168381e-06,
"loss": 0.6597,
"mean_token_accuracy": 0.8198227107524871,
"num_tokens": 762882518.0,
"step": 8610
},
{
"entropy": 0.966015625,
"epoch": 1.1780906170983394,
"grad_norm": 0.804858673815998,
"learning_rate": 2.1188530364942938e-06,
"loss": 0.7079,
"mean_token_accuracy": 0.8111963868141174,
"num_tokens": 763770291.0,
"step": 8620
},
{
"entropy": 0.912890625,
"epoch": 1.1794573908289483,
"grad_norm": 0.6902757043787224,
"learning_rate": 2.115330421304777e-06,
"loss": 0.6491,
"mean_token_accuracy": 0.8205517023801804,
"num_tokens": 764654419.0,
"step": 8630
},
{
"entropy": 0.941015625,
"epoch": 1.1808241645595572,
"grad_norm": 0.7490151722513108,
"learning_rate": 2.1118078061152604e-06,
"loss": 0.6605,
"mean_token_accuracy": 0.8229810237884522,
"num_tokens": 765587578.0,
"step": 8640
},
{
"entropy": 0.9333984375,
"epoch": 1.182190938290166,
"grad_norm": 0.735478146770789,
"learning_rate": 2.1082851909257435e-06,
"loss": 0.6534,
"mean_token_accuracy": 0.8201815336942673,
"num_tokens": 766486320.0,
"step": 8650
},
{
"entropy": 0.964453125,
"epoch": 1.183557712020775,
"grad_norm": 0.695097034916649,
"learning_rate": 2.1047625757362267e-06,
"loss": 0.6692,
"mean_token_accuracy": 0.8184451580047607,
"num_tokens": 767398093.0,
"step": 8660
},
{
"entropy": 0.961328125,
"epoch": 1.1849244857513839,
"grad_norm": 0.8214089955178965,
"learning_rate": 2.1012399605467102e-06,
"loss": 0.6618,
"mean_token_accuracy": 0.8175360769033432,
"num_tokens": 768249145.0,
"step": 8670
},
{
"entropy": 0.9806640625,
"epoch": 1.1862912594819928,
"grad_norm": 0.7344205510620965,
"learning_rate": 2.0977173453571933e-06,
"loss": 0.6796,
"mean_token_accuracy": 0.8167102307081222,
"num_tokens": 769095450.0,
"step": 8680
},
{
"entropy": 0.9822265625,
"epoch": 1.1876580332126017,
"grad_norm": 0.8722157816753306,
"learning_rate": 2.0941947301676765e-06,
"loss": 0.729,
"mean_token_accuracy": 0.8037726879119873,
"num_tokens": 769933084.0,
"step": 8690
},
{
"entropy": 0.9529296875,
"epoch": 1.1890248069432106,
"grad_norm": 0.7559134005470002,
"learning_rate": 2.09067211497816e-06,
"loss": 0.6775,
"mean_token_accuracy": 0.8154547184705734,
"num_tokens": 770828053.0,
"step": 8700
},
{
"entropy": 0.97890625,
"epoch": 1.1903915806738194,
"grad_norm": 0.8172020372571642,
"learning_rate": 2.087149499788643e-06,
"loss": 0.6617,
"mean_token_accuracy": 0.8166135132312775,
"num_tokens": 771702147.0,
"step": 8710
},
{
"entropy": 0.976953125,
"epoch": 1.1917583544044283,
"grad_norm": 0.9059910449462063,
"learning_rate": 2.0836268845991263e-06,
"loss": 0.6525,
"mean_token_accuracy": 0.8214358896017074,
"num_tokens": 772586595.0,
"step": 8720
},
{
"entropy": 0.969921875,
"epoch": 1.1931251281350372,
"grad_norm": 0.8793790463744366,
"learning_rate": 2.08010426940961e-06,
"loss": 0.6835,
"mean_token_accuracy": 0.8155566066503525,
"num_tokens": 773492144.0,
"step": 8730
},
{
"entropy": 0.9767578125,
"epoch": 1.1944919018656461,
"grad_norm": 0.79678227980196,
"learning_rate": 2.076581654220093e-06,
"loss": 0.6814,
"mean_token_accuracy": 0.8166280955076217,
"num_tokens": 774384486.0,
"step": 8740
},
{
"entropy": 0.92734375,
"epoch": 1.195858675596255,
"grad_norm": 0.9117226698577896,
"learning_rate": 2.0730590390305765e-06,
"loss": 0.6402,
"mean_token_accuracy": 0.8254548668861389,
"num_tokens": 775257458.0,
"step": 8750
},
{
"entropy": 0.9541015625,
"epoch": 1.197225449326864,
"grad_norm": 0.8628930579857561,
"learning_rate": 2.0695364238410596e-06,
"loss": 0.6784,
"mean_token_accuracy": 0.8146516442298889,
"num_tokens": 776188791.0,
"step": 8760
},
{
"entropy": 0.98125,
"epoch": 1.1985922230574728,
"grad_norm": 0.728610236496586,
"learning_rate": 2.066013808651543e-06,
"loss": 0.6766,
"mean_token_accuracy": 0.8165136456489563,
"num_tokens": 777067934.0,
"step": 8770
},
{
"entropy": 0.976171875,
"epoch": 1.1999589967880817,
"grad_norm": 0.8740499412572409,
"learning_rate": 2.0624911934620263e-06,
"loss": 0.6979,
"mean_token_accuracy": 0.8138171792030334,
"num_tokens": 777918469.0,
"step": 8780
},
{
"entropy": 0.9919921875,
"epoch": 1.2013257705186906,
"grad_norm": 0.7610939324382054,
"learning_rate": 2.05896857827251e-06,
"loss": 0.7236,
"mean_token_accuracy": 0.8068852484226227,
"num_tokens": 778840876.0,
"step": 8790
},
{
"entropy": 0.95703125,
"epoch": 1.2026925442492995,
"grad_norm": 0.7907807416861908,
"learning_rate": 2.055445963082993e-06,
"loss": 0.6497,
"mean_token_accuracy": 0.8224770784378052,
"num_tokens": 779722325.0,
"step": 8800
},
{
"entropy": 0.9259765625,
"epoch": 1.2040593179799084,
"grad_norm": 0.8212679533449498,
"learning_rate": 2.0519233478934765e-06,
"loss": 0.6532,
"mean_token_accuracy": 0.8211388468742371,
"num_tokens": 780628547.0,
"step": 8810
},
{
"entropy": 0.9853515625,
"epoch": 1.2054260917105173,
"grad_norm": 0.8074463519521047,
"learning_rate": 2.0484007327039596e-06,
"loss": 0.6806,
"mean_token_accuracy": 0.8149957120418548,
"num_tokens": 781503881.0,
"step": 8820
},
{
"entropy": 0.9763671875,
"epoch": 1.2067928654411262,
"grad_norm": 0.836526471029694,
"learning_rate": 2.0448781175144427e-06,
"loss": 0.708,
"mean_token_accuracy": 0.8108351320028305,
"num_tokens": 782375550.0,
"step": 8830
},
{
"entropy": 0.9529296875,
"epoch": 1.208159639171735,
"grad_norm": 0.774658200197617,
"learning_rate": 2.0413555023249263e-06,
"loss": 0.6779,
"mean_token_accuracy": 0.8165445327758789,
"num_tokens": 783264796.0,
"step": 8840
},
{
"entropy": 0.9615234375,
"epoch": 1.209526412902344,
"grad_norm": 0.8068916138132963,
"learning_rate": 2.0378328871354094e-06,
"loss": 0.6758,
"mean_token_accuracy": 0.8167554408311843,
"num_tokens": 784182648.0,
"step": 8850
},
{
"entropy": 0.9626953125,
"epoch": 1.2108931866329529,
"grad_norm": 0.7496684620332369,
"learning_rate": 2.034310271945893e-06,
"loss": 0.6701,
"mean_token_accuracy": 0.8185106724500656,
"num_tokens": 785110665.0,
"step": 8860
},
{
"entropy": 0.948046875,
"epoch": 1.2122599603635618,
"grad_norm": 0.766888785388382,
"learning_rate": 2.030787656756376e-06,
"loss": 0.6826,
"mean_token_accuracy": 0.813462033867836,
"num_tokens": 785985140.0,
"step": 8870
},
{
"entropy": 0.935546875,
"epoch": 1.2136267340941707,
"grad_norm": 0.7398200029292056,
"learning_rate": 2.027265041566859e-06,
"loss": 0.6642,
"mean_token_accuracy": 0.8193085104227066,
"num_tokens": 786868691.0,
"step": 8880
},
{
"entropy": 0.946875,
"epoch": 1.2149935078247796,
"grad_norm": 0.7635226626251108,
"learning_rate": 2.0237424263773427e-06,
"loss": 0.6603,
"mean_token_accuracy": 0.8200589835643768,
"num_tokens": 787720812.0,
"step": 8890
},
{
"entropy": 0.93203125,
"epoch": 1.2163602815553884,
"grad_norm": 0.7303815082206011,
"learning_rate": 2.020219811187826e-06,
"loss": 0.6531,
"mean_token_accuracy": 0.8197902649641037,
"num_tokens": 788632265.0,
"step": 8900
},
{
"entropy": 0.97109375,
"epoch": 1.2177270552859973,
"grad_norm": 0.7102312077899564,
"learning_rate": 2.0166971959983094e-06,
"loss": 0.6925,
"mean_token_accuracy": 0.8125487804412842,
"num_tokens": 789527672.0,
"step": 8910
},
{
"entropy": 0.9453125,
"epoch": 1.2190938290166062,
"grad_norm": 0.7440070254878381,
"learning_rate": 2.0131745808087925e-06,
"loss": 0.6533,
"mean_token_accuracy": 0.82179936170578,
"num_tokens": 790430285.0,
"step": 8920
},
{
"entropy": 0.9513671875,
"epoch": 1.2204606027472151,
"grad_norm": 0.8390291444225266,
"learning_rate": 2.009651965619276e-06,
"loss": 0.6413,
"mean_token_accuracy": 0.8234784364700317,
"num_tokens": 791320202.0,
"step": 8930
},
{
"entropy": 0.97890625,
"epoch": 1.221827376477824,
"grad_norm": 0.8171390781757181,
"learning_rate": 2.0061293504297592e-06,
"loss": 0.6786,
"mean_token_accuracy": 0.8156248360872269,
"num_tokens": 792228711.0,
"step": 8940
},
{
"entropy": 0.9388671875,
"epoch": 1.223194150208433,
"grad_norm": 0.8124132026427158,
"learning_rate": 2.0026067352402428e-06,
"loss": 0.6384,
"mean_token_accuracy": 0.8258668661117554,
"num_tokens": 793111439.0,
"step": 8950
},
{
"entropy": 0.958984375,
"epoch": 1.2245609239390418,
"grad_norm": 0.7232707405306672,
"learning_rate": 1.999084120050726e-06,
"loss": 0.7006,
"mean_token_accuracy": 0.8117028534412384,
"num_tokens": 794001756.0,
"step": 8960
},
{
"entropy": 0.9494140625,
"epoch": 1.2259276976696507,
"grad_norm": 0.7844471513073216,
"learning_rate": 1.9955615048612094e-06,
"loss": 0.6558,
"mean_token_accuracy": 0.8204665273427963,
"num_tokens": 794884233.0,
"step": 8970
},
{
"entropy": 0.9560546875,
"epoch": 1.2272944714002596,
"grad_norm": 0.7773991092764364,
"learning_rate": 1.9920388896716926e-06,
"loss": 0.6376,
"mean_token_accuracy": 0.8257721841335297,
"num_tokens": 795772500.0,
"step": 8980
},
{
"entropy": 0.9779296875,
"epoch": 1.2286612451308685,
"grad_norm": 0.7396742309681765,
"learning_rate": 1.9885162744821757e-06,
"loss": 0.6448,
"mean_token_accuracy": 0.8252496510744095,
"num_tokens": 796668379.0,
"step": 8990
},
{
"entropy": 0.9759765625,
"epoch": 1.2300280188614774,
"grad_norm": 0.8097116458116451,
"learning_rate": 1.9849936592926592e-06,
"loss": 0.6534,
"mean_token_accuracy": 0.821374449133873,
"num_tokens": 797490645.0,
"step": 9000
},
{
"entropy": 0.9830078125,
"epoch": 1.2313947925920863,
"grad_norm": 0.7567160727673162,
"learning_rate": 1.9814710441031424e-06,
"loss": 0.6462,
"mean_token_accuracy": 0.8207679867744446,
"num_tokens": 798370052.0,
"step": 9010
},
{
"entropy": 0.931640625,
"epoch": 1.2327615663226954,
"grad_norm": 0.8260273570216762,
"learning_rate": 1.9779484289136255e-06,
"loss": 0.6395,
"mean_token_accuracy": 0.8241051822900772,
"num_tokens": 799229778.0,
"step": 9020
},
{
"entropy": 0.9744140625,
"epoch": 1.2341283400533043,
"grad_norm": 0.7689746568532453,
"learning_rate": 1.974425813724109e-06,
"loss": 0.6544,
"mean_token_accuracy": 0.8220506489276886,
"num_tokens": 800125708.0,
"step": 9030
},
{
"entropy": 0.9603515625,
"epoch": 1.2354951137839132,
"grad_norm": 0.7630390441365026,
"learning_rate": 1.970903198534592e-06,
"loss": 0.6527,
"mean_token_accuracy": 0.8223497718572617,
"num_tokens": 801028037.0,
"step": 9040
},
{
"entropy": 0.955078125,
"epoch": 1.236861887514522,
"grad_norm": 0.8659551769840342,
"learning_rate": 1.9673805833450753e-06,
"loss": 0.6866,
"mean_token_accuracy": 0.8133615493774414,
"num_tokens": 801926801.0,
"step": 9050
},
{
"entropy": 0.9912109375,
"epoch": 1.238228661245131,
"grad_norm": 0.6526832665649603,
"learning_rate": 1.963857968155559e-06,
"loss": 0.663,
"mean_token_accuracy": 0.8183393895626068,
"num_tokens": 802832899.0,
"step": 9060
},
{
"entropy": 0.962890625,
"epoch": 1.2395954349757399,
"grad_norm": 0.9308593904574556,
"learning_rate": 1.960335352966042e-06,
"loss": 0.6626,
"mean_token_accuracy": 0.8201309889554977,
"num_tokens": 803717792.0,
"step": 9070
},
{
"entropy": 0.9732421875,
"epoch": 1.2409622087063488,
"grad_norm": 0.8085957202651565,
"learning_rate": 1.9568127377765255e-06,
"loss": 0.6985,
"mean_token_accuracy": 0.8095783531665802,
"num_tokens": 804630519.0,
"step": 9080
},
{
"entropy": 0.9369140625,
"epoch": 1.2423289824369577,
"grad_norm": 0.7545865523991641,
"learning_rate": 1.9532901225870086e-06,
"loss": 0.643,
"mean_token_accuracy": 0.8243232250213623,
"num_tokens": 805565081.0,
"step": 9090
},
{
"entropy": 0.9591796875,
"epoch": 1.2436957561675666,
"grad_norm": 0.8336220781967275,
"learning_rate": 1.949767507397492e-06,
"loss": 0.6646,
"mean_token_accuracy": 0.819222030043602,
"num_tokens": 806425066.0,
"step": 9100
},
{
"entropy": 0.9806640625,
"epoch": 1.2450625298981755,
"grad_norm": 0.7724797186262227,
"learning_rate": 1.9462448922079753e-06,
"loss": 0.6864,
"mean_token_accuracy": 0.8132601261138916,
"num_tokens": 807299718.0,
"step": 9110
},
{
"entropy": 0.968359375,
"epoch": 1.2464293036287843,
"grad_norm": 0.7447693533514633,
"learning_rate": 1.942722277018459e-06,
"loss": 0.6714,
"mean_token_accuracy": 0.8172522008419036,
"num_tokens": 808167887.0,
"step": 9120
},
{
"entropy": 0.9892578125,
"epoch": 1.2477960773593932,
"grad_norm": 0.8506382496848501,
"learning_rate": 1.939199661828942e-06,
"loss": 0.6945,
"mean_token_accuracy": 0.8113091826438904,
"num_tokens": 809042851.0,
"step": 9130
},
{
"entropy": 0.935546875,
"epoch": 1.2491628510900021,
"grad_norm": 0.8355172737501745,
"learning_rate": 1.9356770466394255e-06,
"loss": 0.6376,
"mean_token_accuracy": 0.8258450776338577,
"num_tokens": 809871246.0,
"step": 9140
},
{
"entropy": 0.92421875,
"epoch": 1.250529624820611,
"grad_norm": 0.7709583016818217,
"learning_rate": 1.9321544314499086e-06,
"loss": 0.6816,
"mean_token_accuracy": 0.8160624414682388,
"num_tokens": 810754067.0,
"step": 9150
},
{
"entropy": 0.9513671875,
"epoch": 1.25189639855122,
"grad_norm": 0.9670709089860952,
"learning_rate": 1.9286318162603917e-06,
"loss": 0.6576,
"mean_token_accuracy": 0.8186320066452026,
"num_tokens": 811652395.0,
"step": 9160
},
{
"entropy": 0.9294921875,
"epoch": 1.2532631722818288,
"grad_norm": 0.7014359894599187,
"learning_rate": 1.9251092010708753e-06,
"loss": 0.6329,
"mean_token_accuracy": 0.828794127702713,
"num_tokens": 812537595.0,
"step": 9170
},
{
"entropy": 0.97734375,
"epoch": 1.2546299460124377,
"grad_norm": 0.6971853593539425,
"learning_rate": 1.9215865858813584e-06,
"loss": 0.6702,
"mean_token_accuracy": 0.8172507077455521,
"num_tokens": 813418052.0,
"step": 9180
},
{
"entropy": 0.988671875,
"epoch": 1.2559967197430466,
"grad_norm": 0.7098656774545408,
"learning_rate": 1.918063970691842e-06,
"loss": 0.6943,
"mean_token_accuracy": 0.8112302988767623,
"num_tokens": 814316613.0,
"step": 9190
},
{
"entropy": 0.9794921875,
"epoch": 1.2573634934736555,
"grad_norm": 0.7386913879649112,
"learning_rate": 1.914541355502325e-06,
"loss": 0.644,
"mean_token_accuracy": 0.822086563706398,
"num_tokens": 815160785.0,
"step": 9200
},
{
"entropy": 0.951171875,
"epoch": 1.2587302672042644,
"grad_norm": 0.6903349282963741,
"learning_rate": 1.9110187403128082e-06,
"loss": 0.6393,
"mean_token_accuracy": 0.8226767152547836,
"num_tokens": 816036366.0,
"step": 9210
},
{
"entropy": 0.9423828125,
"epoch": 1.2600970409348733,
"grad_norm": 0.757315892690813,
"learning_rate": 1.9074961251232918e-06,
"loss": 0.6635,
"mean_token_accuracy": 0.8199472308158875,
"num_tokens": 816965958.0,
"step": 9220
},
{
"entropy": 0.9408203125,
"epoch": 1.2614638146654822,
"grad_norm": 0.8117734805509911,
"learning_rate": 1.903973509933775e-06,
"loss": 0.6434,
"mean_token_accuracy": 0.8201697915792465,
"num_tokens": 817830932.0,
"step": 9230
},
{
"entropy": 1.00234375,
"epoch": 1.262830588396091,
"grad_norm": 0.7894905531522487,
"learning_rate": 1.9004508947442582e-06,
"loss": 0.6794,
"mean_token_accuracy": 0.8162721246480942,
"num_tokens": 818690604.0,
"step": 9240
},
{
"entropy": 0.955859375,
"epoch": 1.2641973621267,
"grad_norm": 0.8151169239008846,
"learning_rate": 1.8969282795547418e-06,
"loss": 0.6432,
"mean_token_accuracy": 0.8246512889862061,
"num_tokens": 819552184.0,
"step": 9250
},
{
"entropy": 0.98671875,
"epoch": 1.2655641358573089,
"grad_norm": 0.746770863450645,
"learning_rate": 1.8934056643652249e-06,
"loss": 0.658,
"mean_token_accuracy": 0.8203916549682617,
"num_tokens": 820423497.0,
"step": 9260
},
{
"entropy": 0.952734375,
"epoch": 1.2669309095879178,
"grad_norm": 0.7105146778102217,
"learning_rate": 1.889883049175708e-06,
"loss": 0.6483,
"mean_token_accuracy": 0.8231089383363723,
"num_tokens": 821295538.0,
"step": 9270
},
{
"entropy": 0.9111328125,
"epoch": 1.2682976833185267,
"grad_norm": 0.7423466710126421,
"learning_rate": 1.8863604339861916e-06,
"loss": 0.6418,
"mean_token_accuracy": 0.8226804554462432,
"num_tokens": 822180580.0,
"step": 9280
},
{
"entropy": 0.9841796875,
"epoch": 1.2696644570491356,
"grad_norm": 0.8646465662489968,
"learning_rate": 1.8828378187966747e-06,
"loss": 0.6914,
"mean_token_accuracy": 0.8134060978889466,
"num_tokens": 823090612.0,
"step": 9290
},
{
"entropy": 0.9267578125,
"epoch": 1.2710312307797444,
"grad_norm": 0.7705069716891149,
"learning_rate": 1.8793152036071582e-06,
"loss": 0.6585,
"mean_token_accuracy": 0.8212078273296356,
"num_tokens": 823965696.0,
"step": 9300
},
{
"entropy": 0.96875,
"epoch": 1.2723980045103533,
"grad_norm": 0.7427883929240834,
"learning_rate": 1.8757925884176414e-06,
"loss": 0.679,
"mean_token_accuracy": 0.8159714341163635,
"num_tokens": 824871297.0,
"step": 9310
},
{
"entropy": 0.976953125,
"epoch": 1.2737647782409622,
"grad_norm": 0.8863523566574623,
"learning_rate": 1.8722699732281247e-06,
"loss": 0.6969,
"mean_token_accuracy": 0.8099540144205093,
"num_tokens": 825811434.0,
"step": 9320
},
{
"entropy": 0.9880859375,
"epoch": 1.2751315519715711,
"grad_norm": 0.768313887783833,
"learning_rate": 1.868747358038608e-06,
"loss": 0.6968,
"mean_token_accuracy": 0.812763637304306,
"num_tokens": 826682534.0,
"step": 9330
},
{
"entropy": 0.9279296875,
"epoch": 1.27649832570218,
"grad_norm": 0.78998816505468,
"learning_rate": 1.8652247428490914e-06,
"loss": 0.6686,
"mean_token_accuracy": 0.8182229697704315,
"num_tokens": 827565061.0,
"step": 9340
},
{
"entropy": 0.991796875,
"epoch": 1.277865099432789,
"grad_norm": 0.7823500948805245,
"learning_rate": 1.8617021276595745e-06,
"loss": 0.6899,
"mean_token_accuracy": 0.8129715740680694,
"num_tokens": 828433291.0,
"step": 9350
},
{
"entropy": 0.9599609375,
"epoch": 1.2792318731633978,
"grad_norm": 0.8450051695644183,
"learning_rate": 1.858179512470058e-06,
"loss": 0.6608,
"mean_token_accuracy": 0.8231855094432831,
"num_tokens": 829312319.0,
"step": 9360
},
{
"entropy": 0.944140625,
"epoch": 1.2805986468940067,
"grad_norm": 0.791048146851969,
"learning_rate": 1.8546568972805412e-06,
"loss": 0.6705,
"mean_token_accuracy": 0.818240362405777,
"num_tokens": 830252194.0,
"step": 9370
},
{
"entropy": 0.9490234375,
"epoch": 1.2819654206246156,
"grad_norm": 0.8178907631437038,
"learning_rate": 1.8511342820910245e-06,
"loss": 0.6349,
"mean_token_accuracy": 0.8235474348068237,
"num_tokens": 831128961.0,
"step": 9380
},
{
"entropy": 0.9287109375,
"epoch": 1.2833321943552245,
"grad_norm": 0.7613514029226971,
"learning_rate": 1.8476116669015078e-06,
"loss": 0.6198,
"mean_token_accuracy": 0.8273909777402878,
"num_tokens": 832029477.0,
"step": 9390
},
{
"entropy": 0.9380859375,
"epoch": 1.2846989680858334,
"grad_norm": 0.7362017155852393,
"learning_rate": 1.8440890517119912e-06,
"loss": 0.6258,
"mean_token_accuracy": 0.8269206523895264,
"num_tokens": 832931731.0,
"step": 9400
},
{
"entropy": 0.9544921875,
"epoch": 1.2860657418164423,
"grad_norm": 0.8129029782642632,
"learning_rate": 1.8405664365224743e-06,
"loss": 0.6621,
"mean_token_accuracy": 0.819603654742241,
"num_tokens": 833785182.0,
"step": 9410
},
{
"entropy": 0.9880859375,
"epoch": 1.2874325155470512,
"grad_norm": 0.8468513660539789,
"learning_rate": 1.8370438213329578e-06,
"loss": 0.6758,
"mean_token_accuracy": 0.8171648412942887,
"num_tokens": 834717467.0,
"step": 9420
},
{
"entropy": 0.9646484375,
"epoch": 1.28879928927766,
"grad_norm": 0.7931214137929057,
"learning_rate": 1.833521206143441e-06,
"loss": 0.6799,
"mean_token_accuracy": 0.8164188325405121,
"num_tokens": 835602297.0,
"step": 9430
},
{
"entropy": 1.00703125,
"epoch": 1.290166063008269,
"grad_norm": 0.7494563659106036,
"learning_rate": 1.8299985909539245e-06,
"loss": 0.6856,
"mean_token_accuracy": 0.8118420153856277,
"num_tokens": 836448213.0,
"step": 9440
},
{
"entropy": 0.9783203125,
"epoch": 1.2915328367388779,
"grad_norm": 0.8143604978157797,
"learning_rate": 1.8264759757644076e-06,
"loss": 0.6624,
"mean_token_accuracy": 0.8184799611568451,
"num_tokens": 837327129.0,
"step": 9450
},
{
"entropy": 0.9529296875,
"epoch": 1.2928996104694868,
"grad_norm": 0.8467418296845711,
"learning_rate": 1.822953360574891e-06,
"loss": 0.6596,
"mean_token_accuracy": 0.8189761310815811,
"num_tokens": 838172987.0,
"step": 9460
},
{
"entropy": 0.9236328125,
"epoch": 1.2942663842000957,
"grad_norm": 0.6792419618837691,
"learning_rate": 1.8194307453853743e-06,
"loss": 0.6335,
"mean_token_accuracy": 0.82454614341259,
"num_tokens": 839057791.0,
"step": 9470
},
{
"entropy": 1.0072265625,
"epoch": 1.2956331579307045,
"grad_norm": 0.8326927413483743,
"learning_rate": 1.8159081301958576e-06,
"loss": 0.6673,
"mean_token_accuracy": 0.818119814991951,
"num_tokens": 839956502.0,
"step": 9480
},
{
"entropy": 0.947265625,
"epoch": 1.2969999316613134,
"grad_norm": 0.653273157744325,
"learning_rate": 1.8123855150063408e-06,
"loss": 0.6491,
"mean_token_accuracy": 0.82373805642128,
"num_tokens": 840854870.0,
"step": 9490
},
{
"entropy": 0.9740234375,
"epoch": 1.2983667053919223,
"grad_norm": 0.7037300449849899,
"learning_rate": 1.8088628998168243e-06,
"loss": 0.6453,
"mean_token_accuracy": 0.8219100713729859,
"num_tokens": 841722038.0,
"step": 9500
},
{
"entropy": 0.93046875,
"epoch": 1.2997334791225312,
"grad_norm": 0.821168595555187,
"learning_rate": 1.8053402846273074e-06,
"loss": 0.6377,
"mean_token_accuracy": 0.8236586302518845,
"num_tokens": 842603404.0,
"step": 9510
},
{
"entropy": 0.90703125,
"epoch": 1.3011002528531401,
"grad_norm": 0.7282425626832384,
"learning_rate": 1.8018176694377906e-06,
"loss": 0.6208,
"mean_token_accuracy": 0.8272628515958786,
"num_tokens": 843468734.0,
"step": 9520
},
{
"entropy": 0.951171875,
"epoch": 1.302467026583749,
"grad_norm": 0.7588620672129183,
"learning_rate": 1.798295054248274e-06,
"loss": 0.6565,
"mean_token_accuracy": 0.8202333509922027,
"num_tokens": 844321422.0,
"step": 9530
},
{
"entropy": 0.9587890625,
"epoch": 1.303833800314358,
"grad_norm": 0.8415938219660634,
"learning_rate": 1.7947724390587572e-06,
"loss": 0.6438,
"mean_token_accuracy": 0.8215611487627029,
"num_tokens": 845185266.0,
"step": 9540
},
{
"entropy": 0.9802734375,
"epoch": 1.3052005740449668,
"grad_norm": 0.8682839706323242,
"learning_rate": 1.7912498238692408e-06,
"loss": 0.6912,
"mean_token_accuracy": 0.8113156139850617,
"num_tokens": 846083764.0,
"step": 9550
},
{
"entropy": 0.9580078125,
"epoch": 1.3065673477755757,
"grad_norm": 0.7245148525553705,
"learning_rate": 1.7877272086797239e-06,
"loss": 0.6689,
"mean_token_accuracy": 0.8182174772024154,
"num_tokens": 846968868.0,
"step": 9560
},
{
"entropy": 0.9751953125,
"epoch": 1.3079341215061846,
"grad_norm": 0.7473023053668696,
"learning_rate": 1.7842045934902072e-06,
"loss": 0.6558,
"mean_token_accuracy": 0.8204163253307343,
"num_tokens": 847867293.0,
"step": 9570
},
{
"entropy": 0.9501953125,
"epoch": 1.3093008952367935,
"grad_norm": 0.7859866543827214,
"learning_rate": 1.7806819783006908e-06,
"loss": 0.6939,
"mean_token_accuracy": 0.8109886109828949,
"num_tokens": 848738392.0,
"step": 9580
},
{
"entropy": 0.9544921875,
"epoch": 1.3106676689674024,
"grad_norm": 0.7970081589393453,
"learning_rate": 1.777159363111174e-06,
"loss": 0.6765,
"mean_token_accuracy": 0.8174547731876374,
"num_tokens": 849620845.0,
"step": 9590
},
{
"entropy": 0.9498046875,
"epoch": 1.3120344426980113,
"grad_norm": 0.7117716758306367,
"learning_rate": 1.773636747921657e-06,
"loss": 0.6518,
"mean_token_accuracy": 0.8188720703125,
"num_tokens": 850463193.0,
"step": 9600
},
{
"entropy": 0.9828125,
"epoch": 1.3134012164286202,
"grad_norm": 0.7970754483535241,
"learning_rate": 1.7701141327321406e-06,
"loss": 0.6677,
"mean_token_accuracy": 0.8178690373897552,
"num_tokens": 851372856.0,
"step": 9610
},
{
"entropy": 0.953125,
"epoch": 1.314767990159229,
"grad_norm": 0.6730724977340734,
"learning_rate": 1.7665915175426237e-06,
"loss": 0.6646,
"mean_token_accuracy": 0.8197014033794403,
"num_tokens": 852308597.0,
"step": 9620
},
{
"entropy": 0.9330078125,
"epoch": 1.316134763889838,
"grad_norm": 0.8073195994992572,
"learning_rate": 1.763068902353107e-06,
"loss": 0.6494,
"mean_token_accuracy": 0.8214480638504028,
"num_tokens": 853172611.0,
"step": 9630
},
{
"entropy": 0.9904296875,
"epoch": 1.3175015376204469,
"grad_norm": 0.8404412117262525,
"learning_rate": 1.7595462871635904e-06,
"loss": 0.7076,
"mean_token_accuracy": 0.8091458350419998,
"num_tokens": 854025251.0,
"step": 9640
},
{
"entropy": 0.9384765625,
"epoch": 1.3188683113510558,
"grad_norm": 0.751388598788113,
"learning_rate": 1.7560236719740737e-06,
"loss": 0.6197,
"mean_token_accuracy": 0.8281062752008438,
"num_tokens": 854918609.0,
"step": 9650
},
{
"entropy": 0.9533203125,
"epoch": 1.3202350850816646,
"grad_norm": 0.714100783300306,
"learning_rate": 1.752501056784557e-06,
"loss": 0.6408,
"mean_token_accuracy": 0.8250426322221756,
"num_tokens": 855840317.0,
"step": 9660
},
{
"entropy": 0.94765625,
"epoch": 1.3216018588122735,
"grad_norm": 0.802118909366715,
"learning_rate": 1.7489784415950404e-06,
"loss": 0.6117,
"mean_token_accuracy": 0.8283844232559204,
"num_tokens": 856694202.0,
"step": 9670
},
{
"entropy": 0.9365234375,
"epoch": 1.3229686325428824,
"grad_norm": 0.6984910113074086,
"learning_rate": 1.7454558264055235e-06,
"loss": 0.6648,
"mean_token_accuracy": 0.8176283597946167,
"num_tokens": 857560944.0,
"step": 9680
},
{
"entropy": 0.9564453125,
"epoch": 1.3243354062734913,
"grad_norm": 0.8218611334984088,
"learning_rate": 1.741933211216007e-06,
"loss": 0.6451,
"mean_token_accuracy": 0.8238427668809891,
"num_tokens": 858425303.0,
"step": 9690
},
{
"entropy": 0.96953125,
"epoch": 1.3257021800041002,
"grad_norm": 0.8088021317669766,
"learning_rate": 1.7384105960264902e-06,
"loss": 0.6818,
"mean_token_accuracy": 0.8152011394500732,
"num_tokens": 859322782.0,
"step": 9700
},
{
"entropy": 1.0068359375,
"epoch": 1.3270689537347091,
"grad_norm": 0.8596617741916261,
"learning_rate": 1.7348879808369735e-06,
"loss": 0.7228,
"mean_token_accuracy": 0.8037614732980728,
"num_tokens": 860228542.0,
"step": 9710
},
{
"entropy": 0.94296875,
"epoch": 1.328435727465318,
"grad_norm": 0.80424570166153,
"learning_rate": 1.7313653656474568e-06,
"loss": 0.6994,
"mean_token_accuracy": 0.8128274440765381,
"num_tokens": 861128996.0,
"step": 9720
},
{
"entropy": 0.9775390625,
"epoch": 1.329802501195927,
"grad_norm": 0.7877134160861172,
"learning_rate": 1.7278427504579402e-06,
"loss": 0.6977,
"mean_token_accuracy": 0.8150995224714279,
"num_tokens": 862045862.0,
"step": 9730
},
{
"entropy": 0.9814453125,
"epoch": 1.3311692749265358,
"grad_norm": 0.7376216580063435,
"learning_rate": 1.7243201352684233e-06,
"loss": 0.6741,
"mean_token_accuracy": 0.8168181240558624,
"num_tokens": 862980918.0,
"step": 9740
},
{
"entropy": 0.9765625,
"epoch": 1.3325360486571447,
"grad_norm": 0.8524490265930417,
"learning_rate": 1.7207975200789068e-06,
"loss": 0.7127,
"mean_token_accuracy": 0.8097017109394073,
"num_tokens": 863864536.0,
"step": 9750
},
{
"entropy": 0.960546875,
"epoch": 1.3339028223877536,
"grad_norm": 0.8063213939838937,
"learning_rate": 1.71727490488939e-06,
"loss": 0.673,
"mean_token_accuracy": 0.8165948808193206,
"num_tokens": 864692208.0,
"step": 9760
},
{
"entropy": 0.968359375,
"epoch": 1.3352695961183625,
"grad_norm": 0.8761133669572009,
"learning_rate": 1.7137522896998735e-06,
"loss": 0.6637,
"mean_token_accuracy": 0.8180168926715851,
"num_tokens": 865583812.0,
"step": 9770
},
{
"entropy": 0.9310546875,
"epoch": 1.3366363698489714,
"grad_norm": 0.7460893078317583,
"learning_rate": 1.7102296745103566e-06,
"loss": 0.6443,
"mean_token_accuracy": 0.8247941672801972,
"num_tokens": 866442714.0,
"step": 9780
},
{
"entropy": 0.9654296875,
"epoch": 1.3380031435795803,
"grad_norm": 0.7994015003783534,
"learning_rate": 1.7067070593208398e-06,
"loss": 0.6681,
"mean_token_accuracy": 0.8185766398906708,
"num_tokens": 867348693.0,
"step": 9790
},
{
"entropy": 0.943359375,
"epoch": 1.3393699173101892,
"grad_norm": 0.7604060312929702,
"learning_rate": 1.7031844441313233e-06,
"loss": 0.65,
"mean_token_accuracy": 0.8231387794017792,
"num_tokens": 868245125.0,
"step": 9800
},
{
"entropy": 0.9525390625,
"epoch": 1.340736691040798,
"grad_norm": 0.7725418805270208,
"learning_rate": 1.6996618289418064e-06,
"loss": 0.6636,
"mean_token_accuracy": 0.8194416135549545,
"num_tokens": 869147634.0,
"step": 9810
},
{
"entropy": 0.96328125,
"epoch": 1.342103464771407,
"grad_norm": 0.7853009649694399,
"learning_rate": 1.6961392137522898e-06,
"loss": 0.6636,
"mean_token_accuracy": 0.8168342262506485,
"num_tokens": 869984569.0,
"step": 9820
},
{
"entropy": 0.9337890625,
"epoch": 1.3434702385020159,
"grad_norm": 0.7407062449932269,
"learning_rate": 1.6926165985627733e-06,
"loss": 0.6697,
"mean_token_accuracy": 0.8188673168420791,
"num_tokens": 870822937.0,
"step": 9830
},
{
"entropy": 0.965234375,
"epoch": 1.3448370122326248,
"grad_norm": 0.7786327613433935,
"learning_rate": 1.6890939833732564e-06,
"loss": 0.717,
"mean_token_accuracy": 0.8061766117811203,
"num_tokens": 871733262.0,
"step": 9840
},
{
"entropy": 0.990625,
"epoch": 1.3462037859632339,
"grad_norm": 0.8038424313035836,
"learning_rate": 1.6855713681837396e-06,
"loss": 0.6912,
"mean_token_accuracy": 0.8132872313261033,
"num_tokens": 872630340.0,
"step": 9850
},
{
"entropy": 0.9525390625,
"epoch": 1.3475705596938428,
"grad_norm": 0.7920641843276333,
"learning_rate": 1.6820487529942231e-06,
"loss": 0.6529,
"mean_token_accuracy": 0.8209212034940719,
"num_tokens": 873508298.0,
"step": 9860
},
{
"entropy": 0.94140625,
"epoch": 1.3489373334244517,
"grad_norm": 0.8969575598203288,
"learning_rate": 1.6785261378047062e-06,
"loss": 0.6416,
"mean_token_accuracy": 0.8238828271627426,
"num_tokens": 874391916.0,
"step": 9870
},
{
"entropy": 0.9572265625,
"epoch": 1.3503041071550606,
"grad_norm": 0.7711816214225254,
"learning_rate": 1.6750035226151898e-06,
"loss": 0.6503,
"mean_token_accuracy": 0.8220989048480988,
"num_tokens": 875231876.0,
"step": 9880
},
{
"entropy": 0.97109375,
"epoch": 1.3516708808856694,
"grad_norm": 0.8468380475800077,
"learning_rate": 1.671480907425673e-06,
"loss": 0.6807,
"mean_token_accuracy": 0.8157043308019638,
"num_tokens": 876127902.0,
"step": 9890
},
{
"entropy": 0.95546875,
"epoch": 1.3530376546162783,
"grad_norm": 0.848617080057057,
"learning_rate": 1.6679582922361562e-06,
"loss": 0.6815,
"mean_token_accuracy": 0.8160282224416733,
"num_tokens": 876985950.0,
"step": 9900
},
{
"entropy": 0.9638671875,
"epoch": 1.3544044283468872,
"grad_norm": 0.8322917980625255,
"learning_rate": 1.6644356770466396e-06,
"loss": 0.6566,
"mean_token_accuracy": 0.8197604805231095,
"num_tokens": 877880193.0,
"step": 9910
},
{
"entropy": 0.969140625,
"epoch": 1.3557712020774961,
"grad_norm": 0.8260890213086692,
"learning_rate": 1.660913061857123e-06,
"loss": 0.6725,
"mean_token_accuracy": 0.8145093828439712,
"num_tokens": 878770858.0,
"step": 9920
},
{
"entropy": 0.9201171875,
"epoch": 1.357137975808105,
"grad_norm": 0.8294174367737678,
"learning_rate": 1.657390446667606e-06,
"loss": 0.6073,
"mean_token_accuracy": 0.8308471530675888,
"num_tokens": 879667998.0,
"step": 9930
},
{
"entropy": 0.939453125,
"epoch": 1.358504749538714,
"grad_norm": 0.758246033445911,
"learning_rate": 1.6538678314780896e-06,
"loss": 0.6563,
"mean_token_accuracy": 0.820938178896904,
"num_tokens": 880544633.0,
"step": 9940
},
{
"entropy": 1.0283203125,
"epoch": 1.3598715232693228,
"grad_norm": 0.8152867482027147,
"learning_rate": 1.6503452162885727e-06,
"loss": 0.7405,
"mean_token_accuracy": 0.8009586274623871,
"num_tokens": 881402825.0,
"step": 9950
},
{
"entropy": 0.9609375,
"epoch": 1.3612382969999317,
"grad_norm": 0.8194289725668792,
"learning_rate": 1.646822601099056e-06,
"loss": 0.6318,
"mean_token_accuracy": 0.8245018213987351,
"num_tokens": 882291149.0,
"step": 9960
},
{
"entropy": 0.93359375,
"epoch": 1.3626050707305406,
"grad_norm": 0.8165889040271594,
"learning_rate": 1.6432999859095394e-06,
"loss": 0.6475,
"mean_token_accuracy": 0.8228644341230392,
"num_tokens": 883176457.0,
"step": 9970
},
{
"entropy": 0.9916015625,
"epoch": 1.3639718444611495,
"grad_norm": 0.847584843605675,
"learning_rate": 1.6397773707200227e-06,
"loss": 0.7116,
"mean_token_accuracy": 0.8081254929304122,
"num_tokens": 884063256.0,
"step": 9980
},
{
"entropy": 0.960546875,
"epoch": 1.3653386181917584,
"grad_norm": 0.7626379193974944,
"learning_rate": 1.636254755530506e-06,
"loss": 0.6561,
"mean_token_accuracy": 0.8207849949598313,
"num_tokens": 884906668.0,
"step": 9990
},
{
"entropy": 0.9041015625,
"epoch": 1.3667053919223673,
"grad_norm": 0.7465630931105545,
"learning_rate": 1.6327321403409894e-06,
"loss": 0.5838,
"mean_token_accuracy": 0.835880360007286,
"num_tokens": 885782940.0,
"step": 10000
},
{
"entropy": 0.9783203125,
"epoch": 1.3680721656529762,
"grad_norm": 0.9539297705692156,
"learning_rate": 1.6292095251514725e-06,
"loss": 0.6509,
"mean_token_accuracy": 0.8213387310504914,
"num_tokens": 886655374.0,
"step": 10010
},
{
"entropy": 0.9908203125,
"epoch": 1.369438939383585,
"grad_norm": 0.8238106927783081,
"learning_rate": 1.625686909961956e-06,
"loss": 0.7146,
"mean_token_accuracy": 0.8075925707817078,
"num_tokens": 887551775.0,
"step": 10020
},
{
"entropy": 0.975390625,
"epoch": 1.370805713114194,
"grad_norm": 0.8634206097629534,
"learning_rate": 1.6221642947724392e-06,
"loss": 0.6931,
"mean_token_accuracy": 0.8118404626846314,
"num_tokens": 888388302.0,
"step": 10030
},
{
"entropy": 0.9837890625,
"epoch": 1.3721724868448029,
"grad_norm": 0.8663527415971405,
"learning_rate": 1.6186416795829223e-06,
"loss": 0.6758,
"mean_token_accuracy": 0.8152767151594162,
"num_tokens": 889290281.0,
"step": 10040
},
{
"entropy": 0.9744140625,
"epoch": 1.3735392605754118,
"grad_norm": 0.7955066513491015,
"learning_rate": 1.6151190643934058e-06,
"loss": 0.6642,
"mean_token_accuracy": 0.8189565509557724,
"num_tokens": 890192489.0,
"step": 10050
},
{
"entropy": 0.936328125,
"epoch": 1.3749060343060207,
"grad_norm": 0.7951911716704625,
"learning_rate": 1.6115964492038892e-06,
"loss": 0.65,
"mean_token_accuracy": 0.8208369106054306,
"num_tokens": 891071923.0,
"step": 10060
},
{
"entropy": 0.952734375,
"epoch": 1.3762728080366295,
"grad_norm": 0.7564968854659422,
"learning_rate": 1.6080738340143723e-06,
"loss": 0.6508,
"mean_token_accuracy": 0.8257468670606614,
"num_tokens": 891978878.0,
"step": 10070
},
{
"entropy": 0.9447265625,
"epoch": 1.3776395817672384,
"grad_norm": 0.7575684841181967,
"learning_rate": 1.6045512188248559e-06,
"loss": 0.6601,
"mean_token_accuracy": 0.821817809343338,
"num_tokens": 892814800.0,
"step": 10080
},
{
"entropy": 0.9646484375,
"epoch": 1.3790063554978473,
"grad_norm": 0.730241686291065,
"learning_rate": 1.601028603635339e-06,
"loss": 0.6569,
"mean_token_accuracy": 0.8189956367015838,
"num_tokens": 893678292.0,
"step": 10090
},
{
"entropy": 0.9578125,
"epoch": 1.3803731292284562,
"grad_norm": 0.870849417951906,
"learning_rate": 1.5975059884458225e-06,
"loss": 0.68,
"mean_token_accuracy": 0.8153973400592804,
"num_tokens": 894600271.0,
"step": 10100
},
{
"entropy": 0.946484375,
"epoch": 1.3817399029590651,
"grad_norm": 0.7593058590697757,
"learning_rate": 1.5939833732563056e-06,
"loss": 0.6406,
"mean_token_accuracy": 0.823324465751648,
"num_tokens": 895489860.0,
"step": 10110
},
{
"entropy": 0.9517578125,
"epoch": 1.383106676689674,
"grad_norm": 0.7212583590781951,
"learning_rate": 1.5904607580667888e-06,
"loss": 0.6834,
"mean_token_accuracy": 0.8153128325939178,
"num_tokens": 896400113.0,
"step": 10120
},
{
"entropy": 0.96171875,
"epoch": 1.384473450420283,
"grad_norm": 0.8703022040460187,
"learning_rate": 1.5869381428772723e-06,
"loss": 0.6926,
"mean_token_accuracy": 0.8125747859477996,
"num_tokens": 897273945.0,
"step": 10130
},
{
"entropy": 0.983984375,
"epoch": 1.3858402241508918,
"grad_norm": 0.8272747073146095,
"learning_rate": 1.5834155276877554e-06,
"loss": 0.6806,
"mean_token_accuracy": 0.8157221674919128,
"num_tokens": 898141031.0,
"step": 10140
},
{
"entropy": 0.9908203125,
"epoch": 1.3872069978815007,
"grad_norm": 0.7430163517957112,
"learning_rate": 1.5798929124982388e-06,
"loss": 0.6894,
"mean_token_accuracy": 0.8135321795940399,
"num_tokens": 898982893.0,
"step": 10150
},
{
"entropy": 0.9263671875,
"epoch": 1.3885737716121096,
"grad_norm": 0.8054988303795223,
"learning_rate": 1.5763702973087221e-06,
"loss": 0.6581,
"mean_token_accuracy": 0.8199193567037583,
"num_tokens": 899851800.0,
"step": 10160
},
{
"entropy": 0.9287109375,
"epoch": 1.3899405453427185,
"grad_norm": 0.7643937094312923,
"learning_rate": 1.5728476821192054e-06,
"loss": 0.6361,
"mean_token_accuracy": 0.8264935493469239,
"num_tokens": 900727352.0,
"step": 10170
},
{
"entropy": 0.98046875,
"epoch": 1.3913073190733274,
"grad_norm": 0.7680625454745106,
"learning_rate": 1.5693250669296886e-06,
"loss": 0.6768,
"mean_token_accuracy": 0.8163814425468445,
"num_tokens": 901637094.0,
"step": 10180
},
{
"entropy": 0.925,
"epoch": 1.3926740928039363,
"grad_norm": 0.7348245888880519,
"learning_rate": 1.5658024517401721e-06,
"loss": 0.6497,
"mean_token_accuracy": 0.8213391155004501,
"num_tokens": 902522639.0,
"step": 10190
},
{
"entropy": 0.9455078125,
"epoch": 1.3940408665345452,
"grad_norm": 0.6873769621071208,
"learning_rate": 1.5622798365506552e-06,
"loss": 0.6554,
"mean_token_accuracy": 0.8234138786792755,
"num_tokens": 903429509.0,
"step": 10200
},
{
"entropy": 0.967578125,
"epoch": 1.395407640265154,
"grad_norm": 0.749370854237694,
"learning_rate": 1.5587572213611388e-06,
"loss": 0.6575,
"mean_token_accuracy": 0.8215772122144699,
"num_tokens": 904317564.0,
"step": 10210
},
{
"entropy": 1.003125,
"epoch": 1.396774413995763,
"grad_norm": 0.8393572608354448,
"learning_rate": 1.555234606171622e-06,
"loss": 0.6702,
"mean_token_accuracy": 0.8174093902111054,
"num_tokens": 905224463.0,
"step": 10220
},
{
"entropy": 0.95859375,
"epoch": 1.3981411877263719,
"grad_norm": 0.7563139672625745,
"learning_rate": 1.5517119909821052e-06,
"loss": 0.6355,
"mean_token_accuracy": 0.8245254963636398,
"num_tokens": 906142295.0,
"step": 10230
},
{
"entropy": 0.97734375,
"epoch": 1.3995079614569808,
"grad_norm": 0.7318004455555398,
"learning_rate": 1.5481893757925886e-06,
"loss": 0.6787,
"mean_token_accuracy": 0.8155497163534164,
"num_tokens": 907054448.0,
"step": 10240
},
{
"entropy": 0.99921875,
"epoch": 1.4008747351875896,
"grad_norm": 0.9232671810471399,
"learning_rate": 1.544666760603072e-06,
"loss": 0.6717,
"mean_token_accuracy": 0.8166177034378052,
"num_tokens": 907895116.0,
"step": 10250
},
{
"entropy": 0.9939453125,
"epoch": 1.4022415089181985,
"grad_norm": 0.8124700632495534,
"learning_rate": 1.541144145413555e-06,
"loss": 0.6931,
"mean_token_accuracy": 0.8115635514259338,
"num_tokens": 908802189.0,
"step": 10260
},
{
"entropy": 0.96875,
"epoch": 1.4036082826488074,
"grad_norm": 0.8521956065736913,
"learning_rate": 1.5376215302240386e-06,
"loss": 0.657,
"mean_token_accuracy": 0.8192612767219544,
"num_tokens": 909755000.0,
"step": 10270
},
{
"entropy": 0.958984375,
"epoch": 1.4049750563794163,
"grad_norm": 0.7409647563530984,
"learning_rate": 1.5340989150345217e-06,
"loss": 0.6518,
"mean_token_accuracy": 0.820759630203247,
"num_tokens": 910650883.0,
"step": 10280
},
{
"entropy": 0.920703125,
"epoch": 1.4063418301100252,
"grad_norm": 0.7061176861218158,
"learning_rate": 1.5305762998450048e-06,
"loss": 0.6098,
"mean_token_accuracy": 0.8324897587299347,
"num_tokens": 911523735.0,
"step": 10290
},
{
"entropy": 1.01953125,
"epoch": 1.4077086038406341,
"grad_norm": 0.927594656570175,
"learning_rate": 1.5270536846554884e-06,
"loss": 0.7097,
"mean_token_accuracy": 0.8104441314935684,
"num_tokens": 912401560.0,
"step": 10300
},
{
"entropy": 0.97109375,
"epoch": 1.409075377571243,
"grad_norm": 0.8958724160591066,
"learning_rate": 1.5235310694659717e-06,
"loss": 0.7231,
"mean_token_accuracy": 0.8046873867511749,
"num_tokens": 913302312.0,
"step": 10310
},
{
"entropy": 0.92421875,
"epoch": 1.410442151301852,
"grad_norm": 0.8257720804123604,
"learning_rate": 1.5200084542764548e-06,
"loss": 0.6177,
"mean_token_accuracy": 0.8280565023422242,
"num_tokens": 914208968.0,
"step": 10320
},
{
"entropy": 0.9640625,
"epoch": 1.4118089250324608,
"grad_norm": 0.8362244806494594,
"learning_rate": 1.5164858390869384e-06,
"loss": 0.6427,
"mean_token_accuracy": 0.821956816315651,
"num_tokens": 915092764.0,
"step": 10330
},
{
"entropy": 1.015625,
"epoch": 1.4131756987630697,
"grad_norm": 0.8058399879846276,
"learning_rate": 1.5129632238974215e-06,
"loss": 0.7441,
"mean_token_accuracy": 0.8028040051460266,
"num_tokens": 915947203.0,
"step": 10340
},
{
"entropy": 0.9595703125,
"epoch": 1.4145424724936786,
"grad_norm": 0.7084314827629299,
"learning_rate": 1.509440608707905e-06,
"loss": 0.6411,
"mean_token_accuracy": 0.8237459361553192,
"num_tokens": 916862972.0,
"step": 10350
},
{
"entropy": 0.99765625,
"epoch": 1.4159092462242875,
"grad_norm": 0.7666839722592226,
"learning_rate": 1.5059179935183882e-06,
"loss": 0.6812,
"mean_token_accuracy": 0.8129668891429901,
"num_tokens": 917739748.0,
"step": 10360
},
{
"entropy": 0.963671875,
"epoch": 1.4172760199548966,
"grad_norm": 0.7679282577416654,
"learning_rate": 1.5023953783288713e-06,
"loss": 0.6602,
"mean_token_accuracy": 0.8187398850917816,
"num_tokens": 918638904.0,
"step": 10370
},
{
"entropy": 0.949609375,
"epoch": 1.4186427936855055,
"grad_norm": 0.7554386614720418,
"learning_rate": 1.4988727631393549e-06,
"loss": 0.6789,
"mean_token_accuracy": 0.8145727455615998,
"num_tokens": 919508022.0,
"step": 10380
},
{
"entropy": 0.9919921875,
"epoch": 1.4200095674161144,
"grad_norm": 0.849973114433764,
"learning_rate": 1.495350147949838e-06,
"loss": 0.6985,
"mean_token_accuracy": 0.8113285273313522,
"num_tokens": 920421270.0,
"step": 10390
},
{
"entropy": 0.9544921875,
"epoch": 1.4213763411467233,
"grad_norm": 0.7876775134606836,
"learning_rate": 1.4918275327603213e-06,
"loss": 0.6335,
"mean_token_accuracy": 0.82599456012249,
"num_tokens": 921268963.0,
"step": 10400
},
{
"entropy": 0.93359375,
"epoch": 1.4227431148773322,
"grad_norm": 0.7053054362065769,
"learning_rate": 1.4883049175708047e-06,
"loss": 0.6551,
"mean_token_accuracy": 0.8209844261407853,
"num_tokens": 922110156.0,
"step": 10410
},
{
"entropy": 0.9728515625,
"epoch": 1.424109888607941,
"grad_norm": 0.7791736424613706,
"learning_rate": 1.484782302381288e-06,
"loss": 0.6929,
"mean_token_accuracy": 0.8120492190122605,
"num_tokens": 923000326.0,
"step": 10420
},
{
"entropy": 0.9611328125,
"epoch": 1.42547666233855,
"grad_norm": 0.8837147447246653,
"learning_rate": 1.4812596871917711e-06,
"loss": 0.6469,
"mean_token_accuracy": 0.8230437755584716,
"num_tokens": 923866880.0,
"step": 10430
},
{
"entropy": 0.947265625,
"epoch": 1.4268434360691589,
"grad_norm": 0.7293642847382654,
"learning_rate": 1.4777370720022547e-06,
"loss": 0.6605,
"mean_token_accuracy": 0.8191932022571564,
"num_tokens": 924736744.0,
"step": 10440
},
{
"entropy": 0.937890625,
"epoch": 1.4282102097997678,
"grad_norm": 0.7986292291296929,
"learning_rate": 1.4742144568127378e-06,
"loss": 0.6678,
"mean_token_accuracy": 0.8162862867116928,
"num_tokens": 925648312.0,
"step": 10450
},
{
"entropy": 0.9416015625,
"epoch": 1.4295769835303767,
"grad_norm": 0.7510696704254763,
"learning_rate": 1.4706918416232213e-06,
"loss": 0.6647,
"mean_token_accuracy": 0.8184339493513108,
"num_tokens": 926529038.0,
"step": 10460
},
{
"entropy": 0.9787109375,
"epoch": 1.4309437572609855,
"grad_norm": 0.7132958484431231,
"learning_rate": 1.4671692264337045e-06,
"loss": 0.6698,
"mean_token_accuracy": 0.8174517601728439,
"num_tokens": 927441768.0,
"step": 10470
},
{
"entropy": 0.9556640625,
"epoch": 1.4323105309915944,
"grad_norm": 0.6862976875372235,
"learning_rate": 1.4636466112441878e-06,
"loss": 0.6838,
"mean_token_accuracy": 0.8152939230203629,
"num_tokens": 928358695.0,
"step": 10480
},
{
"entropy": 0.9466796875,
"epoch": 1.4336773047222033,
"grad_norm": 0.7288002383168906,
"learning_rate": 1.4601239960546711e-06,
"loss": 0.651,
"mean_token_accuracy": 0.821949478983879,
"num_tokens": 929240770.0,
"step": 10490
},
{
"entropy": 0.9794921875,
"epoch": 1.4350440784528122,
"grad_norm": 0.7279164570622747,
"learning_rate": 1.4566013808651545e-06,
"loss": 0.6581,
"mean_token_accuracy": 0.8188939332962036,
"num_tokens": 930170684.0,
"step": 10500
},
{
"entropy": 0.9197265625,
"epoch": 1.4364108521834211,
"grad_norm": 0.8295558276241062,
"learning_rate": 1.4530787656756376e-06,
"loss": 0.6229,
"mean_token_accuracy": 0.8276139855384826,
"num_tokens": 931088028.0,
"step": 10510
},
{
"entropy": 0.961328125,
"epoch": 1.43777762591403,
"grad_norm": 0.7901095474889985,
"learning_rate": 1.4495561504861211e-06,
"loss": 0.6617,
"mean_token_accuracy": 0.8176721006631851,
"num_tokens": 932014724.0,
"step": 10520
},
{
"entropy": 0.973046875,
"epoch": 1.439144399644639,
"grad_norm": 0.7690614980991484,
"learning_rate": 1.4460335352966043e-06,
"loss": 0.6622,
"mean_token_accuracy": 0.8195680171251297,
"num_tokens": 932888906.0,
"step": 10530
},
{
"entropy": 1.0119140625,
"epoch": 1.4405111733752478,
"grad_norm": 0.7962004620121125,
"learning_rate": 1.4425109201070876e-06,
"loss": 0.7585,
"mean_token_accuracy": 0.7984862864017487,
"num_tokens": 933851911.0,
"step": 10540
},
{
"entropy": 0.9646484375,
"epoch": 1.4418779471058567,
"grad_norm": 0.8363004637282712,
"learning_rate": 1.438988304917571e-06,
"loss": 0.6474,
"mean_token_accuracy": 0.8225688695907593,
"num_tokens": 934766976.0,
"step": 10550
},
{
"entropy": 0.96953125,
"epoch": 1.4432447208364656,
"grad_norm": 0.7374827562627625,
"learning_rate": 1.4354656897280543e-06,
"loss": 0.6975,
"mean_token_accuracy": 0.8124494910240173,
"num_tokens": 935658392.0,
"step": 10560
},
{
"entropy": 0.958203125,
"epoch": 1.4446114945670745,
"grad_norm": 0.7963677153871268,
"learning_rate": 1.4319430745385376e-06,
"loss": 0.657,
"mean_token_accuracy": 0.8188937366008758,
"num_tokens": 936532519.0,
"step": 10570
},
{
"entropy": 0.9728515625,
"epoch": 1.4459782682976834,
"grad_norm": 0.7431137266305762,
"learning_rate": 1.428420459349021e-06,
"loss": 0.6582,
"mean_token_accuracy": 0.8216688752174377,
"num_tokens": 937413962.0,
"step": 10580
},
{
"entropy": 0.923046875,
"epoch": 1.4473450420282923,
"grad_norm": 0.8468566977907768,
"learning_rate": 1.424897844159504e-06,
"loss": 0.6119,
"mean_token_accuracy": 0.8305918484926224,
"num_tokens": 938298992.0,
"step": 10590
},
{
"entropy": 0.9552734375,
"epoch": 1.4487118157589012,
"grad_norm": 0.8630060059167414,
"learning_rate": 1.4213752289699876e-06,
"loss": 0.642,
"mean_token_accuracy": 0.824438464641571,
"num_tokens": 939178986.0,
"step": 10600
},
{
"entropy": 0.919140625,
"epoch": 1.45007858948951,
"grad_norm": 0.6763736517485551,
"learning_rate": 1.4178526137804707e-06,
"loss": 0.6404,
"mean_token_accuracy": 0.8247470617294311,
"num_tokens": 940050901.0,
"step": 10610
},
{
"entropy": 0.9583984375,
"epoch": 1.451445363220119,
"grad_norm": 0.7956552817560858,
"learning_rate": 1.4143299985909538e-06,
"loss": 0.6247,
"mean_token_accuracy": 0.827894726395607,
"num_tokens": 940916956.0,
"step": 10620
},
{
"entropy": 0.9421875,
"epoch": 1.4528121369507279,
"grad_norm": 0.80305867860912,
"learning_rate": 1.4108073834014374e-06,
"loss": 0.639,
"mean_token_accuracy": 0.8238210022449494,
"num_tokens": 941829889.0,
"step": 10630
},
{
"entropy": 0.930078125,
"epoch": 1.4541789106813368,
"grad_norm": 0.8416207180417032,
"learning_rate": 1.4072847682119205e-06,
"loss": 0.6473,
"mean_token_accuracy": 0.8219980627298356,
"num_tokens": 942736711.0,
"step": 10640
},
{
"entropy": 0.9396484375,
"epoch": 1.4555456844119457,
"grad_norm": 0.7686098359053929,
"learning_rate": 1.4037621530224039e-06,
"loss": 0.6602,
"mean_token_accuracy": 0.8194265484809875,
"num_tokens": 943671086.0,
"step": 10650
},
{
"entropy": 0.95546875,
"epoch": 1.4569124581425545,
"grad_norm": 0.7369002972566466,
"learning_rate": 1.4002395378328872e-06,
"loss": 0.6051,
"mean_token_accuracy": 0.8304669946432114,
"num_tokens": 944598233.0,
"step": 10660
},
{
"entropy": 0.94453125,
"epoch": 1.4582792318731634,
"grad_norm": 0.8033186316262746,
"learning_rate": 1.3967169226433705e-06,
"loss": 0.6283,
"mean_token_accuracy": 0.8274977445602417,
"num_tokens": 945480002.0,
"step": 10670
},
{
"entropy": 0.9591796875,
"epoch": 1.4596460056037723,
"grad_norm": 0.7715752225739853,
"learning_rate": 1.393194307453854e-06,
"loss": 0.6806,
"mean_token_accuracy": 0.8151980906724929,
"num_tokens": 946388987.0,
"step": 10680
},
{
"entropy": 0.9833984375,
"epoch": 1.4610127793343812,
"grad_norm": 0.8134642620776632,
"learning_rate": 1.3896716922643372e-06,
"loss": 0.6907,
"mean_token_accuracy": 0.8110586494207382,
"num_tokens": 947253180.0,
"step": 10690
},
{
"entropy": 0.9392578125,
"epoch": 1.4623795530649901,
"grad_norm": 0.7313028152106449,
"learning_rate": 1.3861490770748203e-06,
"loss": 0.6321,
"mean_token_accuracy": 0.8257853239774704,
"num_tokens": 948160131.0,
"step": 10700
},
{
"entropy": 0.97265625,
"epoch": 1.463746326795599,
"grad_norm": 0.8768466894213347,
"learning_rate": 1.3826264618853039e-06,
"loss": 0.671,
"mean_token_accuracy": 0.8149594694375992,
"num_tokens": 949000608.0,
"step": 10710
},
{
"entropy": 0.98984375,
"epoch": 1.465113100526208,
"grad_norm": 0.8005119170306007,
"learning_rate": 1.379103846695787e-06,
"loss": 0.6841,
"mean_token_accuracy": 0.8138638973236084,
"num_tokens": 949951490.0,
"step": 10720
},
{
"entropy": 0.9783203125,
"epoch": 1.4664798742568168,
"grad_norm": 0.7783912548005745,
"learning_rate": 1.3755812315062703e-06,
"loss": 0.6703,
"mean_token_accuracy": 0.8163774311542511,
"num_tokens": 950835351.0,
"step": 10730
},
{
"entropy": 0.9427734375,
"epoch": 1.4678466479874257,
"grad_norm": 0.7941900778902427,
"learning_rate": 1.3720586163167537e-06,
"loss": 0.6662,
"mean_token_accuracy": 0.817992776632309,
"num_tokens": 951720959.0,
"step": 10740
},
{
"entropy": 0.9859375,
"epoch": 1.4692134217180346,
"grad_norm": 0.8704632161702176,
"learning_rate": 1.368536001127237e-06,
"loss": 0.6994,
"mean_token_accuracy": 0.8154178112745285,
"num_tokens": 952584602.0,
"step": 10750
},
{
"entropy": 0.935546875,
"epoch": 1.4705801954486435,
"grad_norm": 0.7308370497966208,
"learning_rate": 1.3650133859377201e-06,
"loss": 0.6626,
"mean_token_accuracy": 0.8191232204437255,
"num_tokens": 953459805.0,
"step": 10760
},
{
"entropy": 0.959375,
"epoch": 1.4719469691792524,
"grad_norm": 0.7918669461353366,
"learning_rate": 1.3614907707482037e-06,
"loss": 0.6659,
"mean_token_accuracy": 0.8175280809402465,
"num_tokens": 954324502.0,
"step": 10770
},
{
"entropy": 1.037890625,
"epoch": 1.4733137429098613,
"grad_norm": 0.8160796311332562,
"learning_rate": 1.3579681555586868e-06,
"loss": 0.71,
"mean_token_accuracy": 0.8104464709758759,
"num_tokens": 955233860.0,
"step": 10780
},
{
"entropy": 0.9865234375,
"epoch": 1.4746805166404702,
"grad_norm": 0.6891525925891543,
"learning_rate": 1.3544455403691703e-06,
"loss": 0.669,
"mean_token_accuracy": 0.8181988954544067,
"num_tokens": 956148137.0,
"step": 10790
},
{
"entropy": 0.943359375,
"epoch": 1.476047290371079,
"grad_norm": 0.7614921039389079,
"learning_rate": 1.3509229251796535e-06,
"loss": 0.6354,
"mean_token_accuracy": 0.8241228997707367,
"num_tokens": 956998510.0,
"step": 10800
},
{
"entropy": 0.967578125,
"epoch": 1.477414064101688,
"grad_norm": 0.7399578604093533,
"learning_rate": 1.3474003099901368e-06,
"loss": 0.6682,
"mean_token_accuracy": 0.8173330903053284,
"num_tokens": 957859645.0,
"step": 10810
},
{
"entropy": 0.9416015625,
"epoch": 1.4787808378322969,
"grad_norm": 0.7169440099616258,
"learning_rate": 1.3438776948006201e-06,
"loss": 0.6371,
"mean_token_accuracy": 0.823886376619339,
"num_tokens": 958729483.0,
"step": 10820
},
{
"entropy": 0.919140625,
"epoch": 1.4801476115629058,
"grad_norm": 0.7582233780412316,
"learning_rate": 1.3403550796111035e-06,
"loss": 0.5886,
"mean_token_accuracy": 0.8344971776008606,
"num_tokens": 959617772.0,
"step": 10830
},
{
"entropy": 0.9583984375,
"epoch": 1.4815143852935146,
"grad_norm": 0.7746895158614466,
"learning_rate": 1.3368324644215866e-06,
"loss": 0.6593,
"mean_token_accuracy": 0.8208555668592453,
"num_tokens": 960496565.0,
"step": 10840
},
{
"entropy": 0.928515625,
"epoch": 1.4828811590241235,
"grad_norm": 0.7144434302823176,
"learning_rate": 1.3333098492320701e-06,
"loss": 0.6454,
"mean_token_accuracy": 0.8223114281892776,
"num_tokens": 961390449.0,
"step": 10850
},
{
"entropy": 0.9783203125,
"epoch": 1.4842479327547324,
"grad_norm": 0.8813163232515298,
"learning_rate": 1.3297872340425533e-06,
"loss": 0.7065,
"mean_token_accuracy": 0.8091205298900604,
"num_tokens": 962243695.0,
"step": 10860
},
{
"entropy": 0.9802734375,
"epoch": 1.4856147064853413,
"grad_norm": 0.8045625087258408,
"learning_rate": 1.3262646188530364e-06,
"loss": 0.7041,
"mean_token_accuracy": 0.8118561625480651,
"num_tokens": 963171977.0,
"step": 10870
},
{
"entropy": 0.9587890625,
"epoch": 1.4869814802159502,
"grad_norm": 0.6890845277255891,
"learning_rate": 1.32274200366352e-06,
"loss": 0.671,
"mean_token_accuracy": 0.8183073937892914,
"num_tokens": 964055991.0,
"step": 10880
},
{
"entropy": 0.9427734375,
"epoch": 1.4883482539465591,
"grad_norm": 0.742874264315451,
"learning_rate": 1.319219388474003e-06,
"loss": 0.6537,
"mean_token_accuracy": 0.8223891139030457,
"num_tokens": 964895585.0,
"step": 10890
},
{
"entropy": 0.984375,
"epoch": 1.489715027677168,
"grad_norm": 0.7892788831681387,
"learning_rate": 1.3156967732844866e-06,
"loss": 0.6949,
"mean_token_accuracy": 0.8103797435760498,
"num_tokens": 965796547.0,
"step": 10900
},
{
"entropy": 0.9271484375,
"epoch": 1.491081801407777,
"grad_norm": 0.816100826514434,
"learning_rate": 1.31217415809497e-06,
"loss": 0.6087,
"mean_token_accuracy": 0.8307056248188018,
"num_tokens": 966706376.0,
"step": 10910
},
{
"entropy": 0.99453125,
"epoch": 1.4924485751383858,
"grad_norm": 0.7688967805924429,
"learning_rate": 1.308651542905453e-06,
"loss": 0.7141,
"mean_token_accuracy": 0.8071741610765457,
"num_tokens": 967641987.0,
"step": 10920
},
{
"entropy": 0.931640625,
"epoch": 1.4938153488689947,
"grad_norm": 0.8364485159886648,
"learning_rate": 1.3051289277159366e-06,
"loss": 0.6435,
"mean_token_accuracy": 0.821866637468338,
"num_tokens": 968513222.0,
"step": 10930
},
{
"entropy": 0.950390625,
"epoch": 1.4951821225996036,
"grad_norm": 0.8630278113676902,
"learning_rate": 1.3016063125264197e-06,
"loss": 0.6828,
"mean_token_accuracy": 0.8179793894290924,
"num_tokens": 969379276.0,
"step": 10940
},
{
"entropy": 0.9748046875,
"epoch": 1.4965488963302125,
"grad_norm": 0.8723764880489543,
"learning_rate": 1.2980836973369029e-06,
"loss": 0.7024,
"mean_token_accuracy": 0.8086298435926438,
"num_tokens": 970233883.0,
"step": 10950
},
{
"entropy": 0.93359375,
"epoch": 1.4979156700608214,
"grad_norm": 0.844391898182339,
"learning_rate": 1.2945610821473864e-06,
"loss": 0.6719,
"mean_token_accuracy": 0.8176459312438965,
"num_tokens": 971089571.0,
"step": 10960
},
{
"entropy": 0.9650390625,
"epoch": 1.4992824437914303,
"grad_norm": 0.8024233466306778,
"learning_rate": 1.2910384669578695e-06,
"loss": 0.6187,
"mean_token_accuracy": 0.8291327089071274,
"num_tokens": 971969845.0,
"step": 10970
},
{
"entropy": 0.9349609375,
"epoch": 1.5006492175220392,
"grad_norm": 0.8310944880312985,
"learning_rate": 1.2875158517683529e-06,
"loss": 0.6549,
"mean_token_accuracy": 0.8222586393356324,
"num_tokens": 972830400.0,
"step": 10980
},
{
"entropy": 0.9724609375,
"epoch": 1.502015991252648,
"grad_norm": 0.8522822496714261,
"learning_rate": 1.2839932365788362e-06,
"loss": 0.7042,
"mean_token_accuracy": 0.8085721492767334,
"num_tokens": 973711071.0,
"step": 10990
},
{
"entropy": 0.9419921875,
"epoch": 1.503382764983257,
"grad_norm": 0.7049955473222205,
"learning_rate": 1.2804706213893195e-06,
"loss": 0.6576,
"mean_token_accuracy": 0.8194338798522949,
"num_tokens": 974615703.0,
"step": 11000
},
{
"entropy": 0.9966796875,
"epoch": 1.5047495387138659,
"grad_norm": 0.7463186134392191,
"learning_rate": 1.2769480061998029e-06,
"loss": 0.686,
"mean_token_accuracy": 0.8152054220438003,
"num_tokens": 975506409.0,
"step": 11010
},
{
"entropy": 0.9427734375,
"epoch": 1.5061163124444747,
"grad_norm": 0.7375404332096999,
"learning_rate": 1.2734253910102862e-06,
"loss": 0.6414,
"mean_token_accuracy": 0.8236503362655639,
"num_tokens": 976439918.0,
"step": 11020
},
{
"entropy": 0.96328125,
"epoch": 1.5074830861750836,
"grad_norm": 0.7566781458681471,
"learning_rate": 1.2699027758207693e-06,
"loss": 0.6502,
"mean_token_accuracy": 0.8225803047418594,
"num_tokens": 977319557.0,
"step": 11030
},
{
"entropy": 1.0044921875,
"epoch": 1.5088498599056925,
"grad_norm": 0.771129134254945,
"learning_rate": 1.2663801606312529e-06,
"loss": 0.7236,
"mean_token_accuracy": 0.8035033345222473,
"num_tokens": 978259078.0,
"step": 11040
},
{
"entropy": 0.967578125,
"epoch": 1.5102166336363014,
"grad_norm": 0.8432760640534822,
"learning_rate": 1.262857545441736e-06,
"loss": 0.6325,
"mean_token_accuracy": 0.8261741250753403,
"num_tokens": 979110803.0,
"step": 11050
},
{
"entropy": 0.959765625,
"epoch": 1.5115834073669103,
"grad_norm": 0.7827302780892468,
"learning_rate": 1.2593349302522193e-06,
"loss": 0.6583,
"mean_token_accuracy": 0.8190777778625489,
"num_tokens": 979995472.0,
"step": 11060
},
{
"entropy": 0.95859375,
"epoch": 1.5129501810975192,
"grad_norm": 0.8513674120590754,
"learning_rate": 1.2558123150627027e-06,
"loss": 0.6398,
"mean_token_accuracy": 0.8246334701776504,
"num_tokens": 980896411.0,
"step": 11070
},
{
"entropy": 0.9759765625,
"epoch": 1.5143169548281281,
"grad_norm": 0.9014625570699559,
"learning_rate": 1.252289699873186e-06,
"loss": 0.7078,
"mean_token_accuracy": 0.809856903553009,
"num_tokens": 981782545.0,
"step": 11080
},
{
"entropy": 0.9673828125,
"epoch": 1.515683728558737,
"grad_norm": 0.8835258080869253,
"learning_rate": 1.2487670846836693e-06,
"loss": 0.6817,
"mean_token_accuracy": 0.8119596362113952,
"num_tokens": 982689516.0,
"step": 11090
},
{
"entropy": 1.0013671875,
"epoch": 1.517050502289346,
"grad_norm": 0.8924543768381761,
"learning_rate": 1.2452444694941527e-06,
"loss": 0.7634,
"mean_token_accuracy": 0.7974960654973984,
"num_tokens": 983535293.0,
"step": 11100
},
{
"entropy": 0.9689453125,
"epoch": 1.5184172760199548,
"grad_norm": 0.6564400950449091,
"learning_rate": 1.2417218543046358e-06,
"loss": 0.6768,
"mean_token_accuracy": 0.8165583819150924,
"num_tokens": 984435788.0,
"step": 11110
},
{
"entropy": 0.9515625,
"epoch": 1.5197840497505637,
"grad_norm": 0.8045550683731021,
"learning_rate": 1.2381992391151191e-06,
"loss": 0.6789,
"mean_token_accuracy": 0.8166078746318817,
"num_tokens": 985303807.0,
"step": 11120
},
{
"entropy": 0.9591796875,
"epoch": 1.5211508234811726,
"grad_norm": 0.7622056381173977,
"learning_rate": 1.2346766239256025e-06,
"loss": 0.681,
"mean_token_accuracy": 0.8134086608886719,
"num_tokens": 986218767.0,
"step": 11130
},
{
"entropy": 0.9513671875,
"epoch": 1.5225175972117815,
"grad_norm": 0.767739416746436,
"learning_rate": 1.2311540087360858e-06,
"loss": 0.6409,
"mean_token_accuracy": 0.8231266885995865,
"num_tokens": 987143778.0,
"step": 11140
},
{
"entropy": 0.94609375,
"epoch": 1.5238843709423904,
"grad_norm": 0.7797105541212462,
"learning_rate": 1.2276313935465691e-06,
"loss": 0.6373,
"mean_token_accuracy": 0.8248185783624649,
"num_tokens": 988031175.0,
"step": 11150
},
{
"entropy": 0.94609375,
"epoch": 1.5252511446729993,
"grad_norm": 0.7010049838450927,
"learning_rate": 1.2241087783570525e-06,
"loss": 0.6686,
"mean_token_accuracy": 0.8213648200035095,
"num_tokens": 988923325.0,
"step": 11160
},
{
"entropy": 0.972265625,
"epoch": 1.5266179184036082,
"grad_norm": 0.7034130248841802,
"learning_rate": 1.2205861631675358e-06,
"loss": 0.663,
"mean_token_accuracy": 0.8174349874258041,
"num_tokens": 989811919.0,
"step": 11170
},
{
"entropy": 1.00859375,
"epoch": 1.527984692134217,
"grad_norm": 0.9316870880370537,
"learning_rate": 1.217063547978019e-06,
"loss": 0.7004,
"mean_token_accuracy": 0.8092321068048477,
"num_tokens": 990679973.0,
"step": 11180
},
{
"entropy": 0.958984375,
"epoch": 1.529351465864826,
"grad_norm": 0.7462337797611746,
"learning_rate": 1.2135409327885023e-06,
"loss": 0.6673,
"mean_token_accuracy": 0.8188110262155532,
"num_tokens": 991562442.0,
"step": 11190
},
{
"entropy": 0.9251953125,
"epoch": 1.5307182395954348,
"grad_norm": 0.7453380687394014,
"learning_rate": 1.2100183175989856e-06,
"loss": 0.6289,
"mean_token_accuracy": 0.8250941693782806,
"num_tokens": 992429060.0,
"step": 11200
},
{
"entropy": 0.91953125,
"epoch": 1.5320850133260437,
"grad_norm": 0.764580432720881,
"learning_rate": 1.206495702409469e-06,
"loss": 0.6344,
"mean_token_accuracy": 0.8245521873235703,
"num_tokens": 993299078.0,
"step": 11210
},
{
"entropy": 0.921484375,
"epoch": 1.5334517870566526,
"grad_norm": 0.8312077045594083,
"learning_rate": 1.202973087219952e-06,
"loss": 0.632,
"mean_token_accuracy": 0.8241001218557358,
"num_tokens": 994183754.0,
"step": 11220
},
{
"entropy": 0.933984375,
"epoch": 1.5348185607872615,
"grad_norm": 0.8219386529951143,
"learning_rate": 1.1994504720304354e-06,
"loss": 0.6775,
"mean_token_accuracy": 0.8163053661584854,
"num_tokens": 995090732.0,
"step": 11230
},
{
"entropy": 0.9373046875,
"epoch": 1.5361853345178704,
"grad_norm": 0.7897471619864396,
"learning_rate": 1.1959278568409187e-06,
"loss": 0.6421,
"mean_token_accuracy": 0.8216757148504257,
"num_tokens": 996004572.0,
"step": 11240
},
{
"entropy": 0.93984375,
"epoch": 1.5375521082484793,
"grad_norm": 0.763102361691227,
"learning_rate": 1.192405241651402e-06,
"loss": 0.636,
"mean_token_accuracy": 0.8234814167022705,
"num_tokens": 996890955.0,
"step": 11250
},
{
"entropy": 0.9505859375,
"epoch": 1.5389188819790882,
"grad_norm": 0.8280864568838051,
"learning_rate": 1.1888826264618854e-06,
"loss": 0.6623,
"mean_token_accuracy": 0.8177784532308578,
"num_tokens": 997769284.0,
"step": 11260
},
{
"entropy": 0.9482421875,
"epoch": 1.540285655709697,
"grad_norm": 0.7854098427558845,
"learning_rate": 1.1853600112723687e-06,
"loss": 0.6561,
"mean_token_accuracy": 0.8215617895126343,
"num_tokens": 998677050.0,
"step": 11270
},
{
"entropy": 0.940234375,
"epoch": 1.541652429440306,
"grad_norm": 0.821513428140394,
"learning_rate": 1.181837396082852e-06,
"loss": 0.6315,
"mean_token_accuracy": 0.8254645735025405,
"num_tokens": 999575420.0,
"step": 11280
},
{
"entropy": 0.9271484375,
"epoch": 1.543019203170915,
"grad_norm": 0.7800071272943234,
"learning_rate": 1.1783147808933352e-06,
"loss": 0.6659,
"mean_token_accuracy": 0.8179851233959198,
"num_tokens": 1000467865.0,
"step": 11290
},
{
"entropy": 1.022265625,
"epoch": 1.5443859769015238,
"grad_norm": 0.8109902736763731,
"learning_rate": 1.1747921657038185e-06,
"loss": 0.716,
"mean_token_accuracy": 0.8073104918003082,
"num_tokens": 1001343857.0,
"step": 11300
},
{
"entropy": 0.93046875,
"epoch": 1.5457527506321327,
"grad_norm": 0.7330754174562256,
"learning_rate": 1.1712695505143019e-06,
"loss": 0.611,
"mean_token_accuracy": 0.8295418322086334,
"num_tokens": 1002206031.0,
"step": 11310
},
{
"entropy": 0.9662109375,
"epoch": 1.5471195243627416,
"grad_norm": 0.7420097334075098,
"learning_rate": 1.1677469353247852e-06,
"loss": 0.6461,
"mean_token_accuracy": 0.8235206723213195,
"num_tokens": 1003046032.0,
"step": 11320
},
{
"entropy": 0.96875,
"epoch": 1.5484862980933507,
"grad_norm": 0.7289168419440175,
"learning_rate": 1.1642243201352685e-06,
"loss": 0.6928,
"mean_token_accuracy": 0.8102324843406677,
"num_tokens": 1003968005.0,
"step": 11330
},
{
"entropy": 0.9423828125,
"epoch": 1.5498530718239596,
"grad_norm": 0.8001463544024926,
"learning_rate": 1.1607017049457519e-06,
"loss": 0.6535,
"mean_token_accuracy": 0.820810866355896,
"num_tokens": 1004874914.0,
"step": 11340
},
{
"entropy": 0.9765625,
"epoch": 1.5512198455545685,
"grad_norm": 0.7638892790631915,
"learning_rate": 1.1571790897562352e-06,
"loss": 0.7002,
"mean_token_accuracy": 0.8118520349264144,
"num_tokens": 1005735910.0,
"step": 11350
},
{
"entropy": 0.9265625,
"epoch": 1.5525866192851774,
"grad_norm": 0.785258479038859,
"learning_rate": 1.1536564745667183e-06,
"loss": 0.6543,
"mean_token_accuracy": 0.8207921653985977,
"num_tokens": 1006600794.0,
"step": 11360
},
{
"entropy": 0.97109375,
"epoch": 1.5539533930157863,
"grad_norm": 0.8431834495696394,
"learning_rate": 1.1501338593772017e-06,
"loss": 0.6745,
"mean_token_accuracy": 0.8166909456253052,
"num_tokens": 1007467493.0,
"step": 11370
},
{
"entropy": 0.9755859375,
"epoch": 1.5553201667463952,
"grad_norm": 0.8332416860917695,
"learning_rate": 1.146611244187685e-06,
"loss": 0.6786,
"mean_token_accuracy": 0.8175596565008163,
"num_tokens": 1008379489.0,
"step": 11380
},
{
"entropy": 0.9587890625,
"epoch": 1.556686940477004,
"grad_norm": 0.7959620307807257,
"learning_rate": 1.1430886289981683e-06,
"loss": 0.6887,
"mean_token_accuracy": 0.8133181989192962,
"num_tokens": 1009259295.0,
"step": 11390
},
{
"entropy": 0.9388671875,
"epoch": 1.558053714207613,
"grad_norm": 0.8480983373214046,
"learning_rate": 1.1395660138086517e-06,
"loss": 0.6261,
"mean_token_accuracy": 0.826907280087471,
"num_tokens": 1010088513.0,
"step": 11400
},
{
"entropy": 0.922265625,
"epoch": 1.5594204879382219,
"grad_norm": 0.7943333116902604,
"learning_rate": 1.136043398619135e-06,
"loss": 0.6277,
"mean_token_accuracy": 0.8270593762397767,
"num_tokens": 1010964392.0,
"step": 11410
},
{
"entropy": 1.0060546875,
"epoch": 1.5607872616688307,
"grad_norm": 0.908292947413652,
"learning_rate": 1.1325207834296184e-06,
"loss": 0.7132,
"mean_token_accuracy": 0.8079800069332123,
"num_tokens": 1011863256.0,
"step": 11420
},
{
"entropy": 0.9357421875,
"epoch": 1.5621540353994396,
"grad_norm": 0.7507959579323996,
"learning_rate": 1.1289981682401017e-06,
"loss": 0.63,
"mean_token_accuracy": 0.8250195533037186,
"num_tokens": 1012715405.0,
"step": 11430
},
{
"entropy": 0.9669921875,
"epoch": 1.5635208091300485,
"grad_norm": 0.7505175435777847,
"learning_rate": 1.1254755530505848e-06,
"loss": 0.6578,
"mean_token_accuracy": 0.8224657893180847,
"num_tokens": 1013602800.0,
"step": 11440
},
{
"entropy": 0.9759765625,
"epoch": 1.5648875828606574,
"grad_norm": 0.715406844879015,
"learning_rate": 1.1219529378610681e-06,
"loss": 0.6957,
"mean_token_accuracy": 0.8115205794572831,
"num_tokens": 1014477602.0,
"step": 11450
},
{
"entropy": 0.970703125,
"epoch": 1.5662543565912663,
"grad_norm": 0.826698961243284,
"learning_rate": 1.1184303226715515e-06,
"loss": 0.6734,
"mean_token_accuracy": 0.8161875218153,
"num_tokens": 1015376894.0,
"step": 11460
},
{
"entropy": 0.979296875,
"epoch": 1.5676211303218752,
"grad_norm": 0.8043796316563955,
"learning_rate": 1.1149077074820346e-06,
"loss": 0.6838,
"mean_token_accuracy": 0.8133771777153015,
"num_tokens": 1016269694.0,
"step": 11470
},
{
"entropy": 1.01015625,
"epoch": 1.5689879040524841,
"grad_norm": 0.9015549433901766,
"learning_rate": 1.111385092292518e-06,
"loss": 0.7208,
"mean_token_accuracy": 0.807605329155922,
"num_tokens": 1017136473.0,
"step": 11480
},
{
"entropy": 0.944140625,
"epoch": 1.570354677783093,
"grad_norm": 0.7194342785588438,
"learning_rate": 1.1078624771030013e-06,
"loss": 0.6634,
"mean_token_accuracy": 0.8191560268402099,
"num_tokens": 1018025066.0,
"step": 11490
},
{
"entropy": 0.97421875,
"epoch": 1.571721451513702,
"grad_norm": 0.775656168865284,
"learning_rate": 1.1043398619134846e-06,
"loss": 0.685,
"mean_token_accuracy": 0.8139021992683411,
"num_tokens": 1018946433.0,
"step": 11500
},
{
"entropy": 0.97265625,
"epoch": 1.5730882252443108,
"grad_norm": 0.7056612670251936,
"learning_rate": 1.100817246723968e-06,
"loss": 0.6735,
"mean_token_accuracy": 0.8165207028388977,
"num_tokens": 1019818403.0,
"step": 11510
},
{
"entropy": 0.946484375,
"epoch": 1.5744549989749197,
"grad_norm": 0.7281449490916801,
"learning_rate": 1.0972946315344513e-06,
"loss": 0.6636,
"mean_token_accuracy": 0.8180854260921478,
"num_tokens": 1020734420.0,
"step": 11520
},
{
"entropy": 0.9396484375,
"epoch": 1.5758217727055286,
"grad_norm": 0.7646424109541311,
"learning_rate": 1.0937720163449346e-06,
"loss": 0.6553,
"mean_token_accuracy": 0.8230131804943085,
"num_tokens": 1021610608.0,
"step": 11530
},
{
"entropy": 0.966015625,
"epoch": 1.5771885464361375,
"grad_norm": 0.7356226785869285,
"learning_rate": 1.0902494011554177e-06,
"loss": 0.6822,
"mean_token_accuracy": 0.8132320880889893,
"num_tokens": 1022507878.0,
"step": 11540
},
{
"entropy": 0.9384765625,
"epoch": 1.5785553201667464,
"grad_norm": 0.7734311627471704,
"learning_rate": 1.086726785965901e-06,
"loss": 0.6396,
"mean_token_accuracy": 0.8243215322494507,
"num_tokens": 1023437587.0,
"step": 11550
},
{
"entropy": 0.9884765625,
"epoch": 1.5799220938973553,
"grad_norm": 0.9112030563977418,
"learning_rate": 1.0832041707763844e-06,
"loss": 0.7072,
"mean_token_accuracy": 0.8086029946804046,
"num_tokens": 1024328404.0,
"step": 11560
},
{
"entropy": 0.9740234375,
"epoch": 1.5812888676279642,
"grad_norm": 0.7345102094817575,
"learning_rate": 1.0796815555868678e-06,
"loss": 0.6878,
"mean_token_accuracy": 0.8158864617347718,
"num_tokens": 1025278391.0,
"step": 11570
},
{
"entropy": 0.983984375,
"epoch": 1.582655641358573,
"grad_norm": 0.7152554063888259,
"learning_rate": 1.076158940397351e-06,
"loss": 0.6918,
"mean_token_accuracy": 0.8110932826995849,
"num_tokens": 1026166802.0,
"step": 11580
},
{
"entropy": 0.9603515625,
"epoch": 1.584022415089182,
"grad_norm": 0.7506958458380673,
"learning_rate": 1.0726363252078344e-06,
"loss": 0.6892,
"mean_token_accuracy": 0.8139872908592224,
"num_tokens": 1027024240.0,
"step": 11590
},
{
"entropy": 0.9298828125,
"epoch": 1.5853891888197909,
"grad_norm": 0.7253385947361238,
"learning_rate": 1.0691137100183178e-06,
"loss": 0.6072,
"mean_token_accuracy": 0.8328027814626694,
"num_tokens": 1027904455.0,
"step": 11600
},
{
"entropy": 0.9267578125,
"epoch": 1.5867559625503997,
"grad_norm": 0.816509050180446,
"learning_rate": 1.065591094828801e-06,
"loss": 0.7049,
"mean_token_accuracy": 0.8125887095928193,
"num_tokens": 1028833540.0,
"step": 11610
},
{
"entropy": 0.9466796875,
"epoch": 1.5881227362810086,
"grad_norm": 0.8196682148110378,
"learning_rate": 1.0620684796392842e-06,
"loss": 0.6358,
"mean_token_accuracy": 0.824048325419426,
"num_tokens": 1029655720.0,
"step": 11620
},
{
"entropy": 0.9083984375,
"epoch": 1.5894895100116175,
"grad_norm": 0.7715178921617069,
"learning_rate": 1.0585458644497676e-06,
"loss": 0.6064,
"mean_token_accuracy": 0.8317598760128021,
"num_tokens": 1030505941.0,
"step": 11630
},
{
"entropy": 0.9958984375,
"epoch": 1.5908562837422264,
"grad_norm": 0.6879877784869406,
"learning_rate": 1.0550232492602509e-06,
"loss": 0.672,
"mean_token_accuracy": 0.8167349129915238,
"num_tokens": 1031370313.0,
"step": 11640
},
{
"entropy": 0.9865234375,
"epoch": 1.5922230574728353,
"grad_norm": 0.7237569566024954,
"learning_rate": 1.0515006340707342e-06,
"loss": 0.6727,
"mean_token_accuracy": 0.8165571212768554,
"num_tokens": 1032243138.0,
"step": 11650
},
{
"entropy": 0.929296875,
"epoch": 1.5935898312034442,
"grad_norm": 0.7890766612292245,
"learning_rate": 1.0479780188812176e-06,
"loss": 0.6296,
"mean_token_accuracy": 0.8247863054275513,
"num_tokens": 1033081238.0,
"step": 11660
},
{
"entropy": 0.948046875,
"epoch": 1.5949566049340531,
"grad_norm": 0.737255870297174,
"learning_rate": 1.0444554036917009e-06,
"loss": 0.6596,
"mean_token_accuracy": 0.8201647460460663,
"num_tokens": 1033975506.0,
"step": 11670
},
{
"entropy": 0.9771484375,
"epoch": 1.5963233786646622,
"grad_norm": 0.7345887089859232,
"learning_rate": 1.0409327885021842e-06,
"loss": 0.677,
"mean_token_accuracy": 0.8137074142694474,
"num_tokens": 1034879253.0,
"step": 11680
},
{
"entropy": 0.9275390625,
"epoch": 1.5976901523952711,
"grad_norm": 0.8117094305660145,
"learning_rate": 1.0374101733126674e-06,
"loss": 0.6419,
"mean_token_accuracy": 0.8245418727397918,
"num_tokens": 1035764543.0,
"step": 11690
},
{
"entropy": 0.9376953125,
"epoch": 1.59905692612588,
"grad_norm": 0.8715548192847319,
"learning_rate": 1.0338875581231507e-06,
"loss": 0.627,
"mean_token_accuracy": 0.8275120288133622,
"num_tokens": 1036597658.0,
"step": 11700
},
{
"entropy": 0.9861328125,
"epoch": 1.600423699856489,
"grad_norm": 0.7842040416038775,
"learning_rate": 1.030364942933634e-06,
"loss": 0.6974,
"mean_token_accuracy": 0.8096868962049484,
"num_tokens": 1037461178.0,
"step": 11710
},
{
"entropy": 0.945703125,
"epoch": 1.6017904735870978,
"grad_norm": 0.644424652582594,
"learning_rate": 1.0268423277441174e-06,
"loss": 0.6649,
"mean_token_accuracy": 0.8212746411561966,
"num_tokens": 1038320543.0,
"step": 11720
},
{
"entropy": 0.9583984375,
"epoch": 1.6031572473177067,
"grad_norm": 0.7882832314602223,
"learning_rate": 1.0233197125546005e-06,
"loss": 0.6601,
"mean_token_accuracy": 0.8192164868116378,
"num_tokens": 1039158322.0,
"step": 11730
},
{
"entropy": 0.9595703125,
"epoch": 1.6045240210483156,
"grad_norm": 0.7832014444809149,
"learning_rate": 1.0197970973650838e-06,
"loss": 0.6523,
"mean_token_accuracy": 0.8204129457473754,
"num_tokens": 1040044664.0,
"step": 11740
},
{
"entropy": 0.9474609375,
"epoch": 1.6058907947789245,
"grad_norm": 0.8665204415138003,
"learning_rate": 1.0162744821755674e-06,
"loss": 0.6857,
"mean_token_accuracy": 0.8151431918144226,
"num_tokens": 1040933733.0,
"step": 11750
},
{
"entropy": 0.93125,
"epoch": 1.6072575685095334,
"grad_norm": 0.7782346843862404,
"learning_rate": 1.0127518669860505e-06,
"loss": 0.6331,
"mean_token_accuracy": 0.8259228497743607,
"num_tokens": 1041821739.0,
"step": 11760
},
{
"entropy": 0.9744140625,
"epoch": 1.6086243422401423,
"grad_norm": 0.8708220504366528,
"learning_rate": 1.0092292517965338e-06,
"loss": 0.7028,
"mean_token_accuracy": 0.8125641196966171,
"num_tokens": 1042734726.0,
"step": 11770
},
{
"entropy": 0.96953125,
"epoch": 1.6099911159707512,
"grad_norm": 0.7739730326778256,
"learning_rate": 1.0057066366070172e-06,
"loss": 0.6588,
"mean_token_accuracy": 0.820311638712883,
"num_tokens": 1043605674.0,
"step": 11780
},
{
"entropy": 0.9275390625,
"epoch": 1.61135788970136,
"grad_norm": 0.7814796070931435,
"learning_rate": 1.0021840214175005e-06,
"loss": 0.647,
"mean_token_accuracy": 0.8233354300260544,
"num_tokens": 1044489733.0,
"step": 11790
},
{
"entropy": 0.9640625,
"epoch": 1.612724663431969,
"grad_norm": 0.8358220089712634,
"learning_rate": 9.986614062279836e-07,
"loss": 0.6695,
"mean_token_accuracy": 0.8181432873010636,
"num_tokens": 1045386114.0,
"step": 11800
},
{
"entropy": 0.926953125,
"epoch": 1.6140914371625779,
"grad_norm": 0.7650188614525529,
"learning_rate": 9.95138791038467e-07,
"loss": 0.6266,
"mean_token_accuracy": 0.8256363153457642,
"num_tokens": 1046281529.0,
"step": 11810
},
{
"entropy": 0.9232421875,
"epoch": 1.6154582108931868,
"grad_norm": 0.7799471947734157,
"learning_rate": 9.916161758489503e-07,
"loss": 0.6237,
"mean_token_accuracy": 0.8287038296461106,
"num_tokens": 1047172601.0,
"step": 11820
},
{
"entropy": 0.98046875,
"epoch": 1.6168249846237956,
"grad_norm": 0.7477910273927951,
"learning_rate": 9.880935606594336e-07,
"loss": 0.6564,
"mean_token_accuracy": 0.8205193400382995,
"num_tokens": 1048111551.0,
"step": 11830
},
{
"entropy": 0.95,
"epoch": 1.6181917583544045,
"grad_norm": 0.6816837471832393,
"learning_rate": 9.84570945469917e-07,
"loss": 0.654,
"mean_token_accuracy": 0.8223404705524444,
"num_tokens": 1048993658.0,
"step": 11840
},
{
"entropy": 0.9666015625,
"epoch": 1.6195585320850134,
"grad_norm": 0.7558658992943872,
"learning_rate": 9.810483302804003e-07,
"loss": 0.6419,
"mean_token_accuracy": 0.8216461807489395,
"num_tokens": 1049881721.0,
"step": 11850
},
{
"entropy": 0.959375,
"epoch": 1.6209253058156223,
"grad_norm": 0.7771213922423909,
"learning_rate": 9.775257150908836e-07,
"loss": 0.6458,
"mean_token_accuracy": 0.8215840369462967,
"num_tokens": 1050738642.0,
"step": 11860
},
{
"entropy": 0.9375,
"epoch": 1.6222920795462312,
"grad_norm": 0.8171821869326382,
"learning_rate": 9.740030999013668e-07,
"loss": 0.6616,
"mean_token_accuracy": 0.8210203379392624,
"num_tokens": 1051607144.0,
"step": 11870
},
{
"entropy": 0.9279296875,
"epoch": 1.6236588532768401,
"grad_norm": 0.7599604481423757,
"learning_rate": 9.7048048471185e-07,
"loss": 0.6439,
"mean_token_accuracy": 0.820563793182373,
"num_tokens": 1052460566.0,
"step": 11880
},
{
"entropy": 0.9572265625,
"epoch": 1.625025627007449,
"grad_norm": 0.6686325352156924,
"learning_rate": 9.669578695223334e-07,
"loss": 0.6918,
"mean_token_accuracy": 0.8113070666790009,
"num_tokens": 1053364568.0,
"step": 11890
},
{
"entropy": 0.9453125,
"epoch": 1.626392400738058,
"grad_norm": 0.8203931221063542,
"learning_rate": 9.634352543328168e-07,
"loss": 0.6547,
"mean_token_accuracy": 0.8192985773086547,
"num_tokens": 1054222654.0,
"step": 11900
},
{
"entropy": 0.9578125,
"epoch": 1.6277591744686668,
"grad_norm": 0.8359158507146092,
"learning_rate": 9.599126391433e-07,
"loss": 0.635,
"mean_token_accuracy": 0.8264235258102417,
"num_tokens": 1055112095.0,
"step": 11910
},
{
"entropy": 0.9759765625,
"epoch": 1.6291259481992757,
"grad_norm": 0.7887556040727974,
"learning_rate": 9.563900239537834e-07,
"loss": 0.6807,
"mean_token_accuracy": 0.8164149075746536,
"num_tokens": 1055990084.0,
"step": 11920
},
{
"entropy": 0.93203125,
"epoch": 1.6304927219298846,
"grad_norm": 0.8434668377889049,
"learning_rate": 9.528674087642667e-07,
"loss": 0.6468,
"mean_token_accuracy": 0.820684164762497,
"num_tokens": 1056906052.0,
"step": 11930
},
{
"entropy": 0.9068359375,
"epoch": 1.6318594956604935,
"grad_norm": 0.7608496830338868,
"learning_rate": 9.4934479357475e-07,
"loss": 0.6161,
"mean_token_accuracy": 0.8309248119592667,
"num_tokens": 1057804689.0,
"step": 11940
},
{
"entropy": 1.0056640625,
"epoch": 1.6332262693911024,
"grad_norm": 0.7272592992633382,
"learning_rate": 9.458221783852332e-07,
"loss": 0.6964,
"mean_token_accuracy": 0.8110878139734268,
"num_tokens": 1058720254.0,
"step": 11950
},
{
"entropy": 0.9548828125,
"epoch": 1.6345930431217113,
"grad_norm": 0.7588203392576517,
"learning_rate": 9.422995631957166e-07,
"loss": 0.6691,
"mean_token_accuracy": 0.8183633238077164,
"num_tokens": 1059561064.0,
"step": 11960
},
{
"entropy": 0.9650390625,
"epoch": 1.6359598168523202,
"grad_norm": 0.8407476551754808,
"learning_rate": 9.387769480061999e-07,
"loss": 0.6966,
"mean_token_accuracy": 0.8111886948347091,
"num_tokens": 1060429575.0,
"step": 11970
},
{
"entropy": 0.9763671875,
"epoch": 1.637326590582929,
"grad_norm": 0.7612907402167215,
"learning_rate": 9.352543328166831e-07,
"loss": 0.6603,
"mean_token_accuracy": 0.8181328684091568,
"num_tokens": 1061352006.0,
"step": 11980
},
{
"entropy": 0.9486328125,
"epoch": 1.638693364313538,
"grad_norm": 0.7034354888803953,
"learning_rate": 9.317317176271665e-07,
"loss": 0.6504,
"mean_token_accuracy": 0.822641983628273,
"num_tokens": 1062196979.0,
"step": 11990
},
{
"entropy": 0.9787109375,
"epoch": 1.6400601380441469,
"grad_norm": 0.7083510006752894,
"learning_rate": 9.282091024376498e-07,
"loss": 0.7175,
"mean_token_accuracy": 0.8077957332134247,
"num_tokens": 1063136300.0,
"step": 12000
},
{
"entropy": 1.010546875,
"epoch": 1.6414269117747557,
"grad_norm": 0.8629914322711024,
"learning_rate": 9.246864872481331e-07,
"loss": 0.7329,
"mean_token_accuracy": 0.8017515569925309,
"num_tokens": 1063984611.0,
"step": 12010
},
{
"entropy": 0.96015625,
"epoch": 1.6427936855053646,
"grad_norm": 0.7662066054326685,
"learning_rate": 9.211638720586164e-07,
"loss": 0.6571,
"mean_token_accuracy": 0.8182473540306091,
"num_tokens": 1064831535.0,
"step": 12020
},
{
"entropy": 0.959765625,
"epoch": 1.6441604592359735,
"grad_norm": 0.8940408171768202,
"learning_rate": 9.176412568690997e-07,
"loss": 0.6672,
"mean_token_accuracy": 0.818760946393013,
"num_tokens": 1065725334.0,
"step": 12030
},
{
"entropy": 0.9458984375,
"epoch": 1.6455272329665824,
"grad_norm": 0.8582001865177349,
"learning_rate": 9.14118641679583e-07,
"loss": 0.7001,
"mean_token_accuracy": 0.8106323152780532,
"num_tokens": 1066626508.0,
"step": 12040
},
{
"entropy": 0.9400390625,
"epoch": 1.6468940066971913,
"grad_norm": 0.754130019259656,
"learning_rate": 9.105960264900663e-07,
"loss": 0.685,
"mean_token_accuracy": 0.8141816943883896,
"num_tokens": 1067460843.0,
"step": 12050
},
{
"entropy": 0.940234375,
"epoch": 1.6482607804278002,
"grad_norm": 0.7564207764101079,
"learning_rate": 9.070734113005496e-07,
"loss": 0.662,
"mean_token_accuracy": 0.8204574137926102,
"num_tokens": 1068399394.0,
"step": 12060
},
{
"entropy": 0.992578125,
"epoch": 1.6496275541584091,
"grad_norm": 0.7888605057244279,
"learning_rate": 9.035507961110329e-07,
"loss": 0.7028,
"mean_token_accuracy": 0.8090411722660065,
"num_tokens": 1069269064.0,
"step": 12070
},
{
"entropy": 0.9419921875,
"epoch": 1.650994327889018,
"grad_norm": 0.781017594793854,
"learning_rate": 9.000281809215163e-07,
"loss": 0.6332,
"mean_token_accuracy": 0.8254497170448303,
"num_tokens": 1070151947.0,
"step": 12080
},
{
"entropy": 0.9484375,
"epoch": 1.652361101619627,
"grad_norm": 0.8288975143110983,
"learning_rate": 8.965055657319995e-07,
"loss": 0.6482,
"mean_token_accuracy": 0.8231672495603561,
"num_tokens": 1071027874.0,
"step": 12090
},
{
"entropy": 0.9447265625,
"epoch": 1.6537278753502358,
"grad_norm": 0.672559414103624,
"learning_rate": 8.929829505424828e-07,
"loss": 0.667,
"mean_token_accuracy": 0.8183021605014801,
"num_tokens": 1071914954.0,
"step": 12100
},
{
"entropy": 0.93671875,
"epoch": 1.6550946490808447,
"grad_norm": 0.7535378650371815,
"learning_rate": 8.894603353529662e-07,
"loss": 0.6281,
"mean_token_accuracy": 0.8273297011852264,
"num_tokens": 1072794976.0,
"step": 12110
},
{
"entropy": 0.9359375,
"epoch": 1.6564614228114536,
"grad_norm": 0.7215564747054025,
"learning_rate": 8.859377201634495e-07,
"loss": 0.6319,
"mean_token_accuracy": 0.8246982097625732,
"num_tokens": 1073716171.0,
"step": 12120
},
{
"entropy": 1.0083984375,
"epoch": 1.6578281965420625,
"grad_norm": 0.7472598562175439,
"learning_rate": 8.824151049739326e-07,
"loss": 0.7082,
"mean_token_accuracy": 0.807920977473259,
"num_tokens": 1074611877.0,
"step": 12130
},
{
"entropy": 0.9212890625,
"epoch": 1.6591949702726714,
"grad_norm": 0.7837632178566025,
"learning_rate": 8.788924897844161e-07,
"loss": 0.6311,
"mean_token_accuracy": 0.8264628678560257,
"num_tokens": 1075528862.0,
"step": 12140
},
{
"entropy": 0.935546875,
"epoch": 1.6605617440032803,
"grad_norm": 0.8166257172846414,
"learning_rate": 8.753698745948994e-07,
"loss": 0.6447,
"mean_token_accuracy": 0.823177257180214,
"num_tokens": 1076437769.0,
"step": 12150
},
{
"entropy": 0.93359375,
"epoch": 1.6619285177338892,
"grad_norm": 0.7769023945644657,
"learning_rate": 8.718472594053825e-07,
"loss": 0.6286,
"mean_token_accuracy": 0.825096708536148,
"num_tokens": 1077282147.0,
"step": 12160
},
{
"entropy": 0.972265625,
"epoch": 1.663295291464498,
"grad_norm": 0.8065906785787484,
"learning_rate": 8.683246442158659e-07,
"loss": 0.6809,
"mean_token_accuracy": 0.8172381192445755,
"num_tokens": 1078141716.0,
"step": 12170
},
{
"entropy": 0.9451171875,
"epoch": 1.664662065195107,
"grad_norm": 0.7055964452157324,
"learning_rate": 8.648020290263492e-07,
"loss": 0.6312,
"mean_token_accuracy": 0.8261988937854767,
"num_tokens": 1078987019.0,
"step": 12180
},
{
"entropy": 0.9681640625,
"epoch": 1.6660288389257158,
"grad_norm": 0.8347227133949184,
"learning_rate": 8.612794138368325e-07,
"loss": 0.6646,
"mean_token_accuracy": 0.8191884607076645,
"num_tokens": 1079883292.0,
"step": 12190
},
{
"entropy": 0.9681640625,
"epoch": 1.6673956126563247,
"grad_norm": 0.8244268779778339,
"learning_rate": 8.577567986473158e-07,
"loss": 0.6572,
"mean_token_accuracy": 0.8213898926973343,
"num_tokens": 1080734797.0,
"step": 12200
},
{
"entropy": 0.93828125,
"epoch": 1.6687623863869336,
"grad_norm": 0.7944788894555347,
"learning_rate": 8.542341834577991e-07,
"loss": 0.6433,
"mean_token_accuracy": 0.8242859244346619,
"num_tokens": 1081602784.0,
"step": 12210
},
{
"entropy": 0.9486328125,
"epoch": 1.6701291601175425,
"grad_norm": 0.8118986030714206,
"learning_rate": 8.507115682682824e-07,
"loss": 0.6607,
"mean_token_accuracy": 0.8202723979949951,
"num_tokens": 1082468814.0,
"step": 12220
},
{
"entropy": 0.975,
"epoch": 1.6714959338481514,
"grad_norm": 0.8467503693479339,
"learning_rate": 8.471889530787658e-07,
"loss": 0.7108,
"mean_token_accuracy": 0.8095838516950608,
"num_tokens": 1083378491.0,
"step": 12230
},
{
"entropy": 0.9505859375,
"epoch": 1.6728627075787603,
"grad_norm": 0.7608262378379579,
"learning_rate": 8.43666337889249e-07,
"loss": 0.6519,
"mean_token_accuracy": 0.8232457756996154,
"num_tokens": 1084256436.0,
"step": 12240
},
{
"entropy": 0.9361328125,
"epoch": 1.6742294813093692,
"grad_norm": 0.7117550594320742,
"learning_rate": 8.401437226997323e-07,
"loss": 0.6664,
"mean_token_accuracy": 0.8190992772579193,
"num_tokens": 1085167492.0,
"step": 12250
},
{
"entropy": 0.9609375,
"epoch": 1.675596255039978,
"grad_norm": 0.8254684831761284,
"learning_rate": 8.366211075102157e-07,
"loss": 0.655,
"mean_token_accuracy": 0.8200739413499832,
"num_tokens": 1085994348.0,
"step": 12260
},
{
"entropy": 0.9451171875,
"epoch": 1.676963028770587,
"grad_norm": 0.7908176987005098,
"learning_rate": 8.330984923206989e-07,
"loss": 0.6467,
"mean_token_accuracy": 0.8227890372276306,
"num_tokens": 1086909354.0,
"step": 12270
},
{
"entropy": 0.966015625,
"epoch": 1.678329802501196,
"grad_norm": 0.775012436580268,
"learning_rate": 8.295758771311822e-07,
"loss": 0.6624,
"mean_token_accuracy": 0.8178764283657074,
"num_tokens": 1087731177.0,
"step": 12280
},
{
"entropy": 0.9537109375,
"epoch": 1.6796965762318048,
"grad_norm": 0.6845038384399972,
"learning_rate": 8.260532619416656e-07,
"loss": 0.6738,
"mean_token_accuracy": 0.8177630752325058,
"num_tokens": 1088587754.0,
"step": 12290
},
{
"entropy": 0.94296875,
"epoch": 1.6810633499624137,
"grad_norm": 0.7490262649805889,
"learning_rate": 8.225306467521489e-07,
"loss": 0.6276,
"mean_token_accuracy": 0.8283274829387665,
"num_tokens": 1089479095.0,
"step": 12300
},
{
"entropy": 0.9630859375,
"epoch": 1.6824301236930226,
"grad_norm": 0.7965800382081495,
"learning_rate": 8.190080315626321e-07,
"loss": 0.6781,
"mean_token_accuracy": 0.815346148610115,
"num_tokens": 1090337044.0,
"step": 12310
},
{
"entropy": 0.9568359375,
"epoch": 1.6837968974236315,
"grad_norm": 0.736393150251633,
"learning_rate": 8.154854163731155e-07,
"loss": 0.6595,
"mean_token_accuracy": 0.8208300828933716,
"num_tokens": 1091244361.0,
"step": 12320
},
{
"entropy": 0.969921875,
"epoch": 1.6851636711542404,
"grad_norm": 0.8210906755178136,
"learning_rate": 8.119628011835988e-07,
"loss": 0.6835,
"mean_token_accuracy": 0.8131062805652618,
"num_tokens": 1092118184.0,
"step": 12330
},
{
"entropy": 0.940625,
"epoch": 1.6865304448848493,
"grad_norm": 0.7610678794497258,
"learning_rate": 8.084401859940821e-07,
"loss": 0.6403,
"mean_token_accuracy": 0.8213399529457093,
"num_tokens": 1092980010.0,
"step": 12340
},
{
"entropy": 0.966015625,
"epoch": 1.6878972186154582,
"grad_norm": 0.7882703948728238,
"learning_rate": 8.049175708045654e-07,
"loss": 0.6803,
"mean_token_accuracy": 0.8165583550930023,
"num_tokens": 1093871943.0,
"step": 12350
},
{
"entropy": 0.9408203125,
"epoch": 1.689263992346067,
"grad_norm": 0.6968684728186172,
"learning_rate": 8.013949556150487e-07,
"loss": 0.6481,
"mean_token_accuracy": 0.821254700422287,
"num_tokens": 1094708515.0,
"step": 12360
},
{
"entropy": 0.9412109375,
"epoch": 1.690630766076676,
"grad_norm": 0.7617989510533331,
"learning_rate": 7.97872340425532e-07,
"loss": 0.6567,
"mean_token_accuracy": 0.8200385451316834,
"num_tokens": 1095605283.0,
"step": 12370
},
{
"entropy": 0.9615234375,
"epoch": 1.6919975398072848,
"grad_norm": 0.9693381660138017,
"learning_rate": 7.943497252360153e-07,
"loss": 0.6985,
"mean_token_accuracy": 0.8121289014816284,
"num_tokens": 1096471521.0,
"step": 12380
},
{
"entropy": 0.9787109375,
"epoch": 1.6933643135378937,
"grad_norm": 0.7782153518548922,
"learning_rate": 7.908271100464986e-07,
"loss": 0.7305,
"mean_token_accuracy": 0.8023481637239456,
"num_tokens": 1097380895.0,
"step": 12390
},
{
"entropy": 0.9677734375,
"epoch": 1.6947310872685026,
"grad_norm": 0.8386679352134524,
"learning_rate": 7.873044948569819e-07,
"loss": 0.6653,
"mean_token_accuracy": 0.8191162437200546,
"num_tokens": 1098264769.0,
"step": 12400
},
{
"entropy": 0.976171875,
"epoch": 1.6960978609991115,
"grad_norm": 0.7620721398312705,
"learning_rate": 7.837818796674653e-07,
"loss": 0.6672,
"mean_token_accuracy": 0.8190709054470062,
"num_tokens": 1099166534.0,
"step": 12410
},
{
"entropy": 0.942578125,
"epoch": 1.6974646347297204,
"grad_norm": 0.8540066552525449,
"learning_rate": 7.802592644779484e-07,
"loss": 0.6259,
"mean_token_accuracy": 0.8257846444845199,
"num_tokens": 1100092895.0,
"step": 12420
},
{
"entropy": 0.9623046875,
"epoch": 1.6988314084603293,
"grad_norm": 0.8154429202809911,
"learning_rate": 7.767366492884317e-07,
"loss": 0.6714,
"mean_token_accuracy": 0.8148260653018952,
"num_tokens": 1100961965.0,
"step": 12430
},
{
"entropy": 0.94140625,
"epoch": 1.7001981821909382,
"grad_norm": 0.8317514266807388,
"learning_rate": 7.732140340989152e-07,
"loss": 0.6765,
"mean_token_accuracy": 0.8146194487810134,
"num_tokens": 1101887589.0,
"step": 12440
},
{
"entropy": 0.958203125,
"epoch": 1.701564955921547,
"grad_norm": 0.7903551790584575,
"learning_rate": 7.696914189093985e-07,
"loss": 0.6474,
"mean_token_accuracy": 0.823235347867012,
"num_tokens": 1102768436.0,
"step": 12450
},
{
"entropy": 0.9474609375,
"epoch": 1.702931729652156,
"grad_norm": 0.6925102977141541,
"learning_rate": 7.661688037198816e-07,
"loss": 0.6717,
"mean_token_accuracy": 0.817302280664444,
"num_tokens": 1103662460.0,
"step": 12460
},
{
"entropy": 0.9279296875,
"epoch": 1.704298503382765,
"grad_norm": 0.7374592856948675,
"learning_rate": 7.62646188530365e-07,
"loss": 0.6194,
"mean_token_accuracy": 0.828275591135025,
"num_tokens": 1104539373.0,
"step": 12470
},
{
"entropy": 0.9005859375,
"epoch": 1.7056652771133738,
"grad_norm": 0.7217870656747233,
"learning_rate": 7.591235733408483e-07,
"loss": 0.6164,
"mean_token_accuracy": 0.8284176290035248,
"num_tokens": 1105434657.0,
"step": 12480
},
{
"entropy": 0.9119140625,
"epoch": 1.7070320508439827,
"grad_norm": 0.8033723401400529,
"learning_rate": 7.556009581513315e-07,
"loss": 0.6393,
"mean_token_accuracy": 0.8241325974464416,
"num_tokens": 1106311018.0,
"step": 12490
},
{
"entropy": 0.92109375,
"epoch": 1.7083988245745916,
"grad_norm": 0.7491355767305014,
"learning_rate": 7.520783429618149e-07,
"loss": 0.6134,
"mean_token_accuracy": 0.8306571543216705,
"num_tokens": 1107229488.0,
"step": 12500
},
{
"entropy": 0.9181640625,
"epoch": 1.7097655983052005,
"grad_norm": 0.7572819113197354,
"learning_rate": 7.485557277722982e-07,
"loss": 0.6573,
"mean_token_accuracy": 0.8217572391033172,
"num_tokens": 1108129586.0,
"step": 12510
},
{
"entropy": 0.9544921875,
"epoch": 1.7111323720358094,
"grad_norm": 0.7540724880104969,
"learning_rate": 7.450331125827815e-07,
"loss": 0.6701,
"mean_token_accuracy": 0.8162120044231415,
"num_tokens": 1109025467.0,
"step": 12520
},
{
"entropy": 0.9771484375,
"epoch": 1.7124991457664183,
"grad_norm": 0.8143666014754177,
"learning_rate": 7.415104973932648e-07,
"loss": 0.68,
"mean_token_accuracy": 0.8150248467922211,
"num_tokens": 1109928602.0,
"step": 12530
},
{
"entropy": 0.9248046875,
"epoch": 1.7138659194970272,
"grad_norm": 0.7892359715837209,
"learning_rate": 7.379878822037481e-07,
"loss": 0.6086,
"mean_token_accuracy": 0.8291657865047455,
"num_tokens": 1110759388.0,
"step": 12540
},
{
"entropy": 0.9787109375,
"epoch": 1.715232693227636,
"grad_norm": 0.8238787192919966,
"learning_rate": 7.344652670142314e-07,
"loss": 0.6867,
"mean_token_accuracy": 0.8146190881729126,
"num_tokens": 1111652033.0,
"step": 12550
},
{
"entropy": 0.961328125,
"epoch": 1.716599466958245,
"grad_norm": 0.748764519879943,
"learning_rate": 7.309426518247147e-07,
"loss": 0.679,
"mean_token_accuracy": 0.8153380513191223,
"num_tokens": 1112559338.0,
"step": 12560
},
{
"entropy": 0.975,
"epoch": 1.7179662406888538,
"grad_norm": 0.7339385446649659,
"learning_rate": 7.27420036635198e-07,
"loss": 0.6693,
"mean_token_accuracy": 0.8172598004341125,
"num_tokens": 1113422543.0,
"step": 12570
},
{
"entropy": 0.9775390625,
"epoch": 1.7193330144194627,
"grad_norm": 0.7707000958845633,
"learning_rate": 7.238974214456813e-07,
"loss": 0.6909,
"mean_token_accuracy": 0.8112324059009552,
"num_tokens": 1114320790.0,
"step": 12580
},
{
"entropy": 0.96328125,
"epoch": 1.7206997881500716,
"grad_norm": 0.8041455604643656,
"learning_rate": 7.203748062561647e-07,
"loss": 0.6325,
"mean_token_accuracy": 0.8271142810583114,
"num_tokens": 1115279578.0,
"step": 12590
},
{
"entropy": 0.9580078125,
"epoch": 1.7220665618806805,
"grad_norm": 0.789869748481753,
"learning_rate": 7.168521910666479e-07,
"loss": 0.6844,
"mean_token_accuracy": 0.8177184104919434,
"num_tokens": 1116158991.0,
"step": 12600
},
{
"entropy": 0.951953125,
"epoch": 1.7234333356112894,
"grad_norm": 0.7728870083765823,
"learning_rate": 7.133295758771312e-07,
"loss": 0.6866,
"mean_token_accuracy": 0.8137596756219864,
"num_tokens": 1117041491.0,
"step": 12610
},
{
"entropy": 0.93515625,
"epoch": 1.7248001093418983,
"grad_norm": 0.8019466841042552,
"learning_rate": 7.098069606876146e-07,
"loss": 0.6615,
"mean_token_accuracy": 0.8188187211751938,
"num_tokens": 1117918004.0,
"step": 12620
},
{
"entropy": 0.91640625,
"epoch": 1.7261668830725072,
"grad_norm": 0.8635241856605076,
"learning_rate": 7.062843454980979e-07,
"loss": 0.6222,
"mean_token_accuracy": 0.8293952405452728,
"num_tokens": 1118785040.0,
"step": 12630
},
{
"entropy": 0.941015625,
"epoch": 1.727533656803116,
"grad_norm": 0.7031520644673506,
"learning_rate": 7.027617303085811e-07,
"loss": 0.6522,
"mean_token_accuracy": 0.8219096451997757,
"num_tokens": 1119691878.0,
"step": 12640
},
{
"entropy": 0.9412109375,
"epoch": 1.728900430533725,
"grad_norm": 0.7953078341953487,
"learning_rate": 6.992391151190645e-07,
"loss": 0.6341,
"mean_token_accuracy": 0.8258995234966278,
"num_tokens": 1120588565.0,
"step": 12650
},
{
"entropy": 0.9724609375,
"epoch": 1.730267204264334,
"grad_norm": 0.8898749308477133,
"learning_rate": 6.957164999295478e-07,
"loss": 0.6787,
"mean_token_accuracy": 0.8162712723016738,
"num_tokens": 1121432713.0,
"step": 12660
},
{
"entropy": 0.9587890625,
"epoch": 1.7316339779949428,
"grad_norm": 0.8300546067850609,
"learning_rate": 6.921938847400309e-07,
"loss": 0.6705,
"mean_token_accuracy": 0.8147017627954483,
"num_tokens": 1122273894.0,
"step": 12670
},
{
"entropy": 0.9470703125,
"epoch": 1.7330007517255517,
"grad_norm": 0.8010819470401588,
"learning_rate": 6.886712695505144e-07,
"loss": 0.6648,
"mean_token_accuracy": 0.8196389436721802,
"num_tokens": 1123187375.0,
"step": 12680
},
{
"entropy": 0.980078125,
"epoch": 1.7343675254561606,
"grad_norm": 0.7774334374432298,
"learning_rate": 6.851486543609977e-07,
"loss": 0.7044,
"mean_token_accuracy": 0.8112076759338379,
"num_tokens": 1124085975.0,
"step": 12690
},
{
"entropy": 0.98984375,
"epoch": 1.7357342991867695,
"grad_norm": 0.758822902127302,
"learning_rate": 6.816260391714811e-07,
"loss": 0.6708,
"mean_token_accuracy": 0.8167396485805511,
"num_tokens": 1124957939.0,
"step": 12700
},
{
"entropy": 0.956640625,
"epoch": 1.7371010729173784,
"grad_norm": 0.81317377619929,
"learning_rate": 6.781034239819642e-07,
"loss": 0.689,
"mean_token_accuracy": 0.8127140671014785,
"num_tokens": 1125845835.0,
"step": 12710
},
{
"entropy": 0.9435546875,
"epoch": 1.7384678466479875,
"grad_norm": 0.8241558268641221,
"learning_rate": 6.745808087924475e-07,
"loss": 0.647,
"mean_token_accuracy": 0.8213272660970687,
"num_tokens": 1126726139.0,
"step": 12720
},
{
"entropy": 0.9873046875,
"epoch": 1.7398346203785964,
"grad_norm": 0.8081318916164657,
"learning_rate": 6.710581936029308e-07,
"loss": 0.6578,
"mean_token_accuracy": 0.8201178312301636,
"num_tokens": 1127591299.0,
"step": 12730
},
{
"entropy": 0.93828125,
"epoch": 1.7412013941092053,
"grad_norm": 0.7839329457780168,
"learning_rate": 6.675355784134143e-07,
"loss": 0.6621,
"mean_token_accuracy": 0.8193959295749664,
"num_tokens": 1128505092.0,
"step": 12740
},
{
"entropy": 0.9615234375,
"epoch": 1.7425681678398142,
"grad_norm": 0.7808842603374061,
"learning_rate": 6.640129632238974e-07,
"loss": 0.6478,
"mean_token_accuracy": 0.8224100649356842,
"num_tokens": 1129395781.0,
"step": 12750
},
{
"entropy": 0.967578125,
"epoch": 1.743934941570423,
"grad_norm": 0.7268919076526645,
"learning_rate": 6.604903480343807e-07,
"loss": 0.6532,
"mean_token_accuracy": 0.8185953348875046,
"num_tokens": 1130301530.0,
"step": 12760
},
{
"entropy": 0.93671875,
"epoch": 1.745301715301032,
"grad_norm": 0.8707006541240871,
"learning_rate": 6.569677328448641e-07,
"loss": 0.6544,
"mean_token_accuracy": 0.8214032769203186,
"num_tokens": 1131161646.0,
"step": 12770
},
{
"entropy": 0.9189453125,
"epoch": 1.7466684890316408,
"grad_norm": 0.7536232663117115,
"learning_rate": 6.534451176553473e-07,
"loss": 0.6337,
"mean_token_accuracy": 0.8268335223197937,
"num_tokens": 1132051345.0,
"step": 12780
},
{
"entropy": 0.9345703125,
"epoch": 1.7480352627622497,
"grad_norm": 0.7873571971082018,
"learning_rate": 6.499225024658306e-07,
"loss": 0.6641,
"mean_token_accuracy": 0.8178780108690262,
"num_tokens": 1132962042.0,
"step": 12790
},
{
"entropy": 0.95859375,
"epoch": 1.7494020364928586,
"grad_norm": 0.7306151096122245,
"learning_rate": 6.46399887276314e-07,
"loss": 0.6358,
"mean_token_accuracy": 0.8253840148448944,
"num_tokens": 1133802402.0,
"step": 12800
},
{
"entropy": 0.9955078125,
"epoch": 1.7507688102234675,
"grad_norm": 0.8831337122218317,
"learning_rate": 6.428772720867973e-07,
"loss": 0.6877,
"mean_token_accuracy": 0.813293731212616,
"num_tokens": 1134670099.0,
"step": 12810
},
{
"entropy": 0.9435546875,
"epoch": 1.7521355839540764,
"grad_norm": 0.7682794018739316,
"learning_rate": 6.393546568972805e-07,
"loss": 0.6336,
"mean_token_accuracy": 0.8245344877243042,
"num_tokens": 1135525749.0,
"step": 12820
},
{
"entropy": 0.973046875,
"epoch": 1.7535023576846853,
"grad_norm": 0.7608888070605316,
"learning_rate": 6.358320417077639e-07,
"loss": 0.6704,
"mean_token_accuracy": 0.8160347014665603,
"num_tokens": 1136407810.0,
"step": 12830
},
{
"entropy": 0.94296875,
"epoch": 1.7548691314152942,
"grad_norm": 0.7537778541212985,
"learning_rate": 6.323094265182472e-07,
"loss": 0.6225,
"mean_token_accuracy": 0.8255596667528152,
"num_tokens": 1137249574.0,
"step": 12840
},
{
"entropy": 0.9263671875,
"epoch": 1.756235905145903,
"grad_norm": 0.7435018562419017,
"learning_rate": 6.287868113287306e-07,
"loss": 0.6364,
"mean_token_accuracy": 0.82479829788208,
"num_tokens": 1138164235.0,
"step": 12850
},
{
"entropy": 0.9318359375,
"epoch": 1.757602678876512,
"grad_norm": 0.7819507572636784,
"learning_rate": 6.252641961392138e-07,
"loss": 0.6615,
"mean_token_accuracy": 0.8203050643205643,
"num_tokens": 1139067214.0,
"step": 12860
},
{
"entropy": 0.898046875,
"epoch": 1.758969452607121,
"grad_norm": 0.7508796250484295,
"learning_rate": 6.217415809496971e-07,
"loss": 0.6224,
"mean_token_accuracy": 0.8282010912895202,
"num_tokens": 1139953795.0,
"step": 12870
},
{
"entropy": 0.961328125,
"epoch": 1.7603362263377298,
"grad_norm": 0.7768694340438211,
"learning_rate": 6.182189657601804e-07,
"loss": 0.6917,
"mean_token_accuracy": 0.813293719291687,
"num_tokens": 1140862640.0,
"step": 12880
},
{
"entropy": 0.945703125,
"epoch": 1.7617030000683387,
"grad_norm": 0.8222819704507756,
"learning_rate": 6.146963505706637e-07,
"loss": 0.6445,
"mean_token_accuracy": 0.8235606282949448,
"num_tokens": 1141743254.0,
"step": 12890
},
{
"entropy": 0.95,
"epoch": 1.7630697737989476,
"grad_norm": 0.7132403286249662,
"learning_rate": 6.11173735381147e-07,
"loss": 0.6969,
"mean_token_accuracy": 0.8116786181926727,
"num_tokens": 1142600515.0,
"step": 12900
},
{
"entropy": 0.9791015625,
"epoch": 1.7644365475295565,
"grad_norm": 0.7637966903671388,
"learning_rate": 6.076511201916304e-07,
"loss": 0.6853,
"mean_token_accuracy": 0.8164904713630676,
"num_tokens": 1143475412.0,
"step": 12910
},
{
"entropy": 0.9484375,
"epoch": 1.7658033212601654,
"grad_norm": 0.895285515857494,
"learning_rate": 6.041285050021136e-07,
"loss": 0.6812,
"mean_token_accuracy": 0.8110013723373413,
"num_tokens": 1144302196.0,
"step": 12920
},
{
"entropy": 0.9712890625,
"epoch": 1.7671700949907743,
"grad_norm": 0.7221405361067177,
"learning_rate": 6.006058898125969e-07,
"loss": 0.6632,
"mean_token_accuracy": 0.8191487312316894,
"num_tokens": 1145232557.0,
"step": 12930
},
{
"entropy": 0.9791015625,
"epoch": 1.7685368687213832,
"grad_norm": 0.8081899817172545,
"learning_rate": 5.970832746230803e-07,
"loss": 0.6984,
"mean_token_accuracy": 0.8124122738838195,
"num_tokens": 1146194324.0,
"step": 12940
},
{
"entropy": 0.963671875,
"epoch": 1.769903642451992,
"grad_norm": 0.8396377470590368,
"learning_rate": 5.935606594335636e-07,
"loss": 0.6466,
"mean_token_accuracy": 0.8227294713258744,
"num_tokens": 1147126259.0,
"step": 12950
},
{
"entropy": 0.9544921875,
"epoch": 1.771270416182601,
"grad_norm": 0.7529061963086654,
"learning_rate": 5.900380442440468e-07,
"loss": 0.6662,
"mean_token_accuracy": 0.8208934605121613,
"num_tokens": 1147998971.0,
"step": 12960
},
{
"entropy": 0.985546875,
"epoch": 1.7726371899132098,
"grad_norm": 0.7916579105749608,
"learning_rate": 5.8651542905453e-07,
"loss": 0.7233,
"mean_token_accuracy": 0.8071961611509323,
"num_tokens": 1148904341.0,
"step": 12970
},
{
"entropy": 0.9365234375,
"epoch": 1.7740039636438187,
"grad_norm": 0.8217461580298846,
"learning_rate": 5.829928138650135e-07,
"loss": 0.6477,
"mean_token_accuracy": 0.8236705273389816,
"num_tokens": 1149783289.0,
"step": 12980
},
{
"entropy": 0.944140625,
"epoch": 1.7753707373744276,
"grad_norm": 0.7150790644025758,
"learning_rate": 5.794701986754967e-07,
"loss": 0.6334,
"mean_token_accuracy": 0.8263595163822174,
"num_tokens": 1150669938.0,
"step": 12990
},
{
"entropy": 0.9689453125,
"epoch": 1.7767375111050365,
"grad_norm": 0.800120736973675,
"learning_rate": 5.759475834859801e-07,
"loss": 0.6628,
"mean_token_accuracy": 0.8195582628250122,
"num_tokens": 1151559575.0,
"step": 13000
},
{
"entropy": 0.9876953125,
"epoch": 1.7781042848356454,
"grad_norm": 0.7696629458143894,
"learning_rate": 5.724249682964633e-07,
"loss": 0.6947,
"mean_token_accuracy": 0.8113165736198426,
"num_tokens": 1152414438.0,
"step": 13010
},
{
"entropy": 0.96328125,
"epoch": 1.7794710585662543,
"grad_norm": 0.7673227291282869,
"learning_rate": 5.689023531069466e-07,
"loss": 0.6456,
"mean_token_accuracy": 0.8217849284410477,
"num_tokens": 1153345079.0,
"step": 13020
},
{
"entropy": 0.9576171875,
"epoch": 1.7808378322968632,
"grad_norm": 0.7891326980051113,
"learning_rate": 5.6537973791743e-07,
"loss": 0.6673,
"mean_token_accuracy": 0.8175371259450912,
"num_tokens": 1154226611.0,
"step": 13030
},
{
"entropy": 0.915625,
"epoch": 1.782204606027472,
"grad_norm": 0.7190515199848061,
"learning_rate": 5.618571227279133e-07,
"loss": 0.6531,
"mean_token_accuracy": 0.8225264340639115,
"num_tokens": 1155145591.0,
"step": 13040
},
{
"entropy": 0.9025390625,
"epoch": 1.783571379758081,
"grad_norm": 0.8003112964765491,
"learning_rate": 5.583345075383965e-07,
"loss": 0.6347,
"mean_token_accuracy": 0.8242779195308685,
"num_tokens": 1156065067.0,
"step": 13050
},
{
"entropy": 0.94140625,
"epoch": 1.78493815348869,
"grad_norm": 0.8437048987948685,
"learning_rate": 5.548118923488799e-07,
"loss": 0.6096,
"mean_token_accuracy": 0.8308081269264221,
"num_tokens": 1156911394.0,
"step": 13060
},
{
"entropy": 0.9361328125,
"epoch": 1.786304927219299,
"grad_norm": 0.7583209890444264,
"learning_rate": 5.512892771593632e-07,
"loss": 0.6299,
"mean_token_accuracy": 0.8269709169864654,
"num_tokens": 1157807466.0,
"step": 13070
},
{
"entropy": 0.986328125,
"epoch": 1.787671700949908,
"grad_norm": 0.6621120783852985,
"learning_rate": 5.477666619698464e-07,
"loss": 0.7114,
"mean_token_accuracy": 0.8087771832942963,
"num_tokens": 1158708008.0,
"step": 13080
},
{
"entropy": 0.96796875,
"epoch": 1.7890384746805168,
"grad_norm": 0.7269944828641891,
"learning_rate": 5.442440467803298e-07,
"loss": 0.6635,
"mean_token_accuracy": 0.8176913946866989,
"num_tokens": 1159583725.0,
"step": 13090
},
{
"entropy": 0.9669921875,
"epoch": 1.7904052484111257,
"grad_norm": 0.8378209922179136,
"learning_rate": 5.407214315908131e-07,
"loss": 0.6341,
"mean_token_accuracy": 0.8241803497076035,
"num_tokens": 1160491493.0,
"step": 13100
},
{
"entropy": 0.94765625,
"epoch": 1.7917720221417346,
"grad_norm": 0.7933005273060361,
"learning_rate": 5.371988164012964e-07,
"loss": 0.6798,
"mean_token_accuracy": 0.8135666817426681,
"num_tokens": 1161409970.0,
"step": 13110
},
{
"entropy": 0.9109375,
"epoch": 1.7931387958723435,
"grad_norm": 0.7231413397214086,
"learning_rate": 5.336762012117797e-07,
"loss": 0.6297,
"mean_token_accuracy": 0.8261041581630707,
"num_tokens": 1162313347.0,
"step": 13120
},
{
"entropy": 0.9697265625,
"epoch": 1.7945055696029524,
"grad_norm": 0.8610229472462243,
"learning_rate": 5.30153586022263e-07,
"loss": 0.6597,
"mean_token_accuracy": 0.8197608828544617,
"num_tokens": 1163220881.0,
"step": 13130
},
{
"entropy": 0.95546875,
"epoch": 1.7958723433335613,
"grad_norm": 0.7138376843888816,
"learning_rate": 5.266309708327462e-07,
"loss": 0.6638,
"mean_token_accuracy": 0.8186276644468308,
"num_tokens": 1164094809.0,
"step": 13140
},
{
"entropy": 0.929296875,
"epoch": 1.7972391170641702,
"grad_norm": 0.7651644914779859,
"learning_rate": 5.231083556432296e-07,
"loss": 0.6301,
"mean_token_accuracy": 0.8285431891679764,
"num_tokens": 1164971338.0,
"step": 13150
},
{
"entropy": 0.9662109375,
"epoch": 1.798605890794779,
"grad_norm": 0.7679783393930418,
"learning_rate": 5.195857404537129e-07,
"loss": 0.733,
"mean_token_accuracy": 0.8123619258403778,
"num_tokens": 1165850482.0,
"step": 13160
},
{
"entropy": 0.97109375,
"epoch": 1.799972664525388,
"grad_norm": 0.8566984016986944,
"learning_rate": 5.160631252641961e-07,
"loss": 0.6502,
"mean_token_accuracy": 0.823721495270729,
"num_tokens": 1166706374.0,
"step": 13170
},
{
"entropy": 0.937890625,
"epoch": 1.8013394382559968,
"grad_norm": 0.9238297281825161,
"learning_rate": 5.125405100746795e-07,
"loss": 0.6609,
"mean_token_accuracy": 0.8206489533185959,
"num_tokens": 1167624146.0,
"step": 13180
},
{
"entropy": 0.9509765625,
"epoch": 1.8027062119866057,
"grad_norm": 0.772110988570183,
"learning_rate": 5.090178948851628e-07,
"loss": 0.6919,
"mean_token_accuracy": 0.8121084183454513,
"num_tokens": 1168541396.0,
"step": 13190
},
{
"entropy": 0.98671875,
"epoch": 1.8040729857172146,
"grad_norm": 0.8735388494350315,
"learning_rate": 5.054952796956461e-07,
"loss": 0.6897,
"mean_token_accuracy": 0.8113891005516052,
"num_tokens": 1169402817.0,
"step": 13200
},
{
"entropy": 0.9181640625,
"epoch": 1.8054397594478235,
"grad_norm": 0.733567387361352,
"learning_rate": 5.019726645061294e-07,
"loss": 0.5973,
"mean_token_accuracy": 0.8347584933042527,
"num_tokens": 1170317297.0,
"step": 13210
},
{
"entropy": 0.9826171875,
"epoch": 1.8068065331784324,
"grad_norm": 0.8487922545813343,
"learning_rate": 4.984500493166127e-07,
"loss": 0.6647,
"mean_token_accuracy": 0.8190629184246063,
"num_tokens": 1171217665.0,
"step": 13220
},
{
"entropy": 0.91015625,
"epoch": 1.8081733069090413,
"grad_norm": 0.7669531906679693,
"learning_rate": 4.94927434127096e-07,
"loss": 0.6015,
"mean_token_accuracy": 0.8341245621442794,
"num_tokens": 1172087543.0,
"step": 13230
},
{
"entropy": 0.9267578125,
"epoch": 1.8095400806396502,
"grad_norm": 0.8765399744780126,
"learning_rate": 4.914048189375794e-07,
"loss": 0.6685,
"mean_token_accuracy": 0.8180892527103424,
"num_tokens": 1172967896.0,
"step": 13240
},
{
"entropy": 0.9708984375,
"epoch": 1.8109068543702591,
"grad_norm": 0.7592766072329505,
"learning_rate": 4.878822037480626e-07,
"loss": 0.6769,
"mean_token_accuracy": 0.8158371716737747,
"num_tokens": 1173765974.0,
"step": 13250
},
{
"entropy": 0.9435546875,
"epoch": 1.812273628100868,
"grad_norm": 0.7573516214770482,
"learning_rate": 4.843595885585459e-07,
"loss": 0.6649,
"mean_token_accuracy": 0.8186886966228485,
"num_tokens": 1174621145.0,
"step": 13260
},
{
"entropy": 0.9560546875,
"epoch": 1.813640401831477,
"grad_norm": 0.8553703762529432,
"learning_rate": 4.808369733690292e-07,
"loss": 0.6784,
"mean_token_accuracy": 0.81549331843853,
"num_tokens": 1175529749.0,
"step": 13270
},
{
"entropy": 0.9912109375,
"epoch": 1.8150071755620858,
"grad_norm": 0.789312079658026,
"learning_rate": 4.773143581795125e-07,
"loss": 0.7387,
"mean_token_accuracy": 0.8042643427848816,
"num_tokens": 1176461343.0,
"step": 13280
},
{
"entropy": 0.9146484375,
"epoch": 1.8163739492926947,
"grad_norm": 0.7483234482678393,
"learning_rate": 4.7379174298999583e-07,
"loss": 0.6221,
"mean_token_accuracy": 0.828419753909111,
"num_tokens": 1177334280.0,
"step": 13290
},
{
"entropy": 0.9583984375,
"epoch": 1.8177407230233036,
"grad_norm": 0.8054616476112986,
"learning_rate": 4.702691278004791e-07,
"loss": 0.6707,
"mean_token_accuracy": 0.8190458834171295,
"num_tokens": 1178258174.0,
"step": 13300
},
{
"entropy": 0.928125,
"epoch": 1.8191074967539125,
"grad_norm": 0.8712198265123247,
"learning_rate": 4.6674651261096245e-07,
"loss": 0.6354,
"mean_token_accuracy": 0.8255772829055786,
"num_tokens": 1179152856.0,
"step": 13310
},
{
"entropy": 0.919921875,
"epoch": 1.8204742704845214,
"grad_norm": 0.8136808297969498,
"learning_rate": 4.632238974214457e-07,
"loss": 0.6406,
"mean_token_accuracy": 0.8255934387445449,
"num_tokens": 1180048625.0,
"step": 13320
},
{
"entropy": 0.920703125,
"epoch": 1.8218410442151303,
"grad_norm": 0.685271908639999,
"learning_rate": 4.5970128223192907e-07,
"loss": 0.6431,
"mean_token_accuracy": 0.8241287857294083,
"num_tokens": 1180938231.0,
"step": 13330
},
{
"entropy": 0.914453125,
"epoch": 1.8232078179457392,
"grad_norm": 0.7495318385131343,
"learning_rate": 4.561786670424123e-07,
"loss": 0.624,
"mean_token_accuracy": 0.826156759262085,
"num_tokens": 1181813930.0,
"step": 13340
},
{
"entropy": 0.9544921875,
"epoch": 1.824574591676348,
"grad_norm": 0.7933197159715784,
"learning_rate": 4.5265605185289563e-07,
"loss": 0.6412,
"mean_token_accuracy": 0.8227519720792771,
"num_tokens": 1182727375.0,
"step": 13350
},
{
"entropy": 0.9361328125,
"epoch": 1.825941365406957,
"grad_norm": 0.7537956073946802,
"learning_rate": 4.491334366633789e-07,
"loss": 0.6395,
"mean_token_accuracy": 0.822977215051651,
"num_tokens": 1183623201.0,
"step": 13360
},
{
"entropy": 0.9580078125,
"epoch": 1.8273081391375658,
"grad_norm": 0.8263457509071437,
"learning_rate": 4.456108214738622e-07,
"loss": 0.6695,
"mean_token_accuracy": 0.8185583978891373,
"num_tokens": 1184513685.0,
"step": 13370
},
{
"entropy": 0.9384765625,
"epoch": 1.8286749128681747,
"grad_norm": 0.8091238562947478,
"learning_rate": 4.4208820628434553e-07,
"loss": 0.6518,
"mean_token_accuracy": 0.8209544122219086,
"num_tokens": 1185400077.0,
"step": 13380
},
{
"entropy": 0.9701171875,
"epoch": 1.8300416865987836,
"grad_norm": 0.8016608998514188,
"learning_rate": 4.385655910948288e-07,
"loss": 0.6997,
"mean_token_accuracy": 0.8090697258710862,
"num_tokens": 1186291424.0,
"step": 13390
},
{
"entropy": 0.9640625,
"epoch": 1.8314084603293925,
"grad_norm": 0.7922895798903973,
"learning_rate": 4.3504297590531215e-07,
"loss": 0.712,
"mean_token_accuracy": 0.8097701162099838,
"num_tokens": 1187194562.0,
"step": 13400
},
{
"entropy": 0.9115234375,
"epoch": 1.8327752340600014,
"grad_norm": 0.7945673490081661,
"learning_rate": 4.3152036071579543e-07,
"loss": 0.6483,
"mean_token_accuracy": 0.8248016238212585,
"num_tokens": 1188098751.0,
"step": 13410
},
{
"entropy": 0.915625,
"epoch": 1.8341420077906103,
"grad_norm": 0.7492012477852826,
"learning_rate": 4.2799774552627877e-07,
"loss": 0.6381,
"mean_token_accuracy": 0.8251723229885102,
"num_tokens": 1189005385.0,
"step": 13420
},
{
"entropy": 0.9173828125,
"epoch": 1.8355087815212192,
"grad_norm": 0.9005472924188173,
"learning_rate": 4.2447513033676205e-07,
"loss": 0.6427,
"mean_token_accuracy": 0.8226012170314789,
"num_tokens": 1189882489.0,
"step": 13430
},
{
"entropy": 0.9521484375,
"epoch": 1.836875555251828,
"grad_norm": 0.8689391989170056,
"learning_rate": 4.209525151472454e-07,
"loss": 0.6563,
"mean_token_accuracy": 0.820982140302658,
"num_tokens": 1190750704.0,
"step": 13440
},
{
"entropy": 0.967578125,
"epoch": 1.838242328982437,
"grad_norm": 0.8064615749782322,
"learning_rate": 4.1742989995772867e-07,
"loss": 0.6923,
"mean_token_accuracy": 0.8118867725133896,
"num_tokens": 1191660625.0,
"step": 13450
},
{
"entropy": 0.9859375,
"epoch": 1.839609102713046,
"grad_norm": 0.8042164371524786,
"learning_rate": 4.13907284768212e-07,
"loss": 0.6893,
"mean_token_accuracy": 0.8135886698961258,
"num_tokens": 1192508000.0,
"step": 13460
},
{
"entropy": 0.990625,
"epoch": 1.8409758764436548,
"grad_norm": 0.8023599200929666,
"learning_rate": 4.1038466957869523e-07,
"loss": 0.6634,
"mean_token_accuracy": 0.8192778617143631,
"num_tokens": 1193362849.0,
"step": 13470
},
{
"entropy": 0.9478515625,
"epoch": 1.8423426501742637,
"grad_norm": 0.7755681557532604,
"learning_rate": 4.068620543891785e-07,
"loss": 0.659,
"mean_token_accuracy": 0.8195610582828522,
"num_tokens": 1194316124.0,
"step": 13480
},
{
"entropy": 0.9486328125,
"epoch": 1.8437094239048726,
"grad_norm": 0.7874340899573737,
"learning_rate": 4.0333943919966185e-07,
"loss": 0.6789,
"mean_token_accuracy": 0.8151122897863388,
"num_tokens": 1195255727.0,
"step": 13490
},
{
"entropy": 0.9291015625,
"epoch": 1.8450761976354815,
"grad_norm": 0.772777436566429,
"learning_rate": 3.9981682401014513e-07,
"loss": 0.642,
"mean_token_accuracy": 0.8226383328437805,
"num_tokens": 1196182466.0,
"step": 13500
},
{
"entropy": 0.9109375,
"epoch": 1.8464429713660904,
"grad_norm": 0.7503026434297624,
"learning_rate": 3.9629420882062847e-07,
"loss": 0.6179,
"mean_token_accuracy": 0.827944228053093,
"num_tokens": 1197059992.0,
"step": 13510
},
{
"entropy": 0.9685546875,
"epoch": 1.8478097450966993,
"grad_norm": 0.7385641732788377,
"learning_rate": 3.9277159363111175e-07,
"loss": 0.6587,
"mean_token_accuracy": 0.8194995641708374,
"num_tokens": 1197948995.0,
"step": 13520
},
{
"entropy": 0.9451171875,
"epoch": 1.8491765188273082,
"grad_norm": 0.8264244483211547,
"learning_rate": 3.892489784415951e-07,
"loss": 0.6593,
"mean_token_accuracy": 0.8188575744628906,
"num_tokens": 1198824386.0,
"step": 13530
},
{
"entropy": 0.9705078125,
"epoch": 1.850543292557917,
"grad_norm": 0.8706827566600667,
"learning_rate": 3.8572636325207837e-07,
"loss": 0.6789,
"mean_token_accuracy": 0.8156465351581573,
"num_tokens": 1199644308.0,
"step": 13540
},
{
"entropy": 0.9927734375,
"epoch": 1.851910066288526,
"grad_norm": 0.7970084942230706,
"learning_rate": 3.822037480625617e-07,
"loss": 0.6587,
"mean_token_accuracy": 0.8197364956140518,
"num_tokens": 1200529892.0,
"step": 13550
},
{
"entropy": 0.913671875,
"epoch": 1.8532768400191348,
"grad_norm": 0.7413107635134959,
"learning_rate": 3.78681132873045e-07,
"loss": 0.6197,
"mean_token_accuracy": 0.8296514332294465,
"num_tokens": 1201485843.0,
"step": 13560
},
{
"entropy": 0.96484375,
"epoch": 1.8546436137497437,
"grad_norm": 0.7147213300350369,
"learning_rate": 3.7515851768352827e-07,
"loss": 0.6665,
"mean_token_accuracy": 0.8180717408657074,
"num_tokens": 1202365458.0,
"step": 13570
},
{
"entropy": 0.9880859375,
"epoch": 1.8560103874803526,
"grad_norm": 0.8846255755274991,
"learning_rate": 3.716359024940116e-07,
"loss": 0.6751,
"mean_token_accuracy": 0.815267214179039,
"num_tokens": 1203210624.0,
"step": 13580
},
{
"entropy": 0.96796875,
"epoch": 1.8573771612109615,
"grad_norm": 0.8166791050381155,
"learning_rate": 3.6811328730449484e-07,
"loss": 0.6643,
"mean_token_accuracy": 0.8179858058691025,
"num_tokens": 1204109870.0,
"step": 13590
},
{
"entropy": 0.962109375,
"epoch": 1.8587439349415704,
"grad_norm": 0.7596291217473398,
"learning_rate": 3.645906721149782e-07,
"loss": 0.6295,
"mean_token_accuracy": 0.8268264472484589,
"num_tokens": 1204941911.0,
"step": 13600
},
{
"entropy": 0.9765625,
"epoch": 1.8601107086721793,
"grad_norm": 0.7863428538014702,
"learning_rate": 3.6106805692546145e-07,
"loss": 0.6624,
"mean_token_accuracy": 0.8202401369810104,
"num_tokens": 1205867891.0,
"step": 13610
},
{
"entropy": 0.974609375,
"epoch": 1.8614774824027882,
"grad_norm": 0.7660769563974467,
"learning_rate": 3.575454417359448e-07,
"loss": 0.6648,
"mean_token_accuracy": 0.8195932060480118,
"num_tokens": 1206743074.0,
"step": 13620
},
{
"entropy": 0.958203125,
"epoch": 1.862844256133397,
"grad_norm": 0.7841436281297082,
"learning_rate": 3.5402282654642807e-07,
"loss": 0.6195,
"mean_token_accuracy": 0.8284026384353638,
"num_tokens": 1207606486.0,
"step": 13630
},
{
"entropy": 0.947265625,
"epoch": 1.864211029864006,
"grad_norm": 0.7675944927517009,
"learning_rate": 3.505002113569114e-07,
"loss": 0.6849,
"mean_token_accuracy": 0.8139402687549591,
"num_tokens": 1208513818.0,
"step": 13640
},
{
"entropy": 0.9734375,
"epoch": 1.865577803594615,
"grad_norm": 0.9054850683091435,
"learning_rate": 3.469775961673947e-07,
"loss": 0.6684,
"mean_token_accuracy": 0.8189633846282959,
"num_tokens": 1209364561.0,
"step": 13650
},
{
"entropy": 0.927734375,
"epoch": 1.8669445773252238,
"grad_norm": 0.7253451425738825,
"learning_rate": 3.43454980977878e-07,
"loss": 0.6435,
"mean_token_accuracy": 0.8249411016702652,
"num_tokens": 1210238798.0,
"step": 13660
},
{
"entropy": 0.9779296875,
"epoch": 1.8683113510558327,
"grad_norm": 0.814241479339585,
"learning_rate": 3.399323657883613e-07,
"loss": 0.6665,
"mean_token_accuracy": 0.8174272984266281,
"num_tokens": 1211114581.0,
"step": 13670
},
{
"entropy": 0.96875,
"epoch": 1.8696781247864416,
"grad_norm": 0.7741570802994397,
"learning_rate": 3.364097505988446e-07,
"loss": 0.6859,
"mean_token_accuracy": 0.8140438944101334,
"num_tokens": 1212024219.0,
"step": 13680
},
{
"entropy": 0.955078125,
"epoch": 1.8710448985170505,
"grad_norm": 0.7243692126923005,
"learning_rate": 3.328871354093279e-07,
"loss": 0.6578,
"mean_token_accuracy": 0.8190436035394668,
"num_tokens": 1212946824.0,
"step": 13690
},
{
"entropy": 0.9705078125,
"epoch": 1.8724116722476594,
"grad_norm": 0.7702697584281489,
"learning_rate": 3.293645202198112e-07,
"loss": 0.6862,
"mean_token_accuracy": 0.813627976179123,
"num_tokens": 1213839645.0,
"step": 13700
},
{
"entropy": 0.964453125,
"epoch": 1.8737784459782683,
"grad_norm": 0.7263615864309159,
"learning_rate": 3.2584190503029454e-07,
"loss": 0.6604,
"mean_token_accuracy": 0.8197873920202255,
"num_tokens": 1214692776.0,
"step": 13710
},
{
"entropy": 0.9416015625,
"epoch": 1.8751452197088772,
"grad_norm": 0.79028363387204,
"learning_rate": 3.223192898407778e-07,
"loss": 0.6665,
"mean_token_accuracy": 0.8201498419046402,
"num_tokens": 1215588902.0,
"step": 13720
},
{
"entropy": 0.93828125,
"epoch": 1.876511993439486,
"grad_norm": 0.7347100290335417,
"learning_rate": 3.1879667465126116e-07,
"loss": 0.6266,
"mean_token_accuracy": 0.8292582154273986,
"num_tokens": 1216493221.0,
"step": 13730
},
{
"entropy": 0.9388671875,
"epoch": 1.877878767170095,
"grad_norm": 0.6450516801101849,
"learning_rate": 3.152740594617444e-07,
"loss": 0.6592,
"mean_token_accuracy": 0.8222676634788513,
"num_tokens": 1217440087.0,
"step": 13740
},
{
"entropy": 0.9431640625,
"epoch": 1.8792455409007038,
"grad_norm": 0.8070275480526404,
"learning_rate": 3.117514442722277e-07,
"loss": 0.6469,
"mean_token_accuracy": 0.8234017282724381,
"num_tokens": 1218357098.0,
"step": 13750
},
{
"entropy": 0.91640625,
"epoch": 1.8806123146313127,
"grad_norm": 0.7179255006011223,
"learning_rate": 3.08228829082711e-07,
"loss": 0.5863,
"mean_token_accuracy": 0.8367748916149139,
"num_tokens": 1219287510.0,
"step": 13760
},
{
"entropy": 0.94921875,
"epoch": 1.8819790883619216,
"grad_norm": 0.8297021160217399,
"learning_rate": 3.0470621389319434e-07,
"loss": 0.6478,
"mean_token_accuracy": 0.821247810125351,
"num_tokens": 1220145054.0,
"step": 13770
},
{
"entropy": 0.970703125,
"epoch": 1.8833458620925305,
"grad_norm": 0.7074829493559098,
"learning_rate": 3.0118359870367763e-07,
"loss": 0.6522,
"mean_token_accuracy": 0.8226251363754272,
"num_tokens": 1220986637.0,
"step": 13780
},
{
"entropy": 0.9263671875,
"epoch": 1.8847126358231394,
"grad_norm": 0.7753510936334449,
"learning_rate": 2.9766098351416096e-07,
"loss": 0.6077,
"mean_token_accuracy": 0.831348466873169,
"num_tokens": 1221882549.0,
"step": 13790
},
{
"entropy": 0.969140625,
"epoch": 1.8860794095537483,
"grad_norm": 0.7346969746993359,
"learning_rate": 2.9413836832464424e-07,
"loss": 0.67,
"mean_token_accuracy": 0.8175934851169586,
"num_tokens": 1222799616.0,
"step": 13800
},
{
"entropy": 0.9880859375,
"epoch": 1.8874461832843572,
"grad_norm": 0.821548655131653,
"learning_rate": 2.906157531351276e-07,
"loss": 0.6999,
"mean_token_accuracy": 0.8120456814765931,
"num_tokens": 1223692723.0,
"step": 13810
},
{
"entropy": 0.9427734375,
"epoch": 1.888812957014966,
"grad_norm": 0.8225711241382246,
"learning_rate": 2.8709313794561086e-07,
"loss": 0.6324,
"mean_token_accuracy": 0.8246808171272277,
"num_tokens": 1224602296.0,
"step": 13820
},
{
"entropy": 0.963671875,
"epoch": 1.890179730745575,
"grad_norm": 0.7850502501814712,
"learning_rate": 2.8357052275609415e-07,
"loss": 0.6885,
"mean_token_accuracy": 0.8121801316738129,
"num_tokens": 1225444602.0,
"step": 13830
},
{
"entropy": 0.99296875,
"epoch": 1.8915465044761839,
"grad_norm": 0.7383709368799191,
"learning_rate": 2.8004790756657743e-07,
"loss": 0.7093,
"mean_token_accuracy": 0.8113682985305786,
"num_tokens": 1226326865.0,
"step": 13840
},
{
"entropy": 0.93984375,
"epoch": 1.8929132782067928,
"grad_norm": 0.7286428852963185,
"learning_rate": 2.7652529237706076e-07,
"loss": 0.6557,
"mean_token_accuracy": 0.8202292352914811,
"num_tokens": 1227216431.0,
"step": 13850
},
{
"entropy": 0.9359375,
"epoch": 1.8942800519374017,
"grad_norm": 0.729584411143629,
"learning_rate": 2.7300267718754405e-07,
"loss": 0.6707,
"mean_token_accuracy": 0.820206680893898,
"num_tokens": 1228073184.0,
"step": 13860
},
{
"entropy": 0.89609375,
"epoch": 1.8956468256680106,
"grad_norm": 0.7725263234237149,
"learning_rate": 2.694800619980274e-07,
"loss": 0.6396,
"mean_token_accuracy": 0.8263342320919037,
"num_tokens": 1228961273.0,
"step": 13870
},
{
"entropy": 0.933984375,
"epoch": 1.8970135993986195,
"grad_norm": 0.8748857598799675,
"learning_rate": 2.6595744680851066e-07,
"loss": 0.6548,
"mean_token_accuracy": 0.8219392538070679,
"num_tokens": 1229885119.0,
"step": 13880
},
{
"entropy": 0.9857421875,
"epoch": 1.8983803731292284,
"grad_norm": 0.7575859085377316,
"learning_rate": 2.6243483161899395e-07,
"loss": 0.6647,
"mean_token_accuracy": 0.8183341473340988,
"num_tokens": 1230756421.0,
"step": 13890
},
{
"entropy": 0.9537109375,
"epoch": 1.8997471468598373,
"grad_norm": 0.7692959488128956,
"learning_rate": 2.589122164294773e-07,
"loss": 0.6795,
"mean_token_accuracy": 0.816574165225029,
"num_tokens": 1231686900.0,
"step": 13900
},
{
"entropy": 0.952734375,
"epoch": 1.9011139205904461,
"grad_norm": 0.8168450820217483,
"learning_rate": 2.5538960123996056e-07,
"loss": 0.6469,
"mean_token_accuracy": 0.8233403831720352,
"num_tokens": 1232568957.0,
"step": 13910
},
{
"entropy": 0.979296875,
"epoch": 1.902480694321055,
"grad_norm": 0.8285439021128614,
"learning_rate": 2.518669860504439e-07,
"loss": 0.6622,
"mean_token_accuracy": 0.8176115065813064,
"num_tokens": 1233407782.0,
"step": 13920
},
{
"entropy": 0.96640625,
"epoch": 1.903847468051664,
"grad_norm": 0.7867602625601082,
"learning_rate": 2.483443708609272e-07,
"loss": 0.6725,
"mean_token_accuracy": 0.818142193555832,
"num_tokens": 1234277152.0,
"step": 13930
},
{
"entropy": 0.92265625,
"epoch": 1.9052142417822728,
"grad_norm": 0.946460147901081,
"learning_rate": 2.4482175567141046e-07,
"loss": 0.6174,
"mean_token_accuracy": 0.8317802727222443,
"num_tokens": 1235185409.0,
"step": 13940
},
{
"entropy": 0.9689453125,
"epoch": 1.9065810155128817,
"grad_norm": 0.742262311379154,
"learning_rate": 2.4129914048189375e-07,
"loss": 0.682,
"mean_token_accuracy": 0.8140867471694946,
"num_tokens": 1236117547.0,
"step": 13950
},
{
"entropy": 0.96484375,
"epoch": 1.9079477892434906,
"grad_norm": 0.7309586084853469,
"learning_rate": 2.3777652529237708e-07,
"loss": 0.6737,
"mean_token_accuracy": 0.8166314333677291,
"num_tokens": 1237019384.0,
"step": 13960
},
{
"entropy": 0.9796875,
"epoch": 1.9093145629740995,
"grad_norm": 0.8496626780289076,
"learning_rate": 2.342539101028604e-07,
"loss": 0.7136,
"mean_token_accuracy": 0.8082188099622727,
"num_tokens": 1237946995.0,
"step": 13970
},
{
"entropy": 0.984765625,
"epoch": 1.9106813367047084,
"grad_norm": 0.8826350226620421,
"learning_rate": 2.307312949133437e-07,
"loss": 0.6921,
"mean_token_accuracy": 0.8130437284708023,
"num_tokens": 1238820204.0,
"step": 13980
},
{
"entropy": 0.941015625,
"epoch": 1.9120481104353173,
"grad_norm": 0.7329638259980651,
"learning_rate": 2.2720867972382698e-07,
"loss": 0.6546,
"mean_token_accuracy": 0.8214570641517639,
"num_tokens": 1239710908.0,
"step": 13990
},
{
"entropy": 0.9587890625,
"epoch": 1.9134148841659262,
"grad_norm": 0.7809183414378078,
"learning_rate": 2.236860645343103e-07,
"loss": 0.6611,
"mean_token_accuracy": 0.8183962583541871,
"num_tokens": 1240579007.0,
"step": 14000
},
{
"entropy": 0.926171875,
"epoch": 1.914781657896535,
"grad_norm": 0.847332914001071,
"learning_rate": 2.201634493447936e-07,
"loss": 0.6353,
"mean_token_accuracy": 0.8230259865522385,
"num_tokens": 1241438818.0,
"step": 14010
},
{
"entropy": 0.935546875,
"epoch": 1.916148431627144,
"grad_norm": 0.8061892522421268,
"learning_rate": 2.166408341552769e-07,
"loss": 0.6883,
"mean_token_accuracy": 0.8139110952615738,
"num_tokens": 1242302339.0,
"step": 14020
},
{
"entropy": 0.9490234375,
"epoch": 1.9175152053577529,
"grad_norm": 0.791229061428933,
"learning_rate": 2.131182189657602e-07,
"loss": 0.6253,
"mean_token_accuracy": 0.8277128159999847,
"num_tokens": 1243178877.0,
"step": 14030
},
{
"entropy": 0.946875,
"epoch": 1.9188819790883618,
"grad_norm": 0.8222931992954303,
"learning_rate": 2.095956037762435e-07,
"loss": 0.6696,
"mean_token_accuracy": 0.8147591412067413,
"num_tokens": 1244068968.0,
"step": 14040
},
{
"entropy": 0.9640625,
"epoch": 1.9202487528189707,
"grad_norm": 0.6928304122286961,
"learning_rate": 2.0607298858672678e-07,
"loss": 0.6299,
"mean_token_accuracy": 0.8253059297800064,
"num_tokens": 1244958623.0,
"step": 14050
},
{
"entropy": 0.9279296875,
"epoch": 1.9216155265495796,
"grad_norm": 0.8407502158881753,
"learning_rate": 2.025503733972101e-07,
"loss": 0.6509,
"mean_token_accuracy": 0.8226454585790635,
"num_tokens": 1245856273.0,
"step": 14060
},
{
"entropy": 0.9861328125,
"epoch": 1.9229823002801885,
"grad_norm": 0.7588174578793612,
"learning_rate": 1.990277582076934e-07,
"loss": 0.702,
"mean_token_accuracy": 0.8120182633399964,
"num_tokens": 1246770150.0,
"step": 14070
},
{
"entropy": 0.9580078125,
"epoch": 1.9243490740107974,
"grad_norm": 0.7460783772201656,
"learning_rate": 1.955051430181767e-07,
"loss": 0.6474,
"mean_token_accuracy": 0.8213987946510315,
"num_tokens": 1247607077.0,
"step": 14080
},
{
"entropy": 0.94765625,
"epoch": 1.9257158477414063,
"grad_norm": 0.8398872726414371,
"learning_rate": 1.9198252782866002e-07,
"loss": 0.6314,
"mean_token_accuracy": 0.8271125167608261,
"num_tokens": 1248487610.0,
"step": 14090
},
{
"entropy": 0.9462890625,
"epoch": 1.9270826214720151,
"grad_norm": 0.7257677333261028,
"learning_rate": 1.8845991263914333e-07,
"loss": 0.6222,
"mean_token_accuracy": 0.8272617608308792,
"num_tokens": 1249385378.0,
"step": 14100
},
{
"entropy": 0.9787109375,
"epoch": 1.928449395202624,
"grad_norm": 0.8448676721526622,
"learning_rate": 1.8493729744962664e-07,
"loss": 0.6482,
"mean_token_accuracy": 0.8238463461399078,
"num_tokens": 1250196035.0,
"step": 14110
},
{
"entropy": 0.9744140625,
"epoch": 1.9298161689332332,
"grad_norm": 0.8034565465259439,
"learning_rate": 1.8141468226010995e-07,
"loss": 0.6941,
"mean_token_accuracy": 0.813032129406929,
"num_tokens": 1251106482.0,
"step": 14120
},
{
"entropy": 0.9333984375,
"epoch": 1.931182942663842,
"grad_norm": 0.8091518558301256,
"learning_rate": 1.778920670705932e-07,
"loss": 0.6226,
"mean_token_accuracy": 0.827666112780571,
"num_tokens": 1251995675.0,
"step": 14130
},
{
"entropy": 0.946484375,
"epoch": 1.932549716394451,
"grad_norm": 0.691804762596156,
"learning_rate": 1.743694518810765e-07,
"loss": 0.6337,
"mean_token_accuracy": 0.8239383637905121,
"num_tokens": 1252902201.0,
"step": 14140
},
{
"entropy": 0.9576171875,
"epoch": 1.9339164901250598,
"grad_norm": 0.7847430629480141,
"learning_rate": 1.7084683669155982e-07,
"loss": 0.6092,
"mean_token_accuracy": 0.8298536509275436,
"num_tokens": 1253798317.0,
"step": 14150
},
{
"entropy": 0.9791015625,
"epoch": 1.9352832638556687,
"grad_norm": 0.752875285652692,
"learning_rate": 1.6732422150204313e-07,
"loss": 0.6634,
"mean_token_accuracy": 0.8204936802387237,
"num_tokens": 1254659906.0,
"step": 14160
},
{
"entropy": 0.9337890625,
"epoch": 1.9366500375862776,
"grad_norm": 0.7198438353682447,
"learning_rate": 1.6380160631252644e-07,
"loss": 0.6371,
"mean_token_accuracy": 0.8235215872526169,
"num_tokens": 1255597488.0,
"step": 14170
},
{
"entropy": 0.9412109375,
"epoch": 1.9380168113168865,
"grad_norm": 0.7250358831935214,
"learning_rate": 1.6027899112300975e-07,
"loss": 0.6419,
"mean_token_accuracy": 0.823494964838028,
"num_tokens": 1256472898.0,
"step": 14180
},
{
"entropy": 0.9427734375,
"epoch": 1.9393835850474954,
"grad_norm": 0.6788750510841337,
"learning_rate": 1.5675637593349303e-07,
"loss": 0.6114,
"mean_token_accuracy": 0.83153036236763,
"num_tokens": 1257391119.0,
"step": 14190
},
{
"entropy": 0.9400390625,
"epoch": 1.9407503587781043,
"grad_norm": 0.772457755936928,
"learning_rate": 1.5323376074397634e-07,
"loss": 0.666,
"mean_token_accuracy": 0.816737738251686,
"num_tokens": 1258295967.0,
"step": 14200
},
{
"entropy": 1.021875,
"epoch": 1.9421171325087132,
"grad_norm": 0.8920991151187041,
"learning_rate": 1.4971114555445965e-07,
"loss": 0.7334,
"mean_token_accuracy": 0.8041081428527832,
"num_tokens": 1259195309.0,
"step": 14210
},
{
"entropy": 0.955078125,
"epoch": 1.943483906239322,
"grad_norm": 0.8145372588192612,
"learning_rate": 1.4618853036494293e-07,
"loss": 0.6786,
"mean_token_accuracy": 0.8159528732299804,
"num_tokens": 1260067768.0,
"step": 14220
},
{
"entropy": 0.9537109375,
"epoch": 1.944850679969931,
"grad_norm": 0.7746855567799292,
"learning_rate": 1.4266591517542624e-07,
"loss": 0.6919,
"mean_token_accuracy": 0.8135550439357757,
"num_tokens": 1260932295.0,
"step": 14230
},
{
"entropy": 0.965234375,
"epoch": 1.94621745370054,
"grad_norm": 0.7229168290099631,
"learning_rate": 1.3914329998590955e-07,
"loss": 0.6248,
"mean_token_accuracy": 0.8277728199958801,
"num_tokens": 1261753749.0,
"step": 14240
},
{
"entropy": 0.960546875,
"epoch": 1.9475842274311488,
"grad_norm": 0.772367361778425,
"learning_rate": 1.3562068479639286e-07,
"loss": 0.6585,
"mean_token_accuracy": 0.8193142265081406,
"num_tokens": 1262672908.0,
"step": 14250
},
{
"entropy": 0.9263671875,
"epoch": 1.9489510011617577,
"grad_norm": 0.7719798073673975,
"learning_rate": 1.3209806960687614e-07,
"loss": 0.6276,
"mean_token_accuracy": 0.8267648875713348,
"num_tokens": 1263528802.0,
"step": 14260
},
{
"entropy": 0.9626953125,
"epoch": 1.9503177748923666,
"grad_norm": 0.8262659459515592,
"learning_rate": 1.2857545441735945e-07,
"loss": 0.6449,
"mean_token_accuracy": 0.8220154374837876,
"num_tokens": 1264405077.0,
"step": 14270
},
{
"entropy": 0.9701171875,
"epoch": 1.9516845486229755,
"grad_norm": 0.8239911541255627,
"learning_rate": 1.2505283922784276e-07,
"loss": 0.6621,
"mean_token_accuracy": 0.8212697088718415,
"num_tokens": 1265320781.0,
"step": 14280
},
{
"entropy": 0.9318359375,
"epoch": 1.9530513223535844,
"grad_norm": 0.767643714507108,
"learning_rate": 1.2153022403832607e-07,
"loss": 0.6509,
"mean_token_accuracy": 0.8210615873336792,
"num_tokens": 1266222145.0,
"step": 14290
},
{
"entropy": 0.9427734375,
"epoch": 1.9544180960841933,
"grad_norm": 0.7611833219401223,
"learning_rate": 1.1800760884880937e-07,
"loss": 0.6742,
"mean_token_accuracy": 0.8159293830394745,
"num_tokens": 1267147735.0,
"step": 14300
},
{
"entropy": 0.955078125,
"epoch": 1.9557848698148022,
"grad_norm": 0.7677147429884099,
"learning_rate": 1.1448499365929266e-07,
"loss": 0.6709,
"mean_token_accuracy": 0.8196497529745101,
"num_tokens": 1268033827.0,
"step": 14310
},
{
"entropy": 0.9625,
"epoch": 1.957151643545411,
"grad_norm": 0.7868368889358502,
"learning_rate": 1.1096237846977597e-07,
"loss": 0.6544,
"mean_token_accuracy": 0.8218078464269638,
"num_tokens": 1268910863.0,
"step": 14320
},
{
"entropy": 0.9765625,
"epoch": 1.95851841727602,
"grad_norm": 0.8675284210634553,
"learning_rate": 1.0743976328025928e-07,
"loss": 0.671,
"mean_token_accuracy": 0.8175516873598099,
"num_tokens": 1269814412.0,
"step": 14330
},
{
"entropy": 0.91015625,
"epoch": 1.9598851910066288,
"grad_norm": 0.7984330210815951,
"learning_rate": 1.0391714809074258e-07,
"loss": 0.6475,
"mean_token_accuracy": 0.8222610145807266,
"num_tokens": 1270716800.0,
"step": 14340
},
{
"entropy": 0.9423828125,
"epoch": 1.9612519647372377,
"grad_norm": 0.7934709267976787,
"learning_rate": 1.0039453290122588e-07,
"loss": 0.6706,
"mean_token_accuracy": 0.8168852418661118,
"num_tokens": 1271593925.0,
"step": 14350
},
{
"entropy": 0.920703125,
"epoch": 1.9626187384678466,
"grad_norm": 0.8098612069463952,
"learning_rate": 9.687191771170918e-08,
"loss": 0.6175,
"mean_token_accuracy": 0.829486507177353,
"num_tokens": 1272462660.0,
"step": 14360
},
{
"entropy": 0.9787109375,
"epoch": 1.9639855121984555,
"grad_norm": 0.8744751453199836,
"learning_rate": 9.334930252219248e-08,
"loss": 0.7289,
"mean_token_accuracy": 0.8059668332338333,
"num_tokens": 1273382850.0,
"step": 14370
},
{
"entropy": 0.9541015625,
"epoch": 1.9653522859290644,
"grad_norm": 0.7994812556441311,
"learning_rate": 8.982668733267578e-08,
"loss": 0.6426,
"mean_token_accuracy": 0.8243532568216324,
"num_tokens": 1274290572.0,
"step": 14380
},
{
"entropy": 0.9236328125,
"epoch": 1.9667190596596733,
"grad_norm": 0.7220431245769323,
"learning_rate": 8.630407214315909e-08,
"loss": 0.6312,
"mean_token_accuracy": 0.8280266433954239,
"num_tokens": 1275174274.0,
"step": 14390
},
{
"entropy": 0.9919921875,
"epoch": 1.9680858333902822,
"grad_norm": 0.858531493805102,
"learning_rate": 8.27814569536424e-08,
"loss": 0.6491,
"mean_token_accuracy": 0.8227692872285843,
"num_tokens": 1276074274.0,
"step": 14400
},
{
"entropy": 0.984375,
"epoch": 1.969452607120891,
"grad_norm": 0.8506750451421263,
"learning_rate": 7.925884176412568e-08,
"loss": 0.7149,
"mean_token_accuracy": 0.8099185764789582,
"num_tokens": 1276932931.0,
"step": 14410
},
{
"entropy": 0.994140625,
"epoch": 1.9708193808515,
"grad_norm": 0.7645206187554846,
"learning_rate": 7.573622657460899e-08,
"loss": 0.6852,
"mean_token_accuracy": 0.8149797081947326,
"num_tokens": 1277813012.0,
"step": 14420
},
{
"entropy": 0.9375,
"epoch": 1.9721861545821089,
"grad_norm": 0.7859166807999547,
"learning_rate": 7.22136113850923e-08,
"loss": 0.6444,
"mean_token_accuracy": 0.8234649926424027,
"num_tokens": 1278702374.0,
"step": 14430
},
{
"entropy": 0.9388671875,
"epoch": 1.9735529283127178,
"grad_norm": 0.7816118413065852,
"learning_rate": 6.869099619557561e-08,
"loss": 0.6227,
"mean_token_accuracy": 0.8284009456634521,
"num_tokens": 1279586770.0,
"step": 14440
},
{
"entropy": 0.9357421875,
"epoch": 1.9749197020433267,
"grad_norm": 0.9597381003850985,
"learning_rate": 6.51683810060589e-08,
"loss": 0.6612,
"mean_token_accuracy": 0.8208856165409089,
"num_tokens": 1280510572.0,
"step": 14450
},
{
"entropy": 0.9767578125,
"epoch": 1.9762864757739358,
"grad_norm": 0.8493823240162809,
"learning_rate": 6.164576581654221e-08,
"loss": 0.6801,
"mean_token_accuracy": 0.8152142941951752,
"num_tokens": 1281364572.0,
"step": 14460
},
{
"entropy": 0.95390625,
"epoch": 1.9776532495045447,
"grad_norm": 0.8334782402277503,
"learning_rate": 5.8123150627025515e-08,
"loss": 0.6411,
"mean_token_accuracy": 0.8226537823677063,
"num_tokens": 1282224524.0,
"step": 14470
},
{
"entropy": 0.951171875,
"epoch": 1.9790200232351536,
"grad_norm": 0.7924940881943815,
"learning_rate": 5.460053543750881e-08,
"loss": 0.649,
"mean_token_accuracy": 0.8234307199716568,
"num_tokens": 1283046842.0,
"step": 14480
},
{
"entropy": 0.966015625,
"epoch": 1.9803867969657625,
"grad_norm": 0.8320536725988,
"learning_rate": 5.107792024799211e-08,
"loss": 0.6838,
"mean_token_accuracy": 0.8158549159765244,
"num_tokens": 1283926789.0,
"step": 14490
},
{
"entropy": 1.0130859375,
"epoch": 1.9817535706963714,
"grad_norm": 0.8150037205059998,
"learning_rate": 4.7555305058475415e-08,
"loss": 0.7204,
"mean_token_accuracy": 0.8063102692365647,
"num_tokens": 1284778482.0,
"step": 14500
},
{
"entropy": 0.9533203125,
"epoch": 1.9831203444269803,
"grad_norm": 0.7518599002709906,
"learning_rate": 4.403268986895872e-08,
"loss": 0.6779,
"mean_token_accuracy": 0.817624905705452,
"num_tokens": 1285717361.0,
"step": 14510
},
{
"entropy": 0.957421875,
"epoch": 1.9844871181575892,
"grad_norm": 0.8225036063448422,
"learning_rate": 4.0510074679442026e-08,
"loss": 0.6745,
"mean_token_accuracy": 0.816216179728508,
"num_tokens": 1286605150.0,
"step": 14520
},
{
"entropy": 0.988671875,
"epoch": 1.985853891888198,
"grad_norm": 0.8582795185245784,
"learning_rate": 3.698745948992532e-08,
"loss": 0.6996,
"mean_token_accuracy": 0.8121593505144119,
"num_tokens": 1287429714.0,
"step": 14530
},
{
"entropy": 0.95,
"epoch": 1.987220665618807,
"grad_norm": 0.7884845356115721,
"learning_rate": 3.3464844300408624e-08,
"loss": 0.6465,
"mean_token_accuracy": 0.8228154748678207,
"num_tokens": 1288274856.0,
"step": 14540
},
{
"entropy": 0.92890625,
"epoch": 1.9885874393494158,
"grad_norm": 0.7767235176594873,
"learning_rate": 2.994222911089193e-08,
"loss": 0.6362,
"mean_token_accuracy": 0.8253120213747025,
"num_tokens": 1289184678.0,
"step": 14550
},
{
"entropy": 0.9712890625,
"epoch": 1.9899542130800247,
"grad_norm": 0.7790827150409058,
"learning_rate": 2.6419613921375232e-08,
"loss": 0.6539,
"mean_token_accuracy": 0.8205474495887757,
"num_tokens": 1290104163.0,
"step": 14560
},
{
"entropy": 0.9138671875,
"epoch": 1.9913209868106336,
"grad_norm": 0.8289755397749429,
"learning_rate": 2.289699873185853e-08,
"loss": 0.6493,
"mean_token_accuracy": 0.8209035515785217,
"num_tokens": 1290984420.0,
"step": 14570
},
{
"entropy": 1.02109375,
"epoch": 1.9926877605412425,
"grad_norm": 0.7621902391937213,
"learning_rate": 1.9374383542341837e-08,
"loss": 0.7211,
"mean_token_accuracy": 0.8061020493507385,
"num_tokens": 1291827527.0,
"step": 14580
},
{
"entropy": 0.9134765625,
"epoch": 1.9940545342718514,
"grad_norm": 0.7625955242488311,
"learning_rate": 1.585176835282514e-08,
"loss": 0.6191,
"mean_token_accuracy": 0.8286906272172928,
"num_tokens": 1292724554.0,
"step": 14590
},
{
"entropy": 0.9986328125,
"epoch": 1.9954213080024603,
"grad_norm": 0.8101909387850501,
"learning_rate": 1.2329153163308442e-08,
"loss": 0.7269,
"mean_token_accuracy": 0.8060611605644226,
"num_tokens": 1293618300.0,
"step": 14600
},
{
"entropy": 0.9521484375,
"epoch": 1.9967880817330692,
"grad_norm": 0.8266510589625489,
"learning_rate": 8.806537973791744e-09,
"loss": 0.6728,
"mean_token_accuracy": 0.8170237243175507,
"num_tokens": 1294534796.0,
"step": 14610
},
{
"entropy": 0.93671875,
"epoch": 1.998154855463678,
"grad_norm": 0.8158418148993896,
"learning_rate": 5.2839227842750465e-09,
"loss": 0.6585,
"mean_token_accuracy": 0.8195945680141449,
"num_tokens": 1295448543.0,
"step": 14620
},
{
"entropy": 0.9552734375,
"epoch": 1.999521629194287,
"grad_norm": 0.7792009234166845,
"learning_rate": 1.7613075947583486e-09,
"loss": 0.6714,
"mean_token_accuracy": 0.8161415457725525,
"num_tokens": 1296340286.0,
"step": 14630
},
{
"entropy": 0.9732142857142857,
"epoch": 2.0,
"mean_token_accuracy": 0.8097479854311261,
"num_tokens": 1296663702.0,
"step": 14634,
"total_flos": 2054605533544448.0,
"train_loss": 0.6844859028735806,
"train_runtime": 36489.0201,
"train_samples_per_second": 51.331,
"train_steps_per_second": 0.401
}
],
"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": 2054605533544448.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink