fpadovani/hin-deva-10mb-after-ppt-Dp-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
374d25c765a0b11caeb9f3108d2e59a4d97ffa9c
hin-deva-10mb-after-ppt-Dp-…/checkpoint-6000/trainer_state.json
ModelHub XC 374d25c765 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/hin-deva-10mb-after-ppt-Dp-100mb-ckpt500_seed3407
Source: Original Platform
2026-07-10 05:50:13 +08:00

12057 lines
329 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5763688760806917,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 7.307085371017456,
"epoch": 0.0004803073967339097,
"grad_norm": 1.203125,
"learning_rate": 2e-06,
"loss": 6.6711,
"mean_token_accuracy": 0.11963684856891632,
"num_tokens": 10855.0,
"step": 5
},
{
"entropy": 7.261483955383301,
"epoch": 0.0009606147934678194,
"grad_norm": 1.09375,
"learning_rate": 4.5e-06,
"loss": 6.611,
"mean_token_accuracy": 0.12699446231126785,
"num_tokens": 24110.0,
"step": 10
},
{
"entropy": 7.322568655014038,
"epoch": 0.001440922190201729,
"grad_norm": 1.171875,
"learning_rate": 7e-06,
"loss": 6.712,
"mean_token_accuracy": 0.11978077217936516,
"num_tokens": 35984.0,
"step": 15
},
{
"entropy": 7.314567613601684,
"epoch": 0.0019212295869356388,
"grad_norm": 1.0859375,
"learning_rate": 9.5e-06,
"loss": 6.8292,
"mean_token_accuracy": 0.11400551423430443,
"num_tokens": 48152.0,
"step": 20
},
{
"entropy": 7.27342529296875,
"epoch": 0.0024015369836695487,
"grad_norm": 1.125,
"learning_rate": 1.2e-05,
"loss": 6.7211,
"mean_token_accuracy": 0.12296620458364486,
"num_tokens": 59810.0,
"step": 25
},
{
"entropy": 7.184682607650757,
"epoch": 0.002881844380403458,
"grad_norm": 1.0,
"learning_rate": 1.4500000000000002e-05,
"loss": 6.5527,
"mean_token_accuracy": 0.12848290130496026,
"num_tokens": 70852.0,
"step": 30
},
{
"entropy": 7.165819931030273,
"epoch": 0.0033621517771373678,
"grad_norm": 1.1171875,
"learning_rate": 1.7000000000000003e-05,
"loss": 6.6676,
"mean_token_accuracy": 0.12854331284761428,
"num_tokens": 83378.0,
"step": 35
},
{
"entropy": 7.211078310012818,
"epoch": 0.0038424591738712775,
"grad_norm": 1.0859375,
"learning_rate": 1.95e-05,
"loss": 6.7318,
"mean_token_accuracy": 0.12352305576205254,
"num_tokens": 95505.0,
"step": 40
},
{
"entropy": 7.143542957305908,
"epoch": 0.004322766570605188,
"grad_norm": 0.94921875,
"learning_rate": 2.2e-05,
"loss": 6.6624,
"mean_token_accuracy": 0.13083189800381662,
"num_tokens": 106812.0,
"step": 45
},
{
"entropy": 7.1696985244750975,
"epoch": 0.004803073967339097,
"grad_norm": 0.94140625,
"learning_rate": 2.4500000000000003e-05,
"loss": 6.6905,
"mean_token_accuracy": 0.12702840194106102,
"num_tokens": 118572.0,
"step": 50
},
{
"entropy": 7.134132242202758,
"epoch": 0.005283381364073006,
"grad_norm": 1.0234375,
"learning_rate": 2.7e-05,
"loss": 6.6333,
"mean_token_accuracy": 0.13059643805027008,
"num_tokens": 130051.0,
"step": 55
},
{
"entropy": 7.118882560729981,
"epoch": 0.005763688760806916,
"grad_norm": 1.1015625,
"learning_rate": 2.95e-05,
"loss": 6.7479,
"mean_token_accuracy": 0.12762305215001107,
"num_tokens": 141920.0,
"step": 60
},
{
"entropy": 7.106608819961548,
"epoch": 0.006243996157540826,
"grad_norm": 0.91015625,
"learning_rate": 3.2e-05,
"loss": 6.6002,
"mean_token_accuracy": 0.13259410262107849,
"num_tokens": 152706.0,
"step": 65
},
{
"entropy": 7.095578145980835,
"epoch": 0.0067243035542747355,
"grad_norm": 0.9140625,
"learning_rate": 3.4500000000000005e-05,
"loss": 6.6396,
"mean_token_accuracy": 0.1325068600475788,
"num_tokens": 165253.0,
"step": 70
},
{
"entropy": 7.086773347854614,
"epoch": 0.007204610951008645,
"grad_norm": 0.88671875,
"learning_rate": 3.7e-05,
"loss": 6.5959,
"mean_token_accuracy": 0.13050568848848343,
"num_tokens": 176708.0,
"step": 75
},
{
"entropy": 7.126611709594727,
"epoch": 0.007684918347742555,
"grad_norm": 1.125,
"learning_rate": 3.95e-05,
"loss": 6.69,
"mean_token_accuracy": 0.1262853056192398,
"num_tokens": 188240.0,
"step": 80
},
{
"entropy": 7.042779111862183,
"epoch": 0.008165225744476465,
"grad_norm": 1.078125,
"learning_rate": 4.2000000000000004e-05,
"loss": 6.496,
"mean_token_accuracy": 0.14095853343605996,
"num_tokens": 198355.0,
"step": 85
},
{
"entropy": 7.078471279144287,
"epoch": 0.008645533141210375,
"grad_norm": 1.125,
"learning_rate": 4.45e-05,
"loss": 6.586,
"mean_token_accuracy": 0.12663363963365554,
"num_tokens": 209497.0,
"step": 90
},
{
"entropy": 7.093791770935058,
"epoch": 0.009125840537944284,
"grad_norm": 1.09375,
"learning_rate": 4.7000000000000004e-05,
"loss": 6.6097,
"mean_token_accuracy": 0.1298912487924099,
"num_tokens": 220859.0,
"step": 95
},
{
"entropy": 7.139126348495483,
"epoch": 0.009606147934678195,
"grad_norm": 1.015625,
"learning_rate": 4.9500000000000004e-05,
"loss": 6.7121,
"mean_token_accuracy": 0.12592171207070352,
"num_tokens": 231550.0,
"step": 100
},
{
"entropy": 7.0799497127532955,
"epoch": 0.010086455331412104,
"grad_norm": 0.96875,
"learning_rate": 5.2e-05,
"loss": 6.6082,
"mean_token_accuracy": 0.12669628784060477,
"num_tokens": 244210.0,
"step": 105
},
{
"entropy": 7.108440780639649,
"epoch": 0.010566762728146013,
"grad_norm": 1.03125,
"learning_rate": 5.45e-05,
"loss": 6.5846,
"mean_token_accuracy": 0.13026389181613923,
"num_tokens": 255745.0,
"step": 110
},
{
"entropy": 7.094820356369018,
"epoch": 0.011047070124879923,
"grad_norm": 1.1328125,
"learning_rate": 5.7e-05,
"loss": 6.581,
"mean_token_accuracy": 0.1345028355717659,
"num_tokens": 266180.0,
"step": 115
},
{
"entropy": 7.168249034881592,
"epoch": 0.011527377521613832,
"grad_norm": 1.0,
"learning_rate": 5.9499999999999996e-05,
"loss": 6.6584,
"mean_token_accuracy": 0.13001366779208184,
"num_tokens": 277736.0,
"step": 120
},
{
"entropy": 7.125609493255615,
"epoch": 0.012007684918347743,
"grad_norm": 1.234375,
"learning_rate": 6.2e-05,
"loss": 6.6671,
"mean_token_accuracy": 0.1278264619410038,
"num_tokens": 289069.0,
"step": 125
},
{
"entropy": 7.0424315452575685,
"epoch": 0.012487992315081652,
"grad_norm": 0.95703125,
"learning_rate": 6.450000000000001e-05,
"loss": 6.6483,
"mean_token_accuracy": 0.1312246046960354,
"num_tokens": 300240.0,
"step": 130
},
{
"entropy": 7.105067348480224,
"epoch": 0.012968299711815562,
"grad_norm": 1.0625,
"learning_rate": 6.7e-05,
"loss": 6.6951,
"mean_token_accuracy": 0.13201999366283418,
"num_tokens": 311698.0,
"step": 135
},
{
"entropy": 7.135102462768555,
"epoch": 0.013448607108549471,
"grad_norm": 1.03125,
"learning_rate": 6.950000000000001e-05,
"loss": 6.652,
"mean_token_accuracy": 0.129649805277586,
"num_tokens": 322844.0,
"step": 140
},
{
"entropy": 7.104988622665405,
"epoch": 0.013928914505283382,
"grad_norm": 1.03125,
"learning_rate": 7.2e-05,
"loss": 6.6544,
"mean_token_accuracy": 0.13054293021559715,
"num_tokens": 335092.0,
"step": 145
},
{
"entropy": 7.036244344711304,
"epoch": 0.01440922190201729,
"grad_norm": 0.97265625,
"learning_rate": 7.45e-05,
"loss": 6.6388,
"mean_token_accuracy": 0.13229672610759735,
"num_tokens": 347033.0,
"step": 150
},
{
"entropy": 7.0629229068756105,
"epoch": 0.014889529298751201,
"grad_norm": 1.1953125,
"learning_rate": 7.7e-05,
"loss": 6.6368,
"mean_token_accuracy": 0.12350815683603286,
"num_tokens": 358696.0,
"step": 155
},
{
"entropy": 7.103863048553467,
"epoch": 0.01536983669548511,
"grad_norm": 1.1171875,
"learning_rate": 7.950000000000001e-05,
"loss": 6.6527,
"mean_token_accuracy": 0.12880626693367958,
"num_tokens": 369390.0,
"step": 160
},
{
"entropy": 7.0600316524505615,
"epoch": 0.01585014409221902,
"grad_norm": 1.015625,
"learning_rate": 8.2e-05,
"loss": 6.5942,
"mean_token_accuracy": 0.12720233500003814,
"num_tokens": 380540.0,
"step": 165
},
{
"entropy": 7.0939164638519285,
"epoch": 0.01633045148895293,
"grad_norm": 1.15625,
"learning_rate": 8.450000000000001e-05,
"loss": 6.5928,
"mean_token_accuracy": 0.1307683579623699,
"num_tokens": 391243.0,
"step": 170
},
{
"entropy": 7.08874363899231,
"epoch": 0.01681075888568684,
"grad_norm": 0.94921875,
"learning_rate": 8.7e-05,
"loss": 6.5721,
"mean_token_accuracy": 0.1347414717078209,
"num_tokens": 403336.0,
"step": 175
},
{
"entropy": 7.054573011398316,
"epoch": 0.01729106628242075,
"grad_norm": 1.21875,
"learning_rate": 8.95e-05,
"loss": 6.5372,
"mean_token_accuracy": 0.1348336383700371,
"num_tokens": 413886.0,
"step": 180
},
{
"entropy": 7.076615619659424,
"epoch": 0.01777137367915466,
"grad_norm": 1.2265625,
"learning_rate": 9.2e-05,
"loss": 6.6369,
"mean_token_accuracy": 0.12773096263408662,
"num_tokens": 425277.0,
"step": 185
},
{
"entropy": 7.169283103942871,
"epoch": 0.01825168107588857,
"grad_norm": 1.09375,
"learning_rate": 9.45e-05,
"loss": 6.6817,
"mean_token_accuracy": 0.13388459086418153,
"num_tokens": 436868.0,
"step": 190
},
{
"entropy": 7.070893096923828,
"epoch": 0.018731988472622477,
"grad_norm": 1.1328125,
"learning_rate": 9.7e-05,
"loss": 6.5585,
"mean_token_accuracy": 0.13712915256619454,
"num_tokens": 448349.0,
"step": 195
},
{
"entropy": 7.00331335067749,
"epoch": 0.01921229586935639,
"grad_norm": 1.1015625,
"learning_rate": 9.95e-05,
"loss": 6.5787,
"mean_token_accuracy": 0.13295828625559808,
"num_tokens": 459447.0,
"step": 200
},
{
"entropy": 7.017018413543701,
"epoch": 0.0196926032660903,
"grad_norm": 1.09375,
"learning_rate": 0.000102,
"loss": 6.5664,
"mean_token_accuracy": 0.13398413211107255,
"num_tokens": 470734.0,
"step": 205
},
{
"entropy": 7.074815893173218,
"epoch": 0.020172910662824207,
"grad_norm": 1.1484375,
"learning_rate": 0.00010449999999999999,
"loss": 6.6675,
"mean_token_accuracy": 0.1243060939013958,
"num_tokens": 482015.0,
"step": 210
},
{
"entropy": 7.051287984848022,
"epoch": 0.020653218059558116,
"grad_norm": 1.0703125,
"learning_rate": 0.000107,
"loss": 6.5853,
"mean_token_accuracy": 0.12988396733999252,
"num_tokens": 493339.0,
"step": 215
},
{
"entropy": 7.021688842773438,
"epoch": 0.021133525456292025,
"grad_norm": 1.1484375,
"learning_rate": 0.0001095,
"loss": 6.5124,
"mean_token_accuracy": 0.13405952081084252,
"num_tokens": 504924.0,
"step": 220
},
{
"entropy": 7.087656021118164,
"epoch": 0.021613832853025938,
"grad_norm": 0.9921875,
"learning_rate": 0.000112,
"loss": 6.5272,
"mean_token_accuracy": 0.12774203419685365,
"num_tokens": 516603.0,
"step": 225
},
{
"entropy": 6.912145376205444,
"epoch": 0.022094140249759846,
"grad_norm": 1.15625,
"learning_rate": 0.0001145,
"loss": 6.4572,
"mean_token_accuracy": 0.13284713625907899,
"num_tokens": 528347.0,
"step": 230
},
{
"entropy": 7.009493589401245,
"epoch": 0.022574447646493755,
"grad_norm": 1.09375,
"learning_rate": 0.00011700000000000001,
"loss": 6.5941,
"mean_token_accuracy": 0.13424275815486908,
"num_tokens": 539328.0,
"step": 235
},
{
"entropy": 7.143858766555786,
"epoch": 0.023054755043227664,
"grad_norm": 1.171875,
"learning_rate": 0.00011949999999999999,
"loss": 6.5522,
"mean_token_accuracy": 0.1398538187146187,
"num_tokens": 549297.0,
"step": 240
},
{
"entropy": 6.905237674713135,
"epoch": 0.023535062439961577,
"grad_norm": 1.15625,
"learning_rate": 0.000122,
"loss": 6.545,
"mean_token_accuracy": 0.13216322883963585,
"num_tokens": 560306.0,
"step": 245
},
{
"entropy": 7.09604697227478,
"epoch": 0.024015369836695485,
"grad_norm": 1.2265625,
"learning_rate": 0.0001245,
"loss": 6.6323,
"mean_token_accuracy": 0.1321306437253952,
"num_tokens": 571972.0,
"step": 250
},
{
"entropy": 7.053035020828247,
"epoch": 0.024495677233429394,
"grad_norm": 1.21875,
"learning_rate": 0.000127,
"loss": 6.5172,
"mean_token_accuracy": 0.13756523206830024,
"num_tokens": 582962.0,
"step": 255
},
{
"entropy": 6.858367395401001,
"epoch": 0.024975984630163303,
"grad_norm": 1.0,
"learning_rate": 0.0001295,
"loss": 6.5103,
"mean_token_accuracy": 0.13402715027332307,
"num_tokens": 597193.0,
"step": 260
},
{
"entropy": 7.156196880340576,
"epoch": 0.025456292026897216,
"grad_norm": 1.1171875,
"learning_rate": 0.000132,
"loss": 6.5892,
"mean_token_accuracy": 0.12901883125305175,
"num_tokens": 608982.0,
"step": 265
},
{
"entropy": 7.014424085617065,
"epoch": 0.025936599423631124,
"grad_norm": 1.140625,
"learning_rate": 0.00013450000000000002,
"loss": 6.6212,
"mean_token_accuracy": 0.12978230491280557,
"num_tokens": 619953.0,
"step": 270
},
{
"entropy": 7.096269750595093,
"epoch": 0.026416906820365033,
"grad_norm": 1.0703125,
"learning_rate": 0.00013700000000000002,
"loss": 6.6321,
"mean_token_accuracy": 0.12898936718702317,
"num_tokens": 631039.0,
"step": 275
},
{
"entropy": 7.03494119644165,
"epoch": 0.026897214217098942,
"grad_norm": 1.1015625,
"learning_rate": 0.0001395,
"loss": 6.6062,
"mean_token_accuracy": 0.13531998470425605,
"num_tokens": 642656.0,
"step": 280
},
{
"entropy": 6.969197034835815,
"epoch": 0.027377521613832854,
"grad_norm": 1.15625,
"learning_rate": 0.00014199999999999998,
"loss": 6.4944,
"mean_token_accuracy": 0.1461842603981495,
"num_tokens": 653748.0,
"step": 285
},
{
"entropy": 6.980546951293945,
"epoch": 0.027857829010566763,
"grad_norm": 1.2265625,
"learning_rate": 0.0001445,
"loss": 6.4709,
"mean_token_accuracy": 0.1350526325404644,
"num_tokens": 665618.0,
"step": 290
},
{
"entropy": 6.976346349716186,
"epoch": 0.028338136407300672,
"grad_norm": 1.1171875,
"learning_rate": 0.000147,
"loss": 6.5103,
"mean_token_accuracy": 0.13303453028202056,
"num_tokens": 677329.0,
"step": 295
},
{
"entropy": 6.96361517906189,
"epoch": 0.02881844380403458,
"grad_norm": 1.1171875,
"learning_rate": 0.0001495,
"loss": 6.5113,
"mean_token_accuracy": 0.1333675727248192,
"num_tokens": 688278.0,
"step": 300
},
{
"entropy": 6.986789083480835,
"epoch": 0.029298751200768493,
"grad_norm": 1.1171875,
"learning_rate": 0.000152,
"loss": 6.5677,
"mean_token_accuracy": 0.13771163299679756,
"num_tokens": 700739.0,
"step": 305
},
{
"entropy": 6.966262865066528,
"epoch": 0.029779058597502402,
"grad_norm": 1.15625,
"learning_rate": 0.00015450000000000001,
"loss": 6.5046,
"mean_token_accuracy": 0.13294478878378868,
"num_tokens": 712527.0,
"step": 310
},
{
"entropy": 6.959698724746704,
"epoch": 0.03025936599423631,
"grad_norm": 1.0625,
"learning_rate": 0.000157,
"loss": 6.5456,
"mean_token_accuracy": 0.13097512274980544,
"num_tokens": 724514.0,
"step": 315
},
{
"entropy": 7.038305854797363,
"epoch": 0.03073967339097022,
"grad_norm": 1.2265625,
"learning_rate": 0.0001595,
"loss": 6.5092,
"mean_token_accuracy": 0.13729717060923577,
"num_tokens": 735679.0,
"step": 320
},
{
"entropy": 6.877438640594482,
"epoch": 0.03121998078770413,
"grad_norm": 0.984375,
"learning_rate": 0.000162,
"loss": 6.4986,
"mean_token_accuracy": 0.13833012878894807,
"num_tokens": 747896.0,
"step": 325
},
{
"entropy": 7.030805873870849,
"epoch": 0.03170028818443804,
"grad_norm": 1.125,
"learning_rate": 0.00016450000000000001,
"loss": 6.5506,
"mean_token_accuracy": 0.13383892253041269,
"num_tokens": 759081.0,
"step": 330
},
{
"entropy": 6.936169385910034,
"epoch": 0.03218059558117195,
"grad_norm": 1.234375,
"learning_rate": 0.00016700000000000002,
"loss": 6.502,
"mean_token_accuracy": 0.1336835578083992,
"num_tokens": 770459.0,
"step": 335
},
{
"entropy": 6.922156190872192,
"epoch": 0.03266090297790586,
"grad_norm": 1.0546875,
"learning_rate": 0.00016950000000000003,
"loss": 6.4557,
"mean_token_accuracy": 0.13467910513281822,
"num_tokens": 783960.0,
"step": 340
},
{
"entropy": 6.945554304122925,
"epoch": 0.03314121037463977,
"grad_norm": 1.1484375,
"learning_rate": 0.00017199999999999998,
"loss": 6.5724,
"mean_token_accuracy": 0.13674034401774407,
"num_tokens": 795425.0,
"step": 345
},
{
"entropy": 6.9888596534729,
"epoch": 0.03362151777137368,
"grad_norm": 1.2578125,
"learning_rate": 0.00017449999999999999,
"loss": 6.5241,
"mean_token_accuracy": 0.13427431806921958,
"num_tokens": 807536.0,
"step": 350
},
{
"entropy": 7.014977931976318,
"epoch": 0.034101825168107586,
"grad_norm": 1.1796875,
"learning_rate": 0.000177,
"loss": 6.5882,
"mean_token_accuracy": 0.13020211309194565,
"num_tokens": 818801.0,
"step": 355
},
{
"entropy": 6.9366690158844,
"epoch": 0.0345821325648415,
"grad_norm": 1.2265625,
"learning_rate": 0.0001795,
"loss": 6.4507,
"mean_token_accuracy": 0.13554891422390938,
"num_tokens": 831497.0,
"step": 360
},
{
"entropy": 6.8712386131286625,
"epoch": 0.03506243996157541,
"grad_norm": 1.0390625,
"learning_rate": 0.000182,
"loss": 6.4424,
"mean_token_accuracy": 0.13640913739800453,
"num_tokens": 842491.0,
"step": 365
},
{
"entropy": 7.009519624710083,
"epoch": 0.03554274735830932,
"grad_norm": 1.1171875,
"learning_rate": 0.0001845,
"loss": 6.523,
"mean_token_accuracy": 0.12829140126705169,
"num_tokens": 854560.0,
"step": 370
},
{
"entropy": 6.867582225799561,
"epoch": 0.03602305475504323,
"grad_norm": 1.1015625,
"learning_rate": 0.000187,
"loss": 6.4306,
"mean_token_accuracy": 0.14245563745498657,
"num_tokens": 866688.0,
"step": 375
},
{
"entropy": 6.9899519920349125,
"epoch": 0.03650336215177714,
"grad_norm": 1.15625,
"learning_rate": 0.0001895,
"loss": 6.5572,
"mean_token_accuracy": 0.12518782839179038,
"num_tokens": 879484.0,
"step": 380
},
{
"entropy": 6.972244501113892,
"epoch": 0.036983669548511046,
"grad_norm": 1.25,
"learning_rate": 0.000192,
"loss": 6.5485,
"mean_token_accuracy": 0.12844221517443657,
"num_tokens": 890807.0,
"step": 385
},
{
"entropy": 6.982624101638794,
"epoch": 0.037463976945244955,
"grad_norm": 1.359375,
"learning_rate": 0.0001945,
"loss": 6.4655,
"mean_token_accuracy": 0.1351318120956421,
"num_tokens": 901759.0,
"step": 390
},
{
"entropy": 6.820593404769897,
"epoch": 0.037944284341978864,
"grad_norm": 1.2421875,
"learning_rate": 0.00019700000000000002,
"loss": 6.4239,
"mean_token_accuracy": 0.13946242481470109,
"num_tokens": 912212.0,
"step": 395
},
{
"entropy": 6.997915410995484,
"epoch": 0.03842459173871278,
"grad_norm": 1.1953125,
"learning_rate": 0.00019950000000000002,
"loss": 6.5082,
"mean_token_accuracy": 0.1373220317065716,
"num_tokens": 923947.0,
"step": 400
},
{
"entropy": 6.887757110595703,
"epoch": 0.03890489913544669,
"grad_norm": 1.2578125,
"learning_rate": 0.000202,
"loss": 6.4624,
"mean_token_accuracy": 0.13495285138487817,
"num_tokens": 935732.0,
"step": 405
},
{
"entropy": 6.860993576049805,
"epoch": 0.0393852065321806,
"grad_norm": 1.3203125,
"learning_rate": 0.00020449999999999998,
"loss": 6.4234,
"mean_token_accuracy": 0.13301072269678116,
"num_tokens": 948261.0,
"step": 410
},
{
"entropy": 6.998539113998413,
"epoch": 0.039865513928914506,
"grad_norm": 1.359375,
"learning_rate": 0.000207,
"loss": 6.5408,
"mean_token_accuracy": 0.13510932326316832,
"num_tokens": 959574.0,
"step": 415
},
{
"entropy": 6.97111144065857,
"epoch": 0.040345821325648415,
"grad_norm": 1.09375,
"learning_rate": 0.0002095,
"loss": 6.6136,
"mean_token_accuracy": 0.12621866688132286,
"num_tokens": 970329.0,
"step": 420
},
{
"entropy": 6.985296583175659,
"epoch": 0.040826128722382324,
"grad_norm": 1.1640625,
"learning_rate": 0.000212,
"loss": 6.4701,
"mean_token_accuracy": 0.14006434381008148,
"num_tokens": 982037.0,
"step": 425
},
{
"entropy": 6.804930353164673,
"epoch": 0.04130643611911623,
"grad_norm": 1.078125,
"learning_rate": 0.0002145,
"loss": 6.4271,
"mean_token_accuracy": 0.13567045107483863,
"num_tokens": 994612.0,
"step": 430
},
{
"entropy": 6.955875444412231,
"epoch": 0.04178674351585014,
"grad_norm": 1.078125,
"learning_rate": 0.00021700000000000002,
"loss": 6.5617,
"mean_token_accuracy": 0.13497636914253236,
"num_tokens": 1005960.0,
"step": 435
},
{
"entropy": 6.869253540039063,
"epoch": 0.04226705091258405,
"grad_norm": 1.0625,
"learning_rate": 0.0002195,
"loss": 6.4174,
"mean_token_accuracy": 0.13668350800871848,
"num_tokens": 1017618.0,
"step": 440
},
{
"entropy": 6.8875102519989015,
"epoch": 0.042747358309317966,
"grad_norm": 1.1953125,
"learning_rate": 0.000222,
"loss": 6.48,
"mean_token_accuracy": 0.13301057890057563,
"num_tokens": 1029307.0,
"step": 445
},
{
"entropy": 6.912690162658691,
"epoch": 0.043227665706051875,
"grad_norm": 1.3125,
"learning_rate": 0.0002245,
"loss": 6.4972,
"mean_token_accuracy": 0.13038710951805116,
"num_tokens": 1042027.0,
"step": 450
},
{
"entropy": 6.889484977722168,
"epoch": 0.043707973102785784,
"grad_norm": 1.140625,
"learning_rate": 0.00022700000000000002,
"loss": 6.4819,
"mean_token_accuracy": 0.13361246138811111,
"num_tokens": 1053125.0,
"step": 455
},
{
"entropy": 6.828107833862305,
"epoch": 0.04418828049951969,
"grad_norm": 1.171875,
"learning_rate": 0.00022950000000000002,
"loss": 6.4905,
"mean_token_accuracy": 0.13674674183130264,
"num_tokens": 1064908.0,
"step": 460
},
{
"entropy": 6.9115313529968265,
"epoch": 0.0446685878962536,
"grad_norm": 1.203125,
"learning_rate": 0.00023200000000000003,
"loss": 6.3581,
"mean_token_accuracy": 0.14502883404493333,
"num_tokens": 1076328.0,
"step": 465
},
{
"entropy": 6.896181726455689,
"epoch": 0.04514889529298751,
"grad_norm": 1.328125,
"learning_rate": 0.00023449999999999998,
"loss": 6.5895,
"mean_token_accuracy": 0.1260202206671238,
"num_tokens": 1088469.0,
"step": 470
},
{
"entropy": 6.823418760299683,
"epoch": 0.04562920268972142,
"grad_norm": 1.1875,
"learning_rate": 0.000237,
"loss": 6.391,
"mean_token_accuracy": 0.1406915917992592,
"num_tokens": 1099408.0,
"step": 475
},
{
"entropy": 6.765524959564209,
"epoch": 0.04610951008645533,
"grad_norm": 1.2109375,
"learning_rate": 0.0002395,
"loss": 6.3342,
"mean_token_accuracy": 0.14521979689598083,
"num_tokens": 1111101.0,
"step": 480
},
{
"entropy": 6.744671630859375,
"epoch": 0.046589817483189244,
"grad_norm": 1.15625,
"learning_rate": 0.000242,
"loss": 6.3479,
"mean_token_accuracy": 0.14791416078805925,
"num_tokens": 1122877.0,
"step": 485
},
{
"entropy": 6.7991739273071286,
"epoch": 0.04707012487992315,
"grad_norm": 1.25,
"learning_rate": 0.0002445,
"loss": 6.454,
"mean_token_accuracy": 0.1279207557439804,
"num_tokens": 1133956.0,
"step": 490
},
{
"entropy": 6.823109483718872,
"epoch": 0.04755043227665706,
"grad_norm": 1.2265625,
"learning_rate": 0.000247,
"loss": 6.3641,
"mean_token_accuracy": 0.13983470499515532,
"num_tokens": 1146101.0,
"step": 495
},
{
"entropy": 6.855337381362915,
"epoch": 0.04803073967339097,
"grad_norm": 1.296875,
"learning_rate": 0.0002495,
"loss": 6.4502,
"mean_token_accuracy": 0.14883086159825326,
"num_tokens": 1157432.0,
"step": 500
},
{
"entropy": 6.822028875350952,
"epoch": 0.04851104707012488,
"grad_norm": 1.1953125,
"learning_rate": 0.000252,
"loss": 6.4439,
"mean_token_accuracy": 0.14014339298009873,
"num_tokens": 1167601.0,
"step": 505
},
{
"entropy": 6.793749713897705,
"epoch": 0.04899135446685879,
"grad_norm": 1.1015625,
"learning_rate": 0.0002545,
"loss": 6.3433,
"mean_token_accuracy": 0.14071932584047317,
"num_tokens": 1178818.0,
"step": 510
},
{
"entropy": 6.961408185958862,
"epoch": 0.0494716618635927,
"grad_norm": 1.109375,
"learning_rate": 0.000257,
"loss": 6.5897,
"mean_token_accuracy": 0.13712238147854805,
"num_tokens": 1189977.0,
"step": 515
},
{
"entropy": 6.825463342666626,
"epoch": 0.049951969260326606,
"grad_norm": 1.1640625,
"learning_rate": 0.0002595,
"loss": 6.4273,
"mean_token_accuracy": 0.13879103511571883,
"num_tokens": 1201039.0,
"step": 520
},
{
"entropy": 6.885947799682617,
"epoch": 0.05043227665706052,
"grad_norm": 1.1171875,
"learning_rate": 0.000262,
"loss": 6.4418,
"mean_token_accuracy": 0.13869711384177208,
"num_tokens": 1212573.0,
"step": 525
},
{
"entropy": 6.831294298171997,
"epoch": 0.05091258405379443,
"grad_norm": 1.34375,
"learning_rate": 0.00026450000000000003,
"loss": 6.3877,
"mean_token_accuracy": 0.14377668723464013,
"num_tokens": 1223382.0,
"step": 530
},
{
"entropy": 6.8336584091186525,
"epoch": 0.05139289145052834,
"grad_norm": 1.0,
"learning_rate": 0.00026700000000000004,
"loss": 6.4756,
"mean_token_accuracy": 0.1397824764251709,
"num_tokens": 1236501.0,
"step": 535
},
{
"entropy": 6.854681539535522,
"epoch": 0.05187319884726225,
"grad_norm": 1.2421875,
"learning_rate": 0.00026950000000000005,
"loss": 6.4854,
"mean_token_accuracy": 0.1359047219157219,
"num_tokens": 1246798.0,
"step": 540
},
{
"entropy": 6.818022680282593,
"epoch": 0.05235350624399616,
"grad_norm": 1.203125,
"learning_rate": 0.00027200000000000005,
"loss": 6.3993,
"mean_token_accuracy": 0.1383517175912857,
"num_tokens": 1258182.0,
"step": 545
},
{
"entropy": 6.725576877593994,
"epoch": 0.052833813640730067,
"grad_norm": 1.25,
"learning_rate": 0.0002745,
"loss": 6.3529,
"mean_token_accuracy": 0.13154633566737176,
"num_tokens": 1270273.0,
"step": 550
},
{
"entropy": 6.824942541122437,
"epoch": 0.053314121037463975,
"grad_norm": 1.2734375,
"learning_rate": 0.000277,
"loss": 6.4187,
"mean_token_accuracy": 0.13395696729421616,
"num_tokens": 1281136.0,
"step": 555
},
{
"entropy": 6.798213529586792,
"epoch": 0.053794428434197884,
"grad_norm": 1.0859375,
"learning_rate": 0.0002795,
"loss": 6.388,
"mean_token_accuracy": 0.13719379529356956,
"num_tokens": 1294488.0,
"step": 560
},
{
"entropy": 6.690643072128296,
"epoch": 0.05427473583093179,
"grad_norm": 1.2265625,
"learning_rate": 0.00028199999999999997,
"loss": 6.342,
"mean_token_accuracy": 0.13738902136683465,
"num_tokens": 1304113.0,
"step": 565
},
{
"entropy": 6.811877346038818,
"epoch": 0.05475504322766571,
"grad_norm": 1.3203125,
"learning_rate": 0.0002845,
"loss": 6.3715,
"mean_token_accuracy": 0.1435349702835083,
"num_tokens": 1315417.0,
"step": 570
},
{
"entropy": 6.705857038497925,
"epoch": 0.05523535062439962,
"grad_norm": 1.2734375,
"learning_rate": 0.000287,
"loss": 6.3487,
"mean_token_accuracy": 0.13941997289657593,
"num_tokens": 1328084.0,
"step": 575
},
{
"entropy": 6.830380821228028,
"epoch": 0.05571565802113353,
"grad_norm": 1.0703125,
"learning_rate": 0.0002895,
"loss": 6.4948,
"mean_token_accuracy": 0.13673060759902,
"num_tokens": 1338696.0,
"step": 580
},
{
"entropy": 6.816842555999756,
"epoch": 0.056195965417867436,
"grad_norm": 1.046875,
"learning_rate": 0.000292,
"loss": 6.4362,
"mean_token_accuracy": 0.13864873573184014,
"num_tokens": 1350240.0,
"step": 585
},
{
"entropy": 6.8044596195220945,
"epoch": 0.056676272814601344,
"grad_norm": 1.203125,
"learning_rate": 0.0002945,
"loss": 6.3876,
"mean_token_accuracy": 0.13224058821797371,
"num_tokens": 1361720.0,
"step": 590
},
{
"entropy": 6.702146100997925,
"epoch": 0.05715658021133525,
"grad_norm": 1.375,
"learning_rate": 0.000297,
"loss": 6.388,
"mean_token_accuracy": 0.14646777287125587,
"num_tokens": 1373286.0,
"step": 595
},
{
"entropy": 6.792327165603638,
"epoch": 0.05763688760806916,
"grad_norm": 1.28125,
"learning_rate": 0.0002995,
"loss": 6.3357,
"mean_token_accuracy": 0.14013056233525276,
"num_tokens": 1384274.0,
"step": 600
},
{
"entropy": 6.695337629318237,
"epoch": 0.05811719500480307,
"grad_norm": 1.15625,
"learning_rate": 0.000302,
"loss": 6.3866,
"mean_token_accuracy": 0.13806435391306876,
"num_tokens": 1395355.0,
"step": 605
},
{
"entropy": 6.67833251953125,
"epoch": 0.05859750240153699,
"grad_norm": 1.2890625,
"learning_rate": 0.0003045,
"loss": 6.3185,
"mean_token_accuracy": 0.14838905334472657,
"num_tokens": 1406664.0,
"step": 610
},
{
"entropy": 6.84452452659607,
"epoch": 0.059077809798270896,
"grad_norm": 1.1875,
"learning_rate": 0.000307,
"loss": 6.4734,
"mean_token_accuracy": 0.1365173079073429,
"num_tokens": 1418450.0,
"step": 615
},
{
"entropy": 6.745831251144409,
"epoch": 0.059558117195004805,
"grad_norm": 1.2578125,
"learning_rate": 0.0003095,
"loss": 6.365,
"mean_token_accuracy": 0.1397569701075554,
"num_tokens": 1430048.0,
"step": 620
},
{
"entropy": 6.811714601516724,
"epoch": 0.060038424591738714,
"grad_norm": 1.1640625,
"learning_rate": 0.000312,
"loss": 6.4099,
"mean_token_accuracy": 0.13645492121577263,
"num_tokens": 1441820.0,
"step": 625
},
{
"entropy": 6.656697416305542,
"epoch": 0.06051873198847262,
"grad_norm": 1.2109375,
"learning_rate": 0.0003145,
"loss": 6.2619,
"mean_token_accuracy": 0.14607831984758377,
"num_tokens": 1453209.0,
"step": 630
},
{
"entropy": 6.677804946899414,
"epoch": 0.06099903938520653,
"grad_norm": 1.171875,
"learning_rate": 0.000317,
"loss": 6.2859,
"mean_token_accuracy": 0.139230377972126,
"num_tokens": 1465423.0,
"step": 635
},
{
"entropy": 6.6918044090271,
"epoch": 0.06147934678194044,
"grad_norm": 1.2734375,
"learning_rate": 0.0003195,
"loss": 6.2776,
"mean_token_accuracy": 0.14577876925468444,
"num_tokens": 1476575.0,
"step": 640
},
{
"entropy": 6.6395305633544925,
"epoch": 0.06195965417867435,
"grad_norm": 1.296875,
"learning_rate": 0.000322,
"loss": 6.2269,
"mean_token_accuracy": 0.14229501113295556,
"num_tokens": 1486932.0,
"step": 645
},
{
"entropy": 6.6645488262176515,
"epoch": 0.06243996157540826,
"grad_norm": 1.125,
"learning_rate": 0.00032450000000000003,
"loss": 6.2634,
"mean_token_accuracy": 0.1412537418305874,
"num_tokens": 1498494.0,
"step": 650
},
{
"entropy": 6.728798723220825,
"epoch": 0.06292026897214217,
"grad_norm": 1.1171875,
"learning_rate": 0.00032700000000000003,
"loss": 6.3314,
"mean_token_accuracy": 0.1367237947881222,
"num_tokens": 1509937.0,
"step": 655
},
{
"entropy": 6.730763149261475,
"epoch": 0.06340057636887608,
"grad_norm": 1.390625,
"learning_rate": 0.00032950000000000004,
"loss": 6.3493,
"mean_token_accuracy": 0.14772669598460197,
"num_tokens": 1519823.0,
"step": 660
},
{
"entropy": 6.728877162933349,
"epoch": 0.06388088376560999,
"grad_norm": 1.2421875,
"learning_rate": 0.00033200000000000005,
"loss": 6.377,
"mean_token_accuracy": 0.1385086067020893,
"num_tokens": 1529943.0,
"step": 665
},
{
"entropy": 6.646539354324341,
"epoch": 0.0643611911623439,
"grad_norm": 1.21875,
"learning_rate": 0.00033450000000000005,
"loss": 6.218,
"mean_token_accuracy": 0.1462228536605835,
"num_tokens": 1540618.0,
"step": 670
},
{
"entropy": 6.621165466308594,
"epoch": 0.06484149855907781,
"grad_norm": 1.2109375,
"learning_rate": 0.000337,
"loss": 6.3349,
"mean_token_accuracy": 0.13347838148474694,
"num_tokens": 1553208.0,
"step": 675
},
{
"entropy": 6.75033278465271,
"epoch": 0.06532180595581172,
"grad_norm": 1.2578125,
"learning_rate": 0.0003395,
"loss": 6.3079,
"mean_token_accuracy": 0.14778160005807878,
"num_tokens": 1563975.0,
"step": 680
},
{
"entropy": 6.730441284179688,
"epoch": 0.06580211335254563,
"grad_norm": 1.25,
"learning_rate": 0.000342,
"loss": 6.3601,
"mean_token_accuracy": 0.13537456542253495,
"num_tokens": 1575998.0,
"step": 685
},
{
"entropy": 6.659586429595947,
"epoch": 0.06628242074927954,
"grad_norm": 1.25,
"learning_rate": 0.00034449999999999997,
"loss": 6.263,
"mean_token_accuracy": 0.14849890321493148,
"num_tokens": 1586041.0,
"step": 690
},
{
"entropy": 6.605801343917847,
"epoch": 0.06676272814601344,
"grad_norm": 1.34375,
"learning_rate": 0.000347,
"loss": 6.324,
"mean_token_accuracy": 0.14218652099370957,
"num_tokens": 1597531.0,
"step": 695
},
{
"entropy": 6.669734811782837,
"epoch": 0.06724303554274735,
"grad_norm": 1.28125,
"learning_rate": 0.0003495,
"loss": 6.3116,
"mean_token_accuracy": 0.13976099193096161,
"num_tokens": 1609255.0,
"step": 700
},
{
"entropy": 6.609857988357544,
"epoch": 0.06772334293948126,
"grad_norm": 1.1015625,
"learning_rate": 0.000352,
"loss": 6.2983,
"mean_token_accuracy": 0.14360181987285614,
"num_tokens": 1621098.0,
"step": 705
},
{
"entropy": 6.6083897113800045,
"epoch": 0.06820365033621517,
"grad_norm": 1.09375,
"learning_rate": 0.0003545,
"loss": 6.189,
"mean_token_accuracy": 0.15515244603157044,
"num_tokens": 1631941.0,
"step": 710
},
{
"entropy": 6.57919979095459,
"epoch": 0.0686839577329491,
"grad_norm": 1.1953125,
"learning_rate": 0.000357,
"loss": 6.2074,
"mean_token_accuracy": 0.14852219969034194,
"num_tokens": 1643117.0,
"step": 715
},
{
"entropy": 6.576744127273559,
"epoch": 0.069164265129683,
"grad_norm": 1.15625,
"learning_rate": 0.0003595,
"loss": 6.2372,
"mean_token_accuracy": 0.1395713485777378,
"num_tokens": 1653595.0,
"step": 720
},
{
"entropy": 6.663236331939697,
"epoch": 0.06964457252641691,
"grad_norm": 1.2421875,
"learning_rate": 0.000362,
"loss": 6.2792,
"mean_token_accuracy": 0.14607299417257308,
"num_tokens": 1664495.0,
"step": 725
},
{
"entropy": 6.556261014938355,
"epoch": 0.07012487992315082,
"grad_norm": 1.1015625,
"learning_rate": 0.0003645,
"loss": 6.2639,
"mean_token_accuracy": 0.1435217887163162,
"num_tokens": 1674923.0,
"step": 730
},
{
"entropy": 6.618684720993042,
"epoch": 0.07060518731988473,
"grad_norm": 1.2890625,
"learning_rate": 0.000367,
"loss": 6.1667,
"mean_token_accuracy": 0.1526644468307495,
"num_tokens": 1685904.0,
"step": 735
},
{
"entropy": 6.627186965942383,
"epoch": 0.07108549471661864,
"grad_norm": 0.984375,
"learning_rate": 0.0003695,
"loss": 6.3877,
"mean_token_accuracy": 0.13355778381228448,
"num_tokens": 1699133.0,
"step": 740
},
{
"entropy": 6.548422527313233,
"epoch": 0.07156580211335255,
"grad_norm": 1.0625,
"learning_rate": 0.000372,
"loss": 6.2018,
"mean_token_accuracy": 0.14151064530014992,
"num_tokens": 1711559.0,
"step": 745
},
{
"entropy": 6.550561428070068,
"epoch": 0.07204610951008646,
"grad_norm": 1.171875,
"learning_rate": 0.0003745,
"loss": 6.1238,
"mean_token_accuracy": 0.1509994685649872,
"num_tokens": 1722526.0,
"step": 750
},
{
"entropy": 6.532330465316773,
"epoch": 0.07252641690682037,
"grad_norm": 1.203125,
"learning_rate": 0.000377,
"loss": 6.1889,
"mean_token_accuracy": 0.14655965566635132,
"num_tokens": 1734261.0,
"step": 755
},
{
"entropy": 6.53015832901001,
"epoch": 0.07300672430355427,
"grad_norm": 1.2421875,
"learning_rate": 0.0003795,
"loss": 6.1803,
"mean_token_accuracy": 0.15340202897787095,
"num_tokens": 1745151.0,
"step": 760
},
{
"entropy": 6.69319715499878,
"epoch": 0.07348703170028818,
"grad_norm": 1.3046875,
"learning_rate": 0.000382,
"loss": 6.3355,
"mean_token_accuracy": 0.1429338976740837,
"num_tokens": 1755463.0,
"step": 765
},
{
"entropy": 6.623394536972046,
"epoch": 0.07396733909702209,
"grad_norm": 1.2265625,
"learning_rate": 0.0003845,
"loss": 6.2828,
"mean_token_accuracy": 0.13968928754329682,
"num_tokens": 1767717.0,
"step": 770
},
{
"entropy": 6.437202119827271,
"epoch": 0.074447646493756,
"grad_norm": 1.3203125,
"learning_rate": 0.00038700000000000003,
"loss": 6.1355,
"mean_token_accuracy": 0.15134122371673583,
"num_tokens": 1779115.0,
"step": 775
},
{
"entropy": 6.419523668289185,
"epoch": 0.07492795389048991,
"grad_norm": 1.1953125,
"learning_rate": 0.00038950000000000003,
"loss": 6.087,
"mean_token_accuracy": 0.15281283408403395,
"num_tokens": 1789644.0,
"step": 780
},
{
"entropy": 6.6132158756256105,
"epoch": 0.07540826128722382,
"grad_norm": 1.2578125,
"learning_rate": 0.00039200000000000004,
"loss": 6.2997,
"mean_token_accuracy": 0.14345607832074164,
"num_tokens": 1800606.0,
"step": 785
},
{
"entropy": 6.524192380905151,
"epoch": 0.07588856868395773,
"grad_norm": 1.125,
"learning_rate": 0.00039450000000000005,
"loss": 6.301,
"mean_token_accuracy": 0.13897362276911734,
"num_tokens": 1812168.0,
"step": 790
},
{
"entropy": 6.73820424079895,
"epoch": 0.07636887608069164,
"grad_norm": 1.0546875,
"learning_rate": 0.00039700000000000005,
"loss": 6.2386,
"mean_token_accuracy": 0.14347319155931473,
"num_tokens": 1823830.0,
"step": 795
},
{
"entropy": 6.445359086990356,
"epoch": 0.07684918347742556,
"grad_norm": 1.1328125,
"learning_rate": 0.0003995,
"loss": 6.2403,
"mean_token_accuracy": 0.1467433527112007,
"num_tokens": 1835611.0,
"step": 800
},
{
"entropy": 6.597848510742187,
"epoch": 0.07732949087415947,
"grad_norm": 1.09375,
"learning_rate": 0.000402,
"loss": 6.3054,
"mean_token_accuracy": 0.13762886747717856,
"num_tokens": 1847036.0,
"step": 805
},
{
"entropy": 6.562679290771484,
"epoch": 0.07780979827089338,
"grad_norm": 1.3046875,
"learning_rate": 0.0004045,
"loss": 6.2378,
"mean_token_accuracy": 0.14037195593118668,
"num_tokens": 1857476.0,
"step": 810
},
{
"entropy": 6.5669573783874515,
"epoch": 0.07829010566762729,
"grad_norm": 1.1328125,
"learning_rate": 0.00040699999999999997,
"loss": 6.2621,
"mean_token_accuracy": 0.14760917276144028,
"num_tokens": 1869073.0,
"step": 815
},
{
"entropy": 6.387801313400269,
"epoch": 0.0787704130643612,
"grad_norm": 0.98828125,
"learning_rate": 0.0004095,
"loss": 6.1412,
"mean_token_accuracy": 0.1509813189506531,
"num_tokens": 1880439.0,
"step": 820
},
{
"entropy": 6.699515724182129,
"epoch": 0.0792507204610951,
"grad_norm": 1.1953125,
"learning_rate": 0.000412,
"loss": 6.2976,
"mean_token_accuracy": 0.14118361324071885,
"num_tokens": 1891600.0,
"step": 825
},
{
"entropy": 6.5128673076629635,
"epoch": 0.07973102785782901,
"grad_norm": 1.2578125,
"learning_rate": 0.0004145,
"loss": 6.2569,
"mean_token_accuracy": 0.14543531611561775,
"num_tokens": 1903126.0,
"step": 830
},
{
"entropy": 6.672891998291016,
"epoch": 0.08021133525456292,
"grad_norm": 1.109375,
"learning_rate": 0.000417,
"loss": 6.226,
"mean_token_accuracy": 0.14782591313123702,
"num_tokens": 1913913.0,
"step": 835
},
{
"entropy": 6.366758394241333,
"epoch": 0.08069164265129683,
"grad_norm": 1.0859375,
"learning_rate": 0.0004195,
"loss": 6.1501,
"mean_token_accuracy": 0.1501932203769684,
"num_tokens": 1924961.0,
"step": 840
},
{
"entropy": 6.692445278167725,
"epoch": 0.08117195004803074,
"grad_norm": 1.265625,
"learning_rate": 0.000422,
"loss": 6.2407,
"mean_token_accuracy": 0.1381278119981289,
"num_tokens": 1936773.0,
"step": 845
},
{
"entropy": 6.406199932098389,
"epoch": 0.08165225744476465,
"grad_norm": 1.1640625,
"learning_rate": 0.0004245,
"loss": 6.1643,
"mean_token_accuracy": 0.1459251821041107,
"num_tokens": 1948190.0,
"step": 850
},
{
"entropy": 6.452398300170898,
"epoch": 0.08213256484149856,
"grad_norm": 1.171875,
"learning_rate": 0.000427,
"loss": 6.1575,
"mean_token_accuracy": 0.14858585894107817,
"num_tokens": 1960038.0,
"step": 855
},
{
"entropy": 6.57236967086792,
"epoch": 0.08261287223823247,
"grad_norm": 1.234375,
"learning_rate": 0.0004295,
"loss": 6.2384,
"mean_token_accuracy": 0.14626160115003586,
"num_tokens": 1970535.0,
"step": 860
},
{
"entropy": 6.537123394012451,
"epoch": 0.08309317963496637,
"grad_norm": 1.171875,
"learning_rate": 0.000432,
"loss": 6.2265,
"mean_token_accuracy": 0.142296252399683,
"num_tokens": 1981386.0,
"step": 865
},
{
"entropy": 6.4313184261322025,
"epoch": 0.08357348703170028,
"grad_norm": 1.046875,
"learning_rate": 0.0004345,
"loss": 6.1375,
"mean_token_accuracy": 0.14900052845478057,
"num_tokens": 1993196.0,
"step": 870
},
{
"entropy": 6.4812110424041744,
"epoch": 0.08405379442843419,
"grad_norm": 1.046875,
"learning_rate": 0.000437,
"loss": 6.1562,
"mean_token_accuracy": 0.15512351393699647,
"num_tokens": 2004756.0,
"step": 875
},
{
"entropy": 6.437265253067016,
"epoch": 0.0845341018251681,
"grad_norm": 1.2265625,
"learning_rate": 0.0004395,
"loss": 6.1392,
"mean_token_accuracy": 0.14693072885274888,
"num_tokens": 2016020.0,
"step": 880
},
{
"entropy": 6.458638095855713,
"epoch": 0.08501440922190202,
"grad_norm": 1.125,
"learning_rate": 0.000442,
"loss": 6.1858,
"mean_token_accuracy": 0.14666839689016342,
"num_tokens": 2027747.0,
"step": 885
},
{
"entropy": 6.551916837692261,
"epoch": 0.08549471661863593,
"grad_norm": 1.3828125,
"learning_rate": 0.0004445,
"loss": 6.1604,
"mean_token_accuracy": 0.14383722916245462,
"num_tokens": 2038841.0,
"step": 890
},
{
"entropy": 6.44727840423584,
"epoch": 0.08597502401536984,
"grad_norm": 1.1953125,
"learning_rate": 0.000447,
"loss": 6.2109,
"mean_token_accuracy": 0.14256930723786354,
"num_tokens": 2049905.0,
"step": 895
},
{
"entropy": 6.512820863723755,
"epoch": 0.08645533141210375,
"grad_norm": 1.234375,
"learning_rate": 0.00044950000000000003,
"loss": 6.2817,
"mean_token_accuracy": 0.14085745438933372,
"num_tokens": 2062492.0,
"step": 900
},
{
"entropy": 6.474342775344849,
"epoch": 0.08693563880883766,
"grad_norm": 1.2265625,
"learning_rate": 0.00045200000000000004,
"loss": 6.216,
"mean_token_accuracy": 0.14198037534952163,
"num_tokens": 2073840.0,
"step": 905
},
{
"entropy": 6.469565916061401,
"epoch": 0.08741594620557157,
"grad_norm": 1.296875,
"learning_rate": 0.00045450000000000004,
"loss": 6.1233,
"mean_token_accuracy": 0.14328503012657165,
"num_tokens": 2085720.0,
"step": 910
},
{
"entropy": 6.508474731445313,
"epoch": 0.08789625360230548,
"grad_norm": 1.1015625,
"learning_rate": 0.00045700000000000005,
"loss": 6.2079,
"mean_token_accuracy": 0.14646697416901588,
"num_tokens": 2096649.0,
"step": 915
},
{
"entropy": 6.526723051071167,
"epoch": 0.08837656099903939,
"grad_norm": 1.015625,
"learning_rate": 0.00045950000000000006,
"loss": 6.2087,
"mean_token_accuracy": 0.14670456051826478,
"num_tokens": 2109030.0,
"step": 920
},
{
"entropy": 6.422917222976684,
"epoch": 0.0888568683957733,
"grad_norm": 1.1015625,
"learning_rate": 0.000462,
"loss": 6.1155,
"mean_token_accuracy": 0.15043116360902786,
"num_tokens": 2121384.0,
"step": 925
},
{
"entropy": 6.429614877700805,
"epoch": 0.0893371757925072,
"grad_norm": 1.1875,
"learning_rate": 0.0004645,
"loss": 6.0734,
"mean_token_accuracy": 0.15336936861276626,
"num_tokens": 2131377.0,
"step": 930
},
{
"entropy": 6.468717241287232,
"epoch": 0.08981748318924111,
"grad_norm": 1.140625,
"learning_rate": 0.000467,
"loss": 6.2611,
"mean_token_accuracy": 0.14346477389335632,
"num_tokens": 2142364.0,
"step": 935
},
{
"entropy": 6.449568080902099,
"epoch": 0.09029779058597502,
"grad_norm": 1.171875,
"learning_rate": 0.0004695,
"loss": 6.1302,
"mean_token_accuracy": 0.14927371144294738,
"num_tokens": 2153040.0,
"step": 940
},
{
"entropy": 6.434846210479736,
"epoch": 0.09077809798270893,
"grad_norm": 1.0859375,
"learning_rate": 0.000472,
"loss": 6.1895,
"mean_token_accuracy": 0.15058233961462975,
"num_tokens": 2165571.0,
"step": 945
},
{
"entropy": 6.456413507461548,
"epoch": 0.09125840537944284,
"grad_norm": 1.15625,
"learning_rate": 0.0004745,
"loss": 6.1612,
"mean_token_accuracy": 0.14620548188686372,
"num_tokens": 2177241.0,
"step": 950
},
{
"entropy": 6.455846166610717,
"epoch": 0.09173871277617675,
"grad_norm": 1.1484375,
"learning_rate": 0.000477,
"loss": 6.1925,
"mean_token_accuracy": 0.13922726288437842,
"num_tokens": 2187475.0,
"step": 955
},
{
"entropy": 6.435314750671386,
"epoch": 0.09221902017291066,
"grad_norm": 1.078125,
"learning_rate": 0.0004795,
"loss": 6.2183,
"mean_token_accuracy": 0.1472764365375042,
"num_tokens": 2198185.0,
"step": 960
},
{
"entropy": 6.411659860610962,
"epoch": 0.09269932756964457,
"grad_norm": 1.125,
"learning_rate": 0.000482,
"loss": 6.1554,
"mean_token_accuracy": 0.14865822941064835,
"num_tokens": 2210404.0,
"step": 965
},
{
"entropy": 6.521021556854248,
"epoch": 0.09317963496637849,
"grad_norm": 1.0859375,
"learning_rate": 0.0004845,
"loss": 6.1382,
"mean_token_accuracy": 0.14820332527160646,
"num_tokens": 2222188.0,
"step": 970
},
{
"entropy": 6.401497077941895,
"epoch": 0.0936599423631124,
"grad_norm": 1.09375,
"learning_rate": 0.000487,
"loss": 6.1825,
"mean_token_accuracy": 0.1503289483487606,
"num_tokens": 2233418.0,
"step": 975
},
{
"entropy": 6.386258792877197,
"epoch": 0.0941402497598463,
"grad_norm": 1.203125,
"learning_rate": 0.0004895,
"loss": 6.1549,
"mean_token_accuracy": 0.15336488783359528,
"num_tokens": 2245053.0,
"step": 980
},
{
"entropy": 6.497596883773804,
"epoch": 0.09462055715658022,
"grad_norm": 1.140625,
"learning_rate": 0.000492,
"loss": 6.1799,
"mean_token_accuracy": 0.1518462046980858,
"num_tokens": 2256375.0,
"step": 985
},
{
"entropy": 6.297069692611695,
"epoch": 0.09510086455331412,
"grad_norm": 1.140625,
"learning_rate": 0.0004945,
"loss": 6.0222,
"mean_token_accuracy": 0.1536286234855652,
"num_tokens": 2267074.0,
"step": 990
},
{
"entropy": 6.3477600574493405,
"epoch": 0.09558117195004803,
"grad_norm": 1.1484375,
"learning_rate": 0.000497,
"loss": 6.0739,
"mean_token_accuracy": 0.15113484859466553,
"num_tokens": 2277168.0,
"step": 995
},
{
"entropy": 6.337840461730957,
"epoch": 0.09606147934678194,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995,
"loss": 6.0562,
"mean_token_accuracy": 0.15290841460227966,
"num_tokens": 2288178.0,
"step": 1000
},
{
"entropy": 6.316500282287597,
"epoch": 0.09654178674351585,
"grad_norm": 1.125,
"learning_rate": 0.0004999999983283737,
"loss": 6.0619,
"mean_token_accuracy": 0.1515717551112175,
"num_tokens": 2299765.0,
"step": 1005
},
{
"entropy": 6.519514703750611,
"epoch": 0.09702209414024976,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999999915373924,
"loss": 6.1835,
"mean_token_accuracy": 0.13907498270273208,
"num_tokens": 2312047.0,
"step": 1010
},
{
"entropy": 6.364095211029053,
"epoch": 0.09750240153698367,
"grad_norm": 1.2265625,
"learning_rate": 0.0004999999795225793,
"loss": 6.1611,
"mean_token_accuracy": 0.14061366766691208,
"num_tokens": 2324118.0,
"step": 1015
},
{
"entropy": 6.4445971012115475,
"epoch": 0.09798270893371758,
"grad_norm": 1.09375,
"learning_rate": 0.0004999999622839347,
"loss": 6.1611,
"mean_token_accuracy": 0.14967067837715148,
"num_tokens": 2335171.0,
"step": 1020
},
{
"entropy": 6.398542642593384,
"epoch": 0.09846301633045149,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999999398214593,
"loss": 6.0743,
"mean_token_accuracy": 0.1492784783244133,
"num_tokens": 2346338.0,
"step": 1025
},
{
"entropy": 6.369733762741089,
"epoch": 0.0989433237271854,
"grad_norm": 1.078125,
"learning_rate": 0.0004999999121351532,
"loss": 6.1011,
"mean_token_accuracy": 0.15191911309957504,
"num_tokens": 2357185.0,
"step": 1030
},
{
"entropy": 6.378099775314331,
"epoch": 0.0994236311239193,
"grad_norm": 1.109375,
"learning_rate": 0.0004999998792250173,
"loss": 6.0369,
"mean_token_accuracy": 0.15464141294360162,
"num_tokens": 2368494.0,
"step": 1035
},
{
"entropy": 6.379990863800049,
"epoch": 0.09990393852065321,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999998410910524,
"loss": 6.2349,
"mean_token_accuracy": 0.14129094332456588,
"num_tokens": 2380800.0,
"step": 1040
},
{
"entropy": 6.467003870010376,
"epoch": 0.10038424591738712,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999997977332592,
"loss": 6.1819,
"mean_token_accuracy": 0.1439610406756401,
"num_tokens": 2391753.0,
"step": 1045
},
{
"entropy": 6.27177791595459,
"epoch": 0.10086455331412104,
"grad_norm": 1.2890625,
"learning_rate": 0.0004999997491516389,
"loss": 6.0646,
"mean_token_accuracy": 0.14684219062328338,
"num_tokens": 2403324.0,
"step": 1050
},
{
"entropy": 6.4170433521270756,
"epoch": 0.10134486071085495,
"grad_norm": 1.109375,
"learning_rate": 0.0004999996953461925,
"loss": 6.1568,
"mean_token_accuracy": 0.1471158519387245,
"num_tokens": 2414873.0,
"step": 1055
},
{
"entropy": 6.226349878311157,
"epoch": 0.10182516810758886,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999996363169212,
"loss": 5.9522,
"mean_token_accuracy": 0.15969930738210678,
"num_tokens": 2425308.0,
"step": 1060
},
{
"entropy": 6.280011129379273,
"epoch": 0.10230547550432277,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999995720638266,
"loss": 5.9956,
"mean_token_accuracy": 0.15745754390954972,
"num_tokens": 2436835.0,
"step": 1065
},
{
"entropy": 6.318831586837769,
"epoch": 0.10278578290105668,
"grad_norm": 1.21875,
"learning_rate": 0.00049999950258691,
"loss": 6.1016,
"mean_token_accuracy": 0.15324254631996154,
"num_tokens": 2446798.0,
"step": 1070
},
{
"entropy": 6.259381628036499,
"epoch": 0.10326609029779059,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999994278861731,
"loss": 6.0097,
"mean_token_accuracy": 0.15251461267471314,
"num_tokens": 2457308.0,
"step": 1075
},
{
"entropy": 6.345122861862182,
"epoch": 0.1037463976945245,
"grad_norm": 0.984375,
"learning_rate": 0.0004999993479616175,
"loss": 6.0775,
"mean_token_accuracy": 0.14187726974487305,
"num_tokens": 2468917.0,
"step": 1080
},
{
"entropy": 6.341993236541748,
"epoch": 0.1042267050912584,
"grad_norm": 1.171875,
"learning_rate": 0.0004999992628132451,
"loss": 6.0947,
"mean_token_accuracy": 0.145965027064085,
"num_tokens": 2481363.0,
"step": 1085
},
{
"entropy": 6.322693538665772,
"epoch": 0.10470701248799232,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999991724410582,
"loss": 6.0875,
"mean_token_accuracy": 0.15120186507701874,
"num_tokens": 2493082.0,
"step": 1090
},
{
"entropy": 6.3057647228240965,
"epoch": 0.10518731988472622,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999990768450583,
"loss": 6.0435,
"mean_token_accuracy": 0.15195624977350236,
"num_tokens": 2503849.0,
"step": 1095
},
{
"entropy": 6.302676057815551,
"epoch": 0.10566762728146013,
"grad_norm": 1.15625,
"learning_rate": 0.0004999989760252482,
"loss": 6.0935,
"mean_token_accuracy": 0.15190254151821136,
"num_tokens": 2514528.0,
"step": 1100
},
{
"entropy": 6.300235176086426,
"epoch": 0.10614793467819404,
"grad_norm": 1.296875,
"learning_rate": 0.0004999988699816299,
"loss": 6.0734,
"mean_token_accuracy": 0.15657310485839843,
"num_tokens": 2524971.0,
"step": 1105
},
{
"entropy": 6.231389999389648,
"epoch": 0.10662824207492795,
"grad_norm": 1.03125,
"learning_rate": 0.0004999987587142058,
"loss": 5.9787,
"mean_token_accuracy": 0.1494237408041954,
"num_tokens": 2535674.0,
"step": 1110
},
{
"entropy": 6.4110383033752445,
"epoch": 0.10710854947166186,
"grad_norm": 1.15625,
"learning_rate": 0.0004999986422229789,
"loss": 6.2173,
"mean_token_accuracy": 0.14188248962163924,
"num_tokens": 2547108.0,
"step": 1115
},
{
"entropy": 6.270652484893799,
"epoch": 0.10758885686839577,
"grad_norm": 1.015625,
"learning_rate": 0.0004999985205079514,
"loss": 6.0312,
"mean_token_accuracy": 0.14654332846403123,
"num_tokens": 2559474.0,
"step": 1120
},
{
"entropy": 6.178351354598999,
"epoch": 0.10806916426512968,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999983935691265,
"loss": 5.877,
"mean_token_accuracy": 0.16631512492895126,
"num_tokens": 2571264.0,
"step": 1125
},
{
"entropy": 6.240466594696045,
"epoch": 0.10854947166186359,
"grad_norm": 1.0703125,
"learning_rate": 0.000499998261406507,
"loss": 6.0402,
"mean_token_accuracy": 0.1550907276570797,
"num_tokens": 2583731.0,
"step": 1130
},
{
"entropy": 6.455864334106446,
"epoch": 0.10902977905859751,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999981240200958,
"loss": 6.0865,
"mean_token_accuracy": 0.15334258675575257,
"num_tokens": 2595497.0,
"step": 1135
},
{
"entropy": 6.156699085235596,
"epoch": 0.10951008645533142,
"grad_norm": 1.015625,
"learning_rate": 0.0004999979814098966,
"loss": 6.0553,
"mean_token_accuracy": 0.1544922299683094,
"num_tokens": 2607358.0,
"step": 1140
},
{
"entropy": 6.313141107559204,
"epoch": 0.10999039385206533,
"grad_norm": 1.09375,
"learning_rate": 0.0004999978335759121,
"loss": 5.9824,
"mean_token_accuracy": 0.16003676876425743,
"num_tokens": 2618936.0,
"step": 1145
},
{
"entropy": 6.253233003616333,
"epoch": 0.11047070124879924,
"grad_norm": 1.140625,
"learning_rate": 0.0004999976805181461,
"loss": 6.1273,
"mean_token_accuracy": 0.1448728360235691,
"num_tokens": 2631840.0,
"step": 1150
},
{
"entropy": 6.283588409423828,
"epoch": 0.11095100864553314,
"grad_norm": 1.15625,
"learning_rate": 0.000499997522236602,
"loss": 6.0696,
"mean_token_accuracy": 0.15917213410139083,
"num_tokens": 2642412.0,
"step": 1155
},
{
"entropy": 6.2728049755096436,
"epoch": 0.11143131604226705,
"grad_norm": 1.03125,
"learning_rate": 0.0004999973587312837,
"loss": 6.0413,
"mean_token_accuracy": 0.15381678342819213,
"num_tokens": 2653890.0,
"step": 1160
},
{
"entropy": 6.398951053619385,
"epoch": 0.11191162343900096,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999971900021947,
"loss": 6.0976,
"mean_token_accuracy": 0.1579543687403202,
"num_tokens": 2664888.0,
"step": 1165
},
{
"entropy": 6.157506227493286,
"epoch": 0.11239193083573487,
"grad_norm": 0.984375,
"learning_rate": 0.0004999970160493391,
"loss": 6.0067,
"mean_token_accuracy": 0.1552426412701607,
"num_tokens": 2675550.0,
"step": 1170
},
{
"entropy": 6.209622144699097,
"epoch": 0.11287223823246878,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999968368727209,
"loss": 6.01,
"mean_token_accuracy": 0.15791884660720826,
"num_tokens": 2688022.0,
"step": 1175
},
{
"entropy": 6.314128446578979,
"epoch": 0.11335254562920269,
"grad_norm": 1.0625,
"learning_rate": 0.0004999966524723442,
"loss": 6.0035,
"mean_token_accuracy": 0.15661349296569824,
"num_tokens": 2698737.0,
"step": 1180
},
{
"entropy": 6.17528281211853,
"epoch": 0.1138328530259366,
"grad_norm": 1.109375,
"learning_rate": 0.0004999964628482135,
"loss": 5.9803,
"mean_token_accuracy": 0.1625732421875,
"num_tokens": 2709844.0,
"step": 1185
},
{
"entropy": 6.311813545227051,
"epoch": 0.1143131604226705,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999962680003328,
"loss": 6.0243,
"mean_token_accuracy": 0.15992799401283264,
"num_tokens": 2720273.0,
"step": 1190
},
{
"entropy": 6.231012916564941,
"epoch": 0.11479346781940442,
"grad_norm": 1.125,
"learning_rate": 0.000499996067928707,
"loss": 6.0555,
"mean_token_accuracy": 0.1537340298295021,
"num_tokens": 2731354.0,
"step": 1195
},
{
"entropy": 6.250294923782349,
"epoch": 0.11527377521613832,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999958626333406,
"loss": 6.0383,
"mean_token_accuracy": 0.15496493577957154,
"num_tokens": 2742966.0,
"step": 1200
},
{
"entropy": 6.222562694549561,
"epoch": 0.11575408261287223,
"grad_norm": 1.1875,
"learning_rate": 0.0004999956521142383,
"loss": 5.9677,
"mean_token_accuracy": 0.1614091530442238,
"num_tokens": 2755010.0,
"step": 1205
},
{
"entropy": 6.238761901855469,
"epoch": 0.11623439000960614,
"grad_norm": 1.109375,
"learning_rate": 0.0004999954363714051,
"loss": 5.9963,
"mean_token_accuracy": 0.1529179558157921,
"num_tokens": 2766176.0,
"step": 1210
},
{
"entropy": 6.375168800354004,
"epoch": 0.11671469740634005,
"grad_norm": 1.09375,
"learning_rate": 0.0004999952154048459,
"loss": 6.1416,
"mean_token_accuracy": 0.1510852135717869,
"num_tokens": 2777861.0,
"step": 1215
},
{
"entropy": 6.179434299468994,
"epoch": 0.11719500480307397,
"grad_norm": 1.0234375,
"learning_rate": 0.000499994989214566,
"loss": 5.9563,
"mean_token_accuracy": 0.15724676698446274,
"num_tokens": 2788725.0,
"step": 1220
},
{
"entropy": 6.156388998031616,
"epoch": 0.11767531219980788,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999947578005705,
"loss": 5.9829,
"mean_token_accuracy": 0.1558725953102112,
"num_tokens": 2801613.0,
"step": 1225
},
{
"entropy": 6.306823587417602,
"epoch": 0.11815561959654179,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999945211628648,
"loss": 6.0591,
"mean_token_accuracy": 0.15092146396636963,
"num_tokens": 2812474.0,
"step": 1230
},
{
"entropy": 6.149747133255005,
"epoch": 0.1186359269932757,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999942793014544,
"loss": 5.9613,
"mean_token_accuracy": 0.15863570123910903,
"num_tokens": 2823178.0,
"step": 1235
},
{
"entropy": 6.148617744445801,
"epoch": 0.11911623439000961,
"grad_norm": 0.9140625,
"learning_rate": 0.000499994032216345,
"loss": 5.9582,
"mean_token_accuracy": 0.15689361244440078,
"num_tokens": 2836486.0,
"step": 1240
},
{
"entropy": 6.2577982425689695,
"epoch": 0.11959654178674352,
"grad_norm": 1.0625,
"learning_rate": 0.0004999937799075422,
"loss": 6.0325,
"mean_token_accuracy": 0.1582626670598984,
"num_tokens": 2847902.0,
"step": 1245
},
{
"entropy": 6.061917066574097,
"epoch": 0.12007684918347743,
"grad_norm": 0.9921875,
"learning_rate": 0.000499993522375052,
"loss": 5.9155,
"mean_token_accuracy": 0.15855048298835756,
"num_tokens": 2859991.0,
"step": 1250
},
{
"entropy": 6.334481048583984,
"epoch": 0.12055715658021134,
"grad_norm": 1.0625,
"learning_rate": 0.0004999932596188802,
"loss": 6.103,
"mean_token_accuracy": 0.14992666617035866,
"num_tokens": 2870269.0,
"step": 1255
},
{
"entropy": 6.16114387512207,
"epoch": 0.12103746397694524,
"grad_norm": 1.140625,
"learning_rate": 0.0004999929916390331,
"loss": 5.974,
"mean_token_accuracy": 0.15265119820833206,
"num_tokens": 2882191.0,
"step": 1260
},
{
"entropy": 6.185744333267212,
"epoch": 0.12151777137367915,
"grad_norm": 1.03125,
"learning_rate": 0.0004999927184355169,
"loss": 5.9814,
"mean_token_accuracy": 0.15251912623643876,
"num_tokens": 2892775.0,
"step": 1265
},
{
"entropy": 6.141828775405884,
"epoch": 0.12199807877041306,
"grad_norm": 1.046875,
"learning_rate": 0.0004999924400083377,
"loss": 5.9617,
"mean_token_accuracy": 0.15787012949585916,
"num_tokens": 2904750.0,
"step": 1270
},
{
"entropy": 6.286509704589844,
"epoch": 0.12247838616714697,
"grad_norm": 1.015625,
"learning_rate": 0.0004999921563575022,
"loss": 6.0399,
"mean_token_accuracy": 0.1519160158932209,
"num_tokens": 2916150.0,
"step": 1275
},
{
"entropy": 6.238752746582032,
"epoch": 0.12295869356388088,
"grad_norm": 1.015625,
"learning_rate": 0.0004999918674830169,
"loss": 6.0001,
"mean_token_accuracy": 0.15371812880039215,
"num_tokens": 2928452.0,
"step": 1280
},
{
"entropy": 6.158025789260864,
"epoch": 0.12343900096061479,
"grad_norm": 1.0,
"learning_rate": 0.0004999915733848886,
"loss": 5.999,
"mean_token_accuracy": 0.15188550055027009,
"num_tokens": 2940577.0,
"step": 1285
},
{
"entropy": 6.150114059448242,
"epoch": 0.1239193083573487,
"grad_norm": 1.015625,
"learning_rate": 0.000499991274063124,
"loss": 5.9678,
"mean_token_accuracy": 0.1559542365372181,
"num_tokens": 2952302.0,
"step": 1290
},
{
"entropy": 6.235889863967896,
"epoch": 0.12439961575408261,
"grad_norm": 1.1875,
"learning_rate": 0.0004999909695177301,
"loss": 6.029,
"mean_token_accuracy": 0.15738558918237686,
"num_tokens": 2964611.0,
"step": 1295
},
{
"entropy": 6.222271394729614,
"epoch": 0.12487992315081652,
"grad_norm": 1.1015625,
"learning_rate": 0.000499990659748714,
"loss": 6.0135,
"mean_token_accuracy": 0.1533857174217701,
"num_tokens": 2975668.0,
"step": 1300
},
{
"entropy": 6.2896746635437015,
"epoch": 0.12536023054755044,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999903447560828,
"loss": 6.0706,
"mean_token_accuracy": 0.15294826924800872,
"num_tokens": 2987303.0,
"step": 1305
},
{
"entropy": 6.250040912628174,
"epoch": 0.12584053794428435,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999900245398439,
"loss": 5.9698,
"mean_token_accuracy": 0.15828934013843537,
"num_tokens": 3000400.0,
"step": 1310
},
{
"entropy": 6.187452220916748,
"epoch": 0.12632084534101826,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999896991000047,
"loss": 5.9104,
"mean_token_accuracy": 0.1555943712592125,
"num_tokens": 3012336.0,
"step": 1315
},
{
"entropy": 6.189295625686645,
"epoch": 0.12680115273775217,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999893684365729,
"loss": 5.9769,
"mean_token_accuracy": 0.15539684668183326,
"num_tokens": 3023004.0,
"step": 1320
},
{
"entropy": 6.154639005661011,
"epoch": 0.12728146013448607,
"grad_norm": 1.015625,
"learning_rate": 0.0004999890325495559,
"loss": 6.0444,
"mean_token_accuracy": 0.15357957929372787,
"num_tokens": 3035147.0,
"step": 1325
},
{
"entropy": 6.1711499214172365,
"epoch": 0.12776176753121998,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999886914389617,
"loss": 5.8703,
"mean_token_accuracy": 0.15838308334350587,
"num_tokens": 3045611.0,
"step": 1330
},
{
"entropy": 6.135772943496704,
"epoch": 0.1282420749279539,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999883451047981,
"loss": 5.8891,
"mean_token_accuracy": 0.1591949164867401,
"num_tokens": 3056420.0,
"step": 1335
},
{
"entropy": 6.1266110897064205,
"epoch": 0.1287223823246878,
"grad_norm": 0.953125,
"learning_rate": 0.0004999879935470733,
"loss": 5.8705,
"mean_token_accuracy": 0.16175246238708496,
"num_tokens": 3068770.0,
"step": 1340
},
{
"entropy": 6.193431091308594,
"epoch": 0.1292026897214217,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999876367657954,
"loss": 6.0132,
"mean_token_accuracy": 0.14821962267160416,
"num_tokens": 3080806.0,
"step": 1345
},
{
"entropy": 6.241663312911987,
"epoch": 0.12968299711815562,
"grad_norm": 1.046875,
"learning_rate": 0.0004999872747609725,
"loss": 6.0259,
"mean_token_accuracy": 0.15041830465197564,
"num_tokens": 3091769.0,
"step": 1350
},
{
"entropy": 6.161892700195312,
"epoch": 0.13016330451488953,
"grad_norm": 1.1796875,
"learning_rate": 0.0004999869075326132,
"loss": 5.9515,
"mean_token_accuracy": 0.15427336543798448,
"num_tokens": 3103121.0,
"step": 1355
},
{
"entropy": 6.167838478088379,
"epoch": 0.13064361191162344,
"grad_norm": 0.8984375,
"learning_rate": 0.000499986535080726,
"loss": 5.9257,
"mean_token_accuracy": 0.1637582778930664,
"num_tokens": 3115606.0,
"step": 1360
},
{
"entropy": 6.070174360275269,
"epoch": 0.13112391930835735,
"grad_norm": 1.03125,
"learning_rate": 0.0004999861574053196,
"loss": 5.8353,
"mean_token_accuracy": 0.15959180593490602,
"num_tokens": 3127961.0,
"step": 1365
},
{
"entropy": 6.0476092338562015,
"epoch": 0.13160422670509125,
"grad_norm": 1.109375,
"learning_rate": 0.0004999857745064027,
"loss": 5.8378,
"mean_token_accuracy": 0.16003572195768356,
"num_tokens": 3138316.0,
"step": 1370
},
{
"entropy": 6.09722580909729,
"epoch": 0.13208453410182516,
"grad_norm": 0.953125,
"learning_rate": 0.000499985386383984,
"loss": 5.8327,
"mean_token_accuracy": 0.16383897215127946,
"num_tokens": 3150818.0,
"step": 1375
},
{
"entropy": 6.173034286499023,
"epoch": 0.13256484149855907,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999849930380729,
"loss": 5.9896,
"mean_token_accuracy": 0.15335828065872192,
"num_tokens": 3162066.0,
"step": 1380
},
{
"entropy": 6.091350126266479,
"epoch": 0.13304514889529298,
"grad_norm": 1.140625,
"learning_rate": 0.0004999845944686781,
"loss": 5.9441,
"mean_token_accuracy": 0.15436633825302123,
"num_tokens": 3172209.0,
"step": 1385
},
{
"entropy": 6.033679103851318,
"epoch": 0.1335254562920269,
"grad_norm": 1.03125,
"learning_rate": 0.0004999841906758093,
"loss": 5.7728,
"mean_token_accuracy": 0.17072101235389708,
"num_tokens": 3183248.0,
"step": 1390
},
{
"entropy": 6.115644407272339,
"epoch": 0.1340057636887608,
"grad_norm": 1.078125,
"learning_rate": 0.0004999837816594757,
"loss": 5.8801,
"mean_token_accuracy": 0.1629835158586502,
"num_tokens": 3194748.0,
"step": 1395
},
{
"entropy": 6.074833250045776,
"epoch": 0.1344860710854947,
"grad_norm": 1.09375,
"learning_rate": 0.0004999833674196865,
"loss": 5.8424,
"mean_token_accuracy": 0.16948780417442322,
"num_tokens": 3205669.0,
"step": 1400
},
{
"entropy": 6.105074739456176,
"epoch": 0.13496637848222862,
"grad_norm": 1.09375,
"learning_rate": 0.0004999829479564518,
"loss": 5.9361,
"mean_token_accuracy": 0.1520385757088661,
"num_tokens": 3216035.0,
"step": 1405
},
{
"entropy": 6.140946578979492,
"epoch": 0.13544668587896252,
"grad_norm": 1.1328125,
"learning_rate": 0.000499982523269781,
"loss": 5.9206,
"mean_token_accuracy": 0.15891541689634323,
"num_tokens": 3227192.0,
"step": 1410
},
{
"entropy": 6.1240339279174805,
"epoch": 0.13592699327569643,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999820933596842,
"loss": 5.9534,
"mean_token_accuracy": 0.1581968516111374,
"num_tokens": 3240237.0,
"step": 1415
},
{
"entropy": 6.090185928344726,
"epoch": 0.13640730067243034,
"grad_norm": 1.0625,
"learning_rate": 0.000499981658226171,
"loss": 5.8259,
"mean_token_accuracy": 0.16865789741277695,
"num_tokens": 3251963.0,
"step": 1420
},
{
"entropy": 6.063898086547852,
"epoch": 0.13688760806916425,
"grad_norm": 1.0625,
"learning_rate": 0.000499981217869252,
"loss": 5.9439,
"mean_token_accuracy": 0.1649617299437523,
"num_tokens": 3263101.0,
"step": 1425
},
{
"entropy": 6.1431561470031735,
"epoch": 0.1373679154658982,
"grad_norm": 1.1953125,
"learning_rate": 0.000499980772288937,
"loss": 5.8366,
"mean_token_accuracy": 0.16629794836044312,
"num_tokens": 3275100.0,
"step": 1430
},
{
"entropy": 6.106657552719116,
"epoch": 0.1378482228626321,
"grad_norm": 0.984375,
"learning_rate": 0.0004999803214852367,
"loss": 5.9272,
"mean_token_accuracy": 0.15512363463640214,
"num_tokens": 3287025.0,
"step": 1435
},
{
"entropy": 6.167088365554809,
"epoch": 0.138328530259366,
"grad_norm": 0.984375,
"learning_rate": 0.0004999798654581613,
"loss": 5.9322,
"mean_token_accuracy": 0.15951334238052367,
"num_tokens": 3299867.0,
"step": 1440
},
{
"entropy": 6.086508417129517,
"epoch": 0.13880883765609991,
"grad_norm": 1.140625,
"learning_rate": 0.0004999794042077214,
"loss": 5.8733,
"mean_token_accuracy": 0.15924615412950516,
"num_tokens": 3311183.0,
"step": 1445
},
{
"entropy": 6.092142724990845,
"epoch": 0.13928914505283382,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999789377339279,
"loss": 5.9375,
"mean_token_accuracy": 0.1590658202767372,
"num_tokens": 3322247.0,
"step": 1450
},
{
"entropy": 6.110898780822754,
"epoch": 0.13976945244956773,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999784660367915,
"loss": 5.8514,
"mean_token_accuracy": 0.16008831560611725,
"num_tokens": 3333369.0,
"step": 1455
},
{
"entropy": 6.026701641082764,
"epoch": 0.14024975984630164,
"grad_norm": 1.140625,
"learning_rate": 0.0004999779891163231,
"loss": 5.8907,
"mean_token_accuracy": 0.1609531357884407,
"num_tokens": 3345876.0,
"step": 1460
},
{
"entropy": 6.066957569122314,
"epoch": 0.14073006724303555,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999775069725339,
"loss": 5.778,
"mean_token_accuracy": 0.16528390794992448,
"num_tokens": 3357323.0,
"step": 1465
},
{
"entropy": 6.024584341049194,
"epoch": 0.14121037463976946,
"grad_norm": 1.1640625,
"learning_rate": 0.000499977019605435,
"loss": 5.8666,
"mean_token_accuracy": 0.1610391244292259,
"num_tokens": 3367689.0,
"step": 1470
},
{
"entropy": 5.971542263031006,
"epoch": 0.14169068203650337,
"grad_norm": 1.046875,
"learning_rate": 0.0004999765270150378,
"loss": 5.8225,
"mean_token_accuracy": 0.16424132734537125,
"num_tokens": 3379472.0,
"step": 1475
},
{
"entropy": 6.139827394485474,
"epoch": 0.14217098943323728,
"grad_norm": 1.03125,
"learning_rate": 0.0004999760292013536,
"loss": 5.8726,
"mean_token_accuracy": 0.15778491795063018,
"num_tokens": 3390929.0,
"step": 1480
},
{
"entropy": 6.217739486694336,
"epoch": 0.14265129682997119,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999755261643941,
"loss": 5.8688,
"mean_token_accuracy": 0.16537885516881942,
"num_tokens": 3401242.0,
"step": 1485
},
{
"entropy": 5.921372890472412,
"epoch": 0.1431316042267051,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999750179041709,
"loss": 5.8606,
"mean_token_accuracy": 0.16257801353931428,
"num_tokens": 3411169.0,
"step": 1490
},
{
"entropy": 6.042217111587524,
"epoch": 0.143611911623439,
"grad_norm": 1.140625,
"learning_rate": 0.0004999745044206959,
"loss": 5.6965,
"mean_token_accuracy": 0.16992068588733672,
"num_tokens": 3423265.0,
"step": 1495
},
{
"entropy": 5.984377431869507,
"epoch": 0.1440922190201729,
"grad_norm": 1.03125,
"learning_rate": 0.0004999739857139809,
"loss": 5.8066,
"mean_token_accuracy": 0.1671621523797512,
"num_tokens": 3434793.0,
"step": 1500
},
{
"entropy": 5.961929225921631,
"epoch": 0.14457252641690682,
"grad_norm": 1.015625,
"learning_rate": 0.000499973461784038,
"loss": 5.7414,
"mean_token_accuracy": 0.17664009779691697,
"num_tokens": 3445732.0,
"step": 1505
},
{
"entropy": 6.045278692245484,
"epoch": 0.14505283381364073,
"grad_norm": 1.015625,
"learning_rate": 0.0004999729326308792,
"loss": 5.9086,
"mean_token_accuracy": 0.16037625521421434,
"num_tokens": 3457090.0,
"step": 1510
},
{
"entropy": 6.174110746383667,
"epoch": 0.14553314121037464,
"grad_norm": 1.2109375,
"learning_rate": 0.000499972398254517,
"loss": 5.8945,
"mean_token_accuracy": 0.15567544400691985,
"num_tokens": 3468087.0,
"step": 1515
},
{
"entropy": 6.075960445404053,
"epoch": 0.14601344860710855,
"grad_norm": 1.1484375,
"learning_rate": 0.000499971858654964,
"loss": 5.8393,
"mean_token_accuracy": 0.16277710646390914,
"num_tokens": 3478820.0,
"step": 1520
},
{
"entropy": 6.011015510559082,
"epoch": 0.14649375600384246,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999713138322321,
"loss": 5.864,
"mean_token_accuracy": 0.15902443826198578,
"num_tokens": 3489878.0,
"step": 1525
},
{
"entropy": 6.096222686767578,
"epoch": 0.14697406340057637,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999707637863346,
"loss": 5.849,
"mean_token_accuracy": 0.16232544481754302,
"num_tokens": 3500944.0,
"step": 1530
},
{
"entropy": 5.966727018356323,
"epoch": 0.14745437079731027,
"grad_norm": 1.15625,
"learning_rate": 0.0004999702085172838,
"loss": 5.8314,
"mean_token_accuracy": 0.16640263050794601,
"num_tokens": 3511383.0,
"step": 1535
},
{
"entropy": 6.095454359054566,
"epoch": 0.14793467819404418,
"grad_norm": 1.0,
"learning_rate": 0.0004999696480250929,
"loss": 5.921,
"mean_token_accuracy": 0.1610638305544853,
"num_tokens": 3523300.0,
"step": 1540
},
{
"entropy": 6.080443000793457,
"epoch": 0.1484149855907781,
"grad_norm": 1.25,
"learning_rate": 0.0004999690823097747,
"loss": 5.8507,
"mean_token_accuracy": 0.15561971068382263,
"num_tokens": 3534371.0,
"step": 1545
},
{
"entropy": 6.039649724960327,
"epoch": 0.148895292987512,
"grad_norm": 1.171875,
"learning_rate": 0.0004999685113713426,
"loss": 5.8226,
"mean_token_accuracy": 0.16583748608827592,
"num_tokens": 3544847.0,
"step": 1550
},
{
"entropy": 6.0535636901855465,
"epoch": 0.1493756003842459,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999679352098096,
"loss": 5.7929,
"mean_token_accuracy": 0.16670662462711333,
"num_tokens": 3555859.0,
"step": 1555
},
{
"entropy": 5.992221164703369,
"epoch": 0.14985590778097982,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999673538251891,
"loss": 5.8236,
"mean_token_accuracy": 0.166217839717865,
"num_tokens": 3568283.0,
"step": 1560
},
{
"entropy": 5.987648916244507,
"epoch": 0.15033621517771373,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999667672174947,
"loss": 5.8857,
"mean_token_accuracy": 0.15877313315868377,
"num_tokens": 3581442.0,
"step": 1565
},
{
"entropy": 6.158979511260986,
"epoch": 0.15081652257444764,
"grad_norm": 1.078125,
"learning_rate": 0.00049996617538674,
"loss": 5.9279,
"mean_token_accuracy": 0.15592343956232071,
"num_tokens": 3594055.0,
"step": 1570
},
{
"entropy": 6.06167197227478,
"epoch": 0.15129682997118155,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999655783329386,
"loss": 5.8897,
"mean_token_accuracy": 0.15909540206193923,
"num_tokens": 3605952.0,
"step": 1575
},
{
"entropy": 6.10714282989502,
"epoch": 0.15177713736791545,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999649760561046,
"loss": 5.9099,
"mean_token_accuracy": 0.1594424292445183,
"num_tokens": 3618544.0,
"step": 1580
},
{
"entropy": 6.059627628326416,
"epoch": 0.15225744476464936,
"grad_norm": 1.03125,
"learning_rate": 0.0004999643685562519,
"loss": 5.8529,
"mean_token_accuracy": 0.164482718706131,
"num_tokens": 3630445.0,
"step": 1585
},
{
"entropy": 6.061960506439209,
"epoch": 0.15273775216138327,
"grad_norm": 1.15625,
"learning_rate": 0.0004999637558333945,
"loss": 5.8683,
"mean_token_accuracy": 0.16716394126415252,
"num_tokens": 3642516.0,
"step": 1590
},
{
"entropy": 5.999404573440552,
"epoch": 0.15321805955811718,
"grad_norm": 1.078125,
"learning_rate": 0.0004999631378875467,
"loss": 5.7843,
"mean_token_accuracy": 0.1711465001106262,
"num_tokens": 3654425.0,
"step": 1595
},
{
"entropy": 5.989068651199341,
"epoch": 0.15369836695485112,
"grad_norm": 1.015625,
"learning_rate": 0.0004999625147187228,
"loss": 5.8055,
"mean_token_accuracy": 0.1635123535990715,
"num_tokens": 3666521.0,
"step": 1600
},
{
"entropy": 6.065861177444458,
"epoch": 0.15417867435158503,
"grad_norm": 1.0,
"learning_rate": 0.0004999618863269373,
"loss": 5.8508,
"mean_token_accuracy": 0.15598922967910767,
"num_tokens": 3679121.0,
"step": 1605
},
{
"entropy": 6.048035669326782,
"epoch": 0.15465898174831894,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999612527122049,
"loss": 5.8645,
"mean_token_accuracy": 0.15967089980840682,
"num_tokens": 3691095.0,
"step": 1610
},
{
"entropy": 5.985746097564697,
"epoch": 0.15513928914505284,
"grad_norm": 0.9375,
"learning_rate": 0.0004999606138745402,
"loss": 5.8272,
"mean_token_accuracy": 0.16446878761053085,
"num_tokens": 3703426.0,
"step": 1615
},
{
"entropy": 6.084816551208496,
"epoch": 0.15561959654178675,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999599698139581,
"loss": 5.8925,
"mean_token_accuracy": 0.16607261896133424,
"num_tokens": 3715429.0,
"step": 1620
},
{
"entropy": 6.091441202163696,
"epoch": 0.15609990393852066,
"grad_norm": 1.046875,
"learning_rate": 0.0004999593205304734,
"loss": 5.8779,
"mean_token_accuracy": 0.16095825433731079,
"num_tokens": 3726327.0,
"step": 1625
},
{
"entropy": 6.048316621780396,
"epoch": 0.15658021133525457,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999586660241012,
"loss": 5.8316,
"mean_token_accuracy": 0.15777417570352553,
"num_tokens": 3736818.0,
"step": 1630
},
{
"entropy": 6.080630207061768,
"epoch": 0.15706051873198848,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999580062948569,
"loss": 5.8335,
"mean_token_accuracy": 0.16248745918273927,
"num_tokens": 3747776.0,
"step": 1635
},
{
"entropy": 5.891999816894531,
"epoch": 0.1575408261287224,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999573413427556,
"loss": 5.6917,
"mean_token_accuracy": 0.1702808439731598,
"num_tokens": 3758990.0,
"step": 1640
},
{
"entropy": 5.972564077377319,
"epoch": 0.1580211335254563,
"grad_norm": 1.015625,
"learning_rate": 0.0004999566711678128,
"loss": 5.7612,
"mean_token_accuracy": 0.165286822617054,
"num_tokens": 3769686.0,
"step": 1645
},
{
"entropy": 6.024297380447388,
"epoch": 0.1585014409221902,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999559957700442,
"loss": 5.8409,
"mean_token_accuracy": 0.16243061274290085,
"num_tokens": 3781815.0,
"step": 1650
},
{
"entropy": 6.015322732925415,
"epoch": 0.15898174831892412,
"grad_norm": 1.046875,
"learning_rate": 0.0004999553151494653,
"loss": 5.8773,
"mean_token_accuracy": 0.1623807817697525,
"num_tokens": 3793392.0,
"step": 1655
},
{
"entropy": 6.032832622528076,
"epoch": 0.15946205571565802,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999546293060919,
"loss": 5.826,
"mean_token_accuracy": 0.16092282980680467,
"num_tokens": 3804974.0,
"step": 1660
},
{
"entropy": 5.871552133560181,
"epoch": 0.15994236311239193,
"grad_norm": 0.96875,
"learning_rate": 0.00049995393823994,
"loss": 5.6694,
"mean_token_accuracy": 0.1742493450641632,
"num_tokens": 3817166.0,
"step": 1665
},
{
"entropy": 6.0467840194702145,
"epoch": 0.16042267050912584,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999532419510255,
"loss": 5.798,
"mean_token_accuracy": 0.16097687929868698,
"num_tokens": 3828151.0,
"step": 1670
},
{
"entropy": 5.955699825286866,
"epoch": 0.16090297790585975,
"grad_norm": 0.99609375,
"learning_rate": 0.000499952540439365,
"loss": 5.81,
"mean_token_accuracy": 0.164180888235569,
"num_tokens": 3839439.0,
"step": 1675
},
{
"entropy": 6.0351283073425295,
"epoch": 0.16138328530259366,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999518337049743,
"loss": 5.8468,
"mean_token_accuracy": 0.16053693741559982,
"num_tokens": 3851694.0,
"step": 1680
},
{
"entropy": 5.977187013626098,
"epoch": 0.16186359269932757,
"grad_norm": 0.9765625,
"learning_rate": 0.00049995112174787,
"loss": 5.8331,
"mean_token_accuracy": 0.1628563791513443,
"num_tokens": 3863593.0,
"step": 1685
},
{
"entropy": 5.929252243041992,
"epoch": 0.16234390009606148,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999504045680687,
"loss": 5.7497,
"mean_token_accuracy": 0.17348818629980087,
"num_tokens": 3874588.0,
"step": 1690
},
{
"entropy": 6.014736175537109,
"epoch": 0.1628242074927954,
"grad_norm": 1.0625,
"learning_rate": 0.0004999496821655869,
"loss": 5.8407,
"mean_token_accuracy": 0.163246688246727,
"num_tokens": 3884662.0,
"step": 1695
},
{
"entropy": 6.109212112426758,
"epoch": 0.1633045148895293,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999489545404414,
"loss": 5.9364,
"mean_token_accuracy": 0.1559630572795868,
"num_tokens": 3896569.0,
"step": 1700
},
{
"entropy": 6.078909492492675,
"epoch": 0.1637848222862632,
"grad_norm": 1.03125,
"learning_rate": 0.0004999482216926493,
"loss": 5.7962,
"mean_token_accuracy": 0.16575282365083693,
"num_tokens": 3907691.0,
"step": 1705
},
{
"entropy": 6.001922178268432,
"epoch": 0.1642651296829971,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999474836222273,
"loss": 5.8099,
"mean_token_accuracy": 0.16902267709374427,
"num_tokens": 3918794.0,
"step": 1710
},
{
"entropy": 5.976260614395142,
"epoch": 0.16474543707973102,
"grad_norm": 1.015625,
"learning_rate": 0.0004999467403291928,
"loss": 5.7941,
"mean_token_accuracy": 0.17384963780641555,
"num_tokens": 3929773.0,
"step": 1715
},
{
"entropy": 5.973721170425415,
"epoch": 0.16522574447646493,
"grad_norm": 1.125,
"learning_rate": 0.0004999459918135628,
"loss": 5.8159,
"mean_token_accuracy": 0.16380798667669297,
"num_tokens": 3940264.0,
"step": 1720
},
{
"entropy": 5.907494354248047,
"epoch": 0.16570605187319884,
"grad_norm": 1.09375,
"learning_rate": 0.000499945238075355,
"loss": 5.7114,
"mean_token_accuracy": 0.1722415491938591,
"num_tokens": 3951500.0,
"step": 1725
},
{
"entropy": 6.023552942276001,
"epoch": 0.16618635926993275,
"grad_norm": 1.09375,
"learning_rate": 0.0004999444791145865,
"loss": 5.7824,
"mean_token_accuracy": 0.1692255288362503,
"num_tokens": 3963580.0,
"step": 1730
},
{
"entropy": 5.8660492420196535,
"epoch": 0.16666666666666666,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999437149312754,
"loss": 5.7285,
"mean_token_accuracy": 0.17743493169546126,
"num_tokens": 3975994.0,
"step": 1735
},
{
"entropy": 6.00585618019104,
"epoch": 0.16714697406340057,
"grad_norm": 1.0390625,
"learning_rate": 0.000499942945525439,
"loss": 5.745,
"mean_token_accuracy": 0.16235984563827516,
"num_tokens": 3987897.0,
"step": 1740
},
{
"entropy": 5.995152854919434,
"epoch": 0.16762728146013448,
"grad_norm": 1.09375,
"learning_rate": 0.0004999421708970954,
"loss": 5.8941,
"mean_token_accuracy": 0.15724924057722092,
"num_tokens": 3999829.0,
"step": 1745
},
{
"entropy": 5.933386898040771,
"epoch": 0.16810758885686838,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999413910462625,
"loss": 5.7299,
"mean_token_accuracy": 0.17098538130521773,
"num_tokens": 4010882.0,
"step": 1750
},
{
"entropy": 5.917968273162842,
"epoch": 0.1685878962536023,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999406059729586,
"loss": 5.7206,
"mean_token_accuracy": 0.16878145337104797,
"num_tokens": 4021423.0,
"step": 1755
},
{
"entropy": 6.074040222167969,
"epoch": 0.1690682036503362,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999398156772016,
"loss": 5.8527,
"mean_token_accuracy": 0.1553002655506134,
"num_tokens": 4033590.0,
"step": 1760
},
{
"entropy": 5.8794393062591555,
"epoch": 0.16954851104707014,
"grad_norm": 1.125,
"learning_rate": 0.00049993902015901,
"loss": 5.7225,
"mean_token_accuracy": 0.17117420434951783,
"num_tokens": 4043978.0,
"step": 1765
},
{
"entropy": 6.0071815013885494,
"epoch": 0.17002881844380405,
"grad_norm": 1.140625,
"learning_rate": 0.0004999382194184023,
"loss": 5.8558,
"mean_token_accuracy": 0.16220006048679353,
"num_tokens": 4054513.0,
"step": 1770
},
{
"entropy": 6.068150615692138,
"epoch": 0.17050912584053796,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999374134553972,
"loss": 5.7949,
"mean_token_accuracy": 0.16809019297361374,
"num_tokens": 4066019.0,
"step": 1775
},
{
"entropy": 5.9036009311676025,
"epoch": 0.17098943323727187,
"grad_norm": 1.0,
"learning_rate": 0.0004999366022700131,
"loss": 5.7542,
"mean_token_accuracy": 0.17212767601013185,
"num_tokens": 4077688.0,
"step": 1780
},
{
"entropy": 6.022014522552491,
"epoch": 0.17146974063400577,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999357858622691,
"loss": 5.8467,
"mean_token_accuracy": 0.16922961324453353,
"num_tokens": 4089803.0,
"step": 1785
},
{
"entropy": 5.994244956970215,
"epoch": 0.17195004803073968,
"grad_norm": 0.984375,
"learning_rate": 0.0004999349642321842,
"loss": 5.7916,
"mean_token_accuracy": 0.1677403450012207,
"num_tokens": 4101969.0,
"step": 1790
},
{
"entropy": 5.974350070953369,
"epoch": 0.1724303554274736,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999341373797772,
"loss": 5.7811,
"mean_token_accuracy": 0.16280766427516938,
"num_tokens": 4113567.0,
"step": 1795
},
{
"entropy": 5.9534135341644285,
"epoch": 0.1729106628242075,
"grad_norm": 1.046875,
"learning_rate": 0.0004999333053050675,
"loss": 5.7357,
"mean_token_accuracy": 0.1677995041012764,
"num_tokens": 4125191.0,
"step": 1800
},
{
"entropy": 5.986348867416382,
"epoch": 0.1733909702209414,
"grad_norm": 1.203125,
"learning_rate": 0.0004999324680080744,
"loss": 5.7731,
"mean_token_accuracy": 0.16389087438583375,
"num_tokens": 4135050.0,
"step": 1805
},
{
"entropy": 5.944538688659668,
"epoch": 0.17387127761767532,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999316254888172,
"loss": 5.8327,
"mean_token_accuracy": 0.1667800709605217,
"num_tokens": 4146874.0,
"step": 1810
},
{
"entropy": 5.925647783279419,
"epoch": 0.17435158501440923,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999307777473157,
"loss": 5.7567,
"mean_token_accuracy": 0.16681902706623078,
"num_tokens": 4158118.0,
"step": 1815
},
{
"entropy": 5.988029193878174,
"epoch": 0.17483189241114314,
"grad_norm": 1.2421875,
"learning_rate": 0.0004999299247835893,
"loss": 5.7486,
"mean_token_accuracy": 0.17824842482805253,
"num_tokens": 4169035.0,
"step": 1820
},
{
"entropy": 5.898885297775268,
"epoch": 0.17531219980787704,
"grad_norm": 1.1015625,
"learning_rate": 0.000499929066597658,
"loss": 5.7262,
"mean_token_accuracy": 0.17178142070770264,
"num_tokens": 4180314.0,
"step": 1825
},
{
"entropy": 6.002191352844238,
"epoch": 0.17579250720461095,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999282031895418,
"loss": 5.8002,
"mean_token_accuracy": 0.1674297496676445,
"num_tokens": 4192238.0,
"step": 1830
},
{
"entropy": 5.8907599449157715,
"epoch": 0.17627281460134486,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999273345592604,
"loss": 5.7252,
"mean_token_accuracy": 0.16727315485477448,
"num_tokens": 4203346.0,
"step": 1835
},
{
"entropy": 5.895097637176514,
"epoch": 0.17675312199807877,
"grad_norm": 1.03125,
"learning_rate": 0.0004999264607068343,
"loss": 5.7977,
"mean_token_accuracy": 0.1713564246892929,
"num_tokens": 4213763.0,
"step": 1840
},
{
"entropy": 5.920028400421143,
"epoch": 0.17723342939481268,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999255816322837,
"loss": 5.754,
"mean_token_accuracy": 0.16816631257534026,
"num_tokens": 4225553.0,
"step": 1845
},
{
"entropy": 6.013068199157715,
"epoch": 0.1777137367915466,
"grad_norm": 1.0546875,
"learning_rate": 0.000499924697335629,
"loss": 5.6628,
"mean_token_accuracy": 0.17668255865573884,
"num_tokens": 4236058.0,
"step": 1850
},
{
"entropy": 5.8163008213043215,
"epoch": 0.1781940441882805,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999238078168906,
"loss": 5.7584,
"mean_token_accuracy": 0.17156918495893478,
"num_tokens": 4248299.0,
"step": 1855
},
{
"entropy": 5.956279039382935,
"epoch": 0.1786743515850144,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999229130760894,
"loss": 5.6868,
"mean_token_accuracy": 0.17354945242404937,
"num_tokens": 4259704.0,
"step": 1860
},
{
"entropy": 5.824860429763794,
"epoch": 0.17915465898174832,
"grad_norm": 1.0546875,
"learning_rate": 0.000499922013113246,
"loss": 5.5708,
"mean_token_accuracy": 0.1824245736002922,
"num_tokens": 4270480.0,
"step": 1865
},
{
"entropy": 5.922730875015259,
"epoch": 0.17963496637848222,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999211079283814,
"loss": 5.8211,
"mean_token_accuracy": 0.16856807470321655,
"num_tokens": 4282104.0,
"step": 1870
},
{
"entropy": 6.019510555267334,
"epoch": 0.18011527377521613,
"grad_norm": 0.984375,
"learning_rate": 0.0004999201975215164,
"loss": 5.7974,
"mean_token_accuracy": 0.16135748475790024,
"num_tokens": 4294251.0,
"step": 1875
},
{
"entropy": 5.900371026992798,
"epoch": 0.18059558117195004,
"grad_norm": 1.0,
"learning_rate": 0.0004999192818926725,
"loss": 5.6804,
"mean_token_accuracy": 0.1701715335249901,
"num_tokens": 4305569.0,
"step": 1880
},
{
"entropy": 5.941131639480591,
"epoch": 0.18107588856868395,
"grad_norm": 1.125,
"learning_rate": 0.0004999183610418706,
"loss": 5.8057,
"mean_token_accuracy": 0.1641265444457531,
"num_tokens": 4317845.0,
"step": 1885
},
{
"entropy": 5.888107252120972,
"epoch": 0.18155619596541786,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999174349691322,
"loss": 5.6809,
"mean_token_accuracy": 0.17280717492103576,
"num_tokens": 4329987.0,
"step": 1890
},
{
"entropy": 5.836309289932251,
"epoch": 0.18203650336215177,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999165036744788,
"loss": 5.6926,
"mean_token_accuracy": 0.17172161340713502,
"num_tokens": 4341628.0,
"step": 1895
},
{
"entropy": 5.965258646011352,
"epoch": 0.18251681075888568,
"grad_norm": 1.140625,
"learning_rate": 0.0004999155671579322,
"loss": 5.7545,
"mean_token_accuracy": 0.16519593745470046,
"num_tokens": 4352379.0,
"step": 1900
},
{
"entropy": 5.800844955444336,
"epoch": 0.1829971181556196,
"grad_norm": 1.1171875,
"learning_rate": 0.000499914625419514,
"loss": 5.7045,
"mean_token_accuracy": 0.17391971200704576,
"num_tokens": 4364800.0,
"step": 1905
},
{
"entropy": 5.960105085372925,
"epoch": 0.1834774255523535,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999136784592459,
"loss": 5.7164,
"mean_token_accuracy": 0.1672212541103363,
"num_tokens": 4376048.0,
"step": 1910
},
{
"entropy": 5.888760232925415,
"epoch": 0.1839577329490874,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999127262771502,
"loss": 5.7058,
"mean_token_accuracy": 0.17140705287456512,
"num_tokens": 4388072.0,
"step": 1915
},
{
"entropy": 5.9937965869903564,
"epoch": 0.1844380403458213,
"grad_norm": 1.1640625,
"learning_rate": 0.0004999117688732487,
"loss": 5.8004,
"mean_token_accuracy": 0.1623227506875992,
"num_tokens": 4399843.0,
"step": 1920
},
{
"entropy": 5.803120040893555,
"epoch": 0.18491834774255522,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999108062475638,
"loss": 5.6574,
"mean_token_accuracy": 0.17487162053585054,
"num_tokens": 4411373.0,
"step": 1925
},
{
"entropy": 5.846405506134033,
"epoch": 0.18539865513928913,
"grad_norm": 1.1171875,
"learning_rate": 0.000499909838400118,
"loss": 5.6564,
"mean_token_accuracy": 0.17340814918279648,
"num_tokens": 4421857.0,
"step": 1930
},
{
"entropy": 5.95341854095459,
"epoch": 0.18587896253602307,
"grad_norm": 1.03125,
"learning_rate": 0.0004999088653309334,
"loss": 5.7408,
"mean_token_accuracy": 0.16888684779405594,
"num_tokens": 4432728.0,
"step": 1935
},
{
"entropy": 5.844529104232788,
"epoch": 0.18635926993275698,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999078870400329,
"loss": 5.6742,
"mean_token_accuracy": 0.1769302397966385,
"num_tokens": 4444683.0,
"step": 1940
},
{
"entropy": 5.96117753982544,
"epoch": 0.18683957732949089,
"grad_norm": 0.984375,
"learning_rate": 0.0004999069035274391,
"loss": 5.8016,
"mean_token_accuracy": 0.16333086788654327,
"num_tokens": 4456961.0,
"step": 1945
},
{
"entropy": 5.875500202178955,
"epoch": 0.1873198847262248,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999059147931747,
"loss": 5.6392,
"mean_token_accuracy": 0.17937855124473573,
"num_tokens": 4468424.0,
"step": 1950
},
{
"entropy": 5.871504831314087,
"epoch": 0.1878001921229587,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999049208372629,
"loss": 5.8396,
"mean_token_accuracy": 0.15889065861701965,
"num_tokens": 4479813.0,
"step": 1955
},
{
"entropy": 6.061702728271484,
"epoch": 0.1882804995196926,
"grad_norm": 1.1875,
"learning_rate": 0.0004999039216597267,
"loss": 5.8364,
"mean_token_accuracy": 0.16994454562664033,
"num_tokens": 4491172.0,
"step": 1960
},
{
"entropy": 5.844229984283447,
"epoch": 0.18876080691642652,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999029172605892,
"loss": 5.7157,
"mean_token_accuracy": 0.17361883521080018,
"num_tokens": 4503063.0,
"step": 1965
},
{
"entropy": 5.998745012283325,
"epoch": 0.18924111431316043,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999019076398738,
"loss": 5.8005,
"mean_token_accuracy": 0.1544879749417305,
"num_tokens": 4514188.0,
"step": 1970
},
{
"entropy": 5.911010074615478,
"epoch": 0.18972142170989434,
"grad_norm": 1.0390625,
"learning_rate": 0.000499900892797604,
"loss": 5.7089,
"mean_token_accuracy": 0.17505204379558564,
"num_tokens": 4525293.0,
"step": 1975
},
{
"entropy": 5.906450510025024,
"epoch": 0.19020172910662825,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998998727338031,
"loss": 5.785,
"mean_token_accuracy": 0.16963134258985518,
"num_tokens": 4536589.0,
"step": 1980
},
{
"entropy": 5.855934715270996,
"epoch": 0.19068203650336216,
"grad_norm": 1.0625,
"learning_rate": 0.0004998988474484952,
"loss": 5.5373,
"mean_token_accuracy": 0.1919272780418396,
"num_tokens": 4547594.0,
"step": 1985
},
{
"entropy": 5.856652116775512,
"epoch": 0.19116234390009607,
"grad_norm": 0.953125,
"learning_rate": 0.0004998978169417038,
"loss": 5.7494,
"mean_token_accuracy": 0.17164870202541352,
"num_tokens": 4559850.0,
"step": 1990
},
{
"entropy": 5.886022758483887,
"epoch": 0.19164265129682997,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998967812134529,
"loss": 5.6822,
"mean_token_accuracy": 0.1702727809548378,
"num_tokens": 4570727.0,
"step": 1995
},
{
"entropy": 5.75973162651062,
"epoch": 0.19212295869356388,
"grad_norm": 1.109375,
"learning_rate": 0.0004998957402637664,
"loss": 5.6355,
"mean_token_accuracy": 0.174292254447937,
"num_tokens": 4582248.0,
"step": 2000
},
{
"entropy": 5.904182481765747,
"epoch": 0.1926032660902978,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998946940926687,
"loss": 5.6548,
"mean_token_accuracy": 0.17286083698272706,
"num_tokens": 4592604.0,
"step": 2005
},
{
"entropy": 5.794434881210327,
"epoch": 0.1930835734870317,
"grad_norm": 1.046875,
"learning_rate": 0.000499893642700184,
"loss": 5.6982,
"mean_token_accuracy": 0.17326663583517074,
"num_tokens": 4604398.0,
"step": 2010
},
{
"entropy": 5.889737224578857,
"epoch": 0.1935638808837656,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998925860863368,
"loss": 5.7798,
"mean_token_accuracy": 0.16947566717863083,
"num_tokens": 4616434.0,
"step": 2015
},
{
"entropy": 5.953923463821411,
"epoch": 0.19404418828049952,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998915242511516,
"loss": 5.7198,
"mean_token_accuracy": 0.17970746457576753,
"num_tokens": 4627577.0,
"step": 2020
},
{
"entropy": 5.960914373397827,
"epoch": 0.19452449567723343,
"grad_norm": 1.0625,
"learning_rate": 0.0004998904571946528,
"loss": 5.7831,
"mean_token_accuracy": 0.1678330808877945,
"num_tokens": 4639698.0,
"step": 2025
},
{
"entropy": 5.958297538757324,
"epoch": 0.19500480307396734,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998893849168655,
"loss": 5.7914,
"mean_token_accuracy": 0.16515226066112518,
"num_tokens": 4650643.0,
"step": 2030
},
{
"entropy": 5.87400336265564,
"epoch": 0.19548511047070125,
"grad_norm": 0.96875,
"learning_rate": 0.0004998883074178144,
"loss": 5.7203,
"mean_token_accuracy": 0.1693568304181099,
"num_tokens": 4662897.0,
"step": 2035
},
{
"entropy": 5.9375998973846436,
"epoch": 0.19596541786743515,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998872246975247,
"loss": 5.7942,
"mean_token_accuracy": 0.17414003014564514,
"num_tokens": 4673701.0,
"step": 2040
},
{
"entropy": 5.967000341415405,
"epoch": 0.19644572526416906,
"grad_norm": 1.0,
"learning_rate": 0.0004998861367560213,
"loss": 5.7591,
"mean_token_accuracy": 0.16811660975217818,
"num_tokens": 4685873.0,
"step": 2045
},
{
"entropy": 5.925240468978882,
"epoch": 0.19692603266090297,
"grad_norm": 1.03125,
"learning_rate": 0.0004998850435933296,
"loss": 5.6411,
"mean_token_accuracy": 0.17622891664505005,
"num_tokens": 4697179.0,
"step": 2050
},
{
"entropy": 5.831101989746093,
"epoch": 0.19740634005763688,
"grad_norm": 1.09375,
"learning_rate": 0.0004998839452094749,
"loss": 5.6862,
"mean_token_accuracy": 0.1738536462187767,
"num_tokens": 4707752.0,
"step": 2055
},
{
"entropy": 5.746025085449219,
"epoch": 0.1978866474543708,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998828416044829,
"loss": 5.5372,
"mean_token_accuracy": 0.18245317041873932,
"num_tokens": 4718413.0,
"step": 2060
},
{
"entropy": 5.855678606033325,
"epoch": 0.1983669548511047,
"grad_norm": 1.1328125,
"learning_rate": 0.000499881732778379,
"loss": 5.7395,
"mean_token_accuracy": 0.16523946076631546,
"num_tokens": 4730033.0,
"step": 2065
},
{
"entropy": 5.843948650360107,
"epoch": 0.1988472622478386,
"grad_norm": 0.921875,
"learning_rate": 0.000499880618731189,
"loss": 5.586,
"mean_token_accuracy": 0.17743310034275056,
"num_tokens": 4742084.0,
"step": 2070
},
{
"entropy": 5.899618673324585,
"epoch": 0.19932756964457252,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998794994629388,
"loss": 5.8189,
"mean_token_accuracy": 0.16559504121541976,
"num_tokens": 4753885.0,
"step": 2075
},
{
"entropy": 5.831307220458984,
"epoch": 0.19980787704130643,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998783749736545,
"loss": 5.6661,
"mean_token_accuracy": 0.17495889812707902,
"num_tokens": 4765686.0,
"step": 2080
},
{
"entropy": 5.899104452133178,
"epoch": 0.20028818443804033,
"grad_norm": 1.015625,
"learning_rate": 0.0004998772452633619,
"loss": 5.7021,
"mean_token_accuracy": 0.16899442821741104,
"num_tokens": 4777157.0,
"step": 2085
},
{
"entropy": 5.828914642333984,
"epoch": 0.20076849183477424,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998761103320876,
"loss": 5.6397,
"mean_token_accuracy": 0.17771946042776107,
"num_tokens": 4788583.0,
"step": 2090
},
{
"entropy": 5.88235239982605,
"epoch": 0.20124879923150815,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998749701798577,
"loss": 5.755,
"mean_token_accuracy": 0.16652097702026367,
"num_tokens": 4800749.0,
"step": 2095
},
{
"entropy": 5.819902420043945,
"epoch": 0.2017291066282421,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998738248066986,
"loss": 5.6686,
"mean_token_accuracy": 0.17305531948804856,
"num_tokens": 4812488.0,
"step": 2100
},
{
"entropy": 5.922385215759277,
"epoch": 0.202209414024976,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998726742126372,
"loss": 5.6469,
"mean_token_accuracy": 0.17342206686735154,
"num_tokens": 4823495.0,
"step": 2105
},
{
"entropy": 5.770005416870117,
"epoch": 0.2026897214217099,
"grad_norm": 1.09375,
"learning_rate": 0.0004998715183976999,
"loss": 5.7124,
"mean_token_accuracy": 0.16901031583547593,
"num_tokens": 4834450.0,
"step": 2110
},
{
"entropy": 5.839266920089722,
"epoch": 0.20317002881844382,
"grad_norm": 0.9375,
"learning_rate": 0.0004998703573619137,
"loss": 5.6129,
"mean_token_accuracy": 0.18550968468189238,
"num_tokens": 4846826.0,
"step": 2115
},
{
"entropy": 5.970982599258423,
"epoch": 0.20365033621517772,
"grad_norm": 1.015625,
"learning_rate": 0.0004998691911053056,
"loss": 5.7977,
"mean_token_accuracy": 0.16186073422431946,
"num_tokens": 4859668.0,
"step": 2120
},
{
"entropy": 5.834139776229859,
"epoch": 0.20413064361191163,
"grad_norm": 1.109375,
"learning_rate": 0.0004998680196279026,
"loss": 5.6906,
"mean_token_accuracy": 0.17062558084726334,
"num_tokens": 4871727.0,
"step": 2125
},
{
"entropy": 5.955994033813477,
"epoch": 0.20461095100864554,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998668429297319,
"loss": 5.7366,
"mean_token_accuracy": 0.1794666439294815,
"num_tokens": 4882191.0,
"step": 2130
},
{
"entropy": 5.847929811477661,
"epoch": 0.20509125840537945,
"grad_norm": 1.1875,
"learning_rate": 0.0004998656610108208,
"loss": 5.6706,
"mean_token_accuracy": 0.17324328422546387,
"num_tokens": 4892416.0,
"step": 2135
},
{
"entropy": 5.728474712371826,
"epoch": 0.20557156580211336,
"grad_norm": 1.109375,
"learning_rate": 0.0004998644738711969,
"loss": 5.6471,
"mean_token_accuracy": 0.16920143961906434,
"num_tokens": 4903572.0,
"step": 2140
},
{
"entropy": 5.977149581909179,
"epoch": 0.20605187319884727,
"grad_norm": 0.890625,
"learning_rate": 0.0004998632815108874,
"loss": 5.7446,
"mean_token_accuracy": 0.17007820159196854,
"num_tokens": 4915417.0,
"step": 2145
},
{
"entropy": 5.810878467559815,
"epoch": 0.20653218059558118,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998620839299203,
"loss": 5.6138,
"mean_token_accuracy": 0.17479929476976394,
"num_tokens": 4926943.0,
"step": 2150
},
{
"entropy": 5.80943193435669,
"epoch": 0.2070124879923151,
"grad_norm": 1.015625,
"learning_rate": 0.0004998608811283233,
"loss": 5.5847,
"mean_token_accuracy": 0.17787374258041383,
"num_tokens": 4937724.0,
"step": 2155
},
{
"entropy": 5.957506895065308,
"epoch": 0.207492795389049,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998596731061244,
"loss": 5.7368,
"mean_token_accuracy": 0.1671774461865425,
"num_tokens": 4949970.0,
"step": 2160
},
{
"entropy": 5.856918811798096,
"epoch": 0.2079731027857829,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998584598633516,
"loss": 5.7383,
"mean_token_accuracy": 0.16863747537136078,
"num_tokens": 4961389.0,
"step": 2165
},
{
"entropy": 5.930571174621582,
"epoch": 0.2084534101825168,
"grad_norm": 1.078125,
"learning_rate": 0.0004998572414000329,
"loss": 5.787,
"mean_token_accuracy": 0.16769310384988784,
"num_tokens": 4973888.0,
"step": 2170
},
{
"entropy": 5.846202564239502,
"epoch": 0.20893371757925072,
"grad_norm": 1.125,
"learning_rate": 0.0004998560177161969,
"loss": 5.7358,
"mean_token_accuracy": 0.16144881695508956,
"num_tokens": 4985423.0,
"step": 2175
},
{
"entropy": 5.814085817337036,
"epoch": 0.20941402497598463,
"grad_norm": 0.984375,
"learning_rate": 0.0004998547888118718,
"loss": 5.7026,
"mean_token_accuracy": 0.16614219099283217,
"num_tokens": 4997711.0,
"step": 2180
},
{
"entropy": 5.891442108154297,
"epoch": 0.20989433237271854,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998535546870862,
"loss": 5.7239,
"mean_token_accuracy": 0.17229223549365996,
"num_tokens": 5009633.0,
"step": 2185
},
{
"entropy": 5.876784420013427,
"epoch": 0.21037463976945245,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998523153418687,
"loss": 5.6628,
"mean_token_accuracy": 0.17913604974746705,
"num_tokens": 5021523.0,
"step": 2190
},
{
"entropy": 5.9075572967529295,
"epoch": 0.21085494716618636,
"grad_norm": 1.0625,
"learning_rate": 0.0004998510707762481,
"loss": 5.7561,
"mean_token_accuracy": 0.1698111966252327,
"num_tokens": 5033513.0,
"step": 2195
},
{
"entropy": 5.961580562591553,
"epoch": 0.21133525456292027,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998498209902533,
"loss": 5.7608,
"mean_token_accuracy": 0.16819328367710112,
"num_tokens": 5047055.0,
"step": 2200
},
{
"entropy": 5.845474815368652,
"epoch": 0.21181556195965417,
"grad_norm": 1.125,
"learning_rate": 0.0004998485659839134,
"loss": 5.6379,
"mean_token_accuracy": 0.17564828246831893,
"num_tokens": 5057613.0,
"step": 2205
},
{
"entropy": 5.840123701095581,
"epoch": 0.21229586935638808,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998473057572575,
"loss": 5.7558,
"mean_token_accuracy": 0.16806377619504928,
"num_tokens": 5068886.0,
"step": 2210
},
{
"entropy": 5.85970664024353,
"epoch": 0.212776176753122,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998460403103146,
"loss": 5.7283,
"mean_token_accuracy": 0.167040803283453,
"num_tokens": 5079978.0,
"step": 2215
},
{
"entropy": 5.902343082427978,
"epoch": 0.2132564841498559,
"grad_norm": 1.078125,
"learning_rate": 0.0004998447696431146,
"loss": 5.6819,
"mean_token_accuracy": 0.1737098440527916,
"num_tokens": 5091021.0,
"step": 2220
},
{
"entropy": 5.76426329612732,
"epoch": 0.2137367915465898,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998434937556865,
"loss": 5.5926,
"mean_token_accuracy": 0.1794767439365387,
"num_tokens": 5101483.0,
"step": 2225
},
{
"entropy": 5.874709224700927,
"epoch": 0.21421709894332372,
"grad_norm": 1.046875,
"learning_rate": 0.0004998422126480602,
"loss": 5.712,
"mean_token_accuracy": 0.16954506933689117,
"num_tokens": 5113116.0,
"step": 2230
},
{
"entropy": 5.8735052108764645,
"epoch": 0.21469740634005763,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998409263202653,
"loss": 5.6648,
"mean_token_accuracy": 0.16956303715705873,
"num_tokens": 5124824.0,
"step": 2235
},
{
"entropy": 5.7920280456542965,
"epoch": 0.21517771373679154,
"grad_norm": 1.046875,
"learning_rate": 0.0004998396347723318,
"loss": 5.5907,
"mean_token_accuracy": 0.1711305171251297,
"num_tokens": 5137567.0,
"step": 2240
},
{
"entropy": 5.7892828464508055,
"epoch": 0.21565802113352545,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998383380042895,
"loss": 5.6577,
"mean_token_accuracy": 0.17161570936441423,
"num_tokens": 5149016.0,
"step": 2245
},
{
"entropy": 5.763986873626709,
"epoch": 0.21613832853025935,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998370360161688,
"loss": 5.5473,
"mean_token_accuracy": 0.17783051282167434,
"num_tokens": 5160356.0,
"step": 2250
},
{
"entropy": 5.848583841323853,
"epoch": 0.21661863592699326,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998357288079996,
"loss": 5.7387,
"mean_token_accuracy": 0.16622594892978668,
"num_tokens": 5172100.0,
"step": 2255
},
{
"entropy": 5.9171970844268795,
"epoch": 0.21709894332372717,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998344163798125,
"loss": 5.7146,
"mean_token_accuracy": 0.16663871705532074,
"num_tokens": 5183984.0,
"step": 2260
},
{
"entropy": 5.842860412597656,
"epoch": 0.21757925072046108,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998330987316379,
"loss": 5.6805,
"mean_token_accuracy": 0.16949303299188614,
"num_tokens": 5195853.0,
"step": 2265
},
{
"entropy": 5.7196276664733885,
"epoch": 0.21805955811719502,
"grad_norm": 1.03125,
"learning_rate": 0.0004998317758635062,
"loss": 5.5312,
"mean_token_accuracy": 0.17934199124574662,
"num_tokens": 5206995.0,
"step": 2270
},
{
"entropy": 5.686021423339843,
"epoch": 0.21853986551392893,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998304477754484,
"loss": 5.5664,
"mean_token_accuracy": 0.17936804592609407,
"num_tokens": 5219291.0,
"step": 2275
},
{
"entropy": 5.80483112335205,
"epoch": 0.21902017291066284,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998291144674952,
"loss": 5.6432,
"mean_token_accuracy": 0.17933146953582763,
"num_tokens": 5230856.0,
"step": 2280
},
{
"entropy": 5.694868040084839,
"epoch": 0.21950048030739674,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998277759396776,
"loss": 5.5055,
"mean_token_accuracy": 0.1853230893611908,
"num_tokens": 5242871.0,
"step": 2285
},
{
"entropy": 5.757525062561035,
"epoch": 0.21998078770413065,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998264321920265,
"loss": 5.6145,
"mean_token_accuracy": 0.17693334370851516,
"num_tokens": 5253835.0,
"step": 2290
},
{
"entropy": 5.757721900939941,
"epoch": 0.22046109510086456,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998250832245734,
"loss": 5.5914,
"mean_token_accuracy": 0.1778374120593071,
"num_tokens": 5266195.0,
"step": 2295
},
{
"entropy": 5.770969820022583,
"epoch": 0.22094140249759847,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998237290373494,
"loss": 5.5701,
"mean_token_accuracy": 0.17990957647562028,
"num_tokens": 5277499.0,
"step": 2300
},
{
"entropy": 5.856108522415161,
"epoch": 0.22142170989433238,
"grad_norm": 0.9921875,
"learning_rate": 0.000499822369630386,
"loss": 5.6911,
"mean_token_accuracy": 0.1632763683795929,
"num_tokens": 5288622.0,
"step": 2305
},
{
"entropy": 5.837642621994019,
"epoch": 0.2219020172910663,
"grad_norm": 1.265625,
"learning_rate": 0.0004998210050037148,
"loss": 5.7363,
"mean_token_accuracy": 0.1674153745174408,
"num_tokens": 5299664.0,
"step": 2310
},
{
"entropy": 5.8238342761993405,
"epoch": 0.2223823246878002,
"grad_norm": 1.078125,
"learning_rate": 0.0004998196351573674,
"loss": 5.6326,
"mean_token_accuracy": 0.1764681726694107,
"num_tokens": 5311627.0,
"step": 2315
},
{
"entropy": 5.69834566116333,
"epoch": 0.2228626320845341,
"grad_norm": 1.015625,
"learning_rate": 0.0004998182600913757,
"loss": 5.5328,
"mean_token_accuracy": 0.1816352590918541,
"num_tokens": 5323000.0,
"step": 2320
},
{
"entropy": 5.7694638729095455,
"epoch": 0.22334293948126802,
"grad_norm": 1.03125,
"learning_rate": 0.0004998168798057715,
"loss": 5.5689,
"mean_token_accuracy": 0.18260359317064284,
"num_tokens": 5333811.0,
"step": 2325
},
{
"entropy": 5.779385423660278,
"epoch": 0.22382324687800192,
"grad_norm": 1.046875,
"learning_rate": 0.000499815494300587,
"loss": 5.5901,
"mean_token_accuracy": 0.17485559582710267,
"num_tokens": 5344762.0,
"step": 2330
},
{
"entropy": 5.806505537033081,
"epoch": 0.22430355427473583,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998141035758542,
"loss": 5.5754,
"mean_token_accuracy": 0.18179302662611008,
"num_tokens": 5356112.0,
"step": 2335
},
{
"entropy": 5.832508087158203,
"epoch": 0.22478386167146974,
"grad_norm": 1.2265625,
"learning_rate": 0.0004998127076316054,
"loss": 5.6984,
"mean_token_accuracy": 0.1719257727265358,
"num_tokens": 5367339.0,
"step": 2340
},
{
"entropy": 5.7249836921691895,
"epoch": 0.22526416906820365,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998113064678734,
"loss": 5.6361,
"mean_token_accuracy": 0.18138026893138887,
"num_tokens": 5378627.0,
"step": 2345
},
{
"entropy": 5.868769311904908,
"epoch": 0.22574447646493756,
"grad_norm": 1.125,
"learning_rate": 0.0004998099000846901,
"loss": 5.6632,
"mean_token_accuracy": 0.17547774761915208,
"num_tokens": 5390209.0,
"step": 2350
},
{
"entropy": 5.81194429397583,
"epoch": 0.22622478386167147,
"grad_norm": 1.2578125,
"learning_rate": 0.0004998084884820887,
"loss": 5.6538,
"mean_token_accuracy": 0.1727759525179863,
"num_tokens": 5401578.0,
"step": 2355
},
{
"entropy": 5.7616051197052,
"epoch": 0.22670509125840538,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998070716601016,
"loss": 5.575,
"mean_token_accuracy": 0.17865398079156875,
"num_tokens": 5413831.0,
"step": 2360
},
{
"entropy": 5.82645959854126,
"epoch": 0.2271853986551393,
"grad_norm": 1.078125,
"learning_rate": 0.0004998056496187618,
"loss": 5.6293,
"mean_token_accuracy": 0.17196177393198014,
"num_tokens": 5425430.0,
"step": 2365
},
{
"entropy": 5.588806056976319,
"epoch": 0.2276657060518732,
"grad_norm": 1.03125,
"learning_rate": 0.0004998042223581025,
"loss": 5.4715,
"mean_token_accuracy": 0.18928515762090684,
"num_tokens": 5435353.0,
"step": 2370
},
{
"entropy": 5.8202369689941404,
"epoch": 0.2281460134486071,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998027898781565,
"loss": 5.6808,
"mean_token_accuracy": 0.1762748748064041,
"num_tokens": 5446925.0,
"step": 2375
},
{
"entropy": 5.709089374542236,
"epoch": 0.228626320845341,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998013521789574,
"loss": 5.5578,
"mean_token_accuracy": 0.18197699040174484,
"num_tokens": 5456613.0,
"step": 2380
},
{
"entropy": 5.799993944168091,
"epoch": 0.22910662824207492,
"grad_norm": 1.171875,
"learning_rate": 0.0004997999092605384,
"loss": 5.5941,
"mean_token_accuracy": 0.17672255784273147,
"num_tokens": 5467790.0,
"step": 2385
},
{
"entropy": 5.814759111404419,
"epoch": 0.22958693563880883,
"grad_norm": 1.0390625,
"learning_rate": 0.000499798461122933,
"loss": 5.5703,
"mean_token_accuracy": 0.1812552884221077,
"num_tokens": 5479166.0,
"step": 2390
},
{
"entropy": 5.6896861553192135,
"epoch": 0.23006724303554274,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997970077661748,
"loss": 5.5596,
"mean_token_accuracy": 0.18635332137346267,
"num_tokens": 5490186.0,
"step": 2395
},
{
"entropy": 5.738391304016114,
"epoch": 0.23054755043227665,
"grad_norm": 1.015625,
"learning_rate": 0.0004997955491902977,
"loss": 5.5424,
"mean_token_accuracy": 0.17801296710968018,
"num_tokens": 5500416.0,
"step": 2400
},
{
"entropy": 5.766972875595092,
"epoch": 0.23102785782901056,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997940853953354,
"loss": 5.6202,
"mean_token_accuracy": 0.1773225635290146,
"num_tokens": 5512189.0,
"step": 2405
},
{
"entropy": 5.711191701889038,
"epoch": 0.23150816522574447,
"grad_norm": 0.98046875,
"learning_rate": 0.000499792616381322,
"loss": 5.4809,
"mean_token_accuracy": 0.18552721440792083,
"num_tokens": 5523631.0,
"step": 2410
},
{
"entropy": 5.715830135345459,
"epoch": 0.23198847262247838,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997911421482914,
"loss": 5.5429,
"mean_token_accuracy": 0.1831541910767555,
"num_tokens": 5535637.0,
"step": 2415
},
{
"entropy": 5.736978912353516,
"epoch": 0.23246878001921228,
"grad_norm": 1.0859375,
"learning_rate": 0.000499789662696278,
"loss": 5.5562,
"mean_token_accuracy": 0.1847958281636238,
"num_tokens": 5546470.0,
"step": 2420
},
{
"entropy": 5.8512872695922855,
"epoch": 0.2329490874159462,
"grad_norm": 1.015625,
"learning_rate": 0.0004997881780253162,
"loss": 5.7116,
"mean_token_accuracy": 0.1715322643518448,
"num_tokens": 5558633.0,
"step": 2425
},
{
"entropy": 5.771648311614991,
"epoch": 0.2334293948126801,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997866881354403,
"loss": 5.6272,
"mean_token_accuracy": 0.17206788063049316,
"num_tokens": 5570427.0,
"step": 2430
},
{
"entropy": 5.735009336471558,
"epoch": 0.23390970220941404,
"grad_norm": 1.0234375,
"learning_rate": 0.000499785193026685,
"loss": 5.6008,
"mean_token_accuracy": 0.17571118026971816,
"num_tokens": 5580991.0,
"step": 2435
},
{
"entropy": 5.840723323822021,
"epoch": 0.23439000960614795,
"grad_norm": 1.0625,
"learning_rate": 0.0004997836926990851,
"loss": 5.6681,
"mean_token_accuracy": 0.17219943851232528,
"num_tokens": 5592777.0,
"step": 2440
},
{
"entropy": 5.714860486984253,
"epoch": 0.23487031700288186,
"grad_norm": 1.0625,
"learning_rate": 0.0004997821871526752,
"loss": 5.5575,
"mean_token_accuracy": 0.1838330253958702,
"num_tokens": 5603326.0,
"step": 2445
},
{
"entropy": 5.682803630828857,
"epoch": 0.23535062439961577,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997806763874905,
"loss": 5.542,
"mean_token_accuracy": 0.18075062334537506,
"num_tokens": 5614504.0,
"step": 2450
},
{
"entropy": 5.7689934253692625,
"epoch": 0.23583093179634967,
"grad_norm": 1.0,
"learning_rate": 0.0004997791604035659,
"loss": 5.6115,
"mean_token_accuracy": 0.18290183395147325,
"num_tokens": 5625150.0,
"step": 2455
},
{
"entropy": 5.732327699661255,
"epoch": 0.23631123919308358,
"grad_norm": 1.0625,
"learning_rate": 0.0004997776392009366,
"loss": 5.6133,
"mean_token_accuracy": 0.17373958230018616,
"num_tokens": 5636815.0,
"step": 2460
},
{
"entropy": 5.774140119552612,
"epoch": 0.2367915465898175,
"grad_norm": 0.96875,
"learning_rate": 0.0004997761127796381,
"loss": 5.6077,
"mean_token_accuracy": 0.17858083546161652,
"num_tokens": 5648272.0,
"step": 2465
},
{
"entropy": 5.742091226577759,
"epoch": 0.2372718539865514,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997745811397056,
"loss": 5.5252,
"mean_token_accuracy": 0.18127477020025254,
"num_tokens": 5659227.0,
"step": 2470
},
{
"entropy": 5.748807668685913,
"epoch": 0.2377521613832853,
"grad_norm": 1.09375,
"learning_rate": 0.0004997730442811748,
"loss": 5.6483,
"mean_token_accuracy": 0.17652334049344062,
"num_tokens": 5670411.0,
"step": 2475
},
{
"entropy": 5.703849220275879,
"epoch": 0.23823246878001922,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997715022040814,
"loss": 5.4874,
"mean_token_accuracy": 0.18095139116048814,
"num_tokens": 5681570.0,
"step": 2480
},
{
"entropy": 5.688530683517456,
"epoch": 0.23871277617675313,
"grad_norm": 0.9921875,
"learning_rate": 0.000499769954908461,
"loss": 5.4769,
"mean_token_accuracy": 0.18896123319864272,
"num_tokens": 5693021.0,
"step": 2485
},
{
"entropy": 5.6701000213623045,
"epoch": 0.23919308357348704,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997684023943498,
"loss": 5.565,
"mean_token_accuracy": 0.1783182293176651,
"num_tokens": 5704043.0,
"step": 2490
},
{
"entropy": 5.731194639205933,
"epoch": 0.23967339097022095,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997668446617837,
"loss": 5.6502,
"mean_token_accuracy": 0.17267276644706725,
"num_tokens": 5715735.0,
"step": 2495
},
{
"entropy": 5.862462711334229,
"epoch": 0.24015369836695485,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997652817107989,
"loss": 5.6205,
"mean_token_accuracy": 0.1759555384516716,
"num_tokens": 5725778.0,
"step": 2500
},
{
"entropy": 5.732063245773316,
"epoch": 0.24063400576368876,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997637135414315,
"loss": 5.6296,
"mean_token_accuracy": 0.17513870894908906,
"num_tokens": 5737224.0,
"step": 2505
},
{
"entropy": 5.8137345790863035,
"epoch": 0.24111431316042267,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997621401537183,
"loss": 5.6583,
"mean_token_accuracy": 0.17411693036556244,
"num_tokens": 5749226.0,
"step": 2510
},
{
"entropy": 5.8183518409729,
"epoch": 0.24159462055715658,
"grad_norm": 1.125,
"learning_rate": 0.0004997605615476955,
"loss": 5.6312,
"mean_token_accuracy": 0.17504109144210817,
"num_tokens": 5760282.0,
"step": 2515
},
{
"entropy": 5.663213491439819,
"epoch": 0.2420749279538905,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997589777234,
"loss": 5.5309,
"mean_token_accuracy": 0.1830235943198204,
"num_tokens": 5771756.0,
"step": 2520
},
{
"entropy": 5.753743457794189,
"epoch": 0.2425552353506244,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997573886808684,
"loss": 5.5671,
"mean_token_accuracy": 0.1693316325545311,
"num_tokens": 5783237.0,
"step": 2525
},
{
"entropy": 5.788538599014283,
"epoch": 0.2430355427473583,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997557944201375,
"loss": 5.6374,
"mean_token_accuracy": 0.17653653472661973,
"num_tokens": 5794825.0,
"step": 2530
},
{
"entropy": 5.7611926078796385,
"epoch": 0.24351585014409222,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997541949412445,
"loss": 5.5517,
"mean_token_accuracy": 0.1838692843914032,
"num_tokens": 5805578.0,
"step": 2535
},
{
"entropy": 5.798102807998657,
"epoch": 0.24399615754082613,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997525902442266,
"loss": 5.6398,
"mean_token_accuracy": 0.17110283374786378,
"num_tokens": 5818201.0,
"step": 2540
},
{
"entropy": 5.721447992324829,
"epoch": 0.24447646493756003,
"grad_norm": 1.0,
"learning_rate": 0.0004997509803291207,
"loss": 5.5639,
"mean_token_accuracy": 0.1799464926123619,
"num_tokens": 5830319.0,
"step": 2545
},
{
"entropy": 5.658524894714356,
"epoch": 0.24495677233429394,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997493651959647,
"loss": 5.5081,
"mean_token_accuracy": 0.18168970346450805,
"num_tokens": 5840638.0,
"step": 2550
},
{
"entropy": 5.725489377975464,
"epoch": 0.24543707973102785,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997477448447955,
"loss": 5.5401,
"mean_token_accuracy": 0.17641098499298097,
"num_tokens": 5852472.0,
"step": 2555
},
{
"entropy": 5.8019109725952145,
"epoch": 0.24591738712776176,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997461192756512,
"loss": 5.5784,
"mean_token_accuracy": 0.17450501769781113,
"num_tokens": 5863455.0,
"step": 2560
},
{
"entropy": 5.637508058547974,
"epoch": 0.24639769452449567,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997444884885694,
"loss": 5.4857,
"mean_token_accuracy": 0.18074321746826172,
"num_tokens": 5873141.0,
"step": 2565
},
{
"entropy": 5.688677740097046,
"epoch": 0.24687800192122958,
"grad_norm": 1.21875,
"learning_rate": 0.0004997428524835879,
"loss": 5.5911,
"mean_token_accuracy": 0.1751626744866371,
"num_tokens": 5884363.0,
"step": 2570
},
{
"entropy": 5.763658571243286,
"epoch": 0.2473583093179635,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997412112607446,
"loss": 5.6443,
"mean_token_accuracy": 0.1757771074771881,
"num_tokens": 5895856.0,
"step": 2575
},
{
"entropy": 5.676737451553345,
"epoch": 0.2478386167146974,
"grad_norm": 1.046875,
"learning_rate": 0.0004997395648200778,
"loss": 5.4743,
"mean_token_accuracy": 0.1850918784737587,
"num_tokens": 5906657.0,
"step": 2580
},
{
"entropy": 5.710775470733642,
"epoch": 0.2483189241114313,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997379131616257,
"loss": 5.5895,
"mean_token_accuracy": 0.17889538109302522,
"num_tokens": 5919496.0,
"step": 2585
},
{
"entropy": 5.750392150878906,
"epoch": 0.24879923150816521,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997362562854266,
"loss": 5.6576,
"mean_token_accuracy": 0.16503879725933074,
"num_tokens": 5932593.0,
"step": 2590
},
{
"entropy": 5.737170696258545,
"epoch": 0.24927953890489912,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997345941915187,
"loss": 5.5772,
"mean_token_accuracy": 0.17579198330640794,
"num_tokens": 5944080.0,
"step": 2595
},
{
"entropy": 5.748187398910522,
"epoch": 0.24975984630163303,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997329268799412,
"loss": 5.5536,
"mean_token_accuracy": 0.18597260415554046,
"num_tokens": 5955703.0,
"step": 2600
},
{
"entropy": 5.728941822052002,
"epoch": 0.25024015369836694,
"grad_norm": 1.03125,
"learning_rate": 0.0004997312543507322,
"loss": 5.6317,
"mean_token_accuracy": 0.17275256365537645,
"num_tokens": 5966979.0,
"step": 2605
},
{
"entropy": 5.767943668365478,
"epoch": 0.2507204610951009,
"grad_norm": 1.03125,
"learning_rate": 0.0004997295766039309,
"loss": 5.5183,
"mean_token_accuracy": 0.1809815689921379,
"num_tokens": 5978808.0,
"step": 2610
},
{
"entropy": 5.720742416381836,
"epoch": 0.25120076849183476,
"grad_norm": 1.0,
"learning_rate": 0.0004997278936395761,
"loss": 5.6934,
"mean_token_accuracy": 0.1722423776984215,
"num_tokens": 5992145.0,
"step": 2615
},
{
"entropy": 5.732787466049194,
"epoch": 0.2516810758885687,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997262054577071,
"loss": 5.5557,
"mean_token_accuracy": 0.17480167895555496,
"num_tokens": 6003723.0,
"step": 2620
},
{
"entropy": 5.763635778427124,
"epoch": 0.2521613832853026,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997245120583627,
"loss": 5.5786,
"mean_token_accuracy": 0.18482747673988342,
"num_tokens": 6014064.0,
"step": 2625
},
{
"entropy": 5.620661401748658,
"epoch": 0.2526416906820365,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997228134415825,
"loss": 5.4966,
"mean_token_accuracy": 0.18726035058498383,
"num_tokens": 6025455.0,
"step": 2630
},
{
"entropy": 5.755416059494019,
"epoch": 0.2531219980787704,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997211096074059,
"loss": 5.6073,
"mean_token_accuracy": 0.16819422841072082,
"num_tokens": 6037347.0,
"step": 2635
},
{
"entropy": 5.697003602981567,
"epoch": 0.25360230547550433,
"grad_norm": 1.03125,
"learning_rate": 0.0004997194005558722,
"loss": 5.4925,
"mean_token_accuracy": 0.18332696408033372,
"num_tokens": 6049236.0,
"step": 2640
},
{
"entropy": 5.591462659835815,
"epoch": 0.2540826128722382,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997176862870216,
"loss": 5.5117,
"mean_token_accuracy": 0.1839895710349083,
"num_tokens": 6060982.0,
"step": 2645
},
{
"entropy": 5.785092067718506,
"epoch": 0.25456292026897215,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997159668008933,
"loss": 5.5093,
"mean_token_accuracy": 0.183223856985569,
"num_tokens": 6070925.0,
"step": 2650
},
{
"entropy": 5.637601613998413,
"epoch": 0.25504322766570603,
"grad_norm": 1.03125,
"learning_rate": 0.0004997142420975277,
"loss": 5.4662,
"mean_token_accuracy": 0.18813761025667192,
"num_tokens": 6081279.0,
"step": 2655
},
{
"entropy": 5.6759350299835205,
"epoch": 0.25552353506243997,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997125121769647,
"loss": 5.5709,
"mean_token_accuracy": 0.18180250078439714,
"num_tokens": 6091797.0,
"step": 2660
},
{
"entropy": 5.803980493545533,
"epoch": 0.25600384245917385,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997107770392444,
"loss": 5.5878,
"mean_token_accuracy": 0.18288871198892592,
"num_tokens": 6103435.0,
"step": 2665
},
{
"entropy": 5.7139026641845705,
"epoch": 0.2564841498559078,
"grad_norm": 0.953125,
"learning_rate": 0.000499709036684407,
"loss": 5.6485,
"mean_token_accuracy": 0.1780319407582283,
"num_tokens": 6114531.0,
"step": 2670
},
{
"entropy": 5.635207653045654,
"epoch": 0.25696445725264166,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997072911124932,
"loss": 5.4849,
"mean_token_accuracy": 0.1845824345946312,
"num_tokens": 6126110.0,
"step": 2675
},
{
"entropy": 5.746867656707764,
"epoch": 0.2574447646493756,
"grad_norm": 1.078125,
"learning_rate": 0.0004997055403235432,
"loss": 5.5801,
"mean_token_accuracy": 0.17680004984140396,
"num_tokens": 6137114.0,
"step": 2680
},
{
"entropy": 5.682720375061035,
"epoch": 0.2579250720461095,
"grad_norm": 1.0,
"learning_rate": 0.0004997037843175978,
"loss": 5.5435,
"mean_token_accuracy": 0.1826479434967041,
"num_tokens": 6148696.0,
"step": 2685
},
{
"entropy": 5.668850469589233,
"epoch": 0.2584053794428434,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997020230946978,
"loss": 5.5377,
"mean_token_accuracy": 0.18099813014268876,
"num_tokens": 6160235.0,
"step": 2690
},
{
"entropy": 5.727720451354981,
"epoch": 0.25888568683957736,
"grad_norm": 1.03125,
"learning_rate": 0.0004997002566548841,
"loss": 5.5204,
"mean_token_accuracy": 0.18033260554075242,
"num_tokens": 6172031.0,
"step": 2695
},
{
"entropy": 5.558167171478272,
"epoch": 0.25936599423631124,
"grad_norm": 1.015625,
"learning_rate": 0.0004996984849981976,
"loss": 5.3993,
"mean_token_accuracy": 0.19067923128604888,
"num_tokens": 6183547.0,
"step": 2700
},
{
"entropy": 5.730989408493042,
"epoch": 0.2598463016330452,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996967081246794,
"loss": 5.5889,
"mean_token_accuracy": 0.17200068533420562,
"num_tokens": 6194768.0,
"step": 2705
},
{
"entropy": 5.720772361755371,
"epoch": 0.26032660902977905,
"grad_norm": 1.015625,
"learning_rate": 0.0004996949260343711,
"loss": 5.5941,
"mean_token_accuracy": 0.17187345325946807,
"num_tokens": 6206099.0,
"step": 2710
},
{
"entropy": 5.782607173919677,
"epoch": 0.260806916426513,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996931387273137,
"loss": 5.585,
"mean_token_accuracy": 0.18310263007879257,
"num_tokens": 6217530.0,
"step": 2715
},
{
"entropy": 5.787333059310913,
"epoch": 0.2612872238232469,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996913462035487,
"loss": 5.6054,
"mean_token_accuracy": 0.1762217938899994,
"num_tokens": 6228564.0,
"step": 2720
},
{
"entropy": 5.637451457977295,
"epoch": 0.2617675312199808,
"grad_norm": 1.0234375,
"learning_rate": 0.000499689548463118,
"loss": 5.4915,
"mean_token_accuracy": 0.17907891869544984,
"num_tokens": 6239945.0,
"step": 2725
},
{
"entropy": 5.713913059234619,
"epoch": 0.2622478386167147,
"grad_norm": 1.0,
"learning_rate": 0.0004996877455060631,
"loss": 5.6031,
"mean_token_accuracy": 0.17165847718715668,
"num_tokens": 6251829.0,
"step": 2730
},
{
"entropy": 5.83690390586853,
"epoch": 0.2627281460134486,
"grad_norm": 1.015625,
"learning_rate": 0.0004996859373324259,
"loss": 5.6997,
"mean_token_accuracy": 0.16278707683086396,
"num_tokens": 6264823.0,
"step": 2735
},
{
"entropy": 5.652474355697632,
"epoch": 0.2632084534101825,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996841239422485,
"loss": 5.3794,
"mean_token_accuracy": 0.18814935386180878,
"num_tokens": 6276247.0,
"step": 2740
},
{
"entropy": 5.562798309326172,
"epoch": 0.26368876080691644,
"grad_norm": 1.0,
"learning_rate": 0.0004996823053355729,
"loss": 5.4945,
"mean_token_accuracy": 0.18557627648115158,
"num_tokens": 6287593.0,
"step": 2745
},
{
"entropy": 5.779765367507935,
"epoch": 0.2641690682036503,
"grad_norm": 1.0,
"learning_rate": 0.0004996804815124413,
"loss": 5.6677,
"mean_token_accuracy": 0.17469124346971512,
"num_tokens": 6299918.0,
"step": 2750
},
{
"entropy": 5.652080011367798,
"epoch": 0.26464937560038426,
"grad_norm": 1.125,
"learning_rate": 0.0004996786524728962,
"loss": 5.5093,
"mean_token_accuracy": 0.18570335656404496,
"num_tokens": 6311147.0,
"step": 2755
},
{
"entropy": 5.57460036277771,
"epoch": 0.26512968299711814,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996768182169797,
"loss": 5.4136,
"mean_token_accuracy": 0.1906318724155426,
"num_tokens": 6323239.0,
"step": 2760
},
{
"entropy": 5.755684471130371,
"epoch": 0.2656099903938521,
"grad_norm": 1.015625,
"learning_rate": 0.0004996749787447349,
"loss": 5.5058,
"mean_token_accuracy": 0.1824209213256836,
"num_tokens": 6334625.0,
"step": 2765
},
{
"entropy": 5.676227712631226,
"epoch": 0.26609029779058596,
"grad_norm": 1.078125,
"learning_rate": 0.000499673134056204,
"loss": 5.5537,
"mean_token_accuracy": 0.17851219475269317,
"num_tokens": 6346068.0,
"step": 2770
},
{
"entropy": 5.677485752105713,
"epoch": 0.2665706051873199,
"grad_norm": 1.125,
"learning_rate": 0.0004996712841514303,
"loss": 5.5272,
"mean_token_accuracy": 0.17988724559545516,
"num_tokens": 6357097.0,
"step": 2775
},
{
"entropy": 5.752181959152222,
"epoch": 0.2670509125840538,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996694290304563,
"loss": 5.5887,
"mean_token_accuracy": 0.16941747516393663,
"num_tokens": 6367481.0,
"step": 2780
},
{
"entropy": 5.618309497833252,
"epoch": 0.2675312199807877,
"grad_norm": 1.0,
"learning_rate": 0.0004996675686933255,
"loss": 5.4942,
"mean_token_accuracy": 0.18622181862592696,
"num_tokens": 6378873.0,
"step": 2785
},
{
"entropy": 5.736754608154297,
"epoch": 0.2680115273775216,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996657031400807,
"loss": 5.5408,
"mean_token_accuracy": 0.17920784205198287,
"num_tokens": 6390651.0,
"step": 2790
},
{
"entropy": 5.613878059387207,
"epoch": 0.26849183477425553,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996638323707655,
"loss": 5.4195,
"mean_token_accuracy": 0.19438366293907167,
"num_tokens": 6401631.0,
"step": 2795
},
{
"entropy": 5.575555801391602,
"epoch": 0.2689721421709894,
"grad_norm": 1.0,
"learning_rate": 0.0004996619563854232,
"loss": 5.505,
"mean_token_accuracy": 0.18674360662698747,
"num_tokens": 6413875.0,
"step": 2800
},
{
"entropy": 5.771028232574463,
"epoch": 0.26945244956772335,
"grad_norm": 1.0625,
"learning_rate": 0.0004996600751840974,
"loss": 5.541,
"mean_token_accuracy": 0.17710736989974976,
"num_tokens": 6425764.0,
"step": 2805
},
{
"entropy": 5.621025276184082,
"epoch": 0.26993275696445723,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996581887668317,
"loss": 5.4427,
"mean_token_accuracy": 0.18712297081947327,
"num_tokens": 6437911.0,
"step": 2810
},
{
"entropy": 5.597506284713745,
"epoch": 0.27041306436119117,
"grad_norm": 1.125,
"learning_rate": 0.00049965629713367,
"loss": 5.4682,
"mean_token_accuracy": 0.18105038553476333,
"num_tokens": 6449942.0,
"step": 2815
},
{
"entropy": 5.707987642288208,
"epoch": 0.27089337175792505,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996544002846561,
"loss": 5.5698,
"mean_token_accuracy": 0.18570097982883454,
"num_tokens": 6461729.0,
"step": 2820
},
{
"entropy": 5.729322052001953,
"epoch": 0.271373679154659,
"grad_norm": 1.015625,
"learning_rate": 0.0004996524982198343,
"loss": 5.553,
"mean_token_accuracy": 0.18336791172623634,
"num_tokens": 6472046.0,
"step": 2825
},
{
"entropy": 5.643385314941407,
"epoch": 0.27185398655139287,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996505909392485,
"loss": 5.5346,
"mean_token_accuracy": 0.17891884595155716,
"num_tokens": 6483308.0,
"step": 2830
},
{
"entropy": 5.559644651412964,
"epoch": 0.2723342939481268,
"grad_norm": 1.03125,
"learning_rate": 0.0004996486784429429,
"loss": 5.4033,
"mean_token_accuracy": 0.18978026360273362,
"num_tokens": 6495093.0,
"step": 2835
},
{
"entropy": 5.695954370498657,
"epoch": 0.2728146013448607,
"grad_norm": 1.1484375,
"learning_rate": 0.0004996467607309622,
"loss": 5.4873,
"mean_token_accuracy": 0.18494741916656493,
"num_tokens": 6505933.0,
"step": 2840
},
{
"entropy": 5.6591239929199215,
"epoch": 0.2732949087415946,
"grad_norm": 1.0625,
"learning_rate": 0.0004996448378033507,
"loss": 5.5485,
"mean_token_accuracy": 0.17702232003211976,
"num_tokens": 6517280.0,
"step": 2845
},
{
"entropy": 5.6757430076599125,
"epoch": 0.2737752161383285,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996429096601532,
"loss": 5.5964,
"mean_token_accuracy": 0.17370087206363677,
"num_tokens": 6528980.0,
"step": 2850
},
{
"entropy": 5.688251590728759,
"epoch": 0.27425552353506244,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996409763014144,
"loss": 5.5803,
"mean_token_accuracy": 0.18222061842679976,
"num_tokens": 6540670.0,
"step": 2855
},
{
"entropy": 5.640368938446045,
"epoch": 0.2747358309317964,
"grad_norm": 1.046875,
"learning_rate": 0.0004996390377271791,
"loss": 5.545,
"mean_token_accuracy": 0.1811893939971924,
"num_tokens": 6551302.0,
"step": 2860
},
{
"entropy": 5.676407194137573,
"epoch": 0.27521613832853026,
"grad_norm": 1.078125,
"learning_rate": 0.0004996370939374924,
"loss": 5.4962,
"mean_token_accuracy": 0.17508159577846527,
"num_tokens": 6563177.0,
"step": 2865
},
{
"entropy": 5.826377010345459,
"epoch": 0.2756964457252642,
"grad_norm": 1.2265625,
"learning_rate": 0.0004996351449323994,
"loss": 5.6048,
"mean_token_accuracy": 0.17976640462875365,
"num_tokens": 6573323.0,
"step": 2870
},
{
"entropy": 5.664675855636597,
"epoch": 0.2761767531219981,
"grad_norm": 1.03125,
"learning_rate": 0.0004996331907119455,
"loss": 5.5386,
"mean_token_accuracy": 0.17416994124650956,
"num_tokens": 6585382.0,
"step": 2875
},
{
"entropy": 5.553443956375122,
"epoch": 0.276657060518732,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996312312761758,
"loss": 5.4332,
"mean_token_accuracy": 0.1913624182343483,
"num_tokens": 6596629.0,
"step": 2880
},
{
"entropy": 5.703445148468018,
"epoch": 0.2771373679154659,
"grad_norm": 1.0625,
"learning_rate": 0.000499629266625136,
"loss": 5.5392,
"mean_token_accuracy": 0.1806390777230263,
"num_tokens": 6608408.0,
"step": 2885
},
{
"entropy": 5.732683229446411,
"epoch": 0.27761767531219983,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996272967588715,
"loss": 5.549,
"mean_token_accuracy": 0.1764448568224907,
"num_tokens": 6619375.0,
"step": 2890
},
{
"entropy": 5.728315591812134,
"epoch": 0.2780979827089337,
"grad_norm": 1.09375,
"learning_rate": 0.0004996253216774283,
"loss": 5.6042,
"mean_token_accuracy": 0.17850979417562485,
"num_tokens": 6631317.0,
"step": 2895
},
{
"entropy": 5.662656164169311,
"epoch": 0.27857829010566765,
"grad_norm": 1.15625,
"learning_rate": 0.0004996233413808521,
"loss": 5.4637,
"mean_token_accuracy": 0.18195123523473739,
"num_tokens": 6642009.0,
"step": 2900
},
{
"entropy": 5.683246278762818,
"epoch": 0.27905859750240153,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996213558691889,
"loss": 5.6133,
"mean_token_accuracy": 0.172495736181736,
"num_tokens": 6654713.0,
"step": 2905
},
{
"entropy": 5.6839197158813475,
"epoch": 0.27953890489913547,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996193651424848,
"loss": 5.5781,
"mean_token_accuracy": 0.1797851651906967,
"num_tokens": 6667157.0,
"step": 2910
},
{
"entropy": 5.665137672424317,
"epoch": 0.28001921229586935,
"grad_norm": 0.9140625,
"learning_rate": 0.000499617369200786,
"loss": 5.5158,
"mean_token_accuracy": 0.1891197681427002,
"num_tokens": 6679573.0,
"step": 2915
},
{
"entropy": 5.741504096984864,
"epoch": 0.2804995196926033,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996153680441389,
"loss": 5.5958,
"mean_token_accuracy": 0.17464567720890045,
"num_tokens": 6691768.0,
"step": 2920
},
{
"entropy": 5.721974849700928,
"epoch": 0.28097982708933716,
"grad_norm": 1.046875,
"learning_rate": 0.00049961336167259,
"loss": 5.556,
"mean_token_accuracy": 0.17392836958169938,
"num_tokens": 6701964.0,
"step": 2925
},
{
"entropy": 5.678666019439698,
"epoch": 0.2814601344860711,
"grad_norm": 1.046875,
"learning_rate": 0.0004996113500861857,
"loss": 5.5394,
"mean_token_accuracy": 0.1783669501543045,
"num_tokens": 6713506.0,
"step": 2930
},
{
"entropy": 5.680933046340942,
"epoch": 0.281940441882805,
"grad_norm": 1.09375,
"learning_rate": 0.0004996093332849729,
"loss": 5.5647,
"mean_token_accuracy": 0.1761838361620903,
"num_tokens": 6724616.0,
"step": 2935
},
{
"entropy": 5.70354528427124,
"epoch": 0.2824207492795389,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996073112689983,
"loss": 5.5319,
"mean_token_accuracy": 0.18018208295106888,
"num_tokens": 6735054.0,
"step": 2940
},
{
"entropy": 5.704903030395508,
"epoch": 0.2829010566762728,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996052840383088,
"loss": 5.589,
"mean_token_accuracy": 0.18162762224674225,
"num_tokens": 6746756.0,
"step": 2945
},
{
"entropy": 5.67103385925293,
"epoch": 0.28338136407300674,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996032515929516,
"loss": 5.454,
"mean_token_accuracy": 0.18400953710079193,
"num_tokens": 6759566.0,
"step": 2950
},
{
"entropy": 5.649921560287476,
"epoch": 0.2838616714697406,
"grad_norm": 1.015625,
"learning_rate": 0.0004996012139329738,
"loss": 5.4738,
"mean_token_accuracy": 0.18114203959703445,
"num_tokens": 6771375.0,
"step": 2955
},
{
"entropy": 5.739535665512085,
"epoch": 0.28434197886647455,
"grad_norm": 1.09375,
"learning_rate": 0.0004995991710584228,
"loss": 5.5935,
"mean_token_accuracy": 0.1726598024368286,
"num_tokens": 6783252.0,
"step": 2960
},
{
"entropy": 5.614906644821167,
"epoch": 0.28482228626320844,
"grad_norm": 1.0,
"learning_rate": 0.0004995971229693459,
"loss": 5.5751,
"mean_token_accuracy": 0.17417452335357667,
"num_tokens": 6795525.0,
"step": 2965
},
{
"entropy": 5.7367006778717045,
"epoch": 0.28530259365994237,
"grad_norm": 0.9375,
"learning_rate": 0.0004995950696657909,
"loss": 5.4922,
"mean_token_accuracy": 0.18741291463375093,
"num_tokens": 6807212.0,
"step": 2970
},
{
"entropy": 5.595437860488891,
"epoch": 0.28578290105667625,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995930111478051,
"loss": 5.4442,
"mean_token_accuracy": 0.1817941963672638,
"num_tokens": 6819367.0,
"step": 2975
},
{
"entropy": 5.637096786499024,
"epoch": 0.2862632084534102,
"grad_norm": 1.03125,
"learning_rate": 0.0004995909474154365,
"loss": 5.5174,
"mean_token_accuracy": 0.18265894204378127,
"num_tokens": 6830405.0,
"step": 2980
},
{
"entropy": 5.62329068183899,
"epoch": 0.28674351585014407,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995888784687331,
"loss": 5.4952,
"mean_token_accuracy": 0.1821579173207283,
"num_tokens": 6841479.0,
"step": 2985
},
{
"entropy": 5.622311162948608,
"epoch": 0.287223823246878,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995868043077428,
"loss": 5.5304,
"mean_token_accuracy": 0.1803388327360153,
"num_tokens": 6851585.0,
"step": 2990
},
{
"entropy": 5.658223676681518,
"epoch": 0.2877041306436119,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995847249325137,
"loss": 5.5059,
"mean_token_accuracy": 0.1853910952806473,
"num_tokens": 6863176.0,
"step": 2995
},
{
"entropy": 5.698240089416504,
"epoch": 0.2881844380403458,
"grad_norm": 1.109375,
"learning_rate": 0.0004995826403430942,
"loss": 5.5613,
"mean_token_accuracy": 0.18125778138637544,
"num_tokens": 6874021.0,
"step": 3000
},
{
"epoch": 0.2881844380403458,
"eval_entropy": 5.4392175834856635,
"eval_loss": 5.524839878082275,
"eval_mean_token_accuracy": 0.1890813923326899,
"eval_num_tokens": 6874021.0,
"eval_runtime": 20.7338,
"eval_samples_per_second": 1582.678,
"eval_steps_per_second": 197.841,
"step": 3000
},
{
"entropy": 5.700993347167969,
"epoch": 0.2886647454370797,
"grad_norm": 1.03125,
"learning_rate": 0.0004995805505395328,
"loss": 5.5182,
"mean_token_accuracy": 0.17984089851379395,
"num_tokens": 6884999.0,
"step": 3005
},
{
"entropy": 5.678546047210693,
"epoch": 0.28914505283381364,
"grad_norm": 1.1875,
"learning_rate": 0.0004995784555218778,
"loss": 5.51,
"mean_token_accuracy": 0.18251369148492813,
"num_tokens": 6897021.0,
"step": 3010
},
{
"entropy": 5.577228927612305,
"epoch": 0.2896253602305475,
"grad_norm": 1.0625,
"learning_rate": 0.0004995763552901779,
"loss": 5.5079,
"mean_token_accuracy": 0.1816476598381996,
"num_tokens": 6908320.0,
"step": 3015
},
{
"entropy": 5.693873882293701,
"epoch": 0.29010566762728146,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995742498444818,
"loss": 5.4779,
"mean_token_accuracy": 0.18809738755226135,
"num_tokens": 6919957.0,
"step": 3020
},
{
"entropy": 5.694812393188476,
"epoch": 0.2905859750240154,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995721391848387,
"loss": 5.4474,
"mean_token_accuracy": 0.1813945174217224,
"num_tokens": 6930531.0,
"step": 3025
},
{
"entropy": 5.677816247940063,
"epoch": 0.2910662824207493,
"grad_norm": 1.03125,
"learning_rate": 0.0004995700233112972,
"loss": 5.5802,
"mean_token_accuracy": 0.17819592505693435,
"num_tokens": 6942556.0,
"step": 3030
},
{
"entropy": 5.662592124938965,
"epoch": 0.2915465898174832,
"grad_norm": 0.94140625,
"learning_rate": 0.0004995679022239066,
"loss": 5.5308,
"mean_token_accuracy": 0.18774571269750595,
"num_tokens": 6954410.0,
"step": 3035
},
{
"entropy": 5.652567958831787,
"epoch": 0.2920268972142171,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995657759227162,
"loss": 5.5406,
"mean_token_accuracy": 0.17681581825017928,
"num_tokens": 6964970.0,
"step": 3040
},
{
"entropy": 5.637769889831543,
"epoch": 0.29250720461095103,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995636444077751,
"loss": 5.4309,
"mean_token_accuracy": 0.1918710544705391,
"num_tokens": 6976016.0,
"step": 3045
},
{
"entropy": 5.602601861953735,
"epoch": 0.2929875120076849,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995615076791333,
"loss": 5.4599,
"mean_token_accuracy": 0.1865808606147766,
"num_tokens": 6987199.0,
"step": 3050
},
{
"entropy": 5.6089927673339846,
"epoch": 0.29346781940441885,
"grad_norm": 1.0,
"learning_rate": 0.0004995593657368399,
"loss": 5.4809,
"mean_token_accuracy": 0.19195563346147537,
"num_tokens": 6999174.0,
"step": 3055
},
{
"entropy": 5.653221988677979,
"epoch": 0.29394812680115273,
"grad_norm": 0.9765625,
"learning_rate": 0.000499557218580945,
"loss": 5.5526,
"mean_token_accuracy": 0.1775237575173378,
"num_tokens": 7012148.0,
"step": 3060
},
{
"entropy": 5.551672124862671,
"epoch": 0.29442843419788667,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995550662114981,
"loss": 5.4211,
"mean_token_accuracy": 0.1860169753432274,
"num_tokens": 7023238.0,
"step": 3065
},
{
"entropy": 5.6040300846099855,
"epoch": 0.29490874159462055,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995529086285495,
"loss": 5.5052,
"mean_token_accuracy": 0.18213031440973282,
"num_tokens": 7034944.0,
"step": 3070
},
{
"entropy": 5.660197782516479,
"epoch": 0.2953890489913545,
"grad_norm": 0.9921875,
"learning_rate": 0.000499550745832149,
"loss": 5.3774,
"mean_token_accuracy": 0.1897702306509018,
"num_tokens": 7046880.0,
"step": 3075
},
{
"entropy": 5.606032466888427,
"epoch": 0.29586935638808837,
"grad_norm": 1.046875,
"learning_rate": 0.0004995485778223471,
"loss": 5.4575,
"mean_token_accuracy": 0.18246425986289977,
"num_tokens": 7057678.0,
"step": 3080
},
{
"entropy": 5.530951976776123,
"epoch": 0.2963496637848223,
"grad_norm": 1.078125,
"learning_rate": 0.0004995464045991939,
"loss": 5.4289,
"mean_token_accuracy": 0.18984924107789994,
"num_tokens": 7068336.0,
"step": 3085
},
{
"entropy": 5.612528896331787,
"epoch": 0.2968299711815562,
"grad_norm": 1.0,
"learning_rate": 0.00049954422616274,
"loss": 5.5012,
"mean_token_accuracy": 0.1788366675376892,
"num_tokens": 7080341.0,
"step": 3090
},
{
"entropy": 5.7698705196380615,
"epoch": 0.2973102785782901,
"grad_norm": 1.1796875,
"learning_rate": 0.0004995420425130359,
"loss": 5.6488,
"mean_token_accuracy": 0.17575520128011704,
"num_tokens": 7090618.0,
"step": 3095
},
{
"entropy": 5.592949104309082,
"epoch": 0.297790585975024,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995398536501324,
"loss": 5.3945,
"mean_token_accuracy": 0.19016729593276976,
"num_tokens": 7101843.0,
"step": 3100
},
{
"entropy": 5.547535753250122,
"epoch": 0.29827089337175794,
"grad_norm": 1.1484375,
"learning_rate": 0.0004995376595740801,
"loss": 5.467,
"mean_token_accuracy": 0.18589547425508499,
"num_tokens": 7112014.0,
"step": 3105
},
{
"entropy": 5.645080947875977,
"epoch": 0.2987512007684918,
"grad_norm": 1.09375,
"learning_rate": 0.0004995354602849302,
"loss": 5.5408,
"mean_token_accuracy": 0.17661455273628235,
"num_tokens": 7123860.0,
"step": 3110
},
{
"entropy": 5.68672022819519,
"epoch": 0.29923150816522576,
"grad_norm": 1.078125,
"learning_rate": 0.0004995332557827337,
"loss": 5.517,
"mean_token_accuracy": 0.1843189463019371,
"num_tokens": 7135901.0,
"step": 3115
},
{
"entropy": 5.60950345993042,
"epoch": 0.29971181556195964,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995310460675416,
"loss": 5.4764,
"mean_token_accuracy": 0.19353332966566086,
"num_tokens": 7148743.0,
"step": 3120
},
{
"entropy": 5.700359869003296,
"epoch": 0.3001921229586936,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995288311394053,
"loss": 5.522,
"mean_token_accuracy": 0.1854911118745804,
"num_tokens": 7160731.0,
"step": 3125
},
{
"entropy": 5.662644529342652,
"epoch": 0.30067243035542746,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995266109983764,
"loss": 5.5154,
"mean_token_accuracy": 0.18299192190170288,
"num_tokens": 7172861.0,
"step": 3130
},
{
"entropy": 5.613388013839722,
"epoch": 0.3011527377521614,
"grad_norm": 1.015625,
"learning_rate": 0.0004995243856445062,
"loss": 5.4631,
"mean_token_accuracy": 0.180388443171978,
"num_tokens": 7183954.0,
"step": 3135
},
{
"entropy": 5.5486509799957275,
"epoch": 0.3016330451488953,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995221550778466,
"loss": 5.4437,
"mean_token_accuracy": 0.18174690902233123,
"num_tokens": 7195466.0,
"step": 3140
},
{
"entropy": 5.612729692459107,
"epoch": 0.3021133525456292,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995199192984491,
"loss": 5.422,
"mean_token_accuracy": 0.19040560573339463,
"num_tokens": 7207173.0,
"step": 3145
},
{
"entropy": 5.657872819900513,
"epoch": 0.3025936599423631,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995176783063657,
"loss": 5.5693,
"mean_token_accuracy": 0.18644312173128127,
"num_tokens": 7220095.0,
"step": 3150
},
{
"entropy": 5.569757556915283,
"epoch": 0.30307396733909703,
"grad_norm": 1.0625,
"learning_rate": 0.0004995154321016487,
"loss": 5.4857,
"mean_token_accuracy": 0.1882634222507477,
"num_tokens": 7230664.0,
"step": 3155
},
{
"entropy": 5.642200469970703,
"epoch": 0.3035542747358309,
"grad_norm": 1.03125,
"learning_rate": 0.0004995131806843499,
"loss": 5.4315,
"mean_token_accuracy": 0.18630266934633255,
"num_tokens": 7241278.0,
"step": 3160
},
{
"entropy": 5.545270013809204,
"epoch": 0.30403458213256485,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995109240545218,
"loss": 5.5928,
"mean_token_accuracy": 0.1777158111333847,
"num_tokens": 7252999.0,
"step": 3165
},
{
"entropy": 5.653474807739258,
"epoch": 0.3045148895292987,
"grad_norm": 1.09375,
"learning_rate": 0.0004995086622122167,
"loss": 5.4169,
"mean_token_accuracy": 0.1864437311887741,
"num_tokens": 7263949.0,
"step": 3170
},
{
"entropy": 5.639758348464966,
"epoch": 0.30499519692603266,
"grad_norm": 0.984375,
"learning_rate": 0.0004995063951574871,
"loss": 5.4714,
"mean_token_accuracy": 0.18542171269655228,
"num_tokens": 7275467.0,
"step": 3175
},
{
"entropy": 5.585669755935669,
"epoch": 0.30547550432276654,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995041228903856,
"loss": 5.4421,
"mean_token_accuracy": 0.19193329364061357,
"num_tokens": 7285534.0,
"step": 3180
},
{
"entropy": 5.6947675228118895,
"epoch": 0.3059558117195005,
"grad_norm": 1.0,
"learning_rate": 0.000499501845410965,
"loss": 5.5569,
"mean_token_accuracy": 0.1815490484237671,
"num_tokens": 7297252.0,
"step": 3185
},
{
"entropy": 5.607858324050904,
"epoch": 0.30643611911623436,
"grad_norm": 0.9375,
"learning_rate": 0.0004994995627192781,
"loss": 5.4299,
"mean_token_accuracy": 0.19153623431921005,
"num_tokens": 7308492.0,
"step": 3190
},
{
"entropy": 5.569702529907227,
"epoch": 0.3069164265129683,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994972748153781,
"loss": 5.4581,
"mean_token_accuracy": 0.18491947501897812,
"num_tokens": 7319703.0,
"step": 3195
},
{
"entropy": 5.674600601196289,
"epoch": 0.30739673390970224,
"grad_norm": 1.078125,
"learning_rate": 0.000499494981699318,
"loss": 5.4274,
"mean_token_accuracy": 0.1937693625688553,
"num_tokens": 7331022.0,
"step": 3200
},
{
"entropy": 5.56959900856018,
"epoch": 0.3078770413064361,
"grad_norm": 0.984375,
"learning_rate": 0.000499492683371151,
"loss": 5.4772,
"mean_token_accuracy": 0.18553764522075653,
"num_tokens": 7342977.0,
"step": 3205
},
{
"entropy": 5.66892671585083,
"epoch": 0.30835734870317005,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994903798309306,
"loss": 5.4562,
"mean_token_accuracy": 0.18036793321371078,
"num_tokens": 7353227.0,
"step": 3210
},
{
"entropy": 5.670237398147583,
"epoch": 0.30883765609990393,
"grad_norm": 1.078125,
"learning_rate": 0.0004994880710787102,
"loss": 5.5121,
"mean_token_accuracy": 0.1748569056391716,
"num_tokens": 7364165.0,
"step": 3215
},
{
"entropy": 5.555734395980835,
"epoch": 0.30931796349663787,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994857571145432,
"loss": 5.4433,
"mean_token_accuracy": 0.18816640377044677,
"num_tokens": 7374800.0,
"step": 3220
},
{
"entropy": 5.496872520446777,
"epoch": 0.30979827089337175,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994834379384837,
"loss": 5.3932,
"mean_token_accuracy": 0.1932916134595871,
"num_tokens": 7386360.0,
"step": 3225
},
{
"entropy": 5.606559085845947,
"epoch": 0.3102785782901057,
"grad_norm": 0.9921875,
"learning_rate": 0.0004994811135505851,
"loss": 5.4317,
"mean_token_accuracy": 0.18842349499464034,
"num_tokens": 7397066.0,
"step": 3230
},
{
"entropy": 5.683773517608643,
"epoch": 0.31075888568683957,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994787839509018,
"loss": 5.5188,
"mean_token_accuracy": 0.1811457246541977,
"num_tokens": 7408349.0,
"step": 3235
},
{
"entropy": 5.629186201095581,
"epoch": 0.3112391930835735,
"grad_norm": 1.078125,
"learning_rate": 0.0004994764491394876,
"loss": 5.5533,
"mean_token_accuracy": 0.17683551460504532,
"num_tokens": 7420343.0,
"step": 3240
},
{
"entropy": 5.65942530632019,
"epoch": 0.3117195004803074,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994741091163969,
"loss": 5.4622,
"mean_token_accuracy": 0.18691066205501555,
"num_tokens": 7431683.0,
"step": 3245
},
{
"entropy": 5.466778755187988,
"epoch": 0.3121998078770413,
"grad_norm": 1.03125,
"learning_rate": 0.000499471763881684,
"loss": 5.3551,
"mean_token_accuracy": 0.1865973576903343,
"num_tokens": 7443327.0,
"step": 3250
},
{
"entropy": 5.634066247940064,
"epoch": 0.3126801152737752,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994694134354031,
"loss": 5.4648,
"mean_token_accuracy": 0.18632208704948425,
"num_tokens": 7454002.0,
"step": 3255
},
{
"entropy": 5.543521547317505,
"epoch": 0.31316042267050914,
"grad_norm": 1.0859375,
"learning_rate": 0.000499467057777609,
"loss": 5.4556,
"mean_token_accuracy": 0.18688408136367798,
"num_tokens": 7464074.0,
"step": 3260
},
{
"entropy": 5.55138783454895,
"epoch": 0.313640730067243,
"grad_norm": 1.078125,
"learning_rate": 0.0004994646969083565,
"loss": 5.3983,
"mean_token_accuracy": 0.19326114803552627,
"num_tokens": 7475543.0,
"step": 3265
},
{
"entropy": 5.635407209396362,
"epoch": 0.31412103746397696,
"grad_norm": 1.046875,
"learning_rate": 0.0004994623308277002,
"loss": 5.4507,
"mean_token_accuracy": 0.18384750932455063,
"num_tokens": 7486818.0,
"step": 3270
},
{
"entropy": 5.62159948348999,
"epoch": 0.31460134486071084,
"grad_norm": 1.0078125,
"learning_rate": 0.000499459959535695,
"loss": 5.4938,
"mean_token_accuracy": 0.18008041083812715,
"num_tokens": 7499046.0,
"step": 3275
},
{
"entropy": 5.600229930877686,
"epoch": 0.3150816522574448,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994575830323962,
"loss": 5.4287,
"mean_token_accuracy": 0.18267867863178253,
"num_tokens": 7509853.0,
"step": 3280
},
{
"entropy": 5.477385759353638,
"epoch": 0.31556195965417866,
"grad_norm": 1.015625,
"learning_rate": 0.0004994552013178586,
"loss": 5.2819,
"mean_token_accuracy": 0.1905076265335083,
"num_tokens": 7521091.0,
"step": 3285
},
{
"entropy": 5.560299873352051,
"epoch": 0.3160422670509126,
"grad_norm": 1.078125,
"learning_rate": 0.000499452814392138,
"loss": 5.4164,
"mean_token_accuracy": 0.1956005111336708,
"num_tokens": 7531317.0,
"step": 3290
},
{
"entropy": 5.59887285232544,
"epoch": 0.3165225744476465,
"grad_norm": 0.96875,
"learning_rate": 0.0004994504222552894,
"loss": 5.5365,
"mean_token_accuracy": 0.18225086703896523,
"num_tokens": 7542822.0,
"step": 3295
},
{
"entropy": 5.675664710998535,
"epoch": 0.3170028818443804,
"grad_norm": 1.0390625,
"learning_rate": 0.0004994480249073684,
"loss": 5.4889,
"mean_token_accuracy": 0.17975055873394014,
"num_tokens": 7552434.0,
"step": 3300
},
{
"entropy": 5.594845724105835,
"epoch": 0.3174831892411143,
"grad_norm": 0.96875,
"learning_rate": 0.0004994456223484308,
"loss": 5.3642,
"mean_token_accuracy": 0.19091499149799346,
"num_tokens": 7563895.0,
"step": 3305
},
{
"entropy": 5.409893989562988,
"epoch": 0.31796349663784823,
"grad_norm": 1.0,
"learning_rate": 0.0004994432145785323,
"loss": 5.3848,
"mean_token_accuracy": 0.19512586593627929,
"num_tokens": 7575391.0,
"step": 3310
},
{
"entropy": 5.639042043685913,
"epoch": 0.3184438040345821,
"grad_norm": 1.0625,
"learning_rate": 0.0004994408015977288,
"loss": 5.5321,
"mean_token_accuracy": 0.19067961126565933,
"num_tokens": 7587119.0,
"step": 3315
},
{
"entropy": 5.650514125823975,
"epoch": 0.31892411143131605,
"grad_norm": 1.078125,
"learning_rate": 0.0004994383834060764,
"loss": 5.5053,
"mean_token_accuracy": 0.18139485269784927,
"num_tokens": 7598615.0,
"step": 3320
},
{
"entropy": 5.662230205535889,
"epoch": 0.31940441882804993,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994359600036311,
"loss": 5.4847,
"mean_token_accuracy": 0.19163189232349395,
"num_tokens": 7610159.0,
"step": 3325
},
{
"entropy": 5.627119827270508,
"epoch": 0.31988472622478387,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994335313904493,
"loss": 5.451,
"mean_token_accuracy": 0.1873135432600975,
"num_tokens": 7620922.0,
"step": 3330
},
{
"entropy": 5.546042776107788,
"epoch": 0.32036503362151775,
"grad_norm": 0.90625,
"learning_rate": 0.0004994310975665873,
"loss": 5.3591,
"mean_token_accuracy": 0.19363973736763002,
"num_tokens": 7632343.0,
"step": 3335
},
{
"entropy": 5.685207319259644,
"epoch": 0.3208453410182517,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994286585321017,
"loss": 5.568,
"mean_token_accuracy": 0.17316349148750304,
"num_tokens": 7644748.0,
"step": 3340
},
{
"entropy": 5.689049005508423,
"epoch": 0.32132564841498557,
"grad_norm": 1.1015625,
"learning_rate": 0.000499426214287049,
"loss": 5.5029,
"mean_token_accuracy": 0.19029539227485656,
"num_tokens": 7655449.0,
"step": 3345
},
{
"entropy": 5.639818286895752,
"epoch": 0.3218059558117195,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994237648314862,
"loss": 5.4696,
"mean_token_accuracy": 0.18956416994333267,
"num_tokens": 7665623.0,
"step": 3350
},
{
"entropy": 5.548650217056275,
"epoch": 0.3222862632084534,
"grad_norm": 1.0,
"learning_rate": 0.0004994213101654697,
"loss": 5.3623,
"mean_token_accuracy": 0.19147212505340577,
"num_tokens": 7676860.0,
"step": 3355
},
{
"entropy": 5.647919225692749,
"epoch": 0.3227665706051873,
"grad_norm": 1.0,
"learning_rate": 0.000499418850289057,
"loss": 5.5559,
"mean_token_accuracy": 0.18325860053300858,
"num_tokens": 7687778.0,
"step": 3360
},
{
"entropy": 5.633452653884888,
"epoch": 0.32324687800192126,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994163852023048,
"loss": 5.4644,
"mean_token_accuracy": 0.18514435291290282,
"num_tokens": 7699154.0,
"step": 3365
},
{
"entropy": 5.59450101852417,
"epoch": 0.32372718539865514,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994139149052706,
"loss": 5.4779,
"mean_token_accuracy": 0.18767904341220856,
"num_tokens": 7711010.0,
"step": 3370
},
{
"entropy": 5.627596235275268,
"epoch": 0.3242074927953891,
"grad_norm": 1.03125,
"learning_rate": 0.0004994114393980117,
"loss": 5.4976,
"mean_token_accuracy": 0.18083676546812058,
"num_tokens": 7721969.0,
"step": 3375
},
{
"entropy": 5.615005588531494,
"epoch": 0.32468780019212296,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994089586805856,
"loss": 5.4211,
"mean_token_accuracy": 0.18771904110908508,
"num_tokens": 7733762.0,
"step": 3380
},
{
"entropy": 5.603797101974488,
"epoch": 0.3251681075888569,
"grad_norm": 1.046875,
"learning_rate": 0.0004994064727530496,
"loss": 5.4356,
"mean_token_accuracy": 0.1839058965444565,
"num_tokens": 7744614.0,
"step": 3385
},
{
"entropy": 5.560338926315308,
"epoch": 0.3256484149855908,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994039816154618,
"loss": 5.3845,
"mean_token_accuracy": 0.1955147460103035,
"num_tokens": 7755799.0,
"step": 3390
},
{
"entropy": 5.513249158859253,
"epoch": 0.3261287223823247,
"grad_norm": 0.9921875,
"learning_rate": 0.00049940148526788,
"loss": 5.4396,
"mean_token_accuracy": 0.1901847615838051,
"num_tokens": 7768140.0,
"step": 3395
},
{
"entropy": 5.557335901260376,
"epoch": 0.3266090297790586,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993989837103618,
"loss": 5.4366,
"mean_token_accuracy": 0.19033351689577102,
"num_tokens": 7778494.0,
"step": 3400
},
{
"entropy": 5.683393812179565,
"epoch": 0.3270893371757925,
"grad_norm": 1.078125,
"learning_rate": 0.0004993964769429657,
"loss": 5.5064,
"mean_token_accuracy": 0.1925898015499115,
"num_tokens": 7789234.0,
"step": 3405
},
{
"entropy": 5.557097578048706,
"epoch": 0.3275696445725264,
"grad_norm": 0.921875,
"learning_rate": 0.0004993939649657498,
"loss": 5.4937,
"mean_token_accuracy": 0.18996243476867675,
"num_tokens": 7800602.0,
"step": 3410
},
{
"entropy": 5.510222053527832,
"epoch": 0.32804995196926034,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993914477787721,
"loss": 5.3159,
"mean_token_accuracy": 0.19364566057920457,
"num_tokens": 7812803.0,
"step": 3415
},
{
"entropy": 5.749662780761719,
"epoch": 0.3285302593659942,
"grad_norm": 1.015625,
"learning_rate": 0.0004993889253820915,
"loss": 5.614,
"mean_token_accuracy": 0.17690031826496125,
"num_tokens": 7825432.0,
"step": 3420
},
{
"entropy": 5.600550270080566,
"epoch": 0.32901056676272816,
"grad_norm": 1.09375,
"learning_rate": 0.0004993863977757663,
"loss": 5.4376,
"mean_token_accuracy": 0.18812805712223052,
"num_tokens": 7837258.0,
"step": 3425
},
{
"entropy": 5.522640132904053,
"epoch": 0.32949087415946204,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993838649598552,
"loss": 5.3049,
"mean_token_accuracy": 0.1966621220111847,
"num_tokens": 7847573.0,
"step": 3430
},
{
"entropy": 5.63411316871643,
"epoch": 0.329971181556196,
"grad_norm": 1.0859375,
"learning_rate": 0.0004993813269344171,
"loss": 5.4645,
"mean_token_accuracy": 0.18168777078390122,
"num_tokens": 7857957.0,
"step": 3435
},
{
"entropy": 5.580896186828613,
"epoch": 0.33045148895292986,
"grad_norm": 1.0234375,
"learning_rate": 0.0004993787836995108,
"loss": 5.3706,
"mean_token_accuracy": 0.1964847505092621,
"num_tokens": 7867996.0,
"step": 3440
},
{
"entropy": 5.501073503494263,
"epoch": 0.3309317963496638,
"grad_norm": 1.046875,
"learning_rate": 0.0004993762352551954,
"loss": 5.4281,
"mean_token_accuracy": 0.1835190922021866,
"num_tokens": 7879245.0,
"step": 3445
},
{
"entropy": 5.646635055541992,
"epoch": 0.3314121037463977,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993736816015301,
"loss": 5.5,
"mean_token_accuracy": 0.17995858192443848,
"num_tokens": 7891186.0,
"step": 3450
},
{
"entropy": 5.6572953224182125,
"epoch": 0.3318924111431316,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993711227385742,
"loss": 5.5248,
"mean_token_accuracy": 0.19139863401651383,
"num_tokens": 7902231.0,
"step": 3455
},
{
"entropy": 5.600442361831665,
"epoch": 0.3323727185398655,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993685586663871,
"loss": 5.488,
"mean_token_accuracy": 0.18677273094654084,
"num_tokens": 7913364.0,
"step": 3460
},
{
"entropy": 5.820091962814331,
"epoch": 0.33285302593659943,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993659893850281,
"loss": 5.6781,
"mean_token_accuracy": 0.1703316792845726,
"num_tokens": 7925217.0,
"step": 3465
},
{
"entropy": 5.581819820404053,
"epoch": 0.3333333333333333,
"grad_norm": 0.8984375,
"learning_rate": 0.0004993634148945573,
"loss": 5.4198,
"mean_token_accuracy": 0.1796149030327797,
"num_tokens": 7937636.0,
"step": 3470
},
{
"entropy": 5.565121173858643,
"epoch": 0.33381364073006725,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993608351950341,
"loss": 5.4536,
"mean_token_accuracy": 0.18050845563411713,
"num_tokens": 7948958.0,
"step": 3475
},
{
"entropy": 5.6448219299316404,
"epoch": 0.33429394812680113,
"grad_norm": 0.9921875,
"learning_rate": 0.0004993582502865185,
"loss": 5.4741,
"mean_token_accuracy": 0.1872809574007988,
"num_tokens": 7960013.0,
"step": 3480
},
{
"entropy": 5.595401859283447,
"epoch": 0.33477425552353507,
"grad_norm": 1.015625,
"learning_rate": 0.0004993556601690706,
"loss": 5.5031,
"mean_token_accuracy": 0.18324995785951614,
"num_tokens": 7971041.0,
"step": 3485
},
{
"entropy": 5.632128286361694,
"epoch": 0.33525456292026895,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993530648427505,
"loss": 5.5098,
"mean_token_accuracy": 0.17839510440826417,
"num_tokens": 7982752.0,
"step": 3490
},
{
"entropy": 5.637563943862915,
"epoch": 0.3357348703170029,
"grad_norm": 1.0625,
"learning_rate": 0.0004993504643076184,
"loss": 5.3971,
"mean_token_accuracy": 0.18504412919282914,
"num_tokens": 7993681.0,
"step": 3495
},
{
"entropy": 5.538132953643799,
"epoch": 0.33621517771373677,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993478585637347,
"loss": 5.4184,
"mean_token_accuracy": 0.18824413269758225,
"num_tokens": 8004727.0,
"step": 3500
},
{
"entropy": 5.585167694091797,
"epoch": 0.3366954851104707,
"grad_norm": 0.98046875,
"learning_rate": 0.0004993452476111599,
"loss": 5.4432,
"mean_token_accuracy": 0.1928482621908188,
"num_tokens": 8015423.0,
"step": 3505
},
{
"entropy": 5.588391065597534,
"epoch": 0.3371757925072046,
"grad_norm": 0.890625,
"learning_rate": 0.0004993426314499546,
"loss": 5.4039,
"mean_token_accuracy": 0.19309359788894653,
"num_tokens": 8027911.0,
"step": 3510
},
{
"entropy": 5.594974565505981,
"epoch": 0.3376560999039385,
"grad_norm": 0.953125,
"learning_rate": 0.0004993400100801796,
"loss": 5.4243,
"mean_token_accuracy": 0.18888370990753173,
"num_tokens": 8038831.0,
"step": 3515
},
{
"entropy": 5.477750301361084,
"epoch": 0.3381364073006724,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993373835018956,
"loss": 5.3115,
"mean_token_accuracy": 0.19790522754192352,
"num_tokens": 8049906.0,
"step": 3520
},
{
"entropy": 5.484724855422973,
"epoch": 0.33861671469740634,
"grad_norm": 0.92578125,
"learning_rate": 0.0004993347517151638,
"loss": 5.4212,
"mean_token_accuracy": 0.18728912770748138,
"num_tokens": 8061158.0,
"step": 3525
},
{
"entropy": 5.6147716522216795,
"epoch": 0.3390970220941403,
"grad_norm": 1.046875,
"learning_rate": 0.0004993321147200452,
"loss": 5.3792,
"mean_token_accuracy": 0.18561409711837767,
"num_tokens": 8071958.0,
"step": 3530
},
{
"entropy": 5.567511987686157,
"epoch": 0.33957732949087416,
"grad_norm": 1.0390625,
"learning_rate": 0.000499329472516601,
"loss": 5.38,
"mean_token_accuracy": 0.18467946201562882,
"num_tokens": 8084068.0,
"step": 3535
},
{
"entropy": 5.463061094284058,
"epoch": 0.3400576368876081,
"grad_norm": 0.98828125,
"learning_rate": 0.0004993268251048925,
"loss": 5.2918,
"mean_token_accuracy": 0.20272473245859146,
"num_tokens": 8096132.0,
"step": 3540
},
{
"entropy": 5.551108026504517,
"epoch": 0.340537944284342,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993241724849814,
"loss": 5.4552,
"mean_token_accuracy": 0.1927696093916893,
"num_tokens": 8107327.0,
"step": 3545
},
{
"entropy": 5.619629001617431,
"epoch": 0.3410182516810759,
"grad_norm": 1.0,
"learning_rate": 0.000499321514656929,
"loss": 5.4397,
"mean_token_accuracy": 0.18920674920082092,
"num_tokens": 8118584.0,
"step": 3550
},
{
"entropy": 5.625332069396973,
"epoch": 0.3414985590778098,
"grad_norm": 1.046875,
"learning_rate": 0.0004993188516207972,
"loss": 5.478,
"mean_token_accuracy": 0.1815608412027359,
"num_tokens": 8130081.0,
"step": 3555
},
{
"entropy": 5.557459163665771,
"epoch": 0.34197886647454373,
"grad_norm": 1.046875,
"learning_rate": 0.0004993161833766478,
"loss": 5.441,
"mean_token_accuracy": 0.18856000006198884,
"num_tokens": 8141463.0,
"step": 3560
},
{
"entropy": 5.623035001754761,
"epoch": 0.3424591738712776,
"grad_norm": 1.109375,
"learning_rate": 0.0004993135099245426,
"loss": 5.479,
"mean_token_accuracy": 0.18447892665863036,
"num_tokens": 8153863.0,
"step": 3565
},
{
"entropy": 5.488258695602417,
"epoch": 0.34293948126801155,
"grad_norm": 1.109375,
"learning_rate": 0.0004993108312645438,
"loss": 5.3974,
"mean_token_accuracy": 0.18443325012922288,
"num_tokens": 8165695.0,
"step": 3570
},
{
"entropy": 5.602726173400879,
"epoch": 0.34341978866474543,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993081473967135,
"loss": 5.457,
"mean_token_accuracy": 0.1891469821333885,
"num_tokens": 8176456.0,
"step": 3575
},
{
"entropy": 5.6182811737060545,
"epoch": 0.34390009606147937,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993054583211143,
"loss": 5.4567,
"mean_token_accuracy": 0.1888190761208534,
"num_tokens": 8189050.0,
"step": 3580
},
{
"entropy": 5.482992887496948,
"epoch": 0.34438040345821325,
"grad_norm": 0.90625,
"learning_rate": 0.0004993027640378081,
"loss": 5.3534,
"mean_token_accuracy": 0.19064465761184693,
"num_tokens": 8200011.0,
"step": 3585
},
{
"entropy": 5.55635633468628,
"epoch": 0.3448607108549472,
"grad_norm": 0.99609375,
"learning_rate": 0.000499300064546858,
"loss": 5.3727,
"mean_token_accuracy": 0.19231539517641066,
"num_tokens": 8211770.0,
"step": 3590
},
{
"entropy": 5.568660497665405,
"epoch": 0.34534101825168106,
"grad_norm": 1.15625,
"learning_rate": 0.0004992973598483264,
"loss": 5.408,
"mean_token_accuracy": 0.19162147790193557,
"num_tokens": 8223582.0,
"step": 3595
},
{
"entropy": 5.642741394042969,
"epoch": 0.345821325648415,
"grad_norm": 1.078125,
"learning_rate": 0.000499294649942276,
"loss": 5.5309,
"mean_token_accuracy": 0.18496304154396057,
"num_tokens": 8234336.0,
"step": 3600
},
{
"entropy": 5.624276733398437,
"epoch": 0.3463016330451489,
"grad_norm": 1.0625,
"learning_rate": 0.0004992919348287699,
"loss": 5.4281,
"mean_token_accuracy": 0.18882226347923278,
"num_tokens": 8244605.0,
"step": 3605
},
{
"entropy": 5.608572101593017,
"epoch": 0.3467819404418828,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992892145078711,
"loss": 5.4808,
"mean_token_accuracy": 0.18347105085849763,
"num_tokens": 8255876.0,
"step": 3610
},
{
"entropy": 5.549061632156372,
"epoch": 0.3472622478386167,
"grad_norm": 0.984375,
"learning_rate": 0.0004992864889796427,
"loss": 5.3814,
"mean_token_accuracy": 0.19023108333349228,
"num_tokens": 8266602.0,
"step": 3615
},
{
"entropy": 5.546348810195923,
"epoch": 0.34774255523535064,
"grad_norm": 0.921875,
"learning_rate": 0.0004992837582441481,
"loss": 5.3625,
"mean_token_accuracy": 0.19489242136478424,
"num_tokens": 8279804.0,
"step": 3620
},
{
"entropy": 5.629941320419311,
"epoch": 0.3482228626320845,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992810223014506,
"loss": 5.4606,
"mean_token_accuracy": 0.1908916175365448,
"num_tokens": 8291020.0,
"step": 3625
},
{
"entropy": 5.618665075302124,
"epoch": 0.34870317002881845,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992782811516137,
"loss": 5.4171,
"mean_token_accuracy": 0.19629807025194168,
"num_tokens": 8302192.0,
"step": 3630
},
{
"entropy": 5.548134994506836,
"epoch": 0.34918347742555234,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992755347947011,
"loss": 5.3865,
"mean_token_accuracy": 0.1896643429994583,
"num_tokens": 8313649.0,
"step": 3635
},
{
"entropy": 5.500442028045654,
"epoch": 0.34966378482228627,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992727832307766,
"loss": 5.385,
"mean_token_accuracy": 0.1900986060500145,
"num_tokens": 8324694.0,
"step": 3640
},
{
"entropy": 5.660983419418335,
"epoch": 0.35014409221902015,
"grad_norm": 1.03125,
"learning_rate": 0.0004992700264599039,
"loss": 5.5324,
"mean_token_accuracy": 0.18272839337587357,
"num_tokens": 8336517.0,
"step": 3645
},
{
"entropy": 5.593430089950561,
"epoch": 0.3506243996157541,
"grad_norm": 0.91015625,
"learning_rate": 0.0004992672644821473,
"loss": 5.4967,
"mean_token_accuracy": 0.1796460419893265,
"num_tokens": 8349001.0,
"step": 3650
},
{
"entropy": 5.585159015655518,
"epoch": 0.35110470701248797,
"grad_norm": 1.015625,
"learning_rate": 0.0004992644972975707,
"loss": 5.4108,
"mean_token_accuracy": 0.1927414983510971,
"num_tokens": 8361230.0,
"step": 3655
},
{
"entropy": 5.5338012218475345,
"epoch": 0.3515850144092219,
"grad_norm": 1.015625,
"learning_rate": 0.0004992617249062383,
"loss": 5.3264,
"mean_token_accuracy": 0.19598029255867006,
"num_tokens": 8372159.0,
"step": 3660
},
{
"entropy": 5.557580852508545,
"epoch": 0.3520653218059558,
"grad_norm": 1.03125,
"learning_rate": 0.0004992589473082147,
"loss": 5.4485,
"mean_token_accuracy": 0.19024865478277206,
"num_tokens": 8383228.0,
"step": 3665
},
{
"entropy": 5.60696702003479,
"epoch": 0.3525456292026897,
"grad_norm": 1.03125,
"learning_rate": 0.0004992561645035641,
"loss": 5.3997,
"mean_token_accuracy": 0.18926765471696855,
"num_tokens": 8394582.0,
"step": 3670
},
{
"entropy": 5.584086465835571,
"epoch": 0.3530259365994236,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992533764923515,
"loss": 5.4094,
"mean_token_accuracy": 0.18775029927492143,
"num_tokens": 8406784.0,
"step": 3675
},
{
"entropy": 5.534383869171142,
"epoch": 0.35350624399615754,
"grad_norm": 1.0,
"learning_rate": 0.0004992505832746412,
"loss": 5.3932,
"mean_token_accuracy": 0.194275338947773,
"num_tokens": 8418405.0,
"step": 3680
},
{
"entropy": 5.582950401306152,
"epoch": 0.3539865513928914,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992477848504983,
"loss": 5.3445,
"mean_token_accuracy": 0.19232997596263884,
"num_tokens": 8430432.0,
"step": 3685
},
{
"entropy": 5.566782712936401,
"epoch": 0.35446685878962536,
"grad_norm": 0.9140625,
"learning_rate": 0.0004992449812199877,
"loss": 5.4912,
"mean_token_accuracy": 0.18145429342985153,
"num_tokens": 8442423.0,
"step": 3690
},
{
"entropy": 5.585980749130249,
"epoch": 0.3549471661863593,
"grad_norm": 0.921875,
"learning_rate": 0.0004992421723831745,
"loss": 5.4931,
"mean_token_accuracy": 0.18717589229345322,
"num_tokens": 8454951.0,
"step": 3695
},
{
"entropy": 5.527453804016114,
"epoch": 0.3554274735830932,
"grad_norm": 0.95703125,
"learning_rate": 0.0004992393583401239,
"loss": 5.3484,
"mean_token_accuracy": 0.1946762263774872,
"num_tokens": 8467758.0,
"step": 3700
},
{
"entropy": 5.585886812210083,
"epoch": 0.3559077809798271,
"grad_norm": 1.015625,
"learning_rate": 0.0004992365390909014,
"loss": 5.4409,
"mean_token_accuracy": 0.18967984169721602,
"num_tokens": 8479728.0,
"step": 3705
},
{
"entropy": 5.6186473846435545,
"epoch": 0.356388088376561,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992337146355721,
"loss": 5.4939,
"mean_token_accuracy": 0.18735775500535964,
"num_tokens": 8492099.0,
"step": 3710
},
{
"entropy": 5.608113765716553,
"epoch": 0.35686839577329493,
"grad_norm": 0.99609375,
"learning_rate": 0.0004992308849742019,
"loss": 5.4456,
"mean_token_accuracy": 0.1802011474967003,
"num_tokens": 8504657.0,
"step": 3715
},
{
"entropy": 5.57549901008606,
"epoch": 0.3573487031700288,
"grad_norm": 1.046875,
"learning_rate": 0.0004992280501068563,
"loss": 5.4066,
"mean_token_accuracy": 0.19084483236074448,
"num_tokens": 8514728.0,
"step": 3720
},
{
"entropy": 5.635108757019043,
"epoch": 0.35782901056676275,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992252100336012,
"loss": 5.5219,
"mean_token_accuracy": 0.18584865778684617,
"num_tokens": 8525588.0,
"step": 3725
},
{
"entropy": 5.581790542602539,
"epoch": 0.35830931796349663,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992223647545027,
"loss": 5.4877,
"mean_token_accuracy": 0.18325926661491393,
"num_tokens": 8537468.0,
"step": 3730
},
{
"entropy": 5.597702932357788,
"epoch": 0.35878962536023057,
"grad_norm": 1.0078125,
"learning_rate": 0.0004992195142696266,
"loss": 5.3977,
"mean_token_accuracy": 0.19316060990095138,
"num_tokens": 8548598.0,
"step": 3735
},
{
"entropy": 5.409280014038086,
"epoch": 0.35926993275696445,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992166585790391,
"loss": 5.2728,
"mean_token_accuracy": 0.20413144379854203,
"num_tokens": 8560301.0,
"step": 3740
},
{
"entropy": 5.53164644241333,
"epoch": 0.3597502401536984,
"grad_norm": 1.078125,
"learning_rate": 0.0004992137976828067,
"loss": 5.368,
"mean_token_accuracy": 0.1967419356107712,
"num_tokens": 8571186.0,
"step": 3745
},
{
"entropy": 5.545560312271118,
"epoch": 0.36023054755043227,
"grad_norm": 1.2109375,
"learning_rate": 0.0004992109315809955,
"loss": 5.3598,
"mean_token_accuracy": 0.19761675149202346,
"num_tokens": 8580725.0,
"step": 3750
},
{
"entropy": 5.526527690887451,
"epoch": 0.3607108549471662,
"grad_norm": 0.90234375,
"learning_rate": 0.0004992080602736725,
"loss": 5.5078,
"mean_token_accuracy": 0.1807180315256119,
"num_tokens": 8594598.0,
"step": 3755
},
{
"entropy": 5.6951195240020756,
"epoch": 0.3611911623439001,
"grad_norm": 0.953125,
"learning_rate": 0.0004992051837609039,
"loss": 5.4955,
"mean_token_accuracy": 0.18285927772521973,
"num_tokens": 8606733.0,
"step": 3760
},
{
"entropy": 5.609960460662842,
"epoch": 0.361671469740634,
"grad_norm": 0.9765625,
"learning_rate": 0.0004992023020427568,
"loss": 5.4313,
"mean_token_accuracy": 0.1881916657090187,
"num_tokens": 8618863.0,
"step": 3765
},
{
"entropy": 5.424804210662842,
"epoch": 0.3621517771373679,
"grad_norm": 1.0,
"learning_rate": 0.0004991994151192979,
"loss": 5.2702,
"mean_token_accuracy": 0.1962995320558548,
"num_tokens": 8629270.0,
"step": 3770
},
{
"entropy": 5.53543701171875,
"epoch": 0.36263208453410184,
"grad_norm": 1.046875,
"learning_rate": 0.0004991965229905943,
"loss": 5.4789,
"mean_token_accuracy": 0.19050425440073013,
"num_tokens": 8641363.0,
"step": 3775
},
{
"entropy": 5.642627811431884,
"epoch": 0.3631123919308357,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991936256567133,
"loss": 5.4365,
"mean_token_accuracy": 0.1923316612839699,
"num_tokens": 8653233.0,
"step": 3780
},
{
"entropy": 5.533376264572143,
"epoch": 0.36359269932756966,
"grad_norm": 0.89453125,
"learning_rate": 0.000499190723117722,
"loss": 5.4265,
"mean_token_accuracy": 0.18614102751016617,
"num_tokens": 8665192.0,
"step": 3785
},
{
"entropy": 5.642783832550049,
"epoch": 0.36407300672430354,
"grad_norm": 1.078125,
"learning_rate": 0.0004991878153736877,
"loss": 5.5196,
"mean_token_accuracy": 0.18100242018699647,
"num_tokens": 8677669.0,
"step": 3790
},
{
"entropy": 5.474308347702026,
"epoch": 0.3645533141210375,
"grad_norm": 1.0625,
"learning_rate": 0.0004991849024246781,
"loss": 5.314,
"mean_token_accuracy": 0.19570468962192536,
"num_tokens": 8688002.0,
"step": 3795
},
{
"entropy": 5.516477632522583,
"epoch": 0.36503362151777136,
"grad_norm": 1.03125,
"learning_rate": 0.0004991819842707608,
"loss": 5.367,
"mean_token_accuracy": 0.18995878249406814,
"num_tokens": 8698396.0,
"step": 3800
},
{
"entropy": 5.623310947418213,
"epoch": 0.3655139289145053,
"grad_norm": 1.0,
"learning_rate": 0.0004991790609120035,
"loss": 5.3885,
"mean_token_accuracy": 0.19130971878767014,
"num_tokens": 8711135.0,
"step": 3805
},
{
"entropy": 5.50729808807373,
"epoch": 0.3659942363112392,
"grad_norm": 0.98828125,
"learning_rate": 0.000499176132348474,
"loss": 5.4236,
"mean_token_accuracy": 0.19352682530879975,
"num_tokens": 8723707.0,
"step": 3810
},
{
"entropy": 5.572604703903198,
"epoch": 0.3664745437079731,
"grad_norm": 1.09375,
"learning_rate": 0.0004991731985802405,
"loss": 5.3627,
"mean_token_accuracy": 0.1964678570628166,
"num_tokens": 8734193.0,
"step": 3815
},
{
"entropy": 5.517657279968262,
"epoch": 0.366954851104707,
"grad_norm": 1.078125,
"learning_rate": 0.0004991702596073708,
"loss": 5.4572,
"mean_token_accuracy": 0.18392777889966966,
"num_tokens": 8745619.0,
"step": 3820
},
{
"entropy": 5.462949419021607,
"epoch": 0.36743515850144093,
"grad_norm": 0.96484375,
"learning_rate": 0.0004991673154299335,
"loss": 5.3582,
"mean_token_accuracy": 0.19223012924194335,
"num_tokens": 8757331.0,
"step": 3825
},
{
"entropy": 5.527809762954712,
"epoch": 0.3679154658981748,
"grad_norm": 0.93359375,
"learning_rate": 0.0004991643660479967,
"loss": 5.3655,
"mean_token_accuracy": 0.19106081277132034,
"num_tokens": 8768840.0,
"step": 3830
},
{
"entropy": 5.562812089920044,
"epoch": 0.36839577329490875,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991614114616289,
"loss": 5.3195,
"mean_token_accuracy": 0.19561284482479097,
"num_tokens": 8781214.0,
"step": 3835
},
{
"entropy": 5.584686708450318,
"epoch": 0.3688760806916426,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991584516708988,
"loss": 5.4033,
"mean_token_accuracy": 0.1916833609342575,
"num_tokens": 8791645.0,
"step": 3840
},
{
"entropy": 5.599692010879517,
"epoch": 0.36935638808837656,
"grad_norm": 1.015625,
"learning_rate": 0.0004991554866758751,
"loss": 5.5787,
"mean_token_accuracy": 0.17887818589806556,
"num_tokens": 8803286.0,
"step": 3845
},
{
"entropy": 5.589219093322754,
"epoch": 0.36983669548511044,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991525164766265,
"loss": 5.3662,
"mean_token_accuracy": 0.18918564915657043,
"num_tokens": 8814207.0,
"step": 3850
},
{
"entropy": 5.53467755317688,
"epoch": 0.3703170028818444,
"grad_norm": 0.9296875,
"learning_rate": 0.0004991495410732222,
"loss": 5.417,
"mean_token_accuracy": 0.18368727713823318,
"num_tokens": 8825540.0,
"step": 3855
},
{
"entropy": 5.591689157485962,
"epoch": 0.37079731027857826,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991465604657311,
"loss": 5.547,
"mean_token_accuracy": 0.17688291668891906,
"num_tokens": 8838182.0,
"step": 3860
},
{
"entropy": 5.570683383941651,
"epoch": 0.3712776176753122,
"grad_norm": 0.96875,
"learning_rate": 0.0004991435746542224,
"loss": 5.4198,
"mean_token_accuracy": 0.19610581248998643,
"num_tokens": 8850211.0,
"step": 3865
},
{
"entropy": 5.517876291275025,
"epoch": 0.37175792507204614,
"grad_norm": 0.94921875,
"learning_rate": 0.0004991405836387655,
"loss": 5.457,
"mean_token_accuracy": 0.182083423435688,
"num_tokens": 8862804.0,
"step": 3870
},
{
"entropy": 5.559181594848633,
"epoch": 0.37223823246878,
"grad_norm": 1.0625,
"learning_rate": 0.0004991375874194298,
"loss": 5.405,
"mean_token_accuracy": 0.18535222709178925,
"num_tokens": 8874112.0,
"step": 3875
},
{
"entropy": 5.540090560913086,
"epoch": 0.37271853986551395,
"grad_norm": 1.015625,
"learning_rate": 0.000499134585996285,
"loss": 5.4274,
"mean_token_accuracy": 0.18555488735437392,
"num_tokens": 8885114.0,
"step": 3880
},
{
"entropy": 5.529178762435913,
"epoch": 0.37319884726224783,
"grad_norm": 1.0,
"learning_rate": 0.0004991315793694004,
"loss": 5.3039,
"mean_token_accuracy": 0.19386369735002518,
"num_tokens": 8895555.0,
"step": 3885
},
{
"entropy": 5.464783573150635,
"epoch": 0.37367915465898177,
"grad_norm": 1.0,
"learning_rate": 0.0004991285675388463,
"loss": 5.3138,
"mean_token_accuracy": 0.20528001785278321,
"num_tokens": 8906073.0,
"step": 3890
},
{
"entropy": 5.548325109481811,
"epoch": 0.37415946205571565,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991255505046922,
"loss": 5.4661,
"mean_token_accuracy": 0.18886378407478333,
"num_tokens": 8916587.0,
"step": 3895
},
{
"entropy": 5.586668920516968,
"epoch": 0.3746397694524496,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991225282670083,
"loss": 5.3657,
"mean_token_accuracy": 0.1910933658480644,
"num_tokens": 8927923.0,
"step": 3900
},
{
"entropy": 5.475469779968262,
"epoch": 0.37512007684918347,
"grad_norm": 1.0078125,
"learning_rate": 0.000499119500825865,
"loss": 5.405,
"mean_token_accuracy": 0.1913732722401619,
"num_tokens": 8939939.0,
"step": 3905
},
{
"entropy": 5.515975379943848,
"epoch": 0.3756003842459174,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991164681813323,
"loss": 5.3909,
"mean_token_accuracy": 0.19644259959459304,
"num_tokens": 8951748.0,
"step": 3910
},
{
"entropy": 5.5175879955291744,
"epoch": 0.3760806916426513,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991134303334807,
"loss": 5.2979,
"mean_token_accuracy": 0.201249860227108,
"num_tokens": 8962922.0,
"step": 3915
},
{
"entropy": 5.4236226081848145,
"epoch": 0.3765609990393852,
"grad_norm": 0.953125,
"learning_rate": 0.0004991103872823807,
"loss": 5.2715,
"mean_token_accuracy": 0.1982575923204422,
"num_tokens": 8974013.0,
"step": 3920
},
{
"entropy": 5.478136730194092,
"epoch": 0.3770413064361191,
"grad_norm": 0.94921875,
"learning_rate": 0.000499107339028103,
"loss": 5.3664,
"mean_token_accuracy": 0.19188770800828933,
"num_tokens": 8986032.0,
"step": 3925
},
{
"entropy": 5.5743492603302,
"epoch": 0.37752161383285304,
"grad_norm": 1.015625,
"learning_rate": 0.0004991042855707184,
"loss": 5.3564,
"mean_token_accuracy": 0.19044692516326905,
"num_tokens": 8996889.0,
"step": 3930
},
{
"entropy": 5.498291397094727,
"epoch": 0.3780019212295869,
"grad_norm": 1.0390625,
"learning_rate": 0.0004991012269102977,
"loss": 5.3435,
"mean_token_accuracy": 0.19032660573720933,
"num_tokens": 9007594.0,
"step": 3935
},
{
"entropy": 5.414751243591309,
"epoch": 0.37848222862632086,
"grad_norm": 1.0390625,
"learning_rate": 0.0004990981630469119,
"loss": 5.33,
"mean_token_accuracy": 0.19158338755369186,
"num_tokens": 9018097.0,
"step": 3940
},
{
"entropy": 5.559832191467285,
"epoch": 0.37896253602305474,
"grad_norm": 0.984375,
"learning_rate": 0.0004990950939806323,
"loss": 5.4603,
"mean_token_accuracy": 0.1812780097126961,
"num_tokens": 9029554.0,
"step": 3945
},
{
"entropy": 5.55887622833252,
"epoch": 0.3794428434197887,
"grad_norm": 0.95703125,
"learning_rate": 0.00049909201971153,
"loss": 5.3256,
"mean_token_accuracy": 0.19056572914123535,
"num_tokens": 9042518.0,
"step": 3950
},
{
"entropy": 5.506370973587036,
"epoch": 0.37992315081652256,
"grad_norm": 0.99609375,
"learning_rate": 0.0004990889402396763,
"loss": 5.3644,
"mean_token_accuracy": 0.19444247484207153,
"num_tokens": 9054524.0,
"step": 3955
},
{
"entropy": 5.506084156036377,
"epoch": 0.3804034582132565,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990858555651431,
"loss": 5.3391,
"mean_token_accuracy": 0.18973129987716675,
"num_tokens": 9065375.0,
"step": 3960
},
{
"entropy": 5.506417083740234,
"epoch": 0.3808837656099904,
"grad_norm": 1.09375,
"learning_rate": 0.0004990827656880015,
"loss": 5.2982,
"mean_token_accuracy": 0.19565317034721375,
"num_tokens": 9076338.0,
"step": 3965
},
{
"entropy": 5.502223873138428,
"epoch": 0.3813640730067243,
"grad_norm": 0.9609375,
"learning_rate": 0.0004990796706083235,
"loss": 5.3421,
"mean_token_accuracy": 0.19122555553913118,
"num_tokens": 9088407.0,
"step": 3970
},
{
"entropy": 5.494334125518799,
"epoch": 0.3818443804034582,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990765703261809,
"loss": 5.3036,
"mean_token_accuracy": 0.19051105678081512,
"num_tokens": 9099833.0,
"step": 3975
},
{
"entropy": 5.3872777938842775,
"epoch": 0.38232468780019213,
"grad_norm": 0.9375,
"learning_rate": 0.0004990734648416458,
"loss": 5.2685,
"mean_token_accuracy": 0.19974401593208313,
"num_tokens": 9111126.0,
"step": 3980
},
{
"entropy": 5.513306760787964,
"epoch": 0.382804995196926,
"grad_norm": 1.0,
"learning_rate": 0.0004990703541547901,
"loss": 5.3689,
"mean_token_accuracy": 0.1958555817604065,
"num_tokens": 9121979.0,
"step": 3985
},
{
"entropy": 5.5826314926147464,
"epoch": 0.38328530259365995,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990672382656863,
"loss": 5.3873,
"mean_token_accuracy": 0.1911478102207184,
"num_tokens": 9132929.0,
"step": 3990
},
{
"entropy": 5.551815366744995,
"epoch": 0.38376560999039383,
"grad_norm": 0.98046875,
"learning_rate": 0.0004990641171744064,
"loss": 5.3495,
"mean_token_accuracy": 0.19590957462787628,
"num_tokens": 9143903.0,
"step": 3995
},
{
"entropy": 5.567360591888428,
"epoch": 0.38424591738712777,
"grad_norm": 1.09375,
"learning_rate": 0.0004990609908810231,
"loss": 5.4273,
"mean_token_accuracy": 0.19291835129261017,
"num_tokens": 9154416.0,
"step": 4000
},
{
"entropy": 5.54535870552063,
"epoch": 0.38472622478386165,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990578593856089,
"loss": 5.409,
"mean_token_accuracy": 0.19129804521799088,
"num_tokens": 9165613.0,
"step": 4005
},
{
"entropy": 5.470605134963989,
"epoch": 0.3852065321805956,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990547226882366,
"loss": 5.3711,
"mean_token_accuracy": 0.1947087824344635,
"num_tokens": 9177884.0,
"step": 4010
},
{
"entropy": 5.600809383392334,
"epoch": 0.38568683957732947,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990515807889788,
"loss": 5.4902,
"mean_token_accuracy": 0.1800956666469574,
"num_tokens": 9190041.0,
"step": 4015
},
{
"entropy": 5.623603248596192,
"epoch": 0.3861671469740634,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990484336879087,
"loss": 5.3883,
"mean_token_accuracy": 0.19113666117191314,
"num_tokens": 9202390.0,
"step": 4020
},
{
"entropy": 5.474275207519531,
"epoch": 0.3866474543707973,
"grad_norm": 1.0859375,
"learning_rate": 0.0004990452813850992,
"loss": 5.3836,
"mean_token_accuracy": 0.19006328135728837,
"num_tokens": 9213437.0,
"step": 4025
},
{
"entropy": 5.555831718444824,
"epoch": 0.3871277617675312,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990421238806236,
"loss": 5.4479,
"mean_token_accuracy": 0.1827810600399971,
"num_tokens": 9226310.0,
"step": 4030
},
{
"entropy": 5.534276008605957,
"epoch": 0.38760806916426516,
"grad_norm": 1.03125,
"learning_rate": 0.0004990389611745551,
"loss": 5.3874,
"mean_token_accuracy": 0.1896564930677414,
"num_tokens": 9236271.0,
"step": 4035
},
{
"entropy": 5.5598194122314455,
"epoch": 0.38808837656099904,
"grad_norm": 1.015625,
"learning_rate": 0.0004990357932669672,
"loss": 5.4553,
"mean_token_accuracy": 0.1913477137684822,
"num_tokens": 9247755.0,
"step": 4040
},
{
"entropy": 5.499641227722168,
"epoch": 0.388568683957733,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990326201579335,
"loss": 5.3006,
"mean_token_accuracy": 0.20061542093753815,
"num_tokens": 9259821.0,
"step": 4045
},
{
"entropy": 5.5496776580810545,
"epoch": 0.38904899135446686,
"grad_norm": 1.125,
"learning_rate": 0.0004990294418475274,
"loss": 5.4043,
"mean_token_accuracy": 0.18962285965681075,
"num_tokens": 9270663.0,
"step": 4050
},
{
"entropy": 5.5318115234375,
"epoch": 0.3895292987512008,
"grad_norm": 1.28125,
"learning_rate": 0.0004990262583358231,
"loss": 5.4028,
"mean_token_accuracy": 0.1873584508895874,
"num_tokens": 9282588.0,
"step": 4055
},
{
"entropy": 5.505263233184815,
"epoch": 0.3900096061479347,
"grad_norm": 1.015625,
"learning_rate": 0.0004990230696228943,
"loss": 5.3763,
"mean_token_accuracy": 0.18254790604114532,
"num_tokens": 9293368.0,
"step": 4060
},
{
"entropy": 5.559754848480225,
"epoch": 0.3904899135446686,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990198757088149,
"loss": 5.4642,
"mean_token_accuracy": 0.18748684972524643,
"num_tokens": 9305962.0,
"step": 4065
},
{
"entropy": 5.516952753067017,
"epoch": 0.3909702209414025,
"grad_norm": 1.0,
"learning_rate": 0.0004990166765936593,
"loss": 5.3301,
"mean_token_accuracy": 0.19609273076057435,
"num_tokens": 9317955.0,
"step": 4070
},
{
"entropy": 5.497630071640015,
"epoch": 0.3914505283381364,
"grad_norm": 0.98828125,
"learning_rate": 0.0004990134722775016,
"loss": 5.3373,
"mean_token_accuracy": 0.19048384428024293,
"num_tokens": 9329491.0,
"step": 4075
},
{
"entropy": 5.501287317276001,
"epoch": 0.3919308357348703,
"grad_norm": 1.015625,
"learning_rate": 0.0004990102627604162,
"loss": 5.4717,
"mean_token_accuracy": 0.19690733551979064,
"num_tokens": 9341612.0,
"step": 4080
},
{
"entropy": 5.572618722915649,
"epoch": 0.39241114313160425,
"grad_norm": 1.046875,
"learning_rate": 0.0004990070480424778,
"loss": 5.411,
"mean_token_accuracy": 0.18688351958990096,
"num_tokens": 9352302.0,
"step": 4085
},
{
"entropy": 5.529020738601685,
"epoch": 0.3928914505283381,
"grad_norm": 0.95703125,
"learning_rate": 0.0004990038281237608,
"loss": 5.3502,
"mean_token_accuracy": 0.19336572736501695,
"num_tokens": 9363303.0,
"step": 4090
},
{
"entropy": 5.446750688552856,
"epoch": 0.39337175792507206,
"grad_norm": 1.0703125,
"learning_rate": 0.0004990006030043401,
"loss": 5.3064,
"mean_token_accuracy": 0.1869595780968666,
"num_tokens": 9375878.0,
"step": 4095
},
{
"entropy": 5.538749361038208,
"epoch": 0.39385206532180594,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989973726842906,
"loss": 5.3651,
"mean_token_accuracy": 0.1941888928413391,
"num_tokens": 9388342.0,
"step": 4100
},
{
"entropy": 5.456574630737305,
"epoch": 0.3943323727185399,
"grad_norm": 1.109375,
"learning_rate": 0.0004989941371636872,
"loss": 5.2692,
"mean_token_accuracy": 0.19488134682178498,
"num_tokens": 9399047.0,
"step": 4105
},
{
"entropy": 5.524734115600586,
"epoch": 0.39481268011527376,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989908964426051,
"loss": 5.3773,
"mean_token_accuracy": 0.1977031409740448,
"num_tokens": 9410172.0,
"step": 4110
},
{
"entropy": 5.5667112350463865,
"epoch": 0.3952929875120077,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989876505211194,
"loss": 5.5074,
"mean_token_accuracy": 0.18664929568767546,
"num_tokens": 9422287.0,
"step": 4115
},
{
"entropy": 5.591302490234375,
"epoch": 0.3957732949087416,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989843993993056,
"loss": 5.3713,
"mean_token_accuracy": 0.19570473879575728,
"num_tokens": 9433709.0,
"step": 4120
},
{
"entropy": 5.40889949798584,
"epoch": 0.3962536023054755,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989811430772392,
"loss": 5.2625,
"mean_token_accuracy": 0.19686949849128724,
"num_tokens": 9445138.0,
"step": 4125
},
{
"entropy": 5.490507650375366,
"epoch": 0.3967339097022094,
"grad_norm": 1.125,
"learning_rate": 0.0004989778815549957,
"loss": 5.4006,
"mean_token_accuracy": 0.18909133076667786,
"num_tokens": 9455263.0,
"step": 4130
},
{
"entropy": 5.5528974533081055,
"epoch": 0.39721421709894333,
"grad_norm": 1.015625,
"learning_rate": 0.0004989746148326508,
"loss": 5.3507,
"mean_token_accuracy": 0.19381564110517502,
"num_tokens": 9465491.0,
"step": 4135
},
{
"entropy": 5.424878263473511,
"epoch": 0.3976945244956772,
"grad_norm": 0.953125,
"learning_rate": 0.0004989713429102805,
"loss": 5.324,
"mean_token_accuracy": 0.1861468955874443,
"num_tokens": 9477601.0,
"step": 4140
},
{
"entropy": 5.505374145507813,
"epoch": 0.39817483189241115,
"grad_norm": 0.921875,
"learning_rate": 0.0004989680657879607,
"loss": 5.3778,
"mean_token_accuracy": 0.1953922465443611,
"num_tokens": 9489385.0,
"step": 4145
},
{
"entropy": 5.459268808364868,
"epoch": 0.39865513928914503,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989647834657675,
"loss": 5.248,
"mean_token_accuracy": 0.2002091407775879,
"num_tokens": 9501131.0,
"step": 4150
},
{
"entropy": 5.464940404891967,
"epoch": 0.39913544668587897,
"grad_norm": 0.9609375,
"learning_rate": 0.000498961495943777,
"loss": 5.4137,
"mean_token_accuracy": 0.19082237184047698,
"num_tokens": 9513094.0,
"step": 4155
},
{
"entropy": 5.63197512626648,
"epoch": 0.39961575408261285,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989582032220656,
"loss": 5.522,
"mean_token_accuracy": 0.18408708721399308,
"num_tokens": 9524538.0,
"step": 4160
},
{
"entropy": 5.56934609413147,
"epoch": 0.4000960614793468,
"grad_norm": 1.1484375,
"learning_rate": 0.0004989549053007096,
"loss": 5.3291,
"mean_token_accuracy": 0.20425097048282623,
"num_tokens": 9535284.0,
"step": 4165
},
{
"entropy": 5.497316694259643,
"epoch": 0.40057636887608067,
"grad_norm": 1.015625,
"learning_rate": 0.0004989516021797858,
"loss": 5.4125,
"mean_token_accuracy": 0.1898416116833687,
"num_tokens": 9546472.0,
"step": 4170
},
{
"entropy": 5.518751907348633,
"epoch": 0.4010566762728146,
"grad_norm": 0.96484375,
"learning_rate": 0.000498948293859371,
"loss": 5.407,
"mean_token_accuracy": 0.18592777997255325,
"num_tokens": 9558358.0,
"step": 4175
},
{
"entropy": 5.598344326019287,
"epoch": 0.4015369836695485,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989449803395415,
"loss": 5.4303,
"mean_token_accuracy": 0.19122905135154725,
"num_tokens": 9570653.0,
"step": 4180
},
{
"entropy": 5.527167081832886,
"epoch": 0.4020172910662824,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989416616203747,
"loss": 5.3722,
"mean_token_accuracy": 0.19405449777841569,
"num_tokens": 9582150.0,
"step": 4185
},
{
"entropy": 5.54949803352356,
"epoch": 0.4024975984630163,
"grad_norm": 0.97265625,
"learning_rate": 0.0004989383377019476,
"loss": 5.3174,
"mean_token_accuracy": 0.1968682125210762,
"num_tokens": 9592462.0,
"step": 4190
},
{
"entropy": 5.459809732437134,
"epoch": 0.40297790585975024,
"grad_norm": 0.95703125,
"learning_rate": 0.0004989350085843371,
"loss": 5.3186,
"mean_token_accuracy": 0.19509333968162537,
"num_tokens": 9604027.0,
"step": 4195
},
{
"entropy": 5.4236900329589846,
"epoch": 0.4034582132564842,
"grad_norm": 0.8828125,
"learning_rate": 0.0004989316742676207,
"loss": 5.3037,
"mean_token_accuracy": 0.19438586235046387,
"num_tokens": 9616325.0,
"step": 4200
},
{
"entropy": 5.422830820083618,
"epoch": 0.40393852065321806,
"grad_norm": 0.96875,
"learning_rate": 0.0004989283347518757,
"loss": 5.2652,
"mean_token_accuracy": 0.1953098714351654,
"num_tokens": 9628133.0,
"step": 4205
},
{
"entropy": 5.626201915740967,
"epoch": 0.404418828049952,
"grad_norm": 0.984375,
"learning_rate": 0.0004989249900371797,
"loss": 5.4747,
"mean_token_accuracy": 0.18496284484863282,
"num_tokens": 9639686.0,
"step": 4210
},
{
"entropy": 5.500379610061645,
"epoch": 0.4048991354466859,
"grad_norm": 0.96875,
"learning_rate": 0.0004989216401236103,
"loss": 5.3695,
"mean_token_accuracy": 0.1925930693745613,
"num_tokens": 9650222.0,
"step": 4215
},
{
"entropy": 5.430653524398804,
"epoch": 0.4053794428434198,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989182850112455,
"loss": 5.2787,
"mean_token_accuracy": 0.20574321448802949,
"num_tokens": 9661792.0,
"step": 4220
},
{
"entropy": 5.5395571231842045,
"epoch": 0.4058597502401537,
"grad_norm": 0.98828125,
"learning_rate": 0.0004989149247001629,
"loss": 5.3946,
"mean_token_accuracy": 0.18767669349908828,
"num_tokens": 9673000.0,
"step": 4225
},
{
"entropy": 5.5615856647491455,
"epoch": 0.40634005763688763,
"grad_norm": 0.99609375,
"learning_rate": 0.0004989115591904407,
"loss": 5.3381,
"mean_token_accuracy": 0.19707095324993135,
"num_tokens": 9685253.0,
"step": 4230
},
{
"entropy": 5.451553249359131,
"epoch": 0.4068203650336215,
"grad_norm": 0.9609375,
"learning_rate": 0.0004989081884821569,
"loss": 5.3231,
"mean_token_accuracy": 0.19093127846717833,
"num_tokens": 9697245.0,
"step": 4235
},
{
"entropy": 5.4586076736450195,
"epoch": 0.40730067243035545,
"grad_norm": 0.9765625,
"learning_rate": 0.0004989048125753899,
"loss": 5.3483,
"mean_token_accuracy": 0.19377574026584626,
"num_tokens": 9710095.0,
"step": 4240
},
{
"entropy": 5.475905895233154,
"epoch": 0.40778097982708933,
"grad_norm": 1.0,
"learning_rate": 0.000498901431470218,
"loss": 5.2338,
"mean_token_accuracy": 0.20166248977184295,
"num_tokens": 9721488.0,
"step": 4245
},
{
"entropy": 5.341732978820801,
"epoch": 0.40826128722382327,
"grad_norm": 0.96484375,
"learning_rate": 0.0004988980451667198,
"loss": 5.1862,
"mean_token_accuracy": 0.20070856511592866,
"num_tokens": 9733280.0,
"step": 4250
},
{
"entropy": 5.480173921585083,
"epoch": 0.40874159462055715,
"grad_norm": 0.921875,
"learning_rate": 0.0004988946536649737,
"loss": 5.3212,
"mean_token_accuracy": 0.18952195197343827,
"num_tokens": 9744514.0,
"step": 4255
},
{
"entropy": 5.479455947875977,
"epoch": 0.4092219020172911,
"grad_norm": 1.0625,
"learning_rate": 0.0004988912569650585,
"loss": 5.3254,
"mean_token_accuracy": 0.19506131410598754,
"num_tokens": 9754931.0,
"step": 4260
},
{
"entropy": 5.408810377120972,
"epoch": 0.40970220941402496,
"grad_norm": 0.95703125,
"learning_rate": 0.0004988878550670533,
"loss": 5.2945,
"mean_token_accuracy": 0.19845788776874543,
"num_tokens": 9765635.0,
"step": 4265
},
{
"entropy": 5.524256181716919,
"epoch": 0.4101825168107589,
"grad_norm": 1.046875,
"learning_rate": 0.0004988844479710369,
"loss": 5.3987,
"mean_token_accuracy": 0.1866440549492836,
"num_tokens": 9777512.0,
"step": 4270
},
{
"entropy": 5.5750714302062985,
"epoch": 0.4106628242074928,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988810356770884,
"loss": 5.4248,
"mean_token_accuracy": 0.1778908535838127,
"num_tokens": 9790128.0,
"step": 4275
},
{
"entropy": 5.4882549285888675,
"epoch": 0.4111431316042267,
"grad_norm": 0.9375,
"learning_rate": 0.000498877618185287,
"loss": 5.3427,
"mean_token_accuracy": 0.19739877730607985,
"num_tokens": 9802549.0,
"step": 4280
},
{
"entropy": 5.437657594680786,
"epoch": 0.4116234390009606,
"grad_norm": 1.0625,
"learning_rate": 0.0004988741954957121,
"loss": 5.3003,
"mean_token_accuracy": 0.19505833089351654,
"num_tokens": 9813736.0,
"step": 4285
},
{
"entropy": 5.389750909805298,
"epoch": 0.41210374639769454,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988707676084432,
"loss": 5.286,
"mean_token_accuracy": 0.20560728162527084,
"num_tokens": 9823785.0,
"step": 4290
},
{
"entropy": 5.458524465560913,
"epoch": 0.4125840537944284,
"grad_norm": 1.0546875,
"learning_rate": 0.0004988673345235597,
"loss": 5.2571,
"mean_token_accuracy": 0.1967891976237297,
"num_tokens": 9834910.0,
"step": 4295
},
{
"entropy": 5.488246631622315,
"epoch": 0.41306436119116235,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988638962411416,
"loss": 5.2894,
"mean_token_accuracy": 0.19697232395410538,
"num_tokens": 9845593.0,
"step": 4300
},
{
"entropy": 5.402709579467773,
"epoch": 0.41354466858789624,
"grad_norm": 0.9296875,
"learning_rate": 0.0004988604527612685,
"loss": 5.1841,
"mean_token_accuracy": 0.21107767671346664,
"num_tokens": 9856763.0,
"step": 4305
},
{
"entropy": 5.507501029968262,
"epoch": 0.4140249759846302,
"grad_norm": 0.98046875,
"learning_rate": 0.0004988570040840205,
"loss": 5.4395,
"mean_token_accuracy": 0.18462781310081483,
"num_tokens": 9869528.0,
"step": 4310
},
{
"entropy": 5.452263307571411,
"epoch": 0.41450528338136405,
"grad_norm": 1.046875,
"learning_rate": 0.0004988535502094774,
"loss": 5.3162,
"mean_token_accuracy": 0.19700846821069717,
"num_tokens": 9881170.0,
"step": 4315
},
{
"entropy": 5.508918333053589,
"epoch": 0.414985590778098,
"grad_norm": 1.0,
"learning_rate": 0.0004988500911377198,
"loss": 5.4062,
"mean_token_accuracy": 0.18960152417421341,
"num_tokens": 9893119.0,
"step": 4320
},
{
"entropy": 5.370126962661743,
"epoch": 0.41546589817483187,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988466268688276,
"loss": 5.2495,
"mean_token_accuracy": 0.20651557743549348,
"num_tokens": 9905339.0,
"step": 4325
},
{
"entropy": 5.523185920715332,
"epoch": 0.4159462055715658,
"grad_norm": 0.9375,
"learning_rate": 0.0004988431574028814,
"loss": 5.353,
"mean_token_accuracy": 0.19940332025289537,
"num_tokens": 9917500.0,
"step": 4330
},
{
"entropy": 5.49964451789856,
"epoch": 0.4164265129682997,
"grad_norm": 1.1015625,
"learning_rate": 0.0004988396827399618,
"loss": 5.4182,
"mean_token_accuracy": 0.1904998764395714,
"num_tokens": 9929667.0,
"step": 4335
},
{
"entropy": 5.51170916557312,
"epoch": 0.4169068203650336,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988362028801495,
"loss": 5.3248,
"mean_token_accuracy": 0.19802108108997346,
"num_tokens": 9941102.0,
"step": 4340
},
{
"entropy": 5.4384832859039305,
"epoch": 0.4173871277617675,
"grad_norm": 0.921875,
"learning_rate": 0.0004988327178235253,
"loss": 5.2273,
"mean_token_accuracy": 0.20404383093118666,
"num_tokens": 9951986.0,
"step": 4345
},
{
"entropy": 5.39904203414917,
"epoch": 0.41786743515850144,
"grad_norm": 0.89453125,
"learning_rate": 0.0004988292275701699,
"loss": 5.2485,
"mean_token_accuracy": 0.1993297517299652,
"num_tokens": 9964486.0,
"step": 4350
},
{
"entropy": 5.455586671829224,
"epoch": 0.4183477425552353,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988257321201646,
"loss": 5.343,
"mean_token_accuracy": 0.19409458339214325,
"num_tokens": 9975909.0,
"step": 4355
},
{
"entropy": 5.496417570114136,
"epoch": 0.41882804995196926,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988222314735902,
"loss": 5.3475,
"mean_token_accuracy": 0.19777753502130507,
"num_tokens": 9986951.0,
"step": 4360
},
{
"entropy": 5.5789155006408695,
"epoch": 0.41930835734870314,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988187256305284,
"loss": 5.4507,
"mean_token_accuracy": 0.1878629818558693,
"num_tokens": 9999234.0,
"step": 4365
},
{
"entropy": 5.499477052688599,
"epoch": 0.4197886647454371,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988152145910603,
"loss": 5.3196,
"mean_token_accuracy": 0.20171295255422592,
"num_tokens": 10010178.0,
"step": 4370
},
{
"entropy": 5.425793218612671,
"epoch": 0.420268972142171,
"grad_norm": 1.0546875,
"learning_rate": 0.0004988116983552675,
"loss": 5.264,
"mean_token_accuracy": 0.19647194296121598,
"num_tokens": 10021183.0,
"step": 4375
},
{
"entropy": 5.567107248306274,
"epoch": 0.4207492795389049,
"grad_norm": 0.921875,
"learning_rate": 0.0004988081769232317,
"loss": 5.5645,
"mean_token_accuracy": 0.18234878778457642,
"num_tokens": 10033686.0,
"step": 4380
},
{
"entropy": 5.449434423446656,
"epoch": 0.42122958693563883,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988046502950346,
"loss": 5.2652,
"mean_token_accuracy": 0.1972568526864052,
"num_tokens": 10045923.0,
"step": 4385
},
{
"entropy": 5.306301689147949,
"epoch": 0.4217098943323727,
"grad_norm": 0.94921875,
"learning_rate": 0.000498801118470758,
"loss": 5.1858,
"mean_token_accuracy": 0.19959155470132828,
"num_tokens": 10057196.0,
"step": 4390
},
{
"entropy": 5.449332141876221,
"epoch": 0.42219020172910665,
"grad_norm": 1.015625,
"learning_rate": 0.000498797581450484,
"loss": 5.3584,
"mean_token_accuracy": 0.19375586211681367,
"num_tokens": 10069655.0,
"step": 4395
},
{
"entropy": 5.511579084396362,
"epoch": 0.42267050912584053,
"grad_norm": 1.1484375,
"learning_rate": 0.0004987940392342948,
"loss": 5.2428,
"mean_token_accuracy": 0.19924264699220656,
"num_tokens": 10080876.0,
"step": 4400
},
{
"entropy": 5.482829475402832,
"epoch": 0.42315081652257447,
"grad_norm": 0.9375,
"learning_rate": 0.0004987904918222726,
"loss": 5.3524,
"mean_token_accuracy": 0.18975469768047332,
"num_tokens": 10091986.0,
"step": 4405
},
{
"entropy": 5.5305768013000485,
"epoch": 0.42363112391930835,
"grad_norm": 1.03125,
"learning_rate": 0.0004987869392144996,
"loss": 5.4321,
"mean_token_accuracy": 0.19329569041728972,
"num_tokens": 10104027.0,
"step": 4410
},
{
"entropy": 5.5070350646972654,
"epoch": 0.4241114313160423,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987833814110584,
"loss": 5.3005,
"mean_token_accuracy": 0.19417518377304077,
"num_tokens": 10114665.0,
"step": 4415
},
{
"entropy": 5.434443378448487,
"epoch": 0.42459173871277617,
"grad_norm": 0.93359375,
"learning_rate": 0.0004987798184120316,
"loss": 5.312,
"mean_token_accuracy": 0.19974145740270616,
"num_tokens": 10126032.0,
"step": 4420
},
{
"entropy": 5.545246362686157,
"epoch": 0.4250720461095101,
"grad_norm": 0.97265625,
"learning_rate": 0.0004987762502175018,
"loss": 5.3715,
"mean_token_accuracy": 0.19160096794366838,
"num_tokens": 10137256.0,
"step": 4425
},
{
"entropy": 5.383159017562866,
"epoch": 0.425552353506244,
"grad_norm": 0.98046875,
"learning_rate": 0.000498772676827552,
"loss": 5.2353,
"mean_token_accuracy": 0.1996947318315506,
"num_tokens": 10149445.0,
"step": 4430
},
{
"entropy": 5.5012122631073,
"epoch": 0.4260326609029779,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987690982422652,
"loss": 5.3886,
"mean_token_accuracy": 0.19004785567522048,
"num_tokens": 10161607.0,
"step": 4435
},
{
"entropy": 5.553760147094726,
"epoch": 0.4265129682997118,
"grad_norm": 1.0234375,
"learning_rate": 0.0004987655144617243,
"loss": 5.3855,
"mean_token_accuracy": 0.1945369154214859,
"num_tokens": 10173184.0,
"step": 4440
},
{
"entropy": 5.452174377441406,
"epoch": 0.42699327569644574,
"grad_norm": 0.94140625,
"learning_rate": 0.0004987619254860126,
"loss": 5.2738,
"mean_token_accuracy": 0.2024209216237068,
"num_tokens": 10184617.0,
"step": 4445
},
{
"entropy": 5.451268672943115,
"epoch": 0.4274735830931796,
"grad_norm": 0.9921875,
"learning_rate": 0.0004987583313152134,
"loss": 5.2866,
"mean_token_accuracy": 0.1954964816570282,
"num_tokens": 10195608.0,
"step": 4450
},
{
"entropy": 5.481286334991455,
"epoch": 0.42795389048991356,
"grad_norm": 0.94921875,
"learning_rate": 0.0004987547319494104,
"loss": 5.3806,
"mean_token_accuracy": 0.19195548743009566,
"num_tokens": 10206763.0,
"step": 4455
},
{
"entropy": 5.507647132873535,
"epoch": 0.42843419788664744,
"grad_norm": 0.92578125,
"learning_rate": 0.0004987511273886867,
"loss": 5.3352,
"mean_token_accuracy": 0.19549722224473953,
"num_tokens": 10218714.0,
"step": 4460
},
{
"entropy": 5.460297584533691,
"epoch": 0.4289145052833814,
"grad_norm": 0.98828125,
"learning_rate": 0.0004987475176331263,
"loss": 5.3499,
"mean_token_accuracy": 0.19259987473487855,
"num_tokens": 10229902.0,
"step": 4465
},
{
"entropy": 5.434809589385987,
"epoch": 0.42939481268011526,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987439026828129,
"loss": 5.2242,
"mean_token_accuracy": 0.19963788241147995,
"num_tokens": 10241578.0,
"step": 4470
},
{
"entropy": 5.37636399269104,
"epoch": 0.4298751200768492,
"grad_norm": 1.109375,
"learning_rate": 0.0004987402825378305,
"loss": 5.1823,
"mean_token_accuracy": 0.20253763049840928,
"num_tokens": 10252109.0,
"step": 4475
},
{
"entropy": 5.490766906738282,
"epoch": 0.4303554274735831,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987366571982631,
"loss": 5.3707,
"mean_token_accuracy": 0.19313031136989595,
"num_tokens": 10263357.0,
"step": 4480
},
{
"entropy": 5.491276741027832,
"epoch": 0.430835734870317,
"grad_norm": 0.95703125,
"learning_rate": 0.0004987330266641948,
"loss": 5.3569,
"mean_token_accuracy": 0.1924482449889183,
"num_tokens": 10275536.0,
"step": 4485
},
{
"entropy": 5.56570987701416,
"epoch": 0.4313160422670509,
"grad_norm": 0.953125,
"learning_rate": 0.0004987293909357101,
"loss": 5.3442,
"mean_token_accuracy": 0.19930241852998734,
"num_tokens": 10286901.0,
"step": 4490
},
{
"entropy": 5.362359619140625,
"epoch": 0.43179634966378483,
"grad_norm": 0.921875,
"learning_rate": 0.0004987257500128933,
"loss": 5.2413,
"mean_token_accuracy": 0.19796304106712342,
"num_tokens": 10298961.0,
"step": 4495
},
{
"entropy": 5.4794525623321535,
"epoch": 0.4322766570605187,
"grad_norm": 0.94921875,
"learning_rate": 0.0004987221038958288,
"loss": 5.3873,
"mean_token_accuracy": 0.19520987272262574,
"num_tokens": 10310911.0,
"step": 4500
},
{
"entropy": 5.569372749328613,
"epoch": 0.43275696445725265,
"grad_norm": 0.9453125,
"learning_rate": 0.0004987184525846015,
"loss": 5.372,
"mean_token_accuracy": 0.19102310240268708,
"num_tokens": 10322267.0,
"step": 4505
},
{
"entropy": 5.460135221481323,
"epoch": 0.4332372718539865,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987147960792958,
"loss": 5.3853,
"mean_token_accuracy": 0.19574257284402846,
"num_tokens": 10335111.0,
"step": 4510
},
{
"entropy": 5.537425756454468,
"epoch": 0.43371757925072046,
"grad_norm": 0.94921875,
"learning_rate": 0.0004987111343799971,
"loss": 5.328,
"mean_token_accuracy": 0.19893868863582612,
"num_tokens": 10345672.0,
"step": 4515
},
{
"entropy": 5.535245752334594,
"epoch": 0.43419788664745435,
"grad_norm": 0.99609375,
"learning_rate": 0.00049870746748679,
"loss": 5.3078,
"mean_token_accuracy": 0.19822679609060287,
"num_tokens": 10357369.0,
"step": 4520
},
{
"entropy": 5.475355434417724,
"epoch": 0.4346781940441883,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987037953997598,
"loss": 5.3352,
"mean_token_accuracy": 0.19392889589071274,
"num_tokens": 10368842.0,
"step": 4525
},
{
"entropy": 5.417094087600708,
"epoch": 0.43515850144092216,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987001181189918,
"loss": 5.2853,
"mean_token_accuracy": 0.19415317475795746,
"num_tokens": 10380096.0,
"step": 4530
},
{
"entropy": 5.3792015552520756,
"epoch": 0.4356388088376561,
"grad_norm": 0.94140625,
"learning_rate": 0.0004986964356445713,
"loss": 5.3172,
"mean_token_accuracy": 0.19834728091955184,
"num_tokens": 10391996.0,
"step": 4535
},
{
"entropy": 5.506541299819946,
"epoch": 0.43611911623439004,
"grad_norm": 1.046875,
"learning_rate": 0.0004986927479765837,
"loss": 5.2618,
"mean_token_accuracy": 0.1930739849805832,
"num_tokens": 10403607.0,
"step": 4540
},
{
"entropy": 5.444096088409424,
"epoch": 0.4365994236311239,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986890551151148,
"loss": 5.2746,
"mean_token_accuracy": 0.1976367026567459,
"num_tokens": 10413580.0,
"step": 4545
},
{
"entropy": 5.40678129196167,
"epoch": 0.43707973102785785,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986853570602503,
"loss": 5.3125,
"mean_token_accuracy": 0.1981777012348175,
"num_tokens": 10426456.0,
"step": 4550
},
{
"entropy": 5.5582586288452145,
"epoch": 0.43756003842459174,
"grad_norm": 0.921875,
"learning_rate": 0.0004986816538120758,
"loss": 5.343,
"mean_token_accuracy": 0.18815538436174392,
"num_tokens": 10438869.0,
"step": 4555
},
{
"entropy": 5.448095989227295,
"epoch": 0.43804034582132567,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986779453706778,
"loss": 5.3537,
"mean_token_accuracy": 0.18729300647974015,
"num_tokens": 10450672.0,
"step": 4560
},
{
"entropy": 5.459218215942383,
"epoch": 0.43852065321805955,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986742317361419,
"loss": 5.261,
"mean_token_accuracy": 0.20304668992757796,
"num_tokens": 10461890.0,
"step": 4565
},
{
"entropy": 5.565139579772949,
"epoch": 0.4390009606147935,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986705129085546,
"loss": 5.3919,
"mean_token_accuracy": 0.18603334277868272,
"num_tokens": 10473866.0,
"step": 4570
},
{
"entropy": 5.509669589996338,
"epoch": 0.43948126801152737,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986667888880021,
"loss": 5.2954,
"mean_token_accuracy": 0.1933439001441002,
"num_tokens": 10484889.0,
"step": 4575
},
{
"entropy": 5.42793641090393,
"epoch": 0.4399615754082613,
"grad_norm": 0.97265625,
"learning_rate": 0.0004986630596745709,
"loss": 5.3419,
"mean_token_accuracy": 0.19692840725183486,
"num_tokens": 10496108.0,
"step": 4580
},
{
"entropy": 5.463746356964111,
"epoch": 0.4404418828049952,
"grad_norm": 1.1171875,
"learning_rate": 0.0004986593252683477,
"loss": 5.2975,
"mean_token_accuracy": 0.19601958990097046,
"num_tokens": 10505472.0,
"step": 4585
},
{
"entropy": 5.345721244812012,
"epoch": 0.4409221902017291,
"grad_norm": 1.015625,
"learning_rate": 0.0004986555856694191,
"loss": 5.2075,
"mean_token_accuracy": 0.20081722289323806,
"num_tokens": 10516954.0,
"step": 4590
},
{
"entropy": 5.5342426776885985,
"epoch": 0.441402497598463,
"grad_norm": 0.96875,
"learning_rate": 0.0004986518408778718,
"loss": 5.3249,
"mean_token_accuracy": 0.1927899092435837,
"num_tokens": 10528166.0,
"step": 4595
},
{
"entropy": 5.422142553329468,
"epoch": 0.44188280499519694,
"grad_norm": 1.109375,
"learning_rate": 0.0004986480908937929,
"loss": 5.2405,
"mean_token_accuracy": 0.19853255301713943,
"num_tokens": 10538112.0,
"step": 4600
},
{
"entropy": 5.489729166030884,
"epoch": 0.4423631123919308,
"grad_norm": 1.015625,
"learning_rate": 0.0004986443357172695,
"loss": 5.3827,
"mean_token_accuracy": 0.19151799231767655,
"num_tokens": 10549888.0,
"step": 4605
},
{
"entropy": 5.626604223251343,
"epoch": 0.44284341978866476,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986405753483887,
"loss": 5.4524,
"mean_token_accuracy": 0.1831491395831108,
"num_tokens": 10561710.0,
"step": 4610
},
{
"entropy": 5.424169254302979,
"epoch": 0.44332372718539864,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986368097872377,
"loss": 5.3121,
"mean_token_accuracy": 0.1889724239706993,
"num_tokens": 10574564.0,
"step": 4615
},
{
"entropy": 5.4898937225341795,
"epoch": 0.4438040345821326,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986330390339042,
"loss": 5.2872,
"mean_token_accuracy": 0.1978645905852318,
"num_tokens": 10586639.0,
"step": 4620
},
{
"entropy": 5.4653246879577635,
"epoch": 0.44428434197886646,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986292630884755,
"loss": 5.2935,
"mean_token_accuracy": 0.19615231901407243,
"num_tokens": 10598730.0,
"step": 4625
},
{
"entropy": 5.420555067062378,
"epoch": 0.4447646493756004,
"grad_norm": 0.98046875,
"learning_rate": 0.0004986254819510393,
"loss": 5.2016,
"mean_token_accuracy": 0.21116500198841096,
"num_tokens": 10610352.0,
"step": 4630
},
{
"entropy": 5.394788074493408,
"epoch": 0.4452449567723343,
"grad_norm": 0.96875,
"learning_rate": 0.0004986216956216835,
"loss": 5.3007,
"mean_token_accuracy": 0.19723331481218337,
"num_tokens": 10621951.0,
"step": 4635
},
{
"entropy": 5.444929742813111,
"epoch": 0.4457252641690682,
"grad_norm": 1.078125,
"learning_rate": 0.000498617904100496,
"loss": 5.2438,
"mean_token_accuracy": 0.20000688582658768,
"num_tokens": 10633207.0,
"step": 4640
},
{
"entropy": 5.516901731491089,
"epoch": 0.4462055715658021,
"grad_norm": 0.921875,
"learning_rate": 0.0004986141073875646,
"loss": 5.3495,
"mean_token_accuracy": 0.18970742523670198,
"num_tokens": 10645853.0,
"step": 4645
},
{
"entropy": 5.386112785339355,
"epoch": 0.44668587896253603,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986103054829779,
"loss": 5.2482,
"mean_token_accuracy": 0.20123800486326218,
"num_tokens": 10656892.0,
"step": 4650
},
{
"entropy": 5.392650985717774,
"epoch": 0.4471661863592699,
"grad_norm": 0.90625,
"learning_rate": 0.0004986064983868237,
"loss": 5.2359,
"mean_token_accuracy": 0.19400574266910553,
"num_tokens": 10670110.0,
"step": 4655
},
{
"entropy": 5.4809633731842045,
"epoch": 0.44764649375600385,
"grad_norm": 1.015625,
"learning_rate": 0.0004986026860991906,
"loss": 5.366,
"mean_token_accuracy": 0.196006740629673,
"num_tokens": 10681255.0,
"step": 4660
},
{
"entropy": 5.473309946060181,
"epoch": 0.44812680115273773,
"grad_norm": 1.171875,
"learning_rate": 0.0004985988686201672,
"loss": 5.4205,
"mean_token_accuracy": 0.19332450479269028,
"num_tokens": 10692631.0,
"step": 4665
},
{
"entropy": 5.477552318572998,
"epoch": 0.44860710854947167,
"grad_norm": 0.9375,
"learning_rate": 0.0004985950459498419,
"loss": 5.2573,
"mean_token_accuracy": 0.20014705508947372,
"num_tokens": 10704880.0,
"step": 4670
},
{
"entropy": 5.45638484954834,
"epoch": 0.44908741594620555,
"grad_norm": 1.0078125,
"learning_rate": 0.0004985912180883037,
"loss": 5.2484,
"mean_token_accuracy": 0.20326420962810515,
"num_tokens": 10715561.0,
"step": 4675
},
{
"entropy": 5.385212564468384,
"epoch": 0.4495677233429395,
"grad_norm": 1.015625,
"learning_rate": 0.0004985873850356411,
"loss": 5.2633,
"mean_token_accuracy": 0.19535834789276124,
"num_tokens": 10727232.0,
"step": 4680
},
{
"entropy": 5.416689777374268,
"epoch": 0.45004803073967337,
"grad_norm": 0.9765625,
"learning_rate": 0.0004985835467919436,
"loss": 5.2857,
"mean_token_accuracy": 0.2015791967511177,
"num_tokens": 10739404.0,
"step": 4685
},
{
"entropy": 5.449459791183472,
"epoch": 0.4505283381364073,
"grad_norm": 0.98046875,
"learning_rate": 0.0004985797033572999,
"loss": 5.3113,
"mean_token_accuracy": 0.19386572688817977,
"num_tokens": 10751948.0,
"step": 4690
},
{
"entropy": 5.434033584594727,
"epoch": 0.4510086455331412,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985758547317994,
"loss": 5.274,
"mean_token_accuracy": 0.19730297327041627,
"num_tokens": 10764611.0,
"step": 4695
},
{
"entropy": 5.465791463851929,
"epoch": 0.4514889529298751,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985720009155315,
"loss": 5.3065,
"mean_token_accuracy": 0.19460007399320603,
"num_tokens": 10775954.0,
"step": 4700
},
{
"entropy": 5.45864953994751,
"epoch": 0.45196926032660906,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985681419085856,
"loss": 5.319,
"mean_token_accuracy": 0.18940264582633973,
"num_tokens": 10788723.0,
"step": 4705
},
{
"entropy": 5.474675464630127,
"epoch": 0.45244956772334294,
"grad_norm": 1.0390625,
"learning_rate": 0.0004985642777110513,
"loss": 5.3058,
"mean_token_accuracy": 0.19875997453927993,
"num_tokens": 10799879.0,
"step": 4710
},
{
"entropy": 5.400967597961426,
"epoch": 0.4529298751200769,
"grad_norm": 0.890625,
"learning_rate": 0.0004985604083230183,
"loss": 5.2621,
"mean_token_accuracy": 0.1981558918952942,
"num_tokens": 10811838.0,
"step": 4715
},
{
"entropy": 5.49187388420105,
"epoch": 0.45341018251681076,
"grad_norm": 1.078125,
"learning_rate": 0.0004985565337445765,
"loss": 5.2807,
"mean_token_accuracy": 0.1971256822347641,
"num_tokens": 10822910.0,
"step": 4720
},
{
"entropy": 5.49492335319519,
"epoch": 0.4538904899135447,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985526539758158,
"loss": 5.3446,
"mean_token_accuracy": 0.19619156718254088,
"num_tokens": 10835344.0,
"step": 4725
},
{
"entropy": 5.4535280704498295,
"epoch": 0.4543707973102786,
"grad_norm": 0.9140625,
"learning_rate": 0.0004985487690168263,
"loss": 5.309,
"mean_token_accuracy": 0.19785993695259094,
"num_tokens": 10846043.0,
"step": 4730
},
{
"entropy": 5.456455278396606,
"epoch": 0.4548511047070125,
"grad_norm": 1.015625,
"learning_rate": 0.000498544878867698,
"loss": 5.2462,
"mean_token_accuracy": 0.20247492641210557,
"num_tokens": 10857783.0,
"step": 4735
},
{
"entropy": 5.465985870361328,
"epoch": 0.4553314121037464,
"grad_norm": 0.94140625,
"learning_rate": 0.0004985409835285215,
"loss": 5.3029,
"mean_token_accuracy": 0.1979512870311737,
"num_tokens": 10870527.0,
"step": 4740
},
{
"entropy": 5.441158056259155,
"epoch": 0.45581171950048033,
"grad_norm": 0.9765625,
"learning_rate": 0.0004985370829993873,
"loss": 5.2907,
"mean_token_accuracy": 0.2029913693666458,
"num_tokens": 10882285.0,
"step": 4745
},
{
"entropy": 5.489823675155639,
"epoch": 0.4562920268972142,
"grad_norm": 0.9921875,
"learning_rate": 0.0004985331772803857,
"loss": 5.3154,
"mean_token_accuracy": 0.19783458560705186,
"num_tokens": 10895319.0,
"step": 4750
},
{
"entropy": 5.472352361679077,
"epoch": 0.45677233429394815,
"grad_norm": 0.93359375,
"learning_rate": 0.0004985292663716074,
"loss": 5.2957,
"mean_token_accuracy": 0.19923655539751053,
"num_tokens": 10906253.0,
"step": 4755
},
{
"entropy": 5.335994291305542,
"epoch": 0.457252641690682,
"grad_norm": 0.8828125,
"learning_rate": 0.0004985253502731435,
"loss": 5.1855,
"mean_token_accuracy": 0.2086847722530365,
"num_tokens": 10918197.0,
"step": 4760
},
{
"entropy": 5.443401622772217,
"epoch": 0.45773294908741596,
"grad_norm": 0.9375,
"learning_rate": 0.0004985214289850845,
"loss": 5.3816,
"mean_token_accuracy": 0.18888968527317046,
"num_tokens": 10930771.0,
"step": 4765
},
{
"entropy": 5.4748773097991945,
"epoch": 0.45821325648414984,
"grad_norm": 0.8671875,
"learning_rate": 0.0004985175025075217,
"loss": 5.2834,
"mean_token_accuracy": 0.1973588064312935,
"num_tokens": 10942759.0,
"step": 4770
},
{
"entropy": 5.6522383213043215,
"epoch": 0.4586935638808838,
"grad_norm": 0.9765625,
"learning_rate": 0.0004985135708405462,
"loss": 5.4848,
"mean_token_accuracy": 0.18645619004964828,
"num_tokens": 10953557.0,
"step": 4775
},
{
"entropy": 5.401769304275513,
"epoch": 0.45917387127761766,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985096339842493,
"loss": 5.2411,
"mean_token_accuracy": 0.20662152916193008,
"num_tokens": 10963142.0,
"step": 4780
},
{
"entropy": 5.379817962646484,
"epoch": 0.4596541786743516,
"grad_norm": 1.0234375,
"learning_rate": 0.0004985056919387224,
"loss": 5.2184,
"mean_token_accuracy": 0.20357681661844254,
"num_tokens": 10974321.0,
"step": 4785
},
{
"entropy": 5.509200239181519,
"epoch": 0.4601344860710855,
"grad_norm": 1.0703125,
"learning_rate": 0.0004985017447040569,
"loss": 5.4133,
"mean_token_accuracy": 0.19487691521644593,
"num_tokens": 10985524.0,
"step": 4790
},
{
"entropy": 5.494694328308105,
"epoch": 0.4606147934678194,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984977922803447,
"loss": 5.3063,
"mean_token_accuracy": 0.19958257526159287,
"num_tokens": 10997606.0,
"step": 4795
},
{
"entropy": 5.412304258346557,
"epoch": 0.4610951008645533,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984938346676772,
"loss": 5.3013,
"mean_token_accuracy": 0.1943564847111702,
"num_tokens": 11010692.0,
"step": 4800
},
{
"entropy": 5.417695474624634,
"epoch": 0.46157540826128723,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984898718661468,
"loss": 5.2374,
"mean_token_accuracy": 0.20364998728036882,
"num_tokens": 11022517.0,
"step": 4805
},
{
"entropy": 5.4359543800354,
"epoch": 0.4620557156580211,
"grad_norm": 1.1015625,
"learning_rate": 0.0004984859038758451,
"loss": 5.248,
"mean_token_accuracy": 0.1995931401848793,
"num_tokens": 11033141.0,
"step": 4810
},
{
"entropy": 5.396144914627075,
"epoch": 0.46253602305475505,
"grad_norm": 0.984375,
"learning_rate": 0.0004984819306968642,
"loss": 5.2408,
"mean_token_accuracy": 0.19817344099283218,
"num_tokens": 11044619.0,
"step": 4815
},
{
"entropy": 5.519974994659424,
"epoch": 0.46301633045148893,
"grad_norm": 0.90234375,
"learning_rate": 0.0004984779523292966,
"loss": 5.2873,
"mean_token_accuracy": 0.20086241513490677,
"num_tokens": 11055934.0,
"step": 4820
},
{
"entropy": 5.440757846832275,
"epoch": 0.46349663784822287,
"grad_norm": 0.94921875,
"learning_rate": 0.0004984739687732345,
"loss": 5.1751,
"mean_token_accuracy": 0.20307007431983948,
"num_tokens": 11066203.0,
"step": 4825
},
{
"entropy": 5.272681379318238,
"epoch": 0.46397694524495675,
"grad_norm": 0.87890625,
"learning_rate": 0.0004984699800287705,
"loss": 5.13,
"mean_token_accuracy": 0.2047649070620537,
"num_tokens": 11079664.0,
"step": 4830
},
{
"entropy": 5.312569189071655,
"epoch": 0.4644572526416907,
"grad_norm": 0.94921875,
"learning_rate": 0.000498465986095997,
"loss": 5.1823,
"mean_token_accuracy": 0.20340044498443605,
"num_tokens": 11091186.0,
"step": 4835
},
{
"entropy": 5.447326278686523,
"epoch": 0.46493756003842457,
"grad_norm": 1.0703125,
"learning_rate": 0.0004984619869750069,
"loss": 5.3092,
"mean_token_accuracy": 0.19724527150392532,
"num_tokens": 11102710.0,
"step": 4840
},
{
"entropy": 5.387120294570923,
"epoch": 0.4654178674351585,
"grad_norm": 0.9453125,
"learning_rate": 0.000498457982665893,
"loss": 5.2046,
"mean_token_accuracy": 0.2028914585709572,
"num_tokens": 11114746.0,
"step": 4845
},
{
"entropy": 5.423678159713745,
"epoch": 0.4658981748318924,
"grad_norm": 0.95703125,
"learning_rate": 0.0004984539731687483,
"loss": 5.2561,
"mean_token_accuracy": 0.19754694998264313,
"num_tokens": 11126572.0,
"step": 4850
},
{
"entropy": 5.399871730804444,
"epoch": 0.4663784822286263,
"grad_norm": 1.0390625,
"learning_rate": 0.0004984499584836659,
"loss": 5.1642,
"mean_token_accuracy": 0.20589766055345535,
"num_tokens": 11137830.0,
"step": 4855
},
{
"entropy": 5.381054067611695,
"epoch": 0.4668587896253602,
"grad_norm": 0.96484375,
"learning_rate": 0.000498445938610739,
"loss": 5.1978,
"mean_token_accuracy": 0.2022456184029579,
"num_tokens": 11148860.0,
"step": 4860
},
{
"entropy": 5.438592147827149,
"epoch": 0.46733909702209414,
"grad_norm": 0.97265625,
"learning_rate": 0.0004984419135500608,
"loss": 5.3228,
"mean_token_accuracy": 0.19238598942756652,
"num_tokens": 11161311.0,
"step": 4865
},
{
"entropy": 5.425858068466186,
"epoch": 0.4678194044188281,
"grad_norm": 0.984375,
"learning_rate": 0.0004984378833017249,
"loss": 5.2317,
"mean_token_accuracy": 0.19774759262800218,
"num_tokens": 11173124.0,
"step": 4870
},
{
"entropy": 5.395693683624268,
"epoch": 0.46829971181556196,
"grad_norm": 1.03125,
"learning_rate": 0.0004984338478658248,
"loss": 5.3059,
"mean_token_accuracy": 0.1966658964753151,
"num_tokens": 11184879.0,
"step": 4875
},
{
"entropy": 5.4837006568908695,
"epoch": 0.4687800192122959,
"grad_norm": 0.921875,
"learning_rate": 0.0004984298072424542,
"loss": 5.3062,
"mean_token_accuracy": 0.1932589590549469,
"num_tokens": 11196243.0,
"step": 4880
},
{
"entropy": 5.432374286651611,
"epoch": 0.4692603266090298,
"grad_norm": 1.015625,
"learning_rate": 0.000498425761431707,
"loss": 5.1681,
"mean_token_accuracy": 0.20986756533384324,
"num_tokens": 11207485.0,
"step": 4885
},
{
"entropy": 5.388256454467774,
"epoch": 0.4697406340057637,
"grad_norm": 0.94921875,
"learning_rate": 0.000498421710433677,
"loss": 5.2099,
"mean_token_accuracy": 0.19842060953378676,
"num_tokens": 11219891.0,
"step": 4890
},
{
"entropy": 5.479468727111817,
"epoch": 0.4702209414024976,
"grad_norm": 0.921875,
"learning_rate": 0.0004984176542484584,
"loss": 5.3116,
"mean_token_accuracy": 0.19674236327409744,
"num_tokens": 11231329.0,
"step": 4895
},
{
"entropy": 5.403811025619507,
"epoch": 0.47070124879923153,
"grad_norm": 1.0,
"learning_rate": 0.0004984135928761452,
"loss": 5.1932,
"mean_token_accuracy": 0.20610021948814392,
"num_tokens": 11241367.0,
"step": 4900
},
{
"entropy": 5.443480825424194,
"epoch": 0.4711815561959654,
"grad_norm": 0.9375,
"learning_rate": 0.0004984095263168317,
"loss": 5.2832,
"mean_token_accuracy": 0.19220978170633315,
"num_tokens": 11254532.0,
"step": 4905
},
{
"entropy": 5.511846876144409,
"epoch": 0.47166186359269935,
"grad_norm": 0.96875,
"learning_rate": 0.0004984054545706124,
"loss": 5.3603,
"mean_token_accuracy": 0.18875951766967775,
"num_tokens": 11265223.0,
"step": 4910
},
{
"entropy": 5.373107099533081,
"epoch": 0.47214217098943323,
"grad_norm": 0.9375,
"learning_rate": 0.000498401377637582,
"loss": 5.3021,
"mean_token_accuracy": 0.19637542366981506,
"num_tokens": 11278228.0,
"step": 4915
},
{
"entropy": 5.564932346343994,
"epoch": 0.47262247838616717,
"grad_norm": 0.93359375,
"learning_rate": 0.000498397295517835,
"loss": 5.3949,
"mean_token_accuracy": 0.1917308121919632,
"num_tokens": 11289654.0,
"step": 4920
},
{
"entropy": 5.421516752243042,
"epoch": 0.47310278578290105,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983932082114659,
"loss": 5.1726,
"mean_token_accuracy": 0.2053057834506035,
"num_tokens": 11301911.0,
"step": 4925
},
{
"entropy": 5.5105547428131105,
"epoch": 0.473583093179635,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983891157185699,
"loss": 5.3574,
"mean_token_accuracy": 0.191430202126503,
"num_tokens": 11312945.0,
"step": 4930
},
{
"entropy": 5.598050594329834,
"epoch": 0.47406340057636887,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983850180392421,
"loss": 5.4025,
"mean_token_accuracy": 0.18814169615507126,
"num_tokens": 11324126.0,
"step": 4935
},
{
"entropy": 5.4626157760620115,
"epoch": 0.4745437079731028,
"grad_norm": 0.96875,
"learning_rate": 0.0004983809151735775,
"loss": 5.3348,
"mean_token_accuracy": 0.1932709753513336,
"num_tokens": 11336395.0,
"step": 4940
},
{
"entropy": 5.45054349899292,
"epoch": 0.4750240153698367,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983768071216713,
"loss": 5.2371,
"mean_token_accuracy": 0.2002416417002678,
"num_tokens": 11347387.0,
"step": 4945
},
{
"entropy": 5.3878782272338865,
"epoch": 0.4755043227665706,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983726938836189,
"loss": 5.2352,
"mean_token_accuracy": 0.20208385437726975,
"num_tokens": 11358467.0,
"step": 4950
},
{
"entropy": 5.491963720321655,
"epoch": 0.4759846301633045,
"grad_norm": 0.9765625,
"learning_rate": 0.0004983685754595159,
"loss": 5.4037,
"mean_token_accuracy": 0.1896727964282036,
"num_tokens": 11370322.0,
"step": 4955
},
{
"entropy": 5.352085447311401,
"epoch": 0.47646493756003844,
"grad_norm": 0.9921875,
"learning_rate": 0.0004983644518494578,
"loss": 5.2004,
"mean_token_accuracy": 0.20571735352277756,
"num_tokens": 11381719.0,
"step": 4960
},
{
"entropy": 5.39383020401001,
"epoch": 0.4769452449567723,
"grad_norm": 1.046875,
"learning_rate": 0.0004983603230535403,
"loss": 5.2258,
"mean_token_accuracy": 0.19686990678310395,
"num_tokens": 11393561.0,
"step": 4965
},
{
"entropy": 5.495372152328491,
"epoch": 0.47742555235350626,
"grad_norm": 0.89453125,
"learning_rate": 0.0004983561890718594,
"loss": 5.3062,
"mean_token_accuracy": 0.19655826836824417,
"num_tokens": 11405411.0,
"step": 4970
},
{
"entropy": 5.526172113418579,
"epoch": 0.47790585975024014,
"grad_norm": 1.1171875,
"learning_rate": 0.000498352049904511,
"loss": 5.4063,
"mean_token_accuracy": 0.18656282126903534,
"num_tokens": 11417419.0,
"step": 4975
},
{
"entropy": 5.416673946380615,
"epoch": 0.4783861671469741,
"grad_norm": 1.046875,
"learning_rate": 0.0004983479055515914,
"loss": 5.2149,
"mean_token_accuracy": 0.19893506020307541,
"num_tokens": 11428145.0,
"step": 4980
},
{
"entropy": 5.351204967498779,
"epoch": 0.47886647454370795,
"grad_norm": 0.9375,
"learning_rate": 0.0004983437560131964,
"loss": 5.1681,
"mean_token_accuracy": 0.20236289799213408,
"num_tokens": 11439224.0,
"step": 4985
},
{
"entropy": 5.425208616256714,
"epoch": 0.4793467819404419,
"grad_norm": 0.97265625,
"learning_rate": 0.0004983396012894228,
"loss": 5.272,
"mean_token_accuracy": 0.19770564287900924,
"num_tokens": 11451731.0,
"step": 4990
},
{
"entropy": 5.458701133728027,
"epoch": 0.47982708933717577,
"grad_norm": 1.0234375,
"learning_rate": 0.0004983354413803666,
"loss": 5.2414,
"mean_token_accuracy": 0.20528912395238877,
"num_tokens": 11463058.0,
"step": 4995
},
{
"entropy": 5.543280839920044,
"epoch": 0.4803073967339097,
"grad_norm": 1.109375,
"learning_rate": 0.0004983312762861248,
"loss": 5.3528,
"mean_token_accuracy": 0.19466368705034257,
"num_tokens": 11472618.0,
"step": 5000
},
{
"entropy": 5.399996662139893,
"epoch": 0.4807877041306436,
"grad_norm": 0.984375,
"learning_rate": 0.0004983271060067939,
"loss": 5.2373,
"mean_token_accuracy": 0.1996273174881935,
"num_tokens": 11483114.0,
"step": 5005
},
{
"entropy": 5.363938045501709,
"epoch": 0.4812680115273775,
"grad_norm": 1.1328125,
"learning_rate": 0.0004983229305424707,
"loss": 5.1987,
"mean_token_accuracy": 0.20535318106412886,
"num_tokens": 11494281.0,
"step": 5010
},
{
"entropy": 5.440830373764038,
"epoch": 0.4817483189241114,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983187498932522,
"loss": 5.2907,
"mean_token_accuracy": 0.19693202525377274,
"num_tokens": 11505962.0,
"step": 5015
},
{
"entropy": 5.44342532157898,
"epoch": 0.48222862632084534,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983145640592354,
"loss": 5.3485,
"mean_token_accuracy": 0.18968168497085572,
"num_tokens": 11517558.0,
"step": 5020
},
{
"entropy": 5.45598030090332,
"epoch": 0.4827089337175792,
"grad_norm": 0.94921875,
"learning_rate": 0.0004983103730405176,
"loss": 5.3599,
"mean_token_accuracy": 0.19084265679121018,
"num_tokens": 11529184.0,
"step": 5025
},
{
"entropy": 5.452455759048462,
"epoch": 0.48318924111431316,
"grad_norm": 0.97265625,
"learning_rate": 0.000498306176837196,
"loss": 5.257,
"mean_token_accuracy": 0.1939294755458832,
"num_tokens": 11540727.0,
"step": 5030
},
{
"entropy": 5.3856573581695555,
"epoch": 0.48366954851104704,
"grad_norm": 0.953125,
"learning_rate": 0.0004983019754493681,
"loss": 5.1863,
"mean_token_accuracy": 0.20147235542535782,
"num_tokens": 11551510.0,
"step": 5035
},
{
"entropy": 5.441668701171875,
"epoch": 0.484149855907781,
"grad_norm": 0.98046875,
"learning_rate": 0.0004982977688771314,
"loss": 5.3278,
"mean_token_accuracy": 0.20095289647579193,
"num_tokens": 11563203.0,
"step": 5040
},
{
"entropy": 5.4081573486328125,
"epoch": 0.4846301633045149,
"grad_norm": 0.8828125,
"learning_rate": 0.0004982935571205835,
"loss": 5.2104,
"mean_token_accuracy": 0.20497542768716812,
"num_tokens": 11576013.0,
"step": 5045
},
{
"entropy": 5.323009014129639,
"epoch": 0.4851104707012488,
"grad_norm": 1.0703125,
"learning_rate": 0.0004982893401798223,
"loss": 5.1553,
"mean_token_accuracy": 0.21624571532011033,
"num_tokens": 11587535.0,
"step": 5050
},
{
"entropy": 5.364218139648438,
"epoch": 0.48559077809798273,
"grad_norm": 0.94140625,
"learning_rate": 0.0004982851180549456,
"loss": 5.2053,
"mean_token_accuracy": 0.20111326426267623,
"num_tokens": 11598487.0,
"step": 5055
},
{
"entropy": 5.324197387695312,
"epoch": 0.4860710854947166,
"grad_norm": 0.9296875,
"learning_rate": 0.0004982808907460515,
"loss": 5.091,
"mean_token_accuracy": 0.22296716421842575,
"num_tokens": 11609457.0,
"step": 5060
},
{
"entropy": 5.3482780933380125,
"epoch": 0.48655139289145055,
"grad_norm": 0.92578125,
"learning_rate": 0.0004982766582532382,
"loss": 5.1526,
"mean_token_accuracy": 0.2081102043390274,
"num_tokens": 11620251.0,
"step": 5065
},
{
"entropy": 5.358762311935425,
"epoch": 0.48703170028818443,
"grad_norm": 0.984375,
"learning_rate": 0.0004982724205766038,
"loss": 5.1492,
"mean_token_accuracy": 0.20766099393367768,
"num_tokens": 11630956.0,
"step": 5070
},
{
"entropy": 5.420529079437256,
"epoch": 0.48751200768491837,
"grad_norm": 0.8671875,
"learning_rate": 0.0004982681777162468,
"loss": 5.2004,
"mean_token_accuracy": 0.20143322944641112,
"num_tokens": 11642560.0,
"step": 5075
},
{
"entropy": 5.312218856811524,
"epoch": 0.48799231508165225,
"grad_norm": 1.03125,
"learning_rate": 0.0004982639296722657,
"loss": 5.1756,
"mean_token_accuracy": 0.20298496782779693,
"num_tokens": 11654050.0,
"step": 5080
},
{
"entropy": 5.3660177230834964,
"epoch": 0.4884726224783862,
"grad_norm": 1.0,
"learning_rate": 0.0004982596764447591,
"loss": 5.3127,
"mean_token_accuracy": 0.19897778779268266,
"num_tokens": 11664947.0,
"step": 5085
},
{
"entropy": 5.4209997177124025,
"epoch": 0.48895292987512007,
"grad_norm": 1.0,
"learning_rate": 0.0004982554180338258,
"loss": 5.2324,
"mean_token_accuracy": 0.20139615535736083,
"num_tokens": 11676927.0,
"step": 5090
},
{
"entropy": 5.530186462402344,
"epoch": 0.489433237271854,
"grad_norm": 1.0859375,
"learning_rate": 0.0004982511544395646,
"loss": 5.3406,
"mean_token_accuracy": 0.19309221357107162,
"num_tokens": 11688573.0,
"step": 5095
},
{
"entropy": 5.338684368133545,
"epoch": 0.4899135446685879,
"grad_norm": 1.046875,
"learning_rate": 0.0004982468856620745,
"loss": 5.2421,
"mean_token_accuracy": 0.1984547719359398,
"num_tokens": 11698704.0,
"step": 5100
},
{
"entropy": 5.307417392730713,
"epoch": 0.4903938520653218,
"grad_norm": 0.95703125,
"learning_rate": 0.0004982426117014545,
"loss": 5.1715,
"mean_token_accuracy": 0.20310387909412383,
"num_tokens": 11709466.0,
"step": 5105
},
{
"entropy": 5.384412813186645,
"epoch": 0.4908741594620557,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982383325578041,
"loss": 5.2733,
"mean_token_accuracy": 0.19685922116041182,
"num_tokens": 11721120.0,
"step": 5110
},
{
"entropy": 5.509656810760498,
"epoch": 0.49135446685878964,
"grad_norm": 0.94921875,
"learning_rate": 0.0004982340482312226,
"loss": 5.2637,
"mean_token_accuracy": 0.1910458594560623,
"num_tokens": 11732120.0,
"step": 5115
},
{
"entropy": 5.373399448394776,
"epoch": 0.4918347742555235,
"grad_norm": 0.953125,
"learning_rate": 0.0004982297587218092,
"loss": 5.1361,
"mean_token_accuracy": 0.20690827518701554,
"num_tokens": 11743501.0,
"step": 5120
},
{
"entropy": 5.366362190246582,
"epoch": 0.49231508165225746,
"grad_norm": 1.0,
"learning_rate": 0.0004982254640296637,
"loss": 5.2242,
"mean_token_accuracy": 0.20181894153356553,
"num_tokens": 11755051.0,
"step": 5125
},
{
"entropy": 5.41463942527771,
"epoch": 0.49279538904899134,
"grad_norm": 0.93359375,
"learning_rate": 0.0004982211641548857,
"loss": 5.3751,
"mean_token_accuracy": 0.1938163235783577,
"num_tokens": 11767663.0,
"step": 5130
},
{
"entropy": 5.473185110092163,
"epoch": 0.4932756964457253,
"grad_norm": 1.015625,
"learning_rate": 0.0004982168590975752,
"loss": 5.2429,
"mean_token_accuracy": 0.20339857041835785,
"num_tokens": 11778828.0,
"step": 5135
},
{
"entropy": 5.507049894332885,
"epoch": 0.49375600384245916,
"grad_norm": 0.9296875,
"learning_rate": 0.0004982125488578321,
"loss": 5.4109,
"mean_token_accuracy": 0.19114813357591628,
"num_tokens": 11790654.0,
"step": 5140
},
{
"entropy": 5.4209589004516605,
"epoch": 0.4942363112391931,
"grad_norm": 1.0625,
"learning_rate": 0.0004982082334357563,
"loss": 5.2036,
"mean_token_accuracy": 0.19969180077314377,
"num_tokens": 11801489.0,
"step": 5145
},
{
"entropy": 5.366079807281494,
"epoch": 0.494716618635927,
"grad_norm": 1.0078125,
"learning_rate": 0.0004982039128314481,
"loss": 5.2152,
"mean_token_accuracy": 0.20027052909135817,
"num_tokens": 11813818.0,
"step": 5150
},
{
"entropy": 5.347721862792969,
"epoch": 0.4951969260326609,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981995870450079,
"loss": 5.2119,
"mean_token_accuracy": 0.20178951025009156,
"num_tokens": 11824814.0,
"step": 5155
},
{
"entropy": 5.463880491256714,
"epoch": 0.4956772334293948,
"grad_norm": 0.9765625,
"learning_rate": 0.0004981952560765361,
"loss": 5.2617,
"mean_token_accuracy": 0.19569255411624908,
"num_tokens": 11836252.0,
"step": 5160
},
{
"entropy": 5.381760025024414,
"epoch": 0.49615754082612873,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981909199261331,
"loss": 5.1709,
"mean_token_accuracy": 0.20152270197868347,
"num_tokens": 11847715.0,
"step": 5165
},
{
"entropy": 5.327801704406738,
"epoch": 0.4966378482228626,
"grad_norm": 0.97265625,
"learning_rate": 0.0004981865785938998,
"loss": 5.1758,
"mean_token_accuracy": 0.20665301084518434,
"num_tokens": 11860309.0,
"step": 5170
},
{
"entropy": 5.362015056610107,
"epoch": 0.49711815561959655,
"grad_norm": 0.88671875,
"learning_rate": 0.0004981822320799367,
"loss": 5.1851,
"mean_token_accuracy": 0.20086576491594316,
"num_tokens": 11872569.0,
"step": 5175
},
{
"entropy": 5.359728670120239,
"epoch": 0.49759846301633043,
"grad_norm": 0.890625,
"learning_rate": 0.0004981778803843449,
"loss": 5.1747,
"mean_token_accuracy": 0.20249261260032653,
"num_tokens": 11884778.0,
"step": 5180
},
{
"entropy": 5.437938070297241,
"epoch": 0.49807877041306436,
"grad_norm": 0.96875,
"learning_rate": 0.0004981735235072256,
"loss": 5.2555,
"mean_token_accuracy": 0.20191961973905564,
"num_tokens": 11897324.0,
"step": 5185
},
{
"entropy": 5.447718381881714,
"epoch": 0.49855907780979825,
"grad_norm": 0.953125,
"learning_rate": 0.0004981691614486796,
"loss": 5.2865,
"mean_token_accuracy": 0.19850171953439713,
"num_tokens": 11909145.0,
"step": 5190
},
{
"entropy": 5.35249171257019,
"epoch": 0.4990393852065322,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981647942088084,
"loss": 5.1834,
"mean_token_accuracy": 0.20171126276254653,
"num_tokens": 11921021.0,
"step": 5195
},
{
"entropy": 5.519689655303955,
"epoch": 0.49951969260326606,
"grad_norm": 0.9921875,
"learning_rate": 0.0004981604217877135,
"loss": 5.3469,
"mean_token_accuracy": 0.1942518636584282,
"num_tokens": 11932565.0,
"step": 5200
},
{
"entropy": 5.374987077713013,
"epoch": 0.5,
"grad_norm": 0.9609375,
"learning_rate": 0.000498156044185496,
"loss": 5.262,
"mean_token_accuracy": 0.19995938688516618,
"num_tokens": 11943225.0,
"step": 5205
},
{
"entropy": 5.364169502258301,
"epoch": 0.5004803073967339,
"grad_norm": 0.984375,
"learning_rate": 0.0004981516614022579,
"loss": 5.2485,
"mean_token_accuracy": 0.2035745918750763,
"num_tokens": 11954821.0,
"step": 5210
},
{
"entropy": 5.43224196434021,
"epoch": 0.5009606147934679,
"grad_norm": 1.0390625,
"learning_rate": 0.0004981472734381008,
"loss": 5.1909,
"mean_token_accuracy": 0.20336654484272004,
"num_tokens": 11966090.0,
"step": 5215
},
{
"entropy": 5.41751708984375,
"epoch": 0.5014409221902018,
"grad_norm": 0.984375,
"learning_rate": 0.0004981428802931267,
"loss": 5.2322,
"mean_token_accuracy": 0.19972468316555023,
"num_tokens": 11977410.0,
"step": 5220
},
{
"entropy": 5.4079958438873295,
"epoch": 0.5019212295869356,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981384819674375,
"loss": 5.2174,
"mean_token_accuracy": 0.19654285460710524,
"num_tokens": 11989119.0,
"step": 5225
},
{
"entropy": 5.499818134307861,
"epoch": 0.5024015369836695,
"grad_norm": 0.91015625,
"learning_rate": 0.0004981340784611354,
"loss": 5.3181,
"mean_token_accuracy": 0.19646679908037185,
"num_tokens": 12000165.0,
"step": 5230
},
{
"entropy": 5.40293173789978,
"epoch": 0.5028818443804035,
"grad_norm": 0.91796875,
"learning_rate": 0.0004981296697743224,
"loss": 5.2831,
"mean_token_accuracy": 0.1927013948559761,
"num_tokens": 12012118.0,
"step": 5235
},
{
"entropy": 5.442527198791504,
"epoch": 0.5033621517771374,
"grad_norm": 1.0,
"learning_rate": 0.0004981252559071012,
"loss": 5.3205,
"mean_token_accuracy": 0.19446006715297698,
"num_tokens": 12023432.0,
"step": 5240
},
{
"entropy": 5.4181389808654785,
"epoch": 0.5038424591738713,
"grad_norm": 1.0234375,
"learning_rate": 0.0004981208368595739,
"loss": 5.2141,
"mean_token_accuracy": 0.20718919187784196,
"num_tokens": 12034323.0,
"step": 5245
},
{
"entropy": 5.35970139503479,
"epoch": 0.5043227665706052,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981164126318435,
"loss": 5.2285,
"mean_token_accuracy": 0.19909979403018951,
"num_tokens": 12045532.0,
"step": 5250
},
{
"entropy": 5.451566362380982,
"epoch": 0.5048030739673391,
"grad_norm": 0.89453125,
"learning_rate": 0.0004981119832240124,
"loss": 5.2431,
"mean_token_accuracy": 0.19895632565021515,
"num_tokens": 12057346.0,
"step": 5255
},
{
"entropy": 5.349082517623901,
"epoch": 0.505283381364073,
"grad_norm": 0.953125,
"learning_rate": 0.0004981075486361837,
"loss": 5.1845,
"mean_token_accuracy": 0.20472613275051116,
"num_tokens": 12068670.0,
"step": 5260
},
{
"entropy": 5.4529791355133055,
"epoch": 0.5057636887608069,
"grad_norm": 0.9921875,
"learning_rate": 0.0004981031088684601,
"loss": 5.3272,
"mean_token_accuracy": 0.19437083154916762,
"num_tokens": 12079664.0,
"step": 5265
},
{
"entropy": 5.469378185272217,
"epoch": 0.5062439961575408,
"grad_norm": 1.0,
"learning_rate": 0.0004980986639209448,
"loss": 5.2632,
"mean_token_accuracy": 0.2042248487472534,
"num_tokens": 12089984.0,
"step": 5270
},
{
"entropy": 5.319522666931152,
"epoch": 0.5067243035542748,
"grad_norm": 1.0,
"learning_rate": 0.000498094213793741,
"loss": 5.1899,
"mean_token_accuracy": 0.20940718650817872,
"num_tokens": 12101182.0,
"step": 5275
},
{
"entropy": 5.4405517578125,
"epoch": 0.5072046109510087,
"grad_norm": 1.0625,
"learning_rate": 0.000498089758486952,
"loss": 5.2895,
"mean_token_accuracy": 0.19249950349330902,
"num_tokens": 12112002.0,
"step": 5280
},
{
"entropy": 5.489940357208252,
"epoch": 0.5076849183477425,
"grad_norm": 0.90625,
"learning_rate": 0.0004980852980006812,
"loss": 5.356,
"mean_token_accuracy": 0.1845449000597,
"num_tokens": 12124194.0,
"step": 5285
},
{
"entropy": 5.358373641967773,
"epoch": 0.5081652257444764,
"grad_norm": 1.046875,
"learning_rate": 0.0004980808323350323,
"loss": 5.2607,
"mean_token_accuracy": 0.20721071660518647,
"num_tokens": 12133966.0,
"step": 5290
},
{
"entropy": 5.400069761276245,
"epoch": 0.5086455331412104,
"grad_norm": 1.0390625,
"learning_rate": 0.0004980763614901089,
"loss": 5.2166,
"mean_token_accuracy": 0.2029334396123886,
"num_tokens": 12145643.0,
"step": 5295
},
{
"entropy": 5.388980054855347,
"epoch": 0.5091258405379443,
"grad_norm": 1.0,
"learning_rate": 0.0004980718854660146,
"loss": 5.2638,
"mean_token_accuracy": 0.19128015339374543,
"num_tokens": 12156804.0,
"step": 5300
},
{
"entropy": 5.410366010665894,
"epoch": 0.5096061479346782,
"grad_norm": 0.93359375,
"learning_rate": 0.0004980674042628537,
"loss": 5.2201,
"mean_token_accuracy": 0.19946327358484267,
"num_tokens": 12168700.0,
"step": 5305
},
{
"entropy": 5.441403770446778,
"epoch": 0.5100864553314121,
"grad_norm": 0.91015625,
"learning_rate": 0.00049806291788073,
"loss": 5.2344,
"mean_token_accuracy": 0.19958012998104097,
"num_tokens": 12181050.0,
"step": 5310
},
{
"entropy": 5.441706371307373,
"epoch": 0.510566762728146,
"grad_norm": 0.953125,
"learning_rate": 0.0004980584263197477,
"loss": 5.321,
"mean_token_accuracy": 0.1931978940963745,
"num_tokens": 12192001.0,
"step": 5315
},
{
"entropy": 5.559450435638428,
"epoch": 0.5110470701248799,
"grad_norm": 1.1171875,
"learning_rate": 0.0004980539295800111,
"loss": 5.4211,
"mean_token_accuracy": 0.19310695230960845,
"num_tokens": 12202436.0,
"step": 5320
},
{
"entropy": 5.374323940277099,
"epoch": 0.5115273775216138,
"grad_norm": 0.96484375,
"learning_rate": 0.0004980494276616246,
"loss": 5.2081,
"mean_token_accuracy": 0.20342611223459245,
"num_tokens": 12214454.0,
"step": 5325
},
{
"entropy": 5.385732889175415,
"epoch": 0.5120076849183477,
"grad_norm": 0.9296875,
"learning_rate": 0.0004980449205646926,
"loss": 5.2272,
"mean_token_accuracy": 0.20538058280944824,
"num_tokens": 12225924.0,
"step": 5330
},
{
"entropy": 5.471169757843017,
"epoch": 0.5124879923150817,
"grad_norm": 1.0234375,
"learning_rate": 0.00049804040828932,
"loss": 5.2273,
"mean_token_accuracy": 0.20452706813812255,
"num_tokens": 12236456.0,
"step": 5335
},
{
"entropy": 5.417616176605224,
"epoch": 0.5129682997118156,
"grad_norm": 0.9609375,
"learning_rate": 0.0004980358908356113,
"loss": 5.2598,
"mean_token_accuracy": 0.20173069089651108,
"num_tokens": 12247719.0,
"step": 5340
},
{
"entropy": 5.375013160705566,
"epoch": 0.5134486071085494,
"grad_norm": 0.98828125,
"learning_rate": 0.0004980313682036717,
"loss": 5.3013,
"mean_token_accuracy": 0.1980268254876137,
"num_tokens": 12259141.0,
"step": 5345
},
{
"entropy": 5.52657642364502,
"epoch": 0.5139289145052833,
"grad_norm": 1.0625,
"learning_rate": 0.0004980268403936058,
"loss": 5.342,
"mean_token_accuracy": 0.19116934835910798,
"num_tokens": 12269748.0,
"step": 5350
},
{
"entropy": 5.426257944107055,
"epoch": 0.5144092219020173,
"grad_norm": 0.96875,
"learning_rate": 0.0004980223074055189,
"loss": 5.297,
"mean_token_accuracy": 0.2044803574681282,
"num_tokens": 12281456.0,
"step": 5355
},
{
"entropy": 5.440986442565918,
"epoch": 0.5148895292987512,
"grad_norm": 0.98046875,
"learning_rate": 0.0004980177692395164,
"loss": 5.2796,
"mean_token_accuracy": 0.1947730675339699,
"num_tokens": 12293763.0,
"step": 5360
},
{
"entropy": 5.409850406646728,
"epoch": 0.5153698366954851,
"grad_norm": 0.88671875,
"learning_rate": 0.0004980132258957035,
"loss": 5.2026,
"mean_token_accuracy": 0.20839707404375077,
"num_tokens": 12305398.0,
"step": 5365
},
{
"entropy": 5.35020489692688,
"epoch": 0.515850144092219,
"grad_norm": 0.99609375,
"learning_rate": 0.0004980086773741856,
"loss": 5.1854,
"mean_token_accuracy": 0.199608738720417,
"num_tokens": 12316582.0,
"step": 5370
},
{
"entropy": 5.461497783660889,
"epoch": 0.516330451488953,
"grad_norm": 1.140625,
"learning_rate": 0.0004980041236750685,
"loss": 5.3115,
"mean_token_accuracy": 0.1998691514134407,
"num_tokens": 12328463.0,
"step": 5375
},
{
"entropy": 5.471179628372193,
"epoch": 0.5168107588856868,
"grad_norm": 0.9765625,
"learning_rate": 0.0004979995647984577,
"loss": 5.291,
"mean_token_accuracy": 0.2032223239541054,
"num_tokens": 12341040.0,
"step": 5380
},
{
"entropy": 5.324481773376465,
"epoch": 0.5172910662824207,
"grad_norm": 0.8671875,
"learning_rate": 0.0004979950007444593,
"loss": 5.1905,
"mean_token_accuracy": 0.2048783466219902,
"num_tokens": 12353024.0,
"step": 5385
},
{
"entropy": 5.357087421417236,
"epoch": 0.5177713736791547,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979904315131792,
"loss": 5.206,
"mean_token_accuracy": 0.20590624809265137,
"num_tokens": 12366100.0,
"step": 5390
},
{
"entropy": 5.353571033477783,
"epoch": 0.5182516810758886,
"grad_norm": 0.96484375,
"learning_rate": 0.0004979858571047233,
"loss": 5.1981,
"mean_token_accuracy": 0.20849017649888993,
"num_tokens": 12377829.0,
"step": 5395
},
{
"entropy": 5.511527395248413,
"epoch": 0.5187319884726225,
"grad_norm": 1.0234375,
"learning_rate": 0.0004979812775191979,
"loss": 5.315,
"mean_token_accuracy": 0.19967973232269287,
"num_tokens": 12390830.0,
"step": 5400
},
{
"entropy": 5.343885278701782,
"epoch": 0.5192122958693564,
"grad_norm": 0.984375,
"learning_rate": 0.0004979766927567094,
"loss": 5.1545,
"mean_token_accuracy": 0.20680712759494782,
"num_tokens": 12401642.0,
"step": 5405
},
{
"entropy": 5.35053563117981,
"epoch": 0.5196926032660903,
"grad_norm": 0.99609375,
"learning_rate": 0.0004979721028173643,
"loss": 5.2792,
"mean_token_accuracy": 0.20148641914129256,
"num_tokens": 12411653.0,
"step": 5410
},
{
"entropy": 5.414511108398438,
"epoch": 0.5201729106628242,
"grad_norm": 1.0546875,
"learning_rate": 0.000497967507701269,
"loss": 5.1717,
"mean_token_accuracy": 0.2088325709104538,
"num_tokens": 12422891.0,
"step": 5415
},
{
"entropy": 5.461643075942993,
"epoch": 0.5206532180595581,
"grad_norm": 0.97265625,
"learning_rate": 0.0004979629074085303,
"loss": 5.2585,
"mean_token_accuracy": 0.20017822086811066,
"num_tokens": 12434190.0,
"step": 5420
},
{
"entropy": 5.425161838531494,
"epoch": 0.521133525456292,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979583019392548,
"loss": 5.3223,
"mean_token_accuracy": 0.19771393835544587,
"num_tokens": 12445796.0,
"step": 5425
},
{
"entropy": 5.515705680847168,
"epoch": 0.521613832853026,
"grad_norm": 1.0625,
"learning_rate": 0.0004979536912935497,
"loss": 5.3871,
"mean_token_accuracy": 0.19371298253536223,
"num_tokens": 12456212.0,
"step": 5430
},
{
"entropy": 5.369207715988159,
"epoch": 0.5220941402497599,
"grad_norm": 1.015625,
"learning_rate": 0.000497949075471522,
"loss": 5.1118,
"mean_token_accuracy": 0.20107672214508057,
"num_tokens": 12467871.0,
"step": 5435
},
{
"entropy": 5.394469690322876,
"epoch": 0.5225744476464937,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979444544732786,
"loss": 5.2053,
"mean_token_accuracy": 0.1949324682354927,
"num_tokens": 12478626.0,
"step": 5440
},
{
"entropy": 5.365805816650391,
"epoch": 0.5230547550432276,
"grad_norm": 0.93359375,
"learning_rate": 0.000497939828298927,
"loss": 5.2886,
"mean_token_accuracy": 0.19763265550136566,
"num_tokens": 12491487.0,
"step": 5445
},
{
"entropy": 5.461965608596802,
"epoch": 0.5235350624399616,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979351969485747,
"loss": 5.2327,
"mean_token_accuracy": 0.19547561854124068,
"num_tokens": 12503240.0,
"step": 5450
},
{
"entropy": 5.455387878417969,
"epoch": 0.5240153698366955,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979305604223291,
"loss": 5.1923,
"mean_token_accuracy": 0.20037689208984374,
"num_tokens": 12513860.0,
"step": 5455
},
{
"entropy": 5.3650209426879885,
"epoch": 0.5244956772334294,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979259187202978,
"loss": 5.2683,
"mean_token_accuracy": 0.20282870829105376,
"num_tokens": 12525884.0,
"step": 5460
},
{
"entropy": 5.4195245742797855,
"epoch": 0.5249759846301633,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979212718425887,
"loss": 5.1955,
"mean_token_accuracy": 0.20026529282331468,
"num_tokens": 12536709.0,
"step": 5465
},
{
"entropy": 5.410429239273071,
"epoch": 0.5254562920268973,
"grad_norm": 1.0703125,
"learning_rate": 0.0004979166197893096,
"loss": 5.1746,
"mean_token_accuracy": 0.20506655275821686,
"num_tokens": 12549727.0,
"step": 5470
},
{
"entropy": 5.3270786762237545,
"epoch": 0.5259365994236311,
"grad_norm": 0.85546875,
"learning_rate": 0.0004979119625605683,
"loss": 5.2661,
"mean_token_accuracy": 0.19516700655221939,
"num_tokens": 12562053.0,
"step": 5475
},
{
"entropy": 5.345616006851197,
"epoch": 0.526416906820365,
"grad_norm": 0.9921875,
"learning_rate": 0.0004979073001564734,
"loss": 5.1452,
"mean_token_accuracy": 0.20667382925748826,
"num_tokens": 12574096.0,
"step": 5480
},
{
"entropy": 5.460164356231689,
"epoch": 0.5268972142170989,
"grad_norm": 0.94140625,
"learning_rate": 0.0004979026325771328,
"loss": 5.3212,
"mean_token_accuracy": 0.20060795843601226,
"num_tokens": 12585416.0,
"step": 5485
},
{
"entropy": 5.413971662521362,
"epoch": 0.5273775216138329,
"grad_norm": 0.890625,
"learning_rate": 0.0004978979598226549,
"loss": 5.1676,
"mean_token_accuracy": 0.20576133579015732,
"num_tokens": 12596861.0,
"step": 5490
},
{
"entropy": 5.333009815216064,
"epoch": 0.5278578290105668,
"grad_norm": 0.98828125,
"learning_rate": 0.0004978932818931483,
"loss": 5.1927,
"mean_token_accuracy": 0.20716883689165116,
"num_tokens": 12607761.0,
"step": 5495
},
{
"entropy": 5.443383312225341,
"epoch": 0.5283381364073007,
"grad_norm": 1.0,
"learning_rate": 0.0004978885987887216,
"loss": 5.3051,
"mean_token_accuracy": 0.20227987468242645,
"num_tokens": 12619889.0,
"step": 5500
},
{
"entropy": 5.509159994125366,
"epoch": 0.5288184438040345,
"grad_norm": 0.97265625,
"learning_rate": 0.0004978839105094833,
"loss": 5.2662,
"mean_token_accuracy": 0.19618211537599564,
"num_tokens": 12630604.0,
"step": 5505
},
{
"entropy": 5.314917039871216,
"epoch": 0.5292987512007685,
"grad_norm": 0.96484375,
"learning_rate": 0.0004978792170555426,
"loss": 5.1658,
"mean_token_accuracy": 0.20746731758117676,
"num_tokens": 12641172.0,
"step": 5510
},
{
"entropy": 5.312827301025391,
"epoch": 0.5297790585975024,
"grad_norm": 0.984375,
"learning_rate": 0.0004978745184270083,
"loss": 5.0923,
"mean_token_accuracy": 0.21527302861213685,
"num_tokens": 12651731.0,
"step": 5515
},
{
"entropy": 5.386341571807861,
"epoch": 0.5302593659942363,
"grad_norm": 0.96875,
"learning_rate": 0.0004978698146239893,
"loss": 5.2144,
"mean_token_accuracy": 0.20662128329277038,
"num_tokens": 12663050.0,
"step": 5520
},
{
"entropy": 5.479044103622437,
"epoch": 0.5307396733909702,
"grad_norm": 0.9453125,
"learning_rate": 0.0004978651056465952,
"loss": 5.3325,
"mean_token_accuracy": 0.19395910054445267,
"num_tokens": 12674732.0,
"step": 5525
},
{
"entropy": 5.313097810745239,
"epoch": 0.5312199807877042,
"grad_norm": 0.9921875,
"learning_rate": 0.000497860391494935,
"loss": 5.1422,
"mean_token_accuracy": 0.2039245158433914,
"num_tokens": 12685981.0,
"step": 5530
},
{
"entropy": 5.447102403640747,
"epoch": 0.531700288184438,
"grad_norm": 1.0078125,
"learning_rate": 0.0004978556721691183,
"loss": 5.2773,
"mean_token_accuracy": 0.19787385314702988,
"num_tokens": 12697139.0,
"step": 5535
},
{
"entropy": 5.3786839008331295,
"epoch": 0.5321805955811719,
"grad_norm": 0.9375,
"learning_rate": 0.0004978509476692547,
"loss": 5.2202,
"mean_token_accuracy": 0.19959738850593567,
"num_tokens": 12708268.0,
"step": 5540
},
{
"entropy": 5.375888729095459,
"epoch": 0.5326609029779059,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978462179954538,
"loss": 5.2049,
"mean_token_accuracy": 0.20851173847913743,
"num_tokens": 12720715.0,
"step": 5545
},
{
"entropy": 5.367966890335083,
"epoch": 0.5331412103746398,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978414831478253,
"loss": 5.1717,
"mean_token_accuracy": 0.20525323897600173,
"num_tokens": 12732409.0,
"step": 5550
},
{
"entropy": 5.346772003173828,
"epoch": 0.5336215177713737,
"grad_norm": 0.91796875,
"learning_rate": 0.0004978367431264794,
"loss": 5.3142,
"mean_token_accuracy": 0.1936624139547348,
"num_tokens": 12745174.0,
"step": 5555
},
{
"entropy": 5.437586259841919,
"epoch": 0.5341018251681076,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978319979315261,
"loss": 5.2593,
"mean_token_accuracy": 0.20328891277313232,
"num_tokens": 12756116.0,
"step": 5560
},
{
"entropy": 5.431155967712402,
"epoch": 0.5345821325648416,
"grad_norm": 0.99609375,
"learning_rate": 0.0004978272475630752,
"loss": 5.1951,
"mean_token_accuracy": 0.20230283141136168,
"num_tokens": 12768183.0,
"step": 5565
},
{
"entropy": 5.306485891342163,
"epoch": 0.5350624399615754,
"grad_norm": 0.9609375,
"learning_rate": 0.0004978224920212374,
"loss": 5.1953,
"mean_token_accuracy": 0.19942874312400818,
"num_tokens": 12778537.0,
"step": 5570
},
{
"entropy": 5.288288307189942,
"epoch": 0.5355427473583093,
"grad_norm": 1.0234375,
"learning_rate": 0.0004978177313061232,
"loss": 5.2401,
"mean_token_accuracy": 0.2020668715238571,
"num_tokens": 12789691.0,
"step": 5575
},
{
"entropy": 5.450948762893677,
"epoch": 0.5360230547550432,
"grad_norm": 1.0,
"learning_rate": 0.0004978129654178426,
"loss": 5.2781,
"mean_token_accuracy": 0.19842252731323243,
"num_tokens": 12801438.0,
"step": 5580
},
{
"entropy": 5.403745031356811,
"epoch": 0.5365033621517772,
"grad_norm": 0.9375,
"learning_rate": 0.0004978081943565067,
"loss": 5.2039,
"mean_token_accuracy": 0.19873054474592208,
"num_tokens": 12812425.0,
"step": 5585
},
{
"entropy": 5.376599311828613,
"epoch": 0.5369836695485111,
"grad_norm": 0.87109375,
"learning_rate": 0.0004978034181222261,
"loss": 5.1988,
"mean_token_accuracy": 0.19666872173547745,
"num_tokens": 12824735.0,
"step": 5590
},
{
"entropy": 5.448877477645874,
"epoch": 0.537463976945245,
"grad_norm": 0.97265625,
"learning_rate": 0.0004977986367151119,
"loss": 5.2731,
"mean_token_accuracy": 0.2031214103102684,
"num_tokens": 12835454.0,
"step": 5595
},
{
"entropy": 5.4336474418640135,
"epoch": 0.5379442843419788,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977938501352747,
"loss": 5.3367,
"mean_token_accuracy": 0.1929567873477936,
"num_tokens": 12847086.0,
"step": 5600
},
{
"entropy": 5.420505571365356,
"epoch": 0.5384245917387128,
"grad_norm": 0.9375,
"learning_rate": 0.0004977890583828259,
"loss": 5.2644,
"mean_token_accuracy": 0.19988568872213364,
"num_tokens": 12857713.0,
"step": 5605
},
{
"entropy": 5.375241994857788,
"epoch": 0.5389048991354467,
"grad_norm": 0.875,
"learning_rate": 0.0004977842614578768,
"loss": 5.2512,
"mean_token_accuracy": 0.1993047297000885,
"num_tokens": 12869967.0,
"step": 5610
},
{
"entropy": 5.490482473373413,
"epoch": 0.5393852065321806,
"grad_norm": 0.9609375,
"learning_rate": 0.0004977794593605386,
"loss": 5.2826,
"mean_token_accuracy": 0.19814034998416902,
"num_tokens": 12881230.0,
"step": 5615
},
{
"entropy": 5.39257779121399,
"epoch": 0.5398655139289145,
"grad_norm": 0.97265625,
"learning_rate": 0.000497774652090923,
"loss": 5.2414,
"mean_token_accuracy": 0.19702692776918412,
"num_tokens": 12892376.0,
"step": 5620
},
{
"entropy": 5.413067626953125,
"epoch": 0.5403458213256485,
"grad_norm": 0.9921875,
"learning_rate": 0.0004977698396491414,
"loss": 5.2405,
"mean_token_accuracy": 0.20418344736099242,
"num_tokens": 12903709.0,
"step": 5625
},
{
"entropy": 5.319581651687622,
"epoch": 0.5408261287223823,
"grad_norm": 0.92578125,
"learning_rate": 0.0004977650220353055,
"loss": 5.0824,
"mean_token_accuracy": 0.20584484487771987,
"num_tokens": 12914958.0,
"step": 5630
},
{
"entropy": 5.376936054229736,
"epoch": 0.5413064361191162,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977601992495274,
"loss": 5.2056,
"mean_token_accuracy": 0.20099034607410432,
"num_tokens": 12927418.0,
"step": 5635
},
{
"entropy": 5.412064933776856,
"epoch": 0.5417867435158501,
"grad_norm": 0.84765625,
"learning_rate": 0.0004977553712919189,
"loss": 5.2503,
"mean_token_accuracy": 0.20160660594701768,
"num_tokens": 12939874.0,
"step": 5640
},
{
"entropy": 5.416967535018921,
"epoch": 0.5422670509125841,
"grad_norm": 0.921875,
"learning_rate": 0.0004977505381625921,
"loss": 5.2681,
"mean_token_accuracy": 0.19486918598413466,
"num_tokens": 12951113.0,
"step": 5645
},
{
"entropy": 5.315142440795898,
"epoch": 0.542747358309318,
"grad_norm": 0.90625,
"learning_rate": 0.0004977456998616593,
"loss": 5.1594,
"mean_token_accuracy": 0.2023612678050995,
"num_tokens": 12961940.0,
"step": 5650
},
{
"entropy": 5.284940385818482,
"epoch": 0.5432276657060519,
"grad_norm": 0.92578125,
"learning_rate": 0.0004977408563892327,
"loss": 5.1674,
"mean_token_accuracy": 0.19892300814390182,
"num_tokens": 12973938.0,
"step": 5655
},
{
"entropy": 5.412263488769531,
"epoch": 0.5437079731027857,
"grad_norm": 0.9921875,
"learning_rate": 0.0004977360077454249,
"loss": 5.1788,
"mean_token_accuracy": 0.20297273993492126,
"num_tokens": 12985400.0,
"step": 5660
},
{
"entropy": 5.385592126846314,
"epoch": 0.5441882804995197,
"grad_norm": 0.9375,
"learning_rate": 0.0004977311539303483,
"loss": 5.2286,
"mean_token_accuracy": 0.20850644558668135,
"num_tokens": 12996402.0,
"step": 5665
},
{
"entropy": 5.293762063980102,
"epoch": 0.5446685878962536,
"grad_norm": 1.046875,
"learning_rate": 0.0004977262949441158,
"loss": 5.1092,
"mean_token_accuracy": 0.21470348685979843,
"num_tokens": 13006991.0,
"step": 5670
},
{
"entropy": 5.397123908996582,
"epoch": 0.5451488952929875,
"grad_norm": 1.0078125,
"learning_rate": 0.0004977214307868399,
"loss": 5.1999,
"mean_token_accuracy": 0.20438773483037947,
"num_tokens": 13016969.0,
"step": 5675
},
{
"entropy": 5.392855453491211,
"epoch": 0.5456292026897214,
"grad_norm": 1.0859375,
"learning_rate": 0.000497716561458634,
"loss": 5.1437,
"mean_token_accuracy": 0.20777873545885087,
"num_tokens": 13027759.0,
"step": 5680
},
{
"entropy": 5.486040258407593,
"epoch": 0.5461095100864554,
"grad_norm": 0.9140625,
"learning_rate": 0.0004977116869596107,
"loss": 5.3619,
"mean_token_accuracy": 0.1973151370882988,
"num_tokens": 13039881.0,
"step": 5685
},
{
"entropy": 5.378542852401734,
"epoch": 0.5465898174831892,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977068072898834,
"loss": 5.1963,
"mean_token_accuracy": 0.20589208602905273,
"num_tokens": 13051443.0,
"step": 5690
},
{
"entropy": 5.4433074474334715,
"epoch": 0.5470701248799231,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977019224495652,
"loss": 5.2987,
"mean_token_accuracy": 0.1966116562485695,
"num_tokens": 13063474.0,
"step": 5695
},
{
"entropy": 5.370102453231811,
"epoch": 0.547550432276657,
"grad_norm": 0.96484375,
"learning_rate": 0.0004976970324387698,
"loss": 5.1416,
"mean_token_accuracy": 0.2106616035103798,
"num_tokens": 13074365.0,
"step": 5700
},
{
"entropy": 5.358544445037841,
"epoch": 0.548030739673391,
"grad_norm": 0.9921875,
"learning_rate": 0.0004976921372576104,
"loss": 5.2212,
"mean_token_accuracy": 0.20368915945291519,
"num_tokens": 13087354.0,
"step": 5705
},
{
"entropy": 5.370021200180053,
"epoch": 0.5485110470701249,
"grad_norm": 0.875,
"learning_rate": 0.0004976872369062011,
"loss": 5.1967,
"mean_token_accuracy": 0.19930920600891114,
"num_tokens": 13099306.0,
"step": 5710
},
{
"entropy": 5.408384513854981,
"epoch": 0.5489913544668588,
"grad_norm": 0.99609375,
"learning_rate": 0.0004976823313846552,
"loss": 5.3329,
"mean_token_accuracy": 0.19190380424261094,
"num_tokens": 13111259.0,
"step": 5715
},
{
"entropy": 5.45887770652771,
"epoch": 0.5494716618635928,
"grad_norm": 0.9453125,
"learning_rate": 0.0004976774206930869,
"loss": 5.2471,
"mean_token_accuracy": 0.19341034889221193,
"num_tokens": 13123589.0,
"step": 5720
},
{
"entropy": 5.377495574951172,
"epoch": 0.5499519692603266,
"grad_norm": 0.98046875,
"learning_rate": 0.0004976725048316101,
"loss": 5.2444,
"mean_token_accuracy": 0.19959345310926438,
"num_tokens": 13136485.0,
"step": 5725
},
{
"entropy": 5.409855890274048,
"epoch": 0.5504322766570605,
"grad_norm": 0.8984375,
"learning_rate": 0.0004976675838003388,
"loss": 5.2149,
"mean_token_accuracy": 0.19824812114238738,
"num_tokens": 13148067.0,
"step": 5730
},
{
"entropy": 5.3532360076904295,
"epoch": 0.5509125840537944,
"grad_norm": 1.0625,
"learning_rate": 0.0004976626575993877,
"loss": 5.1943,
"mean_token_accuracy": 0.19984031617641448,
"num_tokens": 13159813.0,
"step": 5735
},
{
"entropy": 5.414771270751953,
"epoch": 0.5513928914505284,
"grad_norm": 1.1171875,
"learning_rate": 0.0004976577262288705,
"loss": 5.2672,
"mean_token_accuracy": 0.19723765850067138,
"num_tokens": 13170828.0,
"step": 5740
},
{
"entropy": 5.348534822463989,
"epoch": 0.5518731988472623,
"grad_norm": 0.95703125,
"learning_rate": 0.0004976527896889023,
"loss": 5.0988,
"mean_token_accuracy": 0.21700155436992646,
"num_tokens": 13181883.0,
"step": 5745
},
{
"entropy": 5.332348966598511,
"epoch": 0.5523535062439962,
"grad_norm": 0.96875,
"learning_rate": 0.0004976478479795974,
"loss": 5.1755,
"mean_token_accuracy": 0.20436936616897583,
"num_tokens": 13193530.0,
"step": 5750
},
{
"entropy": 5.461158657073975,
"epoch": 0.55283381364073,
"grad_norm": 1.03125,
"learning_rate": 0.0004976429011010706,
"loss": 5.3936,
"mean_token_accuracy": 0.1911667287349701,
"num_tokens": 13205822.0,
"step": 5755
},
{
"entropy": 5.377006673812867,
"epoch": 0.553314121037464,
"grad_norm": 0.9296875,
"learning_rate": 0.0004976379490534366,
"loss": 5.1143,
"mean_token_accuracy": 0.210676147043705,
"num_tokens": 13216698.0,
"step": 5760
},
{
"entropy": 5.346554327011108,
"epoch": 0.5537944284341979,
"grad_norm": 0.9453125,
"learning_rate": 0.0004976329918368107,
"loss": 5.2225,
"mean_token_accuracy": 0.2032114490866661,
"num_tokens": 13228389.0,
"step": 5765
},
{
"entropy": 5.437006664276123,
"epoch": 0.5542747358309318,
"grad_norm": 1.0,
"learning_rate": 0.0004976280294513079,
"loss": 5.2655,
"mean_token_accuracy": 0.19674807637929917,
"num_tokens": 13239628.0,
"step": 5770
},
{
"entropy": 5.438928031921387,
"epoch": 0.5547550432276657,
"grad_norm": 0.921875,
"learning_rate": 0.0004976230618970431,
"loss": 5.2726,
"mean_token_accuracy": 0.20090280920267106,
"num_tokens": 13251149.0,
"step": 5775
},
{
"entropy": 5.4733867168426515,
"epoch": 0.5552353506243997,
"grad_norm": 0.87109375,
"learning_rate": 0.000497618089174132,
"loss": 5.3166,
"mean_token_accuracy": 0.19709511697292328,
"num_tokens": 13264846.0,
"step": 5780
},
{
"entropy": 5.324634456634522,
"epoch": 0.5557156580211335,
"grad_norm": 0.91796875,
"learning_rate": 0.0004976131112826898,
"loss": 5.1012,
"mean_token_accuracy": 0.2113727256655693,
"num_tokens": 13275409.0,
"step": 5785
},
{
"entropy": 5.231151103973389,
"epoch": 0.5561959654178674,
"grad_norm": 0.97265625,
"learning_rate": 0.0004976081282228323,
"loss": 5.0786,
"mean_token_accuracy": 0.20907817929983138,
"num_tokens": 13287173.0,
"step": 5790
},
{
"entropy": 5.402900981903076,
"epoch": 0.5566762728146013,
"grad_norm": 0.95703125,
"learning_rate": 0.000497603139994675,
"loss": 5.1658,
"mean_token_accuracy": 0.2003258064389229,
"num_tokens": 13298225.0,
"step": 5795
},
{
"entropy": 5.381881332397461,
"epoch": 0.5571565802113353,
"grad_norm": 0.93359375,
"learning_rate": 0.0004975981465983338,
"loss": 5.1716,
"mean_token_accuracy": 0.20382763296365738,
"num_tokens": 13309685.0,
"step": 5800
},
{
"entropy": 5.313303709030151,
"epoch": 0.5576368876080692,
"grad_norm": 0.95703125,
"learning_rate": 0.0004975931480339246,
"loss": 5.1777,
"mean_token_accuracy": 0.20920694172382354,
"num_tokens": 13320837.0,
"step": 5805
},
{
"entropy": 5.373168277740478,
"epoch": 0.5581171950048031,
"grad_norm": 0.890625,
"learning_rate": 0.0004975881443015635,
"loss": 5.279,
"mean_token_accuracy": 0.20333101153373717,
"num_tokens": 13333512.0,
"step": 5810
},
{
"entropy": 5.478464984893799,
"epoch": 0.5585975024015369,
"grad_norm": 0.90625,
"learning_rate": 0.0004975831354013667,
"loss": 5.3068,
"mean_token_accuracy": 0.19963099658489228,
"num_tokens": 13345189.0,
"step": 5815
},
{
"entropy": 5.362034654617309,
"epoch": 0.5590778097982709,
"grad_norm": 0.953125,
"learning_rate": 0.0004975781213334503,
"loss": 5.1646,
"mean_token_accuracy": 0.2086104243993759,
"num_tokens": 13356123.0,
"step": 5820
},
{
"entropy": 5.3880220413208,
"epoch": 0.5595581171950048,
"grad_norm": 1.0546875,
"learning_rate": 0.0004975731020979309,
"loss": 5.2036,
"mean_token_accuracy": 0.20239470899105072,
"num_tokens": 13366902.0,
"step": 5825
},
{
"entropy": 5.459943771362305,
"epoch": 0.5600384245917387,
"grad_norm": 1.078125,
"learning_rate": 0.0004975680776949249,
"loss": 5.2605,
"mean_token_accuracy": 0.200407475233078,
"num_tokens": 13377567.0,
"step": 5830
},
{
"entropy": 5.400122451782226,
"epoch": 0.5605187319884726,
"grad_norm": 0.98828125,
"learning_rate": 0.0004975630481245492,
"loss": 5.2056,
"mean_token_accuracy": 0.2129899948835373,
"num_tokens": 13387297.0,
"step": 5835
},
{
"entropy": 5.408231210708618,
"epoch": 0.5609990393852066,
"grad_norm": 0.99609375,
"learning_rate": 0.0004975580133869202,
"loss": 5.2405,
"mean_token_accuracy": 0.20631884038448334,
"num_tokens": 13397723.0,
"step": 5840
},
{
"entropy": 5.416802930831909,
"epoch": 0.5614793467819404,
"grad_norm": 0.94921875,
"learning_rate": 0.0004975529734821552,
"loss": 5.2988,
"mean_token_accuracy": 0.20144724398851394,
"num_tokens": 13409875.0,
"step": 5845
},
{
"entropy": 5.36104941368103,
"epoch": 0.5619596541786743,
"grad_norm": 0.9609375,
"learning_rate": 0.0004975479284103708,
"loss": 5.1902,
"mean_token_accuracy": 0.20576498061418533,
"num_tokens": 13421338.0,
"step": 5850
},
{
"entropy": 5.426533842086792,
"epoch": 0.5624399615754082,
"grad_norm": 1.03125,
"learning_rate": 0.0004975428781716845,
"loss": 5.2212,
"mean_token_accuracy": 0.20507221519947053,
"num_tokens": 13431373.0,
"step": 5855
},
{
"entropy": 5.410734415054321,
"epoch": 0.5629202689721422,
"grad_norm": 0.890625,
"learning_rate": 0.0004975378227662134,
"loss": 5.2411,
"mean_token_accuracy": 0.20476576089859008,
"num_tokens": 13443158.0,
"step": 5860
},
{
"entropy": 5.4595184326171875,
"epoch": 0.5634005763688761,
"grad_norm": 0.95703125,
"learning_rate": 0.0004975327621940746,
"loss": 5.3079,
"mean_token_accuracy": 0.1933220088481903,
"num_tokens": 13454559.0,
"step": 5865
},
{
"entropy": 5.514381456375122,
"epoch": 0.56388088376561,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975276964553861,
"loss": 5.3656,
"mean_token_accuracy": 0.19711938351392747,
"num_tokens": 13466934.0,
"step": 5870
},
{
"entropy": 5.40661940574646,
"epoch": 0.5643611911623438,
"grad_norm": 0.95703125,
"learning_rate": 0.0004975226255502651,
"loss": 5.1231,
"mean_token_accuracy": 0.21357636600732804,
"num_tokens": 13477770.0,
"step": 5875
},
{
"entropy": 5.451413154602051,
"epoch": 0.5648414985590778,
"grad_norm": 0.953125,
"learning_rate": 0.0004975175494788297,
"loss": 5.3249,
"mean_token_accuracy": 0.1998008131980896,
"num_tokens": 13490093.0,
"step": 5880
},
{
"entropy": 5.453090238571167,
"epoch": 0.5653218059558117,
"grad_norm": 1.2890625,
"learning_rate": 0.0004975124682411974,
"loss": 5.182,
"mean_token_accuracy": 0.20121294111013413,
"num_tokens": 13500663.0,
"step": 5885
},
{
"entropy": 5.405351638793945,
"epoch": 0.5658021133525456,
"grad_norm": 1.0703125,
"learning_rate": 0.0004975073818374863,
"loss": 5.2896,
"mean_token_accuracy": 0.2002902314066887,
"num_tokens": 13512369.0,
"step": 5890
},
{
"entropy": 5.3891054630279545,
"epoch": 0.5662824207492796,
"grad_norm": 1.0,
"learning_rate": 0.0004975022902678145,
"loss": 5.1734,
"mean_token_accuracy": 0.20277402997016908,
"num_tokens": 13523181.0,
"step": 5895
},
{
"entropy": 5.361138439178466,
"epoch": 0.5667627281460135,
"grad_norm": 0.97265625,
"learning_rate": 0.0004974971935323003,
"loss": 5.1224,
"mean_token_accuracy": 0.20029759407043457,
"num_tokens": 13534113.0,
"step": 5900
},
{
"entropy": 5.327734184265137,
"epoch": 0.5672430355427474,
"grad_norm": 1.015625,
"learning_rate": 0.0004974920916310619,
"loss": 5.1607,
"mean_token_accuracy": 0.20367970913648606,
"num_tokens": 13545037.0,
"step": 5905
},
{
"entropy": 5.405796194076538,
"epoch": 0.5677233429394812,
"grad_norm": 1.0078125,
"learning_rate": 0.0004974869845642178,
"loss": 5.2249,
"mean_token_accuracy": 0.20131197422742844,
"num_tokens": 13555541.0,
"step": 5910
},
{
"entropy": 5.352486181259155,
"epoch": 0.5682036503362152,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974818723318866,
"loss": 5.1964,
"mean_token_accuracy": 0.20470297634601592,
"num_tokens": 13566951.0,
"step": 5915
},
{
"entropy": 5.4211790561676025,
"epoch": 0.5686839577329491,
"grad_norm": 0.9296875,
"learning_rate": 0.0004974767549341868,
"loss": 5.2705,
"mean_token_accuracy": 0.2000730812549591,
"num_tokens": 13578492.0,
"step": 5920
},
{
"entropy": 5.4369526386260985,
"epoch": 0.569164265129683,
"grad_norm": 1.03125,
"learning_rate": 0.0004974716323712376,
"loss": 5.1561,
"mean_token_accuracy": 0.21389784663915634,
"num_tokens": 13589183.0,
"step": 5925
},
{
"entropy": 5.404523801803589,
"epoch": 0.5696445725264169,
"grad_norm": 0.96875,
"learning_rate": 0.0004974665046431576,
"loss": 5.2862,
"mean_token_accuracy": 0.202628256380558,
"num_tokens": 13600588.0,
"step": 5930
},
{
"entropy": 5.343457460403442,
"epoch": 0.5701248799231509,
"grad_norm": 1.0234375,
"learning_rate": 0.0004974613717500659,
"loss": 5.1779,
"mean_token_accuracy": 0.20688795298337936,
"num_tokens": 13612107.0,
"step": 5935
},
{
"entropy": 5.510196113586426,
"epoch": 0.5706051873198847,
"grad_norm": 1.1328125,
"learning_rate": 0.0004974562336920818,
"loss": 5.3502,
"mean_token_accuracy": 0.1963939294219017,
"num_tokens": 13623973.0,
"step": 5940
},
{
"entropy": 5.374628925323487,
"epoch": 0.5710854947166186,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974510904693245,
"loss": 5.146,
"mean_token_accuracy": 0.20711901485919954,
"num_tokens": 13634994.0,
"step": 5945
},
{
"entropy": 5.426691341400146,
"epoch": 0.5715658021133525,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974459420819134,
"loss": 5.3057,
"mean_token_accuracy": 0.20290742069482803,
"num_tokens": 13646361.0,
"step": 5950
},
{
"entropy": 5.406249523162842,
"epoch": 0.5720461095100865,
"grad_norm": 1.0,
"learning_rate": 0.000497440788529968,
"loss": 5.1866,
"mean_token_accuracy": 0.20263564288616182,
"num_tokens": 13656975.0,
"step": 5955
},
{
"entropy": 5.3955769538879395,
"epoch": 0.5725264169068204,
"grad_norm": 0.98828125,
"learning_rate": 0.0004974356298136081,
"loss": 5.2381,
"mean_token_accuracy": 0.19841890335083007,
"num_tokens": 13668434.0,
"step": 5960
},
{
"entropy": 5.416602754592896,
"epoch": 0.5730067243035543,
"grad_norm": 1.0234375,
"learning_rate": 0.0004974304659329533,
"loss": 5.2127,
"mean_token_accuracy": 0.19981901794672013,
"num_tokens": 13679266.0,
"step": 5965
},
{
"entropy": 5.355310297012329,
"epoch": 0.5734870317002881,
"grad_norm": 0.96875,
"learning_rate": 0.0004974252968881236,
"loss": 5.2388,
"mean_token_accuracy": 0.1961240813136101,
"num_tokens": 13690921.0,
"step": 5970
},
{
"entropy": 5.414062929153443,
"epoch": 0.5739673390970221,
"grad_norm": 0.94921875,
"learning_rate": 0.000497420122679239,
"loss": 5.1833,
"mean_token_accuracy": 0.20539622157812118,
"num_tokens": 13702329.0,
"step": 5975
},
{
"entropy": 5.3050618171691895,
"epoch": 0.574447646493756,
"grad_norm": 0.921875,
"learning_rate": 0.0004974149433064196,
"loss": 5.1471,
"mean_token_accuracy": 0.2154633417725563,
"num_tokens": 13713356.0,
"step": 5980
},
{
"entropy": 5.2760419845581055,
"epoch": 0.5749279538904899,
"grad_norm": 0.94921875,
"learning_rate": 0.0004974097587697856,
"loss": 5.1355,
"mean_token_accuracy": 0.20253586918115615,
"num_tokens": 13724718.0,
"step": 5985
},
{
"entropy": 5.281339883804321,
"epoch": 0.5754082612872238,
"grad_norm": 0.921875,
"learning_rate": 0.0004974045690694575,
"loss": 5.1596,
"mean_token_accuracy": 0.2093519538640976,
"num_tokens": 13736113.0,
"step": 5990
},
{
"entropy": 5.454013776779175,
"epoch": 0.5758885686839578,
"grad_norm": 0.90234375,
"learning_rate": 0.0004973993742055557,
"loss": 5.1872,
"mean_token_accuracy": 0.2056156277656555,
"num_tokens": 13748322.0,
"step": 5995
},
{
"entropy": 5.352500104904175,
"epoch": 0.5763688760806917,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973941741782007,
"loss": 5.1919,
"mean_token_accuracy": 0.19695847779512404,
"num_tokens": 13759433.0,
"step": 6000
},
{
"epoch": 0.5763688760806917,
"eval_entropy": 5.17189993351509,
"eval_loss": 5.2146196365356445,
"eval_mean_token_accuracy": 0.2106242566784959,
"eval_num_tokens": 13759433.0,
"eval_runtime": 20.5416,
"eval_samples_per_second": 1597.487,
"eval_steps_per_second": 199.692,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 104090,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3129573368954880.0,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink