fpadovani/eng-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
8574823600795f6828667f797a55dc9b44fcae56
eng-latn-100mb-after-ppt-Dp…/checkpoint-12000/trainer_state.json
ModelHub XC 8574823600 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/eng-latn-100mb-after-ppt-Dp-100mb-ckpt500_seed3407
Source: Original Platform
2026-08-03 15:47:32 +08:00

24079 lines
660 KiB
JSON
Raw Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.3477284214073117,
"eval_steps": 3000,
"global_step": 12000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 6.907250690460205,
"epoch": 0.0005615769079575448,
"grad_norm": 1.1953125,
"learning_rate": 2e-06,
"loss": 6.6454,
"mean_token_accuracy": 0.12325676903128624,
"num_tokens": 11659.0,
"step": 5
},
{
"entropy": 6.95299882888794,
"epoch": 0.0011231538159150896,
"grad_norm": 1.2578125,
"learning_rate": 4.5e-06,
"loss": 6.6251,
"mean_token_accuracy": 0.12640999108552933,
"num_tokens": 21794.0,
"step": 10
},
{
"entropy": 6.88270936012268,
"epoch": 0.0016847307238726343,
"grad_norm": 1.296875,
"learning_rate": 7e-06,
"loss": 6.6267,
"mean_token_accuracy": 0.11778247579932213,
"num_tokens": 33206.0,
"step": 15
},
{
"entropy": 6.894813537597656,
"epoch": 0.0022463076318301792,
"grad_norm": 1.140625,
"learning_rate": 9.5e-06,
"loss": 6.6672,
"mean_token_accuracy": 0.11579557210206985,
"num_tokens": 44487.0,
"step": 20
},
{
"entropy": 6.8511096954345705,
"epoch": 0.002807884539787724,
"grad_norm": 1.0,
"learning_rate": 1.2e-05,
"loss": 6.7036,
"mean_token_accuracy": 0.12203121483325959,
"num_tokens": 56062.0,
"step": 25
},
{
"entropy": 6.77201623916626,
"epoch": 0.0033694614477452686,
"grad_norm": 1.1484375,
"learning_rate": 1.4500000000000002e-05,
"loss": 6.5797,
"mean_token_accuracy": 0.12473140060901641,
"num_tokens": 67453.0,
"step": 30
},
{
"entropy": 6.868989038467407,
"epoch": 0.003931038355702813,
"grad_norm": 1.03125,
"learning_rate": 1.7000000000000003e-05,
"loss": 6.6964,
"mean_token_accuracy": 0.12546522542834282,
"num_tokens": 78527.0,
"step": 35
},
{
"entropy": 6.834421539306641,
"epoch": 0.0044926152636603585,
"grad_norm": 1.0234375,
"learning_rate": 1.95e-05,
"loss": 6.5568,
"mean_token_accuracy": 0.1299024060368538,
"num_tokens": 89267.0,
"step": 40
},
{
"entropy": 6.802654600143432,
"epoch": 0.005054192171617903,
"grad_norm": 1.0546875,
"learning_rate": 2.2e-05,
"loss": 6.567,
"mean_token_accuracy": 0.12636540681123734,
"num_tokens": 102035.0,
"step": 45
},
{
"entropy": 6.78539891242981,
"epoch": 0.005615769079575448,
"grad_norm": 1.21875,
"learning_rate": 2.4500000000000003e-05,
"loss": 6.6043,
"mean_token_accuracy": 0.12845361977815628,
"num_tokens": 112734.0,
"step": 50
},
{
"entropy": 6.808405780792237,
"epoch": 0.006177345987532993,
"grad_norm": 0.9375,
"learning_rate": 2.7e-05,
"loss": 6.518,
"mean_token_accuracy": 0.13111241459846495,
"num_tokens": 123854.0,
"step": 55
},
{
"entropy": 6.786837434768676,
"epoch": 0.006738922895490537,
"grad_norm": 1.171875,
"learning_rate": 2.95e-05,
"loss": 6.5574,
"mean_token_accuracy": 0.12830817103385925,
"num_tokens": 133444.0,
"step": 60
},
{
"entropy": 6.7932250022888185,
"epoch": 0.007300499803448083,
"grad_norm": 0.9765625,
"learning_rate": 3.2e-05,
"loss": 6.4492,
"mean_token_accuracy": 0.13187738582491876,
"num_tokens": 144678.0,
"step": 65
},
{
"entropy": 6.834688949584961,
"epoch": 0.007862076711405626,
"grad_norm": 1.1171875,
"learning_rate": 3.4500000000000005e-05,
"loss": 6.5573,
"mean_token_accuracy": 0.1290239155292511,
"num_tokens": 155866.0,
"step": 70
},
{
"entropy": 6.833860158920288,
"epoch": 0.008423653619363172,
"grad_norm": 1.0546875,
"learning_rate": 3.7e-05,
"loss": 6.5594,
"mean_token_accuracy": 0.1337614819407463,
"num_tokens": 166965.0,
"step": 75
},
{
"entropy": 6.771542024612427,
"epoch": 0.008985230527320717,
"grad_norm": 1.234375,
"learning_rate": 3.95e-05,
"loss": 6.5349,
"mean_token_accuracy": 0.13214125260710716,
"num_tokens": 177423.0,
"step": 80
},
{
"entropy": 6.810234785079956,
"epoch": 0.00954680743527826,
"grad_norm": 0.95703125,
"learning_rate": 4.2000000000000004e-05,
"loss": 6.5843,
"mean_token_accuracy": 0.13490741923451424,
"num_tokens": 188687.0,
"step": 85
},
{
"entropy": 6.790950298309326,
"epoch": 0.010108384343235806,
"grad_norm": 1.03125,
"learning_rate": 4.45e-05,
"loss": 6.6447,
"mean_token_accuracy": 0.1322058655321598,
"num_tokens": 200173.0,
"step": 90
},
{
"entropy": 6.904566144943237,
"epoch": 0.010669961251193351,
"grad_norm": 0.99609375,
"learning_rate": 4.7000000000000004e-05,
"loss": 6.6097,
"mean_token_accuracy": 0.12937101572752,
"num_tokens": 210877.0,
"step": 95
},
{
"entropy": 6.820630264282227,
"epoch": 0.011231538159150897,
"grad_norm": 1.0,
"learning_rate": 4.9500000000000004e-05,
"loss": 6.4946,
"mean_token_accuracy": 0.1320508226752281,
"num_tokens": 221770.0,
"step": 100
},
{
"entropy": 6.799368095397949,
"epoch": 0.01179311506710844,
"grad_norm": 1.1484375,
"learning_rate": 5.2e-05,
"loss": 6.6108,
"mean_token_accuracy": 0.13751114308834075,
"num_tokens": 232140.0,
"step": 105
},
{
"entropy": 6.9101881980896,
"epoch": 0.012354691975065986,
"grad_norm": 1.09375,
"learning_rate": 5.45e-05,
"loss": 6.6995,
"mean_token_accuracy": 0.1259769469499588,
"num_tokens": 244186.0,
"step": 110
},
{
"entropy": 6.848400115966797,
"epoch": 0.012916268883023531,
"grad_norm": 1.1484375,
"learning_rate": 5.7e-05,
"loss": 6.5407,
"mean_token_accuracy": 0.13194369077682494,
"num_tokens": 254805.0,
"step": 115
},
{
"entropy": 6.698233985900879,
"epoch": 0.013477845790981074,
"grad_norm": 1.0546875,
"learning_rate": 5.9499999999999996e-05,
"loss": 6.5033,
"mean_token_accuracy": 0.13761864304542543,
"num_tokens": 265055.0,
"step": 120
},
{
"entropy": 6.791996097564697,
"epoch": 0.01403942269893862,
"grad_norm": 0.984375,
"learning_rate": 6.2e-05,
"loss": 6.4269,
"mean_token_accuracy": 0.13726359307765962,
"num_tokens": 275850.0,
"step": 125
},
{
"entropy": 6.7355358600616455,
"epoch": 0.014600999606896165,
"grad_norm": 1.0703125,
"learning_rate": 6.450000000000001e-05,
"loss": 6.4766,
"mean_token_accuracy": 0.13020233884453775,
"num_tokens": 286802.0,
"step": 130
},
{
"entropy": 6.776710081100464,
"epoch": 0.015162576514853709,
"grad_norm": 1.1328125,
"learning_rate": 6.7e-05,
"loss": 6.5889,
"mean_token_accuracy": 0.12709178403019905,
"num_tokens": 297070.0,
"step": 135
},
{
"entropy": 6.857462978363037,
"epoch": 0.015724153422811252,
"grad_norm": 1.0078125,
"learning_rate": 6.950000000000001e-05,
"loss": 6.5067,
"mean_token_accuracy": 0.1300993725657463,
"num_tokens": 309617.0,
"step": 140
},
{
"entropy": 6.7578826427459715,
"epoch": 0.0162857303307688,
"grad_norm": 1.25,
"learning_rate": 7.2e-05,
"loss": 6.5446,
"mean_token_accuracy": 0.12609508857131005,
"num_tokens": 321310.0,
"step": 145
},
{
"entropy": 6.758755540847778,
"epoch": 0.016847307238726343,
"grad_norm": 1.15625,
"learning_rate": 7.45e-05,
"loss": 6.4748,
"mean_token_accuracy": 0.1324310265481472,
"num_tokens": 332243.0,
"step": 150
},
{
"entropy": 6.731501817703247,
"epoch": 0.017408884146683887,
"grad_norm": 1.0859375,
"learning_rate": 7.7e-05,
"loss": 6.5059,
"mean_token_accuracy": 0.12957077771425246,
"num_tokens": 342847.0,
"step": 155
},
{
"entropy": 6.806047821044922,
"epoch": 0.017970461054641434,
"grad_norm": 1.1875,
"learning_rate": 7.950000000000001e-05,
"loss": 6.4264,
"mean_token_accuracy": 0.13830571845173836,
"num_tokens": 353570.0,
"step": 160
},
{
"entropy": 6.7714825630187985,
"epoch": 0.018532037962598977,
"grad_norm": 1.3125,
"learning_rate": 8.2e-05,
"loss": 6.4818,
"mean_token_accuracy": 0.1378109060227871,
"num_tokens": 364330.0,
"step": 165
},
{
"entropy": 6.691267204284668,
"epoch": 0.01909361487055652,
"grad_norm": 1.265625,
"learning_rate": 8.450000000000001e-05,
"loss": 6.5244,
"mean_token_accuracy": 0.13161558136343957,
"num_tokens": 374930.0,
"step": 170
},
{
"entropy": 6.84858512878418,
"epoch": 0.019655191778514068,
"grad_norm": 1.234375,
"learning_rate": 8.7e-05,
"loss": 6.4959,
"mean_token_accuracy": 0.1296277239918709,
"num_tokens": 384690.0,
"step": 175
},
{
"entropy": 6.797068119049072,
"epoch": 0.020216768686471612,
"grad_norm": 1.3359375,
"learning_rate": 8.95e-05,
"loss": 6.5289,
"mean_token_accuracy": 0.13395457416772844,
"num_tokens": 395753.0,
"step": 180
},
{
"entropy": 6.726787281036377,
"epoch": 0.020778345594429155,
"grad_norm": 1.296875,
"learning_rate": 9.2e-05,
"loss": 6.4848,
"mean_token_accuracy": 0.1342456102371216,
"num_tokens": 407049.0,
"step": 185
},
{
"entropy": 6.7613801002502445,
"epoch": 0.021339922502386702,
"grad_norm": 1.34375,
"learning_rate": 9.45e-05,
"loss": 6.511,
"mean_token_accuracy": 0.13599891290068628,
"num_tokens": 417942.0,
"step": 190
},
{
"entropy": 6.74392523765564,
"epoch": 0.021901499410344246,
"grad_norm": 1.28125,
"learning_rate": 9.7e-05,
"loss": 6.5347,
"mean_token_accuracy": 0.13084860444068908,
"num_tokens": 428151.0,
"step": 195
},
{
"entropy": 6.873997688293457,
"epoch": 0.022463076318301793,
"grad_norm": 1.125,
"learning_rate": 9.95e-05,
"loss": 6.6172,
"mean_token_accuracy": 0.12304027304053307,
"num_tokens": 439837.0,
"step": 200
},
{
"entropy": 6.6801533699035645,
"epoch": 0.023024653226259337,
"grad_norm": 1.1015625,
"learning_rate": 0.000102,
"loss": 6.4234,
"mean_token_accuracy": 0.129472017288208,
"num_tokens": 450357.0,
"step": 205
},
{
"entropy": 6.85008544921875,
"epoch": 0.02358623013421688,
"grad_norm": 1.2578125,
"learning_rate": 0.00010449999999999999,
"loss": 6.4674,
"mean_token_accuracy": 0.1346777245402336,
"num_tokens": 461174.0,
"step": 210
},
{
"entropy": 6.803839302062988,
"epoch": 0.024147807042174427,
"grad_norm": 1.140625,
"learning_rate": 0.000107,
"loss": 6.5461,
"mean_token_accuracy": 0.132686547935009,
"num_tokens": 472785.0,
"step": 215
},
{
"entropy": 6.726331377029419,
"epoch": 0.02470938395013197,
"grad_norm": 1.109375,
"learning_rate": 0.0001095,
"loss": 6.5429,
"mean_token_accuracy": 0.1351382851600647,
"num_tokens": 483248.0,
"step": 220
},
{
"entropy": 6.723070573806763,
"epoch": 0.025270960858089515,
"grad_norm": 1.171875,
"learning_rate": 0.000112,
"loss": 6.3822,
"mean_token_accuracy": 0.1378464035689831,
"num_tokens": 494690.0,
"step": 225
},
{
"entropy": 6.7689587593078615,
"epoch": 0.025832537766047062,
"grad_norm": 1.09375,
"learning_rate": 0.0001145,
"loss": 6.6013,
"mean_token_accuracy": 0.1281190924346447,
"num_tokens": 506192.0,
"step": 230
},
{
"entropy": 6.8111066818237305,
"epoch": 0.026394114674004605,
"grad_norm": 1.1875,
"learning_rate": 0.00011700000000000001,
"loss": 6.5057,
"mean_token_accuracy": 0.13290760517120362,
"num_tokens": 517152.0,
"step": 235
},
{
"entropy": 6.639764785766602,
"epoch": 0.02695569158196215,
"grad_norm": 1.203125,
"learning_rate": 0.00011949999999999999,
"loss": 6.4202,
"mean_token_accuracy": 0.13568630814552307,
"num_tokens": 527836.0,
"step": 240
},
{
"entropy": 6.822380447387696,
"epoch": 0.027517268489919696,
"grad_norm": 1.1015625,
"learning_rate": 0.000122,
"loss": 6.5896,
"mean_token_accuracy": 0.13045021742582322,
"num_tokens": 539491.0,
"step": 245
},
{
"entropy": 6.886163139343262,
"epoch": 0.02807884539787724,
"grad_norm": 1.015625,
"learning_rate": 0.0001245,
"loss": 6.5679,
"mean_token_accuracy": 0.13087331652641296,
"num_tokens": 550816.0,
"step": 250
},
{
"entropy": 6.840306997299194,
"epoch": 0.028640422305834783,
"grad_norm": 1.140625,
"learning_rate": 0.000127,
"loss": 6.624,
"mean_token_accuracy": 0.12609125748276712,
"num_tokens": 561566.0,
"step": 255
},
{
"entropy": 6.74591097831726,
"epoch": 0.02920199921379233,
"grad_norm": 1.171875,
"learning_rate": 0.0001295,
"loss": 6.5495,
"mean_token_accuracy": 0.12940011620521547,
"num_tokens": 573032.0,
"step": 260
},
{
"entropy": 6.815591430664062,
"epoch": 0.029763576121749874,
"grad_norm": 1.0078125,
"learning_rate": 0.000132,
"loss": 6.4639,
"mean_token_accuracy": 0.13757430762052536,
"num_tokens": 584140.0,
"step": 265
},
{
"entropy": 6.750753736495971,
"epoch": 0.030325153029707418,
"grad_norm": 1.21875,
"learning_rate": 0.00013450000000000002,
"loss": 6.5428,
"mean_token_accuracy": 0.13629817366600036,
"num_tokens": 595097.0,
"step": 270
},
{
"entropy": 6.8167486667633055,
"epoch": 0.030886729937664965,
"grad_norm": 1.2890625,
"learning_rate": 0.00013700000000000002,
"loss": 6.5745,
"mean_token_accuracy": 0.13026740923523902,
"num_tokens": 606797.0,
"step": 275
},
{
"entropy": 6.7974999904632565,
"epoch": 0.031448306845622505,
"grad_norm": 1.1875,
"learning_rate": 0.0001395,
"loss": 6.5443,
"mean_token_accuracy": 0.1298480361700058,
"num_tokens": 618187.0,
"step": 280
},
{
"entropy": 6.741901540756226,
"epoch": 0.032009883753580055,
"grad_norm": 1.0546875,
"learning_rate": 0.00014199999999999998,
"loss": 6.4842,
"mean_token_accuracy": 0.13356513679027557,
"num_tokens": 629829.0,
"step": 285
},
{
"entropy": 6.724414396286011,
"epoch": 0.0325714606615376,
"grad_norm": 1.203125,
"learning_rate": 0.0001445,
"loss": 6.5091,
"mean_token_accuracy": 0.13159143030643464,
"num_tokens": 640914.0,
"step": 290
},
{
"entropy": 6.815596628189087,
"epoch": 0.03313303756949514,
"grad_norm": 1.2421875,
"learning_rate": 0.000147,
"loss": 6.5293,
"mean_token_accuracy": 0.13000186085700988,
"num_tokens": 653057.0,
"step": 295
},
{
"entropy": 6.669460153579712,
"epoch": 0.033694614477452686,
"grad_norm": 1.1796875,
"learning_rate": 0.0001495,
"loss": 6.4485,
"mean_token_accuracy": 0.14079594984650612,
"num_tokens": 663437.0,
"step": 300
},
{
"entropy": 6.751016330718994,
"epoch": 0.03425619138541023,
"grad_norm": 1.3046875,
"learning_rate": 0.000152,
"loss": 6.5369,
"mean_token_accuracy": 0.13079639598727227,
"num_tokens": 675416.0,
"step": 305
},
{
"entropy": 6.6445365905761715,
"epoch": 0.03481776829336777,
"grad_norm": 1.203125,
"learning_rate": 0.00015450000000000001,
"loss": 6.3375,
"mean_token_accuracy": 0.13328307271003723,
"num_tokens": 686633.0,
"step": 310
},
{
"entropy": 6.61797342300415,
"epoch": 0.035379345201325324,
"grad_norm": 1.171875,
"learning_rate": 0.000157,
"loss": 6.4539,
"mean_token_accuracy": 0.13328708410263063,
"num_tokens": 698065.0,
"step": 315
},
{
"entropy": 6.7716223239898685,
"epoch": 0.03594092210928287,
"grad_norm": 1.1875,
"learning_rate": 0.0001595,
"loss": 6.4572,
"mean_token_accuracy": 0.13270723596215248,
"num_tokens": 708652.0,
"step": 320
},
{
"entropy": 6.836631536483765,
"epoch": 0.03650249901724041,
"grad_norm": 1.234375,
"learning_rate": 0.000162,
"loss": 6.633,
"mean_token_accuracy": 0.12807837575674058,
"num_tokens": 720199.0,
"step": 325
},
{
"entropy": 6.755860805511475,
"epoch": 0.037064075925197955,
"grad_norm": 1.0859375,
"learning_rate": 0.00016450000000000001,
"loss": 6.594,
"mean_token_accuracy": 0.12865597382187843,
"num_tokens": 732964.0,
"step": 330
},
{
"entropy": 6.682922077178955,
"epoch": 0.0376256528331555,
"grad_norm": 1.25,
"learning_rate": 0.00016700000000000002,
"loss": 6.3655,
"mean_token_accuracy": 0.13459469601511956,
"num_tokens": 743528.0,
"step": 335
},
{
"entropy": 6.717207574844361,
"epoch": 0.03818722974111304,
"grad_norm": 1.203125,
"learning_rate": 0.00016950000000000003,
"loss": 6.4622,
"mean_token_accuracy": 0.13458281084895135,
"num_tokens": 754365.0,
"step": 340
},
{
"entropy": 6.578363466262817,
"epoch": 0.03874880664907059,
"grad_norm": 1.1640625,
"learning_rate": 0.00017199999999999998,
"loss": 6.3886,
"mean_token_accuracy": 0.12879974991083146,
"num_tokens": 766094.0,
"step": 345
},
{
"entropy": 6.680249881744385,
"epoch": 0.039310383557028136,
"grad_norm": 1.21875,
"learning_rate": 0.00017449999999999999,
"loss": 6.4493,
"mean_token_accuracy": 0.13100041821599007,
"num_tokens": 777275.0,
"step": 350
},
{
"entropy": 6.72165961265564,
"epoch": 0.03987196046498568,
"grad_norm": 1.1328125,
"learning_rate": 0.000177,
"loss": 6.5285,
"mean_token_accuracy": 0.13037424385547638,
"num_tokens": 789436.0,
"step": 355
},
{
"entropy": 6.754672527313232,
"epoch": 0.040433537372943223,
"grad_norm": 1.1953125,
"learning_rate": 0.0001795,
"loss": 6.5537,
"mean_token_accuracy": 0.1309170365333557,
"num_tokens": 800164.0,
"step": 360
},
{
"entropy": 6.621752738952637,
"epoch": 0.04099511428090077,
"grad_norm": 1.1484375,
"learning_rate": 0.000182,
"loss": 6.4161,
"mean_token_accuracy": 0.13131387978792192,
"num_tokens": 811797.0,
"step": 365
},
{
"entropy": 6.678555488586426,
"epoch": 0.04155669118885831,
"grad_norm": 1.1796875,
"learning_rate": 0.0001845,
"loss": 6.4265,
"mean_token_accuracy": 0.13252343758940696,
"num_tokens": 824158.0,
"step": 370
},
{
"entropy": 6.68765778541565,
"epoch": 0.04211826809681586,
"grad_norm": 1.0703125,
"learning_rate": 0.000187,
"loss": 6.4324,
"mean_token_accuracy": 0.13959610983729362,
"num_tokens": 835803.0,
"step": 375
},
{
"entropy": 6.701789617538452,
"epoch": 0.042679845004773405,
"grad_norm": 1.078125,
"learning_rate": 0.0001895,
"loss": 6.5153,
"mean_token_accuracy": 0.12811891362071037,
"num_tokens": 848106.0,
"step": 380
},
{
"entropy": 6.676432752609253,
"epoch": 0.04324142191273095,
"grad_norm": 1.203125,
"learning_rate": 0.000192,
"loss": 6.5047,
"mean_token_accuracy": 0.13056822791695594,
"num_tokens": 859101.0,
"step": 385
},
{
"entropy": 6.634853458404541,
"epoch": 0.04380299882068849,
"grad_norm": 1.109375,
"learning_rate": 0.0001945,
"loss": 6.363,
"mean_token_accuracy": 0.13974561989307405,
"num_tokens": 870309.0,
"step": 390
},
{
"entropy": 6.6745370388031,
"epoch": 0.044364575728646036,
"grad_norm": 1.2890625,
"learning_rate": 0.00019700000000000002,
"loss": 6.4732,
"mean_token_accuracy": 0.13085832819342613,
"num_tokens": 881515.0,
"step": 395
},
{
"entropy": 6.707674884796143,
"epoch": 0.044926152636603586,
"grad_norm": 1.2734375,
"learning_rate": 0.00019950000000000002,
"loss": 6.4392,
"mean_token_accuracy": 0.13565397635102272,
"num_tokens": 892129.0,
"step": 400
},
{
"entropy": 6.609837865829467,
"epoch": 0.04548772954456113,
"grad_norm": 1.2109375,
"learning_rate": 0.000202,
"loss": 6.46,
"mean_token_accuracy": 0.13184903636574746,
"num_tokens": 903186.0,
"step": 405
},
{
"entropy": 6.642455530166626,
"epoch": 0.046049306452518673,
"grad_norm": 1.375,
"learning_rate": 0.00020449999999999998,
"loss": 6.4299,
"mean_token_accuracy": 0.13762376606464385,
"num_tokens": 913471.0,
"step": 410
},
{
"entropy": 6.7312219619750975,
"epoch": 0.04661088336047622,
"grad_norm": 1.3359375,
"learning_rate": 0.000207,
"loss": 6.4316,
"mean_token_accuracy": 0.13134424537420272,
"num_tokens": 924283.0,
"step": 415
},
{
"entropy": 6.647944927215576,
"epoch": 0.04717246026843376,
"grad_norm": 1.1640625,
"learning_rate": 0.0002095,
"loss": 6.4697,
"mean_token_accuracy": 0.13021936267614365,
"num_tokens": 934975.0,
"step": 420
},
{
"entropy": 6.60395565032959,
"epoch": 0.047734037176391304,
"grad_norm": 1.3515625,
"learning_rate": 0.000212,
"loss": 6.3365,
"mean_token_accuracy": 0.14171250090003013,
"num_tokens": 945587.0,
"step": 425
},
{
"entropy": 6.608612155914306,
"epoch": 0.048295614084348855,
"grad_norm": 1.2421875,
"learning_rate": 0.0002145,
"loss": 6.3925,
"mean_token_accuracy": 0.13692179098725318,
"num_tokens": 955935.0,
"step": 430
},
{
"entropy": 6.589728116989136,
"epoch": 0.0488571909923064,
"grad_norm": 1.109375,
"learning_rate": 0.00021700000000000002,
"loss": 6.389,
"mean_token_accuracy": 0.13498958870768546,
"num_tokens": 967960.0,
"step": 435
},
{
"entropy": 6.663047790527344,
"epoch": 0.04941876790026394,
"grad_norm": 1.1015625,
"learning_rate": 0.0002195,
"loss": 6.4281,
"mean_token_accuracy": 0.13640725910663604,
"num_tokens": 979680.0,
"step": 440
},
{
"entropy": 6.576627540588379,
"epoch": 0.049980344808221486,
"grad_norm": 0.99609375,
"learning_rate": 0.000222,
"loss": 6.512,
"mean_token_accuracy": 0.13283532932400705,
"num_tokens": 990600.0,
"step": 445
},
{
"entropy": 6.744882965087891,
"epoch": 0.05054192171617903,
"grad_norm": 1.2890625,
"learning_rate": 0.0002245,
"loss": 6.5571,
"mean_token_accuracy": 0.13367009460926055,
"num_tokens": 1002331.0,
"step": 450
},
{
"entropy": 6.614725875854492,
"epoch": 0.05110349862413657,
"grad_norm": 1.2734375,
"learning_rate": 0.00022700000000000002,
"loss": 6.3702,
"mean_token_accuracy": 0.1379753105342388,
"num_tokens": 1013473.0,
"step": 455
},
{
"entropy": 6.682910823822022,
"epoch": 0.051665075532094124,
"grad_norm": 1.1171875,
"learning_rate": 0.00022950000000000002,
"loss": 6.4372,
"mean_token_accuracy": 0.1283923380076885,
"num_tokens": 1024434.0,
"step": 460
},
{
"entropy": 6.680650281906128,
"epoch": 0.05222665244005167,
"grad_norm": 1.1015625,
"learning_rate": 0.00023200000000000003,
"loss": 6.468,
"mean_token_accuracy": 0.12814455404877662,
"num_tokens": 1035452.0,
"step": 465
},
{
"entropy": 6.5947794914245605,
"epoch": 0.05278822934800921,
"grad_norm": 1.1328125,
"learning_rate": 0.00023449999999999998,
"loss": 6.3441,
"mean_token_accuracy": 0.13335954174399375,
"num_tokens": 1046213.0,
"step": 470
},
{
"entropy": 6.454158306121826,
"epoch": 0.053349806255966754,
"grad_norm": 1.21875,
"learning_rate": 0.000237,
"loss": 6.3879,
"mean_token_accuracy": 0.1338328018784523,
"num_tokens": 1057469.0,
"step": 475
},
{
"entropy": 6.801644802093506,
"epoch": 0.0539113831639243,
"grad_norm": 1.15625,
"learning_rate": 0.0002395,
"loss": 6.5276,
"mean_token_accuracy": 0.13049064353108406,
"num_tokens": 1069543.0,
"step": 480
},
{
"entropy": 6.555569219589233,
"epoch": 0.05447296007188184,
"grad_norm": 1.1171875,
"learning_rate": 0.000242,
"loss": 6.3959,
"mean_token_accuracy": 0.13704427033662797,
"num_tokens": 1080476.0,
"step": 485
},
{
"entropy": 6.722979974746704,
"epoch": 0.05503453697983939,
"grad_norm": 1.1875,
"learning_rate": 0.0002445,
"loss": 6.514,
"mean_token_accuracy": 0.13329028636217116,
"num_tokens": 1091700.0,
"step": 490
},
{
"entropy": 6.5727941513061525,
"epoch": 0.055596113887796936,
"grad_norm": 1.09375,
"learning_rate": 0.000247,
"loss": 6.4015,
"mean_token_accuracy": 0.1319067433476448,
"num_tokens": 1103056.0,
"step": 495
},
{
"entropy": 6.674144220352173,
"epoch": 0.05615769079575448,
"grad_norm": 1.2265625,
"learning_rate": 0.0002495,
"loss": 6.4715,
"mean_token_accuracy": 0.13558261916041375,
"num_tokens": 1114836.0,
"step": 500
},
{
"entropy": 6.547731256484985,
"epoch": 0.05671926770371202,
"grad_norm": 1.1640625,
"learning_rate": 0.000252,
"loss": 6.3698,
"mean_token_accuracy": 0.13027856275439262,
"num_tokens": 1125993.0,
"step": 505
},
{
"entropy": 6.658946466445923,
"epoch": 0.05728084461166957,
"grad_norm": 1.296875,
"learning_rate": 0.0002545,
"loss": 6.4507,
"mean_token_accuracy": 0.13343697637319565,
"num_tokens": 1136719.0,
"step": 510
},
{
"entropy": 6.56524429321289,
"epoch": 0.05784242151962711,
"grad_norm": 1.3046875,
"learning_rate": 0.000257,
"loss": 6.338,
"mean_token_accuracy": 0.137813039124012,
"num_tokens": 1146947.0,
"step": 515
},
{
"entropy": 6.607427501678467,
"epoch": 0.05840399842758466,
"grad_norm": 1.4140625,
"learning_rate": 0.0002595,
"loss": 6.4617,
"mean_token_accuracy": 0.1326425388455391,
"num_tokens": 1157974.0,
"step": 520
},
{
"entropy": 6.560697078704834,
"epoch": 0.058965575335542204,
"grad_norm": 1.265625,
"learning_rate": 0.000262,
"loss": 6.4036,
"mean_token_accuracy": 0.13357415795326233,
"num_tokens": 1168573.0,
"step": 525
},
{
"entropy": 6.574508571624756,
"epoch": 0.05952715224349975,
"grad_norm": 1.21875,
"learning_rate": 0.00026450000000000003,
"loss": 6.2877,
"mean_token_accuracy": 0.13817285895347595,
"num_tokens": 1178446.0,
"step": 530
},
{
"entropy": 6.566018009185791,
"epoch": 0.06008872915145729,
"grad_norm": 1.1484375,
"learning_rate": 0.00026700000000000004,
"loss": 6.3405,
"mean_token_accuracy": 0.13772412464022638,
"num_tokens": 1190558.0,
"step": 535
},
{
"entropy": 6.564810800552368,
"epoch": 0.060650306059414835,
"grad_norm": 1.3125,
"learning_rate": 0.00026950000000000005,
"loss": 6.4939,
"mean_token_accuracy": 0.1318053610622883,
"num_tokens": 1201929.0,
"step": 540
},
{
"entropy": 6.502597904205322,
"epoch": 0.06121188296737238,
"grad_norm": 1.109375,
"learning_rate": 0.00027200000000000005,
"loss": 6.3444,
"mean_token_accuracy": 0.13646928146481513,
"num_tokens": 1213155.0,
"step": 545
},
{
"entropy": 6.680653238296509,
"epoch": 0.06177345987532993,
"grad_norm": 1.25,
"learning_rate": 0.0002745,
"loss": 6.4053,
"mean_token_accuracy": 0.13270772770047187,
"num_tokens": 1223961.0,
"step": 550
},
{
"entropy": 6.381764841079712,
"epoch": 0.06233503678328747,
"grad_norm": 1.2421875,
"learning_rate": 0.000277,
"loss": 6.2988,
"mean_token_accuracy": 0.1437389425933361,
"num_tokens": 1233888.0,
"step": 555
},
{
"entropy": 6.593340253829956,
"epoch": 0.06289661369124501,
"grad_norm": 1.1875,
"learning_rate": 0.0002795,
"loss": 6.4209,
"mean_token_accuracy": 0.12833445593714715,
"num_tokens": 1244405.0,
"step": 560
},
{
"entropy": 6.558970785140991,
"epoch": 0.06345819059920256,
"grad_norm": 0.9453125,
"learning_rate": 0.00028199999999999997,
"loss": 6.3654,
"mean_token_accuracy": 0.13667553216218947,
"num_tokens": 1257099.0,
"step": 565
},
{
"entropy": 6.631209850311279,
"epoch": 0.06401976750716011,
"grad_norm": 1.28125,
"learning_rate": 0.0002845,
"loss": 6.4699,
"mean_token_accuracy": 0.13471307903528212,
"num_tokens": 1267965.0,
"step": 570
},
{
"entropy": 6.516822481155396,
"epoch": 0.06458134441511765,
"grad_norm": 1.171875,
"learning_rate": 0.000287,
"loss": 6.3936,
"mean_token_accuracy": 0.13442024663090707,
"num_tokens": 1280075.0,
"step": 575
},
{
"entropy": 6.609190845489502,
"epoch": 0.0651429213230752,
"grad_norm": 1.421875,
"learning_rate": 0.0002895,
"loss": 6.3864,
"mean_token_accuracy": 0.13422752767801285,
"num_tokens": 1290574.0,
"step": 580
},
{
"entropy": 6.4985980033874515,
"epoch": 0.06570449823103273,
"grad_norm": 1.140625,
"learning_rate": 0.000292,
"loss": 6.3732,
"mean_token_accuracy": 0.13692231178283693,
"num_tokens": 1302680.0,
"step": 585
},
{
"entropy": 6.57489275932312,
"epoch": 0.06626607513899029,
"grad_norm": 1.1171875,
"learning_rate": 0.0002945,
"loss": 6.3562,
"mean_token_accuracy": 0.13580932915210725,
"num_tokens": 1314577.0,
"step": 590
},
{
"entropy": 6.555920886993408,
"epoch": 0.06682765204694784,
"grad_norm": 1.109375,
"learning_rate": 0.000297,
"loss": 6.4958,
"mean_token_accuracy": 0.13412591591477394,
"num_tokens": 1326560.0,
"step": 595
},
{
"entropy": 6.718612575531006,
"epoch": 0.06738922895490537,
"grad_norm": 1.125,
"learning_rate": 0.0002995,
"loss": 6.553,
"mean_token_accuracy": 0.12826075702905654,
"num_tokens": 1339105.0,
"step": 600
},
{
"entropy": 6.350273942947387,
"epoch": 0.06795080586286292,
"grad_norm": 1.125,
"learning_rate": 0.000302,
"loss": 6.2291,
"mean_token_accuracy": 0.13848018050193786,
"num_tokens": 1349914.0,
"step": 605
},
{
"entropy": 6.579149675369263,
"epoch": 0.06851238277082046,
"grad_norm": 1.15625,
"learning_rate": 0.0003045,
"loss": 6.4524,
"mean_token_accuracy": 0.13018497154116632,
"num_tokens": 1361700.0,
"step": 610
},
{
"entropy": 6.584105443954468,
"epoch": 0.06907395967877801,
"grad_norm": 1.140625,
"learning_rate": 0.000307,
"loss": 6.456,
"mean_token_accuracy": 0.13250951543450357,
"num_tokens": 1373103.0,
"step": 615
},
{
"entropy": 6.421159172058106,
"epoch": 0.06963553658673555,
"grad_norm": 1.34375,
"learning_rate": 0.0003095,
"loss": 6.2674,
"mean_token_accuracy": 0.13772882893681526,
"num_tokens": 1383742.0,
"step": 620
},
{
"entropy": 6.502809143066406,
"epoch": 0.0701971134946931,
"grad_norm": 1.03125,
"learning_rate": 0.000312,
"loss": 6.4221,
"mean_token_accuracy": 0.13373162150382994,
"num_tokens": 1396656.0,
"step": 625
},
{
"entropy": 6.531112861633301,
"epoch": 0.07075869040265065,
"grad_norm": 1.1640625,
"learning_rate": 0.0003145,
"loss": 6.2977,
"mean_token_accuracy": 0.1519811674952507,
"num_tokens": 1406709.0,
"step": 630
},
{
"entropy": 6.507521104812622,
"epoch": 0.07132026731060818,
"grad_norm": 1.09375,
"learning_rate": 0.000317,
"loss": 6.3298,
"mean_token_accuracy": 0.14185064658522606,
"num_tokens": 1417793.0,
"step": 635
},
{
"entropy": 6.518734645843506,
"epoch": 0.07188184421856574,
"grad_norm": 1.0234375,
"learning_rate": 0.0003195,
"loss": 6.418,
"mean_token_accuracy": 0.14041453301906587,
"num_tokens": 1429093.0,
"step": 640
},
{
"entropy": 6.450685930252075,
"epoch": 0.07244342112652327,
"grad_norm": 1.015625,
"learning_rate": 0.000322,
"loss": 6.3521,
"mean_token_accuracy": 0.14151380732655525,
"num_tokens": 1440907.0,
"step": 645
},
{
"entropy": 6.5532149314880375,
"epoch": 0.07300499803448082,
"grad_norm": 1.15625,
"learning_rate": 0.00032450000000000003,
"loss": 6.4215,
"mean_token_accuracy": 0.13366219252347947,
"num_tokens": 1452089.0,
"step": 650
},
{
"entropy": 6.611659145355224,
"epoch": 0.07356657494243837,
"grad_norm": 1.078125,
"learning_rate": 0.00032700000000000003,
"loss": 6.432,
"mean_token_accuracy": 0.1317155674099922,
"num_tokens": 1464378.0,
"step": 655
},
{
"entropy": 6.506606101989746,
"epoch": 0.07412815185039591,
"grad_norm": 1.125,
"learning_rate": 0.00032950000000000004,
"loss": 6.3608,
"mean_token_accuracy": 0.13717821538448333,
"num_tokens": 1476678.0,
"step": 660
},
{
"entropy": 6.418495512008667,
"epoch": 0.07468972875835346,
"grad_norm": 1.1796875,
"learning_rate": 0.00033200000000000005,
"loss": 6.199,
"mean_token_accuracy": 0.13983268588781356,
"num_tokens": 1486808.0,
"step": 665
},
{
"entropy": 6.513941431045533,
"epoch": 0.075251305666311,
"grad_norm": 1.0390625,
"learning_rate": 0.00033450000000000005,
"loss": 6.5064,
"mean_token_accuracy": 0.13336510956287384,
"num_tokens": 1499281.0,
"step": 670
},
{
"entropy": 6.550311517715454,
"epoch": 0.07581288257426855,
"grad_norm": 1.4296875,
"learning_rate": 0.000337,
"loss": 6.3971,
"mean_token_accuracy": 0.13824842870235443,
"num_tokens": 1510202.0,
"step": 675
},
{
"entropy": 6.478364896774292,
"epoch": 0.07637445948222608,
"grad_norm": 1.0546875,
"learning_rate": 0.0003395,
"loss": 6.2777,
"mean_token_accuracy": 0.14033918008208274,
"num_tokens": 1521492.0,
"step": 680
},
{
"entropy": 6.363365268707275,
"epoch": 0.07693603639018363,
"grad_norm": 1.0625,
"learning_rate": 0.000342,
"loss": 6.2901,
"mean_token_accuracy": 0.14578976929187776,
"num_tokens": 1532856.0,
"step": 685
},
{
"entropy": 6.4714092254638675,
"epoch": 0.07749761329814119,
"grad_norm": 1.109375,
"learning_rate": 0.00034449999999999997,
"loss": 6.293,
"mean_token_accuracy": 0.13803935647010804,
"num_tokens": 1543408.0,
"step": 690
},
{
"entropy": 6.445067262649536,
"epoch": 0.07805919020609872,
"grad_norm": 1.171875,
"learning_rate": 0.000347,
"loss": 6.3314,
"mean_token_accuracy": 0.13593096658587456,
"num_tokens": 1555474.0,
"step": 695
},
{
"entropy": 6.49391450881958,
"epoch": 0.07862076711405627,
"grad_norm": 1.0859375,
"learning_rate": 0.0003495,
"loss": 6.4259,
"mean_token_accuracy": 0.1368790291249752,
"num_tokens": 1567484.0,
"step": 700
},
{
"entropy": 6.461107492446899,
"epoch": 0.07918234402201381,
"grad_norm": 1.2109375,
"learning_rate": 0.000352,
"loss": 6.2683,
"mean_token_accuracy": 0.14710718691349028,
"num_tokens": 1578204.0,
"step": 705
},
{
"entropy": 6.4600510597229,
"epoch": 0.07974392092997136,
"grad_norm": 1.2734375,
"learning_rate": 0.0003545,
"loss": 6.2575,
"mean_token_accuracy": 0.14133917540311813,
"num_tokens": 1589012.0,
"step": 710
},
{
"entropy": 6.342525625228882,
"epoch": 0.08030549783792891,
"grad_norm": 1.2734375,
"learning_rate": 0.000357,
"loss": 6.2677,
"mean_token_accuracy": 0.14088504314422606,
"num_tokens": 1601099.0,
"step": 715
},
{
"entropy": 6.518351411819458,
"epoch": 0.08086707474588645,
"grad_norm": 1.203125,
"learning_rate": 0.0003595,
"loss": 6.4327,
"mean_token_accuracy": 0.13333295062184333,
"num_tokens": 1612814.0,
"step": 720
},
{
"entropy": 6.471275472640992,
"epoch": 0.081428651653844,
"grad_norm": 1.265625,
"learning_rate": 0.000362,
"loss": 6.3316,
"mean_token_accuracy": 0.13668572530150414,
"num_tokens": 1624953.0,
"step": 725
},
{
"entropy": 6.470378446578979,
"epoch": 0.08199022856180153,
"grad_norm": 1.1171875,
"learning_rate": 0.0003645,
"loss": 6.3809,
"mean_token_accuracy": 0.13890909254550934,
"num_tokens": 1635673.0,
"step": 730
},
{
"entropy": 6.376466226577759,
"epoch": 0.08255180546975908,
"grad_norm": 1.0859375,
"learning_rate": 0.000367,
"loss": 6.2719,
"mean_token_accuracy": 0.1436118446290493,
"num_tokens": 1647282.0,
"step": 735
},
{
"entropy": 6.441848707199097,
"epoch": 0.08311338237771662,
"grad_norm": 1.140625,
"learning_rate": 0.0003695,
"loss": 6.2867,
"mean_token_accuracy": 0.14489989280700682,
"num_tokens": 1658426.0,
"step": 740
},
{
"entropy": 6.345063209533691,
"epoch": 0.08367495928567417,
"grad_norm": 1.265625,
"learning_rate": 0.000372,
"loss": 6.3131,
"mean_token_accuracy": 0.14454299807548524,
"num_tokens": 1669958.0,
"step": 745
},
{
"entropy": 6.474502325057983,
"epoch": 0.08423653619363172,
"grad_norm": 1.125,
"learning_rate": 0.0003745,
"loss": 6.3825,
"mean_token_accuracy": 0.13400474563241005,
"num_tokens": 1682409.0,
"step": 750
},
{
"entropy": 6.442701625823974,
"epoch": 0.08479811310158926,
"grad_norm": 1.09375,
"learning_rate": 0.000377,
"loss": 6.3045,
"mean_token_accuracy": 0.14752128720283508,
"num_tokens": 1692409.0,
"step": 755
},
{
"entropy": 6.501216983795166,
"epoch": 0.08535969000954681,
"grad_norm": 1.1015625,
"learning_rate": 0.0003795,
"loss": 6.3863,
"mean_token_accuracy": 0.13712989687919616,
"num_tokens": 1705018.0,
"step": 760
},
{
"entropy": 6.35915904045105,
"epoch": 0.08592126691750435,
"grad_norm": 1.109375,
"learning_rate": 0.000382,
"loss": 6.3271,
"mean_token_accuracy": 0.1366030417382717,
"num_tokens": 1715625.0,
"step": 765
},
{
"entropy": 6.459257507324219,
"epoch": 0.0864828438254619,
"grad_norm": 1.1640625,
"learning_rate": 0.0003845,
"loss": 6.3949,
"mean_token_accuracy": 0.13145335242152215,
"num_tokens": 1727045.0,
"step": 770
},
{
"entropy": 6.523573160171509,
"epoch": 0.08704442073341945,
"grad_norm": 1.2109375,
"learning_rate": 0.00038700000000000003,
"loss": 6.4586,
"mean_token_accuracy": 0.13751968517899513,
"num_tokens": 1737374.0,
"step": 775
},
{
"entropy": 6.468965530395508,
"epoch": 0.08760599764137698,
"grad_norm": 1.109375,
"learning_rate": 0.00038950000000000003,
"loss": 6.4517,
"mean_token_accuracy": 0.1402681902050972,
"num_tokens": 1747788.0,
"step": 780
},
{
"entropy": 6.297554206848145,
"epoch": 0.08816757454933453,
"grad_norm": 1.1484375,
"learning_rate": 0.00039200000000000004,
"loss": 6.1546,
"mean_token_accuracy": 0.14846104085445405,
"num_tokens": 1757769.0,
"step": 785
},
{
"entropy": 6.381324434280396,
"epoch": 0.08872915145729207,
"grad_norm": 1.1171875,
"learning_rate": 0.00039450000000000005,
"loss": 6.2715,
"mean_token_accuracy": 0.13741729855537416,
"num_tokens": 1769246.0,
"step": 790
},
{
"entropy": 6.461723899841308,
"epoch": 0.08929072836524962,
"grad_norm": 1.140625,
"learning_rate": 0.00039700000000000005,
"loss": 6.4011,
"mean_token_accuracy": 0.13841823786497115,
"num_tokens": 1781306.0,
"step": 795
},
{
"entropy": 6.396069002151489,
"epoch": 0.08985230527320717,
"grad_norm": 1.046875,
"learning_rate": 0.0003995,
"loss": 6.2957,
"mean_token_accuracy": 0.13979703485965728,
"num_tokens": 1792888.0,
"step": 800
},
{
"entropy": 6.42486801147461,
"epoch": 0.09041388218116471,
"grad_norm": 1.21875,
"learning_rate": 0.000402,
"loss": 6.3201,
"mean_token_accuracy": 0.14163311570882797,
"num_tokens": 1802938.0,
"step": 805
},
{
"entropy": 6.467221689224243,
"epoch": 0.09097545908912226,
"grad_norm": 1.078125,
"learning_rate": 0.0004045,
"loss": 6.3228,
"mean_token_accuracy": 0.13640945702791213,
"num_tokens": 1814738.0,
"step": 810
},
{
"entropy": 6.423444128036499,
"epoch": 0.0915370359970798,
"grad_norm": 1.109375,
"learning_rate": 0.00040699999999999997,
"loss": 6.4079,
"mean_token_accuracy": 0.1382846400141716,
"num_tokens": 1825100.0,
"step": 815
},
{
"entropy": 6.2810968399047855,
"epoch": 0.09209861290503735,
"grad_norm": 1.203125,
"learning_rate": 0.0004095,
"loss": 6.2361,
"mean_token_accuracy": 0.1458885870873928,
"num_tokens": 1835879.0,
"step": 820
},
{
"entropy": 6.548891592025757,
"epoch": 0.09266018981299488,
"grad_norm": 1.1484375,
"learning_rate": 0.000412,
"loss": 6.3787,
"mean_token_accuracy": 0.13275464922189711,
"num_tokens": 1848352.0,
"step": 825
},
{
"entropy": 6.358658218383789,
"epoch": 0.09322176672095243,
"grad_norm": 1.15625,
"learning_rate": 0.0004145,
"loss": 6.3266,
"mean_token_accuracy": 0.13629675209522246,
"num_tokens": 1859687.0,
"step": 830
},
{
"entropy": 6.3769364833831785,
"epoch": 0.09378334362890998,
"grad_norm": 1.0703125,
"learning_rate": 0.000417,
"loss": 6.3259,
"mean_token_accuracy": 0.13939538821578026,
"num_tokens": 1871438.0,
"step": 835
},
{
"entropy": 6.295517444610596,
"epoch": 0.09434492053686752,
"grad_norm": 1.1484375,
"learning_rate": 0.0004195,
"loss": 6.195,
"mean_token_accuracy": 0.1484709143638611,
"num_tokens": 1881519.0,
"step": 840
},
{
"entropy": 6.5199895858764645,
"epoch": 0.09490649744482507,
"grad_norm": 1.109375,
"learning_rate": 0.000422,
"loss": 6.4049,
"mean_token_accuracy": 0.13696413338184357,
"num_tokens": 1892428.0,
"step": 845
},
{
"entropy": 6.450248670578003,
"epoch": 0.09546807435278261,
"grad_norm": 1.109375,
"learning_rate": 0.0004245,
"loss": 6.2804,
"mean_token_accuracy": 0.1426387369632721,
"num_tokens": 1903760.0,
"step": 850
},
{
"entropy": 6.360091924667358,
"epoch": 0.09602965126074016,
"grad_norm": 0.9765625,
"learning_rate": 0.000427,
"loss": 6.359,
"mean_token_accuracy": 0.13211985379457475,
"num_tokens": 1915430.0,
"step": 855
},
{
"entropy": 6.364841318130493,
"epoch": 0.09659122816869771,
"grad_norm": 1.2421875,
"learning_rate": 0.0004295,
"loss": 6.312,
"mean_token_accuracy": 0.13486656472086905,
"num_tokens": 1925709.0,
"step": 860
},
{
"entropy": 6.435139513015747,
"epoch": 0.09715280507665525,
"grad_norm": 1.0,
"learning_rate": 0.000432,
"loss": 6.3714,
"mean_token_accuracy": 0.13526980876922606,
"num_tokens": 1938148.0,
"step": 865
},
{
"entropy": 6.1814793109893795,
"epoch": 0.0977143819846128,
"grad_norm": 1.0,
"learning_rate": 0.0004345,
"loss": 6.2161,
"mean_token_accuracy": 0.14539682120084763,
"num_tokens": 1949028.0,
"step": 870
},
{
"entropy": 6.381429719924927,
"epoch": 0.09827595889257033,
"grad_norm": 1.0703125,
"learning_rate": 0.000437,
"loss": 6.2716,
"mean_token_accuracy": 0.14102880284190178,
"num_tokens": 1960353.0,
"step": 875
},
{
"entropy": 6.478352308273315,
"epoch": 0.09883753580052788,
"grad_norm": 1.1484375,
"learning_rate": 0.0004395,
"loss": 6.4072,
"mean_token_accuracy": 0.1384950503706932,
"num_tokens": 1972335.0,
"step": 880
},
{
"entropy": 6.287540864944458,
"epoch": 0.09939911270848542,
"grad_norm": 1.15625,
"learning_rate": 0.000442,
"loss": 6.2435,
"mean_token_accuracy": 0.13890341594815253,
"num_tokens": 1982851.0,
"step": 885
},
{
"entropy": 6.266198635101318,
"epoch": 0.09996068961644297,
"grad_norm": 1.0703125,
"learning_rate": 0.0004445,
"loss": 6.2305,
"mean_token_accuracy": 0.14235106781125068,
"num_tokens": 1993931.0,
"step": 890
},
{
"entropy": 6.41697473526001,
"epoch": 0.10052226652440052,
"grad_norm": 1.0390625,
"learning_rate": 0.000447,
"loss": 6.3631,
"mean_token_accuracy": 0.1405455969274044,
"num_tokens": 2004838.0,
"step": 895
},
{
"entropy": 6.252490043640137,
"epoch": 0.10108384343235806,
"grad_norm": 1.2265625,
"learning_rate": 0.00044950000000000003,
"loss": 6.1584,
"mean_token_accuracy": 0.14193521440029144,
"num_tokens": 2015620.0,
"step": 900
},
{
"entropy": 6.32470383644104,
"epoch": 0.10164542034031561,
"grad_norm": 1.140625,
"learning_rate": 0.00045200000000000004,
"loss": 6.2855,
"mean_token_accuracy": 0.1429427184164524,
"num_tokens": 2027489.0,
"step": 905
},
{
"entropy": 6.2411714553833,
"epoch": 0.10220699724827315,
"grad_norm": 1.2578125,
"learning_rate": 0.00045450000000000004,
"loss": 6.1741,
"mean_token_accuracy": 0.14678896516561507,
"num_tokens": 2038475.0,
"step": 910
},
{
"entropy": 6.2395162105560305,
"epoch": 0.1027685741562307,
"grad_norm": 1.09375,
"learning_rate": 0.00045700000000000005,
"loss": 6.1705,
"mean_token_accuracy": 0.1411990463733673,
"num_tokens": 2048938.0,
"step": 915
},
{
"entropy": 6.229062700271607,
"epoch": 0.10333015106418825,
"grad_norm": 0.98046875,
"learning_rate": 0.00045950000000000006,
"loss": 6.2775,
"mean_token_accuracy": 0.13909629583358765,
"num_tokens": 2060372.0,
"step": 920
},
{
"entropy": 6.308560037612915,
"epoch": 0.10389172797214578,
"grad_norm": 1.0625,
"learning_rate": 0.000462,
"loss": 6.2487,
"mean_token_accuracy": 0.13983388617634773,
"num_tokens": 2071558.0,
"step": 925
},
{
"entropy": 6.3064391136169435,
"epoch": 0.10445330488010333,
"grad_norm": 1.0703125,
"learning_rate": 0.0004645,
"loss": 6.2987,
"mean_token_accuracy": 0.1368556186556816,
"num_tokens": 2082350.0,
"step": 930
},
{
"entropy": 6.3746119976043705,
"epoch": 0.10501488178806087,
"grad_norm": 1.0390625,
"learning_rate": 0.000467,
"loss": 6.3368,
"mean_token_accuracy": 0.1350345179438591,
"num_tokens": 2093448.0,
"step": 935
},
{
"entropy": 6.493036365509033,
"epoch": 0.10557645869601842,
"grad_norm": 1.1328125,
"learning_rate": 0.0004695,
"loss": 6.3681,
"mean_token_accuracy": 0.13741556107997893,
"num_tokens": 2104171.0,
"step": 940
},
{
"entropy": 6.2003154277801515,
"epoch": 0.10613803560397596,
"grad_norm": 1.3125,
"learning_rate": 0.000472,
"loss": 6.2705,
"mean_token_accuracy": 0.1359943650662899,
"num_tokens": 2114496.0,
"step": 945
},
{
"entropy": 6.376204109191894,
"epoch": 0.10669961251193351,
"grad_norm": 1.1015625,
"learning_rate": 0.0004745,
"loss": 6.2566,
"mean_token_accuracy": 0.1408710964024067,
"num_tokens": 2125930.0,
"step": 950
},
{
"entropy": 6.257950973510742,
"epoch": 0.10726118941989106,
"grad_norm": 1.1640625,
"learning_rate": 0.000477,
"loss": 6.2212,
"mean_token_accuracy": 0.14402413070201875,
"num_tokens": 2136452.0,
"step": 955
},
{
"entropy": 6.282464265823364,
"epoch": 0.1078227663278486,
"grad_norm": 1.0546875,
"learning_rate": 0.0004795,
"loss": 6.3404,
"mean_token_accuracy": 0.13901666328310966,
"num_tokens": 2148233.0,
"step": 960
},
{
"entropy": 6.241464614868164,
"epoch": 0.10838434323580615,
"grad_norm": 1.1015625,
"learning_rate": 0.000482,
"loss": 6.1153,
"mean_token_accuracy": 0.1443318985402584,
"num_tokens": 2159475.0,
"step": 965
},
{
"entropy": 6.1993176460266115,
"epoch": 0.10894592014376368,
"grad_norm": 1.1484375,
"learning_rate": 0.0004845,
"loss": 6.2461,
"mean_token_accuracy": 0.1433636873960495,
"num_tokens": 2170042.0,
"step": 970
},
{
"entropy": 6.288603162765503,
"epoch": 0.10950749705172123,
"grad_norm": 1.2265625,
"learning_rate": 0.000487,
"loss": 6.2437,
"mean_token_accuracy": 0.13873123675584792,
"num_tokens": 2182234.0,
"step": 975
},
{
"entropy": 6.20459680557251,
"epoch": 0.11006907395967878,
"grad_norm": 1.1484375,
"learning_rate": 0.0004895,
"loss": 6.2355,
"mean_token_accuracy": 0.14040311500430108,
"num_tokens": 2192477.0,
"step": 980
},
{
"entropy": 6.208745098114013,
"epoch": 0.11063065086763632,
"grad_norm": 1.2265625,
"learning_rate": 0.000492,
"loss": 6.2087,
"mean_token_accuracy": 0.13950579315423967,
"num_tokens": 2203082.0,
"step": 985
},
{
"entropy": 6.242075681686401,
"epoch": 0.11119222777559387,
"grad_norm": 1.0234375,
"learning_rate": 0.0004945,
"loss": 6.178,
"mean_token_accuracy": 0.14390679448843002,
"num_tokens": 2213684.0,
"step": 990
},
{
"entropy": 6.218676996231079,
"epoch": 0.11175380468355141,
"grad_norm": 1.0703125,
"learning_rate": 0.000497,
"loss": 6.209,
"mean_token_accuracy": 0.13738821148872377,
"num_tokens": 2225524.0,
"step": 995
},
{
"entropy": 6.2625988006591795,
"epoch": 0.11231538159150896,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995,
"loss": 6.1674,
"mean_token_accuracy": 0.14143801629543304,
"num_tokens": 2236004.0,
"step": 1000
},
{
"entropy": 6.335353660583496,
"epoch": 0.1128769584994665,
"grad_norm": 1.0625,
"learning_rate": 0.000499999997707492,
"loss": 6.3267,
"mean_token_accuracy": 0.13301377221941948,
"num_tokens": 2247730.0,
"step": 1005
},
{
"entropy": 6.29675931930542,
"epoch": 0.11343853540742405,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999999883941787,
"loss": 6.2836,
"mean_token_accuracy": 0.14176394045352936,
"num_tokens": 2260132.0,
"step": 1010
},
{
"entropy": 6.2044346809387205,
"epoch": 0.1140001123153816,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999999719167785,
"loss": 6.2477,
"mean_token_accuracy": 0.14374205321073533,
"num_tokens": 2270060.0,
"step": 1015
},
{
"entropy": 6.291596841812134,
"epoch": 0.11456168922333913,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999999482752917,
"loss": 6.2676,
"mean_token_accuracy": 0.14092471450567245,
"num_tokens": 2281568.0,
"step": 1020
},
{
"entropy": 6.234738063812256,
"epoch": 0.11512326613129668,
"grad_norm": 0.984375,
"learning_rate": 0.0004999999174697195,
"loss": 6.2891,
"mean_token_accuracy": 0.13759731277823448,
"num_tokens": 2293325.0,
"step": 1025
},
{
"entropy": 6.364012956619263,
"epoch": 0.11568484303925422,
"grad_norm": 1.0625,
"learning_rate": 0.0004999998795000627,
"loss": 6.279,
"mean_token_accuracy": 0.142484587430954,
"num_tokens": 2304780.0,
"step": 1030
},
{
"entropy": 6.17555603981018,
"epoch": 0.11624641994721177,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999998343663223,
"loss": 6.2167,
"mean_token_accuracy": 0.13944805189967155,
"num_tokens": 2315999.0,
"step": 1035
},
{
"entropy": 6.316880798339843,
"epoch": 0.11680799685516932,
"grad_norm": 1.09375,
"learning_rate": 0.0004999997820685,
"loss": 6.3785,
"mean_token_accuracy": 0.1423812247812748,
"num_tokens": 2326695.0,
"step": 1040
},
{
"entropy": 6.304249572753906,
"epoch": 0.11736957376312686,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999997226065974,
"loss": 6.2032,
"mean_token_accuracy": 0.14872457087039948,
"num_tokens": 2336925.0,
"step": 1045
},
{
"entropy": 6.186109495162964,
"epoch": 0.11793115067108441,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999996559806163,
"loss": 6.1183,
"mean_token_accuracy": 0.14349379688501357,
"num_tokens": 2347431.0,
"step": 1050
},
{
"entropy": 6.174486351013184,
"epoch": 0.11849272757904195,
"grad_norm": 0.984375,
"learning_rate": 0.0004999995821905589,
"loss": 6.222,
"mean_token_accuracy": 0.14372190833091736,
"num_tokens": 2359894.0,
"step": 1055
},
{
"entropy": 6.233391332626343,
"epoch": 0.1190543044869995,
"grad_norm": 1.0625,
"learning_rate": 0.0004999995012364276,
"loss": 6.2421,
"mean_token_accuracy": 0.14038106054067612,
"num_tokens": 2371181.0,
"step": 1060
},
{
"entropy": 6.264003944396973,
"epoch": 0.11961588139495703,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999994131182248,
"loss": 6.1404,
"mean_token_accuracy": 0.14172154366970063,
"num_tokens": 2381531.0,
"step": 1065
},
{
"entropy": 6.283302164077758,
"epoch": 0.12017745830291458,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999993178359535,
"loss": 6.2758,
"mean_token_accuracy": 0.14069122895598413,
"num_tokens": 2391573.0,
"step": 1070
},
{
"entropy": 6.221704006195068,
"epoch": 0.12073903521087213,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999992153896166,
"loss": 6.2438,
"mean_token_accuracy": 0.14414030984044074,
"num_tokens": 2402822.0,
"step": 1075
},
{
"entropy": 6.1785337924957275,
"epoch": 0.12130061211882967,
"grad_norm": 1.203125,
"learning_rate": 0.0004999991057792175,
"loss": 6.1556,
"mean_token_accuracy": 0.1388797678053379,
"num_tokens": 2414322.0,
"step": 1080
},
{
"entropy": 6.271573257446289,
"epoch": 0.12186218902678722,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999989890047596,
"loss": 6.3473,
"mean_token_accuracy": 0.14235838428139685,
"num_tokens": 2426560.0,
"step": 1085
},
{
"entropy": 6.270335340499878,
"epoch": 0.12242376593474476,
"grad_norm": 1.125,
"learning_rate": 0.0004999988650662467,
"loss": 6.2468,
"mean_token_accuracy": 0.14222736656665802,
"num_tokens": 2437723.0,
"step": 1090
},
{
"entropy": 6.153625774383545,
"epoch": 0.12298534284270231,
"grad_norm": 1.078125,
"learning_rate": 0.0004999987339636825,
"loss": 6.1991,
"mean_token_accuracy": 0.14895833060145378,
"num_tokens": 2448497.0,
"step": 1095
},
{
"entropy": 6.157371854782104,
"epoch": 0.12354691975065986,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999985956970713,
"loss": 6.0728,
"mean_token_accuracy": 0.15379315465688706,
"num_tokens": 2459089.0,
"step": 1100
},
{
"entropy": 6.129484939575195,
"epoch": 0.1241084966586174,
"grad_norm": 1.078125,
"learning_rate": 0.0004999984502664178,
"loss": 6.1597,
"mean_token_accuracy": 0.14950370639562607,
"num_tokens": 2469882.0,
"step": 1105
},
{
"entropy": 6.1923255443573,
"epoch": 0.12467007356657495,
"grad_norm": 1.0,
"learning_rate": 0.0004999982976717261,
"loss": 6.1939,
"mean_token_accuracy": 0.14724715948104858,
"num_tokens": 2482120.0,
"step": 1110
},
{
"entropy": 6.231283187866211,
"epoch": 0.1252316504745325,
"grad_norm": 1.09375,
"learning_rate": 0.0004999981379130016,
"loss": 6.2066,
"mean_token_accuracy": 0.14431151151657104,
"num_tokens": 2492626.0,
"step": 1115
},
{
"entropy": 6.266522312164307,
"epoch": 0.12579322738249002,
"grad_norm": 0.95703125,
"learning_rate": 0.000499997970990249,
"loss": 6.2584,
"mean_token_accuracy": 0.1434864081442356,
"num_tokens": 2506347.0,
"step": 1120
},
{
"entropy": 6.067158842086792,
"epoch": 0.12635480429044757,
"grad_norm": 1.015625,
"learning_rate": 0.0004999977969034737,
"loss": 6.1402,
"mean_token_accuracy": 0.15051724016666412,
"num_tokens": 2516584.0,
"step": 1125
},
{
"entropy": 6.22009768486023,
"epoch": 0.12691638119840512,
"grad_norm": 1.03125,
"learning_rate": 0.0004999976156526814,
"loss": 6.1856,
"mean_token_accuracy": 0.14873197600245475,
"num_tokens": 2526731.0,
"step": 1130
},
{
"entropy": 6.165716361999512,
"epoch": 0.12747795810636267,
"grad_norm": 1.03125,
"learning_rate": 0.0004999974272378776,
"loss": 6.0988,
"mean_token_accuracy": 0.1502756729722023,
"num_tokens": 2537180.0,
"step": 1135
},
{
"entropy": 6.150550699234008,
"epoch": 0.12803953501432022,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999972316590685,
"loss": 6.1675,
"mean_token_accuracy": 0.14633068516850473,
"num_tokens": 2547617.0,
"step": 1140
},
{
"entropy": 6.15447096824646,
"epoch": 0.12860111192227774,
"grad_norm": 1.015625,
"learning_rate": 0.0004999970289162604,
"loss": 6.0589,
"mean_token_accuracy": 0.15604657232761382,
"num_tokens": 2557743.0,
"step": 1145
},
{
"entropy": 6.099272203445435,
"epoch": 0.1291626888302353,
"grad_norm": 1.03125,
"learning_rate": 0.0004999968190094595,
"loss": 6.1532,
"mean_token_accuracy": 0.14685856103897094,
"num_tokens": 2568351.0,
"step": 1150
},
{
"entropy": 6.13305516242981,
"epoch": 0.12972426573819285,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999966019386727,
"loss": 6.22,
"mean_token_accuracy": 0.14849678426980972,
"num_tokens": 2579703.0,
"step": 1155
},
{
"entropy": 6.29257230758667,
"epoch": 0.1302858426461504,
"grad_norm": 1.046875,
"learning_rate": 0.0004999963777039067,
"loss": 6.1955,
"mean_token_accuracy": 0.14582142382860183,
"num_tokens": 2591197.0,
"step": 1160
},
{
"entropy": 6.143500089645386,
"epoch": 0.13084741955410795,
"grad_norm": 1.109375,
"learning_rate": 0.000499996146305169,
"loss": 6.1668,
"mean_token_accuracy": 0.14330864027142526,
"num_tokens": 2601836.0,
"step": 1165
},
{
"entropy": 6.083126783370972,
"epoch": 0.13140899646206547,
"grad_norm": 0.96875,
"learning_rate": 0.0004999959077424665,
"loss": 6.1079,
"mean_token_accuracy": 0.1498291090130806,
"num_tokens": 2613513.0,
"step": 1170
},
{
"entropy": 6.128898954391479,
"epoch": 0.13197057337002302,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999956620158072,
"loss": 6.1413,
"mean_token_accuracy": 0.14912581145763398,
"num_tokens": 2625246.0,
"step": 1175
},
{
"entropy": 6.137507438659668,
"epoch": 0.13253215027798057,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999954091251987,
"loss": 6.1053,
"mean_token_accuracy": 0.14960974007844924,
"num_tokens": 2636375.0,
"step": 1180
},
{
"entropy": 6.159454727172852,
"epoch": 0.13309372718593812,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999951490706491,
"loss": 6.2012,
"mean_token_accuracy": 0.14142030403017997,
"num_tokens": 2648586.0,
"step": 1185
},
{
"entropy": 6.1870311260223385,
"epoch": 0.13365530409389567,
"grad_norm": 1.015625,
"learning_rate": 0.0004999948818521667,
"loss": 6.2085,
"mean_token_accuracy": 0.15162340849637984,
"num_tokens": 2660847.0,
"step": 1190
},
{
"entropy": 6.215510368347168,
"epoch": 0.1342168810018532,
"grad_norm": 0.99609375,
"learning_rate": 0.00049999460746976,
"loss": 6.1379,
"mean_token_accuracy": 0.14824316501617432,
"num_tokens": 2671160.0,
"step": 1195
},
{
"entropy": 6.205865049362183,
"epoch": 0.13477845790981074,
"grad_norm": 1.09375,
"learning_rate": 0.0004999943259234377,
"loss": 6.249,
"mean_token_accuracy": 0.1429877385497093,
"num_tokens": 2681622.0,
"step": 1200
},
{
"entropy": 6.0475321292877195,
"epoch": 0.1353400348177683,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999940372132089,
"loss": 6.05,
"mean_token_accuracy": 0.15074699968099595,
"num_tokens": 2691503.0,
"step": 1205
},
{
"entropy": 6.133896541595459,
"epoch": 0.13590161172572585,
"grad_norm": 1.015625,
"learning_rate": 0.0004999937413390826,
"loss": 6.1482,
"mean_token_accuracy": 0.14561435282230378,
"num_tokens": 2702207.0,
"step": 1210
},
{
"entropy": 6.135827493667603,
"epoch": 0.1364631886336834,
"grad_norm": 0.9375,
"learning_rate": 0.0004999934383010683,
"loss": 6.1481,
"mean_token_accuracy": 0.1448237679898739,
"num_tokens": 2713178.0,
"step": 1215
},
{
"entropy": 6.2300678253173825,
"epoch": 0.13702476554164092,
"grad_norm": 1.140625,
"learning_rate": 0.0004999931280991757,
"loss": 6.1363,
"mean_token_accuracy": 0.15154407769441605,
"num_tokens": 2723717.0,
"step": 1220
},
{
"entropy": 6.124025821685791,
"epoch": 0.13758634244959847,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999928107334146,
"loss": 6.2217,
"mean_token_accuracy": 0.14136773943901063,
"num_tokens": 2735669.0,
"step": 1225
},
{
"entropy": 6.099333381652832,
"epoch": 0.13814791935755602,
"grad_norm": 1.046875,
"learning_rate": 0.0004999924862037952,
"loss": 6.1166,
"mean_token_accuracy": 0.15113521963357926,
"num_tokens": 2746087.0,
"step": 1230
},
{
"entropy": 6.119935989379883,
"epoch": 0.13870949626551357,
"grad_norm": 1.0625,
"learning_rate": 0.0004999921545103278,
"loss": 6.1391,
"mean_token_accuracy": 0.1446000136435032,
"num_tokens": 2757143.0,
"step": 1235
},
{
"entropy": 6.234173107147217,
"epoch": 0.1392710731734711,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999918156530229,
"loss": 6.2428,
"mean_token_accuracy": 0.14500228315591812,
"num_tokens": 2769340.0,
"step": 1240
},
{
"entropy": 6.175450468063355,
"epoch": 0.13983265008142864,
"grad_norm": 1.03125,
"learning_rate": 0.0004999914696318913,
"loss": 6.1589,
"mean_token_accuracy": 0.14806368499994277,
"num_tokens": 2779514.0,
"step": 1245
},
{
"entropy": 6.119281053543091,
"epoch": 0.1403942269893862,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999911164469441,
"loss": 6.0888,
"mean_token_accuracy": 0.15193744301795958,
"num_tokens": 2790801.0,
"step": 1250
},
{
"entropy": 6.1100088596344,
"epoch": 0.14095580389734375,
"grad_norm": 1.140625,
"learning_rate": 0.0004999907560981925,
"loss": 6.0847,
"mean_token_accuracy": 0.14798859357833863,
"num_tokens": 2801082.0,
"step": 1255
},
{
"entropy": 6.083402299880982,
"epoch": 0.1415173808053013,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999903885856479,
"loss": 6.1062,
"mean_token_accuracy": 0.14728771150112152,
"num_tokens": 2812719.0,
"step": 1260
},
{
"entropy": 6.076613521575927,
"epoch": 0.14207895771325882,
"grad_norm": 1.046875,
"learning_rate": 0.000499990013909322,
"loss": 6.1067,
"mean_token_accuracy": 0.1431245759129524,
"num_tokens": 2823766.0,
"step": 1265
},
{
"entropy": 6.133080291748047,
"epoch": 0.14264053462121637,
"grad_norm": 0.98828125,
"learning_rate": 0.000499989632069227,
"loss": 6.1957,
"mean_token_accuracy": 0.14594514071941375,
"num_tokens": 2834801.0,
"step": 1270
},
{
"entropy": 6.081394481658935,
"epoch": 0.14320211152917392,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999892430653747,
"loss": 6.0072,
"mean_token_accuracy": 0.15184544026851654,
"num_tokens": 2845769.0,
"step": 1275
},
{
"entropy": 6.042043972015381,
"epoch": 0.14376368843713147,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999888468977776,
"loss": 6.0533,
"mean_token_accuracy": 0.15463033318519592,
"num_tokens": 2856001.0,
"step": 1280
},
{
"entropy": 6.135152864456177,
"epoch": 0.14432526534508902,
"grad_norm": 1.015625,
"learning_rate": 0.0004999884435664485,
"loss": 6.0896,
"mean_token_accuracy": 0.14773025661706923,
"num_tokens": 2867703.0,
"step": 1285
},
{
"entropy": 5.996362161636353,
"epoch": 0.14488684225304654,
"grad_norm": 1.171875,
"learning_rate": 0.0004999880330714,
"loss": 6.0314,
"mean_token_accuracy": 0.15192831605672835,
"num_tokens": 2878671.0,
"step": 1290
},
{
"entropy": 6.103845834732056,
"epoch": 0.1454484191610041,
"grad_norm": 1.015625,
"learning_rate": 0.0004999876154126453,
"loss": 6.1468,
"mean_token_accuracy": 0.14183202311396598,
"num_tokens": 2890137.0,
"step": 1295
},
{
"entropy": 6.1058008670806885,
"epoch": 0.14600999606896164,
"grad_norm": 1.0625,
"learning_rate": 0.0004999871905901977,
"loss": 6.0485,
"mean_token_accuracy": 0.15308912694454194,
"num_tokens": 2900344.0,
"step": 1300
},
{
"entropy": 6.0173821449279785,
"epoch": 0.1465715729769192,
"grad_norm": 1.046875,
"learning_rate": 0.0004999867586040706,
"loss": 6.0786,
"mean_token_accuracy": 0.14948148727416993,
"num_tokens": 2912270.0,
"step": 1305
},
{
"entropy": 6.176186656951904,
"epoch": 0.14713314988487675,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999863194542779,
"loss": 6.1416,
"mean_token_accuracy": 0.14145402163267135,
"num_tokens": 2924677.0,
"step": 1310
},
{
"entropy": 5.98449125289917,
"epoch": 0.14769472679283427,
"grad_norm": 1.078125,
"learning_rate": 0.0004999858731408336,
"loss": 6.0758,
"mean_token_accuracy": 0.14689501821994783,
"num_tokens": 2935794.0,
"step": 1315
},
{
"entropy": 6.144368124008179,
"epoch": 0.14825630370079182,
"grad_norm": 1.140625,
"learning_rate": 0.0004999854196637518,
"loss": 6.1803,
"mean_token_accuracy": 0.1436318762600422,
"num_tokens": 2945958.0,
"step": 1320
},
{
"entropy": 6.140670251846314,
"epoch": 0.14881788060874937,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999849590230469,
"loss": 6.1384,
"mean_token_accuracy": 0.14359039589762687,
"num_tokens": 2957536.0,
"step": 1325
},
{
"entropy": 6.109699487686157,
"epoch": 0.14937945751670692,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999844912187336,
"loss": 6.1209,
"mean_token_accuracy": 0.145068246871233,
"num_tokens": 2967787.0,
"step": 1330
},
{
"entropy": 6.110361862182617,
"epoch": 0.14994103442466447,
"grad_norm": 0.97265625,
"learning_rate": 0.000499984016250827,
"loss": 6.137,
"mean_token_accuracy": 0.1437262400984764,
"num_tokens": 2980522.0,
"step": 1335
},
{
"entropy": 5.966966819763184,
"epoch": 0.150502611332622,
"grad_norm": 1.03125,
"learning_rate": 0.000499983534119342,
"loss": 6.0733,
"mean_token_accuracy": 0.15085284113883973,
"num_tokens": 2990497.0,
"step": 1340
},
{
"entropy": 6.219952869415283,
"epoch": 0.15106418824057954,
"grad_norm": 1.1875,
"learning_rate": 0.0004999830448242941,
"loss": 6.163,
"mean_token_accuracy": 0.14726221784949303,
"num_tokens": 3002931.0,
"step": 1345
},
{
"entropy": 6.013589143753052,
"epoch": 0.1516257651485371,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999825483656986,
"loss": 6.0739,
"mean_token_accuracy": 0.15152935981750487,
"num_tokens": 3014137.0,
"step": 1350
},
{
"entropy": 6.241993951797485,
"epoch": 0.15218734205649465,
"grad_norm": 0.984375,
"learning_rate": 0.0004999820447435716,
"loss": 6.2617,
"mean_token_accuracy": 0.14130268543958663,
"num_tokens": 3026145.0,
"step": 1355
},
{
"entropy": 6.032399415969849,
"epoch": 0.15274891896445217,
"grad_norm": 1.046875,
"learning_rate": 0.0004999815339579291,
"loss": 6.1122,
"mean_token_accuracy": 0.14616605639457703,
"num_tokens": 3037068.0,
"step": 1360
},
{
"entropy": 6.133391094207764,
"epoch": 0.15331049587240972,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999810160087872,
"loss": 6.034,
"mean_token_accuracy": 0.1539703369140625,
"num_tokens": 3047836.0,
"step": 1365
},
{
"entropy": 6.099278354644776,
"epoch": 0.15387207278036727,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999804908961626,
"loss": 6.1684,
"mean_token_accuracy": 0.14856233298778534,
"num_tokens": 3060309.0,
"step": 1370
},
{
"entropy": 6.092678594589233,
"epoch": 0.15443364968832482,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999799586200719,
"loss": 6.1375,
"mean_token_accuracy": 0.15069601237773894,
"num_tokens": 3071789.0,
"step": 1375
},
{
"entropy": 6.0972583293914795,
"epoch": 0.15499522659628237,
"grad_norm": 1.140625,
"learning_rate": 0.000499979419180532,
"loss": 6.0467,
"mean_token_accuracy": 0.15597807765007018,
"num_tokens": 3081756.0,
"step": 1380
},
{
"entropy": 6.030021715164184,
"epoch": 0.1555568035042399,
"grad_norm": 1.0625,
"learning_rate": 0.0004999788725775601,
"loss": 6.1537,
"mean_token_accuracy": 0.14404376447200776,
"num_tokens": 3093456.0,
"step": 1385
},
{
"entropy": 6.176510953903199,
"epoch": 0.15611838041219744,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999783188111737,
"loss": 6.083,
"mean_token_accuracy": 0.15343996584415437,
"num_tokens": 3104565.0,
"step": 1390
},
{
"entropy": 6.049490928649902,
"epoch": 0.156679957320155,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999777578813904,
"loss": 6.0964,
"mean_token_accuracy": 0.14417065382003785,
"num_tokens": 3115577.0,
"step": 1395
},
{
"entropy": 6.068271827697754,
"epoch": 0.15724153422811254,
"grad_norm": 1.046875,
"learning_rate": 0.000499977189788228,
"loss": 5.9898,
"mean_token_accuracy": 0.1535682797431946,
"num_tokens": 3126019.0,
"step": 1400
},
{
"entropy": 6.11846694946289,
"epoch": 0.1578031111360701,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999766145317046,
"loss": 6.1593,
"mean_token_accuracy": 0.14163137823343278,
"num_tokens": 3136191.0,
"step": 1405
},
{
"entropy": 6.1475354671478275,
"epoch": 0.15836468804402762,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999760321118385,
"loss": 6.0184,
"mean_token_accuracy": 0.14922462552785873,
"num_tokens": 3147737.0,
"step": 1410
},
{
"entropy": 5.963467979431153,
"epoch": 0.15892626495198517,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999754425286483,
"loss": 6.0995,
"mean_token_accuracy": 0.14720706641674042,
"num_tokens": 3158494.0,
"step": 1415
},
{
"entropy": 6.131221580505371,
"epoch": 0.15948784185994272,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999748457821529,
"loss": 6.1288,
"mean_token_accuracy": 0.14769198149442672,
"num_tokens": 3170550.0,
"step": 1420
},
{
"entropy": 5.9402268409729,
"epoch": 0.16004941876790027,
"grad_norm": 0.91015625,
"learning_rate": 0.000499974241872371,
"loss": 6.1076,
"mean_token_accuracy": 0.15461537688970567,
"num_tokens": 3182860.0,
"step": 1425
},
{
"entropy": 6.07914547920227,
"epoch": 0.16061099567585782,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999736307993221,
"loss": 6.105,
"mean_token_accuracy": 0.14613698273897172,
"num_tokens": 3193807.0,
"step": 1430
},
{
"entropy": 5.986855363845825,
"epoch": 0.16117257258381534,
"grad_norm": 0.984375,
"learning_rate": 0.0004999730125630255,
"loss": 5.9808,
"mean_token_accuracy": 0.15065714865922927,
"num_tokens": 3203959.0,
"step": 1435
},
{
"entropy": 6.042674398422241,
"epoch": 0.1617341494917729,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999723871635008,
"loss": 5.9961,
"mean_token_accuracy": 0.15428200960159302,
"num_tokens": 3215416.0,
"step": 1440
},
{
"entropy": 6.139855527877808,
"epoch": 0.16229572639973044,
"grad_norm": 1.03125,
"learning_rate": 0.0004999717546007681,
"loss": 6.1393,
"mean_token_accuracy": 0.14156408235430717,
"num_tokens": 3226938.0,
"step": 1445
},
{
"entropy": 5.931169843673706,
"epoch": 0.162857303307688,
"grad_norm": 1.125,
"learning_rate": 0.0004999711148748476,
"loss": 5.956,
"mean_token_accuracy": 0.1564284533262253,
"num_tokens": 3238406.0,
"step": 1450
},
{
"entropy": 5.968686246871949,
"epoch": 0.16341888021564555,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999704679857594,
"loss": 5.9748,
"mean_token_accuracy": 0.14992773681879043,
"num_tokens": 3249695.0,
"step": 1455
},
{
"entropy": 6.125838279724121,
"epoch": 0.16398045712360307,
"grad_norm": 1.0625,
"learning_rate": 0.0004999698139335243,
"loss": 6.1147,
"mean_token_accuracy": 0.15428986698389052,
"num_tokens": 3259656.0,
"step": 1460
},
{
"entropy": 5.994536018371582,
"epoch": 0.16454203403156062,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999691527181632,
"loss": 5.9514,
"mean_token_accuracy": 0.1599651977419853,
"num_tokens": 3269712.0,
"step": 1465
},
{
"entropy": 5.873442840576172,
"epoch": 0.16510361093951817,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999684843396968,
"loss": 5.948,
"mean_token_accuracy": 0.16050935834646224,
"num_tokens": 3280343.0,
"step": 1470
},
{
"entropy": 6.029610300064087,
"epoch": 0.16566518784747572,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999678087981467,
"loss": 5.9863,
"mean_token_accuracy": 0.15649282932281494,
"num_tokens": 3291649.0,
"step": 1475
},
{
"entropy": 6.136847448348999,
"epoch": 0.16622676475543324,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999671260935343,
"loss": 6.1547,
"mean_token_accuracy": 0.15154551565647126,
"num_tokens": 3304190.0,
"step": 1480
},
{
"entropy": 5.872249698638916,
"epoch": 0.1667883416633908,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999664362258814,
"loss": 5.8936,
"mean_token_accuracy": 0.1636761337518692,
"num_tokens": 3313966.0,
"step": 1485
},
{
"entropy": 6.072989749908447,
"epoch": 0.16734991857134834,
"grad_norm": 1.03125,
"learning_rate": 0.0004999657391952099,
"loss": 6.1474,
"mean_token_accuracy": 0.1493370659649372,
"num_tokens": 3326017.0,
"step": 1490
},
{
"entropy": 6.061523818969727,
"epoch": 0.1679114954793059,
"grad_norm": 0.90625,
"learning_rate": 0.000499965035001542,
"loss": 6.0294,
"mean_token_accuracy": 0.14847790449857712,
"num_tokens": 3338264.0,
"step": 1495
},
{
"entropy": 6.020901775360107,
"epoch": 0.16847307238726345,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999643236449001,
"loss": 6.0681,
"mean_token_accuracy": 0.1578044191002846,
"num_tokens": 3348901.0,
"step": 1500
},
{
"entropy": 6.05039758682251,
"epoch": 0.16903464929522097,
"grad_norm": 1.015625,
"learning_rate": 0.0004999636051253069,
"loss": 6.0115,
"mean_token_accuracy": 0.15544183254241944,
"num_tokens": 3358441.0,
"step": 1505
},
{
"entropy": 6.062154722213745,
"epoch": 0.16959622620317852,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999628794427854,
"loss": 6.0913,
"mean_token_accuracy": 0.15557602494955064,
"num_tokens": 3369334.0,
"step": 1510
},
{
"entropy": 5.989631223678589,
"epoch": 0.17015780311113607,
"grad_norm": 1.09375,
"learning_rate": 0.0004999621465973583,
"loss": 6.0789,
"mean_token_accuracy": 0.1499525025486946,
"num_tokens": 3380698.0,
"step": 1515
},
{
"entropy": 6.214061117172241,
"epoch": 0.17071938001909362,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999614065890495,
"loss": 6.0942,
"mean_token_accuracy": 0.1456768698990345,
"num_tokens": 3391061.0,
"step": 1520
},
{
"entropy": 6.012625885009766,
"epoch": 0.17128095692705117,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999606594178821,
"loss": 6.1471,
"mean_token_accuracy": 0.14754906371235849,
"num_tokens": 3403983.0,
"step": 1525
},
{
"entropy": 6.164724922180175,
"epoch": 0.1718425338350087,
"grad_norm": 1.125,
"learning_rate": 0.00049995990508388,
"loss": 6.0759,
"mean_token_accuracy": 0.15379567444324493,
"num_tokens": 3413838.0,
"step": 1530
},
{
"entropy": 5.909680509567261,
"epoch": 0.17240411074296624,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999591435870674,
"loss": 5.9074,
"mean_token_accuracy": 0.15706708282232285,
"num_tokens": 3423978.0,
"step": 1535
},
{
"entropy": 6.045839118957519,
"epoch": 0.1729656876509238,
"grad_norm": 1.15625,
"learning_rate": 0.0004999583749274684,
"loss": 6.0766,
"mean_token_accuracy": 0.15526383817195893,
"num_tokens": 3433416.0,
"step": 1540
},
{
"entropy": 5.899442911148071,
"epoch": 0.17352726455888134,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999575991051074,
"loss": 5.9403,
"mean_token_accuracy": 0.15632218718528748,
"num_tokens": 3444983.0,
"step": 1545
},
{
"entropy": 5.97276668548584,
"epoch": 0.1740888414668389,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999568161200093,
"loss": 5.9251,
"mean_token_accuracy": 0.15755780041217804,
"num_tokens": 3455808.0,
"step": 1550
},
{
"entropy": 5.955643653869629,
"epoch": 0.17465041837479642,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999560259721989,
"loss": 6.0733,
"mean_token_accuracy": 0.14496413096785546,
"num_tokens": 3467503.0,
"step": 1555
},
{
"entropy": 6.2084818363189695,
"epoch": 0.17521199528275397,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999552286617014,
"loss": 6.2662,
"mean_token_accuracy": 0.1477560058236122,
"num_tokens": 3479941.0,
"step": 1560
},
{
"entropy": 5.958353185653687,
"epoch": 0.17577357219071152,
"grad_norm": 1.09375,
"learning_rate": 0.0004999544241885421,
"loss": 5.9351,
"mean_token_accuracy": 0.16224518567323684,
"num_tokens": 3491069.0,
"step": 1565
},
{
"entropy": 6.06676082611084,
"epoch": 0.17633514909866907,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999536125527467,
"loss": 6.1282,
"mean_token_accuracy": 0.1457708492875099,
"num_tokens": 3502065.0,
"step": 1570
},
{
"entropy": 6.07559905052185,
"epoch": 0.17689672600662662,
"grad_norm": 0.98046875,
"learning_rate": 0.000499952793754341,
"loss": 6.0865,
"mean_token_accuracy": 0.15441752076148987,
"num_tokens": 3514228.0,
"step": 1575
},
{
"entropy": 5.995469331741333,
"epoch": 0.17745830291458414,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999519677933512,
"loss": 6.0211,
"mean_token_accuracy": 0.15433061122894287,
"num_tokens": 3525801.0,
"step": 1580
},
{
"entropy": 5.998544597625733,
"epoch": 0.1780198798225417,
"grad_norm": 1.046875,
"learning_rate": 0.0004999511346698035,
"loss": 5.9454,
"mean_token_accuracy": 0.15480177700519562,
"num_tokens": 3536431.0,
"step": 1585
},
{
"entropy": 5.968015098571778,
"epoch": 0.17858145673049924,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999502943837244,
"loss": 6.0659,
"mean_token_accuracy": 0.15099444910883902,
"num_tokens": 3549098.0,
"step": 1590
},
{
"entropy": 6.096561050415039,
"epoch": 0.1791430336384568,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999494469351406,
"loss": 6.0114,
"mean_token_accuracy": 0.152137491106987,
"num_tokens": 3560532.0,
"step": 1595
},
{
"entropy": 5.961009120941162,
"epoch": 0.17970461054641435,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999485923240793,
"loss": 6.0303,
"mean_token_accuracy": 0.15316579714417458,
"num_tokens": 3572175.0,
"step": 1600
},
{
"entropy": 6.033007192611694,
"epoch": 0.18026618745437187,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999477305505675,
"loss": 5.9451,
"mean_token_accuracy": 0.17299906462430953,
"num_tokens": 3582862.0,
"step": 1605
},
{
"entropy": 6.030126714706421,
"epoch": 0.18082776436232942,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999468616146328,
"loss": 6.1329,
"mean_token_accuracy": 0.15195471569895744,
"num_tokens": 3594685.0,
"step": 1610
},
{
"entropy": 5.915484094619751,
"epoch": 0.18138934127028697,
"grad_norm": 1.078125,
"learning_rate": 0.0004999459855163026,
"loss": 5.9167,
"mean_token_accuracy": 0.1637891188263893,
"num_tokens": 3604847.0,
"step": 1615
},
{
"entropy": 6.032248640060425,
"epoch": 0.18195091817824452,
"grad_norm": 0.98828125,
"learning_rate": 0.000499945102255605,
"loss": 5.9967,
"mean_token_accuracy": 0.15529887825250627,
"num_tokens": 3615549.0,
"step": 1620
},
{
"entropy": 5.955660629272461,
"epoch": 0.18251249508620204,
"grad_norm": 1.0625,
"learning_rate": 0.0004999442118325682,
"loss": 5.9328,
"mean_token_accuracy": 0.16356256008148193,
"num_tokens": 3626300.0,
"step": 1625
},
{
"entropy": 5.940017890930176,
"epoch": 0.1830740719941596,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999433142472205,
"loss": 5.9533,
"mean_token_accuracy": 0.15708972960710527,
"num_tokens": 3637832.0,
"step": 1630
},
{
"entropy": 6.009549951553344,
"epoch": 0.18363564890211714,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999424094995902,
"loss": 5.9815,
"mean_token_accuracy": 0.15054063647985458,
"num_tokens": 3649179.0,
"step": 1635
},
{
"entropy": 5.970891094207763,
"epoch": 0.1841972258100747,
"grad_norm": 0.96875,
"learning_rate": 0.0004999414975897066,
"loss": 6.0249,
"mean_token_accuracy": 0.15206027776002884,
"num_tokens": 3660011.0,
"step": 1640
},
{
"entropy": 6.120618104934692,
"epoch": 0.18475880271803224,
"grad_norm": 0.921875,
"learning_rate": 0.0004999405785175983,
"loss": 6.1014,
"mean_token_accuracy": 0.1466618612408638,
"num_tokens": 3672791.0,
"step": 1645
},
{
"entropy": 5.9426877975463865,
"epoch": 0.18532037962598977,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999396522832949,
"loss": 5.9301,
"mean_token_accuracy": 0.14880775511264802,
"num_tokens": 3684709.0,
"step": 1650
},
{
"entropy": 5.8846558094024655,
"epoch": 0.18588195653394732,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999387188868256,
"loss": 5.9009,
"mean_token_accuracy": 0.1574036091566086,
"num_tokens": 3694675.0,
"step": 1655
},
{
"entropy": 5.897461223602295,
"epoch": 0.18644353344190487,
"grad_norm": 0.953125,
"learning_rate": 0.0004999377783282203,
"loss": 5.9052,
"mean_token_accuracy": 0.15229580998420716,
"num_tokens": 3706384.0,
"step": 1660
},
{
"entropy": 6.019663524627686,
"epoch": 0.18700511034986242,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999368306075089,
"loss": 5.9767,
"mean_token_accuracy": 0.15526307225227357,
"num_tokens": 3717419.0,
"step": 1665
},
{
"entropy": 5.97164831161499,
"epoch": 0.18756668725781997,
"grad_norm": 1.015625,
"learning_rate": 0.0004999358757247215,
"loss": 6.0001,
"mean_token_accuracy": 0.15804468095302582,
"num_tokens": 3727306.0,
"step": 1670
},
{
"entropy": 6.075372171401978,
"epoch": 0.1881282641657775,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999349136798886,
"loss": 6.1408,
"mean_token_accuracy": 0.1482490971684456,
"num_tokens": 3739476.0,
"step": 1675
},
{
"entropy": 5.917860984802246,
"epoch": 0.18868984107373504,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999339444730409,
"loss": 5.8937,
"mean_token_accuracy": 0.16344992816448212,
"num_tokens": 3751035.0,
"step": 1680
},
{
"entropy": 5.984053277969361,
"epoch": 0.1892514179816926,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999329681042092,
"loss": 5.9851,
"mean_token_accuracy": 0.15741585195064545,
"num_tokens": 3762881.0,
"step": 1685
},
{
"entropy": 5.960563278198242,
"epoch": 0.18981299488965014,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999319845734243,
"loss": 5.999,
"mean_token_accuracy": 0.15340179800987244,
"num_tokens": 3774180.0,
"step": 1690
},
{
"entropy": 5.944005250930786,
"epoch": 0.1903745717976077,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999309938807181,
"loss": 5.9978,
"mean_token_accuracy": 0.16235336810350418,
"num_tokens": 3784864.0,
"step": 1695
},
{
"entropy": 6.061145496368408,
"epoch": 0.19093614870556522,
"grad_norm": 1.0,
"learning_rate": 0.0004999299960261216,
"loss": 5.983,
"mean_token_accuracy": 0.15478956401348115,
"num_tokens": 3797693.0,
"step": 1700
},
{
"entropy": 5.940720462799073,
"epoch": 0.19149772561352277,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999289910096669,
"loss": 6.0048,
"mean_token_accuracy": 0.15954993665218353,
"num_tokens": 3809379.0,
"step": 1705
},
{
"entropy": 5.869456434249878,
"epoch": 0.19205930252148032,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999279788313858,
"loss": 5.8466,
"mean_token_accuracy": 0.1634460434317589,
"num_tokens": 3819473.0,
"step": 1710
},
{
"entropy": 5.839190578460693,
"epoch": 0.19262087942943787,
"grad_norm": 1.0,
"learning_rate": 0.0004999269594913107,
"loss": 5.8447,
"mean_token_accuracy": 0.16184932887554168,
"num_tokens": 3831037.0,
"step": 1715
},
{
"entropy": 5.890625619888306,
"epoch": 0.19318245633739542,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999259329894739,
"loss": 5.9796,
"mean_token_accuracy": 0.15089921206235885,
"num_tokens": 3841767.0,
"step": 1720
},
{
"entropy": 5.863656806945801,
"epoch": 0.19374403324535294,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999248993259081,
"loss": 5.8179,
"mean_token_accuracy": 0.1636965826153755,
"num_tokens": 3851629.0,
"step": 1725
},
{
"entropy": 5.859853363037109,
"epoch": 0.1943056101533105,
"grad_norm": 1.109375,
"learning_rate": 0.0004999238585006463,
"loss": 5.8875,
"mean_token_accuracy": 0.15624083131551741,
"num_tokens": 3863247.0,
"step": 1730
},
{
"entropy": 5.972808027267456,
"epoch": 0.19486718706126804,
"grad_norm": 1.0,
"learning_rate": 0.0004999228105137218,
"loss": 5.8814,
"mean_token_accuracy": 0.1572691485285759,
"num_tokens": 3873949.0,
"step": 1735
},
{
"entropy": 5.924859476089478,
"epoch": 0.1954287639692256,
"grad_norm": 1.03125,
"learning_rate": 0.0004999217553651676,
"loss": 5.9895,
"mean_token_accuracy": 0.1545493260025978,
"num_tokens": 3884992.0,
"step": 1740
},
{
"entropy": 6.055422115325928,
"epoch": 0.19599034087718312,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999206930550175,
"loss": 6.1164,
"mean_token_accuracy": 0.15783915370702745,
"num_tokens": 3897029.0,
"step": 1745
},
{
"entropy": 5.9067707538604735,
"epoch": 0.19655191778514067,
"grad_norm": 1.0,
"learning_rate": 0.0004999196235833054,
"loss": 5.9658,
"mean_token_accuracy": 0.15075957030057907,
"num_tokens": 3909510.0,
"step": 1750
},
{
"entropy": 5.964732027053833,
"epoch": 0.19711349469309822,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999185469500651,
"loss": 5.8633,
"mean_token_accuracy": 0.15894850939512253,
"num_tokens": 3920593.0,
"step": 1755
},
{
"entropy": 5.965554332733154,
"epoch": 0.19767507160105577,
"grad_norm": 0.984375,
"learning_rate": 0.000499917463155331,
"loss": 5.9988,
"mean_token_accuracy": 0.150234617292881,
"num_tokens": 3932215.0,
"step": 1760
},
{
"entropy": 5.989650630950928,
"epoch": 0.19823664850901332,
"grad_norm": 1.078125,
"learning_rate": 0.0004999163721991378,
"loss": 5.9462,
"mean_token_accuracy": 0.15746484696865082,
"num_tokens": 3943876.0,
"step": 1765
},
{
"entropy": 5.773177862167358,
"epoch": 0.19879822541697084,
"grad_norm": 1.0,
"learning_rate": 0.00049991527408152,
"loss": 5.9507,
"mean_token_accuracy": 0.15623908042907714,
"num_tokens": 3954533.0,
"step": 1770
},
{
"entropy": 6.059681606292725,
"epoch": 0.1993598023249284,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999141688025127,
"loss": 5.9939,
"mean_token_accuracy": 0.16129837930202484,
"num_tokens": 3965722.0,
"step": 1775
},
{
"entropy": 5.901776123046875,
"epoch": 0.19992137923288594,
"grad_norm": 0.921875,
"learning_rate": 0.000499913056362151,
"loss": 5.9075,
"mean_token_accuracy": 0.15883826315402985,
"num_tokens": 3976737.0,
"step": 1780
},
{
"entropy": 6.016579627990723,
"epoch": 0.2004829561408435,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999119367604703,
"loss": 5.9984,
"mean_token_accuracy": 0.15875881016254426,
"num_tokens": 3986102.0,
"step": 1785
},
{
"entropy": 5.9803629398345945,
"epoch": 0.20104453304880104,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999108099975064,
"loss": 5.976,
"mean_token_accuracy": 0.15650469213724136,
"num_tokens": 3996534.0,
"step": 1790
},
{
"entropy": 5.863087463378906,
"epoch": 0.20160610995675857,
"grad_norm": 1.0390625,
"learning_rate": 0.000499909676073295,
"loss": 5.8881,
"mean_token_accuracy": 0.15706568509340285,
"num_tokens": 4006947.0,
"step": 1795
},
{
"entropy": 5.90200252532959,
"epoch": 0.20216768686471612,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999085349878723,
"loss": 5.92,
"mean_token_accuracy": 0.16723052114248277,
"num_tokens": 4017723.0,
"step": 1800
},
{
"entropy": 5.895722150802612,
"epoch": 0.20272926377267367,
"grad_norm": 1.0625,
"learning_rate": 0.0004999073867412746,
"loss": 5.9885,
"mean_token_accuracy": 0.15485088378190995,
"num_tokens": 4028347.0,
"step": 1805
},
{
"entropy": 5.922601175308228,
"epoch": 0.20329084068063122,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999062313335385,
"loss": 5.8643,
"mean_token_accuracy": 0.16550229638814926,
"num_tokens": 4039648.0,
"step": 1810
},
{
"entropy": 5.962918710708618,
"epoch": 0.20385241758858877,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999050687647008,
"loss": 5.9226,
"mean_token_accuracy": 0.15441896617412568,
"num_tokens": 4050786.0,
"step": 1815
},
{
"entropy": 5.971924304962158,
"epoch": 0.2044139944965463,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999038990347984,
"loss": 5.9563,
"mean_token_accuracy": 0.15948153734207154,
"num_tokens": 4061184.0,
"step": 1820
},
{
"entropy": 6.024726390838623,
"epoch": 0.20497557140450384,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999027221438688,
"loss": 6.1013,
"mean_token_accuracy": 0.1498976394534111,
"num_tokens": 4073607.0,
"step": 1825
},
{
"entropy": 5.907567167282105,
"epoch": 0.2055371483124614,
"grad_norm": 1.03125,
"learning_rate": 0.000499901538091949,
"loss": 5.8961,
"mean_token_accuracy": 0.15584656596183777,
"num_tokens": 4083924.0,
"step": 1830
},
{
"entropy": 5.908780908584594,
"epoch": 0.20609872522041894,
"grad_norm": 1.09375,
"learning_rate": 0.0004999003468790773,
"loss": 5.895,
"mean_token_accuracy": 0.16026797592639924,
"num_tokens": 4094083.0,
"step": 1835
},
{
"entropy": 5.952602481842041,
"epoch": 0.2066603021283765,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998991485052911,
"loss": 5.9312,
"mean_token_accuracy": 0.16474931836128234,
"num_tokens": 4105124.0,
"step": 1840
},
{
"entropy": 6.067802858352661,
"epoch": 0.20722187903633402,
"grad_norm": 1.0,
"learning_rate": 0.000499897942970629,
"loss": 6.0896,
"mean_token_accuracy": 0.14841930121183394,
"num_tokens": 4116792.0,
"step": 1845
},
{
"entropy": 6.09526219367981,
"epoch": 0.20778345594429157,
"grad_norm": 0.96875,
"learning_rate": 0.000499896730275129,
"loss": 6.0843,
"mean_token_accuracy": 0.1532771721482277,
"num_tokens": 4127441.0,
"step": 1850
},
{
"entropy": 5.940713548660279,
"epoch": 0.20834503285224912,
"grad_norm": 1.015625,
"learning_rate": 0.00049989551041883,
"loss": 6.0739,
"mean_token_accuracy": 0.15080675035715102,
"num_tokens": 4138696.0,
"step": 1855
},
{
"entropy": 5.9166515350341795,
"epoch": 0.20890660976020667,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998942834017707,
"loss": 5.9439,
"mean_token_accuracy": 0.15614898800849913,
"num_tokens": 4151254.0,
"step": 1860
},
{
"entropy": 5.902920818328857,
"epoch": 0.2094681866681642,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998930492239901,
"loss": 5.8982,
"mean_token_accuracy": 0.15039533078670503,
"num_tokens": 4162046.0,
"step": 1865
},
{
"entropy": 5.854731798171997,
"epoch": 0.21002976357612174,
"grad_norm": 1.078125,
"learning_rate": 0.0004998918078855277,
"loss": 5.8424,
"mean_token_accuracy": 0.1606363594532013,
"num_tokens": 4172409.0,
"step": 1870
},
{
"entropy": 5.904069328308106,
"epoch": 0.2105913404840793,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998905593864227,
"loss": 6.0008,
"mean_token_accuracy": 0.16382352709770204,
"num_tokens": 4182949.0,
"step": 1875
},
{
"entropy": 5.930221652984619,
"epoch": 0.21115291739203684,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998893037267153,
"loss": 5.9673,
"mean_token_accuracy": 0.15838558822870255,
"num_tokens": 4194164.0,
"step": 1880
},
{
"entropy": 6.024990034103394,
"epoch": 0.2117144942999944,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998880409064452,
"loss": 5.9359,
"mean_token_accuracy": 0.16012249439954757,
"num_tokens": 4206056.0,
"step": 1885
},
{
"entropy": 5.89975209236145,
"epoch": 0.21227607120795192,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998867709256527,
"loss": 5.9935,
"mean_token_accuracy": 0.15553135573863983,
"num_tokens": 4216660.0,
"step": 1890
},
{
"entropy": 5.904560232162476,
"epoch": 0.21283764811590947,
"grad_norm": 0.97265625,
"learning_rate": 0.000499885493784378,
"loss": 5.8812,
"mean_token_accuracy": 0.15976034998893737,
"num_tokens": 4228495.0,
"step": 1895
},
{
"entropy": 5.912075519561768,
"epoch": 0.21339922502386702,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998842094826621,
"loss": 5.8696,
"mean_token_accuracy": 0.1601097196340561,
"num_tokens": 4238560.0,
"step": 1900
},
{
"entropy": 6.007304859161377,
"epoch": 0.21396080193182457,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998829180205458,
"loss": 6.0989,
"mean_token_accuracy": 0.15267504751682281,
"num_tokens": 4250553.0,
"step": 1905
},
{
"entropy": 5.997589445114135,
"epoch": 0.21452237883978212,
"grad_norm": 0.984375,
"learning_rate": 0.0004998816193980701,
"loss": 5.9946,
"mean_token_accuracy": 0.15328334122896195,
"num_tokens": 4262050.0,
"step": 1910
},
{
"entropy": 6.066816091537476,
"epoch": 0.21508395574773964,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998803136152764,
"loss": 5.9717,
"mean_token_accuracy": 0.14724281281232834,
"num_tokens": 4273019.0,
"step": 1915
},
{
"entropy": 5.860903739929199,
"epoch": 0.2156455326556972,
"grad_norm": 1.0625,
"learning_rate": 0.0004998790006722063,
"loss": 5.8356,
"mean_token_accuracy": 0.16368041932582855,
"num_tokens": 4283585.0,
"step": 1920
},
{
"entropy": 5.8601470470428465,
"epoch": 0.21620710956365474,
"grad_norm": 1.078125,
"learning_rate": 0.0004998776805689016,
"loss": 5.9195,
"mean_token_accuracy": 0.15469036400318145,
"num_tokens": 4294905.0,
"step": 1925
},
{
"entropy": 6.0024792671203615,
"epoch": 0.2167686864716123,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998763533054045,
"loss": 6.0381,
"mean_token_accuracy": 0.15404224991798401,
"num_tokens": 4305926.0,
"step": 1930
},
{
"entropy": 5.8503828048706055,
"epoch": 0.21733026337956984,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998750188817568,
"loss": 6.0068,
"mean_token_accuracy": 0.1558867573738098,
"num_tokens": 4318297.0,
"step": 1935
},
{
"entropy": 5.923427248001099,
"epoch": 0.21789184028752737,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998736772980015,
"loss": 5.8309,
"mean_token_accuracy": 0.16621251851320268,
"num_tokens": 4328795.0,
"step": 1940
},
{
"entropy": 5.854658794403076,
"epoch": 0.21845341719548492,
"grad_norm": 0.921875,
"learning_rate": 0.000499872328554181,
"loss": 5.8404,
"mean_token_accuracy": 0.1629622608423233,
"num_tokens": 4340279.0,
"step": 1945
},
{
"entropy": 6.01924500465393,
"epoch": 0.21901499410344247,
"grad_norm": 1.0625,
"learning_rate": 0.0004998709726503383,
"loss": 6.0073,
"mean_token_accuracy": 0.15287342071533203,
"num_tokens": 4352157.0,
"step": 1950
},
{
"entropy": 5.869948482513427,
"epoch": 0.21957657101140002,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998696095865168,
"loss": 5.8464,
"mean_token_accuracy": 0.15993180274963378,
"num_tokens": 4361909.0,
"step": 1955
},
{
"entropy": 5.694135141372681,
"epoch": 0.22013814791935757,
"grad_norm": 1.140625,
"learning_rate": 0.0004998682393627595,
"loss": 5.7801,
"mean_token_accuracy": 0.15679869204759597,
"num_tokens": 4372177.0,
"step": 1960
},
{
"entropy": 5.851674175262451,
"epoch": 0.2206997248273151,
"grad_norm": 1.0,
"learning_rate": 0.0004998668619791104,
"loss": 5.8245,
"mean_token_accuracy": 0.16455647349357605,
"num_tokens": 4382872.0,
"step": 1965
},
{
"entropy": 5.910435581207276,
"epoch": 0.22126130173527264,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998654774356131,
"loss": 5.9296,
"mean_token_accuracy": 0.15243469476699828,
"num_tokens": 4393623.0,
"step": 1970
},
{
"entropy": 5.87087049484253,
"epoch": 0.2218228786432302,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998640857323118,
"loss": 5.8629,
"mean_token_accuracy": 0.15855282992124559,
"num_tokens": 4404215.0,
"step": 1975
},
{
"entropy": 5.980984401702881,
"epoch": 0.22238445555118774,
"grad_norm": 1.03125,
"learning_rate": 0.0004998626868692509,
"loss": 6.1252,
"mean_token_accuracy": 0.1530395619571209,
"num_tokens": 4415978.0,
"step": 1980
},
{
"entropy": 5.92978253364563,
"epoch": 0.2229460324591453,
"grad_norm": 1.03125,
"learning_rate": 0.0004998612808464748,
"loss": 5.8779,
"mean_token_accuracy": 0.1633330002427101,
"num_tokens": 4427060.0,
"step": 1985
},
{
"entropy": 5.8504064083099365,
"epoch": 0.22350760936710282,
"grad_norm": 0.953125,
"learning_rate": 0.0004998598676640283,
"loss": 5.9388,
"mean_token_accuracy": 0.1539946474134922,
"num_tokens": 4438145.0,
"step": 1990
},
{
"entropy": 6.1169726848602295,
"epoch": 0.22406918627506037,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998584473219563,
"loss": 6.0259,
"mean_token_accuracy": 0.1604885935783386,
"num_tokens": 4449324.0,
"step": 1995
},
{
"entropy": 5.830072832107544,
"epoch": 0.22463076318301792,
"grad_norm": 1.078125,
"learning_rate": 0.0004998570198203043,
"loss": 5.7461,
"mean_token_accuracy": 0.16123024970293046,
"num_tokens": 4459728.0,
"step": 2000
},
{
"entropy": 5.833692741394043,
"epoch": 0.22519234009097547,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998555851591175,
"loss": 5.9654,
"mean_token_accuracy": 0.15974030271172523,
"num_tokens": 4471164.0,
"step": 2005
},
{
"entropy": 5.977700471878052,
"epoch": 0.225753916998933,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998541433384416,
"loss": 6.0127,
"mean_token_accuracy": 0.15873760730028152,
"num_tokens": 4483349.0,
"step": 2010
},
{
"entropy": 6.0327905178070065,
"epoch": 0.22631549390689054,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998526943583226,
"loss": 5.986,
"mean_token_accuracy": 0.1547345146536827,
"num_tokens": 4495311.0,
"step": 2015
},
{
"entropy": 5.838544273376465,
"epoch": 0.2268770708148481,
"grad_norm": 1.09375,
"learning_rate": 0.0004998512382188066,
"loss": 5.9097,
"mean_token_accuracy": 0.15987676233053208,
"num_tokens": 4507017.0,
"step": 2020
},
{
"entropy": 5.924619531631469,
"epoch": 0.22743864772280564,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998497749199401,
"loss": 5.9415,
"mean_token_accuracy": 0.1556266188621521,
"num_tokens": 4517746.0,
"step": 2025
},
{
"entropy": 5.994407606124878,
"epoch": 0.2280002246307632,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998483044617694,
"loss": 6.0424,
"mean_token_accuracy": 0.1516892984509468,
"num_tokens": 4529149.0,
"step": 2030
},
{
"entropy": 5.9190531253814695,
"epoch": 0.22856180153872072,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998468268443416,
"loss": 5.9182,
"mean_token_accuracy": 0.15801755338907242,
"num_tokens": 4540630.0,
"step": 2035
},
{
"entropy": 5.837906694412231,
"epoch": 0.22912337844667827,
"grad_norm": 1.03125,
"learning_rate": 0.0004998453420677034,
"loss": 5.8474,
"mean_token_accuracy": 0.15756188482046127,
"num_tokens": 4552877.0,
"step": 2040
},
{
"entropy": 5.93119626045227,
"epoch": 0.22968495535463582,
"grad_norm": 1.1640625,
"learning_rate": 0.0004998438501319025,
"loss": 5.9465,
"mean_token_accuracy": 0.16055746376514435,
"num_tokens": 4563634.0,
"step": 2045
},
{
"entropy": 5.885257625579834,
"epoch": 0.23024653226259337,
"grad_norm": 1.109375,
"learning_rate": 0.0004998423510369862,
"loss": 5.8801,
"mean_token_accuracy": 0.16271908730268478,
"num_tokens": 4574007.0,
"step": 2050
},
{
"entropy": 5.919662857055664,
"epoch": 0.23080810917055092,
"grad_norm": 1.109375,
"learning_rate": 0.0004998408447830022,
"loss": 5.8817,
"mean_token_accuracy": 0.15877689719200133,
"num_tokens": 4584146.0,
"step": 2055
},
{
"entropy": 5.917513370513916,
"epoch": 0.23136968607850844,
"grad_norm": 1.078125,
"learning_rate": 0.0004998393313699985,
"loss": 5.8949,
"mean_token_accuracy": 0.15564839094877242,
"num_tokens": 4594839.0,
"step": 2060
},
{
"entropy": 6.0348950862884525,
"epoch": 0.231931262986466,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998378107980233,
"loss": 6.0664,
"mean_token_accuracy": 0.14918494820594788,
"num_tokens": 4608229.0,
"step": 2065
},
{
"entropy": 5.948345947265625,
"epoch": 0.23249283989442354,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998362830671249,
"loss": 5.868,
"mean_token_accuracy": 0.15786196440458297,
"num_tokens": 4619206.0,
"step": 2070
},
{
"entropy": 5.810974550247193,
"epoch": 0.2330544168023811,
"grad_norm": 1.09375,
"learning_rate": 0.0004998347481773522,
"loss": 5.9003,
"mean_token_accuracy": 0.1612463802099228,
"num_tokens": 4628966.0,
"step": 2075
},
{
"entropy": 5.922925519943237,
"epoch": 0.23361599371033864,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998332061287538,
"loss": 5.7708,
"mean_token_accuracy": 0.1643754854798317,
"num_tokens": 4639935.0,
"step": 2080
},
{
"entropy": 5.940003728866577,
"epoch": 0.23417757061829617,
"grad_norm": 1.046875,
"learning_rate": 0.000499831656921379,
"loss": 6.0308,
"mean_token_accuracy": 0.15250603705644608,
"num_tokens": 4651998.0,
"step": 2085
},
{
"entropy": 6.024502515792847,
"epoch": 0.23473914752625372,
"grad_norm": 1.046875,
"learning_rate": 0.000499830100555277,
"loss": 6.0123,
"mean_token_accuracy": 0.15178350806236268,
"num_tokens": 4664002.0,
"step": 2090
},
{
"entropy": 5.8859413146972654,
"epoch": 0.23530072443421127,
"grad_norm": 1.03125,
"learning_rate": 0.0004998285370304974,
"loss": 5.8844,
"mean_token_accuracy": 0.15997599363327025,
"num_tokens": 4675343.0,
"step": 2095
},
{
"entropy": 5.954652452468872,
"epoch": 0.23586230134216882,
"grad_norm": 1.015625,
"learning_rate": 0.00049982696634709,
"loss": 5.9424,
"mean_token_accuracy": 0.1595751538872719,
"num_tokens": 4686410.0,
"step": 2100
},
{
"entropy": 5.892966175079346,
"epoch": 0.23642387825012637,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998253885051048,
"loss": 5.9708,
"mean_token_accuracy": 0.15921736210584642,
"num_tokens": 4697768.0,
"step": 2105
},
{
"entropy": 6.021402645111084,
"epoch": 0.2369854551580839,
"grad_norm": 1.0546875,
"learning_rate": 0.000499823803504592,
"loss": 5.9789,
"mean_token_accuracy": 0.15751142501831056,
"num_tokens": 4709933.0,
"step": 2110
},
{
"entropy": 5.943789482116699,
"epoch": 0.23754703206604144,
"grad_norm": 1.1015625,
"learning_rate": 0.0004998222113456022,
"loss": 6.0011,
"mean_token_accuracy": 0.15058652609586715,
"num_tokens": 4721385.0,
"step": 2115
},
{
"entropy": 5.848850536346435,
"epoch": 0.238108608973999,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998206120281858,
"loss": 5.8405,
"mean_token_accuracy": 0.1643430173397064,
"num_tokens": 4732562.0,
"step": 2120
},
{
"entropy": 5.8342078685760494,
"epoch": 0.23867018588195654,
"grad_norm": 1.0234375,
"learning_rate": 0.000499819005552394,
"loss": 5.845,
"mean_token_accuracy": 0.16131186932325364,
"num_tokens": 4742794.0,
"step": 2125
},
{
"entropy": 5.809117650985717,
"epoch": 0.23923176278991407,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998173919182779,
"loss": 5.8211,
"mean_token_accuracy": 0.16349539011716843,
"num_tokens": 4754903.0,
"step": 2130
},
{
"entropy": 5.895607423782349,
"epoch": 0.23979333969787162,
"grad_norm": 1.0625,
"learning_rate": 0.000499815771125889,
"loss": 5.8638,
"mean_token_accuracy": 0.15568834841251372,
"num_tokens": 4766165.0,
"step": 2135
},
{
"entropy": 5.870055103302002,
"epoch": 0.24035491660582917,
"grad_norm": 1.109375,
"learning_rate": 0.0004998141431752785,
"loss": 5.8605,
"mean_token_accuracy": 0.16405391991138457,
"num_tokens": 4776551.0,
"step": 2140
},
{
"entropy": 5.973432111740112,
"epoch": 0.24091649351378672,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998125080664987,
"loss": 6.0518,
"mean_token_accuracy": 0.1545905739068985,
"num_tokens": 4788344.0,
"step": 2145
},
{
"entropy": 5.9314477920532225,
"epoch": 0.24147807042174427,
"grad_norm": 1.0625,
"learning_rate": 0.0004998108657996012,
"loss": 5.8448,
"mean_token_accuracy": 0.15539077967405318,
"num_tokens": 4799314.0,
"step": 2150
},
{
"entropy": 5.852494192123413,
"epoch": 0.2420396473297018,
"grad_norm": 1.0625,
"learning_rate": 0.0004998092163746387,
"loss": 5.9001,
"mean_token_accuracy": 0.16167995780706407,
"num_tokens": 4809588.0,
"step": 2155
},
{
"entropy": 5.873464822769165,
"epoch": 0.24260122423765934,
"grad_norm": 1.046875,
"learning_rate": 0.0004998075597916636,
"loss": 5.7199,
"mean_token_accuracy": 0.16549544185400009,
"num_tokens": 4820076.0,
"step": 2160
},
{
"entropy": 5.871007013320923,
"epoch": 0.2431628011456169,
"grad_norm": 1.046875,
"learning_rate": 0.0004998058960507286,
"loss": 5.9969,
"mean_token_accuracy": 0.15287212729454042,
"num_tokens": 4831504.0,
"step": 2165
},
{
"entropy": 5.903195571899414,
"epoch": 0.24372437805357444,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998042251518866,
"loss": 5.852,
"mean_token_accuracy": 0.15901218801736833,
"num_tokens": 4842826.0,
"step": 2170
},
{
"entropy": 5.77811713218689,
"epoch": 0.244285954961532,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998025470951908,
"loss": 5.8241,
"mean_token_accuracy": 0.1635367751121521,
"num_tokens": 4853279.0,
"step": 2175
},
{
"entropy": 5.900368165969849,
"epoch": 0.24484753186948952,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998008618806949,
"loss": 5.8797,
"mean_token_accuracy": 0.1670658990740776,
"num_tokens": 4863332.0,
"step": 2180
},
{
"entropy": 5.892571830749512,
"epoch": 0.24540910877744707,
"grad_norm": 1.0546875,
"learning_rate": 0.0004997991695084523,
"loss": 6.0107,
"mean_token_accuracy": 0.14886087477207183,
"num_tokens": 4874217.0,
"step": 2185
},
{
"entropy": 5.966875743865967,
"epoch": 0.24597068568540462,
"grad_norm": 1.0625,
"learning_rate": 0.0004997974699785169,
"loss": 5.8602,
"mean_token_accuracy": 0.16415059491991996,
"num_tokens": 4885322.0,
"step": 2190
},
{
"entropy": 5.766821432113647,
"epoch": 0.24653226259336217,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997957632909429,
"loss": 5.8141,
"mean_token_accuracy": 0.16065592169761658,
"num_tokens": 4895546.0,
"step": 2195
},
{
"entropy": 6.031574392318726,
"epoch": 0.24709383950131972,
"grad_norm": 1.15625,
"learning_rate": 0.0004997940494457846,
"loss": 6.0277,
"mean_token_accuracy": 0.15350462794303893,
"num_tokens": 4906957.0,
"step": 2200
},
{
"entropy": 5.988462448120117,
"epoch": 0.24765541640927724,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997923284430967,
"loss": 5.9386,
"mean_token_accuracy": 0.15301946252584459,
"num_tokens": 4917965.0,
"step": 2205
},
{
"entropy": 5.804724073410034,
"epoch": 0.2482169933172348,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997906002829338,
"loss": 5.8381,
"mean_token_accuracy": 0.16316511780023574,
"num_tokens": 4928585.0,
"step": 2210
},
{
"entropy": 5.896196508407593,
"epoch": 0.24877857022519234,
"grad_norm": 0.9921875,
"learning_rate": 0.000499788864965351,
"loss": 5.8441,
"mean_token_accuracy": 0.16036625653505326,
"num_tokens": 4939818.0,
"step": 2215
},
{
"entropy": 5.913401126861572,
"epoch": 0.2493401471331499,
"grad_norm": 1.0625,
"learning_rate": 0.0004997871224904036,
"loss": 5.9333,
"mean_token_accuracy": 0.15534456223249435,
"num_tokens": 4951752.0,
"step": 2220
},
{
"entropy": 5.7823755741119385,
"epoch": 0.24990172404110744,
"grad_norm": 1.109375,
"learning_rate": 0.0004997853728581468,
"loss": 5.769,
"mean_token_accuracy": 0.16761937439441682,
"num_tokens": 4961846.0,
"step": 2225
},
{
"entropy": 5.883668756484985,
"epoch": 0.250463300949065,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997836160686368,
"loss": 5.9231,
"mean_token_accuracy": 0.15353350937366486,
"num_tokens": 4973693.0,
"step": 2230
},
{
"entropy": 5.913628482818604,
"epoch": 0.25102487785702254,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997818521219291,
"loss": 5.8394,
"mean_token_accuracy": 0.16180141270160675,
"num_tokens": 4986156.0,
"step": 2235
},
{
"entropy": 5.8587078094482425,
"epoch": 0.25158645476498004,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997800810180803,
"loss": 5.92,
"mean_token_accuracy": 0.1600989058613777,
"num_tokens": 4997615.0,
"step": 2240
},
{
"entropy": 5.905812120437622,
"epoch": 0.2521480316729376,
"grad_norm": 1.0625,
"learning_rate": 0.0004997783027571464,
"loss": 5.8608,
"mean_token_accuracy": 0.16093384325504304,
"num_tokens": 5007655.0,
"step": 2245
},
{
"entropy": 5.831983852386474,
"epoch": 0.25270960858089514,
"grad_norm": 1.078125,
"learning_rate": 0.0004997765173391841,
"loss": 5.9228,
"mean_token_accuracy": 0.16303323209285736,
"num_tokens": 5018933.0,
"step": 2250
},
{
"entropy": 6.02240252494812,
"epoch": 0.2532711854888527,
"grad_norm": 1.109375,
"learning_rate": 0.0004997747247642505,
"loss": 5.9194,
"mean_token_accuracy": 0.15804935097694398,
"num_tokens": 5030495.0,
"step": 2255
},
{
"entropy": 5.899349594116211,
"epoch": 0.25383276239681024,
"grad_norm": 1.03125,
"learning_rate": 0.0004997729250324023,
"loss": 5.9707,
"mean_token_accuracy": 0.15581197887659073,
"num_tokens": 5041553.0,
"step": 2260
},
{
"entropy": 5.873454856872558,
"epoch": 0.2543943393047678,
"grad_norm": 1.0625,
"learning_rate": 0.0004997711181436971,
"loss": 5.8922,
"mean_token_accuracy": 0.16303403079509735,
"num_tokens": 5052312.0,
"step": 2265
},
{
"entropy": 5.983870792388916,
"epoch": 0.25495591621272534,
"grad_norm": 1.015625,
"learning_rate": 0.0004997693040981922,
"loss": 6.0817,
"mean_token_accuracy": 0.15263137072324753,
"num_tokens": 5063445.0,
"step": 2270
},
{
"entropy": 5.986553621292114,
"epoch": 0.2555174931206829,
"grad_norm": 1.1484375,
"learning_rate": 0.0004997674828959456,
"loss": 5.8491,
"mean_token_accuracy": 0.16022175401449204,
"num_tokens": 5073052.0,
"step": 2275
},
{
"entropy": 5.865057420730591,
"epoch": 0.25607907002864044,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997656545370151,
"loss": 5.9049,
"mean_token_accuracy": 0.16042996644973756,
"num_tokens": 5084174.0,
"step": 2280
},
{
"entropy": 5.910764932632446,
"epoch": 0.256640646936598,
"grad_norm": 1.078125,
"learning_rate": 0.000499763819021459,
"loss": 5.8863,
"mean_token_accuracy": 0.1606077790260315,
"num_tokens": 5094106.0,
"step": 2285
},
{
"entropy": 5.987953424453735,
"epoch": 0.2572022238445555,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997619763493357,
"loss": 5.9748,
"mean_token_accuracy": 0.15885440707206727,
"num_tokens": 5106963.0,
"step": 2290
},
{
"entropy": 5.774737882614136,
"epoch": 0.25776380075251304,
"grad_norm": 0.9765625,
"learning_rate": 0.000499760126520704,
"loss": 5.7889,
"mean_token_accuracy": 0.15251622200012208,
"num_tokens": 5119476.0,
"step": 2295
},
{
"entropy": 5.9726386070251465,
"epoch": 0.2583253776604706,
"grad_norm": 1.140625,
"learning_rate": 0.0004997582695356225,
"loss": 6.0337,
"mean_token_accuracy": 0.14983026534318925,
"num_tokens": 5132169.0,
"step": 2300
},
{
"entropy": 5.869831800460815,
"epoch": 0.25888695456842814,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997564053941507,
"loss": 5.8598,
"mean_token_accuracy": 0.15989481210708617,
"num_tokens": 5143354.0,
"step": 2305
},
{
"entropy": 5.843227052688599,
"epoch": 0.2594485314763857,
"grad_norm": 1.140625,
"learning_rate": 0.0004997545340963478,
"loss": 5.9274,
"mean_token_accuracy": 0.15680477470159532,
"num_tokens": 5154624.0,
"step": 2310
},
{
"entropy": 5.9521379470825195,
"epoch": 0.26001010838434324,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997526556422733,
"loss": 5.8292,
"mean_token_accuracy": 0.1580936551094055,
"num_tokens": 5165349.0,
"step": 2315
},
{
"entropy": 5.822861623764038,
"epoch": 0.2605716852923008,
"grad_norm": 1.28125,
"learning_rate": 0.0004997507700319872,
"loss": 5.9123,
"mean_token_accuracy": 0.1558194637298584,
"num_tokens": 5175874.0,
"step": 2320
},
{
"entropy": 5.890210199356079,
"epoch": 0.26113326220025834,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997488772655492,
"loss": 5.7924,
"mean_token_accuracy": 0.16859100311994552,
"num_tokens": 5186591.0,
"step": 2325
},
{
"entropy": 5.807303524017334,
"epoch": 0.2616948391082159,
"grad_norm": 1.140625,
"learning_rate": 0.0004997469773430199,
"loss": 5.9036,
"mean_token_accuracy": 0.16755606681108476,
"num_tokens": 5197307.0,
"step": 2330
},
{
"entropy": 5.858344030380249,
"epoch": 0.2622564160161734,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997450702644596,
"loss": 5.8595,
"mean_token_accuracy": 0.15741084516048431,
"num_tokens": 5209868.0,
"step": 2335
},
{
"entropy": 5.90931806564331,
"epoch": 0.26281799292413094,
"grad_norm": 1.03125,
"learning_rate": 0.0004997431560299292,
"loss": 5.8743,
"mean_token_accuracy": 0.16276440173387527,
"num_tokens": 5220365.0,
"step": 2340
},
{
"entropy": 5.818673753738404,
"epoch": 0.2633795698320885,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997412346394894,
"loss": 5.7881,
"mean_token_accuracy": 0.16023485809564592,
"num_tokens": 5231575.0,
"step": 2345
},
{
"entropy": 5.897857189178467,
"epoch": 0.26394114674004604,
"grad_norm": 1.2109375,
"learning_rate": 0.0004997393060932017,
"loss": 5.8785,
"mean_token_accuracy": 0.1601746052503586,
"num_tokens": 5241852.0,
"step": 2350
},
{
"entropy": 5.899834012985229,
"epoch": 0.2645027236480036,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997373703911273,
"loss": 5.9222,
"mean_token_accuracy": 0.16097686886787416,
"num_tokens": 5254536.0,
"step": 2355
},
{
"entropy": 5.932958650588989,
"epoch": 0.26506430055596114,
"grad_norm": 1.1328125,
"learning_rate": 0.0004997354275333277,
"loss": 5.8317,
"mean_token_accuracy": 0.1640108659863472,
"num_tokens": 5264946.0,
"step": 2360
},
{
"entropy": 5.631682395935059,
"epoch": 0.2656258774639187,
"grad_norm": 1.0390625,
"learning_rate": 0.000499733477519865,
"loss": 5.6754,
"mean_token_accuracy": 0.16263261884450914,
"num_tokens": 5275527.0,
"step": 2365
},
{
"entropy": 5.878888082504273,
"epoch": 0.26618745437187624,
"grad_norm": 1.140625,
"learning_rate": 0.0004997315203508013,
"loss": 5.8516,
"mean_token_accuracy": 0.1570986658334732,
"num_tokens": 5285471.0,
"step": 2370
},
{
"entropy": 5.764362859725952,
"epoch": 0.2667490312798338,
"grad_norm": 1.1875,
"learning_rate": 0.0004997295560261988,
"loss": 5.7642,
"mean_token_accuracy": 0.15874749571084976,
"num_tokens": 5295743.0,
"step": 2375
},
{
"entropy": 5.877685689926148,
"epoch": 0.26731060818779134,
"grad_norm": 1.1328125,
"learning_rate": 0.00049972758454612,
"loss": 5.9113,
"mean_token_accuracy": 0.1542122907936573,
"num_tokens": 5307963.0,
"step": 2380
},
{
"entropy": 5.861995458602905,
"epoch": 0.26787218509574884,
"grad_norm": 1.1875,
"learning_rate": 0.0004997256059106279,
"loss": 5.8059,
"mean_token_accuracy": 0.16155965775251388,
"num_tokens": 5318394.0,
"step": 2385
},
{
"entropy": 5.900924968719482,
"epoch": 0.2684337620037064,
"grad_norm": 0.9375,
"learning_rate": 0.0004997236201197853,
"loss": 6.0391,
"mean_token_accuracy": 0.14638192504644393,
"num_tokens": 5331728.0,
"step": 2390
},
{
"entropy": 5.966968250274658,
"epoch": 0.26899533891166394,
"grad_norm": 1.1171875,
"learning_rate": 0.0004997216271736554,
"loss": 5.8965,
"mean_token_accuracy": 0.1681509181857109,
"num_tokens": 5342825.0,
"step": 2395
},
{
"entropy": 5.880633354187012,
"epoch": 0.2695569158196215,
"grad_norm": 1.03125,
"learning_rate": 0.0004997196270723018,
"loss": 5.8227,
"mean_token_accuracy": 0.16092611998319625,
"num_tokens": 5354232.0,
"step": 2400
},
{
"entropy": 5.8500697135925295,
"epoch": 0.27011849272757904,
"grad_norm": 1.171875,
"learning_rate": 0.0004997176198157881,
"loss": 5.898,
"mean_token_accuracy": 0.15567438974976539,
"num_tokens": 5364807.0,
"step": 2405
},
{
"entropy": 5.827037715911866,
"epoch": 0.2706800696355366,
"grad_norm": 1.046875,
"learning_rate": 0.0004997156054041781,
"loss": 5.8845,
"mean_token_accuracy": 0.160299713909626,
"num_tokens": 5375593.0,
"step": 2410
},
{
"entropy": 5.85796971321106,
"epoch": 0.27124164654349414,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997135838375362,
"loss": 5.834,
"mean_token_accuracy": 0.16348498463630676,
"num_tokens": 5386740.0,
"step": 2415
},
{
"entropy": 5.861660146713257,
"epoch": 0.2718032234514517,
"grad_norm": 1.0625,
"learning_rate": 0.0004997115551159267,
"loss": 5.8209,
"mean_token_accuracy": 0.16418611854314805,
"num_tokens": 5397732.0,
"step": 2420
},
{
"entropy": 5.94402961730957,
"epoch": 0.27236480035940924,
"grad_norm": 1.015625,
"learning_rate": 0.000499709519239414,
"loss": 5.8794,
"mean_token_accuracy": 0.15414437800645828,
"num_tokens": 5409392.0,
"step": 2425
},
{
"entropy": 5.664468765258789,
"epoch": 0.2729263772673668,
"grad_norm": 0.96875,
"learning_rate": 0.0004997074762080632,
"loss": 5.8412,
"mean_token_accuracy": 0.16693630516529084,
"num_tokens": 5420903.0,
"step": 2430
},
{
"entropy": 5.802863931655883,
"epoch": 0.2734879541753243,
"grad_norm": 0.9453125,
"learning_rate": 0.000499705426021939,
"loss": 5.7761,
"mean_token_accuracy": 0.15942942947149277,
"num_tokens": 5432631.0,
"step": 2435
},
{
"entropy": 5.864587879180908,
"epoch": 0.27404953108328184,
"grad_norm": 1.0703125,
"learning_rate": 0.000499703368681107,
"loss": 5.7458,
"mean_token_accuracy": 0.16454415321350097,
"num_tokens": 5443463.0,
"step": 2440
},
{
"entropy": 5.8874420642852785,
"epoch": 0.2746111079912394,
"grad_norm": 1.046875,
"learning_rate": 0.0004997013041856326,
"loss": 6.0522,
"mean_token_accuracy": 0.15659836456179618,
"num_tokens": 5455903.0,
"step": 2445
},
{
"entropy": 5.890227794647217,
"epoch": 0.27517268489919694,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996992325355815,
"loss": 5.8903,
"mean_token_accuracy": 0.15483457446098328,
"num_tokens": 5467201.0,
"step": 2450
},
{
"entropy": 5.875626754760742,
"epoch": 0.2757342618071545,
"grad_norm": 0.984375,
"learning_rate": 0.0004996971537310198,
"loss": 5.7529,
"mean_token_accuracy": 0.16340564638376237,
"num_tokens": 5478109.0,
"step": 2455
},
{
"entropy": 5.876842784881592,
"epoch": 0.27629583871511204,
"grad_norm": 1.0,
"learning_rate": 0.0004996950677720133,
"loss": 5.8787,
"mean_token_accuracy": 0.15813624560832978,
"num_tokens": 5489621.0,
"step": 2460
},
{
"entropy": 5.80870246887207,
"epoch": 0.2768574156230696,
"grad_norm": 1.03125,
"learning_rate": 0.000499692974658629,
"loss": 5.8328,
"mean_token_accuracy": 0.1639888569712639,
"num_tokens": 5500606.0,
"step": 2465
},
{
"entropy": 5.865239953994751,
"epoch": 0.27741899253102714,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996908743909331,
"loss": 5.7753,
"mean_token_accuracy": 0.17143337428569794,
"num_tokens": 5510768.0,
"step": 2470
},
{
"entropy": 5.7702569484710695,
"epoch": 0.2779805694389847,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996887669689926,
"loss": 5.8076,
"mean_token_accuracy": 0.16477752476930618,
"num_tokens": 5522210.0,
"step": 2475
},
{
"entropy": 5.871284866333008,
"epoch": 0.2785421463469422,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996866523928747,
"loss": 5.9173,
"mean_token_accuracy": 0.15478019267320633,
"num_tokens": 5533055.0,
"step": 2480
},
{
"entropy": 5.936899280548095,
"epoch": 0.27910372325489974,
"grad_norm": 1.0625,
"learning_rate": 0.0004996845306626467,
"loss": 5.9302,
"mean_token_accuracy": 0.15565606355667114,
"num_tokens": 5544345.0,
"step": 2485
},
{
"entropy": 5.899860382080078,
"epoch": 0.2796653001628573,
"grad_norm": 1.15625,
"learning_rate": 0.0004996824017783759,
"loss": 5.8508,
"mean_token_accuracy": 0.1602659970521927,
"num_tokens": 5556734.0,
"step": 2490
},
{
"entropy": 5.805173587799072,
"epoch": 0.28022687707081484,
"grad_norm": 1.1328125,
"learning_rate": 0.0004996802657401304,
"loss": 5.8803,
"mean_token_accuracy": 0.16296618580818176,
"num_tokens": 5566874.0,
"step": 2495
},
{
"entropy": 5.9315550327301025,
"epoch": 0.2807884539787724,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996781225479781,
"loss": 5.8878,
"mean_token_accuracy": 0.15491053313016892,
"num_tokens": 5578812.0,
"step": 2500
},
{
"entropy": 5.847195196151733,
"epoch": 0.28135003088672994,
"grad_norm": 1.0625,
"learning_rate": 0.0004996759722019872,
"loss": 5.9389,
"mean_token_accuracy": 0.16007010489702225,
"num_tokens": 5590021.0,
"step": 2505
},
{
"entropy": 5.904675149917603,
"epoch": 0.2819116077946875,
"grad_norm": 1.109375,
"learning_rate": 0.0004996738147022262,
"loss": 5.9373,
"mean_token_accuracy": 0.150753453373909,
"num_tokens": 5600921.0,
"step": 2510
},
{
"entropy": 5.90674614906311,
"epoch": 0.28247318470264504,
"grad_norm": 1.0625,
"learning_rate": 0.0004996716500487638,
"loss": 5.8641,
"mean_token_accuracy": 0.15689948797225953,
"num_tokens": 5610615.0,
"step": 2515
},
{
"entropy": 5.765136003494263,
"epoch": 0.2830347616106026,
"grad_norm": 1.078125,
"learning_rate": 0.000499669478241669,
"loss": 5.7885,
"mean_token_accuracy": 0.1646771937608719,
"num_tokens": 5621596.0,
"step": 2520
},
{
"entropy": 5.968653011322021,
"epoch": 0.28359633851856014,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996672992810108,
"loss": 5.8273,
"mean_token_accuracy": 0.16271875649690629,
"num_tokens": 5633047.0,
"step": 2525
},
{
"entropy": 5.8039604187011715,
"epoch": 0.28415791542651764,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996651131668587,
"loss": 5.811,
"mean_token_accuracy": 0.15627873688936234,
"num_tokens": 5644970.0,
"step": 2530
},
{
"entropy": 5.813772583007813,
"epoch": 0.2847194923344752,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996629198992823,
"loss": 5.939,
"mean_token_accuracy": 0.15850594341754914,
"num_tokens": 5656631.0,
"step": 2535
},
{
"entropy": 5.905940103530884,
"epoch": 0.28528106924243274,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996607194783513,
"loss": 5.8919,
"mean_token_accuracy": 0.15616682022809983,
"num_tokens": 5668808.0,
"step": 2540
},
{
"entropy": 5.760390377044677,
"epoch": 0.2858426461503903,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996585119041359,
"loss": 5.8169,
"mean_token_accuracy": 0.1564560815691948,
"num_tokens": 5679493.0,
"step": 2545
},
{
"entropy": 5.950744247436523,
"epoch": 0.28640422305834784,
"grad_norm": 1.046875,
"learning_rate": 0.0004996562971767063,
"loss": 5.8465,
"mean_token_accuracy": 0.16337623000144957,
"num_tokens": 5689879.0,
"step": 2550
},
{
"entropy": 5.8402002334594725,
"epoch": 0.2869657999663054,
"grad_norm": 1.1171875,
"learning_rate": 0.000499654075296133,
"loss": 5.9299,
"mean_token_accuracy": 0.1620929419994354,
"num_tokens": 5699934.0,
"step": 2555
},
{
"entropy": 5.833514976501465,
"epoch": 0.28752737687426294,
"grad_norm": 1.109375,
"learning_rate": 0.000499651846262487,
"loss": 5.847,
"mean_token_accuracy": 0.16643718332052232,
"num_tokens": 5710171.0,
"step": 2560
},
{
"entropy": 5.89036922454834,
"epoch": 0.2880889537822205,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996496100758389,
"loss": 5.8521,
"mean_token_accuracy": 0.15369296669960023,
"num_tokens": 5721190.0,
"step": 2565
},
{
"entropy": 5.9653236865997314,
"epoch": 0.28865053069017804,
"grad_norm": 1.203125,
"learning_rate": 0.0004996473667362604,
"loss": 5.9678,
"mean_token_accuracy": 0.15457451790571214,
"num_tokens": 5733706.0,
"step": 2570
},
{
"entropy": 5.831788444519043,
"epoch": 0.28921210759813554,
"grad_norm": 1.0859375,
"learning_rate": 0.0004996451162438224,
"loss": 5.9374,
"mean_token_accuracy": 0.1601427301764488,
"num_tokens": 5745712.0,
"step": 2575
},
{
"entropy": 5.781694507598877,
"epoch": 0.2897736845060931,
"grad_norm": 1.203125,
"learning_rate": 0.0004996428585985967,
"loss": 5.7341,
"mean_token_accuracy": 0.17258975356817247,
"num_tokens": 5757915.0,
"step": 2580
},
{
"entropy": 5.880732631683349,
"epoch": 0.29033526141405064,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996405938006553,
"loss": 5.8882,
"mean_token_accuracy": 0.16366344690322876,
"num_tokens": 5768788.0,
"step": 2585
},
{
"entropy": 5.860728406906128,
"epoch": 0.2908968383220082,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996383218500703,
"loss": 5.8124,
"mean_token_accuracy": 0.1624564841389656,
"num_tokens": 5779806.0,
"step": 2590
},
{
"entropy": 5.755586099624634,
"epoch": 0.29145841522996574,
"grad_norm": 1.109375,
"learning_rate": 0.000499636042746914,
"loss": 5.7729,
"mean_token_accuracy": 0.1583534672856331,
"num_tokens": 5790177.0,
"step": 2595
},
{
"entropy": 5.843555927276611,
"epoch": 0.2920199921379233,
"grad_norm": 1.0859375,
"learning_rate": 0.000499633756491259,
"loss": 5.8031,
"mean_token_accuracy": 0.16728467345237732,
"num_tokens": 5800251.0,
"step": 2600
},
{
"entropy": 5.909883689880371,
"epoch": 0.29258156904588084,
"grad_norm": 1.03125,
"learning_rate": 0.0004996314630831781,
"loss": 5.8989,
"mean_token_accuracy": 0.1601306065917015,
"num_tokens": 5812210.0,
"step": 2605
},
{
"entropy": 5.9030238628387455,
"epoch": 0.2931431459538384,
"grad_norm": 0.97265625,
"learning_rate": 0.0004996291625227443,
"loss": 5.8644,
"mean_token_accuracy": 0.16422059535980224,
"num_tokens": 5823280.0,
"step": 2610
},
{
"entropy": 5.924188566207886,
"epoch": 0.29370472286179594,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996268548100307,
"loss": 5.9502,
"mean_token_accuracy": 0.15365658402442933,
"num_tokens": 5835076.0,
"step": 2615
},
{
"entropy": 5.797477912902832,
"epoch": 0.2942662997697535,
"grad_norm": 1.0859375,
"learning_rate": 0.000499624539945111,
"loss": 5.7425,
"mean_token_accuracy": 0.16603392660617827,
"num_tokens": 5845376.0,
"step": 2620
},
{
"entropy": 5.711445426940918,
"epoch": 0.294827876677711,
"grad_norm": 1.03125,
"learning_rate": 0.0004996222179280589,
"loss": 5.7702,
"mean_token_accuracy": 0.1608809530735016,
"num_tokens": 5857124.0,
"step": 2625
},
{
"entropy": 5.9171717166900635,
"epoch": 0.29538945358566854,
"grad_norm": 1.046875,
"learning_rate": 0.0004996198887589481,
"loss": 5.8885,
"mean_token_accuracy": 0.15417518019676207,
"num_tokens": 5868681.0,
"step": 2630
},
{
"entropy": 5.88248610496521,
"epoch": 0.2959510304936261,
"grad_norm": 1.015625,
"learning_rate": 0.0004996175524378531,
"loss": 5.8621,
"mean_token_accuracy": 0.16559387296438216,
"num_tokens": 5880532.0,
"step": 2635
},
{
"entropy": 5.840338468551636,
"epoch": 0.29651260740158364,
"grad_norm": 1.15625,
"learning_rate": 0.000499615208964848,
"loss": 5.8342,
"mean_token_accuracy": 0.16637962609529494,
"num_tokens": 5891582.0,
"step": 2640
},
{
"entropy": 5.8416746139526365,
"epoch": 0.2970741843095412,
"grad_norm": 1.1875,
"learning_rate": 0.0004996128583400077,
"loss": 5.8178,
"mean_token_accuracy": 0.15835877358913422,
"num_tokens": 5903543.0,
"step": 2645
},
{
"entropy": 5.731740427017212,
"epoch": 0.29763576121749874,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996105005634066,
"loss": 5.7954,
"mean_token_accuracy": 0.16752633601427078,
"num_tokens": 5913140.0,
"step": 2650
},
{
"entropy": 5.834004831314087,
"epoch": 0.2981973381254563,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996081356351202,
"loss": 5.755,
"mean_token_accuracy": 0.16460932940244674,
"num_tokens": 5924010.0,
"step": 2655
},
{
"entropy": 5.876258516311646,
"epoch": 0.29875891503341384,
"grad_norm": 1.1640625,
"learning_rate": 0.0004996057635552237,
"loss": 5.9686,
"mean_token_accuracy": 0.15854546874761583,
"num_tokens": 5936246.0,
"step": 2660
},
{
"entropy": 5.855793046951294,
"epoch": 0.2993204919413714,
"grad_norm": 1.171875,
"learning_rate": 0.0004996033843237925,
"loss": 5.8589,
"mean_token_accuracy": 0.15378582924604417,
"num_tokens": 5947594.0,
"step": 2665
},
{
"entropy": 5.903825950622559,
"epoch": 0.29988206884932894,
"grad_norm": 1.203125,
"learning_rate": 0.0004996009979409025,
"loss": 5.7911,
"mean_token_accuracy": 0.15631641298532487,
"num_tokens": 5957660.0,
"step": 2670
},
{
"entropy": 5.682236814498902,
"epoch": 0.30044364575728644,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995986044066295,
"loss": 5.6203,
"mean_token_accuracy": 0.17106491774320604,
"num_tokens": 5968881.0,
"step": 2675
},
{
"entropy": 5.778797912597656,
"epoch": 0.301005222665244,
"grad_norm": 1.09375,
"learning_rate": 0.00049959620372105,
"loss": 5.7907,
"mean_token_accuracy": 0.16492329984903337,
"num_tokens": 5980032.0,
"step": 2680
},
{
"entropy": 5.740841341018677,
"epoch": 0.30156679957320154,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995937958842401,
"loss": 5.8008,
"mean_token_accuracy": 0.16221535503864287,
"num_tokens": 5991781.0,
"step": 2685
},
{
"entropy": 5.818138599395752,
"epoch": 0.3021283764811591,
"grad_norm": 1.1328125,
"learning_rate": 0.0004995913808962767,
"loss": 5.8343,
"mean_token_accuracy": 0.1621527448296547,
"num_tokens": 6002303.0,
"step": 2690
},
{
"entropy": 5.8888787746429445,
"epoch": 0.30268995338911664,
"grad_norm": 1.2109375,
"learning_rate": 0.0004995889587572366,
"loss": 5.9065,
"mean_token_accuracy": 0.1600490316748619,
"num_tokens": 6011523.0,
"step": 2695
},
{
"entropy": 5.85958423614502,
"epoch": 0.3032515302970742,
"grad_norm": 1.0390625,
"learning_rate": 0.000499586529467197,
"loss": 5.7173,
"mean_token_accuracy": 0.16480220556259156,
"num_tokens": 6022169.0,
"step": 2700
},
{
"entropy": 5.674841737747192,
"epoch": 0.30381310720503174,
"grad_norm": 1.2109375,
"learning_rate": 0.0004995840930262351,
"loss": 5.7695,
"mean_token_accuracy": 0.16273038387298583,
"num_tokens": 6033393.0,
"step": 2705
},
{
"entropy": 5.830945682525635,
"epoch": 0.3043746841129893,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995816494344287,
"loss": 5.8442,
"mean_token_accuracy": 0.15597543269395828,
"num_tokens": 6045365.0,
"step": 2710
},
{
"entropy": 5.830998420715332,
"epoch": 0.30493626102094684,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995791986918554,
"loss": 5.8675,
"mean_token_accuracy": 0.16169115602970124,
"num_tokens": 6057482.0,
"step": 2715
},
{
"entropy": 5.871564197540283,
"epoch": 0.30549783792890434,
"grad_norm": 1.1015625,
"learning_rate": 0.0004995767407985933,
"loss": 5.7681,
"mean_token_accuracy": 0.1672608807682991,
"num_tokens": 6068194.0,
"step": 2720
},
{
"entropy": 5.78353214263916,
"epoch": 0.3060594148368619,
"grad_norm": 1.09375,
"learning_rate": 0.0004995742757547207,
"loss": 5.8293,
"mean_token_accuracy": 0.16143801659345627,
"num_tokens": 6080540.0,
"step": 2725
},
{
"entropy": 5.745278215408325,
"epoch": 0.30662099174481944,
"grad_norm": 1.0859375,
"learning_rate": 0.000499571803560316,
"loss": 5.8713,
"mean_token_accuracy": 0.15866425335407258,
"num_tokens": 6092464.0,
"step": 2730
},
{
"entropy": 5.87835726737976,
"epoch": 0.307182568652777,
"grad_norm": 1.015625,
"learning_rate": 0.0004995693242154581,
"loss": 5.8209,
"mean_token_accuracy": 0.16295599937438965,
"num_tokens": 6103242.0,
"step": 2735
},
{
"entropy": 5.812991285324097,
"epoch": 0.30774414556073454,
"grad_norm": 1.015625,
"learning_rate": 0.0004995668377202258,
"loss": 5.8634,
"mean_token_accuracy": 0.16028525680303574,
"num_tokens": 6116383.0,
"step": 2740
},
{
"entropy": 5.766334486007691,
"epoch": 0.3083057224686921,
"grad_norm": 1.03125,
"learning_rate": 0.0004995643440746982,
"loss": 5.6643,
"mean_token_accuracy": 0.17048484236001968,
"num_tokens": 6126883.0,
"step": 2745
},
{
"entropy": 5.8382853984832765,
"epoch": 0.30886729937664964,
"grad_norm": 1.171875,
"learning_rate": 0.0004995618432789548,
"loss": 5.861,
"mean_token_accuracy": 0.1548911824822426,
"num_tokens": 6139613.0,
"step": 2750
},
{
"entropy": 5.867930698394775,
"epoch": 0.3094288762846072,
"grad_norm": 1.109375,
"learning_rate": 0.0004995593353330753,
"loss": 5.9304,
"mean_token_accuracy": 0.15693748146295547,
"num_tokens": 6151950.0,
"step": 2755
},
{
"entropy": 5.812861728668213,
"epoch": 0.30999045319256474,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995568202371393,
"loss": 5.7224,
"mean_token_accuracy": 0.17015610337257386,
"num_tokens": 6161563.0,
"step": 2760
},
{
"entropy": 5.749365663528442,
"epoch": 0.3105520301005223,
"grad_norm": 1.21875,
"learning_rate": 0.0004995542979912272,
"loss": 5.7369,
"mean_token_accuracy": 0.1699253425002098,
"num_tokens": 6170966.0,
"step": 2765
},
{
"entropy": 5.7737060546875,
"epoch": 0.3111136070084798,
"grad_norm": 1.1171875,
"learning_rate": 0.0004995517685954192,
"loss": 5.8066,
"mean_token_accuracy": 0.16174303740262985,
"num_tokens": 6182563.0,
"step": 2770
},
{
"entropy": 5.850103712081909,
"epoch": 0.31167518391643734,
"grad_norm": 1.078125,
"learning_rate": 0.0004995492320497958,
"loss": 5.9185,
"mean_token_accuracy": 0.15409982949495316,
"num_tokens": 6194495.0,
"step": 2775
},
{
"entropy": 5.839843225479126,
"epoch": 0.3122367608243949,
"grad_norm": 1.0625,
"learning_rate": 0.0004995466883544376,
"loss": 5.8204,
"mean_token_accuracy": 0.15846606194972992,
"num_tokens": 6206594.0,
"step": 2780
},
{
"entropy": 5.840381479263305,
"epoch": 0.31279833773235244,
"grad_norm": 1.0703125,
"learning_rate": 0.0004995441375094258,
"loss": 5.8248,
"mean_token_accuracy": 0.15951959937810897,
"num_tokens": 6218371.0,
"step": 2785
},
{
"entropy": 5.90253176689148,
"epoch": 0.31335991464031,
"grad_norm": 1.0625,
"learning_rate": 0.0004995415795148417,
"loss": 5.7562,
"mean_token_accuracy": 0.1673472911119461,
"num_tokens": 6228723.0,
"step": 2790
},
{
"entropy": 5.830163764953613,
"epoch": 0.31392149154826754,
"grad_norm": 1.140625,
"learning_rate": 0.0004995390143707665,
"loss": 5.9485,
"mean_token_accuracy": 0.1597628489136696,
"num_tokens": 6240535.0,
"step": 2795
},
{
"entropy": 5.762459945678711,
"epoch": 0.3144830684562251,
"grad_norm": 1.0859375,
"learning_rate": 0.0004995364420772821,
"loss": 5.6412,
"mean_token_accuracy": 0.16819480657577515,
"num_tokens": 6250643.0,
"step": 2800
},
{
"entropy": 5.762458372116089,
"epoch": 0.31504464536418264,
"grad_norm": 1.15625,
"learning_rate": 0.0004995338626344702,
"loss": 5.7843,
"mean_token_accuracy": 0.16505634933710098,
"num_tokens": 6259569.0,
"step": 2805
},
{
"entropy": 5.820625829696655,
"epoch": 0.3156062222721402,
"grad_norm": 1.1953125,
"learning_rate": 0.000499531276042413,
"loss": 5.9333,
"mean_token_accuracy": 0.16434977650642396,
"num_tokens": 6271108.0,
"step": 2810
},
{
"entropy": 5.940713834762573,
"epoch": 0.31616779918009774,
"grad_norm": 1.203125,
"learning_rate": 0.0004995286823011929,
"loss": 5.7569,
"mean_token_accuracy": 0.15814750343561174,
"num_tokens": 6282352.0,
"step": 2815
},
{
"entropy": 5.755598640441894,
"epoch": 0.31672937608805524,
"grad_norm": 1.2734375,
"learning_rate": 0.0004995260814108926,
"loss": 5.8778,
"mean_token_accuracy": 0.15963443219661713,
"num_tokens": 6294946.0,
"step": 2820
},
{
"entropy": 5.729330968856812,
"epoch": 0.3172909529960128,
"grad_norm": 1.09375,
"learning_rate": 0.0004995234733715947,
"loss": 5.6965,
"mean_token_accuracy": 0.16342825293540955,
"num_tokens": 6305552.0,
"step": 2825
},
{
"entropy": 5.801796007156372,
"epoch": 0.31785252990397034,
"grad_norm": 1.03125,
"learning_rate": 0.0004995208581833824,
"loss": 5.7689,
"mean_token_accuracy": 0.17125651091337205,
"num_tokens": 6315815.0,
"step": 2830
},
{
"entropy": 5.7499488353729244,
"epoch": 0.3184141068119279,
"grad_norm": 1.2421875,
"learning_rate": 0.0004995182358463388,
"loss": 5.8121,
"mean_token_accuracy": 0.16185301691293716,
"num_tokens": 6327384.0,
"step": 2835
},
{
"entropy": 5.878792953491211,
"epoch": 0.31897568371988544,
"grad_norm": 1.0625,
"learning_rate": 0.0004995156063605475,
"loss": 5.7812,
"mean_token_accuracy": 0.15886088460683823,
"num_tokens": 6338083.0,
"step": 2840
},
{
"entropy": 5.884696388244629,
"epoch": 0.319537260627843,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995129697260922,
"loss": 5.7806,
"mean_token_accuracy": 0.16254436522722243,
"num_tokens": 6349058.0,
"step": 2845
},
{
"entropy": 5.763200044631958,
"epoch": 0.32009883753580054,
"grad_norm": 1.078125,
"learning_rate": 0.0004995103259430569,
"loss": 5.7963,
"mean_token_accuracy": 0.16049613952636718,
"num_tokens": 6359655.0,
"step": 2850
},
{
"entropy": 5.859290027618409,
"epoch": 0.3206604144437581,
"grad_norm": 1.1875,
"learning_rate": 0.0004995076750115258,
"loss": 5.9237,
"mean_token_accuracy": 0.15698505491018294,
"num_tokens": 6372088.0,
"step": 2855
},
{
"entropy": 5.821562719345093,
"epoch": 0.32122199135171564,
"grad_norm": 1.0546875,
"learning_rate": 0.0004995050169315832,
"loss": 5.8296,
"mean_token_accuracy": 0.16010505557060242,
"num_tokens": 6382946.0,
"step": 2860
},
{
"entropy": 5.839111375808716,
"epoch": 0.32178356825967314,
"grad_norm": 1.046875,
"learning_rate": 0.0004995023517033138,
"loss": 5.8192,
"mean_token_accuracy": 0.15969384908676149,
"num_tokens": 6392591.0,
"step": 2865
},
{
"entropy": 5.756126356124878,
"epoch": 0.3223451451676307,
"grad_norm": 1.09375,
"learning_rate": 0.0004994996793268024,
"loss": 5.7604,
"mean_token_accuracy": 0.16648190021514891,
"num_tokens": 6403159.0,
"step": 2870
},
{
"entropy": 5.839804649353027,
"epoch": 0.32290672207558824,
"grad_norm": 1.1640625,
"learning_rate": 0.0004994969998021342,
"loss": 5.7795,
"mean_token_accuracy": 0.1644168198108673,
"num_tokens": 6414232.0,
"step": 2875
},
{
"entropy": 5.806992721557617,
"epoch": 0.3234682989835458,
"grad_norm": 1.09375,
"learning_rate": 0.0004994943131293944,
"loss": 5.7638,
"mean_token_accuracy": 0.17432548552751542,
"num_tokens": 6425762.0,
"step": 2880
},
{
"entropy": 5.756858587265015,
"epoch": 0.32402987589150334,
"grad_norm": 1.0859375,
"learning_rate": 0.0004994916193086687,
"loss": 5.8069,
"mean_token_accuracy": 0.15830935537815094,
"num_tokens": 6439077.0,
"step": 2885
},
{
"entropy": 5.886305570602417,
"epoch": 0.3245914527994609,
"grad_norm": 1.078125,
"learning_rate": 0.0004994889183400428,
"loss": 5.8773,
"mean_token_accuracy": 0.15929146558046342,
"num_tokens": 6450425.0,
"step": 2890
},
{
"entropy": 5.817720079421997,
"epoch": 0.32515302970741844,
"grad_norm": 1.140625,
"learning_rate": 0.0004994862102236025,
"loss": 5.8412,
"mean_token_accuracy": 0.16064791902899742,
"num_tokens": 6461347.0,
"step": 2895
},
{
"entropy": 5.844183015823364,
"epoch": 0.325714606615376,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994834949594342,
"loss": 5.8254,
"mean_token_accuracy": 0.16305376589298248,
"num_tokens": 6472428.0,
"step": 2900
},
{
"entropy": 5.790859222412109,
"epoch": 0.32627618352333354,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994807725476246,
"loss": 5.7773,
"mean_token_accuracy": 0.17451646625995637,
"num_tokens": 6483393.0,
"step": 2905
},
{
"entropy": 5.800299119949341,
"epoch": 0.3268377604312911,
"grad_norm": 1.171875,
"learning_rate": 0.0004994780429882599,
"loss": 5.7707,
"mean_token_accuracy": 0.16646299362182618,
"num_tokens": 6493520.0,
"step": 2910
},
{
"entropy": 5.733630228042602,
"epoch": 0.3273993373392486,
"grad_norm": 0.984375,
"learning_rate": 0.0004994753062814274,
"loss": 5.7145,
"mean_token_accuracy": 0.1693594589829445,
"num_tokens": 6505282.0,
"step": 2915
},
{
"entropy": 5.871979188919068,
"epoch": 0.32796091424720614,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994725624272139,
"loss": 5.8873,
"mean_token_accuracy": 0.16457537859678267,
"num_tokens": 6516808.0,
"step": 2920
},
{
"entropy": 5.8417102813720705,
"epoch": 0.3285224911551637,
"grad_norm": 1.109375,
"learning_rate": 0.0004994698114257071,
"loss": 5.8478,
"mean_token_accuracy": 0.15840103775262832,
"num_tokens": 6529184.0,
"step": 2925
},
{
"entropy": 5.779627275466919,
"epoch": 0.32908406806312124,
"grad_norm": 1.171875,
"learning_rate": 0.0004994670532769944,
"loss": 5.6786,
"mean_token_accuracy": 0.17033708542585374,
"num_tokens": 6539909.0,
"step": 2930
},
{
"entropy": 5.757491588592529,
"epoch": 0.3296456449710788,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994642879811637,
"loss": 5.7469,
"mean_token_accuracy": 0.16941402852535248,
"num_tokens": 6550870.0,
"step": 2935
},
{
"entropy": 5.731397819519043,
"epoch": 0.33020722187903634,
"grad_norm": 1.0546875,
"learning_rate": 0.000499461515538303,
"loss": 5.7978,
"mean_token_accuracy": 0.16825173050165176,
"num_tokens": 6561784.0,
"step": 2940
},
{
"entropy": 5.695051193237305,
"epoch": 0.3307687987869939,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994587359485006,
"loss": 5.7558,
"mean_token_accuracy": 0.16522381603717803,
"num_tokens": 6573422.0,
"step": 2945
},
{
"entropy": 5.857944059371948,
"epoch": 0.33133037569495144,
"grad_norm": 1.0703125,
"learning_rate": 0.000499455949211845,
"loss": 5.8264,
"mean_token_accuracy": 0.17036428451538085,
"num_tokens": 6584565.0,
"step": 2950
},
{
"entropy": 5.805426692962646,
"epoch": 0.331891952602909,
"grad_norm": 1.1875,
"learning_rate": 0.0004994531553284248,
"loss": 5.7163,
"mean_token_accuracy": 0.16582873910665513,
"num_tokens": 6594238.0,
"step": 2955
},
{
"entropy": 5.6973024845123295,
"epoch": 0.3324535295108665,
"grad_norm": 1.0625,
"learning_rate": 0.0004994503542983293,
"loss": 5.8091,
"mean_token_accuracy": 0.16427911669015885,
"num_tokens": 6606309.0,
"step": 2960
},
{
"entropy": 5.958718585968017,
"epoch": 0.33301510641882404,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994475461216472,
"loss": 5.8964,
"mean_token_accuracy": 0.15852140635252,
"num_tokens": 6617916.0,
"step": 2965
},
{
"entropy": 5.791497230529785,
"epoch": 0.3335766833267816,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994447307984684,
"loss": 5.7402,
"mean_token_accuracy": 0.16254918575286864,
"num_tokens": 6627951.0,
"step": 2970
},
{
"entropy": 5.758692979812622,
"epoch": 0.33413826023473914,
"grad_norm": 1.078125,
"learning_rate": 0.0004994419083288823,
"loss": 5.8127,
"mean_token_accuracy": 0.1611504301428795,
"num_tokens": 6639285.0,
"step": 2975
},
{
"entropy": 5.775280380249024,
"epoch": 0.3346998371426967,
"grad_norm": 1.15625,
"learning_rate": 0.0004994390787129787,
"loss": 5.6681,
"mean_token_accuracy": 0.17186915576457978,
"num_tokens": 6649973.0,
"step": 2980
},
{
"entropy": 5.65654149055481,
"epoch": 0.33526141405065424,
"grad_norm": 1.1484375,
"learning_rate": 0.0004994362419508478,
"loss": 5.6986,
"mean_token_accuracy": 0.16761759668588638,
"num_tokens": 6660625.0,
"step": 2985
},
{
"entropy": 5.793176507949829,
"epoch": 0.3358229909586118,
"grad_norm": 1.125,
"learning_rate": 0.00049943339804258,
"loss": 5.7808,
"mean_token_accuracy": 0.16183299124240874,
"num_tokens": 6673191.0,
"step": 2990
},
{
"entropy": 5.859116125106811,
"epoch": 0.33638456786656934,
"grad_norm": 1.125,
"learning_rate": 0.0004994305469882657,
"loss": 5.7747,
"mean_token_accuracy": 0.16482254564762117,
"num_tokens": 6683993.0,
"step": 2995
},
{
"entropy": 5.732059669494629,
"epoch": 0.3369461447745269,
"grad_norm": 1.1015625,
"learning_rate": 0.0004994276887879957,
"loss": 5.8124,
"mean_token_accuracy": 0.16300738751888275,
"num_tokens": 6694881.0,
"step": 3000
},
{
"epoch": 0.3369461447745269,
"eval_entropy": 5.692985942875998,
"eval_loss": 5.818831920623779,
"eval_mean_token_accuracy": 0.16977359401578598,
"eval_num_tokens": 6694881.0,
"eval_runtime": 24.0447,
"eval_samples_per_second": 1289.804,
"eval_steps_per_second": 161.241,
"step": 3000
},
{
"entropy": 5.854642629623413,
"epoch": 0.33750772168248444,
"grad_norm": 1.125,
"learning_rate": 0.0004994248234418611,
"loss": 5.8063,
"mean_token_accuracy": 0.1622493028640747,
"num_tokens": 6706648.0,
"step": 3005
},
{
"entropy": 5.760342454910278,
"epoch": 0.33806929859044194,
"grad_norm": 1.296875,
"learning_rate": 0.000499421950949953,
"loss": 5.8166,
"mean_token_accuracy": 0.16395463645458222,
"num_tokens": 6716858.0,
"step": 3010
},
{
"entropy": 5.759495258331299,
"epoch": 0.3386308754983995,
"grad_norm": 1.1171875,
"learning_rate": 0.000499419071312363,
"loss": 5.7722,
"mean_token_accuracy": 0.16477779150009156,
"num_tokens": 6726558.0,
"step": 3015
},
{
"entropy": 5.8673491954803465,
"epoch": 0.33919245240635704,
"grad_norm": 1.1328125,
"learning_rate": 0.0004994161845291828,
"loss": 5.7689,
"mean_token_accuracy": 0.15882837325334548,
"num_tokens": 6737819.0,
"step": 3020
},
{
"entropy": 5.770743465423584,
"epoch": 0.3397540293143146,
"grad_norm": 1.09375,
"learning_rate": 0.0004994132906005041,
"loss": 5.6956,
"mean_token_accuracy": 0.16847639083862304,
"num_tokens": 6748593.0,
"step": 3025
},
{
"entropy": 5.662119150161743,
"epoch": 0.34031560622227214,
"grad_norm": 1.171875,
"learning_rate": 0.0004994103895264193,
"loss": 5.808,
"mean_token_accuracy": 0.15632442981004716,
"num_tokens": 6760631.0,
"step": 3030
},
{
"entropy": 5.8305895805358885,
"epoch": 0.3408771831302297,
"grad_norm": 1.125,
"learning_rate": 0.0004994074813070207,
"loss": 5.7277,
"mean_token_accuracy": 0.16029628366231918,
"num_tokens": 6772318.0,
"step": 3035
},
{
"entropy": 5.712289094924927,
"epoch": 0.34143876003818724,
"grad_norm": 1.0546875,
"learning_rate": 0.0004994045659424007,
"loss": 5.7568,
"mean_token_accuracy": 0.16799456626176834,
"num_tokens": 6782684.0,
"step": 3040
},
{
"entropy": 5.805101203918457,
"epoch": 0.3420003369461448,
"grad_norm": 1.1171875,
"learning_rate": 0.0004994016434326524,
"loss": 5.6384,
"mean_token_accuracy": 0.17280945479869841,
"num_tokens": 6793895.0,
"step": 3045
},
{
"entropy": 5.63124098777771,
"epoch": 0.34256191385410234,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993987137778687,
"loss": 5.7098,
"mean_token_accuracy": 0.16750476360321045,
"num_tokens": 6804794.0,
"step": 3050
},
{
"entropy": 5.7920314311981205,
"epoch": 0.3431234907620599,
"grad_norm": 1.0078125,
"learning_rate": 0.000499395776978143,
"loss": 5.8365,
"mean_token_accuracy": 0.16300917863845826,
"num_tokens": 6815976.0,
"step": 3055
},
{
"entropy": 5.908558082580567,
"epoch": 0.3436850676700174,
"grad_norm": 1.15625,
"learning_rate": 0.0004993928330335686,
"loss": 5.8047,
"mean_token_accuracy": 0.15736788883805275,
"num_tokens": 6828027.0,
"step": 3060
},
{
"entropy": 5.7943390846252445,
"epoch": 0.34424664457797494,
"grad_norm": 1.15625,
"learning_rate": 0.0004993898819442395,
"loss": 5.797,
"mean_token_accuracy": 0.1607540503144264,
"num_tokens": 6839195.0,
"step": 3065
},
{
"entropy": 5.65758695602417,
"epoch": 0.3448082214859325,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993869237102494,
"loss": 5.6645,
"mean_token_accuracy": 0.1629405677318573,
"num_tokens": 6851731.0,
"step": 3070
},
{
"entropy": 5.8014930248260494,
"epoch": 0.34536979839389004,
"grad_norm": 1.2890625,
"learning_rate": 0.0004993839583316927,
"loss": 5.7082,
"mean_token_accuracy": 0.16856752187013627,
"num_tokens": 6862916.0,
"step": 3075
},
{
"entropy": 5.851268243789673,
"epoch": 0.3459313753018476,
"grad_norm": 1.125,
"learning_rate": 0.0004993809858086638,
"loss": 5.8821,
"mean_token_accuracy": 0.15647284388542176,
"num_tokens": 6876003.0,
"step": 3080
},
{
"entropy": 5.744989681243896,
"epoch": 0.34649295220980514,
"grad_norm": 1.2421875,
"learning_rate": 0.0004993780061412571,
"loss": 5.8566,
"mean_token_accuracy": 0.16301696449518205,
"num_tokens": 6886040.0,
"step": 3085
},
{
"entropy": 5.859422111511231,
"epoch": 0.3470545291177627,
"grad_norm": 1.1171875,
"learning_rate": 0.0004993750193295678,
"loss": 5.763,
"mean_token_accuracy": 0.16644185930490493,
"num_tokens": 6896758.0,
"step": 3090
},
{
"entropy": 5.861282157897949,
"epoch": 0.34761610602572024,
"grad_norm": 1.140625,
"learning_rate": 0.0004993720253736909,
"loss": 5.8231,
"mean_token_accuracy": 0.16256381422281266,
"num_tokens": 6907287.0,
"step": 3095
},
{
"entropy": 5.739157772064209,
"epoch": 0.3481776829336778,
"grad_norm": 1.265625,
"learning_rate": 0.0004993690242737216,
"loss": 5.8456,
"mean_token_accuracy": 0.1610807254910469,
"num_tokens": 6918992.0,
"step": 3100
},
{
"entropy": 5.980351161956787,
"epoch": 0.3487392598416353,
"grad_norm": 1.46875,
"learning_rate": 0.0004993660160297555,
"loss": 5.8619,
"mean_token_accuracy": 0.1558055192232132,
"num_tokens": 6931493.0,
"step": 3105
},
{
"entropy": 5.767164993286133,
"epoch": 0.34930083674959284,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993630006418885,
"loss": 5.6767,
"mean_token_accuracy": 0.16661891490221023,
"num_tokens": 6941807.0,
"step": 3110
},
{
"entropy": 5.80352258682251,
"epoch": 0.3498624136575504,
"grad_norm": 0.9921875,
"learning_rate": 0.0004993599781102166,
"loss": 5.887,
"mean_token_accuracy": 0.15885087549686433,
"num_tokens": 6953436.0,
"step": 3115
},
{
"entropy": 5.988248872756958,
"epoch": 0.35042399056550794,
"grad_norm": 1.1875,
"learning_rate": 0.0004993569484348359,
"loss": 5.9841,
"mean_token_accuracy": 0.15586867481470107,
"num_tokens": 6965247.0,
"step": 3120
},
{
"entropy": 5.805822563171387,
"epoch": 0.3509855674734655,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993539116158429,
"loss": 5.7704,
"mean_token_accuracy": 0.16849809139966965,
"num_tokens": 6976605.0,
"step": 3125
},
{
"entropy": 5.786275577545166,
"epoch": 0.35154714438142304,
"grad_norm": 1.0703125,
"learning_rate": 0.0004993508676533346,
"loss": 5.7244,
"mean_token_accuracy": 0.17590725123882295,
"num_tokens": 6986205.0,
"step": 3130
},
{
"entropy": 5.7095787525177,
"epoch": 0.3521087212893806,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993478165474074,
"loss": 5.6966,
"mean_token_accuracy": 0.16691629439592362,
"num_tokens": 6997699.0,
"step": 3135
},
{
"entropy": 5.683685970306397,
"epoch": 0.35267029819733814,
"grad_norm": 1.046875,
"learning_rate": 0.0004993447582981588,
"loss": 5.7278,
"mean_token_accuracy": 0.16395678967237473,
"num_tokens": 7008968.0,
"step": 3140
},
{
"entropy": 5.881155633926392,
"epoch": 0.3532318751052957,
"grad_norm": 1.2578125,
"learning_rate": 0.000499341692905686,
"loss": 5.7474,
"mean_token_accuracy": 0.16149352639913558,
"num_tokens": 7020633.0,
"step": 3145
},
{
"entropy": 5.731144523620605,
"epoch": 0.35379345201325324,
"grad_norm": 1.1796875,
"learning_rate": 0.0004993386203700868,
"loss": 5.665,
"mean_token_accuracy": 0.1709931015968323,
"num_tokens": 7031187.0,
"step": 3150
},
{
"entropy": 5.793157529830933,
"epoch": 0.35435502892121074,
"grad_norm": 1.15625,
"learning_rate": 0.0004993355406914589,
"loss": 5.8803,
"mean_token_accuracy": 0.16072463989257812,
"num_tokens": 7042662.0,
"step": 3155
},
{
"entropy": 5.875354909896851,
"epoch": 0.3549166058291683,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993324538699003,
"loss": 5.8742,
"mean_token_accuracy": 0.16741756796836854,
"num_tokens": 7054297.0,
"step": 3160
},
{
"entropy": 5.829679822921753,
"epoch": 0.35547818273712584,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993293599055094,
"loss": 5.7261,
"mean_token_accuracy": 0.16745392680168153,
"num_tokens": 7064961.0,
"step": 3165
},
{
"entropy": 5.842280483245849,
"epoch": 0.3560397596450834,
"grad_norm": 0.97265625,
"learning_rate": 0.0004993262587983848,
"loss": 5.8718,
"mean_token_accuracy": 0.1634344592690468,
"num_tokens": 7076857.0,
"step": 3170
},
{
"entropy": 5.800778865814209,
"epoch": 0.35660133655304094,
"grad_norm": 1.1953125,
"learning_rate": 0.000499323150548625,
"loss": 5.746,
"mean_token_accuracy": 0.16409371495246888,
"num_tokens": 7088926.0,
"step": 3175
},
{
"entropy": 5.678825235366821,
"epoch": 0.3571629134609985,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993200351563291,
"loss": 5.6641,
"mean_token_accuracy": 0.16953629851341248,
"num_tokens": 7099478.0,
"step": 3180
},
{
"entropy": 5.712728023529053,
"epoch": 0.35772449036895604,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993169126215962,
"loss": 5.7167,
"mean_token_accuracy": 0.170328189432621,
"num_tokens": 7111776.0,
"step": 3185
},
{
"entropy": 5.9294391632080075,
"epoch": 0.3582860672769136,
"grad_norm": 1.1015625,
"learning_rate": 0.0004993137829445259,
"loss": 5.7993,
"mean_token_accuracy": 0.16302864849567414,
"num_tokens": 7123084.0,
"step": 3190
},
{
"entropy": 5.741609525680542,
"epoch": 0.35884764418487114,
"grad_norm": 1.6171875,
"learning_rate": 0.0004993106461252177,
"loss": 5.7801,
"mean_token_accuracy": 0.16674443185329438,
"num_tokens": 7134973.0,
"step": 3195
},
{
"entropy": 5.749413919448853,
"epoch": 0.3594092210928287,
"grad_norm": 1.0,
"learning_rate": 0.0004993075021637716,
"loss": 5.809,
"mean_token_accuracy": 0.1610458165407181,
"num_tokens": 7147227.0,
"step": 3200
},
{
"entropy": 5.90180139541626,
"epoch": 0.3599707980007862,
"grad_norm": 1.078125,
"learning_rate": 0.0004993043510602875,
"loss": 5.7751,
"mean_token_accuracy": 0.16475535035133362,
"num_tokens": 7158201.0,
"step": 3205
},
{
"entropy": 5.679529809951783,
"epoch": 0.36053237490874374,
"grad_norm": 0.96875,
"learning_rate": 0.000499301192814866,
"loss": 5.6862,
"mean_token_accuracy": 0.17030281126499175,
"num_tokens": 7168909.0,
"step": 3210
},
{
"entropy": 5.898109769821167,
"epoch": 0.3610939518167013,
"grad_norm": 1.1484375,
"learning_rate": 0.0004992980274276074,
"loss": 5.8473,
"mean_token_accuracy": 0.15956147015094757,
"num_tokens": 7179877.0,
"step": 3215
},
{
"entropy": 5.758398485183716,
"epoch": 0.36165552872465884,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992948548986128,
"loss": 5.7488,
"mean_token_accuracy": 0.1702277645468712,
"num_tokens": 7190823.0,
"step": 3220
},
{
"entropy": 5.669489002227783,
"epoch": 0.3622171056326164,
"grad_norm": 1.15625,
"learning_rate": 0.0004992916752279829,
"loss": 5.6728,
"mean_token_accuracy": 0.16876401752233505,
"num_tokens": 7201700.0,
"step": 3225
},
{
"entropy": 5.790194272994995,
"epoch": 0.36277868254057394,
"grad_norm": 1.0625,
"learning_rate": 0.000499288488415819,
"loss": 5.7321,
"mean_token_accuracy": 0.16968879699707032,
"num_tokens": 7213618.0,
"step": 3230
},
{
"entropy": 5.915035676956177,
"epoch": 0.3633402594485315,
"grad_norm": 1.0625,
"learning_rate": 0.0004992852944622229,
"loss": 5.8987,
"mean_token_accuracy": 0.16299344301223756,
"num_tokens": 7226207.0,
"step": 3235
},
{
"entropy": 5.695443773269654,
"epoch": 0.36390183635648904,
"grad_norm": 1.109375,
"learning_rate": 0.0004992820933672958,
"loss": 5.7733,
"mean_token_accuracy": 0.16770253777503968,
"num_tokens": 7237545.0,
"step": 3240
},
{
"entropy": 5.769032526016235,
"epoch": 0.3644634132644466,
"grad_norm": 1.09375,
"learning_rate": 0.0004992788851311399,
"loss": 5.7229,
"mean_token_accuracy": 0.16769974529743195,
"num_tokens": 7247799.0,
"step": 3245
},
{
"entropy": 5.740810489654541,
"epoch": 0.3650249901724041,
"grad_norm": 1.1875,
"learning_rate": 0.0004992756697538574,
"loss": 5.7188,
"mean_token_accuracy": 0.16097148954868318,
"num_tokens": 7260229.0,
"step": 3250
},
{
"entropy": 5.803533220291138,
"epoch": 0.36558656708036164,
"grad_norm": 1.203125,
"learning_rate": 0.0004992724472355505,
"loss": 5.8319,
"mean_token_accuracy": 0.1603524476289749,
"num_tokens": 7271457.0,
"step": 3255
},
{
"entropy": 5.777963066101075,
"epoch": 0.3661481439883192,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992692175763219,
"loss": 5.724,
"mean_token_accuracy": 0.1639913350343704,
"num_tokens": 7282124.0,
"step": 3260
},
{
"entropy": 5.841952705383301,
"epoch": 0.36670972089627674,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992659807762745,
"loss": 5.8344,
"mean_token_accuracy": 0.1636727675795555,
"num_tokens": 7292981.0,
"step": 3265
},
{
"entropy": 5.9353352069854735,
"epoch": 0.3672712978042343,
"grad_norm": 1.0859375,
"learning_rate": 0.0004992627368355112,
"loss": 5.9831,
"mean_token_accuracy": 0.16069335043430327,
"num_tokens": 7304591.0,
"step": 3270
},
{
"entropy": 5.866574001312256,
"epoch": 0.36783287471219184,
"grad_norm": 1.1796875,
"learning_rate": 0.0004992594857541355,
"loss": 5.7368,
"mean_token_accuracy": 0.16766837984323502,
"num_tokens": 7315918.0,
"step": 3275
},
{
"entropy": 5.781996202468872,
"epoch": 0.3683944516201494,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992562275322506,
"loss": 5.8196,
"mean_token_accuracy": 0.16292733699083328,
"num_tokens": 7327725.0,
"step": 3280
},
{
"entropy": 5.847544145584107,
"epoch": 0.36895602852810694,
"grad_norm": 1.125,
"learning_rate": 0.0004992529621699606,
"loss": 5.8237,
"mean_token_accuracy": 0.16025864481925964,
"num_tokens": 7339787.0,
"step": 3285
},
{
"entropy": 5.834546852111816,
"epoch": 0.3695176054360645,
"grad_norm": 1.0546875,
"learning_rate": 0.0004992496896673692,
"loss": 5.8096,
"mean_token_accuracy": 0.1641978293657303,
"num_tokens": 7351284.0,
"step": 3290
},
{
"entropy": 5.6964123249053955,
"epoch": 0.37007918234402204,
"grad_norm": 1.2421875,
"learning_rate": 0.0004992464100245809,
"loss": 5.635,
"mean_token_accuracy": 0.17318471223115922,
"num_tokens": 7362105.0,
"step": 3295
},
{
"entropy": 5.717716646194458,
"epoch": 0.37064075925197953,
"grad_norm": 1.1015625,
"learning_rate": 0.0004992431232416997,
"loss": 5.7537,
"mean_token_accuracy": 0.16456210613250732,
"num_tokens": 7375204.0,
"step": 3300
},
{
"entropy": 5.950256299972534,
"epoch": 0.3712023361599371,
"grad_norm": 1.1640625,
"learning_rate": 0.0004992398293188307,
"loss": 5.8881,
"mean_token_accuracy": 0.16083506047725676,
"num_tokens": 7385853.0,
"step": 3305
},
{
"entropy": 5.766655445098877,
"epoch": 0.37176391306789464,
"grad_norm": 1.1484375,
"learning_rate": 0.0004992365282560784,
"loss": 5.7557,
"mean_token_accuracy": 0.1588754177093506,
"num_tokens": 7397459.0,
"step": 3310
},
{
"entropy": 5.714860486984253,
"epoch": 0.3723254899758522,
"grad_norm": 1.1328125,
"learning_rate": 0.0004992332200535483,
"loss": 5.7441,
"mean_token_accuracy": 0.1593509331345558,
"num_tokens": 7408996.0,
"step": 3315
},
{
"entropy": 5.871336269378662,
"epoch": 0.37288706688380974,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992299047113452,
"loss": 5.8379,
"mean_token_accuracy": 0.15264829993247986,
"num_tokens": 7419938.0,
"step": 3320
},
{
"entropy": 5.791214656829834,
"epoch": 0.3734486437917673,
"grad_norm": 1.109375,
"learning_rate": 0.0004992265822295752,
"loss": 5.8396,
"mean_token_accuracy": 0.1603675067424774,
"num_tokens": 7430181.0,
"step": 3325
},
{
"entropy": 5.841214656829834,
"epoch": 0.37401022069972484,
"grad_norm": 1.2734375,
"learning_rate": 0.0004992232526083438,
"loss": 5.7907,
"mean_token_accuracy": 0.1604009285569191,
"num_tokens": 7440774.0,
"step": 3330
},
{
"entropy": 5.834426069259644,
"epoch": 0.3745717976076824,
"grad_norm": 1.015625,
"learning_rate": 0.000499219915847757,
"loss": 5.7327,
"mean_token_accuracy": 0.16392534077167512,
"num_tokens": 7451744.0,
"step": 3335
},
{
"entropy": 5.6204368591308596,
"epoch": 0.37513337451563994,
"grad_norm": 1.078125,
"learning_rate": 0.0004992165719479211,
"loss": 5.7748,
"mean_token_accuracy": 0.1650886654853821,
"num_tokens": 7463731.0,
"step": 3340
},
{
"entropy": 5.888685989379883,
"epoch": 0.37569495142359743,
"grad_norm": 1.2265625,
"learning_rate": 0.0004992132209089427,
"loss": 5.7506,
"mean_token_accuracy": 0.1694270506501198,
"num_tokens": 7474392.0,
"step": 3345
},
{
"entropy": 5.728668689727783,
"epoch": 0.376256528331555,
"grad_norm": 1.1953125,
"learning_rate": 0.0004992098627309283,
"loss": 5.6752,
"mean_token_accuracy": 0.1741607367992401,
"num_tokens": 7484860.0,
"step": 3350
},
{
"entropy": 5.673388385772705,
"epoch": 0.37681810523951254,
"grad_norm": 1.078125,
"learning_rate": 0.000499206497413985,
"loss": 5.7317,
"mean_token_accuracy": 0.16711729168891906,
"num_tokens": 7495877.0,
"step": 3355
},
{
"entropy": 5.852774953842163,
"epoch": 0.3773796821474701,
"grad_norm": 1.234375,
"learning_rate": 0.0004992031249582197,
"loss": 5.7878,
"mean_token_accuracy": 0.16651966869831086,
"num_tokens": 7506879.0,
"step": 3360
},
{
"entropy": 5.735845470428467,
"epoch": 0.37794125905542764,
"grad_norm": 1.1640625,
"learning_rate": 0.0004991997453637401,
"loss": 5.6398,
"mean_token_accuracy": 0.17161250114440918,
"num_tokens": 7517405.0,
"step": 3365
},
{
"entropy": 5.72213454246521,
"epoch": 0.3785028359633852,
"grad_norm": 1.0546875,
"learning_rate": 0.0004991963586306536,
"loss": 5.7185,
"mean_token_accuracy": 0.16417621821165085,
"num_tokens": 7528832.0,
"step": 3370
},
{
"entropy": 5.802087879180908,
"epoch": 0.37906441287134274,
"grad_norm": 1.03125,
"learning_rate": 0.0004991929647590682,
"loss": 5.7002,
"mean_token_accuracy": 0.16515496075153352,
"num_tokens": 7541048.0,
"step": 3375
},
{
"entropy": 5.739951086044312,
"epoch": 0.3796259897793003,
"grad_norm": 1.0703125,
"learning_rate": 0.0004991895637490917,
"loss": 5.7056,
"mean_token_accuracy": 0.17090942710638046,
"num_tokens": 7551849.0,
"step": 3380
},
{
"entropy": 5.651166772842407,
"epoch": 0.38018756668725784,
"grad_norm": 1.140625,
"learning_rate": 0.0004991861556008326,
"loss": 5.7374,
"mean_token_accuracy": 0.1727711170911789,
"num_tokens": 7563148.0,
"step": 3385
},
{
"entropy": 5.890611839294434,
"epoch": 0.3807491435952154,
"grad_norm": 1.296875,
"learning_rate": 0.0004991827403143992,
"loss": 5.9224,
"mean_token_accuracy": 0.15950270295143126,
"num_tokens": 7575119.0,
"step": 3390
},
{
"entropy": 5.863496351242065,
"epoch": 0.3813107205031729,
"grad_norm": 1.1796875,
"learning_rate": 0.0004991793178899006,
"loss": 5.7577,
"mean_token_accuracy": 0.167296801507473,
"num_tokens": 7585899.0,
"step": 3395
},
{
"entropy": 5.789218902587891,
"epoch": 0.38187229741113043,
"grad_norm": 1.078125,
"learning_rate": 0.0004991758883274455,
"loss": 5.6604,
"mean_token_accuracy": 0.17499837279319763,
"num_tokens": 7598938.0,
"step": 3400
},
{
"entropy": 5.622494649887085,
"epoch": 0.382433874319088,
"grad_norm": 1.328125,
"learning_rate": 0.0004991724516271432,
"loss": 5.7392,
"mean_token_accuracy": 0.16043092608451842,
"num_tokens": 7610087.0,
"step": 3405
},
{
"entropy": 5.892052268981933,
"epoch": 0.38299545122704554,
"grad_norm": 1.2109375,
"learning_rate": 0.0004991690077891032,
"loss": 5.8658,
"mean_token_accuracy": 0.1590296171605587,
"num_tokens": 7622186.0,
"step": 3410
},
{
"entropy": 5.862188148498535,
"epoch": 0.3835570281350031,
"grad_norm": 1.1171875,
"learning_rate": 0.0004991655568134348,
"loss": 5.8112,
"mean_token_accuracy": 0.16279813647270203,
"num_tokens": 7633501.0,
"step": 3415
},
{
"entropy": 5.642923498153687,
"epoch": 0.38411860504296064,
"grad_norm": 1.1484375,
"learning_rate": 0.0004991620987002483,
"loss": 5.7307,
"mean_token_accuracy": 0.16552092581987382,
"num_tokens": 7643656.0,
"step": 3420
},
{
"entropy": 5.830251979827881,
"epoch": 0.3846801819509182,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991586334496536,
"loss": 5.7469,
"mean_token_accuracy": 0.16577651500701904,
"num_tokens": 7654565.0,
"step": 3425
},
{
"entropy": 5.7995219230651855,
"epoch": 0.38524175885887574,
"grad_norm": 1.1640625,
"learning_rate": 0.0004991551610617611,
"loss": 5.6644,
"mean_token_accuracy": 0.17500552237033845,
"num_tokens": 7664725.0,
"step": 3430
},
{
"entropy": 5.670253467559815,
"epoch": 0.3858033357668333,
"grad_norm": 1.359375,
"learning_rate": 0.0004991516815366812,
"loss": 5.7536,
"mean_token_accuracy": 0.16700845211744308,
"num_tokens": 7674772.0,
"step": 3435
},
{
"entropy": 5.849098968505859,
"epoch": 0.38636491267479084,
"grad_norm": 1.1953125,
"learning_rate": 0.000499148194874525,
"loss": 5.8251,
"mean_token_accuracy": 0.15801840871572495,
"num_tokens": 7686488.0,
"step": 3440
},
{
"entropy": 5.833809089660645,
"epoch": 0.38692648958274833,
"grad_norm": 1.1796875,
"learning_rate": 0.0004991447010754032,
"loss": 5.8096,
"mean_token_accuracy": 0.16452603340148925,
"num_tokens": 7698842.0,
"step": 3445
},
{
"entropy": 5.719552946090698,
"epoch": 0.3874880664907059,
"grad_norm": 1.109375,
"learning_rate": 0.0004991412001394273,
"loss": 5.5925,
"mean_token_accuracy": 0.16661614626646043,
"num_tokens": 7709255.0,
"step": 3450
},
{
"entropy": 5.6269690990448,
"epoch": 0.38804964339866344,
"grad_norm": 1.140625,
"learning_rate": 0.0004991376920667086,
"loss": 5.6411,
"mean_token_accuracy": 0.16555450409650802,
"num_tokens": 7720182.0,
"step": 3455
},
{
"entropy": 5.719253253936768,
"epoch": 0.388611220306621,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991341768573589,
"loss": 5.6592,
"mean_token_accuracy": 0.17378831207752227,
"num_tokens": 7730944.0,
"step": 3460
},
{
"entropy": 5.7774358749389645,
"epoch": 0.38917279721457854,
"grad_norm": 1.140625,
"learning_rate": 0.00049913065451149,
"loss": 5.7339,
"mean_token_accuracy": 0.16520332992076875,
"num_tokens": 7741686.0,
"step": 3465
},
{
"entropy": 5.7580331325531,
"epoch": 0.3897343741225361,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991271250292141,
"loss": 5.7289,
"mean_token_accuracy": 0.16803366243839263,
"num_tokens": 7752245.0,
"step": 3470
},
{
"entropy": 5.634769535064697,
"epoch": 0.39029595103049364,
"grad_norm": 1.1484375,
"learning_rate": 0.0004991235884106437,
"loss": 5.7108,
"mean_token_accuracy": 0.16429706960916518,
"num_tokens": 7762622.0,
"step": 3475
},
{
"entropy": 5.904698276519776,
"epoch": 0.3908575279384512,
"grad_norm": 1.3046875,
"learning_rate": 0.0004991200446558913,
"loss": 5.8229,
"mean_token_accuracy": 0.16702470034360886,
"num_tokens": 7773550.0,
"step": 3480
},
{
"entropy": 5.779788017272949,
"epoch": 0.39141910484640874,
"grad_norm": 1.046875,
"learning_rate": 0.0004991164937650698,
"loss": 5.7287,
"mean_token_accuracy": 0.16326591968536378,
"num_tokens": 7784988.0,
"step": 3485
},
{
"entropy": 5.82987608909607,
"epoch": 0.39198068175436623,
"grad_norm": 1.0625,
"learning_rate": 0.0004991129357382922,
"loss": 5.815,
"mean_token_accuracy": 0.1673831447958946,
"num_tokens": 7797706.0,
"step": 3490
},
{
"entropy": 5.747679376602173,
"epoch": 0.3925422586623238,
"grad_norm": 1.1015625,
"learning_rate": 0.0004991093705756718,
"loss": 5.6905,
"mean_token_accuracy": 0.17106844186782838,
"num_tokens": 7808960.0,
"step": 3495
},
{
"entropy": 5.676139545440674,
"epoch": 0.39310383557028133,
"grad_norm": 1.078125,
"learning_rate": 0.0004991057982773221,
"loss": 5.6591,
"mean_token_accuracy": 0.17813282757997512,
"num_tokens": 7819271.0,
"step": 3500
},
{
"entropy": 5.755858039855957,
"epoch": 0.3936654124782389,
"grad_norm": 1.015625,
"learning_rate": 0.0004991022188433569,
"loss": 5.7929,
"mean_token_accuracy": 0.16215733736753463,
"num_tokens": 7832257.0,
"step": 3505
},
{
"entropy": 5.870551013946534,
"epoch": 0.39422698938619644,
"grad_norm": 1.171875,
"learning_rate": 0.00049909863227389,
"loss": 5.81,
"mean_token_accuracy": 0.16174299120903016,
"num_tokens": 7842684.0,
"step": 3510
},
{
"entropy": 5.7586606502532955,
"epoch": 0.394788566294154,
"grad_norm": 1.015625,
"learning_rate": 0.0004990950385690358,
"loss": 5.8144,
"mean_token_accuracy": 0.1619935393333435,
"num_tokens": 7855031.0,
"step": 3515
},
{
"entropy": 5.68963360786438,
"epoch": 0.39535014320211154,
"grad_norm": 1.0625,
"learning_rate": 0.0004990914377289088,
"loss": 5.6028,
"mean_token_accuracy": 0.16759985089302062,
"num_tokens": 7865291.0,
"step": 3520
},
{
"entropy": 5.646744012832642,
"epoch": 0.3959117201100691,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990878297536233,
"loss": 5.5664,
"mean_token_accuracy": 0.16892427802085877,
"num_tokens": 7876401.0,
"step": 3525
},
{
"entropy": 5.67297773361206,
"epoch": 0.39647329701802664,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990842146432946,
"loss": 5.6256,
"mean_token_accuracy": 0.17110522240400314,
"num_tokens": 7887787.0,
"step": 3530
},
{
"entropy": 5.712070369720459,
"epoch": 0.3970348739259842,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990805923980375,
"loss": 5.6323,
"mean_token_accuracy": 0.1659396395087242,
"num_tokens": 7898550.0,
"step": 3535
},
{
"entropy": 5.665684556961059,
"epoch": 0.3975964508339417,
"grad_norm": 1.1640625,
"learning_rate": 0.0004990769630179673,
"loss": 5.7676,
"mean_token_accuracy": 0.16291757076978683,
"num_tokens": 7909254.0,
"step": 3540
},
{
"entropy": 5.893109703063965,
"epoch": 0.39815802774189923,
"grad_norm": 1.4140625,
"learning_rate": 0.0004990733265032,
"loss": 5.7973,
"mean_token_accuracy": 0.1682746008038521,
"num_tokens": 7920164.0,
"step": 3545
},
{
"entropy": 5.681936979293823,
"epoch": 0.3987196046498568,
"grad_norm": 1.2109375,
"learning_rate": 0.0004990696828538508,
"loss": 5.7464,
"mean_token_accuracy": 0.1638046160340309,
"num_tokens": 7931390.0,
"step": 3550
},
{
"entropy": 5.7912890911102295,
"epoch": 0.39928118155781434,
"grad_norm": 1.171875,
"learning_rate": 0.000499066032070036,
"loss": 5.7501,
"mean_token_accuracy": 0.16561875641345977,
"num_tokens": 7943765.0,
"step": 3555
},
{
"entropy": 5.696421575546265,
"epoch": 0.3998427584657719,
"grad_norm": 1.1484375,
"learning_rate": 0.0004990623741518719,
"loss": 5.5769,
"mean_token_accuracy": 0.17134051471948625,
"num_tokens": 7954621.0,
"step": 3560
},
{
"entropy": 5.629381084442139,
"epoch": 0.40040433537372944,
"grad_norm": 1.109375,
"learning_rate": 0.000499058709099475,
"loss": 5.7987,
"mean_token_accuracy": 0.16370998024940492,
"num_tokens": 7966063.0,
"step": 3565
},
{
"entropy": 5.861378383636475,
"epoch": 0.400965912281687,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990550369129618,
"loss": 5.7126,
"mean_token_accuracy": 0.17164952754974366,
"num_tokens": 7977391.0,
"step": 3570
},
{
"entropy": 5.798222494125366,
"epoch": 0.40152748918964454,
"grad_norm": 1.125,
"learning_rate": 0.0004990513575924494,
"loss": 5.8279,
"mean_token_accuracy": 0.16133795231580733,
"num_tokens": 7988786.0,
"step": 3575
},
{
"entropy": 5.739356803894043,
"epoch": 0.4020890660976021,
"grad_norm": 1.0,
"learning_rate": 0.0004990476711380548,
"loss": 5.6304,
"mean_token_accuracy": 0.1664183497428894,
"num_tokens": 8000848.0,
"step": 3580
},
{
"entropy": 5.730372095108033,
"epoch": 0.40265064300555964,
"grad_norm": 1.0625,
"learning_rate": 0.0004990439775498954,
"loss": 5.7903,
"mean_token_accuracy": 0.1628401517868042,
"num_tokens": 8012621.0,
"step": 3585
},
{
"entropy": 5.897176885604859,
"epoch": 0.40321221991351713,
"grad_norm": 1.09375,
"learning_rate": 0.000499040276828089,
"loss": 5.8352,
"mean_token_accuracy": 0.1639959394931793,
"num_tokens": 8023599.0,
"step": 3590
},
{
"entropy": 5.598546123504638,
"epoch": 0.4037737968214747,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990365689727533,
"loss": 5.6539,
"mean_token_accuracy": 0.17202971875667572,
"num_tokens": 8037151.0,
"step": 3595
},
{
"entropy": 5.7562020778656,
"epoch": 0.40433537372943223,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990328539840062,
"loss": 5.8423,
"mean_token_accuracy": 0.16440727114677428,
"num_tokens": 8048606.0,
"step": 3600
},
{
"entropy": 5.7380390644073485,
"epoch": 0.4048969506373898,
"grad_norm": 1.265625,
"learning_rate": 0.0004990291318619663,
"loss": 5.6297,
"mean_token_accuracy": 0.17538885176181793,
"num_tokens": 8058430.0,
"step": 3605
},
{
"entropy": 5.634087419509887,
"epoch": 0.40545852754534734,
"grad_norm": 1.1171875,
"learning_rate": 0.0004990254026067518,
"loss": 5.6338,
"mean_token_accuracy": 0.1682212010025978,
"num_tokens": 8069554.0,
"step": 3610
},
{
"entropy": 5.83514289855957,
"epoch": 0.4060201044533049,
"grad_norm": 1.0546875,
"learning_rate": 0.0004990216662184817,
"loss": 5.7918,
"mean_token_accuracy": 0.15937150940299033,
"num_tokens": 8082131.0,
"step": 3615
},
{
"entropy": 5.669177007675171,
"epoch": 0.40658168136126244,
"grad_norm": 1.0859375,
"learning_rate": 0.0004990179226972749,
"loss": 5.6369,
"mean_token_accuracy": 0.17017019540071487,
"num_tokens": 8093454.0,
"step": 3620
},
{
"entropy": 5.850167655944825,
"epoch": 0.40714325826922,
"grad_norm": 1.1953125,
"learning_rate": 0.0004990141720432504,
"loss": 5.8959,
"mean_token_accuracy": 0.1559806853532791,
"num_tokens": 8105502.0,
"step": 3625
},
{
"entropy": 5.780977392196656,
"epoch": 0.40770483517717754,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990104142565279,
"loss": 5.7574,
"mean_token_accuracy": 0.15795129239559175,
"num_tokens": 8116524.0,
"step": 3630
},
{
"entropy": 5.783453130722046,
"epoch": 0.40826641208513503,
"grad_norm": 1.1328125,
"learning_rate": 0.000499006649337227,
"loss": 5.7482,
"mean_token_accuracy": 0.17147931903600694,
"num_tokens": 8126949.0,
"step": 3635
},
{
"entropy": 5.756874036788941,
"epoch": 0.4088279889930926,
"grad_norm": 1.171875,
"learning_rate": 0.0004990028772854674,
"loss": 5.6934,
"mean_token_accuracy": 0.17261993736028672,
"num_tokens": 8137346.0,
"step": 3640
},
{
"entropy": 5.6672529697418215,
"epoch": 0.40938956590105013,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989990981013693,
"loss": 5.7154,
"mean_token_accuracy": 0.161580191552639,
"num_tokens": 8148499.0,
"step": 3645
},
{
"entropy": 5.815642881393432,
"epoch": 0.4099511428090077,
"grad_norm": 1.125,
"learning_rate": 0.000498995311785053,
"loss": 5.7648,
"mean_token_accuracy": 0.16477382630109788,
"num_tokens": 8159645.0,
"step": 3650
},
{
"entropy": 5.709488248825073,
"epoch": 0.41051271971696524,
"grad_norm": 1.0703125,
"learning_rate": 0.0004989915183366392,
"loss": 5.6384,
"mean_token_accuracy": 0.17073693871498108,
"num_tokens": 8171033.0,
"step": 3655
},
{
"entropy": 5.665631628036499,
"epoch": 0.4110742966249228,
"grad_norm": 1.1796875,
"learning_rate": 0.0004989877177562485,
"loss": 5.718,
"mean_token_accuracy": 0.1607867658138275,
"num_tokens": 8181582.0,
"step": 3660
},
{
"entropy": 5.709115076065063,
"epoch": 0.41163587353288034,
"grad_norm": 1.1640625,
"learning_rate": 0.000498983910044002,
"loss": 5.6057,
"mean_token_accuracy": 0.1714171588420868,
"num_tokens": 8192432.0,
"step": 3665
},
{
"entropy": 5.808908319473266,
"epoch": 0.4121974504408379,
"grad_norm": 1.0625,
"learning_rate": 0.0004989800952000209,
"loss": 5.7997,
"mean_token_accuracy": 0.16239149272441863,
"num_tokens": 8203995.0,
"step": 3670
},
{
"entropy": 5.714737892150879,
"epoch": 0.41275902734879544,
"grad_norm": 1.234375,
"learning_rate": 0.0004989762732244268,
"loss": 5.6344,
"mean_token_accuracy": 0.17067258954048156,
"num_tokens": 8214323.0,
"step": 3675
},
{
"entropy": 5.671315240859985,
"epoch": 0.413320604256753,
"grad_norm": 1.203125,
"learning_rate": 0.0004989724441173413,
"loss": 5.6992,
"mean_token_accuracy": 0.16343058943748473,
"num_tokens": 8226285.0,
"step": 3680
},
{
"entropy": 5.85388331413269,
"epoch": 0.4138821811647105,
"grad_norm": 1.1015625,
"learning_rate": 0.0004989686078788862,
"loss": 5.8613,
"mean_token_accuracy": 0.1590399593114853,
"num_tokens": 8237961.0,
"step": 3685
},
{
"entropy": 5.66704831123352,
"epoch": 0.41444375807266803,
"grad_norm": 1.078125,
"learning_rate": 0.0004989647645091839,
"loss": 5.6069,
"mean_token_accuracy": 0.17645668387413024,
"num_tokens": 8248463.0,
"step": 3690
},
{
"entropy": 5.705190849304199,
"epoch": 0.4150053349806256,
"grad_norm": 1.171875,
"learning_rate": 0.0004989609140083565,
"loss": 5.7531,
"mean_token_accuracy": 0.1683824598789215,
"num_tokens": 8260342.0,
"step": 3695
},
{
"entropy": 5.820456171035767,
"epoch": 0.41556691188858313,
"grad_norm": 1.078125,
"learning_rate": 0.0004989570563765269,
"loss": 5.8667,
"mean_token_accuracy": 0.16553982496261596,
"num_tokens": 8272055.0,
"step": 3700
},
{
"entropy": 5.9223099708557125,
"epoch": 0.4161284887965407,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989531916138176,
"loss": 5.7958,
"mean_token_accuracy": 0.17000854909420013,
"num_tokens": 8282129.0,
"step": 3705
},
{
"entropy": 5.816414403915405,
"epoch": 0.41669006570449824,
"grad_norm": 1.1953125,
"learning_rate": 0.0004989493197203519,
"loss": 5.8083,
"mean_token_accuracy": 0.16709389835596083,
"num_tokens": 8294663.0,
"step": 3710
},
{
"entropy": 5.690795183181763,
"epoch": 0.4172516426124558,
"grad_norm": 1.0546875,
"learning_rate": 0.000498945440696253,
"loss": 5.6211,
"mean_token_accuracy": 0.17238393127918245,
"num_tokens": 8305882.0,
"step": 3715
},
{
"entropy": 5.682877445220948,
"epoch": 0.41781321952041334,
"grad_norm": 1.109375,
"learning_rate": 0.0004989415545416444,
"loss": 5.7206,
"mean_token_accuracy": 0.1669295072555542,
"num_tokens": 8316604.0,
"step": 3720
},
{
"entropy": 5.841225004196167,
"epoch": 0.4183747964283709,
"grad_norm": 1.078125,
"learning_rate": 0.0004989376612566498,
"loss": 5.8059,
"mean_token_accuracy": 0.16633398383855819,
"num_tokens": 8328126.0,
"step": 3725
},
{
"entropy": 5.700233364105225,
"epoch": 0.4189363733363284,
"grad_norm": 1.1328125,
"learning_rate": 0.0004989337608413933,
"loss": 5.7237,
"mean_token_accuracy": 0.16907681822776793,
"num_tokens": 8339050.0,
"step": 3730
},
{
"entropy": 5.638241767883301,
"epoch": 0.41949795024428593,
"grad_norm": 1.109375,
"learning_rate": 0.0004989298532959989,
"loss": 5.5633,
"mean_token_accuracy": 0.17306438386440276,
"num_tokens": 8350216.0,
"step": 3735
},
{
"entropy": 5.744198417663574,
"epoch": 0.4200595271522435,
"grad_norm": 1.0234375,
"learning_rate": 0.0004989259386205912,
"loss": 5.724,
"mean_token_accuracy": 0.1716397538781166,
"num_tokens": 8362311.0,
"step": 3740
},
{
"entropy": 5.771803522109986,
"epoch": 0.42062110406020103,
"grad_norm": 1.1953125,
"learning_rate": 0.0004989220168152948,
"loss": 5.8952,
"mean_token_accuracy": 0.1681353360414505,
"num_tokens": 8373751.0,
"step": 3745
},
{
"entropy": 5.806397485733032,
"epoch": 0.4211826809681586,
"grad_norm": 1.15625,
"learning_rate": 0.0004989180878802346,
"loss": 5.7191,
"mean_token_accuracy": 0.16910514533519744,
"num_tokens": 8384932.0,
"step": 3750
},
{
"entropy": 5.7420250415802006,
"epoch": 0.42174425787611614,
"grad_norm": 1.1640625,
"learning_rate": 0.0004989141518155354,
"loss": 5.6719,
"mean_token_accuracy": 0.17444310039281846,
"num_tokens": 8394973.0,
"step": 3755
},
{
"entropy": 5.744848775863647,
"epoch": 0.4223058347840737,
"grad_norm": 1.046875,
"learning_rate": 0.0004989102086213229,
"loss": 5.7605,
"mean_token_accuracy": 0.16957378536462783,
"num_tokens": 8406590.0,
"step": 3760
},
{
"entropy": 5.737107515335083,
"epoch": 0.42286741169203124,
"grad_norm": 1.2109375,
"learning_rate": 0.0004989062582977225,
"loss": 5.6729,
"mean_token_accuracy": 0.16458474695682526,
"num_tokens": 8417127.0,
"step": 3765
},
{
"entropy": 5.799488592147827,
"epoch": 0.4234289885999888,
"grad_norm": 1.1875,
"learning_rate": 0.0004989023008448599,
"loss": 5.7788,
"mean_token_accuracy": 0.16813544780015946,
"num_tokens": 8429442.0,
"step": 3770
},
{
"entropy": 5.722933292388916,
"epoch": 0.42399056550794634,
"grad_norm": 1.125,
"learning_rate": 0.0004988983362628614,
"loss": 5.6955,
"mean_token_accuracy": 0.1680734187364578,
"num_tokens": 8439375.0,
"step": 3775
},
{
"entropy": 5.683926296234131,
"epoch": 0.42455214241590383,
"grad_norm": 1.0078125,
"learning_rate": 0.0004988943645518528,
"loss": 5.7076,
"mean_token_accuracy": 0.17047268003225327,
"num_tokens": 8449817.0,
"step": 3780
},
{
"entropy": 5.699952173233032,
"epoch": 0.4251137193238614,
"grad_norm": 1.1953125,
"learning_rate": 0.0004988903857119611,
"loss": 5.5914,
"mean_token_accuracy": 0.17530252039432526,
"num_tokens": 8460096.0,
"step": 3785
},
{
"entropy": 5.709064769744873,
"epoch": 0.42567529623181893,
"grad_norm": 1.296875,
"learning_rate": 0.0004988863997433125,
"loss": 5.7134,
"mean_token_accuracy": 0.16697499454021453,
"num_tokens": 8470241.0,
"step": 3790
},
{
"entropy": 5.762786388397217,
"epoch": 0.4262368731397765,
"grad_norm": 1.1328125,
"learning_rate": 0.0004988824066460341,
"loss": 5.7268,
"mean_token_accuracy": 0.16301118284463884,
"num_tokens": 8481438.0,
"step": 3795
},
{
"entropy": 5.769738817214966,
"epoch": 0.42679845004773403,
"grad_norm": 1.1484375,
"learning_rate": 0.000498878406420253,
"loss": 5.7738,
"mean_token_accuracy": 0.165237495303154,
"num_tokens": 8492588.0,
"step": 3800
},
{
"entropy": 5.7274257183074955,
"epoch": 0.4273600269556916,
"grad_norm": 1.140625,
"learning_rate": 0.0004988743990660967,
"loss": 5.7196,
"mean_token_accuracy": 0.17144177854061127,
"num_tokens": 8502298.0,
"step": 3805
},
{
"entropy": 5.6736900806427,
"epoch": 0.42792160386364914,
"grad_norm": 1.109375,
"learning_rate": 0.0004988703845836927,
"loss": 5.5721,
"mean_token_accuracy": 0.16779429465532303,
"num_tokens": 8513507.0,
"step": 3810
},
{
"entropy": 5.72471432685852,
"epoch": 0.4284831807716067,
"grad_norm": 1.265625,
"learning_rate": 0.0004988663629731687,
"loss": 5.6573,
"mean_token_accuracy": 0.17619085162878037,
"num_tokens": 8523888.0,
"step": 3815
},
{
"entropy": 5.666543865203858,
"epoch": 0.42904475767956424,
"grad_norm": 1.234375,
"learning_rate": 0.000498862334234653,
"loss": 5.613,
"mean_token_accuracy": 0.17636462599039077,
"num_tokens": 8534032.0,
"step": 3820
},
{
"entropy": 5.66659688949585,
"epoch": 0.4296063345875218,
"grad_norm": 1.2890625,
"learning_rate": 0.0004988582983682737,
"loss": 5.7521,
"mean_token_accuracy": 0.16864021867513657,
"num_tokens": 8545544.0,
"step": 3825
},
{
"entropy": 5.8163188934326175,
"epoch": 0.4301679114954793,
"grad_norm": 1.1796875,
"learning_rate": 0.0004988542553741594,
"loss": 5.7189,
"mean_token_accuracy": 0.17160673141479493,
"num_tokens": 8555428.0,
"step": 3830
},
{
"entropy": 5.6663635730743405,
"epoch": 0.43072948840343683,
"grad_norm": 1.25,
"learning_rate": 0.0004988502052524388,
"loss": 5.6103,
"mean_token_accuracy": 0.16832801029086114,
"num_tokens": 8566456.0,
"step": 3835
},
{
"entropy": 5.660061836242676,
"epoch": 0.4312910653113944,
"grad_norm": 1.078125,
"learning_rate": 0.0004988461480032408,
"loss": 5.6227,
"mean_token_accuracy": 0.17229728102684022,
"num_tokens": 8578125.0,
"step": 3840
},
{
"entropy": 5.728463125228882,
"epoch": 0.43185264221935193,
"grad_norm": 1.140625,
"learning_rate": 0.0004988420836266946,
"loss": 5.6879,
"mean_token_accuracy": 0.16244398951530456,
"num_tokens": 8589530.0,
"step": 3845
},
{
"entropy": 5.783083534240722,
"epoch": 0.4324142191273095,
"grad_norm": 1.28125,
"learning_rate": 0.0004988380121229298,
"loss": 5.7869,
"mean_token_accuracy": 0.166258242726326,
"num_tokens": 8601655.0,
"step": 3850
},
{
"entropy": 5.720654153823853,
"epoch": 0.43297579603526704,
"grad_norm": 1.140625,
"learning_rate": 0.0004988339334920757,
"loss": 5.7053,
"mean_token_accuracy": 0.17133706212043762,
"num_tokens": 8613355.0,
"step": 3855
},
{
"entropy": 5.684089469909668,
"epoch": 0.4335373729432246,
"grad_norm": 1.125,
"learning_rate": 0.0004988298477342625,
"loss": 5.6315,
"mean_token_accuracy": 0.17381691038608552,
"num_tokens": 8624154.0,
"step": 3860
},
{
"entropy": 5.719579362869263,
"epoch": 0.43409894985118214,
"grad_norm": 1.09375,
"learning_rate": 0.00049882575484962,
"loss": 5.7142,
"mean_token_accuracy": 0.1678782135248184,
"num_tokens": 8633888.0,
"step": 3865
},
{
"entropy": 5.735606288909912,
"epoch": 0.4346605267591397,
"grad_norm": 1.1796875,
"learning_rate": 0.0004988216548382787,
"loss": 5.6956,
"mean_token_accuracy": 0.167276069521904,
"num_tokens": 8644965.0,
"step": 3870
},
{
"entropy": 5.715915155410767,
"epoch": 0.4352221036670972,
"grad_norm": 1.2265625,
"learning_rate": 0.0004988175477003691,
"loss": 5.6381,
"mean_token_accuracy": 0.17569438219070435,
"num_tokens": 8655483.0,
"step": 3875
},
{
"entropy": 5.719552898406983,
"epoch": 0.43578368057505473,
"grad_norm": 1.2578125,
"learning_rate": 0.000498813433436022,
"loss": 5.6194,
"mean_token_accuracy": 0.1684095546603203,
"num_tokens": 8666443.0,
"step": 3880
},
{
"entropy": 5.779576778411865,
"epoch": 0.4363452574830123,
"grad_norm": 1.265625,
"learning_rate": 0.0004988093120453684,
"loss": 5.7009,
"mean_token_accuracy": 0.16824239492416382,
"num_tokens": 8677198.0,
"step": 3885
},
{
"entropy": 5.698169851303101,
"epoch": 0.43690683439096983,
"grad_norm": 1.2421875,
"learning_rate": 0.0004988051835285394,
"loss": 5.7413,
"mean_token_accuracy": 0.16558287888765336,
"num_tokens": 8688303.0,
"step": 3890
},
{
"entropy": 5.798931407928467,
"epoch": 0.4374684112989274,
"grad_norm": 1.4296875,
"learning_rate": 0.0004988010478856667,
"loss": 5.6982,
"mean_token_accuracy": 0.16919583082199097,
"num_tokens": 8700326.0,
"step": 3895
},
{
"entropy": 5.651122379302978,
"epoch": 0.43802998820688493,
"grad_norm": 1.1328125,
"learning_rate": 0.0004987969051168816,
"loss": 5.6532,
"mean_token_accuracy": 0.17064241766929628,
"num_tokens": 8711101.0,
"step": 3900
},
{
"entropy": 5.7225219249725345,
"epoch": 0.4385915651148425,
"grad_norm": 1.1640625,
"learning_rate": 0.0004987927552223165,
"loss": 5.7874,
"mean_token_accuracy": 0.16373796612024308,
"num_tokens": 8721751.0,
"step": 3905
},
{
"entropy": 5.846100473403931,
"epoch": 0.43915314202280004,
"grad_norm": 1.3203125,
"learning_rate": 0.0004987885982021032,
"loss": 5.8829,
"mean_token_accuracy": 0.15994126051664354,
"num_tokens": 8733550.0,
"step": 3910
},
{
"entropy": 5.780022525787354,
"epoch": 0.4397147189307576,
"grad_norm": 1.2578125,
"learning_rate": 0.0004987844340563741,
"loss": 5.6583,
"mean_token_accuracy": 0.16807118505239488,
"num_tokens": 8744218.0,
"step": 3915
},
{
"entropy": 5.667231273651123,
"epoch": 0.44027629583871514,
"grad_norm": 1.0078125,
"learning_rate": 0.0004987802627852619,
"loss": 5.7045,
"mean_token_accuracy": 0.16555256098508836,
"num_tokens": 8756531.0,
"step": 3920
},
{
"entropy": 5.867982053756714,
"epoch": 0.44083787274667263,
"grad_norm": 1.2265625,
"learning_rate": 0.0004987760843888992,
"loss": 5.7599,
"mean_token_accuracy": 0.17022316455841063,
"num_tokens": 8768114.0,
"step": 3925
},
{
"entropy": 5.72443118095398,
"epoch": 0.4413994496546302,
"grad_norm": 1.21875,
"learning_rate": 0.0004987718988674193,
"loss": 5.7577,
"mean_token_accuracy": 0.1657242715358734,
"num_tokens": 8780010.0,
"step": 3930
},
{
"entropy": 5.842220401763916,
"epoch": 0.44196102656258773,
"grad_norm": 1.1015625,
"learning_rate": 0.0004987677062209552,
"loss": 5.7089,
"mean_token_accuracy": 0.16331320852041245,
"num_tokens": 8792183.0,
"step": 3935
},
{
"entropy": 5.766517972946167,
"epoch": 0.4425226034705453,
"grad_norm": 1.125,
"learning_rate": 0.0004987635064496407,
"loss": 5.8122,
"mean_token_accuracy": 0.1598869666457176,
"num_tokens": 8804513.0,
"step": 3940
},
{
"entropy": 5.6813007354736325,
"epoch": 0.44308418037850283,
"grad_norm": 1.2265625,
"learning_rate": 0.0004987592995536093,
"loss": 5.701,
"mean_token_accuracy": 0.17081207036972046,
"num_tokens": 8814918.0,
"step": 3945
},
{
"entropy": 5.725718259811401,
"epoch": 0.4436457572864604,
"grad_norm": 1.3984375,
"learning_rate": 0.0004987550855329951,
"loss": 5.6457,
"mean_token_accuracy": 0.16755146384239197,
"num_tokens": 8826176.0,
"step": 3950
},
{
"entropy": 5.67350959777832,
"epoch": 0.44420733419441794,
"grad_norm": 1.109375,
"learning_rate": 0.0004987508643879322,
"loss": 5.661,
"mean_token_accuracy": 0.16792258769273757,
"num_tokens": 8837242.0,
"step": 3955
},
{
"entropy": 5.787046289443969,
"epoch": 0.4447689111023755,
"grad_norm": 1.140625,
"learning_rate": 0.0004987466361185549,
"loss": 5.6269,
"mean_token_accuracy": 0.1736079826951027,
"num_tokens": 8848972.0,
"step": 3960
},
{
"entropy": 5.5895074844360355,
"epoch": 0.44533048801033304,
"grad_norm": 1.15625,
"learning_rate": 0.0004987424007249979,
"loss": 5.6224,
"mean_token_accuracy": 0.16402306407690048,
"num_tokens": 8861108.0,
"step": 3965
},
{
"entropy": 5.762013387680054,
"epoch": 0.4458920649182906,
"grad_norm": 1.265625,
"learning_rate": 0.0004987381582073962,
"loss": 5.7189,
"mean_token_accuracy": 0.16729851216077804,
"num_tokens": 8872161.0,
"step": 3970
},
{
"entropy": 5.941451597213745,
"epoch": 0.4464536418262481,
"grad_norm": 1.25,
"learning_rate": 0.0004987339085658848,
"loss": 5.8068,
"mean_token_accuracy": 0.1685393825173378,
"num_tokens": 8882497.0,
"step": 3975
},
{
"entropy": 5.603485059738159,
"epoch": 0.44701521873420563,
"grad_norm": 1.21875,
"learning_rate": 0.0004987296518005988,
"loss": 5.5183,
"mean_token_accuracy": 0.17434909641742707,
"num_tokens": 8893155.0,
"step": 3980
},
{
"entropy": 5.588304758071899,
"epoch": 0.4475767956421632,
"grad_norm": 1.1171875,
"learning_rate": 0.000498725387911674,
"loss": 5.6065,
"mean_token_accuracy": 0.1754014179110527,
"num_tokens": 8904478.0,
"step": 3985
},
{
"entropy": 5.640865135192871,
"epoch": 0.44813837255012073,
"grad_norm": 1.234375,
"learning_rate": 0.0004987211168992461,
"loss": 5.7279,
"mean_token_accuracy": 0.16352857649326324,
"num_tokens": 8915470.0,
"step": 3990
},
{
"entropy": 5.807948732376099,
"epoch": 0.4486999494580783,
"grad_norm": 1.1796875,
"learning_rate": 0.000498716838763451,
"loss": 5.6544,
"mean_token_accuracy": 0.17243108600378038,
"num_tokens": 8925644.0,
"step": 3995
},
{
"entropy": 5.715225267410278,
"epoch": 0.44926152636603583,
"grad_norm": 1.09375,
"learning_rate": 0.000498712553504425,
"loss": 5.6888,
"mean_token_accuracy": 0.16561631858348846,
"num_tokens": 8937000.0,
"step": 4000
},
{
"entropy": 5.715015840530396,
"epoch": 0.4498231032739934,
"grad_norm": 1.1640625,
"learning_rate": 0.0004987082611223046,
"loss": 5.6828,
"mean_token_accuracy": 0.16424830853939057,
"num_tokens": 8948258.0,
"step": 4005
},
{
"entropy": 5.755721092224121,
"epoch": 0.45038468018195094,
"grad_norm": 1.1875,
"learning_rate": 0.0004987039616172265,
"loss": 5.6881,
"mean_token_accuracy": 0.16630871891975402,
"num_tokens": 8960107.0,
"step": 4010
},
{
"entropy": 5.734629678726196,
"epoch": 0.4509462570899085,
"grad_norm": 1.171875,
"learning_rate": 0.0004986996549893273,
"loss": 5.6746,
"mean_token_accuracy": 0.17057377398014067,
"num_tokens": 8969612.0,
"step": 4015
},
{
"entropy": 5.750925445556641,
"epoch": 0.451507833997866,
"grad_norm": 1.0703125,
"learning_rate": 0.0004986953412387445,
"loss": 5.7889,
"mean_token_accuracy": 0.16431803852319718,
"num_tokens": 8980970.0,
"step": 4020
},
{
"entropy": 5.774376440048218,
"epoch": 0.45206941090582353,
"grad_norm": 1.125,
"learning_rate": 0.0004986910203656151,
"loss": 5.7123,
"mean_token_accuracy": 0.1699560835957527,
"num_tokens": 8992218.0,
"step": 4025
},
{
"entropy": 5.628290319442749,
"epoch": 0.4526309878137811,
"grad_norm": 1.1328125,
"learning_rate": 0.0004986866923700769,
"loss": 5.6482,
"mean_token_accuracy": 0.17175580114126204,
"num_tokens": 9003737.0,
"step": 4030
},
{
"entropy": 5.740095758438111,
"epoch": 0.45319256472173863,
"grad_norm": 1.1328125,
"learning_rate": 0.0004986823572522677,
"loss": 5.7098,
"mean_token_accuracy": 0.17027049511671066,
"num_tokens": 9015427.0,
"step": 4035
},
{
"entropy": 5.744700384140015,
"epoch": 0.4537541416296962,
"grad_norm": 1.25,
"learning_rate": 0.0004986780150123254,
"loss": 5.7255,
"mean_token_accuracy": 0.1616039626300335,
"num_tokens": 9025003.0,
"step": 4040
},
{
"entropy": 5.705526113510132,
"epoch": 0.45431571853765373,
"grad_norm": 1.1015625,
"learning_rate": 0.0004986736656503885,
"loss": 5.6698,
"mean_token_accuracy": 0.1695774719119072,
"num_tokens": 9036074.0,
"step": 4045
},
{
"entropy": 5.830773735046387,
"epoch": 0.4548772954456113,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986693091665952,
"loss": 5.6789,
"mean_token_accuracy": 0.16831374317407607,
"num_tokens": 9046878.0,
"step": 4050
},
{
"entropy": 5.785623168945312,
"epoch": 0.45543887235356884,
"grad_norm": 1.078125,
"learning_rate": 0.0004986649455610845,
"loss": 5.7498,
"mean_token_accuracy": 0.1699572652578354,
"num_tokens": 9058366.0,
"step": 4055
},
{
"entropy": 5.7290223121643065,
"epoch": 0.4560004492615264,
"grad_norm": 1.15625,
"learning_rate": 0.000498660574833995,
"loss": 5.7595,
"mean_token_accuracy": 0.16795325577259063,
"num_tokens": 9069542.0,
"step": 4060
},
{
"entropy": 5.81335220336914,
"epoch": 0.45656202616948394,
"grad_norm": 1.046875,
"learning_rate": 0.0004986561969854661,
"loss": 5.9339,
"mean_token_accuracy": 0.15881104171276092,
"num_tokens": 9081219.0,
"step": 4065
},
{
"entropy": 5.803720617294312,
"epoch": 0.45712360307744143,
"grad_norm": 1.1484375,
"learning_rate": 0.0004986518120156373,
"loss": 5.6226,
"mean_token_accuracy": 0.17852383106946945,
"num_tokens": 9092101.0,
"step": 4070
},
{
"entropy": 5.717171287536621,
"epoch": 0.457685179985399,
"grad_norm": 1.2265625,
"learning_rate": 0.000498647419924648,
"loss": 5.6764,
"mean_token_accuracy": 0.1673010155558586,
"num_tokens": 9101900.0,
"step": 4075
},
{
"entropy": 5.778856706619263,
"epoch": 0.45824675689335653,
"grad_norm": 1.1796875,
"learning_rate": 0.000498643020712638,
"loss": 5.8139,
"mean_token_accuracy": 0.16337325274944306,
"num_tokens": 9114369.0,
"step": 4080
},
{
"entropy": 5.762792730331421,
"epoch": 0.4588083338013141,
"grad_norm": 1.2421875,
"learning_rate": 0.0004986386143797475,
"loss": 5.615,
"mean_token_accuracy": 0.17072657644748687,
"num_tokens": 9126548.0,
"step": 4085
},
{
"entropy": 5.7627839088439945,
"epoch": 0.45936991070927163,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986342009261169,
"loss": 5.8107,
"mean_token_accuracy": 0.16945810765028,
"num_tokens": 9138070.0,
"step": 4090
},
{
"entropy": 5.740090227127075,
"epoch": 0.4599314876172292,
"grad_norm": 1.234375,
"learning_rate": 0.0004986297803518865,
"loss": 5.6978,
"mean_token_accuracy": 0.16473543047904968,
"num_tokens": 9149489.0,
"step": 4095
},
{
"entropy": 5.77739839553833,
"epoch": 0.46049306452518673,
"grad_norm": 1.1640625,
"learning_rate": 0.0004986253526571971,
"loss": 5.7085,
"mean_token_accuracy": 0.17158406525850295,
"num_tokens": 9160363.0,
"step": 4100
},
{
"entropy": 5.784992265701294,
"epoch": 0.4610546414331443,
"grad_norm": 1.1953125,
"learning_rate": 0.0004986209178421899,
"loss": 5.7509,
"mean_token_accuracy": 0.16491172164678575,
"num_tokens": 9171619.0,
"step": 4105
},
{
"entropy": 5.644632768630982,
"epoch": 0.46161621834110184,
"grad_norm": 1.203125,
"learning_rate": 0.000498616475907006,
"loss": 5.5969,
"mean_token_accuracy": 0.17256009876728057,
"num_tokens": 9182828.0,
"step": 4110
},
{
"entropy": 5.794235563278198,
"epoch": 0.46217779524905933,
"grad_norm": 1.15625,
"learning_rate": 0.0004986120268517865,
"loss": 5.7765,
"mean_token_accuracy": 0.17010755389928817,
"num_tokens": 9195249.0,
"step": 4115
},
{
"entropy": 5.678744506835938,
"epoch": 0.4627393721570169,
"grad_norm": 1.3828125,
"learning_rate": 0.0004986075706766736,
"loss": 5.5759,
"mean_token_accuracy": 0.17103192806243897,
"num_tokens": 9205770.0,
"step": 4120
},
{
"entropy": 5.6166435241699215,
"epoch": 0.46330094906497443,
"grad_norm": 1.21875,
"learning_rate": 0.0004986031073818088,
"loss": 5.5905,
"mean_token_accuracy": 0.173680616915226,
"num_tokens": 9216367.0,
"step": 4125
},
{
"entropy": 5.659033250808716,
"epoch": 0.463862525972932,
"grad_norm": 1.2109375,
"learning_rate": 0.0004985986369673343,
"loss": 5.705,
"mean_token_accuracy": 0.1619468003511429,
"num_tokens": 9228524.0,
"step": 4130
},
{
"entropy": 5.7106360912323,
"epoch": 0.46442410288088953,
"grad_norm": 1.2734375,
"learning_rate": 0.0004985941594333926,
"loss": 5.6549,
"mean_token_accuracy": 0.17514504939317704,
"num_tokens": 9238907.0,
"step": 4135
},
{
"entropy": 5.769889688491821,
"epoch": 0.4649856797888471,
"grad_norm": 1.28125,
"learning_rate": 0.000498589674780126,
"loss": 5.7966,
"mean_token_accuracy": 0.172238390147686,
"num_tokens": 9250491.0,
"step": 4140
},
{
"entropy": 5.7498846530914305,
"epoch": 0.46554725669680463,
"grad_norm": 1.1875,
"learning_rate": 0.0004985851830076775,
"loss": 5.5834,
"mean_token_accuracy": 0.17644327133893967,
"num_tokens": 9260984.0,
"step": 4145
},
{
"entropy": 5.763900184631348,
"epoch": 0.4661088336047622,
"grad_norm": 1.1796875,
"learning_rate": 0.0004985806841161901,
"loss": 5.7518,
"mean_token_accuracy": 0.1682131789624691,
"num_tokens": 9272416.0,
"step": 4150
},
{
"entropy": 5.792823648452758,
"epoch": 0.46667041051271974,
"grad_norm": 1.3828125,
"learning_rate": 0.000498576178105807,
"loss": 5.7255,
"mean_token_accuracy": 0.1679478943347931,
"num_tokens": 9284513.0,
"step": 4155
},
{
"entropy": 5.809442186355591,
"epoch": 0.4672319874206773,
"grad_norm": 1.2890625,
"learning_rate": 0.0004985716649766717,
"loss": 5.6255,
"mean_token_accuracy": 0.17599722594022751,
"num_tokens": 9293422.0,
"step": 4160
},
{
"entropy": 5.636828804016114,
"epoch": 0.4677935643286348,
"grad_norm": 1.203125,
"learning_rate": 0.0004985671447289277,
"loss": 5.6261,
"mean_token_accuracy": 0.16604670286178588,
"num_tokens": 9304735.0,
"step": 4165
},
{
"entropy": 5.698333883285523,
"epoch": 0.46835514123659233,
"grad_norm": 1.3515625,
"learning_rate": 0.0004985626173627194,
"loss": 5.7275,
"mean_token_accuracy": 0.1659233331680298,
"num_tokens": 9315659.0,
"step": 4170
},
{
"entropy": 5.719541835784912,
"epoch": 0.4689167181445499,
"grad_norm": 1.25,
"learning_rate": 0.0004985580828781904,
"loss": 5.7418,
"mean_token_accuracy": 0.16786241233348848,
"num_tokens": 9327106.0,
"step": 4175
},
{
"entropy": 5.883518838882447,
"epoch": 0.46947829505250743,
"grad_norm": 1.25,
"learning_rate": 0.0004985535412754855,
"loss": 5.7813,
"mean_token_accuracy": 0.1729061335325241,
"num_tokens": 9338762.0,
"step": 4180
},
{
"entropy": 5.691299200057983,
"epoch": 0.470039871960465,
"grad_norm": 1.484375,
"learning_rate": 0.000498548992554749,
"loss": 5.6583,
"mean_token_accuracy": 0.1687624856829643,
"num_tokens": 9349495.0,
"step": 4185
},
{
"entropy": 5.834885692596435,
"epoch": 0.47060144886842253,
"grad_norm": 1.21875,
"learning_rate": 0.0004985444367161259,
"loss": 5.7409,
"mean_token_accuracy": 0.1736970901489258,
"num_tokens": 9360830.0,
"step": 4190
},
{
"entropy": 5.6631816864013675,
"epoch": 0.4711630257763801,
"grad_norm": 1.1953125,
"learning_rate": 0.0004985398737597614,
"loss": 5.631,
"mean_token_accuracy": 0.16964688301086425,
"num_tokens": 9371668.0,
"step": 4195
},
{
"entropy": 5.7335247039794925,
"epoch": 0.47172460268433763,
"grad_norm": 1.28125,
"learning_rate": 0.0004985353036858004,
"loss": 5.7681,
"mean_token_accuracy": 0.16211981773376466,
"num_tokens": 9383437.0,
"step": 4200
},
{
"entropy": 5.664580583572388,
"epoch": 0.4722861795922952,
"grad_norm": 1.140625,
"learning_rate": 0.0004985307264943887,
"loss": 5.599,
"mean_token_accuracy": 0.17601722329854966,
"num_tokens": 9394597.0,
"step": 4205
},
{
"entropy": 5.617810297012329,
"epoch": 0.47284775650025274,
"grad_norm": 1.390625,
"learning_rate": 0.0004985261421856721,
"loss": 5.5168,
"mean_token_accuracy": 0.17707317918539048,
"num_tokens": 9404322.0,
"step": 4210
},
{
"entropy": 5.755496406555176,
"epoch": 0.47340933340821023,
"grad_norm": 1.09375,
"learning_rate": 0.0004985215507597963,
"loss": 5.8892,
"mean_token_accuracy": 0.1571732796728611,
"num_tokens": 9416740.0,
"step": 4215
},
{
"entropy": 5.790236186981201,
"epoch": 0.4739709103161678,
"grad_norm": 1.0859375,
"learning_rate": 0.0004985169522169076,
"loss": 5.703,
"mean_token_accuracy": 0.16906435489654542,
"num_tokens": 9428617.0,
"step": 4220
},
{
"entropy": 5.615524625778198,
"epoch": 0.47453248722412533,
"grad_norm": 1.1953125,
"learning_rate": 0.0004985123465571526,
"loss": 5.6643,
"mean_token_accuracy": 0.17278331369161606,
"num_tokens": 9439686.0,
"step": 4225
},
{
"entropy": 5.729635190963745,
"epoch": 0.4750940641320829,
"grad_norm": 1.296875,
"learning_rate": 0.0004985077337806777,
"loss": 5.6259,
"mean_token_accuracy": 0.17399909645318984,
"num_tokens": 9450378.0,
"step": 4230
},
{
"entropy": 5.782553482055664,
"epoch": 0.47565564104004043,
"grad_norm": 1.296875,
"learning_rate": 0.0004985031138876299,
"loss": 5.6689,
"mean_token_accuracy": 0.1688866823911667,
"num_tokens": 9461659.0,
"step": 4235
},
{
"entropy": 5.720237922668457,
"epoch": 0.476217217947998,
"grad_norm": 0.9765625,
"learning_rate": 0.0004984984868781563,
"loss": 5.706,
"mean_token_accuracy": 0.16742132753133773,
"num_tokens": 9472528.0,
"step": 4240
},
{
"entropy": 5.745385217666626,
"epoch": 0.47677879485595553,
"grad_norm": 1.0859375,
"learning_rate": 0.0004984938527524041,
"loss": 5.6343,
"mean_token_accuracy": 0.17709124386310576,
"num_tokens": 9483388.0,
"step": 4245
},
{
"entropy": 5.53260817527771,
"epoch": 0.4773403717639131,
"grad_norm": 1.171875,
"learning_rate": 0.000498489211510521,
"loss": 5.4964,
"mean_token_accuracy": 0.1782183974981308,
"num_tokens": 9495178.0,
"step": 4250
},
{
"entropy": 5.673513889312744,
"epoch": 0.47790194867187064,
"grad_norm": 1.171875,
"learning_rate": 0.0004984845631526547,
"loss": 5.7266,
"mean_token_accuracy": 0.16570596396923065,
"num_tokens": 9506226.0,
"step": 4255
},
{
"entropy": 5.77213306427002,
"epoch": 0.47846352557982813,
"grad_norm": 1.1015625,
"learning_rate": 0.0004984799076789534,
"loss": 5.6385,
"mean_token_accuracy": 0.16841391623020172,
"num_tokens": 9517671.0,
"step": 4260
},
{
"entropy": 5.687767696380615,
"epoch": 0.4790251024877857,
"grad_norm": 1.3671875,
"learning_rate": 0.0004984752450895649,
"loss": 5.702,
"mean_token_accuracy": 0.1754951685667038,
"num_tokens": 9527637.0,
"step": 4265
},
{
"entropy": 5.798471355438233,
"epoch": 0.47958667939574323,
"grad_norm": 1.3359375,
"learning_rate": 0.000498470575384638,
"loss": 5.7935,
"mean_token_accuracy": 0.1615818075835705,
"num_tokens": 9539686.0,
"step": 4270
},
{
"entropy": 5.790640926361084,
"epoch": 0.4801482563037008,
"grad_norm": 1.2890625,
"learning_rate": 0.0004984658985643214,
"loss": 5.755,
"mean_token_accuracy": 0.1745975509285927,
"num_tokens": 9550775.0,
"step": 4275
},
{
"entropy": 5.795468854904175,
"epoch": 0.48070983321165833,
"grad_norm": 1.1875,
"learning_rate": 0.0004984612146287638,
"loss": 5.6672,
"mean_token_accuracy": 0.16850375533103942,
"num_tokens": 9561760.0,
"step": 4280
},
{
"entropy": 5.6588294982910154,
"epoch": 0.4812714101196159,
"grad_norm": 1.265625,
"learning_rate": 0.0004984565235781147,
"loss": 5.7017,
"mean_token_accuracy": 0.16536185294389724,
"num_tokens": 9573899.0,
"step": 4285
},
{
"entropy": 5.7043373584747314,
"epoch": 0.48183298702757343,
"grad_norm": 1.1796875,
"learning_rate": 0.000498451825412523,
"loss": 5.6448,
"mean_token_accuracy": 0.16699309051036834,
"num_tokens": 9584336.0,
"step": 4290
},
{
"entropy": 5.827418994903565,
"epoch": 0.482394563935531,
"grad_norm": 1.265625,
"learning_rate": 0.0004984471201321386,
"loss": 5.8456,
"mean_token_accuracy": 0.1607780173420906,
"num_tokens": 9596614.0,
"step": 4295
},
{
"entropy": 5.754131650924682,
"epoch": 0.48295614084348854,
"grad_norm": 1.1328125,
"learning_rate": 0.0004984424077371112,
"loss": 5.6801,
"mean_token_accuracy": 0.17277033030986785,
"num_tokens": 9606796.0,
"step": 4300
},
{
"entropy": 5.7239092826843265,
"epoch": 0.4835177177514461,
"grad_norm": 1.703125,
"learning_rate": 0.0004984376882275909,
"loss": 5.668,
"mean_token_accuracy": 0.1688304141163826,
"num_tokens": 9617705.0,
"step": 4305
},
{
"entropy": 5.6378098487854,
"epoch": 0.4840792946594036,
"grad_norm": 1.1640625,
"learning_rate": 0.0004984329616037279,
"loss": 5.6849,
"mean_token_accuracy": 0.1724874347448349,
"num_tokens": 9628098.0,
"step": 4310
},
{
"entropy": 5.690792465209961,
"epoch": 0.48464087156736113,
"grad_norm": 1.2734375,
"learning_rate": 0.0004984282278656727,
"loss": 5.6317,
"mean_token_accuracy": 0.1730911076068878,
"num_tokens": 9637697.0,
"step": 4315
},
{
"entropy": 5.666739749908447,
"epoch": 0.4852024484753187,
"grad_norm": 1.25,
"learning_rate": 0.0004984234870135762,
"loss": 5.5197,
"mean_token_accuracy": 0.1769974336028099,
"num_tokens": 9648221.0,
"step": 4320
},
{
"entropy": 5.773747825622559,
"epoch": 0.48576402538327623,
"grad_norm": 1.2109375,
"learning_rate": 0.0004984187390475891,
"loss": 5.7202,
"mean_token_accuracy": 0.16769949197769166,
"num_tokens": 9660182.0,
"step": 4325
},
{
"entropy": 5.671519470214844,
"epoch": 0.4863256022912338,
"grad_norm": 1.1796875,
"learning_rate": 0.0004984139839678628,
"loss": 5.7196,
"mean_token_accuracy": 0.16675896048545838,
"num_tokens": 9672121.0,
"step": 4330
},
{
"entropy": 5.768838596343994,
"epoch": 0.48688717919919133,
"grad_norm": 1.265625,
"learning_rate": 0.0004984092217745486,
"loss": 5.6232,
"mean_token_accuracy": 0.1730677530169487,
"num_tokens": 9682651.0,
"step": 4335
},
{
"entropy": 5.777725887298584,
"epoch": 0.4874487561071489,
"grad_norm": 1.1875,
"learning_rate": 0.0004984044524677981,
"loss": 5.7831,
"mean_token_accuracy": 0.16424083113670349,
"num_tokens": 9693989.0,
"step": 4340
},
{
"entropy": 5.660239791870117,
"epoch": 0.48801033301510643,
"grad_norm": 1.1875,
"learning_rate": 0.0004983996760477632,
"loss": 5.6261,
"mean_token_accuracy": 0.1726745620369911,
"num_tokens": 9703802.0,
"step": 4345
},
{
"entropy": 5.6591418266296385,
"epoch": 0.488571909923064,
"grad_norm": 1.3515625,
"learning_rate": 0.000498394892514596,
"loss": 5.5721,
"mean_token_accuracy": 0.1818757399916649,
"num_tokens": 9714287.0,
"step": 4350
},
{
"entropy": 5.653743743896484,
"epoch": 0.48913348683102154,
"grad_norm": 1.28125,
"learning_rate": 0.0004983901018684486,
"loss": 5.6098,
"mean_token_accuracy": 0.1732673391699791,
"num_tokens": 9724819.0,
"step": 4355
},
{
"entropy": 5.735607576370239,
"epoch": 0.48969506373897903,
"grad_norm": 1.359375,
"learning_rate": 0.0004983853041094739,
"loss": 5.7434,
"mean_token_accuracy": 0.16624641716480254,
"num_tokens": 9736326.0,
"step": 4360
},
{
"entropy": 5.833954906463623,
"epoch": 0.4902566406469366,
"grad_norm": 1.1484375,
"learning_rate": 0.0004983804992378243,
"loss": 5.7707,
"mean_token_accuracy": 0.16942221522331238,
"num_tokens": 9747883.0,
"step": 4365
},
{
"entropy": 5.686681890487671,
"epoch": 0.49081821755489413,
"grad_norm": 1.59375,
"learning_rate": 0.0004983756872536532,
"loss": 5.6305,
"mean_token_accuracy": 0.17395332753658294,
"num_tokens": 9758191.0,
"step": 4370
},
{
"entropy": 5.771088075637818,
"epoch": 0.4913797944628517,
"grad_norm": 1.3828125,
"learning_rate": 0.0004983708681571134,
"loss": 5.7465,
"mean_token_accuracy": 0.1644330158829689,
"num_tokens": 9768852.0,
"step": 4375
},
{
"entropy": 5.801872968673706,
"epoch": 0.49194137137080923,
"grad_norm": 1.2578125,
"learning_rate": 0.0004983660419483586,
"loss": 5.6736,
"mean_token_accuracy": 0.16890106201171876,
"num_tokens": 9779033.0,
"step": 4380
},
{
"entropy": 5.633499383926392,
"epoch": 0.4925029482787668,
"grad_norm": 1.1796875,
"learning_rate": 0.0004983612086275424,
"loss": 5.5972,
"mean_token_accuracy": 0.17648783475160598,
"num_tokens": 9790777.0,
"step": 4385
},
{
"entropy": 5.873503541946411,
"epoch": 0.49306452518672433,
"grad_norm": 1.2421875,
"learning_rate": 0.0004983563681948187,
"loss": 5.8732,
"mean_token_accuracy": 0.15995177030563354,
"num_tokens": 9802305.0,
"step": 4390
},
{
"entropy": 5.762320137023925,
"epoch": 0.4936261020946819,
"grad_norm": 1.3046875,
"learning_rate": 0.0004983515206503416,
"loss": 5.6528,
"mean_token_accuracy": 0.16844187974929808,
"num_tokens": 9813590.0,
"step": 4395
},
{
"entropy": 5.667017126083374,
"epoch": 0.49418767900263944,
"grad_norm": 1.3828125,
"learning_rate": 0.0004983466659942655,
"loss": 5.7402,
"mean_token_accuracy": 0.16711403727531432,
"num_tokens": 9824080.0,
"step": 4400
},
{
"entropy": 5.791653776168824,
"epoch": 0.49474925591059693,
"grad_norm": 1.21875,
"learning_rate": 0.0004983418042267449,
"loss": 5.6341,
"mean_token_accuracy": 0.16764632910490035,
"num_tokens": 9835441.0,
"step": 4405
},
{
"entropy": 5.6966509342193605,
"epoch": 0.4953108328185545,
"grad_norm": 1.2734375,
"learning_rate": 0.0004983369353479345,
"loss": 5.6224,
"mean_token_accuracy": 0.17434028089046477,
"num_tokens": 9846750.0,
"step": 4410
},
{
"entropy": 5.705007600784302,
"epoch": 0.49587240972651203,
"grad_norm": 1.421875,
"learning_rate": 0.0004983320593579895,
"loss": 5.7672,
"mean_token_accuracy": 0.16393694877624512,
"num_tokens": 9857987.0,
"step": 4415
},
{
"entropy": 5.774279975891114,
"epoch": 0.4964339866344696,
"grad_norm": 1.140625,
"learning_rate": 0.0004983271762570653,
"loss": 5.733,
"mean_token_accuracy": 0.1640989825129509,
"num_tokens": 9869736.0,
"step": 4420
},
{
"entropy": 5.738402318954468,
"epoch": 0.49699556354242713,
"grad_norm": 1.1171875,
"learning_rate": 0.000498322286045317,
"loss": 5.6845,
"mean_token_accuracy": 0.16476065963506697,
"num_tokens": 9881610.0,
"step": 4425
},
{
"entropy": 5.702138423919678,
"epoch": 0.4975571404503847,
"grad_norm": 1.2734375,
"learning_rate": 0.0004983173887229006,
"loss": 5.676,
"mean_token_accuracy": 0.16530685871839523,
"num_tokens": 9893203.0,
"step": 4430
},
{
"entropy": 5.712817287445068,
"epoch": 0.49811871735834223,
"grad_norm": 1.25,
"learning_rate": 0.0004983124842899719,
"loss": 5.6034,
"mean_token_accuracy": 0.16780591905117034,
"num_tokens": 9904392.0,
"step": 4435
},
{
"entropy": 5.551604795455932,
"epoch": 0.4986802942662998,
"grad_norm": 1.359375,
"learning_rate": 0.0004983075727466871,
"loss": 5.6049,
"mean_token_accuracy": 0.17411096394062042,
"num_tokens": 9916184.0,
"step": 4440
},
{
"entropy": 5.741838788986206,
"epoch": 0.49924187117425733,
"grad_norm": 1.1953125,
"learning_rate": 0.0004983026540932026,
"loss": 5.7091,
"mean_token_accuracy": 0.17291998565196992,
"num_tokens": 9927516.0,
"step": 4445
},
{
"entropy": 5.773145580291748,
"epoch": 0.4998034480822149,
"grad_norm": 1.3203125,
"learning_rate": 0.0004982977283296749,
"loss": 5.7322,
"mean_token_accuracy": 0.1653789162635803,
"num_tokens": 9939182.0,
"step": 4450
},
{
"entropy": 5.792496871948242,
"epoch": 0.5003650249901724,
"grad_norm": 1.5546875,
"learning_rate": 0.000498292795456261,
"loss": 5.7542,
"mean_token_accuracy": 0.1686200365424156,
"num_tokens": 9950382.0,
"step": 4455
},
{
"entropy": 5.750992727279663,
"epoch": 0.50092660189813,
"grad_norm": 1.609375,
"learning_rate": 0.0004982878554731178,
"loss": 5.7399,
"mean_token_accuracy": 0.17350716888904572,
"num_tokens": 9961053.0,
"step": 4460
},
{
"entropy": 5.686504983901978,
"epoch": 0.5014881788060875,
"grad_norm": 1.1875,
"learning_rate": 0.0004982829083804028,
"loss": 5.6509,
"mean_token_accuracy": 0.17399145066738128,
"num_tokens": 9971313.0,
"step": 4465
},
{
"entropy": 5.686606454849243,
"epoch": 0.5020497557140451,
"grad_norm": 1.28125,
"learning_rate": 0.0004982779541782733,
"loss": 5.6552,
"mean_token_accuracy": 0.1797782674431801,
"num_tokens": 9981257.0,
"step": 4470
},
{
"entropy": 5.6951133728027346,
"epoch": 0.5026113326220026,
"grad_norm": 1.5546875,
"learning_rate": 0.0004982729928668872,
"loss": 5.567,
"mean_token_accuracy": 0.17247527539730073,
"num_tokens": 9991860.0,
"step": 4475
},
{
"entropy": 5.70995512008667,
"epoch": 0.5031729095299601,
"grad_norm": 1.28125,
"learning_rate": 0.0004982680244464024,
"loss": 5.6897,
"mean_token_accuracy": 0.17129286378622055,
"num_tokens": 10002320.0,
"step": 4480
},
{
"entropy": 5.786509037017822,
"epoch": 0.5037344864379176,
"grad_norm": 1.359375,
"learning_rate": 0.0004982630489169771,
"loss": 5.7862,
"mean_token_accuracy": 0.1639959082007408,
"num_tokens": 10013187.0,
"step": 4485
},
{
"entropy": 5.644416236877442,
"epoch": 0.5042960633458752,
"grad_norm": 1.21875,
"learning_rate": 0.0004982580662787697,
"loss": 5.5165,
"mean_token_accuracy": 0.1790595307946205,
"num_tokens": 10024268.0,
"step": 4490
},
{
"entropy": 5.731760311126709,
"epoch": 0.5048576402538327,
"grad_norm": 1.3515625,
"learning_rate": 0.0004982530765319389,
"loss": 5.7919,
"mean_token_accuracy": 0.16981189101934432,
"num_tokens": 10034738.0,
"step": 4495
},
{
"entropy": 5.814146232604981,
"epoch": 0.5054192171617903,
"grad_norm": 1.4140625,
"learning_rate": 0.0004982480796766436,
"loss": 5.7183,
"mean_token_accuracy": 0.169575534760952,
"num_tokens": 10045666.0,
"step": 4500
},
{
"entropy": 5.667458629608154,
"epoch": 0.5059807940697478,
"grad_norm": 2.109375,
"learning_rate": 0.0004982430757130429,
"loss": 5.6101,
"mean_token_accuracy": 0.17285872995853424,
"num_tokens": 10057084.0,
"step": 4505
},
{
"entropy": 5.660326385498047,
"epoch": 0.5065423709777054,
"grad_norm": 1.3359375,
"learning_rate": 0.0004982380646412959,
"loss": 5.5795,
"mean_token_accuracy": 0.1707088440656662,
"num_tokens": 10066754.0,
"step": 4510
},
{
"entropy": 5.643389797210693,
"epoch": 0.5071039478856629,
"grad_norm": 1.2109375,
"learning_rate": 0.0004982330464615625,
"loss": 5.5879,
"mean_token_accuracy": 0.17634238600730895,
"num_tokens": 10078139.0,
"step": 4515
},
{
"entropy": 5.7179309844970705,
"epoch": 0.5076655247936205,
"grad_norm": 1.203125,
"learning_rate": 0.0004982280211740023,
"loss": 5.6492,
"mean_token_accuracy": 0.1679838240146637,
"num_tokens": 10089652.0,
"step": 4520
},
{
"entropy": 5.5877519130706785,
"epoch": 0.508227101701578,
"grad_norm": 1.2578125,
"learning_rate": 0.0004982229887787753,
"loss": 5.4859,
"mean_token_accuracy": 0.17653454691171647,
"num_tokens": 10100642.0,
"step": 4525
},
{
"entropy": 5.707630443572998,
"epoch": 0.5087886786095356,
"grad_norm": 1.2734375,
"learning_rate": 0.0004982179492760418,
"loss": 5.7087,
"mean_token_accuracy": 0.16735783219337463,
"num_tokens": 10112036.0,
"step": 4530
},
{
"entropy": 5.698233795166016,
"epoch": 0.5093502555174931,
"grad_norm": 1.328125,
"learning_rate": 0.0004982129026659621,
"loss": 5.5799,
"mean_token_accuracy": 0.17062337547540665,
"num_tokens": 10121653.0,
"step": 4535
},
{
"entropy": 5.682953500747681,
"epoch": 0.5099118324254507,
"grad_norm": 1.3046875,
"learning_rate": 0.0004982078489486972,
"loss": 5.6939,
"mean_token_accuracy": 0.17003500163555146,
"num_tokens": 10132271.0,
"step": 4540
},
{
"entropy": 5.639585161209107,
"epoch": 0.5104734093334082,
"grad_norm": 1.296875,
"learning_rate": 0.0004982027881244076,
"loss": 5.5727,
"mean_token_accuracy": 0.17094035893678666,
"num_tokens": 10143411.0,
"step": 4545
},
{
"entropy": 5.773826980590821,
"epoch": 0.5110349862413658,
"grad_norm": 1.25,
"learning_rate": 0.0004981977201932549,
"loss": 5.6731,
"mean_token_accuracy": 0.172907754778862,
"num_tokens": 10154338.0,
"step": 4550
},
{
"entropy": 5.792322492599487,
"epoch": 0.5115965631493233,
"grad_norm": 1.2421875,
"learning_rate": 0.0004981926451554001,
"loss": 5.7105,
"mean_token_accuracy": 0.17222313582897186,
"num_tokens": 10164810.0,
"step": 4555
},
{
"entropy": 5.703101015090942,
"epoch": 0.5121581400572809,
"grad_norm": 1.3046875,
"learning_rate": 0.0004981875630110048,
"loss": 5.6375,
"mean_token_accuracy": 0.1698868989944458,
"num_tokens": 10176323.0,
"step": 4560
},
{
"entropy": 5.686052560806274,
"epoch": 0.5127197169652384,
"grad_norm": 1.3125,
"learning_rate": 0.0004981824737602311,
"loss": 5.6564,
"mean_token_accuracy": 0.17510976791381835,
"num_tokens": 10187110.0,
"step": 4565
},
{
"entropy": 5.63431224822998,
"epoch": 0.513281293873196,
"grad_norm": 1.296875,
"learning_rate": 0.0004981773774032409,
"loss": 5.613,
"mean_token_accuracy": 0.17071798145771028,
"num_tokens": 10199099.0,
"step": 4570
},
{
"entropy": 5.708571243286133,
"epoch": 0.5138428707811534,
"grad_norm": 1.34375,
"learning_rate": 0.0004981722739401962,
"loss": 5.6087,
"mean_token_accuracy": 0.17462070137262345,
"num_tokens": 10209726.0,
"step": 4575
},
{
"entropy": 5.764850616455078,
"epoch": 0.514404447689111,
"grad_norm": 1.46875,
"learning_rate": 0.00049816716337126,
"loss": 5.7211,
"mean_token_accuracy": 0.16945174485445022,
"num_tokens": 10220858.0,
"step": 4580
},
{
"entropy": 5.558522653579712,
"epoch": 0.5149660245970685,
"grad_norm": 1.34375,
"learning_rate": 0.0004981620456965946,
"loss": 5.5423,
"mean_token_accuracy": 0.17813366651535034,
"num_tokens": 10232250.0,
"step": 4585
},
{
"entropy": 5.76922926902771,
"epoch": 0.5155276015050261,
"grad_norm": 1.578125,
"learning_rate": 0.0004981569209163633,
"loss": 5.7328,
"mean_token_accuracy": 0.1641071617603302,
"num_tokens": 10244584.0,
"step": 4590
},
{
"entropy": 5.802131414413452,
"epoch": 0.5160891784129836,
"grad_norm": 1.46875,
"learning_rate": 0.0004981517890307288,
"loss": 5.6212,
"mean_token_accuracy": 0.17408365458250047,
"num_tokens": 10255926.0,
"step": 4595
},
{
"entropy": 5.721409463882447,
"epoch": 0.5166507553209412,
"grad_norm": 1.5390625,
"learning_rate": 0.000498146650039855,
"loss": 5.6891,
"mean_token_accuracy": 0.16892822533845903,
"num_tokens": 10266691.0,
"step": 4600
},
{
"entropy": 5.719908094406128,
"epoch": 0.5172123322288987,
"grad_norm": 1.4375,
"learning_rate": 0.0004981415039439054,
"loss": 5.7592,
"mean_token_accuracy": 0.16662688702344894,
"num_tokens": 10277744.0,
"step": 4605
},
{
"entropy": 5.709627294540406,
"epoch": 0.5177739091368563,
"grad_norm": 1.34375,
"learning_rate": 0.0004981363507430436,
"loss": 5.6052,
"mean_token_accuracy": 0.17523760497570037,
"num_tokens": 10288792.0,
"step": 4610
},
{
"entropy": 5.740534257888794,
"epoch": 0.5183354860448138,
"grad_norm": 1.390625,
"learning_rate": 0.0004981311904374339,
"loss": 5.6415,
"mean_token_accuracy": 0.17211671769618989,
"num_tokens": 10299569.0,
"step": 4615
},
{
"entropy": 5.639956712722778,
"epoch": 0.5188970629527714,
"grad_norm": 1.4140625,
"learning_rate": 0.0004981260230272405,
"loss": 5.6269,
"mean_token_accuracy": 0.17451872825622558,
"num_tokens": 10309742.0,
"step": 4620
},
{
"entropy": 5.790673398971558,
"epoch": 0.5194586398607289,
"grad_norm": 1.734375,
"learning_rate": 0.0004981208485126281,
"loss": 5.6689,
"mean_token_accuracy": 0.16619865447282792,
"num_tokens": 10320603.0,
"step": 4625
},
{
"entropy": 5.68487696647644,
"epoch": 0.5200202167686865,
"grad_norm": 1.3125,
"learning_rate": 0.0004981156668937613,
"loss": 5.6418,
"mean_token_accuracy": 0.1712615206837654,
"num_tokens": 10331498.0,
"step": 4630
},
{
"entropy": 5.6432267189025875,
"epoch": 0.520581793676644,
"grad_norm": 1.484375,
"learning_rate": 0.0004981104781708051,
"loss": 5.6691,
"mean_token_accuracy": 0.16949173659086228,
"num_tokens": 10342002.0,
"step": 4635
},
{
"entropy": 5.706935214996338,
"epoch": 0.5211433705846016,
"grad_norm": 1.5078125,
"learning_rate": 0.0004981052823439247,
"loss": 5.6176,
"mean_token_accuracy": 0.1688291236758232,
"num_tokens": 10351797.0,
"step": 4640
},
{
"entropy": 5.650330400466919,
"epoch": 0.5217049474925591,
"grad_norm": 1.3984375,
"learning_rate": 0.0004981000794132858,
"loss": 5.5573,
"mean_token_accuracy": 0.17336327135562896,
"num_tokens": 10362538.0,
"step": 4645
},
{
"entropy": 5.674770832061768,
"epoch": 0.5222665244005167,
"grad_norm": 1.5859375,
"learning_rate": 0.0004980948693790536,
"loss": 5.5812,
"mean_token_accuracy": 0.17896029800176622,
"num_tokens": 10373365.0,
"step": 4650
},
{
"entropy": 5.622525215148926,
"epoch": 0.5228281013084742,
"grad_norm": 1.3046875,
"learning_rate": 0.0004980896522413943,
"loss": 5.5762,
"mean_token_accuracy": 0.16874225288629532,
"num_tokens": 10385087.0,
"step": 4655
},
{
"entropy": 5.696762609481811,
"epoch": 0.5233896782164318,
"grad_norm": 1.390625,
"learning_rate": 0.0004980844280004739,
"loss": 5.6879,
"mean_token_accuracy": 0.16689287573099137,
"num_tokens": 10396014.0,
"step": 4660
},
{
"entropy": 5.694055700302124,
"epoch": 0.5239512551243893,
"grad_norm": 1.4375,
"learning_rate": 0.0004980791966564589,
"loss": 5.6275,
"mean_token_accuracy": 0.16552235633134843,
"num_tokens": 10407844.0,
"step": 4665
},
{
"entropy": 5.664031982421875,
"epoch": 0.5245128320323468,
"grad_norm": 1.1796875,
"learning_rate": 0.0004980739582095158,
"loss": 5.6913,
"mean_token_accuracy": 0.16484711915254593,
"num_tokens": 10419892.0,
"step": 4670
},
{
"entropy": 5.759139060974121,
"epoch": 0.5250744089403043,
"grad_norm": 1.40625,
"learning_rate": 0.0004980687126598111,
"loss": 5.665,
"mean_token_accuracy": 0.16463309079408645,
"num_tokens": 10430686.0,
"step": 4675
},
{
"entropy": 5.729007196426392,
"epoch": 0.5256359858482619,
"grad_norm": 1.234375,
"learning_rate": 0.0004980634600075123,
"loss": 5.6732,
"mean_token_accuracy": 0.16411852985620498,
"num_tokens": 10443785.0,
"step": 4680
},
{
"entropy": 5.643607139587402,
"epoch": 0.5261975627562194,
"grad_norm": 1.140625,
"learning_rate": 0.0004980582002527863,
"loss": 5.5762,
"mean_token_accuracy": 0.1756297081708908,
"num_tokens": 10454889.0,
"step": 4685
},
{
"entropy": 5.681384086608887,
"epoch": 0.526759139664177,
"grad_norm": 1.328125,
"learning_rate": 0.0004980529333958008,
"loss": 5.7337,
"mean_token_accuracy": 0.16439043134450912,
"num_tokens": 10466525.0,
"step": 4690
},
{
"entropy": 5.817110013961792,
"epoch": 0.5273207165721345,
"grad_norm": 1.234375,
"learning_rate": 0.0004980476594367233,
"loss": 5.663,
"mean_token_accuracy": 0.16700455844402312,
"num_tokens": 10476329.0,
"step": 4695
},
{
"entropy": 5.797155332565308,
"epoch": 0.5278822934800921,
"grad_norm": 1.34375,
"learning_rate": 0.0004980423783757219,
"loss": 5.7329,
"mean_token_accuracy": 0.167078597843647,
"num_tokens": 10487178.0,
"step": 4700
},
{
"entropy": 5.71194429397583,
"epoch": 0.5284438703880496,
"grad_norm": 1.28125,
"learning_rate": 0.0004980370902129646,
"loss": 5.6868,
"mean_token_accuracy": 0.16118710339069367,
"num_tokens": 10499293.0,
"step": 4705
},
{
"entropy": 5.821799802780151,
"epoch": 0.5290054472960072,
"grad_norm": 1.453125,
"learning_rate": 0.00049803179494862,
"loss": 5.8721,
"mean_token_accuracy": 0.1579592101275921,
"num_tokens": 10511762.0,
"step": 4710
},
{
"entropy": 5.701119422912598,
"epoch": 0.5295670242039647,
"grad_norm": 1.5859375,
"learning_rate": 0.0004980264925828563,
"loss": 5.5766,
"mean_token_accuracy": 0.17831983268260956,
"num_tokens": 10521725.0,
"step": 4715
},
{
"entropy": 5.619213342666626,
"epoch": 0.5301286011119223,
"grad_norm": 1.3984375,
"learning_rate": 0.0004980211831158427,
"loss": 5.5586,
"mean_token_accuracy": 0.1704928159713745,
"num_tokens": 10531764.0,
"step": 4720
},
{
"entropy": 5.708400440216065,
"epoch": 0.5306901780198798,
"grad_norm": 1.2578125,
"learning_rate": 0.0004980158665477481,
"loss": 5.6212,
"mean_token_accuracy": 0.16902651339769365,
"num_tokens": 10542127.0,
"step": 4725
},
{
"entropy": 5.7879430770874025,
"epoch": 0.5312517549278374,
"grad_norm": 1.2578125,
"learning_rate": 0.0004980105428787419,
"loss": 5.7019,
"mean_token_accuracy": 0.1671772062778473,
"num_tokens": 10552580.0,
"step": 4730
},
{
"entropy": 5.775068092346191,
"epoch": 0.5318133318357949,
"grad_norm": 1.2890625,
"learning_rate": 0.0004980052121089934,
"loss": 5.6431,
"mean_token_accuracy": 0.16931187361478806,
"num_tokens": 10562961.0,
"step": 4735
},
{
"entropy": 5.723823928833008,
"epoch": 0.5323749087437525,
"grad_norm": 1.2421875,
"learning_rate": 0.0004979998742386725,
"loss": 5.6393,
"mean_token_accuracy": 0.17154961675405503,
"num_tokens": 10573210.0,
"step": 4740
},
{
"entropy": 5.550125026702881,
"epoch": 0.53293648565171,
"grad_norm": 1.078125,
"learning_rate": 0.0004979945292679492,
"loss": 5.6055,
"mean_token_accuracy": 0.17195845246315003,
"num_tokens": 10583878.0,
"step": 4745
},
{
"entropy": 5.736267518997193,
"epoch": 0.5334980625596676,
"grad_norm": 1.234375,
"learning_rate": 0.0004979891771969935,
"loss": 5.6749,
"mean_token_accuracy": 0.16752220094203948,
"num_tokens": 10593420.0,
"step": 4750
},
{
"entropy": 5.72455244064331,
"epoch": 0.5340596394676251,
"grad_norm": 1.5234375,
"learning_rate": 0.000497983818025976,
"loss": 5.7001,
"mean_token_accuracy": 0.17147122621536254,
"num_tokens": 10604213.0,
"step": 4755
},
{
"entropy": 5.733754968643188,
"epoch": 0.5346212163755827,
"grad_norm": 1.3359375,
"learning_rate": 0.0004979784517550671,
"loss": 5.6804,
"mean_token_accuracy": 0.16548150330781936,
"num_tokens": 10615502.0,
"step": 4760
},
{
"entropy": 5.770488023757935,
"epoch": 0.5351827932835401,
"grad_norm": 1.1796875,
"learning_rate": 0.0004979730783844379,
"loss": 5.7339,
"mean_token_accuracy": 0.16314537972211837,
"num_tokens": 10627589.0,
"step": 4765
},
{
"entropy": 5.643162393569947,
"epoch": 0.5357443701914977,
"grad_norm": 1.359375,
"learning_rate": 0.0004979676979142595,
"loss": 5.5656,
"mean_token_accuracy": 0.17168369889259338,
"num_tokens": 10638365.0,
"step": 4770
},
{
"entropy": 5.686746835708618,
"epoch": 0.5363059470994552,
"grad_norm": 1.25,
"learning_rate": 0.000497962310344703,
"loss": 5.6775,
"mean_token_accuracy": 0.1732397511601448,
"num_tokens": 10650664.0,
"step": 4775
},
{
"entropy": 5.774690580368042,
"epoch": 0.5368675240074128,
"grad_norm": 1.1484375,
"learning_rate": 0.00049795691567594,
"loss": 5.7192,
"mean_token_accuracy": 0.17400692254304886,
"num_tokens": 10662418.0,
"step": 4780
},
{
"entropy": 5.727471208572387,
"epoch": 0.5374291009153703,
"grad_norm": 1.390625,
"learning_rate": 0.0004979515139081426,
"loss": 5.5911,
"mean_token_accuracy": 0.1714724600315094,
"num_tokens": 10672793.0,
"step": 4785
},
{
"entropy": 5.6494053363800045,
"epoch": 0.5379906778233279,
"grad_norm": 1.265625,
"learning_rate": 0.0004979461050414822,
"loss": 5.6331,
"mean_token_accuracy": 0.17695551216602326,
"num_tokens": 10684082.0,
"step": 4790
},
{
"entropy": 5.581946754455567,
"epoch": 0.5385522547312854,
"grad_norm": 1.3515625,
"learning_rate": 0.0004979406890761315,
"loss": 5.5812,
"mean_token_accuracy": 0.1715100049972534,
"num_tokens": 10694490.0,
"step": 4795
},
{
"entropy": 5.674076843261719,
"epoch": 0.539113831639243,
"grad_norm": 1.40625,
"learning_rate": 0.0004979352660122629,
"loss": 5.5837,
"mean_token_accuracy": 0.17617941051721572,
"num_tokens": 10704731.0,
"step": 4800
},
{
"entropy": 5.66850790977478,
"epoch": 0.5396754085472005,
"grad_norm": 1.1171875,
"learning_rate": 0.0004979298358500488,
"loss": 5.6032,
"mean_token_accuracy": 0.17590020447969437,
"num_tokens": 10715735.0,
"step": 4805
},
{
"entropy": 5.624487066268921,
"epoch": 0.5402369854551581,
"grad_norm": 1.1484375,
"learning_rate": 0.0004979243985896624,
"loss": 5.583,
"mean_token_accuracy": 0.16958857625722884,
"num_tokens": 10726951.0,
"step": 4810
},
{
"entropy": 5.6196897506713865,
"epoch": 0.5407985623631156,
"grad_norm": 1.21875,
"learning_rate": 0.0004979189542312765,
"loss": 5.6462,
"mean_token_accuracy": 0.16544457972049714,
"num_tokens": 10737609.0,
"step": 4815
},
{
"entropy": 5.708540296554565,
"epoch": 0.5413601392710732,
"grad_norm": 1.2265625,
"learning_rate": 0.0004979135027750649,
"loss": 5.546,
"mean_token_accuracy": 0.17781233191490173,
"num_tokens": 10748957.0,
"step": 4820
},
{
"entropy": 5.692952251434326,
"epoch": 0.5419217161790307,
"grad_norm": 1.234375,
"learning_rate": 0.0004979080442212007,
"loss": 5.7278,
"mean_token_accuracy": 0.16248067915439607,
"num_tokens": 10759398.0,
"step": 4825
},
{
"entropy": 5.722014236450195,
"epoch": 0.5424832930869883,
"grad_norm": 1.2890625,
"learning_rate": 0.0004979025785698581,
"loss": 5.6898,
"mean_token_accuracy": 0.16471544951200484,
"num_tokens": 10770436.0,
"step": 4830
},
{
"entropy": 5.661945915222168,
"epoch": 0.5430448699949458,
"grad_norm": 1.5546875,
"learning_rate": 0.0004978971058212108,
"loss": 5.5268,
"mean_token_accuracy": 0.18072929829359055,
"num_tokens": 10781160.0,
"step": 4835
},
{
"entropy": 5.747938585281372,
"epoch": 0.5436064469029034,
"grad_norm": 1.1328125,
"learning_rate": 0.0004978916259754331,
"loss": 5.6488,
"mean_token_accuracy": 0.17626226991415023,
"num_tokens": 10793170.0,
"step": 4840
},
{
"entropy": 5.622552728652954,
"epoch": 0.5441680238108609,
"grad_norm": 1.2109375,
"learning_rate": 0.0004978861390326998,
"loss": 5.6017,
"mean_token_accuracy": 0.17149905860424042,
"num_tokens": 10803515.0,
"step": 4845
},
{
"entropy": 5.683855104446411,
"epoch": 0.5447296007188185,
"grad_norm": 1.3203125,
"learning_rate": 0.0004978806449931854,
"loss": 5.6508,
"mean_token_accuracy": 0.1706086203455925,
"num_tokens": 10814508.0,
"step": 4850
},
{
"entropy": 5.7445684432983395,
"epoch": 0.545291177626776,
"grad_norm": 1.1953125,
"learning_rate": 0.0004978751438570647,
"loss": 5.7316,
"mean_token_accuracy": 0.16973227858543397,
"num_tokens": 10825924.0,
"step": 4855
},
{
"entropy": 5.806747388839722,
"epoch": 0.5458527545347336,
"grad_norm": 1.2578125,
"learning_rate": 0.0004978696356245129,
"loss": 5.7128,
"mean_token_accuracy": 0.1720080703496933,
"num_tokens": 10837854.0,
"step": 4860
},
{
"entropy": 5.6908331394195555,
"epoch": 0.546414331442691,
"grad_norm": 1.28125,
"learning_rate": 0.0004978641202957057,
"loss": 5.6048,
"mean_token_accuracy": 0.17313567101955413,
"num_tokens": 10848279.0,
"step": 4865
},
{
"entropy": 5.578272819519043,
"epoch": 0.5469759083506486,
"grad_norm": 1.28125,
"learning_rate": 0.0004978585978708183,
"loss": 5.6206,
"mean_token_accuracy": 0.16253421157598497,
"num_tokens": 10859615.0,
"step": 4870
},
{
"entropy": 5.803055715560913,
"epoch": 0.5475374852586061,
"grad_norm": 1.4453125,
"learning_rate": 0.0004978530683500267,
"loss": 5.5867,
"mean_token_accuracy": 0.1725769266486168,
"num_tokens": 10871388.0,
"step": 4875
},
{
"entropy": 5.652014684677124,
"epoch": 0.5480990621665637,
"grad_norm": 1.3828125,
"learning_rate": 0.0004978475317335071,
"loss": 5.6551,
"mean_token_accuracy": 0.17160367518663405,
"num_tokens": 10882411.0,
"step": 4880
},
{
"entropy": 5.646870088577271,
"epoch": 0.5486606390745212,
"grad_norm": 1.5859375,
"learning_rate": 0.0004978419880214356,
"loss": 5.7551,
"mean_token_accuracy": 0.1707601010799408,
"num_tokens": 10894116.0,
"step": 4885
},
{
"entropy": 5.688396406173706,
"epoch": 0.5492222159824788,
"grad_norm": 1.359375,
"learning_rate": 0.0004978364372139887,
"loss": 5.5448,
"mean_token_accuracy": 0.17501346468925477,
"num_tokens": 10904514.0,
"step": 4890
},
{
"entropy": 5.611290693283081,
"epoch": 0.5497837928904363,
"grad_norm": 1.28125,
"learning_rate": 0.0004978308793113434,
"loss": 5.603,
"mean_token_accuracy": 0.17075815349817275,
"num_tokens": 10915602.0,
"step": 4895
},
{
"entropy": 5.676729536056518,
"epoch": 0.5503453697983939,
"grad_norm": 1.78125,
"learning_rate": 0.0004978253143136762,
"loss": 5.5808,
"mean_token_accuracy": 0.1725596696138382,
"num_tokens": 10926172.0,
"step": 4900
},
{
"entropy": 5.761356067657471,
"epoch": 0.5509069467063514,
"grad_norm": 1.40625,
"learning_rate": 0.0004978197422211647,
"loss": 5.7094,
"mean_token_accuracy": 0.16489994525909424,
"num_tokens": 10938435.0,
"step": 4905
},
{
"entropy": 5.622772598266602,
"epoch": 0.551468523614309,
"grad_norm": 1.421875,
"learning_rate": 0.0004978141630339863,
"loss": 5.5783,
"mean_token_accuracy": 0.17229571789503098,
"num_tokens": 10949118.0,
"step": 4910
},
{
"entropy": 5.6643308162689205,
"epoch": 0.5520301005222665,
"grad_norm": 1.328125,
"learning_rate": 0.0004978085767523184,
"loss": 5.6307,
"mean_token_accuracy": 0.16718402355909348,
"num_tokens": 10960120.0,
"step": 4915
},
{
"entropy": 5.6887603282928465,
"epoch": 0.5525916774302241,
"grad_norm": 1.4140625,
"learning_rate": 0.000497802983376339,
"loss": 5.6164,
"mean_token_accuracy": 0.17536334991455077,
"num_tokens": 10971210.0,
"step": 4920
},
{
"entropy": 5.637243032455444,
"epoch": 0.5531532543381816,
"grad_norm": 1.25,
"learning_rate": 0.0004977973829062263,
"loss": 5.6027,
"mean_token_accuracy": 0.18030029237270356,
"num_tokens": 10983446.0,
"step": 4925
},
{
"entropy": 5.72392692565918,
"epoch": 0.5537148312461392,
"grad_norm": 1.3984375,
"learning_rate": 0.0004977917753421585,
"loss": 5.7205,
"mean_token_accuracy": 0.16798515021800994,
"num_tokens": 10994776.0,
"step": 4930
},
{
"entropy": 5.770320081710816,
"epoch": 0.5542764081540967,
"grad_norm": 1.6171875,
"learning_rate": 0.000497786160684314,
"loss": 5.6232,
"mean_token_accuracy": 0.16814448237419127,
"num_tokens": 11005280.0,
"step": 4935
},
{
"entropy": 5.640605020523071,
"epoch": 0.5548379850620543,
"grad_norm": 1.3828125,
"learning_rate": 0.0004977805389328719,
"loss": 5.625,
"mean_token_accuracy": 0.1676290139555931,
"num_tokens": 11017509.0,
"step": 4940
},
{
"entropy": 5.709056663513183,
"epoch": 0.5553995619700118,
"grad_norm": 1.3359375,
"learning_rate": 0.000497774910088011,
"loss": 5.6466,
"mean_token_accuracy": 0.17211585193872453,
"num_tokens": 11029087.0,
"step": 4945
},
{
"entropy": 5.731402254104614,
"epoch": 0.5559611388779694,
"grad_norm": 1.296875,
"learning_rate": 0.0004977692741499106,
"loss": 5.6906,
"mean_token_accuracy": 0.1648757725954056,
"num_tokens": 11041144.0,
"step": 4950
},
{
"entropy": 5.854576158523559,
"epoch": 0.5565227157859269,
"grad_norm": 1.2578125,
"learning_rate": 0.00049776363111875,
"loss": 5.7568,
"mean_token_accuracy": 0.16550998091697694,
"num_tokens": 11054050.0,
"step": 4955
},
{
"entropy": 5.716461849212647,
"epoch": 0.5570842926938844,
"grad_norm": 1.5,
"learning_rate": 0.0004977579809947091,
"loss": 5.5785,
"mean_token_accuracy": 0.17459672689437866,
"num_tokens": 11064257.0,
"step": 4960
},
{
"entropy": 5.706990432739258,
"epoch": 0.5576458696018419,
"grad_norm": 1.3046875,
"learning_rate": 0.0004977523237779676,
"loss": 5.6906,
"mean_token_accuracy": 0.1730833023786545,
"num_tokens": 11075568.0,
"step": 4965
},
{
"entropy": 5.668417358398438,
"epoch": 0.5582074465097995,
"grad_norm": 1.890625,
"learning_rate": 0.0004977466594687057,
"loss": 5.6438,
"mean_token_accuracy": 0.1710194766521454,
"num_tokens": 11087078.0,
"step": 4970
},
{
"entropy": 5.870555448532104,
"epoch": 0.558769023417757,
"grad_norm": 1.453125,
"learning_rate": 0.0004977409880671039,
"loss": 5.7966,
"mean_token_accuracy": 0.16272210329771042,
"num_tokens": 11100655.0,
"step": 4975
},
{
"entropy": 5.721091938018799,
"epoch": 0.5593306003257146,
"grad_norm": 1.296875,
"learning_rate": 0.0004977353095733427,
"loss": 5.6956,
"mean_token_accuracy": 0.16899259388446808,
"num_tokens": 11110675.0,
"step": 4980
},
{
"entropy": 5.67507381439209,
"epoch": 0.5598921772336721,
"grad_norm": 1.265625,
"learning_rate": 0.0004977296239876026,
"loss": 5.6884,
"mean_token_accuracy": 0.17192643135786057,
"num_tokens": 11122337.0,
"step": 4985
},
{
"entropy": 5.79321255683899,
"epoch": 0.5604537541416297,
"grad_norm": 1.5234375,
"learning_rate": 0.000497723931310065,
"loss": 5.6087,
"mean_token_accuracy": 0.17079081982374192,
"num_tokens": 11132672.0,
"step": 4990
},
{
"entropy": 5.695679092407227,
"epoch": 0.5610153310495872,
"grad_norm": 1.2578125,
"learning_rate": 0.0004977182315409111,
"loss": 5.6217,
"mean_token_accuracy": 0.17549654692411423,
"num_tokens": 11143740.0,
"step": 4995
},
{
"entropy": 5.616427898406982,
"epoch": 0.5615769079575448,
"grad_norm": 1.4140625,
"learning_rate": 0.0004977125246803224,
"loss": 5.579,
"mean_token_accuracy": 0.17188211381435395,
"num_tokens": 11153722.0,
"step": 5000
},
{
"entropy": 5.693008613586426,
"epoch": 0.5621384848655023,
"grad_norm": 1.421875,
"learning_rate": 0.0004977068107284803,
"loss": 5.6892,
"mean_token_accuracy": 0.16788404434919357,
"num_tokens": 11165042.0,
"step": 5005
},
{
"entropy": 5.705210399627686,
"epoch": 0.5627000617734599,
"grad_norm": 1.3515625,
"learning_rate": 0.0004977010896855672,
"loss": 5.5626,
"mean_token_accuracy": 0.1795437827706337,
"num_tokens": 11176834.0,
"step": 5010
},
{
"entropy": 5.631637239456177,
"epoch": 0.5632616386814174,
"grad_norm": 1.2734375,
"learning_rate": 0.0004976953615517649,
"loss": 5.6463,
"mean_token_accuracy": 0.17071973234415055,
"num_tokens": 11188442.0,
"step": 5015
},
{
"entropy": 5.6706257343292235,
"epoch": 0.563823215589375,
"grad_norm": 1.2734375,
"learning_rate": 0.000497689626327256,
"loss": 5.5621,
"mean_token_accuracy": 0.17319224178791046,
"num_tokens": 11197855.0,
"step": 5020
},
{
"entropy": 5.786153745651245,
"epoch": 0.5643847924973325,
"grad_norm": 1.375,
"learning_rate": 0.000497683884012223,
"loss": 5.6557,
"mean_token_accuracy": 0.1667248785495758,
"num_tokens": 11209300.0,
"step": 5025
},
{
"entropy": 5.65827751159668,
"epoch": 0.5649463694052901,
"grad_norm": 1.375,
"learning_rate": 0.0004976781346068487,
"loss": 5.6641,
"mean_token_accuracy": 0.16813240200281143,
"num_tokens": 11222175.0,
"step": 5030
},
{
"entropy": 5.778843069076538,
"epoch": 0.5655079463132476,
"grad_norm": 1.53125,
"learning_rate": 0.0004976723781113163,
"loss": 5.6755,
"mean_token_accuracy": 0.16809964030981064,
"num_tokens": 11233277.0,
"step": 5035
},
{
"entropy": 5.749898529052734,
"epoch": 0.5660695232212052,
"grad_norm": 1.4140625,
"learning_rate": 0.0004976666145258089,
"loss": 5.6895,
"mean_token_accuracy": 0.17083817720413208,
"num_tokens": 11245241.0,
"step": 5040
},
{
"entropy": 5.763813304901123,
"epoch": 0.5666311001291627,
"grad_norm": 1.25,
"learning_rate": 0.0004976608438505103,
"loss": 5.7219,
"mean_token_accuracy": 0.16458606868982315,
"num_tokens": 11257359.0,
"step": 5045
},
{
"entropy": 5.654830598831177,
"epoch": 0.5671926770371203,
"grad_norm": 1.2265625,
"learning_rate": 0.0004976550660856041,
"loss": 5.6425,
"mean_token_accuracy": 0.17339012175798416,
"num_tokens": 11270651.0,
"step": 5050
},
{
"entropy": 5.670261240005493,
"epoch": 0.5677542539450777,
"grad_norm": 1.3671875,
"learning_rate": 0.000497649281231274,
"loss": 5.6022,
"mean_token_accuracy": 0.17513128519058227,
"num_tokens": 11280164.0,
"step": 5055
},
{
"entropy": 5.676502513885498,
"epoch": 0.5683158308530353,
"grad_norm": 1.2265625,
"learning_rate": 0.0004976434892877046,
"loss": 5.5416,
"mean_token_accuracy": 0.17301636785268784,
"num_tokens": 11290156.0,
"step": 5060
},
{
"entropy": 5.66754846572876,
"epoch": 0.5688774077609928,
"grad_norm": 1.3046875,
"learning_rate": 0.0004976376902550802,
"loss": 5.5987,
"mean_token_accuracy": 0.17832732051610947,
"num_tokens": 11301929.0,
"step": 5065
},
{
"entropy": 5.627131843566895,
"epoch": 0.5694389846689504,
"grad_norm": 1.2890625,
"learning_rate": 0.0004976318841335854,
"loss": 5.5398,
"mean_token_accuracy": 0.17666608840227127,
"num_tokens": 11312805.0,
"step": 5070
},
{
"entropy": 5.510635185241699,
"epoch": 0.5700005615769079,
"grad_norm": 1.34375,
"learning_rate": 0.0004976260709234049,
"loss": 5.4383,
"mean_token_accuracy": 0.1773639589548111,
"num_tokens": 11322316.0,
"step": 5075
},
{
"entropy": 5.6436632633209225,
"epoch": 0.5705621384848655,
"grad_norm": 1.3515625,
"learning_rate": 0.000497620250624724,
"loss": 5.6184,
"mean_token_accuracy": 0.17366870492696762,
"num_tokens": 11332994.0,
"step": 5080
},
{
"entropy": 5.7490044116973875,
"epoch": 0.571123715392823,
"grad_norm": 1.3046875,
"learning_rate": 0.000497614423237728,
"loss": 5.6038,
"mean_token_accuracy": 0.1674187570810318,
"num_tokens": 11344355.0,
"step": 5085
},
{
"entropy": 5.649996328353882,
"epoch": 0.5716852923007806,
"grad_norm": 1.3203125,
"learning_rate": 0.0004976085887626026,
"loss": 5.6445,
"mean_token_accuracy": 0.1675386518239975,
"num_tokens": 11356275.0,
"step": 5090
},
{
"entropy": 5.660901927947998,
"epoch": 0.5722468692087381,
"grad_norm": 1.1875,
"learning_rate": 0.0004976027471995331,
"loss": 5.6362,
"mean_token_accuracy": 0.167660254240036,
"num_tokens": 11368624.0,
"step": 5095
},
{
"entropy": 5.809105062484742,
"epoch": 0.5728084461166957,
"grad_norm": 1.2109375,
"learning_rate": 0.000497596898548706,
"loss": 5.7482,
"mean_token_accuracy": 0.16694559156894684,
"num_tokens": 11380520.0,
"step": 5100
},
{
"entropy": 5.670925569534302,
"epoch": 0.5733700230246532,
"grad_norm": 1.28125,
"learning_rate": 0.0004975910428103072,
"loss": 5.6605,
"mean_token_accuracy": 0.17296826988458633,
"num_tokens": 11392293.0,
"step": 5105
},
{
"entropy": 5.7538786888122555,
"epoch": 0.5739315999326108,
"grad_norm": 1.3828125,
"learning_rate": 0.0004975851799845233,
"loss": 5.7114,
"mean_token_accuracy": 0.16333020329475403,
"num_tokens": 11403866.0,
"step": 5110
},
{
"entropy": 5.672324466705322,
"epoch": 0.5744931768405683,
"grad_norm": 1.171875,
"learning_rate": 0.000497579310071541,
"loss": 5.5474,
"mean_token_accuracy": 0.1810181826353073,
"num_tokens": 11415334.0,
"step": 5115
},
{
"entropy": 5.647949743270874,
"epoch": 0.5750547537485259,
"grad_norm": 1.375,
"learning_rate": 0.0004975734330715471,
"loss": 5.6714,
"mean_token_accuracy": 0.17573580741882325,
"num_tokens": 11426840.0,
"step": 5120
},
{
"entropy": 5.806167507171631,
"epoch": 0.5756163306564834,
"grad_norm": 1.1484375,
"learning_rate": 0.0004975675489847287,
"loss": 5.8266,
"mean_token_accuracy": 0.1623709037899971,
"num_tokens": 11438497.0,
"step": 5125
},
{
"entropy": 5.711648225784302,
"epoch": 0.576177907564441,
"grad_norm": 1.5703125,
"learning_rate": 0.0004975616578112734,
"loss": 5.5403,
"mean_token_accuracy": 0.17929288148880004,
"num_tokens": 11448705.0,
"step": 5130
},
{
"entropy": 5.764576959609985,
"epoch": 0.5767394844723985,
"grad_norm": 1.3046875,
"learning_rate": 0.0004975557595513684,
"loss": 5.8046,
"mean_token_accuracy": 0.1625811442732811,
"num_tokens": 11460610.0,
"step": 5135
},
{
"entropy": 5.7110974311828615,
"epoch": 0.5773010613803561,
"grad_norm": 1.265625,
"learning_rate": 0.0004975498542052018,
"loss": 5.5736,
"mean_token_accuracy": 0.17118022739887237,
"num_tokens": 11471170.0,
"step": 5140
},
{
"entropy": 5.669892883300781,
"epoch": 0.5778626382883136,
"grad_norm": 1.2890625,
"learning_rate": 0.0004975439417729615,
"loss": 5.5579,
"mean_token_accuracy": 0.1798362225294113,
"num_tokens": 11482090.0,
"step": 5145
},
{
"entropy": 5.597491455078125,
"epoch": 0.5784242151962711,
"grad_norm": 1.2109375,
"learning_rate": 0.0004975380222548358,
"loss": 5.6582,
"mean_token_accuracy": 0.17230407446622847,
"num_tokens": 11493650.0,
"step": 5150
},
{
"entropy": 5.721773099899292,
"epoch": 0.5789857921042286,
"grad_norm": 1.84375,
"learning_rate": 0.0004975320956510131,
"loss": 5.6402,
"mean_token_accuracy": 0.16425044387578963,
"num_tokens": 11505903.0,
"step": 5155
},
{
"entropy": 5.798407459259034,
"epoch": 0.5795473690121862,
"grad_norm": 1.2265625,
"learning_rate": 0.0004975261619616822,
"loss": 5.6683,
"mean_token_accuracy": 0.1673250138759613,
"num_tokens": 11517194.0,
"step": 5160
},
{
"entropy": 5.702363348007202,
"epoch": 0.5801089459201437,
"grad_norm": 1.3359375,
"learning_rate": 0.0004975202211870321,
"loss": 5.7035,
"mean_token_accuracy": 0.17134629487991332,
"num_tokens": 11528683.0,
"step": 5165
},
{
"entropy": 5.740747785568237,
"epoch": 0.5806705228281013,
"grad_norm": 1.328125,
"learning_rate": 0.0004975142733272518,
"loss": 5.7214,
"mean_token_accuracy": 0.1703103080391884,
"num_tokens": 11541940.0,
"step": 5170
},
{
"entropy": 5.706987285614014,
"epoch": 0.5812320997360588,
"grad_norm": 1.203125,
"learning_rate": 0.0004975083183825306,
"loss": 5.5925,
"mean_token_accuracy": 0.17313284575939178,
"num_tokens": 11553045.0,
"step": 5175
},
{
"entropy": 5.617302179336548,
"epoch": 0.5817936766440164,
"grad_norm": 1.3203125,
"learning_rate": 0.0004975023563530584,
"loss": 5.5568,
"mean_token_accuracy": 0.17872895151376725,
"num_tokens": 11564401.0,
"step": 5180
},
{
"entropy": 5.647500705718994,
"epoch": 0.5823552535519739,
"grad_norm": 1.7890625,
"learning_rate": 0.0004974963872390249,
"loss": 5.6178,
"mean_token_accuracy": 0.1722454309463501,
"num_tokens": 11576424.0,
"step": 5185
},
{
"entropy": 5.722530794143677,
"epoch": 0.5829168304599315,
"grad_norm": 1.3203125,
"learning_rate": 0.00049749041104062,
"loss": 5.6224,
"mean_token_accuracy": 0.16687341630458832,
"num_tokens": 11587853.0,
"step": 5190
},
{
"entropy": 5.699846982955933,
"epoch": 0.583478407367889,
"grad_norm": 1.3984375,
"learning_rate": 0.0004974844277580343,
"loss": 5.6021,
"mean_token_accuracy": 0.1791071429848671,
"num_tokens": 11597938.0,
"step": 5195
},
{
"entropy": 5.802369403839111,
"epoch": 0.5840399842758466,
"grad_norm": 1.484375,
"learning_rate": 0.0004974784373914579,
"loss": 5.7831,
"mean_token_accuracy": 0.16783083081245423,
"num_tokens": 11610077.0,
"step": 5200
},
{
"entropy": 5.722145366668701,
"epoch": 0.5846015611838041,
"grad_norm": 1.515625,
"learning_rate": 0.0004974724399410818,
"loss": 5.6463,
"mean_token_accuracy": 0.17028158605098725,
"num_tokens": 11622079.0,
"step": 5205
},
{
"entropy": 5.6010784149169925,
"epoch": 0.5851631380917617,
"grad_norm": 1.671875,
"learning_rate": 0.0004974664354070971,
"loss": 5.5748,
"mean_token_accuracy": 0.18004426509141921,
"num_tokens": 11633047.0,
"step": 5210
},
{
"entropy": 5.612564897537231,
"epoch": 0.5857247149997192,
"grad_norm": 1.28125,
"learning_rate": 0.0004974604237896947,
"loss": 5.5226,
"mean_token_accuracy": 0.17414324283599852,
"num_tokens": 11643858.0,
"step": 5215
},
{
"entropy": 5.617132186889648,
"epoch": 0.5862862919076768,
"grad_norm": 1.28125,
"learning_rate": 0.0004974544050890662,
"loss": 5.5507,
"mean_token_accuracy": 0.1753794565796852,
"num_tokens": 11654550.0,
"step": 5220
},
{
"entropy": 5.649725914001465,
"epoch": 0.5868478688156343,
"grad_norm": 1.2421875,
"learning_rate": 0.000497448379305403,
"loss": 5.5981,
"mean_token_accuracy": 0.16864759474992752,
"num_tokens": 11665787.0,
"step": 5225
},
{
"entropy": 5.775894212722778,
"epoch": 0.5874094457235919,
"grad_norm": 1.234375,
"learning_rate": 0.0004974423464388973,
"loss": 5.6991,
"mean_token_accuracy": 0.167301045358181,
"num_tokens": 11677717.0,
"step": 5230
},
{
"entropy": 5.6776011943817135,
"epoch": 0.5879710226315494,
"grad_norm": 1.515625,
"learning_rate": 0.000497436306489741,
"loss": 5.5956,
"mean_token_accuracy": 0.17099462747573851,
"num_tokens": 11687714.0,
"step": 5235
},
{
"entropy": 5.611685132980346,
"epoch": 0.588532599539507,
"grad_norm": 1.28125,
"learning_rate": 0.0004974302594581264,
"loss": 5.4834,
"mean_token_accuracy": 0.17616420686244966,
"num_tokens": 11698777.0,
"step": 5240
},
{
"entropy": 5.600326061248779,
"epoch": 0.5890941764474645,
"grad_norm": 1.2578125,
"learning_rate": 0.0004974242053442461,
"loss": 5.5777,
"mean_token_accuracy": 0.17680131494998932,
"num_tokens": 11710922.0,
"step": 5245
},
{
"entropy": 5.638203573226929,
"epoch": 0.589655753355422,
"grad_norm": 1.28125,
"learning_rate": 0.0004974181441482927,
"loss": 5.6221,
"mean_token_accuracy": 0.17195345014333724,
"num_tokens": 11721450.0,
"step": 5250
},
{
"entropy": 5.615335893630982,
"epoch": 0.5902173302633795,
"grad_norm": 1.296875,
"learning_rate": 0.0004974120758704595,
"loss": 5.5255,
"mean_token_accuracy": 0.17599792331457137,
"num_tokens": 11732458.0,
"step": 5255
},
{
"entropy": 5.622411012649536,
"epoch": 0.5907789071713371,
"grad_norm": 1.328125,
"learning_rate": 0.0004974060005109394,
"loss": 5.5136,
"mean_token_accuracy": 0.17872669249773027,
"num_tokens": 11742649.0,
"step": 5260
},
{
"entropy": 5.5806248664855955,
"epoch": 0.5913404840792946,
"grad_norm": 1.3125,
"learning_rate": 0.000497399918069926,
"loss": 5.5499,
"mean_token_accuracy": 0.17637165039777755,
"num_tokens": 11753811.0,
"step": 5265
},
{
"entropy": 5.676913595199585,
"epoch": 0.5919020609872522,
"grad_norm": 1.7109375,
"learning_rate": 0.0004973938285476131,
"loss": 5.6426,
"mean_token_accuracy": 0.16756807267665863,
"num_tokens": 11763965.0,
"step": 5270
},
{
"entropy": 5.78911247253418,
"epoch": 0.5924636378952097,
"grad_norm": 1.4921875,
"learning_rate": 0.0004973877319441943,
"loss": 5.6517,
"mean_token_accuracy": 0.17557948976755142,
"num_tokens": 11774323.0,
"step": 5275
},
{
"entropy": 5.662372732162476,
"epoch": 0.5930252148031673,
"grad_norm": 1.546875,
"learning_rate": 0.0004973816282598639,
"loss": 5.6248,
"mean_token_accuracy": 0.16901510804891587,
"num_tokens": 11786157.0,
"step": 5280
},
{
"entropy": 5.6589898586273195,
"epoch": 0.5935867917111248,
"grad_norm": 1.4375,
"learning_rate": 0.0004973755174948162,
"loss": 5.6031,
"mean_token_accuracy": 0.17968236804008483,
"num_tokens": 11797344.0,
"step": 5285
},
{
"entropy": 5.562589979171753,
"epoch": 0.5941483686190824,
"grad_norm": 1.3828125,
"learning_rate": 0.0004973693996492458,
"loss": 5.5474,
"mean_token_accuracy": 0.17363488227128981,
"num_tokens": 11808149.0,
"step": 5290
},
{
"entropy": 5.658795547485352,
"epoch": 0.5947099455270399,
"grad_norm": 1.25,
"learning_rate": 0.0004973632747233475,
"loss": 5.5614,
"mean_token_accuracy": 0.17789566218852998,
"num_tokens": 11820203.0,
"step": 5295
},
{
"entropy": 5.691214466094971,
"epoch": 0.5952715224349975,
"grad_norm": 1.5859375,
"learning_rate": 0.0004973571427173163,
"loss": 5.6368,
"mean_token_accuracy": 0.1706070989370346,
"num_tokens": 11830172.0,
"step": 5300
},
{
"entropy": 5.72310791015625,
"epoch": 0.595833099342955,
"grad_norm": 1.609375,
"learning_rate": 0.0004973510036313473,
"loss": 5.6123,
"mean_token_accuracy": 0.17963551431894303,
"num_tokens": 11840203.0,
"step": 5305
},
{
"entropy": 5.703700017929077,
"epoch": 0.5963946762509126,
"grad_norm": 1.28125,
"learning_rate": 0.0004973448574656362,
"loss": 5.5642,
"mean_token_accuracy": 0.17044158577919005,
"num_tokens": 11851002.0,
"step": 5310
},
{
"entropy": 5.638322210311889,
"epoch": 0.5969562531588701,
"grad_norm": 2.3125,
"learning_rate": 0.0004973387042203786,
"loss": 5.557,
"mean_token_accuracy": 0.17067076861858368,
"num_tokens": 11861559.0,
"step": 5315
},
{
"entropy": 5.650593423843384,
"epoch": 0.5975178300668277,
"grad_norm": 1.3359375,
"learning_rate": 0.0004973325438957704,
"loss": 5.6352,
"mean_token_accuracy": 0.16927488148212433,
"num_tokens": 11872535.0,
"step": 5320
},
{
"entropy": 5.730974721908569,
"epoch": 0.5980794069747852,
"grad_norm": 1.4609375,
"learning_rate": 0.0004973263764920079,
"loss": 5.6381,
"mean_token_accuracy": 0.17522387355566024,
"num_tokens": 11884308.0,
"step": 5325
},
{
"entropy": 5.763652229309082,
"epoch": 0.5986409838827428,
"grad_norm": 2.03125,
"learning_rate": 0.0004973202020092871,
"loss": 5.7064,
"mean_token_accuracy": 0.17014836370944977,
"num_tokens": 11896136.0,
"step": 5330
},
{
"entropy": 5.711992073059082,
"epoch": 0.5992025607907003,
"grad_norm": 1.4453125,
"learning_rate": 0.000497314020447805,
"loss": 5.5416,
"mean_token_accuracy": 0.18177102655172347,
"num_tokens": 11906497.0,
"step": 5335
},
{
"entropy": 5.714601945877075,
"epoch": 0.5997641376986579,
"grad_norm": 1.8671875,
"learning_rate": 0.0004973078318077583,
"loss": 5.657,
"mean_token_accuracy": 0.1648639425635338,
"num_tokens": 11917872.0,
"step": 5340
},
{
"entropy": 5.7148108959198,
"epoch": 0.6003257146066153,
"grad_norm": 1.75,
"learning_rate": 0.000497301636089344,
"loss": 5.7067,
"mean_token_accuracy": 0.16739431023597717,
"num_tokens": 11929727.0,
"step": 5345
},
{
"entropy": 5.703443956375122,
"epoch": 0.6008872915145729,
"grad_norm": 1.203125,
"learning_rate": 0.0004972954332927592,
"loss": 5.4647,
"mean_token_accuracy": 0.17234138697385787,
"num_tokens": 11939996.0,
"step": 5350
},
{
"entropy": 5.6297016620635985,
"epoch": 0.6014488684225304,
"grad_norm": 2.296875,
"learning_rate": 0.0004972892234182018,
"loss": 5.5533,
"mean_token_accuracy": 0.1777931809425354,
"num_tokens": 11950683.0,
"step": 5355
},
{
"entropy": 5.572098159790039,
"epoch": 0.602010445330488,
"grad_norm": 1.265625,
"learning_rate": 0.0004972830064658692,
"loss": 5.5607,
"mean_token_accuracy": 0.16803841143846512,
"num_tokens": 11962447.0,
"step": 5360
},
{
"entropy": 5.670172309875488,
"epoch": 0.6025720222384455,
"grad_norm": 1.4140625,
"learning_rate": 0.0004972767824359596,
"loss": 5.6426,
"mean_token_accuracy": 0.17171704918146133,
"num_tokens": 11972740.0,
"step": 5365
},
{
"entropy": 5.653109931945801,
"epoch": 0.6031335991464031,
"grad_norm": 1.3984375,
"learning_rate": 0.0004972705513286709,
"loss": 5.5351,
"mean_token_accuracy": 0.17810094207525254,
"num_tokens": 11983292.0,
"step": 5370
},
{
"entropy": 5.783536672592163,
"epoch": 0.6036951760543606,
"grad_norm": 1.265625,
"learning_rate": 0.0004972643131442017,
"loss": 5.7516,
"mean_token_accuracy": 0.1622177094221115,
"num_tokens": 11996384.0,
"step": 5375
},
{
"entropy": 5.783068609237671,
"epoch": 0.6042567529623182,
"grad_norm": 1.5078125,
"learning_rate": 0.0004972580678827506,
"loss": 5.6997,
"mean_token_accuracy": 0.17653866857290268,
"num_tokens": 12006938.0,
"step": 5380
},
{
"entropy": 5.556166553497315,
"epoch": 0.6048183298702757,
"grad_norm": 1.375,
"learning_rate": 0.0004972518155445163,
"loss": 5.4786,
"mean_token_accuracy": 0.17628339529037476,
"num_tokens": 12018729.0,
"step": 5385
},
{
"entropy": 5.81696310043335,
"epoch": 0.6053799067782333,
"grad_norm": 1.4140625,
"learning_rate": 0.0004972455561296982,
"loss": 5.829,
"mean_token_accuracy": 0.16385156512260438,
"num_tokens": 12029625.0,
"step": 5390
},
{
"entropy": 5.728468894958496,
"epoch": 0.6059414836861908,
"grad_norm": 1.265625,
"learning_rate": 0.0004972392896384952,
"loss": 5.5372,
"mean_token_accuracy": 0.17295820266008377,
"num_tokens": 12040243.0,
"step": 5395
},
{
"entropy": 5.746753835678101,
"epoch": 0.6065030605941484,
"grad_norm": 1.359375,
"learning_rate": 0.0004972330160711071,
"loss": 5.8157,
"mean_token_accuracy": 0.15855337381362916,
"num_tokens": 12051961.0,
"step": 5400
},
{
"entropy": 5.7441887855529785,
"epoch": 0.6070646375021059,
"grad_norm": 1.328125,
"learning_rate": 0.0004972267354277336,
"loss": 5.6844,
"mean_token_accuracy": 0.17277398109436035,
"num_tokens": 12063969.0,
"step": 5405
},
{
"entropy": 5.754958724975586,
"epoch": 0.6076262144100635,
"grad_norm": 1.3828125,
"learning_rate": 0.0004972204477085746,
"loss": 5.572,
"mean_token_accuracy": 0.1725847214460373,
"num_tokens": 12075682.0,
"step": 5410
},
{
"entropy": 5.71474814414978,
"epoch": 0.608187791318021,
"grad_norm": 1.375,
"learning_rate": 0.0004972141529138303,
"loss": 5.7005,
"mean_token_accuracy": 0.17008328288793564,
"num_tokens": 12086595.0,
"step": 5415
},
{
"entropy": 5.657503509521485,
"epoch": 0.6087493682259786,
"grad_norm": 1.46875,
"learning_rate": 0.0004972078510437013,
"loss": 5.5951,
"mean_token_accuracy": 0.17330475896596909,
"num_tokens": 12099421.0,
"step": 5420
},
{
"entropy": 5.732693338394165,
"epoch": 0.6093109451339361,
"grad_norm": 1.34375,
"learning_rate": 0.0004972015420983882,
"loss": 5.6888,
"mean_token_accuracy": 0.1706646278500557,
"num_tokens": 12109680.0,
"step": 5425
},
{
"entropy": 5.620195960998535,
"epoch": 0.6098725220418937,
"grad_norm": 1.2578125,
"learning_rate": 0.0004971952260780915,
"loss": 5.5687,
"mean_token_accuracy": 0.18063763082027434,
"num_tokens": 12121196.0,
"step": 5430
},
{
"entropy": 5.634686851501465,
"epoch": 0.6104340989498512,
"grad_norm": 1.1875,
"learning_rate": 0.000497188902983013,
"loss": 5.5136,
"mean_token_accuracy": 0.1639615297317505,
"num_tokens": 12132262.0,
"step": 5435
},
{
"entropy": 5.709107875823975,
"epoch": 0.6109956758578087,
"grad_norm": 1.28125,
"learning_rate": 0.0004971825728133535,
"loss": 5.6939,
"mean_token_accuracy": 0.17179398387670516,
"num_tokens": 12143747.0,
"step": 5440
},
{
"entropy": 5.8577793598175045,
"epoch": 0.6115572527657662,
"grad_norm": 1.1484375,
"learning_rate": 0.0004971762355693146,
"loss": 5.8013,
"mean_token_accuracy": 0.16207447946071624,
"num_tokens": 12155887.0,
"step": 5445
},
{
"entropy": 5.644887685775757,
"epoch": 0.6121188296737238,
"grad_norm": 1.3671875,
"learning_rate": 0.0004971698912510983,
"loss": 5.5719,
"mean_token_accuracy": 0.16924232244491577,
"num_tokens": 12166718.0,
"step": 5450
},
{
"entropy": 5.689836120605468,
"epoch": 0.6126804065816813,
"grad_norm": 1.1953125,
"learning_rate": 0.0004971635398589066,
"loss": 5.7254,
"mean_token_accuracy": 0.1709606260061264,
"num_tokens": 12178540.0,
"step": 5455
},
{
"entropy": 5.835622835159302,
"epoch": 0.6132419834896389,
"grad_norm": 1.5234375,
"learning_rate": 0.0004971571813929415,
"loss": 5.7545,
"mean_token_accuracy": 0.16316340267658233,
"num_tokens": 12189820.0,
"step": 5460
},
{
"entropy": 5.733452463150025,
"epoch": 0.6138035603975964,
"grad_norm": 1.3046875,
"learning_rate": 0.0004971508158534056,
"loss": 5.6429,
"mean_token_accuracy": 0.17269133329391478,
"num_tokens": 12201679.0,
"step": 5465
},
{
"entropy": 5.692891263961792,
"epoch": 0.614365137305554,
"grad_norm": 1.2890625,
"learning_rate": 0.0004971444432405015,
"loss": 5.6662,
"mean_token_accuracy": 0.17074246257543563,
"num_tokens": 12213793.0,
"step": 5470
},
{
"entropy": 5.832134485244751,
"epoch": 0.6149267142135115,
"grad_norm": 1.2109375,
"learning_rate": 0.0004971380635544323,
"loss": 5.754,
"mean_token_accuracy": 0.1672576442360878,
"num_tokens": 12224519.0,
"step": 5475
},
{
"entropy": 5.815476036071777,
"epoch": 0.6154882911214691,
"grad_norm": 1.2109375,
"learning_rate": 0.0004971316767954009,
"loss": 5.742,
"mean_token_accuracy": 0.16240453571081162,
"num_tokens": 12236401.0,
"step": 5480
},
{
"entropy": 5.749634885787964,
"epoch": 0.6160498680294266,
"grad_norm": 1.3359375,
"learning_rate": 0.0004971252829636108,
"loss": 5.7604,
"mean_token_accuracy": 0.16711125820875167,
"num_tokens": 12247670.0,
"step": 5485
},
{
"entropy": 5.706438398361206,
"epoch": 0.6166114449373842,
"grad_norm": 1.21875,
"learning_rate": 0.0004971188820592656,
"loss": 5.6283,
"mean_token_accuracy": 0.17230897694826125,
"num_tokens": 12258877.0,
"step": 5490
},
{
"entropy": 5.6592991828918455,
"epoch": 0.6171730218453417,
"grad_norm": 1.4140625,
"learning_rate": 0.000497112474082569,
"loss": 5.6033,
"mean_token_accuracy": 0.17209488153457642,
"num_tokens": 12268507.0,
"step": 5495
},
{
"entropy": 5.677366638183594,
"epoch": 0.6177345987532993,
"grad_norm": 1.4140625,
"learning_rate": 0.000497106059033725,
"loss": 5.6053,
"mean_token_accuracy": 0.16927239298820496,
"num_tokens": 12280420.0,
"step": 5500
},
{
"entropy": 5.638323354721069,
"epoch": 0.6182961756612568,
"grad_norm": 1.640625,
"learning_rate": 0.0004970996369129381,
"loss": 5.5799,
"mean_token_accuracy": 0.17210918068885803,
"num_tokens": 12290876.0,
"step": 5505
},
{
"entropy": 5.6833113670349125,
"epoch": 0.6188577525692144,
"grad_norm": 1.3984375,
"learning_rate": 0.0004970932077204125,
"loss": 5.5084,
"mean_token_accuracy": 0.17870903611183167,
"num_tokens": 12301831.0,
"step": 5510
},
{
"entropy": 5.584318208694458,
"epoch": 0.6194193294771719,
"grad_norm": 1.390625,
"learning_rate": 0.000497086771456353,
"loss": 5.5357,
"mean_token_accuracy": 0.17946565747261048,
"num_tokens": 12312678.0,
"step": 5515
},
{
"entropy": 5.786611175537109,
"epoch": 0.6199809063851295,
"grad_norm": 1.2578125,
"learning_rate": 0.0004970803281209646,
"loss": 5.6981,
"mean_token_accuracy": 0.1650458663702011,
"num_tokens": 12322905.0,
"step": 5520
},
{
"entropy": 5.739599418640137,
"epoch": 0.620542483293087,
"grad_norm": 1.46875,
"learning_rate": 0.0004970738777144524,
"loss": 5.5689,
"mean_token_accuracy": 0.16803901195526122,
"num_tokens": 12333091.0,
"step": 5525
},
{
"entropy": 5.6710591316223145,
"epoch": 0.6211040602010446,
"grad_norm": 1.3203125,
"learning_rate": 0.0004970674202370218,
"loss": 5.5732,
"mean_token_accuracy": 0.17317660748958588,
"num_tokens": 12346310.0,
"step": 5530
},
{
"entropy": 5.625415563583374,
"epoch": 0.621665637109002,
"grad_norm": 1.1953125,
"learning_rate": 0.0004970609556888785,
"loss": 5.5051,
"mean_token_accuracy": 0.17522240728139876,
"num_tokens": 12357138.0,
"step": 5535
},
{
"entropy": 5.753187751770019,
"epoch": 0.6222272140169596,
"grad_norm": 1.2578125,
"learning_rate": 0.0004970544840702282,
"loss": 5.5931,
"mean_token_accuracy": 0.17838772386312485,
"num_tokens": 12367382.0,
"step": 5540
},
{
"entropy": 5.619511842727661,
"epoch": 0.6227887909249171,
"grad_norm": 1.4296875,
"learning_rate": 0.0004970480053812771,
"loss": 5.5927,
"mean_token_accuracy": 0.16931483745574952,
"num_tokens": 12380357.0,
"step": 5545
},
{
"entropy": 5.7702712059021,
"epoch": 0.6233503678328747,
"grad_norm": 1.515625,
"learning_rate": 0.0004970415196222312,
"loss": 5.7789,
"mean_token_accuracy": 0.162600776553154,
"num_tokens": 12392187.0,
"step": 5550
},
{
"entropy": 5.72232723236084,
"epoch": 0.6239119447408322,
"grad_norm": 1.6328125,
"learning_rate": 0.0004970350267932973,
"loss": 5.625,
"mean_token_accuracy": 0.17594385296106338,
"num_tokens": 12402924.0,
"step": 5555
},
{
"entropy": 5.760821294784546,
"epoch": 0.6244735216487898,
"grad_norm": 1.5625,
"learning_rate": 0.000497028526894682,
"loss": 5.7509,
"mean_token_accuracy": 0.16636327356100084,
"num_tokens": 12413554.0,
"step": 5560
},
{
"entropy": 5.710971355438232,
"epoch": 0.6250350985567473,
"grad_norm": 1.8359375,
"learning_rate": 0.0004970220199265923,
"loss": 5.6217,
"mean_token_accuracy": 0.17457333505153655,
"num_tokens": 12424525.0,
"step": 5565
},
{
"entropy": 5.704411888122559,
"epoch": 0.6255966754647049,
"grad_norm": 1.6875,
"learning_rate": 0.0004970155058892354,
"loss": 5.6217,
"mean_token_accuracy": 0.17579136788845062,
"num_tokens": 12436405.0,
"step": 5570
},
{
"entropy": 5.80294828414917,
"epoch": 0.6261582523726624,
"grad_norm": 5.4375,
"learning_rate": 0.0004970089847828187,
"loss": 5.6829,
"mean_token_accuracy": 0.16928793489933014,
"num_tokens": 12446909.0,
"step": 5575
},
{
"entropy": 5.805694341659546,
"epoch": 0.62671982928062,
"grad_norm": 1.265625,
"learning_rate": 0.0004970024566075497,
"loss": 5.6831,
"mean_token_accuracy": 0.1733838587999344,
"num_tokens": 12458739.0,
"step": 5580
},
{
"entropy": 5.667700862884521,
"epoch": 0.6272814061885775,
"grad_norm": 1.546875,
"learning_rate": 0.0004969959213636365,
"loss": 5.5879,
"mean_token_accuracy": 0.17088814824819565,
"num_tokens": 12469529.0,
"step": 5585
},
{
"entropy": 5.7064825057983395,
"epoch": 0.6278429830965351,
"grad_norm": 1.953125,
"learning_rate": 0.0004969893790512871,
"loss": 5.6361,
"mean_token_accuracy": 0.16559593081474305,
"num_tokens": 12480067.0,
"step": 5590
},
{
"entropy": 5.683680152893066,
"epoch": 0.6284045600044926,
"grad_norm": 1.328125,
"learning_rate": 0.0004969828296707097,
"loss": 5.53,
"mean_token_accuracy": 0.17746374309062957,
"num_tokens": 12491043.0,
"step": 5595
},
{
"entropy": 5.591855859756469,
"epoch": 0.6289661369124502,
"grad_norm": 1.3125,
"learning_rate": 0.0004969762732221129,
"loss": 5.4596,
"mean_token_accuracy": 0.17833520472049713,
"num_tokens": 12501852.0,
"step": 5600
},
{
"entropy": 5.721845388412476,
"epoch": 0.6295277138204077,
"grad_norm": 1.625,
"learning_rate": 0.0004969697097057054,
"loss": 5.7432,
"mean_token_accuracy": 0.16794373393058776,
"num_tokens": 12512696.0,
"step": 5605
},
{
"entropy": 5.7294896125793455,
"epoch": 0.6300892907283653,
"grad_norm": 1.6484375,
"learning_rate": 0.0004969631391216964,
"loss": 5.6126,
"mean_token_accuracy": 0.17836637794971466,
"num_tokens": 12523643.0,
"step": 5610
},
{
"entropy": 5.675888442993164,
"epoch": 0.6306508676363228,
"grad_norm": 1.2734375,
"learning_rate": 0.0004969565614702948,
"loss": 5.5904,
"mean_token_accuracy": 0.1714102581143379,
"num_tokens": 12534827.0,
"step": 5615
},
{
"entropy": 5.523678112030029,
"epoch": 0.6312124445442804,
"grad_norm": 1.515625,
"learning_rate": 0.0004969499767517104,
"loss": 5.4672,
"mean_token_accuracy": 0.1736081287264824,
"num_tokens": 12545977.0,
"step": 5620
},
{
"entropy": 5.569380378723144,
"epoch": 0.6317740214522379,
"grad_norm": 1.515625,
"learning_rate": 0.0004969433849661525,
"loss": 5.5281,
"mean_token_accuracy": 0.17083040922880172,
"num_tokens": 12555890.0,
"step": 5625
},
{
"entropy": 5.705742740631104,
"epoch": 0.6323355983601955,
"grad_norm": 1.4375,
"learning_rate": 0.0004969367861138313,
"loss": 5.5944,
"mean_token_accuracy": 0.17643035650253297,
"num_tokens": 12566412.0,
"step": 5630
},
{
"entropy": 5.7321854591369625,
"epoch": 0.6328971752681529,
"grad_norm": 1.328125,
"learning_rate": 0.0004969301801949565,
"loss": 5.6266,
"mean_token_accuracy": 0.1705393522977829,
"num_tokens": 12577175.0,
"step": 5635
},
{
"entropy": 5.609008073806763,
"epoch": 0.6334587521761105,
"grad_norm": 1.140625,
"learning_rate": 0.000496923567209739,
"loss": 5.5396,
"mean_token_accuracy": 0.17606472074985505,
"num_tokens": 12588760.0,
"step": 5640
},
{
"entropy": 5.662323617935181,
"epoch": 0.634020329084068,
"grad_norm": 1.59375,
"learning_rate": 0.0004969169471583889,
"loss": 5.612,
"mean_token_accuracy": 0.1708120733499527,
"num_tokens": 12599899.0,
"step": 5645
},
{
"entropy": 5.7322752475738525,
"epoch": 0.6345819059920256,
"grad_norm": 1.1875,
"learning_rate": 0.0004969103200411171,
"loss": 5.5901,
"mean_token_accuracy": 0.17937593311071395,
"num_tokens": 12610836.0,
"step": 5650
},
{
"entropy": 5.586485290527344,
"epoch": 0.6351434828999831,
"grad_norm": 1.390625,
"learning_rate": 0.0004969036858581346,
"loss": 5.5278,
"mean_token_accuracy": 0.17668813914060594,
"num_tokens": 12621855.0,
"step": 5655
},
{
"entropy": 5.653785037994385,
"epoch": 0.6357050598079407,
"grad_norm": 1.40625,
"learning_rate": 0.0004968970446096527,
"loss": 5.6423,
"mean_token_accuracy": 0.1710375264286995,
"num_tokens": 12632964.0,
"step": 5660
},
{
"entropy": 5.623979997634888,
"epoch": 0.6362666367158982,
"grad_norm": 1.421875,
"learning_rate": 0.000496890396295883,
"loss": 5.4752,
"mean_token_accuracy": 0.18113864064216614,
"num_tokens": 12643717.0,
"step": 5665
},
{
"entropy": 5.643584251403809,
"epoch": 0.6368282136238558,
"grad_norm": 1.34375,
"learning_rate": 0.0004968837409170368,
"loss": 5.5943,
"mean_token_accuracy": 0.16968384236097336,
"num_tokens": 12654491.0,
"step": 5670
},
{
"entropy": 5.676835727691651,
"epoch": 0.6373897905318133,
"grad_norm": 1.5234375,
"learning_rate": 0.0004968770784733263,
"loss": 5.6498,
"mean_token_accuracy": 0.1708448812365532,
"num_tokens": 12666128.0,
"step": 5675
},
{
"entropy": 5.773913192749023,
"epoch": 0.6379513674397709,
"grad_norm": 1.2578125,
"learning_rate": 0.0004968704089649636,
"loss": 5.7227,
"mean_token_accuracy": 0.17005614936351776,
"num_tokens": 12677957.0,
"step": 5680
},
{
"entropy": 5.719657611846924,
"epoch": 0.6385129443477284,
"grad_norm": 1.375,
"learning_rate": 0.0004968637323921611,
"loss": 5.5524,
"mean_token_accuracy": 0.1731327548623085,
"num_tokens": 12687964.0,
"step": 5685
},
{
"entropy": 5.5747850894927975,
"epoch": 0.639074521255686,
"grad_norm": 1.65625,
"learning_rate": 0.0004968570487551312,
"loss": 5.5632,
"mean_token_accuracy": 0.16955589205026628,
"num_tokens": 12698805.0,
"step": 5690
},
{
"entropy": 5.706884717941284,
"epoch": 0.6396360981636435,
"grad_norm": 2.421875,
"learning_rate": 0.0004968503580540869,
"loss": 5.5716,
"mean_token_accuracy": 0.17181707918643951,
"num_tokens": 12709942.0,
"step": 5695
},
{
"entropy": 5.744397354125977,
"epoch": 0.6401976750716011,
"grad_norm": 1.7109375,
"learning_rate": 0.0004968436602892411,
"loss": 5.6626,
"mean_token_accuracy": 0.1701844498515129,
"num_tokens": 12721856.0,
"step": 5700
},
{
"entropy": 5.586921834945679,
"epoch": 0.6407592519795586,
"grad_norm": 1.3359375,
"learning_rate": 0.0004968369554608071,
"loss": 5.4852,
"mean_token_accuracy": 0.17421021014451982,
"num_tokens": 12732619.0,
"step": 5705
},
{
"entropy": 5.64688630104065,
"epoch": 0.6413208288875162,
"grad_norm": 1.625,
"learning_rate": 0.0004968302435689985,
"loss": 5.5731,
"mean_token_accuracy": 0.176441790163517,
"num_tokens": 12743478.0,
"step": 5710
},
{
"entropy": 5.65359525680542,
"epoch": 0.6418824057954737,
"grad_norm": 1.375,
"learning_rate": 0.0004968235246140288,
"loss": 5.537,
"mean_token_accuracy": 0.16710498183965683,
"num_tokens": 12754761.0,
"step": 5715
},
{
"entropy": 5.721187162399292,
"epoch": 0.6424439827034313,
"grad_norm": 1.359375,
"learning_rate": 0.0004968167985961121,
"loss": 5.6531,
"mean_token_accuracy": 0.1721176728606224,
"num_tokens": 12765803.0,
"step": 5720
},
{
"entropy": 5.686595392227173,
"epoch": 0.6430055596113888,
"grad_norm": 1.21875,
"learning_rate": 0.0004968100655154625,
"loss": 5.6614,
"mean_token_accuracy": 0.17171961218118667,
"num_tokens": 12777882.0,
"step": 5725
},
{
"entropy": 5.823597717285156,
"epoch": 0.6435671365193463,
"grad_norm": 1.2421875,
"learning_rate": 0.0004968033253722943,
"loss": 5.6844,
"mean_token_accuracy": 0.16798087805509568,
"num_tokens": 12788268.0,
"step": 5730
},
{
"entropy": 5.711064672470092,
"epoch": 0.6441287134273038,
"grad_norm": 1.4453125,
"learning_rate": 0.0004967965781668223,
"loss": 5.6828,
"mean_token_accuracy": 0.17314098328351973,
"num_tokens": 12800506.0,
"step": 5735
},
{
"entropy": 5.635964393615723,
"epoch": 0.6446902903352614,
"grad_norm": 1.515625,
"learning_rate": 0.0004967898238992613,
"loss": 5.6394,
"mean_token_accuracy": 0.17088195234537124,
"num_tokens": 12813040.0,
"step": 5740
},
{
"entropy": 5.81028904914856,
"epoch": 0.6452518672432189,
"grad_norm": 1.28125,
"learning_rate": 0.0004967830625698262,
"loss": 5.6703,
"mean_token_accuracy": 0.17704896926879882,
"num_tokens": 12825534.0,
"step": 5745
},
{
"entropy": 5.680642986297608,
"epoch": 0.6458134441511765,
"grad_norm": 1.2421875,
"learning_rate": 0.0004967762941787323,
"loss": 5.5697,
"mean_token_accuracy": 0.17251336574554443,
"num_tokens": 12835882.0,
"step": 5750
},
{
"entropy": 5.697614574432373,
"epoch": 0.646375021059134,
"grad_norm": 1.421875,
"learning_rate": 0.0004967695187261952,
"loss": 5.7571,
"mean_token_accuracy": 0.16053135842084884,
"num_tokens": 12847663.0,
"step": 5755
},
{
"entropy": 5.942152643203736,
"epoch": 0.6469365979670916,
"grad_norm": 1.296875,
"learning_rate": 0.0004967627362124306,
"loss": 5.8859,
"mean_token_accuracy": 0.16270481050014496,
"num_tokens": 12860789.0,
"step": 5760
},
{
"entropy": 5.733904647827148,
"epoch": 0.6474981748750491,
"grad_norm": 1.7890625,
"learning_rate": 0.0004967559466376546,
"loss": 5.6171,
"mean_token_accuracy": 0.17376980632543565,
"num_tokens": 12872509.0,
"step": 5765
},
{
"entropy": 5.660063743591309,
"epoch": 0.6480597517830067,
"grad_norm": 1.4453125,
"learning_rate": 0.0004967491500020831,
"loss": 5.6107,
"mean_token_accuracy": 0.17281572371721268,
"num_tokens": 12884321.0,
"step": 5770
},
{
"entropy": 5.559579133987427,
"epoch": 0.6486213286909642,
"grad_norm": 1.265625,
"learning_rate": 0.0004967423463059328,
"loss": 5.4945,
"mean_token_accuracy": 0.17920683622360228,
"num_tokens": 12894559.0,
"step": 5775
},
{
"entropy": 5.673180055618286,
"epoch": 0.6491829055989218,
"grad_norm": 1.3671875,
"learning_rate": 0.0004967355355494201,
"loss": 5.5893,
"mean_token_accuracy": 0.17977261841297149,
"num_tokens": 12904934.0,
"step": 5780
},
{
"entropy": 5.857377004623413,
"epoch": 0.6497444825068793,
"grad_norm": 1.5859375,
"learning_rate": 0.000496728717732762,
"loss": 5.7201,
"mean_token_accuracy": 0.16673576533794404,
"num_tokens": 12916551.0,
"step": 5785
},
{
"entropy": 5.686304044723511,
"epoch": 0.6503060594148369,
"grad_norm": 1.546875,
"learning_rate": 0.0004967218928561755,
"loss": 5.6135,
"mean_token_accuracy": 0.16970376223325728,
"num_tokens": 12928097.0,
"step": 5790
},
{
"entropy": 5.666470575332641,
"epoch": 0.6508676363227944,
"grad_norm": 1.1796875,
"learning_rate": 0.0004967150609198779,
"loss": 5.4941,
"mean_token_accuracy": 0.18141568899154664,
"num_tokens": 12939161.0,
"step": 5795
},
{
"entropy": 5.540322732925415,
"epoch": 0.651429213230752,
"grad_norm": 1.484375,
"learning_rate": 0.0004967082219240869,
"loss": 5.5465,
"mean_token_accuracy": 0.1759266436100006,
"num_tokens": 12950862.0,
"step": 5800
},
{
"entropy": 5.640594577789306,
"epoch": 0.6519907901387095,
"grad_norm": 1.3125,
"learning_rate": 0.0004967013758690199,
"loss": 5.5915,
"mean_token_accuracy": 0.16954418420791625,
"num_tokens": 12962333.0,
"step": 5805
},
{
"entropy": 5.721122884750367,
"epoch": 0.6525523670466671,
"grad_norm": 1.359375,
"learning_rate": 0.0004966945227548953,
"loss": 5.5625,
"mean_token_accuracy": 0.17373017519712447,
"num_tokens": 12973136.0,
"step": 5810
},
{
"entropy": 5.635897636413574,
"epoch": 0.6531139439546246,
"grad_norm": 1.234375,
"learning_rate": 0.0004966876625819309,
"loss": 5.5946,
"mean_token_accuracy": 0.17273072004318238,
"num_tokens": 12983602.0,
"step": 5815
},
{
"entropy": 5.61157808303833,
"epoch": 0.6536755208625822,
"grad_norm": 1.3984375,
"learning_rate": 0.0004966807953503456,
"loss": 5.5054,
"mean_token_accuracy": 0.1837313652038574,
"num_tokens": 12994129.0,
"step": 5820
},
{
"entropy": 5.6429441452026365,
"epoch": 0.6542370977705396,
"grad_norm": 1.4921875,
"learning_rate": 0.0004966739210603578,
"loss": 5.5769,
"mean_token_accuracy": 0.1716706186532974,
"num_tokens": 13005071.0,
"step": 5825
},
{
"entropy": 5.76044659614563,
"epoch": 0.6547986746784972,
"grad_norm": 1.296875,
"learning_rate": 0.0004966670397121862,
"loss": 5.7498,
"mean_token_accuracy": 0.1712135151028633,
"num_tokens": 13016202.0,
"step": 5830
},
{
"entropy": 5.831915235519409,
"epoch": 0.6553602515864547,
"grad_norm": 1.359375,
"learning_rate": 0.0004966601513060502,
"loss": 5.7317,
"mean_token_accuracy": 0.16146057546138765,
"num_tokens": 13028468.0,
"step": 5835
},
{
"entropy": 5.698105955123902,
"epoch": 0.6559218284944123,
"grad_norm": 1.234375,
"learning_rate": 0.000496653255842169,
"loss": 5.6234,
"mean_token_accuracy": 0.16889994144439696,
"num_tokens": 13039667.0,
"step": 5840
},
{
"entropy": 5.6738808155059814,
"epoch": 0.6564834054023698,
"grad_norm": 1.6953125,
"learning_rate": 0.0004966463533207621,
"loss": 5.6781,
"mean_token_accuracy": 0.16560440063476561,
"num_tokens": 13051583.0,
"step": 5845
},
{
"entropy": 5.713420963287353,
"epoch": 0.6570449823103274,
"grad_norm": 1.359375,
"learning_rate": 0.0004966394437420493,
"loss": 5.6184,
"mean_token_accuracy": 0.17039913088083267,
"num_tokens": 13062401.0,
"step": 5850
},
{
"entropy": 5.610844469070434,
"epoch": 0.6576065592182849,
"grad_norm": 1.4296875,
"learning_rate": 0.0004966325271062508,
"loss": 5.5139,
"mean_token_accuracy": 0.17710702270269393,
"num_tokens": 13074219.0,
"step": 5855
},
{
"entropy": 5.713991641998291,
"epoch": 0.6581681361262425,
"grad_norm": 1.453125,
"learning_rate": 0.0004966256034135866,
"loss": 5.7229,
"mean_token_accuracy": 0.16896682530641555,
"num_tokens": 13086073.0,
"step": 5860
},
{
"entropy": 5.725924634933472,
"epoch": 0.6587297130342,
"grad_norm": 1.3046875,
"learning_rate": 0.0004966186726642772,
"loss": 5.6251,
"mean_token_accuracy": 0.17155906260013581,
"num_tokens": 13096340.0,
"step": 5865
},
{
"entropy": 5.686580228805542,
"epoch": 0.6592912899421576,
"grad_norm": 1.265625,
"learning_rate": 0.0004966117348585434,
"loss": 5.5971,
"mean_token_accuracy": 0.1746459811925888,
"num_tokens": 13107273.0,
"step": 5870
},
{
"entropy": 5.642148494720459,
"epoch": 0.6598528668501151,
"grad_norm": 1.2890625,
"learning_rate": 0.0004966047899966059,
"loss": 5.5705,
"mean_token_accuracy": 0.16856825947761536,
"num_tokens": 13117388.0,
"step": 5875
},
{
"entropy": 5.6900513648986815,
"epoch": 0.6604144437580727,
"grad_norm": 1.25,
"learning_rate": 0.000496597838078686,
"loss": 5.6765,
"mean_token_accuracy": 0.16927792578935624,
"num_tokens": 13128720.0,
"step": 5880
},
{
"entropy": 5.757421445846558,
"epoch": 0.6609760206660302,
"grad_norm": 1.6484375,
"learning_rate": 0.000496590879105005,
"loss": 5.5965,
"mean_token_accuracy": 0.1739147871732712,
"num_tokens": 13139405.0,
"step": 5885
},
{
"entropy": 5.706152772903442,
"epoch": 0.6615375975739878,
"grad_norm": 1.4765625,
"learning_rate": 0.0004965839130757845,
"loss": 5.6841,
"mean_token_accuracy": 0.17259052246809006,
"num_tokens": 13151408.0,
"step": 5890
},
{
"entropy": 5.705246162414551,
"epoch": 0.6620991744819453,
"grad_norm": 1.609375,
"learning_rate": 0.0004965769399912462,
"loss": 5.589,
"mean_token_accuracy": 0.1736095055937767,
"num_tokens": 13162592.0,
"step": 5895
},
{
"entropy": 5.750041055679321,
"epoch": 0.6626607513899029,
"grad_norm": 1.265625,
"learning_rate": 0.0004965699598516122,
"loss": 5.6238,
"mean_token_accuracy": 0.1780148133635521,
"num_tokens": 13173040.0,
"step": 5900
},
{
"entropy": 5.678125524520874,
"epoch": 0.6632223282978604,
"grad_norm": 1.421875,
"learning_rate": 0.0004965629726571047,
"loss": 5.7039,
"mean_token_accuracy": 0.16705054491758348,
"num_tokens": 13185475.0,
"step": 5905
},
{
"entropy": 5.660798645019531,
"epoch": 0.663783905205818,
"grad_norm": 1.9921875,
"learning_rate": 0.0004965559784079463,
"loss": 5.5707,
"mean_token_accuracy": 0.1771893247961998,
"num_tokens": 13196481.0,
"step": 5910
},
{
"entropy": 5.6448406219482425,
"epoch": 0.6643454821137755,
"grad_norm": 1.3359375,
"learning_rate": 0.0004965489771043596,
"loss": 5.463,
"mean_token_accuracy": 0.18122804462909697,
"num_tokens": 13207947.0,
"step": 5915
},
{
"entropy": 5.534987926483154,
"epoch": 0.664907059021733,
"grad_norm": 1.65625,
"learning_rate": 0.0004965419687465676,
"loss": 5.5088,
"mean_token_accuracy": 0.18186939060688018,
"num_tokens": 13219392.0,
"step": 5920
},
{
"entropy": 5.5524852752685545,
"epoch": 0.6654686359296905,
"grad_norm": 2.0,
"learning_rate": 0.0004965349533347934,
"loss": 5.5606,
"mean_token_accuracy": 0.17652862668037414,
"num_tokens": 13231556.0,
"step": 5925
},
{
"entropy": 5.682732057571411,
"epoch": 0.6660302128376481,
"grad_norm": 1.5859375,
"learning_rate": 0.0004965279308692602,
"loss": 5.4856,
"mean_token_accuracy": 0.1784966066479683,
"num_tokens": 13241287.0,
"step": 5930
},
{
"entropy": 5.590028429031372,
"epoch": 0.6665917897456056,
"grad_norm": 1.3515625,
"learning_rate": 0.0004965209013501919,
"loss": 5.5347,
"mean_token_accuracy": 0.16716251224279405,
"num_tokens": 13254120.0,
"step": 5935
},
{
"entropy": 5.640761947631836,
"epoch": 0.6671533666535632,
"grad_norm": 1.6484375,
"learning_rate": 0.0004965138647778122,
"loss": 5.617,
"mean_token_accuracy": 0.17144743204116822,
"num_tokens": 13265745.0,
"step": 5940
},
{
"entropy": 5.777652835845947,
"epoch": 0.6677149435615207,
"grad_norm": 1.3203125,
"learning_rate": 0.000496506821152345,
"loss": 5.674,
"mean_token_accuracy": 0.17016586810350418,
"num_tokens": 13277791.0,
"step": 5945
},
{
"entropy": 5.653607511520386,
"epoch": 0.6682765204694783,
"grad_norm": 1.359375,
"learning_rate": 0.0004964997704740147,
"loss": 5.53,
"mean_token_accuracy": 0.1750830665230751,
"num_tokens": 13288078.0,
"step": 5950
},
{
"entropy": 5.609050178527832,
"epoch": 0.6688380973774358,
"grad_norm": 1.546875,
"learning_rate": 0.000496492712743046,
"loss": 5.5828,
"mean_token_accuracy": 0.17747070640325546,
"num_tokens": 13300547.0,
"step": 5955
},
{
"entropy": 5.695092391967774,
"epoch": 0.6693996742853934,
"grad_norm": 2.6875,
"learning_rate": 0.0004964856479596632,
"loss": 5.6278,
"mean_token_accuracy": 0.17592556327581405,
"num_tokens": 13310339.0,
"step": 5960
},
{
"entropy": 5.7985865592956545,
"epoch": 0.6699612511933509,
"grad_norm": 1.359375,
"learning_rate": 0.0004964785761240915,
"loss": 5.6191,
"mean_token_accuracy": 0.17247435599565505,
"num_tokens": 13321294.0,
"step": 5965
},
{
"entropy": 5.734959840774536,
"epoch": 0.6705228281013085,
"grad_norm": 1.390625,
"learning_rate": 0.0004964714972365562,
"loss": 5.6395,
"mean_token_accuracy": 0.1715475797653198,
"num_tokens": 13332608.0,
"step": 5970
},
{
"entropy": 5.684270429611206,
"epoch": 0.671084405009266,
"grad_norm": 1.3203125,
"learning_rate": 0.0004964644112972824,
"loss": 5.5927,
"mean_token_accuracy": 0.18088890761137008,
"num_tokens": 13344169.0,
"step": 5975
},
{
"entropy": 5.598596572875977,
"epoch": 0.6716459819172236,
"grad_norm": 1.3828125,
"learning_rate": 0.0004964573183064959,
"loss": 5.5537,
"mean_token_accuracy": 0.17526524662971496,
"num_tokens": 13355713.0,
"step": 5980
},
{
"entropy": 5.701221179962158,
"epoch": 0.6722075588251811,
"grad_norm": 1.625,
"learning_rate": 0.0004964502182644225,
"loss": 5.6369,
"mean_token_accuracy": 0.17090377658605577,
"num_tokens": 13366644.0,
"step": 5985
},
{
"entropy": 5.704200220108032,
"epoch": 0.6727691357331387,
"grad_norm": 1.7109375,
"learning_rate": 0.0004964431111712883,
"loss": 5.6047,
"mean_token_accuracy": 0.17175456434488295,
"num_tokens": 13378125.0,
"step": 5990
},
{
"entropy": 5.683547067642212,
"epoch": 0.6733307126410962,
"grad_norm": 1.6484375,
"learning_rate": 0.0004964359970273195,
"loss": 5.6559,
"mean_token_accuracy": 0.17148393243551255,
"num_tokens": 13389681.0,
"step": 5995
},
{
"entropy": 5.775100183486939,
"epoch": 0.6738922895490538,
"grad_norm": 1.6640625,
"learning_rate": 0.0004964288758327426,
"loss": 5.635,
"mean_token_accuracy": 0.17012319415807725,
"num_tokens": 13399814.0,
"step": 6000
},
{
"epoch": 0.6738922895490538,
"eval_entropy": 5.647960238767894,
"eval_loss": 5.6351118087768555,
"eval_mean_token_accuracy": 0.17846342232890247,
"eval_num_tokens": 13399814.0,
"eval_runtime": 23.7606,
"eval_samples_per_second": 1305.227,
"eval_steps_per_second": 163.169,
"step": 6000
},
{
"entropy": 5.789413118362427,
"epoch": 0.6744538664570113,
"grad_norm": 1.4140625,
"learning_rate": 0.0004964217475877846,
"loss": 5.651,
"mean_token_accuracy": 0.17309489250183105,
"num_tokens": 13410614.0,
"step": 6005
},
{
"entropy": 5.647662210464477,
"epoch": 0.6750154433649689,
"grad_norm": 1.703125,
"learning_rate": 0.000496414612292672,
"loss": 5.6595,
"mean_token_accuracy": 0.16725679486989975,
"num_tokens": 13421392.0,
"step": 6010
},
{
"entropy": 5.707165765762329,
"epoch": 0.6755770202729264,
"grad_norm": 1.34375,
"learning_rate": 0.0004964074699476325,
"loss": 5.6993,
"mean_token_accuracy": 0.16803635954856871,
"num_tokens": 13432004.0,
"step": 6015
},
{
"entropy": 5.708346605300903,
"epoch": 0.6761385971808839,
"grad_norm": 1.421875,
"learning_rate": 0.0004964003205528933,
"loss": 5.5228,
"mean_token_accuracy": 0.17715829163789748,
"num_tokens": 13442450.0,
"step": 6020
},
{
"entropy": 5.69212908744812,
"epoch": 0.6767001740888414,
"grad_norm": 1.4140625,
"learning_rate": 0.0004963931641086819,
"loss": 5.5887,
"mean_token_accuracy": 0.1682564824819565,
"num_tokens": 13453848.0,
"step": 6025
},
{
"entropy": 5.67762393951416,
"epoch": 0.677261750996799,
"grad_norm": 1.3125,
"learning_rate": 0.0004963860006152262,
"loss": 5.6024,
"mean_token_accuracy": 0.1763468623161316,
"num_tokens": 13465119.0,
"step": 6030
},
{
"entropy": 5.6405839920043945,
"epoch": 0.6778233279047565,
"grad_norm": 1.3515625,
"learning_rate": 0.0004963788300727546,
"loss": 5.5839,
"mean_token_accuracy": 0.17347716689109802,
"num_tokens": 13475937.0,
"step": 6035
},
{
"entropy": 5.612260532379151,
"epoch": 0.6783849048127141,
"grad_norm": 1.4296875,
"learning_rate": 0.0004963716524814951,
"loss": 5.5539,
"mean_token_accuracy": 0.18009347170591355,
"num_tokens": 13487430.0,
"step": 6040
},
{
"entropy": 5.696366596221924,
"epoch": 0.6789464817206716,
"grad_norm": 1.5546875,
"learning_rate": 0.0004963644678416763,
"loss": 5.5599,
"mean_token_accuracy": 0.17814573496580124,
"num_tokens": 13499631.0,
"step": 6045
},
{
"entropy": 5.728988552093506,
"epoch": 0.6795080586286292,
"grad_norm": 1.3984375,
"learning_rate": 0.000496357276153527,
"loss": 5.6381,
"mean_token_accuracy": 0.1726953700184822,
"num_tokens": 13511753.0,
"step": 6050
},
{
"entropy": 5.665334320068359,
"epoch": 0.6800696355365867,
"grad_norm": 1.28125,
"learning_rate": 0.0004963500774172763,
"loss": 5.6919,
"mean_token_accuracy": 0.1653744973242283,
"num_tokens": 13524228.0,
"step": 6055
},
{
"entropy": 5.7868529796600345,
"epoch": 0.6806312124445443,
"grad_norm": 1.3515625,
"learning_rate": 0.000496342871633153,
"loss": 5.7068,
"mean_token_accuracy": 0.17238008230924606,
"num_tokens": 13535700.0,
"step": 6060
},
{
"entropy": 5.744654893875122,
"epoch": 0.6811927893525018,
"grad_norm": 1.734375,
"learning_rate": 0.0004963356588013871,
"loss": 5.685,
"mean_token_accuracy": 0.16285565346479416,
"num_tokens": 13547337.0,
"step": 6065
},
{
"entropy": 5.597080421447754,
"epoch": 0.6817543662604594,
"grad_norm": 1.328125,
"learning_rate": 0.0004963284389222078,
"loss": 5.5463,
"mean_token_accuracy": 0.17075700163841248,
"num_tokens": 13558350.0,
"step": 6070
},
{
"entropy": 5.663588571548462,
"epoch": 0.6823159431684169,
"grad_norm": 1.390625,
"learning_rate": 0.0004963212119958453,
"loss": 5.5867,
"mean_token_accuracy": 0.17460559010505677,
"num_tokens": 13569230.0,
"step": 6075
},
{
"entropy": 5.70565881729126,
"epoch": 0.6828775200763745,
"grad_norm": 1.4296875,
"learning_rate": 0.0004963139780225295,
"loss": 5.6842,
"mean_token_accuracy": 0.17543704062700272,
"num_tokens": 13580502.0,
"step": 6080
},
{
"entropy": 5.731264972686768,
"epoch": 0.683439096984332,
"grad_norm": 1.484375,
"learning_rate": 0.0004963067370024909,
"loss": 5.5818,
"mean_token_accuracy": 0.1747697487473488,
"num_tokens": 13591524.0,
"step": 6085
},
{
"entropy": 5.758881998062134,
"epoch": 0.6840006738922896,
"grad_norm": 1.53125,
"learning_rate": 0.0004962994889359598,
"loss": 5.6044,
"mean_token_accuracy": 0.1821832165122032,
"num_tokens": 13602432.0,
"step": 6090
},
{
"entropy": 5.5485320568084715,
"epoch": 0.6845622508002471,
"grad_norm": 1.453125,
"learning_rate": 0.0004962922338231673,
"loss": 5.543,
"mean_token_accuracy": 0.18179477304220198,
"num_tokens": 13613052.0,
"step": 6095
},
{
"entropy": 5.6250978946685795,
"epoch": 0.6851238277082047,
"grad_norm": 2.171875,
"learning_rate": 0.0004962849716643441,
"loss": 5.5718,
"mean_token_accuracy": 0.1751675471663475,
"num_tokens": 13623961.0,
"step": 6100
},
{
"entropy": 5.637742233276367,
"epoch": 0.6856854046161622,
"grad_norm": 1.5,
"learning_rate": 0.0004962777024597216,
"loss": 5.5527,
"mean_token_accuracy": 0.17164352983236314,
"num_tokens": 13635285.0,
"step": 6105
},
{
"entropy": 5.814977836608887,
"epoch": 0.6862469815241198,
"grad_norm": 1.5546875,
"learning_rate": 0.0004962704262095314,
"loss": 5.7185,
"mean_token_accuracy": 0.167195363342762,
"num_tokens": 13647720.0,
"step": 6110
},
{
"entropy": 5.621568918228149,
"epoch": 0.6868085584320772,
"grad_norm": 1.4453125,
"learning_rate": 0.0004962631429140048,
"loss": 5.5554,
"mean_token_accuracy": 0.17185777127742768,
"num_tokens": 13659726.0,
"step": 6115
},
{
"entropy": 5.7803551197052006,
"epoch": 0.6873701353400348,
"grad_norm": 1.6328125,
"learning_rate": 0.000496255852573374,
"loss": 5.7396,
"mean_token_accuracy": 0.1632569059729576,
"num_tokens": 13670332.0,
"step": 6120
},
{
"entropy": 5.797141695022583,
"epoch": 0.6879317122479923,
"grad_norm": 1.390625,
"learning_rate": 0.0004962485551878711,
"loss": 5.6845,
"mean_token_accuracy": 0.16198990643024444,
"num_tokens": 13682419.0,
"step": 6125
},
{
"entropy": 5.556501197814941,
"epoch": 0.6884932891559499,
"grad_norm": 1.421875,
"learning_rate": 0.0004962412507577283,
"loss": 5.4642,
"mean_token_accuracy": 0.17773663401603698,
"num_tokens": 13692698.0,
"step": 6130
},
{
"entropy": 5.610977458953857,
"epoch": 0.6890548660639074,
"grad_norm": 1.3828125,
"learning_rate": 0.0004962339392831783,
"loss": 5.6492,
"mean_token_accuracy": 0.17161332368850707,
"num_tokens": 13704791.0,
"step": 6135
},
{
"entropy": 5.684554576873779,
"epoch": 0.689616442971865,
"grad_norm": 1.3359375,
"learning_rate": 0.000496226620764454,
"loss": 5.5797,
"mean_token_accuracy": 0.17454966008663178,
"num_tokens": 13715860.0,
"step": 6140
},
{
"entropy": 5.682456588745117,
"epoch": 0.6901780198798225,
"grad_norm": 1.4296875,
"learning_rate": 0.000496219295201788,
"loss": 5.5488,
"mean_token_accuracy": 0.1810535654425621,
"num_tokens": 13727029.0,
"step": 6145
},
{
"entropy": 5.633379650115967,
"epoch": 0.6907395967877801,
"grad_norm": 1.3359375,
"learning_rate": 0.0004962119625954141,
"loss": 5.5336,
"mean_token_accuracy": 0.1844476118683815,
"num_tokens": 13737168.0,
"step": 6150
},
{
"entropy": 5.7368792533874515,
"epoch": 0.6913011736957376,
"grad_norm": 1.421875,
"learning_rate": 0.0004962046229455653,
"loss": 5.6819,
"mean_token_accuracy": 0.16525454223155975,
"num_tokens": 13749828.0,
"step": 6155
},
{
"entropy": 5.662187194824218,
"epoch": 0.6918627506036952,
"grad_norm": 1.515625,
"learning_rate": 0.0004961972762524757,
"loss": 5.5014,
"mean_token_accuracy": 0.17295316755771636,
"num_tokens": 13761617.0,
"step": 6160
},
{
"entropy": 5.7150428771972654,
"epoch": 0.6924243275116527,
"grad_norm": 1.4375,
"learning_rate": 0.000496189922516379,
"loss": 5.564,
"mean_token_accuracy": 0.18064149469137192,
"num_tokens": 13772721.0,
"step": 6165
},
{
"entropy": 5.5662336349487305,
"epoch": 0.6929859044196103,
"grad_norm": 1.671875,
"learning_rate": 0.0004961825617375093,
"loss": 5.5491,
"mean_token_accuracy": 0.17482539415359497,
"num_tokens": 13783215.0,
"step": 6170
},
{
"entropy": 5.597888994216919,
"epoch": 0.6935474813275678,
"grad_norm": 1.453125,
"learning_rate": 0.0004961751939161011,
"loss": 5.4926,
"mean_token_accuracy": 0.18007427752017974,
"num_tokens": 13793496.0,
"step": 6175
},
{
"entropy": 5.562454175949097,
"epoch": 0.6941090582355254,
"grad_norm": 1.390625,
"learning_rate": 0.000496167819052389,
"loss": 5.4132,
"mean_token_accuracy": 0.1751859113574028,
"num_tokens": 13802818.0,
"step": 6180
},
{
"entropy": 5.776867961883545,
"epoch": 0.6946706351434829,
"grad_norm": 1.5078125,
"learning_rate": 0.0004961604371466077,
"loss": 5.7271,
"mean_token_accuracy": 0.16933899670839309,
"num_tokens": 13814449.0,
"step": 6185
},
{
"entropy": 5.715918731689453,
"epoch": 0.6952322120514405,
"grad_norm": 1.4921875,
"learning_rate": 0.0004961530481989923,
"loss": 5.6517,
"mean_token_accuracy": 0.1751192480325699,
"num_tokens": 13825369.0,
"step": 6190
},
{
"entropy": 5.599141407012939,
"epoch": 0.695793788959398,
"grad_norm": 1.28125,
"learning_rate": 0.0004961456522097782,
"loss": 5.5078,
"mean_token_accuracy": 0.1696367621421814,
"num_tokens": 13837992.0,
"step": 6195
},
{
"entropy": 5.72159218788147,
"epoch": 0.6963553658673556,
"grad_norm": 1.890625,
"learning_rate": 0.0004961382491792006,
"loss": 5.613,
"mean_token_accuracy": 0.16797615885734557,
"num_tokens": 13849771.0,
"step": 6200
},
{
"entropy": 5.684537553787232,
"epoch": 0.6969169427753131,
"grad_norm": 1.3359375,
"learning_rate": 0.0004961308391074953,
"loss": 5.5806,
"mean_token_accuracy": 0.17560651004314423,
"num_tokens": 13861997.0,
"step": 6205
},
{
"entropy": 5.743185567855835,
"epoch": 0.6974785196832706,
"grad_norm": 1.265625,
"learning_rate": 0.0004961234219948986,
"loss": 5.6715,
"mean_token_accuracy": 0.16763371229171753,
"num_tokens": 13872972.0,
"step": 6210
},
{
"entropy": 5.574690151214599,
"epoch": 0.6980400965912281,
"grad_norm": 1.5,
"learning_rate": 0.0004961159978416461,
"loss": 5.4557,
"mean_token_accuracy": 0.17770340740680696,
"num_tokens": 13883139.0,
"step": 6215
},
{
"entropy": 5.674945592880249,
"epoch": 0.6986016734991857,
"grad_norm": 1.34375,
"learning_rate": 0.0004961085666479746,
"loss": 5.6238,
"mean_token_accuracy": 0.17407322973012923,
"num_tokens": 13894326.0,
"step": 6220
},
{
"entropy": 5.744280576705933,
"epoch": 0.6991632504071432,
"grad_norm": 1.2578125,
"learning_rate": 0.0004961011284141206,
"loss": 5.6452,
"mean_token_accuracy": 0.17508044242858886,
"num_tokens": 13905525.0,
"step": 6225
},
{
"entropy": 5.6775284767150875,
"epoch": 0.6997248273151008,
"grad_norm": 1.5625,
"learning_rate": 0.0004960936831403209,
"loss": 5.5743,
"mean_token_accuracy": 0.17416172325611115,
"num_tokens": 13915973.0,
"step": 6230
},
{
"entropy": 5.575125598907471,
"epoch": 0.7002864042230583,
"grad_norm": 1.5234375,
"learning_rate": 0.0004960862308268126,
"loss": 5.3974,
"mean_token_accuracy": 0.18169076591730118,
"num_tokens": 13926380.0,
"step": 6235
},
{
"entropy": 5.637649297714233,
"epoch": 0.7008479811310159,
"grad_norm": 1.609375,
"learning_rate": 0.0004960787714738329,
"loss": 5.5939,
"mean_token_accuracy": 0.16871747374534607,
"num_tokens": 13938523.0,
"step": 6240
},
{
"entropy": 5.621913194656372,
"epoch": 0.7014095580389734,
"grad_norm": 1.40625,
"learning_rate": 0.0004960713050816194,
"loss": 5.571,
"mean_token_accuracy": 0.17842501550912857,
"num_tokens": 13949343.0,
"step": 6245
},
{
"entropy": 5.690118980407715,
"epoch": 0.701971134946931,
"grad_norm": 1.390625,
"learning_rate": 0.0004960638316504098,
"loss": 5.6894,
"mean_token_accuracy": 0.17289069592952727,
"num_tokens": 13960056.0,
"step": 6250
},
{
"entropy": 5.82127161026001,
"epoch": 0.7025327118548885,
"grad_norm": 1.5234375,
"learning_rate": 0.0004960563511804421,
"loss": 5.6529,
"mean_token_accuracy": 0.1700647547841072,
"num_tokens": 13971932.0,
"step": 6255
},
{
"entropy": 5.788155841827392,
"epoch": 0.7030942887628461,
"grad_norm": 1.3046875,
"learning_rate": 0.0004960488636719545,
"loss": 5.649,
"mean_token_accuracy": 0.1751314803957939,
"num_tokens": 13981992.0,
"step": 6260
},
{
"entropy": 5.719943380355835,
"epoch": 0.7036558656708036,
"grad_norm": 1.6328125,
"learning_rate": 0.0004960413691251852,
"loss": 5.6,
"mean_token_accuracy": 0.1771599233150482,
"num_tokens": 13992215.0,
"step": 6265
},
{
"entropy": 5.715781736373901,
"epoch": 0.7042174425787612,
"grad_norm": 1.8515625,
"learning_rate": 0.0004960338675403731,
"loss": 5.4964,
"mean_token_accuracy": 0.1881141945719719,
"num_tokens": 14002689.0,
"step": 6270
},
{
"entropy": 5.681948709487915,
"epoch": 0.7047790194867187,
"grad_norm": 1.65625,
"learning_rate": 0.0004960263589177568,
"loss": 5.6175,
"mean_token_accuracy": 0.17318082898855208,
"num_tokens": 14014951.0,
"step": 6275
},
{
"entropy": 5.591769695281982,
"epoch": 0.7053405963946763,
"grad_norm": 1.78125,
"learning_rate": 0.0004960188432575757,
"loss": 5.5601,
"mean_token_accuracy": 0.17742443233728408,
"num_tokens": 14025502.0,
"step": 6280
},
{
"entropy": 5.762725782394409,
"epoch": 0.7059021733026338,
"grad_norm": 1.3359375,
"learning_rate": 0.0004960113205600688,
"loss": 5.6034,
"mean_token_accuracy": 0.17833426892757415,
"num_tokens": 14036576.0,
"step": 6285
},
{
"entropy": 5.7332977771759035,
"epoch": 0.7064637502105914,
"grad_norm": 1.390625,
"learning_rate": 0.0004960037908254757,
"loss": 5.6201,
"mean_token_accuracy": 0.17569523304700851,
"num_tokens": 14046493.0,
"step": 6290
},
{
"entropy": 5.747225427627564,
"epoch": 0.7070253271185489,
"grad_norm": 1.2578125,
"learning_rate": 0.0004959962540540361,
"loss": 5.6728,
"mean_token_accuracy": 0.17666876018047334,
"num_tokens": 14057297.0,
"step": 6295
},
{
"entropy": 5.593193626403808,
"epoch": 0.7075869040265065,
"grad_norm": 1.2578125,
"learning_rate": 0.0004959887102459903,
"loss": 5.4619,
"mean_token_accuracy": 0.17642207443714142,
"num_tokens": 14068947.0,
"step": 6300
},
{
"entropy": 5.627945518493652,
"epoch": 0.7081484809344639,
"grad_norm": 1.4375,
"learning_rate": 0.0004959811594015781,
"loss": 5.5507,
"mean_token_accuracy": 0.1774741992354393,
"num_tokens": 14078979.0,
"step": 6305
},
{
"entropy": 5.598028516769409,
"epoch": 0.7087100578424215,
"grad_norm": 1.3515625,
"learning_rate": 0.0004959736015210401,
"loss": 5.4939,
"mean_token_accuracy": 0.17994166165590286,
"num_tokens": 14089785.0,
"step": 6310
},
{
"entropy": 5.66406216621399,
"epoch": 0.709271634750379,
"grad_norm": 1.3984375,
"learning_rate": 0.0004959660366046169,
"loss": 5.5925,
"mean_token_accuracy": 0.17430199831724166,
"num_tokens": 14099921.0,
"step": 6315
},
{
"entropy": 5.625072622299195,
"epoch": 0.7098332116583366,
"grad_norm": 1.2578125,
"learning_rate": 0.0004959584646525494,
"loss": 5.4987,
"mean_token_accuracy": 0.1763535723090172,
"num_tokens": 14110965.0,
"step": 6320
},
{
"entropy": 5.653204345703125,
"epoch": 0.7103947885662941,
"grad_norm": 1.25,
"learning_rate": 0.0004959508856650788,
"loss": 5.6477,
"mean_token_accuracy": 0.1728035107254982,
"num_tokens": 14122409.0,
"step": 6325
},
{
"entropy": 5.685785961151123,
"epoch": 0.7109563654742517,
"grad_norm": 1.4140625,
"learning_rate": 0.0004959432996424463,
"loss": 5.584,
"mean_token_accuracy": 0.17243881225585939,
"num_tokens": 14134017.0,
"step": 6330
},
{
"entropy": 5.703707075119018,
"epoch": 0.7115179423822092,
"grad_norm": 1.3203125,
"learning_rate": 0.0004959357065848932,
"loss": 5.6929,
"mean_token_accuracy": 0.17194198369979857,
"num_tokens": 14144843.0,
"step": 6335
},
{
"entropy": 5.6564782619476315,
"epoch": 0.7120795192901668,
"grad_norm": 1.4765625,
"learning_rate": 0.0004959281064926618,
"loss": 5.548,
"mean_token_accuracy": 0.17492249310016633,
"num_tokens": 14155944.0,
"step": 6340
},
{
"entropy": 5.683899736404419,
"epoch": 0.7126410961981243,
"grad_norm": 1.2421875,
"learning_rate": 0.0004959204993659936,
"loss": 5.6132,
"mean_token_accuracy": 0.17579726427793502,
"num_tokens": 14167064.0,
"step": 6345
},
{
"entropy": 5.719380855560303,
"epoch": 0.7132026731060819,
"grad_norm": 1.453125,
"learning_rate": 0.0004959128852051311,
"loss": 5.619,
"mean_token_accuracy": 0.17238176017999648,
"num_tokens": 14178583.0,
"step": 6350
},
{
"entropy": 5.62568769454956,
"epoch": 0.7137642500140394,
"grad_norm": 1.3046875,
"learning_rate": 0.0004959052640103167,
"loss": 5.52,
"mean_token_accuracy": 0.17467750161886214,
"num_tokens": 14188752.0,
"step": 6355
},
{
"entropy": 5.60446982383728,
"epoch": 0.714325826921997,
"grad_norm": 1.4921875,
"learning_rate": 0.000495897635781793,
"loss": 5.513,
"mean_token_accuracy": 0.17688787132501602,
"num_tokens": 14199962.0,
"step": 6360
},
{
"entropy": 5.645568132400513,
"epoch": 0.7148874038299545,
"grad_norm": 1.8125,
"learning_rate": 0.0004958900005198029,
"loss": 5.5254,
"mean_token_accuracy": 0.17567192763090134,
"num_tokens": 14209877.0,
"step": 6365
},
{
"entropy": 5.63578691482544,
"epoch": 0.7154489807379121,
"grad_norm": 2.3125,
"learning_rate": 0.0004958823582245894,
"loss": 5.5938,
"mean_token_accuracy": 0.1752375215291977,
"num_tokens": 14220672.0,
"step": 6370
},
{
"entropy": 5.781293249130249,
"epoch": 0.7160105576458696,
"grad_norm": 1.5,
"learning_rate": 0.0004958747088963961,
"loss": 5.5871,
"mean_token_accuracy": 0.1764242798089981,
"num_tokens": 14232348.0,
"step": 6375
},
{
"entropy": 5.636292600631714,
"epoch": 0.7165721345538272,
"grad_norm": 1.6015625,
"learning_rate": 0.0004958670525354663,
"loss": 5.6076,
"mean_token_accuracy": 0.16509898155927658,
"num_tokens": 14243985.0,
"step": 6380
},
{
"entropy": 5.575510883331299,
"epoch": 0.7171337114617847,
"grad_norm": 1.203125,
"learning_rate": 0.000495859389142044,
"loss": 5.4575,
"mean_token_accuracy": 0.1797981232404709,
"num_tokens": 14253548.0,
"step": 6385
},
{
"entropy": 5.7398152351379395,
"epoch": 0.7176952883697423,
"grad_norm": 1.453125,
"learning_rate": 0.000495851718716373,
"loss": 5.6135,
"mean_token_accuracy": 0.17006600499153138,
"num_tokens": 14264757.0,
"step": 6390
},
{
"entropy": 5.601899766921997,
"epoch": 0.7182568652776998,
"grad_norm": 1.1875,
"learning_rate": 0.0004958440412586977,
"loss": 5.517,
"mean_token_accuracy": 0.18686744570732117,
"num_tokens": 14275946.0,
"step": 6395
},
{
"entropy": 5.585464763641357,
"epoch": 0.7188184421856574,
"grad_norm": 1.359375,
"learning_rate": 0.0004958363567692624,
"loss": 5.5876,
"mean_token_accuracy": 0.16820885688066484,
"num_tokens": 14287006.0,
"step": 6400
},
{
"entropy": 5.779633092880249,
"epoch": 0.7193800190936148,
"grad_norm": 1.296875,
"learning_rate": 0.0004958286652483118,
"loss": 5.6398,
"mean_token_accuracy": 0.16703630983829498,
"num_tokens": 14299328.0,
"step": 6405
},
{
"entropy": 5.711787748336792,
"epoch": 0.7199415960015724,
"grad_norm": 1.2734375,
"learning_rate": 0.0004958209666960909,
"loss": 5.6084,
"mean_token_accuracy": 0.17494853734970092,
"num_tokens": 14310048.0,
"step": 6410
},
{
"entropy": 5.565041589736938,
"epoch": 0.7205031729095299,
"grad_norm": 1.3125,
"learning_rate": 0.0004958132611128448,
"loss": 5.5351,
"mean_token_accuracy": 0.18220226615667343,
"num_tokens": 14321107.0,
"step": 6415
},
{
"entropy": 5.618595027923584,
"epoch": 0.7210647498174875,
"grad_norm": 1.2734375,
"learning_rate": 0.0004958055484988188,
"loss": 5.4544,
"mean_token_accuracy": 0.17759331315755844,
"num_tokens": 14332561.0,
"step": 6420
},
{
"entropy": 5.713983678817749,
"epoch": 0.721626326725445,
"grad_norm": 1.4375,
"learning_rate": 0.0004957978288542585,
"loss": 5.6547,
"mean_token_accuracy": 0.17716152369976043,
"num_tokens": 14343238.0,
"step": 6425
},
{
"entropy": 5.630457258224487,
"epoch": 0.7221879036334026,
"grad_norm": 1.25,
"learning_rate": 0.0004957901021794098,
"loss": 5.5522,
"mean_token_accuracy": 0.18020768761634826,
"num_tokens": 14356067.0,
"step": 6430
},
{
"entropy": 5.626893949508667,
"epoch": 0.7227494805413601,
"grad_norm": 1.3828125,
"learning_rate": 0.0004957823684745184,
"loss": 5.5691,
"mean_token_accuracy": 0.1788392499089241,
"num_tokens": 14367302.0,
"step": 6435
},
{
"entropy": 5.669112777709961,
"epoch": 0.7233110574493177,
"grad_norm": 1.4453125,
"learning_rate": 0.0004957746277398309,
"loss": 5.5122,
"mean_token_accuracy": 0.1818000316619873,
"num_tokens": 14377535.0,
"step": 6440
},
{
"entropy": 5.725162029266357,
"epoch": 0.7238726343572752,
"grad_norm": 1.609375,
"learning_rate": 0.0004957668799755937,
"loss": 5.6543,
"mean_token_accuracy": 0.16769258230924605,
"num_tokens": 14389493.0,
"step": 6445
},
{
"entropy": 5.655407428741455,
"epoch": 0.7244342112652328,
"grad_norm": 1.6015625,
"learning_rate": 0.0004957591251820532,
"loss": 5.5567,
"mean_token_accuracy": 0.17890584468841553,
"num_tokens": 14401564.0,
"step": 6450
},
{
"entropy": 5.708035230636597,
"epoch": 0.7249957881731903,
"grad_norm": 1.4453125,
"learning_rate": 0.0004957513633594567,
"loss": 5.6117,
"mean_token_accuracy": 0.1804538518190384,
"num_tokens": 14412123.0,
"step": 6455
},
{
"entropy": 5.654670763015747,
"epoch": 0.7255573650811479,
"grad_norm": 1.21875,
"learning_rate": 0.0004957435945080512,
"loss": 5.5957,
"mean_token_accuracy": 0.177344973385334,
"num_tokens": 14423614.0,
"step": 6460
},
{
"entropy": 5.717520666122437,
"epoch": 0.7261189419891054,
"grad_norm": 1.5859375,
"learning_rate": 0.0004957358186280839,
"loss": 5.5816,
"mean_token_accuracy": 0.1718784213066101,
"num_tokens": 14434830.0,
"step": 6465
},
{
"entropy": 5.6881279945373535,
"epoch": 0.726680518897063,
"grad_norm": 1.4140625,
"learning_rate": 0.0004957280357198026,
"loss": 5.6281,
"mean_token_accuracy": 0.1757378563284874,
"num_tokens": 14445015.0,
"step": 6470
},
{
"entropy": 5.67712926864624,
"epoch": 0.7272420958050205,
"grad_norm": 1.5390625,
"learning_rate": 0.0004957202457834551,
"loss": 5.7051,
"mean_token_accuracy": 0.16345522552728653,
"num_tokens": 14455809.0,
"step": 6475
},
{
"entropy": 5.684776067733765,
"epoch": 0.7278036727129781,
"grad_norm": 1.5234375,
"learning_rate": 0.0004957124488192893,
"loss": 5.4971,
"mean_token_accuracy": 0.18258208334445952,
"num_tokens": 14466565.0,
"step": 6480
},
{
"entropy": 5.684683609008789,
"epoch": 0.7283652496209356,
"grad_norm": 5.0,
"learning_rate": 0.0004957046448275535,
"loss": 5.5932,
"mean_token_accuracy": 0.1817127451300621,
"num_tokens": 14477383.0,
"step": 6485
},
{
"entropy": 5.665232229232788,
"epoch": 0.7289268265288932,
"grad_norm": 1.3046875,
"learning_rate": 0.0004956968338084961,
"loss": 5.5752,
"mean_token_accuracy": 0.1728922724723816,
"num_tokens": 14489968.0,
"step": 6490
},
{
"entropy": 5.7213369846344,
"epoch": 0.7294884034368507,
"grad_norm": 1.65625,
"learning_rate": 0.0004956890157623661,
"loss": 5.6268,
"mean_token_accuracy": 0.17355639338493348,
"num_tokens": 14500808.0,
"step": 6495
},
{
"entropy": 5.63701434135437,
"epoch": 0.7300499803448082,
"grad_norm": 1.9375,
"learning_rate": 0.0004956811906894122,
"loss": 5.619,
"mean_token_accuracy": 0.17685243785381316,
"num_tokens": 14511444.0,
"step": 6500
},
{
"entropy": 5.638969278335571,
"epoch": 0.7306115572527657,
"grad_norm": 1.5625,
"learning_rate": 0.0004956733585898836,
"loss": 5.5391,
"mean_token_accuracy": 0.17560631036758423,
"num_tokens": 14522562.0,
"step": 6505
},
{
"entropy": 5.653846740722656,
"epoch": 0.7311731341607233,
"grad_norm": 1.4609375,
"learning_rate": 0.0004956655194640297,
"loss": 5.5634,
"mean_token_accuracy": 0.17611929327249526,
"num_tokens": 14533497.0,
"step": 6510
},
{
"entropy": 5.669759464263916,
"epoch": 0.7317347110686808,
"grad_norm": 1.5078125,
"learning_rate": 0.0004956576733121,
"loss": 5.603,
"mean_token_accuracy": 0.16803350895643235,
"num_tokens": 14545637.0,
"step": 6515
},
{
"entropy": 5.711380434036255,
"epoch": 0.7322962879766384,
"grad_norm": 1.2578125,
"learning_rate": 0.0004956498201343444,
"loss": 5.5943,
"mean_token_accuracy": 0.1739874631166458,
"num_tokens": 14556826.0,
"step": 6520
},
{
"entropy": 5.614512300491333,
"epoch": 0.7328578648845959,
"grad_norm": 1.703125,
"learning_rate": 0.000495641959931013,
"loss": 5.5319,
"mean_token_accuracy": 0.17449793368577957,
"num_tokens": 14567315.0,
"step": 6525
},
{
"entropy": 5.626665687561035,
"epoch": 0.7334194417925535,
"grad_norm": 1.625,
"learning_rate": 0.0004956340927023561,
"loss": 5.6004,
"mean_token_accuracy": 0.17751851975917815,
"num_tokens": 14577685.0,
"step": 6530
},
{
"entropy": 5.699931049346924,
"epoch": 0.733981018700511,
"grad_norm": 1.4140625,
"learning_rate": 0.000495626218448624,
"loss": 5.5628,
"mean_token_accuracy": 0.17922184467315674,
"num_tokens": 14588885.0,
"step": 6535
},
{
"entropy": 5.662351083755493,
"epoch": 0.7345425956084686,
"grad_norm": 2.0625,
"learning_rate": 0.0004956183371700676,
"loss": 5.5797,
"mean_token_accuracy": 0.1780926913022995,
"num_tokens": 14599774.0,
"step": 6540
},
{
"entropy": 5.752463150024414,
"epoch": 0.7351041725164261,
"grad_norm": 1.5703125,
"learning_rate": 0.0004956104488669377,
"loss": 5.6278,
"mean_token_accuracy": 0.17140088528394698,
"num_tokens": 14612101.0,
"step": 6545
},
{
"entropy": 5.7867025375366214,
"epoch": 0.7356657494243837,
"grad_norm": 2.0,
"learning_rate": 0.0004956025535394856,
"loss": 5.7046,
"mean_token_accuracy": 0.16080682575702668,
"num_tokens": 14622151.0,
"step": 6550
},
{
"entropy": 5.68880443572998,
"epoch": 0.7362273263323412,
"grad_norm": 1.328125,
"learning_rate": 0.0004955946511879626,
"loss": 5.5951,
"mean_token_accuracy": 0.16880016028881073,
"num_tokens": 14634631.0,
"step": 6555
},
{
"entropy": 5.625009202957154,
"epoch": 0.7367889032402988,
"grad_norm": 1.390625,
"learning_rate": 0.0004955867418126205,
"loss": 5.5388,
"mean_token_accuracy": 0.17859316021203994,
"num_tokens": 14645429.0,
"step": 6560
},
{
"entropy": 5.603000259399414,
"epoch": 0.7373504801482563,
"grad_norm": 1.53125,
"learning_rate": 0.0004955788254137108,
"loss": 5.544,
"mean_token_accuracy": 0.18018025755882264,
"num_tokens": 14657309.0,
"step": 6565
},
{
"entropy": 5.640166854858398,
"epoch": 0.7379120570562139,
"grad_norm": 1.5703125,
"learning_rate": 0.0004955709019914857,
"loss": 5.4801,
"mean_token_accuracy": 0.17528376281261443,
"num_tokens": 14668578.0,
"step": 6570
},
{
"entropy": 5.720955514907837,
"epoch": 0.7384736339641714,
"grad_norm": 1.8359375,
"learning_rate": 0.0004955629715461977,
"loss": 5.6918,
"mean_token_accuracy": 0.1695711076259613,
"num_tokens": 14679120.0,
"step": 6575
},
{
"entropy": 5.7782378673553465,
"epoch": 0.739035210872129,
"grad_norm": 1.25,
"learning_rate": 0.0004955550340780991,
"loss": 5.6371,
"mean_token_accuracy": 0.17302252501249313,
"num_tokens": 14690849.0,
"step": 6580
},
{
"entropy": 5.662907314300537,
"epoch": 0.7395967877800865,
"grad_norm": 1.140625,
"learning_rate": 0.0004955470895874427,
"loss": 5.5982,
"mean_token_accuracy": 0.1746624603867531,
"num_tokens": 14702431.0,
"step": 6585
},
{
"entropy": 5.546743249893188,
"epoch": 0.7401583646880441,
"grad_norm": 1.6015625,
"learning_rate": 0.0004955391380744813,
"loss": 5.4996,
"mean_token_accuracy": 0.1823481425642967,
"num_tokens": 14714271.0,
"step": 6590
},
{
"entropy": 5.656334733963012,
"epoch": 0.7407199415960015,
"grad_norm": 1.6171875,
"learning_rate": 0.0004955311795394684,
"loss": 5.6346,
"mean_token_accuracy": 0.17327096909284592,
"num_tokens": 14724796.0,
"step": 6595
},
{
"entropy": 5.654460525512695,
"epoch": 0.7412815185039591,
"grad_norm": 1.3203125,
"learning_rate": 0.000495523213982657,
"loss": 5.5467,
"mean_token_accuracy": 0.1740984171628952,
"num_tokens": 14735331.0,
"step": 6600
},
{
"entropy": 5.718264532089234,
"epoch": 0.7418430954119166,
"grad_norm": 1.2265625,
"learning_rate": 0.0004955152414043011,
"loss": 5.622,
"mean_token_accuracy": 0.17427025288343428,
"num_tokens": 14748595.0,
"step": 6605
},
{
"entropy": 5.730298662185669,
"epoch": 0.7424046723198742,
"grad_norm": 1.421875,
"learning_rate": 0.0004955072618046543,
"loss": 5.5554,
"mean_token_accuracy": 0.17893725484609604,
"num_tokens": 14759501.0,
"step": 6610
},
{
"entropy": 5.576894760131836,
"epoch": 0.7429662492278317,
"grad_norm": 1.2265625,
"learning_rate": 0.0004954992751839708,
"loss": 5.6067,
"mean_token_accuracy": 0.17271900177001953,
"num_tokens": 14769752.0,
"step": 6615
},
{
"entropy": 5.649269533157349,
"epoch": 0.7435278261357893,
"grad_norm": 1.53125,
"learning_rate": 0.0004954912815425048,
"loss": 5.5402,
"mean_token_accuracy": 0.1787666454911232,
"num_tokens": 14781166.0,
"step": 6620
},
{
"entropy": 5.606461000442505,
"epoch": 0.7440894030437468,
"grad_norm": 1.25,
"learning_rate": 0.000495483280880511,
"loss": 5.4479,
"mean_token_accuracy": 0.182544481754303,
"num_tokens": 14790536.0,
"step": 6625
},
{
"entropy": 5.710970211029053,
"epoch": 0.7446509799517044,
"grad_norm": 1.40625,
"learning_rate": 0.000495475273198244,
"loss": 5.6592,
"mean_token_accuracy": 0.1696382686495781,
"num_tokens": 14801673.0,
"step": 6630
},
{
"entropy": 5.645382738113403,
"epoch": 0.7452125568596619,
"grad_norm": 1.3203125,
"learning_rate": 0.0004954672584959588,
"loss": 5.5587,
"mean_token_accuracy": 0.17158541679382325,
"num_tokens": 14812345.0,
"step": 6635
},
{
"entropy": 5.752765321731568,
"epoch": 0.7457741337676195,
"grad_norm": 1.421875,
"learning_rate": 0.0004954592367739104,
"loss": 5.7287,
"mean_token_accuracy": 0.1673116758465767,
"num_tokens": 14823620.0,
"step": 6640
},
{
"entropy": 5.695698976516724,
"epoch": 0.746335710675577,
"grad_norm": 1.328125,
"learning_rate": 0.0004954512080323546,
"loss": 5.5237,
"mean_token_accuracy": 0.1829296514391899,
"num_tokens": 14833487.0,
"step": 6645
},
{
"entropy": 5.50823655128479,
"epoch": 0.7468972875835346,
"grad_norm": 1.328125,
"learning_rate": 0.0004954431722715468,
"loss": 5.4009,
"mean_token_accuracy": 0.1812967523932457,
"num_tokens": 14844421.0,
"step": 6650
},
{
"entropy": 5.552810001373291,
"epoch": 0.7474588644914921,
"grad_norm": 1.3203125,
"learning_rate": 0.0004954351294917428,
"loss": 5.5788,
"mean_token_accuracy": 0.1808403879404068,
"num_tokens": 14855525.0,
"step": 6655
},
{
"entropy": 5.781732559204102,
"epoch": 0.7480204413994497,
"grad_norm": 1.546875,
"learning_rate": 0.0004954270796931989,
"loss": 5.6251,
"mean_token_accuracy": 0.16996127367019653,
"num_tokens": 14866717.0,
"step": 6660
},
{
"entropy": 5.637063026428223,
"epoch": 0.7485820183074072,
"grad_norm": 1.40625,
"learning_rate": 0.0004954190228761712,
"loss": 5.5286,
"mean_token_accuracy": 0.17126020938158035,
"num_tokens": 14877372.0,
"step": 6665
},
{
"entropy": 5.645986270904541,
"epoch": 0.7491435952153648,
"grad_norm": 1.5234375,
"learning_rate": 0.0004954109590409164,
"loss": 5.5854,
"mean_token_accuracy": 0.175726780295372,
"num_tokens": 14888618.0,
"step": 6670
},
{
"entropy": 5.605580997467041,
"epoch": 0.7497051721233223,
"grad_norm": 1.3203125,
"learning_rate": 0.0004954028881876911,
"loss": 5.6016,
"mean_token_accuracy": 0.1725963309407234,
"num_tokens": 14900421.0,
"step": 6675
},
{
"entropy": 5.665360307693481,
"epoch": 0.7502667490312799,
"grad_norm": 1.3671875,
"learning_rate": 0.0004953948103167524,
"loss": 5.559,
"mean_token_accuracy": 0.1724382147192955,
"num_tokens": 14910634.0,
"step": 6680
},
{
"entropy": 5.769053840637207,
"epoch": 0.7508283259392374,
"grad_norm": 1.3515625,
"learning_rate": 0.0004953867254283575,
"loss": 5.6231,
"mean_token_accuracy": 0.17022716104984284,
"num_tokens": 14921408.0,
"step": 6685
},
{
"entropy": 5.672090148925781,
"epoch": 0.7513899028471949,
"grad_norm": 1.328125,
"learning_rate": 0.0004953786335227636,
"loss": 5.6063,
"mean_token_accuracy": 0.17685625553131104,
"num_tokens": 14933435.0,
"step": 6690
},
{
"entropy": 5.643823146820068,
"epoch": 0.7519514797551524,
"grad_norm": 1.5859375,
"learning_rate": 0.0004953705346002287,
"loss": 5.6382,
"mean_token_accuracy": 0.16704337745904924,
"num_tokens": 14946488.0,
"step": 6695
},
{
"entropy": 5.613405656814575,
"epoch": 0.75251305666311,
"grad_norm": 2.09375,
"learning_rate": 0.0004953624286610106,
"loss": 5.4853,
"mean_token_accuracy": 0.17981879264116288,
"num_tokens": 14958053.0,
"step": 6700
},
{
"entropy": 5.620806646347046,
"epoch": 0.7530746335710675,
"grad_norm": 1.484375,
"learning_rate": 0.0004953543157053672,
"loss": 5.5685,
"mean_token_accuracy": 0.1722116619348526,
"num_tokens": 14968922.0,
"step": 6705
},
{
"entropy": 5.681433057785034,
"epoch": 0.7536362104790251,
"grad_norm": 1.4140625,
"learning_rate": 0.0004953461957335569,
"loss": 5.5681,
"mean_token_accuracy": 0.17288730889558793,
"num_tokens": 14979541.0,
"step": 6710
},
{
"entropy": 5.677426815032959,
"epoch": 0.7541977873869826,
"grad_norm": 1.375,
"learning_rate": 0.0004953380687458382,
"loss": 5.5925,
"mean_token_accuracy": 0.16874028295278548,
"num_tokens": 14990009.0,
"step": 6715
},
{
"entropy": 5.611559820175171,
"epoch": 0.7547593642949402,
"grad_norm": 1.453125,
"learning_rate": 0.0004953299347424701,
"loss": 5.6063,
"mean_token_accuracy": 0.17147741615772247,
"num_tokens": 15001712.0,
"step": 6720
},
{
"entropy": 5.711595582962036,
"epoch": 0.7553209412028977,
"grad_norm": 1.3515625,
"learning_rate": 0.0004953217937237114,
"loss": 5.5591,
"mean_token_accuracy": 0.18007346391677856,
"num_tokens": 15012574.0,
"step": 6725
},
{
"entropy": 5.575225877761841,
"epoch": 0.7558825181108553,
"grad_norm": 1.328125,
"learning_rate": 0.0004953136456898212,
"loss": 5.5404,
"mean_token_accuracy": 0.17802151888608933,
"num_tokens": 15023034.0,
"step": 6730
},
{
"entropy": 5.669577693939209,
"epoch": 0.7564440950188128,
"grad_norm": 1.1875,
"learning_rate": 0.0004953054906410593,
"loss": 5.6023,
"mean_token_accuracy": 0.17056265622377395,
"num_tokens": 15033925.0,
"step": 6735
},
{
"entropy": 5.660538673400879,
"epoch": 0.7570056719267704,
"grad_norm": 1.2109375,
"learning_rate": 0.0004952973285776848,
"loss": 5.5833,
"mean_token_accuracy": 0.17850170135498047,
"num_tokens": 15044947.0,
"step": 6740
},
{
"entropy": 5.65308198928833,
"epoch": 0.7575672488347279,
"grad_norm": 1.359375,
"learning_rate": 0.0004952891594999582,
"loss": 5.4883,
"mean_token_accuracy": 0.18364972174167632,
"num_tokens": 15054784.0,
"step": 6745
},
{
"entropy": 5.578481435775757,
"epoch": 0.7581288257426855,
"grad_norm": 1.203125,
"learning_rate": 0.0004952809834081392,
"loss": 5.5649,
"mean_token_accuracy": 0.17120407074689864,
"num_tokens": 15066272.0,
"step": 6750
},
{
"entropy": 5.7180320739746096,
"epoch": 0.758690402650643,
"grad_norm": 1.421875,
"learning_rate": 0.0004952728003024883,
"loss": 5.6511,
"mean_token_accuracy": 0.17361172735691072,
"num_tokens": 15079504.0,
"step": 6755
},
{
"entropy": 5.629498052597046,
"epoch": 0.7592519795586006,
"grad_norm": 1.28125,
"learning_rate": 0.0004952646101832659,
"loss": 5.5799,
"mean_token_accuracy": 0.1806523859500885,
"num_tokens": 15089623.0,
"step": 6760
},
{
"entropy": 5.591461181640625,
"epoch": 0.7598135564665581,
"grad_norm": 1.3828125,
"learning_rate": 0.0004952564130507331,
"loss": 5.5002,
"mean_token_accuracy": 0.17521500140428542,
"num_tokens": 15100816.0,
"step": 6765
},
{
"entropy": 5.621237897872925,
"epoch": 0.7603751333745157,
"grad_norm": 1.3984375,
"learning_rate": 0.0004952482089051506,
"loss": 5.6017,
"mean_token_accuracy": 0.17664288878440856,
"num_tokens": 15111697.0,
"step": 6770
},
{
"entropy": 5.683472728729248,
"epoch": 0.7609367102824732,
"grad_norm": 1.3046875,
"learning_rate": 0.0004952399977467796,
"loss": 5.6577,
"mean_token_accuracy": 0.1735858991742134,
"num_tokens": 15122869.0,
"step": 6775
},
{
"entropy": 5.620102405548096,
"epoch": 0.7614982871904308,
"grad_norm": 1.375,
"learning_rate": 0.0004952317795758817,
"loss": 5.4624,
"mean_token_accuracy": 0.17760494947433472,
"num_tokens": 15134709.0,
"step": 6780
},
{
"entropy": 5.606343555450439,
"epoch": 0.7620598640983883,
"grad_norm": 1.3828125,
"learning_rate": 0.0004952235543927185,
"loss": 5.4953,
"mean_token_accuracy": 0.18179071545600892,
"num_tokens": 15145319.0,
"step": 6785
},
{
"entropy": 5.6012715816497805,
"epoch": 0.7626214410063458,
"grad_norm": 1.3203125,
"learning_rate": 0.000495215322197552,
"loss": 5.528,
"mean_token_accuracy": 0.17106836438179016,
"num_tokens": 15156463.0,
"step": 6790
},
{
"entropy": 5.690052270889282,
"epoch": 0.7631830179143033,
"grad_norm": 1.3046875,
"learning_rate": 0.0004952070829906442,
"loss": 5.6671,
"mean_token_accuracy": 0.16552183628082276,
"num_tokens": 15167618.0,
"step": 6795
},
{
"entropy": 5.694620513916016,
"epoch": 0.7637445948222609,
"grad_norm": 1.3046875,
"learning_rate": 0.0004951988367722574,
"loss": 5.5992,
"mean_token_accuracy": 0.17493318766355515,
"num_tokens": 15179312.0,
"step": 6800
},
{
"entropy": 5.594572973251343,
"epoch": 0.7643061717302184,
"grad_norm": 1.3125,
"learning_rate": 0.0004951905835426544,
"loss": 5.5437,
"mean_token_accuracy": 0.17916350811719894,
"num_tokens": 15191370.0,
"step": 6805
},
{
"entropy": 5.690248918533325,
"epoch": 0.764867748638176,
"grad_norm": 1.3515625,
"learning_rate": 0.0004951823233020976,
"loss": 5.5016,
"mean_token_accuracy": 0.1812632903456688,
"num_tokens": 15201060.0,
"step": 6810
},
{
"entropy": 5.581395959854126,
"epoch": 0.7654293255461335,
"grad_norm": 1.3203125,
"learning_rate": 0.0004951740560508504,
"loss": 5.436,
"mean_token_accuracy": 0.17881249785423278,
"num_tokens": 15212186.0,
"step": 6815
},
{
"entropy": 5.5898966789245605,
"epoch": 0.7659909024540911,
"grad_norm": 1.421875,
"learning_rate": 0.0004951657817891758,
"loss": 5.4652,
"mean_token_accuracy": 0.18410132378339766,
"num_tokens": 15223301.0,
"step": 6820
},
{
"entropy": 5.633123874664307,
"epoch": 0.7665524793620486,
"grad_norm": 1.171875,
"learning_rate": 0.0004951575005173373,
"loss": 5.6174,
"mean_token_accuracy": 0.17449728548526763,
"num_tokens": 15234475.0,
"step": 6825
},
{
"entropy": 5.647843217849731,
"epoch": 0.7671140562700062,
"grad_norm": 1.125,
"learning_rate": 0.0004951492122355986,
"loss": 5.5837,
"mean_token_accuracy": 0.17357583343982697,
"num_tokens": 15245877.0,
"step": 6830
},
{
"entropy": 5.702678775787353,
"epoch": 0.7676756331779637,
"grad_norm": 1.453125,
"learning_rate": 0.0004951409169442236,
"loss": 5.5878,
"mean_token_accuracy": 0.17982097864151,
"num_tokens": 15257375.0,
"step": 6835
},
{
"entropy": 5.659814786911011,
"epoch": 0.7682372100859213,
"grad_norm": 1.203125,
"learning_rate": 0.0004951326146434765,
"loss": 5.5907,
"mean_token_accuracy": 0.17140717506408693,
"num_tokens": 15268079.0,
"step": 6840
},
{
"entropy": 5.617394495010376,
"epoch": 0.7687987869938788,
"grad_norm": 1.40625,
"learning_rate": 0.0004951243053336214,
"loss": 5.5724,
"mean_token_accuracy": 0.1755111485719681,
"num_tokens": 15278678.0,
"step": 6845
},
{
"entropy": 5.727441596984863,
"epoch": 0.7693603639018364,
"grad_norm": 1.421875,
"learning_rate": 0.0004951159890149232,
"loss": 5.6861,
"mean_token_accuracy": 0.16761346608400346,
"num_tokens": 15290080.0,
"step": 6850
},
{
"entropy": 5.69748740196228,
"epoch": 0.7699219408097939,
"grad_norm": 1.390625,
"learning_rate": 0.0004951076656876466,
"loss": 5.5925,
"mean_token_accuracy": 0.16981475949287414,
"num_tokens": 15300602.0,
"step": 6855
},
{
"entropy": 5.7041943073272705,
"epoch": 0.7704835177177515,
"grad_norm": 1.2265625,
"learning_rate": 0.0004950993353520564,
"loss": 5.617,
"mean_token_accuracy": 0.17697811722755433,
"num_tokens": 15312823.0,
"step": 6860
},
{
"entropy": 5.69435224533081,
"epoch": 0.771045094625709,
"grad_norm": 1.1484375,
"learning_rate": 0.000495090998008418,
"loss": 5.6123,
"mean_token_accuracy": 0.17509002089500428,
"num_tokens": 15323870.0,
"step": 6865
},
{
"entropy": 5.614969205856323,
"epoch": 0.7716066715336666,
"grad_norm": 1.21875,
"learning_rate": 0.000495082653656997,
"loss": 5.5105,
"mean_token_accuracy": 0.176980821788311,
"num_tokens": 15334887.0,
"step": 6870
},
{
"entropy": 5.722302484512329,
"epoch": 0.7721682484416241,
"grad_norm": 1.328125,
"learning_rate": 0.0004950743022980587,
"loss": 5.6106,
"mean_token_accuracy": 0.16959810853004456,
"num_tokens": 15345871.0,
"step": 6875
},
{
"entropy": 5.586536026000976,
"epoch": 0.7727298253495817,
"grad_norm": 1.3203125,
"learning_rate": 0.0004950659439318696,
"loss": 5.4624,
"mean_token_accuracy": 0.17703748494386673,
"num_tokens": 15356530.0,
"step": 6880
},
{
"entropy": 5.61054539680481,
"epoch": 0.7732914022575391,
"grad_norm": 1.25,
"learning_rate": 0.0004950575785586952,
"loss": 5.5424,
"mean_token_accuracy": 0.17756601125001908,
"num_tokens": 15368756.0,
"step": 6885
},
{
"entropy": 5.7294392585754395,
"epoch": 0.7738529791654967,
"grad_norm": 1.2890625,
"learning_rate": 0.0004950492061788023,
"loss": 5.6218,
"mean_token_accuracy": 0.1781395137310028,
"num_tokens": 15379450.0,
"step": 6890
},
{
"entropy": 5.61515965461731,
"epoch": 0.7744145560734542,
"grad_norm": 1.2109375,
"learning_rate": 0.0004950408267924573,
"loss": 5.5333,
"mean_token_accuracy": 0.17131865173578262,
"num_tokens": 15389501.0,
"step": 6895
},
{
"entropy": 5.5294106006622314,
"epoch": 0.7749761329814118,
"grad_norm": 2.328125,
"learning_rate": 0.0004950324403999271,
"loss": 5.549,
"mean_token_accuracy": 0.17641455233097075,
"num_tokens": 15400608.0,
"step": 6900
},
{
"entropy": 5.751321268081665,
"epoch": 0.7755377098893693,
"grad_norm": 1.40625,
"learning_rate": 0.0004950240470014785,
"loss": 5.6216,
"mean_token_accuracy": 0.17117204666137695,
"num_tokens": 15411709.0,
"step": 6905
},
{
"entropy": 5.611541652679444,
"epoch": 0.7760992867973269,
"grad_norm": 1.3046875,
"learning_rate": 0.0004950156465973791,
"loss": 5.4792,
"mean_token_accuracy": 0.18414486944675446,
"num_tokens": 15422292.0,
"step": 6910
},
{
"entropy": 5.685716915130615,
"epoch": 0.7766608637052844,
"grad_norm": 1.265625,
"learning_rate": 0.000495007239187896,
"loss": 5.6344,
"mean_token_accuracy": 0.1741434559226036,
"num_tokens": 15434439.0,
"step": 6915
},
{
"entropy": 5.66471037864685,
"epoch": 0.777222440613242,
"grad_norm": 1.703125,
"learning_rate": 0.0004949988247732972,
"loss": 5.5693,
"mean_token_accuracy": 0.1746375307440758,
"num_tokens": 15445480.0,
"step": 6920
},
{
"entropy": 5.623708629608155,
"epoch": 0.7777840175211995,
"grad_norm": 1.3984375,
"learning_rate": 0.0004949904033538504,
"loss": 5.5543,
"mean_token_accuracy": 0.17277151942253113,
"num_tokens": 15456960.0,
"step": 6925
},
{
"entropy": 5.6369672298431395,
"epoch": 0.7783455944291571,
"grad_norm": 1.5,
"learning_rate": 0.000494981974929824,
"loss": 5.5595,
"mean_token_accuracy": 0.17388996630907058,
"num_tokens": 15468169.0,
"step": 6930
},
{
"entropy": 5.676503467559814,
"epoch": 0.7789071713371146,
"grad_norm": 1.53125,
"learning_rate": 0.000494973539501486,
"loss": 5.5397,
"mean_token_accuracy": 0.17739749997854232,
"num_tokens": 15478740.0,
"step": 6935
},
{
"entropy": 5.599559688568116,
"epoch": 0.7794687482450722,
"grad_norm": 1.359375,
"learning_rate": 0.0004949650970691052,
"loss": 5.6318,
"mean_token_accuracy": 0.17098047733306884,
"num_tokens": 15489705.0,
"step": 6940
},
{
"entropy": 5.578276538848877,
"epoch": 0.7800303251530297,
"grad_norm": 1.3203125,
"learning_rate": 0.0004949566476329505,
"loss": 5.4604,
"mean_token_accuracy": 0.17657845616340637,
"num_tokens": 15500549.0,
"step": 6945
},
{
"entropy": 5.801419973373413,
"epoch": 0.7805919020609873,
"grad_norm": 2.0,
"learning_rate": 0.0004949481911932908,
"loss": 5.6195,
"mean_token_accuracy": 0.16939314752817153,
"num_tokens": 15511124.0,
"step": 6950
},
{
"entropy": 5.624944019317627,
"epoch": 0.7811534789689448,
"grad_norm": 1.1953125,
"learning_rate": 0.0004949397277503954,
"loss": 5.5192,
"mean_token_accuracy": 0.17904226183891297,
"num_tokens": 15521810.0,
"step": 6955
},
{
"entropy": 5.619323778152466,
"epoch": 0.7817150558769024,
"grad_norm": 1.203125,
"learning_rate": 0.0004949312573045338,
"loss": 5.5598,
"mean_token_accuracy": 0.17706193923950195,
"num_tokens": 15533108.0,
"step": 6960
},
{
"entropy": 5.6386102676391605,
"epoch": 0.7822766327848599,
"grad_norm": 1.3984375,
"learning_rate": 0.0004949227798559757,
"loss": 5.5896,
"mean_token_accuracy": 0.17438544034957887,
"num_tokens": 15544841.0,
"step": 6965
},
{
"entropy": 5.667905616760254,
"epoch": 0.7828382096928175,
"grad_norm": 1.25,
"learning_rate": 0.0004949142954049909,
"loss": 5.5738,
"mean_token_accuracy": 0.1699427530169487,
"num_tokens": 15556297.0,
"step": 6970
},
{
"entropy": 5.608936309814453,
"epoch": 0.783399786600775,
"grad_norm": 1.1796875,
"learning_rate": 0.0004949058039518496,
"loss": 5.466,
"mean_token_accuracy": 0.17948467880487443,
"num_tokens": 15566607.0,
"step": 6975
},
{
"entropy": 5.6763256072998045,
"epoch": 0.7839613635087325,
"grad_norm": 1.2265625,
"learning_rate": 0.0004948973054968222,
"loss": 5.6339,
"mean_token_accuracy": 0.17235210984945298,
"num_tokens": 15577435.0,
"step": 6980
},
{
"entropy": 5.6864330768585205,
"epoch": 0.78452294041669,
"grad_norm": 1.578125,
"learning_rate": 0.0004948888000401794,
"loss": 5.5385,
"mean_token_accuracy": 0.17783611118793488,
"num_tokens": 15587844.0,
"step": 6985
},
{
"entropy": 5.653660917282105,
"epoch": 0.7850845173246476,
"grad_norm": 1.4921875,
"learning_rate": 0.0004948802875821918,
"loss": 5.6096,
"mean_token_accuracy": 0.17504583597183226,
"num_tokens": 15599586.0,
"step": 6990
},
{
"entropy": 5.57503309249878,
"epoch": 0.7856460942326051,
"grad_norm": 1.4140625,
"learning_rate": 0.0004948717681231308,
"loss": 5.4893,
"mean_token_accuracy": 0.1764574870467186,
"num_tokens": 15610838.0,
"step": 6995
},
{
"entropy": 5.704991436004638,
"epoch": 0.7862076711405627,
"grad_norm": 1.3359375,
"learning_rate": 0.0004948632416632672,
"loss": 5.5773,
"mean_token_accuracy": 0.17896959483623504,
"num_tokens": 15622652.0,
"step": 7000
},
{
"entropy": 5.749733257293701,
"epoch": 0.7867692480485202,
"grad_norm": 1.4921875,
"learning_rate": 0.0004948547082028728,
"loss": 5.6619,
"mean_token_accuracy": 0.17040863633155823,
"num_tokens": 15634280.0,
"step": 7005
},
{
"entropy": 5.545354843139648,
"epoch": 0.7873308249564778,
"grad_norm": 1.2578125,
"learning_rate": 0.0004948461677422193,
"loss": 5.3864,
"mean_token_accuracy": 0.1879497066140175,
"num_tokens": 15644088.0,
"step": 7010
},
{
"entropy": 5.598902750015259,
"epoch": 0.7878924018644353,
"grad_norm": 1.328125,
"learning_rate": 0.0004948376202815785,
"loss": 5.6464,
"mean_token_accuracy": 0.16963843405246734,
"num_tokens": 15654968.0,
"step": 7015
},
{
"entropy": 5.639025545120239,
"epoch": 0.7884539787723929,
"grad_norm": 1.5625,
"learning_rate": 0.0004948290658212225,
"loss": 5.5409,
"mean_token_accuracy": 0.17644733488559722,
"num_tokens": 15666758.0,
"step": 7020
},
{
"entropy": 5.5757989406585695,
"epoch": 0.7890155556803504,
"grad_norm": 1.390625,
"learning_rate": 0.0004948205043614239,
"loss": 5.393,
"mean_token_accuracy": 0.1842886060476303,
"num_tokens": 15677295.0,
"step": 7025
},
{
"entropy": 5.612846231460571,
"epoch": 0.789577132588308,
"grad_norm": 1.328125,
"learning_rate": 0.0004948119359024551,
"loss": 5.6049,
"mean_token_accuracy": 0.18194248229265214,
"num_tokens": 15689508.0,
"step": 7030
},
{
"entropy": 5.684176445007324,
"epoch": 0.7901387094962655,
"grad_norm": 1.3359375,
"learning_rate": 0.0004948033604445891,
"loss": 5.5654,
"mean_token_accuracy": 0.17480146139860153,
"num_tokens": 15701289.0,
"step": 7035
},
{
"entropy": 5.5590503215789795,
"epoch": 0.7907002864042231,
"grad_norm": 1.3125,
"learning_rate": 0.0004947947779880988,
"loss": 5.4962,
"mean_token_accuracy": 0.17311416417360306,
"num_tokens": 15711579.0,
"step": 7040
},
{
"entropy": 5.616797590255738,
"epoch": 0.7912618633121806,
"grad_norm": 1.2578125,
"learning_rate": 0.0004947861885332576,
"loss": 5.6041,
"mean_token_accuracy": 0.17735150903463365,
"num_tokens": 15722629.0,
"step": 7045
},
{
"entropy": 5.660729646682739,
"epoch": 0.7918234402201382,
"grad_norm": 1.25,
"learning_rate": 0.0004947775920803388,
"loss": 5.5344,
"mean_token_accuracy": 0.16928197890520097,
"num_tokens": 15735818.0,
"step": 7050
},
{
"entropy": 5.580780839920044,
"epoch": 0.7923850171280957,
"grad_norm": 1.4140625,
"learning_rate": 0.0004947689886296162,
"loss": 5.6003,
"mean_token_accuracy": 0.17081615328788757,
"num_tokens": 15746914.0,
"step": 7055
},
{
"entropy": 5.58987717628479,
"epoch": 0.7929465940360533,
"grad_norm": 1.4375,
"learning_rate": 0.000494760378181364,
"loss": 5.4935,
"mean_token_accuracy": 0.17170348018407822,
"num_tokens": 15757597.0,
"step": 7060
},
{
"entropy": 5.567567014694214,
"epoch": 0.7935081709440108,
"grad_norm": 1.2734375,
"learning_rate": 0.0004947517607358559,
"loss": 5.4412,
"mean_token_accuracy": 0.18403613269329072,
"num_tokens": 15768860.0,
"step": 7065
},
{
"entropy": 5.586142492294312,
"epoch": 0.7940697478519684,
"grad_norm": 1.203125,
"learning_rate": 0.0004947431362933666,
"loss": 5.5499,
"mean_token_accuracy": 0.17376096397638321,
"num_tokens": 15779476.0,
"step": 7070
},
{
"entropy": 5.593358564376831,
"epoch": 0.7946313247599258,
"grad_norm": 1.390625,
"learning_rate": 0.0004947345048541706,
"loss": 5.5978,
"mean_token_accuracy": 0.17086222916841506,
"num_tokens": 15792142.0,
"step": 7075
},
{
"entropy": 5.627873659133911,
"epoch": 0.7951929016678834,
"grad_norm": 1.21875,
"learning_rate": 0.0004947258664185427,
"loss": 5.5612,
"mean_token_accuracy": 0.17813314646482467,
"num_tokens": 15802307.0,
"step": 7080
},
{
"entropy": 5.574973487854004,
"epoch": 0.7957544785758409,
"grad_norm": 1.3984375,
"learning_rate": 0.0004947172209867581,
"loss": 5.5376,
"mean_token_accuracy": 0.175410096347332,
"num_tokens": 15812986.0,
"step": 7085
},
{
"entropy": 5.608232688903809,
"epoch": 0.7963160554837985,
"grad_norm": 1.75,
"learning_rate": 0.0004947085685590919,
"loss": 5.4891,
"mean_token_accuracy": 0.1814771920442581,
"num_tokens": 15823938.0,
"step": 7090
},
{
"entropy": 5.548796844482422,
"epoch": 0.796877632391756,
"grad_norm": 1.34375,
"learning_rate": 0.0004946999091358196,
"loss": 5.4601,
"mean_token_accuracy": 0.17969344258308412,
"num_tokens": 15835813.0,
"step": 7095
},
{
"entropy": 5.629628658294678,
"epoch": 0.7974392092997136,
"grad_norm": 1.3359375,
"learning_rate": 0.000494691242717217,
"loss": 5.5217,
"mean_token_accuracy": 0.18037265986204148,
"num_tokens": 15846203.0,
"step": 7100
},
{
"entropy": 5.608344888687133,
"epoch": 0.7980007862076711,
"grad_norm": 1.1484375,
"learning_rate": 0.0004946825693035601,
"loss": 5.5094,
"mean_token_accuracy": 0.18162015080451965,
"num_tokens": 15856919.0,
"step": 7105
},
{
"entropy": 5.624166631698609,
"epoch": 0.7985623631156287,
"grad_norm": 1.265625,
"learning_rate": 0.000494673888895125,
"loss": 5.6254,
"mean_token_accuracy": 0.1729702666401863,
"num_tokens": 15867149.0,
"step": 7110
},
{
"entropy": 5.587703895568848,
"epoch": 0.7991239400235862,
"grad_norm": 1.5234375,
"learning_rate": 0.0004946652014921881,
"loss": 5.5234,
"mean_token_accuracy": 0.1791645109653473,
"num_tokens": 15877229.0,
"step": 7115
},
{
"entropy": 5.675305700302124,
"epoch": 0.7996855169315438,
"grad_norm": 1.296875,
"learning_rate": 0.0004946565070950259,
"loss": 5.4923,
"mean_token_accuracy": 0.1787053793668747,
"num_tokens": 15888799.0,
"step": 7120
},
{
"entropy": 5.540292453765869,
"epoch": 0.8002470938395013,
"grad_norm": 1.2734375,
"learning_rate": 0.0004946478057039153,
"loss": 5.4961,
"mean_token_accuracy": 0.1854872927069664,
"num_tokens": 15900144.0,
"step": 7125
},
{
"entropy": 5.680820035934448,
"epoch": 0.8008086707474589,
"grad_norm": 1.296875,
"learning_rate": 0.0004946390973191335,
"loss": 5.6666,
"mean_token_accuracy": 0.1647701993584633,
"num_tokens": 15912030.0,
"step": 7130
},
{
"entropy": 5.640228366851806,
"epoch": 0.8013702476554164,
"grad_norm": 1.2578125,
"learning_rate": 0.0004946303819409575,
"loss": 5.5258,
"mean_token_accuracy": 0.17563342601060866,
"num_tokens": 15923606.0,
"step": 7135
},
{
"entropy": 5.561696338653564,
"epoch": 0.801931824563374,
"grad_norm": 1.109375,
"learning_rate": 0.0004946216595696652,
"loss": 5.5222,
"mean_token_accuracy": 0.17987987995147706,
"num_tokens": 15935077.0,
"step": 7140
},
{
"entropy": 5.55102572441101,
"epoch": 0.8024934014713315,
"grad_norm": 1.421875,
"learning_rate": 0.0004946129302055338,
"loss": 5.4509,
"mean_token_accuracy": 0.18315509408712388,
"num_tokens": 15945331.0,
"step": 7145
},
{
"entropy": 5.613504028320312,
"epoch": 0.8030549783792891,
"grad_norm": 1.28125,
"learning_rate": 0.0004946041938488417,
"loss": 5.554,
"mean_token_accuracy": 0.17014957815408707,
"num_tokens": 15957031.0,
"step": 7150
},
{
"entropy": 5.656913232803345,
"epoch": 0.8036165552872466,
"grad_norm": 1.28125,
"learning_rate": 0.0004945954504998668,
"loss": 5.5871,
"mean_token_accuracy": 0.17657895237207413,
"num_tokens": 15967545.0,
"step": 7155
},
{
"entropy": 5.742512130737305,
"epoch": 0.8041781321952042,
"grad_norm": 1.2578125,
"learning_rate": 0.0004945867001588877,
"loss": 5.6416,
"mean_token_accuracy": 0.1659504219889641,
"num_tokens": 15978700.0,
"step": 7160
},
{
"entropy": 5.5622330665588375,
"epoch": 0.8047397091031617,
"grad_norm": 1.2109375,
"learning_rate": 0.0004945779428261827,
"loss": 5.535,
"mean_token_accuracy": 0.18144425600767136,
"num_tokens": 15989712.0,
"step": 7165
},
{
"entropy": 5.695117473602295,
"epoch": 0.8053012860111193,
"grad_norm": 1.265625,
"learning_rate": 0.000494569178502031,
"loss": 5.7256,
"mean_token_accuracy": 0.168807815015316,
"num_tokens": 16002380.0,
"step": 7170
},
{
"entropy": 5.741608715057373,
"epoch": 0.8058628629190767,
"grad_norm": 1.2109375,
"learning_rate": 0.0004945604071867114,
"loss": 5.5036,
"mean_token_accuracy": 0.17376484870910644,
"num_tokens": 16013749.0,
"step": 7175
},
{
"entropy": 5.648626518249512,
"epoch": 0.8064244398270343,
"grad_norm": 1.25,
"learning_rate": 0.0004945516288805035,
"loss": 5.6753,
"mean_token_accuracy": 0.17325000762939452,
"num_tokens": 16025131.0,
"step": 7180
},
{
"entropy": 5.688432312011718,
"epoch": 0.8069860167349918,
"grad_norm": 1.1640625,
"learning_rate": 0.0004945428435836864,
"loss": 5.5551,
"mean_token_accuracy": 0.17331638038158417,
"num_tokens": 16037269.0,
"step": 7185
},
{
"entropy": 5.785771179199219,
"epoch": 0.8075475936429494,
"grad_norm": 1.3515625,
"learning_rate": 0.0004945340512965401,
"loss": 5.7537,
"mean_token_accuracy": 0.1645844742655754,
"num_tokens": 16050992.0,
"step": 7190
},
{
"entropy": 5.611382865905762,
"epoch": 0.8081091705509069,
"grad_norm": 1.390625,
"learning_rate": 0.0004945252520193443,
"loss": 5.5678,
"mean_token_accuracy": 0.16939788907766343,
"num_tokens": 16062309.0,
"step": 7195
},
{
"entropy": 5.717952156066895,
"epoch": 0.8086707474588645,
"grad_norm": 1.4453125,
"learning_rate": 0.0004945164457523795,
"loss": 5.6069,
"mean_token_accuracy": 0.17542664408683778,
"num_tokens": 16073079.0,
"step": 7200
},
{
"entropy": 5.636781930923462,
"epoch": 0.809232324366822,
"grad_norm": 1.296875,
"learning_rate": 0.0004945076324959258,
"loss": 5.4983,
"mean_token_accuracy": 0.17701818495988847,
"num_tokens": 16082885.0,
"step": 7205
},
{
"entropy": 5.523493957519531,
"epoch": 0.8097939012747796,
"grad_norm": 1.2265625,
"learning_rate": 0.0004944988122502641,
"loss": 5.4604,
"mean_token_accuracy": 0.18318521678447724,
"num_tokens": 16094173.0,
"step": 7210
},
{
"entropy": 5.606122255325317,
"epoch": 0.8103554781827371,
"grad_norm": 1.46875,
"learning_rate": 0.000494489985015675,
"loss": 5.4997,
"mean_token_accuracy": 0.17629220932722092,
"num_tokens": 16106351.0,
"step": 7215
},
{
"entropy": 5.518839168548584,
"epoch": 0.8109170550906947,
"grad_norm": 1.3515625,
"learning_rate": 0.0004944811507924396,
"loss": 5.4944,
"mean_token_accuracy": 0.17697875201702118,
"num_tokens": 16116892.0,
"step": 7220
},
{
"entropy": 5.711473798751831,
"epoch": 0.8114786319986522,
"grad_norm": 1.5234375,
"learning_rate": 0.0004944723095808393,
"loss": 5.6444,
"mean_token_accuracy": 0.17179988324642181,
"num_tokens": 16128316.0,
"step": 7225
},
{
"entropy": 5.62243881225586,
"epoch": 0.8120402089066098,
"grad_norm": 1.3203125,
"learning_rate": 0.0004944634613811555,
"loss": 5.4668,
"mean_token_accuracy": 0.17707161903381347,
"num_tokens": 16139245.0,
"step": 7230
},
{
"entropy": 5.573305988311768,
"epoch": 0.8126017858145673,
"grad_norm": 1.5078125,
"learning_rate": 0.00049445460619367,
"loss": 5.5134,
"mean_token_accuracy": 0.1800749585032463,
"num_tokens": 16150893.0,
"step": 7235
},
{
"entropy": 5.689392137527466,
"epoch": 0.8131633627225249,
"grad_norm": 1.7109375,
"learning_rate": 0.0004944457440186648,
"loss": 5.6641,
"mean_token_accuracy": 0.1602585047483444,
"num_tokens": 16163430.0,
"step": 7240
},
{
"entropy": 5.752180337905884,
"epoch": 0.8137249396304824,
"grad_norm": 1.2734375,
"learning_rate": 0.0004944368748564219,
"loss": 5.5326,
"mean_token_accuracy": 0.17394165098667144,
"num_tokens": 16173954.0,
"step": 7245
},
{
"entropy": 5.625596761703491,
"epoch": 0.81428651653844,
"grad_norm": 1.2578125,
"learning_rate": 0.0004944279987072238,
"loss": 5.5994,
"mean_token_accuracy": 0.1711202770471573,
"num_tokens": 16185491.0,
"step": 7250
},
{
"entropy": 5.513287830352783,
"epoch": 0.8148480934463975,
"grad_norm": 1.4296875,
"learning_rate": 0.0004944191155713531,
"loss": 5.4368,
"mean_token_accuracy": 0.18979976773262025,
"num_tokens": 16195301.0,
"step": 7255
},
{
"entropy": 5.625479507446289,
"epoch": 0.8154096703543551,
"grad_norm": 1.3125,
"learning_rate": 0.0004944102254490928,
"loss": 5.5718,
"mean_token_accuracy": 0.16931443363428117,
"num_tokens": 16206620.0,
"step": 7260
},
{
"entropy": 5.666644763946533,
"epoch": 0.8159712472623126,
"grad_norm": 1.2578125,
"learning_rate": 0.0004944013283407257,
"loss": 5.6648,
"mean_token_accuracy": 0.1700120151042938,
"num_tokens": 16218693.0,
"step": 7265
},
{
"entropy": 5.696663188934326,
"epoch": 0.8165328241702701,
"grad_norm": 1.375,
"learning_rate": 0.0004943924242465352,
"loss": 5.5333,
"mean_token_accuracy": 0.17683588862419128,
"num_tokens": 16229124.0,
"step": 7270
},
{
"entropy": 5.598269510269165,
"epoch": 0.8170944010782276,
"grad_norm": 1.28125,
"learning_rate": 0.0004943835131668049,
"loss": 5.5159,
"mean_token_accuracy": 0.17627435326576232,
"num_tokens": 16239999.0,
"step": 7275
},
{
"entropy": 5.590861892700195,
"epoch": 0.8176559779861852,
"grad_norm": 1.6875,
"learning_rate": 0.0004943745951018184,
"loss": 5.5899,
"mean_token_accuracy": 0.1771497517824173,
"num_tokens": 16251027.0,
"step": 7280
},
{
"entropy": 5.644129705429077,
"epoch": 0.8182175548941427,
"grad_norm": 1.90625,
"learning_rate": 0.0004943656700518597,
"loss": 5.562,
"mean_token_accuracy": 0.17446047216653823,
"num_tokens": 16262924.0,
"step": 7285
},
{
"entropy": 5.777271699905396,
"epoch": 0.8187791318021003,
"grad_norm": 1.9453125,
"learning_rate": 0.0004943567380172129,
"loss": 5.6598,
"mean_token_accuracy": 0.17346756011247635,
"num_tokens": 16274481.0,
"step": 7290
},
{
"entropy": 5.663027095794678,
"epoch": 0.8193407087100578,
"grad_norm": 1.65625,
"learning_rate": 0.0004943477989981627,
"loss": 5.5872,
"mean_token_accuracy": 0.17727357149124146,
"num_tokens": 16285214.0,
"step": 7295
},
{
"entropy": 5.638958692550659,
"epoch": 0.8199022856180154,
"grad_norm": 1.3828125,
"learning_rate": 0.0004943388529949933,
"loss": 5.6549,
"mean_token_accuracy": 0.17275621443986894,
"num_tokens": 16296077.0,
"step": 7300
},
{
"entropy": 5.694351625442505,
"epoch": 0.8204638625259729,
"grad_norm": 1.4765625,
"learning_rate": 0.0004943299000079899,
"loss": 5.5658,
"mean_token_accuracy": 0.17576992213726045,
"num_tokens": 16307514.0,
"step": 7305
},
{
"entropy": 5.664661645889282,
"epoch": 0.8210254394339305,
"grad_norm": 1.1953125,
"learning_rate": 0.0004943209400374373,
"loss": 5.5648,
"mean_token_accuracy": 0.17784671485424042,
"num_tokens": 16318734.0,
"step": 7310
},
{
"entropy": 5.544511318206787,
"epoch": 0.821587016341888,
"grad_norm": 1.234375,
"learning_rate": 0.000494311973083621,
"loss": 5.6007,
"mean_token_accuracy": 0.1734233930706978,
"num_tokens": 16329057.0,
"step": 7315
},
{
"entropy": 5.614696598052978,
"epoch": 0.8221485932498456,
"grad_norm": 1.4453125,
"learning_rate": 0.0004943029991468264,
"loss": 5.5067,
"mean_token_accuracy": 0.18081510961055755,
"num_tokens": 16339646.0,
"step": 7320
},
{
"entropy": 5.686064720153809,
"epoch": 0.8227101701578031,
"grad_norm": 1.421875,
"learning_rate": 0.0004942940182273393,
"loss": 5.5808,
"mean_token_accuracy": 0.17242059260606765,
"num_tokens": 16349783.0,
"step": 7325
},
{
"entropy": 5.667108964920044,
"epoch": 0.8232717470657607,
"grad_norm": 1.421875,
"learning_rate": 0.0004942850303254455,
"loss": 5.5213,
"mean_token_accuracy": 0.17718788385391235,
"num_tokens": 16361078.0,
"step": 7330
},
{
"entropy": 5.5600062847137455,
"epoch": 0.8238333239737182,
"grad_norm": 1.2421875,
"learning_rate": 0.0004942760354414313,
"loss": 5.5197,
"mean_token_accuracy": 0.1786741554737091,
"num_tokens": 16371308.0,
"step": 7335
},
{
"entropy": 5.631235837936401,
"epoch": 0.8243949008816758,
"grad_norm": 1.375,
"learning_rate": 0.0004942670335755831,
"loss": 5.5776,
"mean_token_accuracy": 0.17667640298604964,
"num_tokens": 16382682.0,
"step": 7340
},
{
"entropy": 5.658403491973877,
"epoch": 0.8249564777896333,
"grad_norm": 1.1953125,
"learning_rate": 0.0004942580247281876,
"loss": 5.5919,
"mean_token_accuracy": 0.17763372212648393,
"num_tokens": 16393715.0,
"step": 7345
},
{
"entropy": 5.522956800460816,
"epoch": 0.8255180546975909,
"grad_norm": 1.53125,
"learning_rate": 0.0004942490088995315,
"loss": 5.5685,
"mean_token_accuracy": 0.17523156702518464,
"num_tokens": 16406332.0,
"step": 7350
},
{
"entropy": 5.615347194671631,
"epoch": 0.8260796316055484,
"grad_norm": 1.140625,
"learning_rate": 0.0004942399860899018,
"loss": 5.5081,
"mean_token_accuracy": 0.17780206948518754,
"num_tokens": 16417702.0,
"step": 7355
},
{
"entropy": 5.634275007247925,
"epoch": 0.826641208513506,
"grad_norm": 1.5625,
"learning_rate": 0.0004942309562995862,
"loss": 5.546,
"mean_token_accuracy": 0.1753668487071991,
"num_tokens": 16428497.0,
"step": 7360
},
{
"entropy": 5.652381181716919,
"epoch": 0.8272027854214634,
"grad_norm": 1.46875,
"learning_rate": 0.0004942219195288716,
"loss": 5.5574,
"mean_token_accuracy": 0.17919475287199021,
"num_tokens": 16440637.0,
"step": 7365
},
{
"entropy": 5.643956327438355,
"epoch": 0.827764362329421,
"grad_norm": 1.5625,
"learning_rate": 0.0004942128757780463,
"loss": 5.5779,
"mean_token_accuracy": 0.16893937960267066,
"num_tokens": 16453213.0,
"step": 7370
},
{
"entropy": 5.6806666374206545,
"epoch": 0.8283259392373785,
"grad_norm": 1.53125,
"learning_rate": 0.0004942038250473979,
"loss": 5.5881,
"mean_token_accuracy": 0.17754160463809968,
"num_tokens": 16464345.0,
"step": 7375
},
{
"entropy": 5.599323654174805,
"epoch": 0.8288875161453361,
"grad_norm": 1.3125,
"learning_rate": 0.0004941947673372147,
"loss": 5.4625,
"mean_token_accuracy": 0.18206140398979187,
"num_tokens": 16475399.0,
"step": 7380
},
{
"entropy": 5.54746823310852,
"epoch": 0.8294490930532936,
"grad_norm": 1.2734375,
"learning_rate": 0.0004941857026477851,
"loss": 5.4422,
"mean_token_accuracy": 0.17859069705009462,
"num_tokens": 16485439.0,
"step": 7385
},
{
"entropy": 5.557292175292969,
"epoch": 0.8300106699612512,
"grad_norm": 1.296875,
"learning_rate": 0.0004941766309793978,
"loss": 5.4375,
"mean_token_accuracy": 0.18442922979593276,
"num_tokens": 16496765.0,
"step": 7390
},
{
"entropy": 5.522080659866333,
"epoch": 0.8305722468692087,
"grad_norm": 1.171875,
"learning_rate": 0.0004941675523323414,
"loss": 5.4854,
"mean_token_accuracy": 0.1782712459564209,
"num_tokens": 16509199.0,
"step": 7395
},
{
"entropy": 5.662637615203858,
"epoch": 0.8311338237771663,
"grad_norm": 1.171875,
"learning_rate": 0.0004941584667069052,
"loss": 5.6228,
"mean_token_accuracy": 0.17423093616962432,
"num_tokens": 16520073.0,
"step": 7400
},
{
"entropy": 5.573271036148071,
"epoch": 0.8316954006851238,
"grad_norm": 1.1171875,
"learning_rate": 0.0004941493741033784,
"loss": 5.5116,
"mean_token_accuracy": 0.17435383349657058,
"num_tokens": 16530919.0,
"step": 7405
},
{
"entropy": 5.64184193611145,
"epoch": 0.8322569775930814,
"grad_norm": 1.59375,
"learning_rate": 0.0004941402745220507,
"loss": 5.5352,
"mean_token_accuracy": 0.17571932673454285,
"num_tokens": 16542591.0,
"step": 7410
},
{
"entropy": 5.65180344581604,
"epoch": 0.8328185545010389,
"grad_norm": 1.390625,
"learning_rate": 0.0004941311679632115,
"loss": 5.5356,
"mean_token_accuracy": 0.1789597228169441,
"num_tokens": 16554984.0,
"step": 7415
},
{
"entropy": 5.5525611400604244,
"epoch": 0.8333801314089965,
"grad_norm": 1.1796875,
"learning_rate": 0.000494122054427151,
"loss": 5.457,
"mean_token_accuracy": 0.1813283607363701,
"num_tokens": 16566598.0,
"step": 7420
},
{
"entropy": 5.675726747512817,
"epoch": 0.833941708316954,
"grad_norm": 1.4609375,
"learning_rate": 0.0004941129339141592,
"loss": 5.5919,
"mean_token_accuracy": 0.17330875098705292,
"num_tokens": 16577868.0,
"step": 7425
},
{
"entropy": 5.639889430999756,
"epoch": 0.8345032852249116,
"grad_norm": 1.1640625,
"learning_rate": 0.0004941038064245267,
"loss": 5.5306,
"mean_token_accuracy": 0.17521563321352004,
"num_tokens": 16587891.0,
"step": 7430
},
{
"entropy": 5.590980434417725,
"epoch": 0.8350648621328691,
"grad_norm": 1.25,
"learning_rate": 0.000494094671958544,
"loss": 5.5002,
"mean_token_accuracy": 0.18300278037786483,
"num_tokens": 16599924.0,
"step": 7435
},
{
"entropy": 5.536389112472534,
"epoch": 0.8356264390408267,
"grad_norm": 1.1953125,
"learning_rate": 0.0004940855305165019,
"loss": 5.4341,
"mean_token_accuracy": 0.18307463973760604,
"num_tokens": 16609958.0,
"step": 7440
},
{
"entropy": 5.602199649810791,
"epoch": 0.8361880159487842,
"grad_norm": 1.21875,
"learning_rate": 0.0004940763820986916,
"loss": 5.5691,
"mean_token_accuracy": 0.17436801195144652,
"num_tokens": 16621159.0,
"step": 7445
},
{
"entropy": 5.658262634277344,
"epoch": 0.8367495928567418,
"grad_norm": 1.40625,
"learning_rate": 0.0004940672267054044,
"loss": 5.6578,
"mean_token_accuracy": 0.1736158013343811,
"num_tokens": 16632263.0,
"step": 7450
},
{
"entropy": 5.658412265777588,
"epoch": 0.8373111697646993,
"grad_norm": 1.2109375,
"learning_rate": 0.0004940580643369316,
"loss": 5.5403,
"mean_token_accuracy": 0.17218570560216903,
"num_tokens": 16644117.0,
"step": 7455
},
{
"entropy": 5.513913202285766,
"epoch": 0.8378727466726568,
"grad_norm": 2.0625,
"learning_rate": 0.0004940488949935651,
"loss": 5.4624,
"mean_token_accuracy": 0.1835022285580635,
"num_tokens": 16654529.0,
"step": 7460
},
{
"entropy": 5.592473459243775,
"epoch": 0.8384343235806143,
"grad_norm": 1.328125,
"learning_rate": 0.0004940397186755968,
"loss": 5.4831,
"mean_token_accuracy": 0.1790134757757187,
"num_tokens": 16666191.0,
"step": 7465
},
{
"entropy": 5.62993049621582,
"epoch": 0.8389959004885719,
"grad_norm": 1.375,
"learning_rate": 0.0004940305353833191,
"loss": 5.468,
"mean_token_accuracy": 0.1802850767970085,
"num_tokens": 16677252.0,
"step": 7470
},
{
"entropy": 5.558318948745727,
"epoch": 0.8395574773965294,
"grad_norm": 1.4140625,
"learning_rate": 0.000494021345117024,
"loss": 5.5245,
"mean_token_accuracy": 0.177115735411644,
"num_tokens": 16689487.0,
"step": 7475
},
{
"entropy": 5.517132043838501,
"epoch": 0.840119054304487,
"grad_norm": 1.15625,
"learning_rate": 0.0004940121478770044,
"loss": 5.4129,
"mean_token_accuracy": 0.18159824013710021,
"num_tokens": 16699955.0,
"step": 7480
},
{
"entropy": 5.480426025390625,
"epoch": 0.8406806312124445,
"grad_norm": 1.09375,
"learning_rate": 0.0004940029436635531,
"loss": 5.4677,
"mean_token_accuracy": 0.1777705207467079,
"num_tokens": 16711071.0,
"step": 7485
},
{
"entropy": 5.5705982685089115,
"epoch": 0.8412422081204021,
"grad_norm": 1.2734375,
"learning_rate": 0.0004939937324769631,
"loss": 5.4843,
"mean_token_accuracy": 0.18347815722227095,
"num_tokens": 16722115.0,
"step": 7490
},
{
"entropy": 5.575274133682251,
"epoch": 0.8418037850283596,
"grad_norm": 1.1875,
"learning_rate": 0.0004939845143175276,
"loss": 5.5142,
"mean_token_accuracy": 0.17353819608688353,
"num_tokens": 16733146.0,
"step": 7495
},
{
"entropy": 5.551184749603271,
"epoch": 0.8423653619363172,
"grad_norm": 1.328125,
"learning_rate": 0.0004939752891855404,
"loss": 5.4961,
"mean_token_accuracy": 0.17520865350961684,
"num_tokens": 16745543.0,
"step": 7500
},
{
"entropy": 5.610068368911743,
"epoch": 0.8429269388442747,
"grad_norm": 1.328125,
"learning_rate": 0.0004939660570812951,
"loss": 5.5468,
"mean_token_accuracy": 0.178034807741642,
"num_tokens": 16755782.0,
"step": 7505
},
{
"entropy": 5.667875194549561,
"epoch": 0.8434885157522323,
"grad_norm": 1.5390625,
"learning_rate": 0.0004939568180050855,
"loss": 5.5843,
"mean_token_accuracy": 0.17520271986722946,
"num_tokens": 16767509.0,
"step": 7510
},
{
"entropy": 5.54585223197937,
"epoch": 0.8440500926601898,
"grad_norm": 1.34375,
"learning_rate": 0.0004939475719572059,
"loss": 5.473,
"mean_token_accuracy": 0.1772061690688133,
"num_tokens": 16779257.0,
"step": 7515
},
{
"entropy": 5.511939907073975,
"epoch": 0.8446116695681474,
"grad_norm": 1.46875,
"learning_rate": 0.0004939383189379507,
"loss": 5.5053,
"mean_token_accuracy": 0.18014951050281525,
"num_tokens": 16790769.0,
"step": 7520
},
{
"entropy": 5.564608478546143,
"epoch": 0.8451732464761049,
"grad_norm": 1.6796875,
"learning_rate": 0.0004939290589476145,
"loss": 5.4972,
"mean_token_accuracy": 0.17935482561588287,
"num_tokens": 16800054.0,
"step": 7525
},
{
"entropy": 5.71444001197815,
"epoch": 0.8457348233840625,
"grad_norm": 1.2734375,
"learning_rate": 0.0004939197919864922,
"loss": 5.5906,
"mean_token_accuracy": 0.17285536974668503,
"num_tokens": 16811485.0,
"step": 7530
},
{
"entropy": 5.725357866287231,
"epoch": 0.84629640029202,
"grad_norm": 1.6015625,
"learning_rate": 0.0004939105180548788,
"loss": 5.6387,
"mean_token_accuracy": 0.1729096233844757,
"num_tokens": 16822683.0,
"step": 7535
},
{
"entropy": 5.642815732955933,
"epoch": 0.8468579771999776,
"grad_norm": 1.3359375,
"learning_rate": 0.0004939012371530696,
"loss": 5.5223,
"mean_token_accuracy": 0.17263311743736268,
"num_tokens": 16834200.0,
"step": 7540
},
{
"entropy": 5.683886909484864,
"epoch": 0.8474195541079351,
"grad_norm": 1.34375,
"learning_rate": 0.0004938919492813601,
"loss": 5.5694,
"mean_token_accuracy": 0.17125529646873475,
"num_tokens": 16846399.0,
"step": 7545
},
{
"entropy": 5.7869508266448975,
"epoch": 0.8479811310158927,
"grad_norm": 1.2578125,
"learning_rate": 0.0004938826544400461,
"loss": 5.7227,
"mean_token_accuracy": 0.1707400768995285,
"num_tokens": 16858322.0,
"step": 7550
},
{
"entropy": 5.636294269561768,
"epoch": 0.8485427079238502,
"grad_norm": 1.7734375,
"learning_rate": 0.0004938733526294234,
"loss": 5.5338,
"mean_token_accuracy": 0.179585063457489,
"num_tokens": 16869537.0,
"step": 7555
},
{
"entropy": 5.700771713256836,
"epoch": 0.8491042848318077,
"grad_norm": 1.3125,
"learning_rate": 0.0004938640438497883,
"loss": 5.5867,
"mean_token_accuracy": 0.17592836767435074,
"num_tokens": 16881566.0,
"step": 7560
},
{
"entropy": 5.580954933166504,
"epoch": 0.8496658617397652,
"grad_norm": 1.3984375,
"learning_rate": 0.0004938547281014372,
"loss": 5.4869,
"mean_token_accuracy": 0.18331398367881774,
"num_tokens": 16892959.0,
"step": 7565
},
{
"entropy": 5.624685525894165,
"epoch": 0.8502274386477228,
"grad_norm": 1.5390625,
"learning_rate": 0.0004938454053846666,
"loss": 5.5622,
"mean_token_accuracy": 0.17130545526742935,
"num_tokens": 16904809.0,
"step": 7570
},
{
"entropy": 5.668943977355957,
"epoch": 0.8507890155556803,
"grad_norm": 1.5078125,
"learning_rate": 0.0004938360756997735,
"loss": 5.5898,
"mean_token_accuracy": 0.16893067359924316,
"num_tokens": 16916969.0,
"step": 7575
},
{
"entropy": 5.544728565216064,
"epoch": 0.8513505924636379,
"grad_norm": 1.234375,
"learning_rate": 0.0004938267390470547,
"loss": 5.4715,
"mean_token_accuracy": 0.17910400182008743,
"num_tokens": 16928253.0,
"step": 7580
},
{
"entropy": 5.5811975479125975,
"epoch": 0.8519121693715954,
"grad_norm": 1.453125,
"learning_rate": 0.0004938173954268079,
"loss": 5.5193,
"mean_token_accuracy": 0.17128382474184037,
"num_tokens": 16939875.0,
"step": 7585
},
{
"entropy": 5.705540990829467,
"epoch": 0.852473746279553,
"grad_norm": 1.234375,
"learning_rate": 0.0004938080448393302,
"loss": 5.6293,
"mean_token_accuracy": 0.17560341358184814,
"num_tokens": 16950173.0,
"step": 7590
},
{
"entropy": 5.571158695220947,
"epoch": 0.8530353231875105,
"grad_norm": 1.25,
"learning_rate": 0.0004937986872849195,
"loss": 5.5521,
"mean_token_accuracy": 0.17518737763166428,
"num_tokens": 16960933.0,
"step": 7595
},
{
"entropy": 5.540823221206665,
"epoch": 0.8535969000954681,
"grad_norm": 1.328125,
"learning_rate": 0.0004937893227638738,
"loss": 5.4403,
"mean_token_accuracy": 0.17984343469142913,
"num_tokens": 16972284.0,
"step": 7600
},
{
"entropy": 5.637776184082031,
"epoch": 0.8541584770034256,
"grad_norm": 1.3671875,
"learning_rate": 0.0004937799512764912,
"loss": 5.4656,
"mean_token_accuracy": 0.18278502374887468,
"num_tokens": 16983541.0,
"step": 7605
},
{
"entropy": 5.654651689529419,
"epoch": 0.8547200539113832,
"grad_norm": 1.3515625,
"learning_rate": 0.00049377057282307,
"loss": 5.6113,
"mean_token_accuracy": 0.17200611531734467,
"num_tokens": 16993773.0,
"step": 7610
},
{
"entropy": 5.6675492286682125,
"epoch": 0.8552816308193407,
"grad_norm": 1.3984375,
"learning_rate": 0.000493761187403909,
"loss": 5.6082,
"mean_token_accuracy": 0.1749629333615303,
"num_tokens": 17005287.0,
"step": 7615
},
{
"entropy": 5.670874834060669,
"epoch": 0.8558432077272983,
"grad_norm": 1.7734375,
"learning_rate": 0.000493751795019307,
"loss": 5.5302,
"mean_token_accuracy": 0.17562350630760193,
"num_tokens": 17015664.0,
"step": 7620
},
{
"entropy": 5.589368152618408,
"epoch": 0.8564047846352558,
"grad_norm": 1.3203125,
"learning_rate": 0.0004937423956695629,
"loss": 5.4452,
"mean_token_accuracy": 0.17568252831697465,
"num_tokens": 17026876.0,
"step": 7625
},
{
"entropy": 5.534160041809082,
"epoch": 0.8569663615432134,
"grad_norm": 1.453125,
"learning_rate": 0.0004937329893549762,
"loss": 5.4838,
"mean_token_accuracy": 0.19195917546749114,
"num_tokens": 17037321.0,
"step": 7630
},
{
"entropy": 5.635166215896606,
"epoch": 0.8575279384511709,
"grad_norm": 1.6640625,
"learning_rate": 0.0004937235760758462,
"loss": 5.5699,
"mean_token_accuracy": 0.17798927128314973,
"num_tokens": 17047506.0,
"step": 7635
},
{
"entropy": 5.689205980300903,
"epoch": 0.8580895153591285,
"grad_norm": 1.2578125,
"learning_rate": 0.0004937141558324726,
"loss": 5.6256,
"mean_token_accuracy": 0.17240417003631592,
"num_tokens": 17059710.0,
"step": 7640
},
{
"entropy": 5.695173454284668,
"epoch": 0.858651092267086,
"grad_norm": 1.6484375,
"learning_rate": 0.0004937047286251556,
"loss": 5.6207,
"mean_token_accuracy": 0.16439640372991562,
"num_tokens": 17070956.0,
"step": 7645
},
{
"entropy": 5.697442436218262,
"epoch": 0.8592126691750436,
"grad_norm": 1.34375,
"learning_rate": 0.0004936952944541952,
"loss": 5.5905,
"mean_token_accuracy": 0.17429565787315368,
"num_tokens": 17081404.0,
"step": 7650
},
{
"entropy": 5.725706768035889,
"epoch": 0.859774246083001,
"grad_norm": 1.21875,
"learning_rate": 0.0004936858533198917,
"loss": 5.5997,
"mean_token_accuracy": 0.17281186878681182,
"num_tokens": 17092692.0,
"step": 7655
},
{
"entropy": 5.556993293762207,
"epoch": 0.8603358229909586,
"grad_norm": 1.5703125,
"learning_rate": 0.000493676405222546,
"loss": 5.4512,
"mean_token_accuracy": 0.18004993349313736,
"num_tokens": 17103730.0,
"step": 7660
},
{
"entropy": 5.641902351379395,
"epoch": 0.8608973998989161,
"grad_norm": 1.390625,
"learning_rate": 0.0004936669501624586,
"loss": 5.5148,
"mean_token_accuracy": 0.17758802622556685,
"num_tokens": 17115086.0,
"step": 7665
},
{
"entropy": 5.647478532791138,
"epoch": 0.8614589768068737,
"grad_norm": 1.34375,
"learning_rate": 0.0004936574881399308,
"loss": 5.529,
"mean_token_accuracy": 0.17794135510921477,
"num_tokens": 17125928.0,
"step": 7670
},
{
"entropy": 5.625667524337769,
"epoch": 0.8620205537148312,
"grad_norm": 1.234375,
"learning_rate": 0.0004936480191552637,
"loss": 5.5538,
"mean_token_accuracy": 0.17434030771255493,
"num_tokens": 17137328.0,
"step": 7675
},
{
"entropy": 5.664294862747193,
"epoch": 0.8625821306227888,
"grad_norm": 1.359375,
"learning_rate": 0.0004936385432087589,
"loss": 5.5297,
"mean_token_accuracy": 0.17809005081653595,
"num_tokens": 17148274.0,
"step": 7680
},
{
"entropy": 5.654427337646484,
"epoch": 0.8631437075307463,
"grad_norm": 1.296875,
"learning_rate": 0.0004936290603007181,
"loss": 5.5596,
"mean_token_accuracy": 0.17495724856853484,
"num_tokens": 17158834.0,
"step": 7685
},
{
"entropy": 5.623084259033203,
"epoch": 0.8637052844387039,
"grad_norm": 1.234375,
"learning_rate": 0.0004936195704314432,
"loss": 5.4305,
"mean_token_accuracy": 0.17616414725780488,
"num_tokens": 17169276.0,
"step": 7690
},
{
"entropy": 5.582148790359497,
"epoch": 0.8642668613466614,
"grad_norm": 1.59375,
"learning_rate": 0.0004936100736012363,
"loss": 5.4733,
"mean_token_accuracy": 0.1812399610877037,
"num_tokens": 17180642.0,
"step": 7695
},
{
"entropy": 5.53903341293335,
"epoch": 0.864828438254619,
"grad_norm": 1.421875,
"learning_rate": 0.0004936005698104,
"loss": 5.5384,
"mean_token_accuracy": 0.18119236081838608,
"num_tokens": 17192477.0,
"step": 7700
},
{
"entropy": 5.702814388275146,
"epoch": 0.8653900151625765,
"grad_norm": 1.5703125,
"learning_rate": 0.0004935910590592367,
"loss": 5.6577,
"mean_token_accuracy": 0.17858854085206985,
"num_tokens": 17202653.0,
"step": 7705
},
{
"entropy": 5.5267924785614015,
"epoch": 0.8659515920705341,
"grad_norm": 1.2265625,
"learning_rate": 0.0004935815413480494,
"loss": 5.4058,
"mean_token_accuracy": 0.18510970920324327,
"num_tokens": 17212607.0,
"step": 7710
},
{
"entropy": 5.637176418304444,
"epoch": 0.8665131689784916,
"grad_norm": 1.546875,
"learning_rate": 0.000493572016677141,
"loss": 5.5599,
"mean_token_accuracy": 0.18122070282697678,
"num_tokens": 17223701.0,
"step": 7715
},
{
"entropy": 5.649902868270874,
"epoch": 0.8670747458864492,
"grad_norm": 1.2734375,
"learning_rate": 0.0004935624850468148,
"loss": 5.4543,
"mean_token_accuracy": 0.1848278358578682,
"num_tokens": 17234878.0,
"step": 7720
},
{
"entropy": 5.545849084854126,
"epoch": 0.8676363227944067,
"grad_norm": 1.5859375,
"learning_rate": 0.0004935529464573742,
"loss": 5.5245,
"mean_token_accuracy": 0.1810195416212082,
"num_tokens": 17245813.0,
"step": 7725
},
{
"entropy": 5.616622447967529,
"epoch": 0.8681978997023643,
"grad_norm": 1.4609375,
"learning_rate": 0.0004935434009091232,
"loss": 5.6174,
"mean_token_accuracy": 0.17362616658210756,
"num_tokens": 17257378.0,
"step": 7730
},
{
"entropy": 5.615325307846069,
"epoch": 0.8687594766103218,
"grad_norm": 1.2734375,
"learning_rate": 0.0004935338484023655,
"loss": 5.5245,
"mean_token_accuracy": 0.17634663581848145,
"num_tokens": 17269108.0,
"step": 7735
},
{
"entropy": 5.630329751968384,
"epoch": 0.8693210535182794,
"grad_norm": 1.5390625,
"learning_rate": 0.0004935242889374052,
"loss": 5.5108,
"mean_token_accuracy": 0.17099627256393432,
"num_tokens": 17280643.0,
"step": 7740
},
{
"entropy": 5.731979513168335,
"epoch": 0.8698826304262369,
"grad_norm": 1.4375,
"learning_rate": 0.0004935147225145469,
"loss": 5.5552,
"mean_token_accuracy": 0.17678023278713226,
"num_tokens": 17291236.0,
"step": 7745
},
{
"entropy": 5.692935657501221,
"epoch": 0.8704442073341944,
"grad_norm": 1.3515625,
"learning_rate": 0.0004935051491340951,
"loss": 5.6523,
"mean_token_accuracy": 0.16854152977466583,
"num_tokens": 17303547.0,
"step": 7750
},
{
"entropy": 5.687366342544555,
"epoch": 0.8710057842421519,
"grad_norm": 1.421875,
"learning_rate": 0.0004934955687963546,
"loss": 5.6241,
"mean_token_accuracy": 0.16987596005201339,
"num_tokens": 17314803.0,
"step": 7755
},
{
"entropy": 5.609250402450561,
"epoch": 0.8715673611501095,
"grad_norm": 1.40625,
"learning_rate": 0.0004934859815016304,
"loss": 5.4931,
"mean_token_accuracy": 0.17658208757638932,
"num_tokens": 17326009.0,
"step": 7760
},
{
"entropy": 5.615665006637573,
"epoch": 0.872128938058067,
"grad_norm": 1.296875,
"learning_rate": 0.0004934763872502278,
"loss": 5.4989,
"mean_token_accuracy": 0.1838627815246582,
"num_tokens": 17337078.0,
"step": 7765
},
{
"entropy": 5.663925695419311,
"epoch": 0.8726905149660246,
"grad_norm": 1.6015625,
"learning_rate": 0.0004934667860424523,
"loss": 5.6273,
"mean_token_accuracy": 0.1725619524717331,
"num_tokens": 17349266.0,
"step": 7770
},
{
"entropy": 5.5061936378479,
"epoch": 0.8732520918739821,
"grad_norm": 1.296875,
"learning_rate": 0.0004934571778786096,
"loss": 5.415,
"mean_token_accuracy": 0.1802388235926628,
"num_tokens": 17360327.0,
"step": 7775
},
{
"entropy": 5.711481142044067,
"epoch": 0.8738136687819397,
"grad_norm": 1.484375,
"learning_rate": 0.0004934475627590057,
"loss": 5.579,
"mean_token_accuracy": 0.17152512520551683,
"num_tokens": 17372727.0,
"step": 7780
},
{
"entropy": 5.7374217987060545,
"epoch": 0.8743752456898972,
"grad_norm": 2.6875,
"learning_rate": 0.0004934379406839466,
"loss": 5.6376,
"mean_token_accuracy": 0.17415147572755812,
"num_tokens": 17384160.0,
"step": 7785
},
{
"entropy": 5.603996801376343,
"epoch": 0.8749368225978548,
"grad_norm": 1.25,
"learning_rate": 0.0004934283116537387,
"loss": 5.548,
"mean_token_accuracy": 0.17670674175024031,
"num_tokens": 17396009.0,
"step": 7790
},
{
"entropy": 5.627090883255005,
"epoch": 0.8754983995058123,
"grad_norm": 1.28125,
"learning_rate": 0.0004934186756686888,
"loss": 5.5572,
"mean_token_accuracy": 0.17788248509168625,
"num_tokens": 17406903.0,
"step": 7795
},
{
"entropy": 5.568345642089843,
"epoch": 0.8760599764137699,
"grad_norm": 1.515625,
"learning_rate": 0.0004934090327291034,
"loss": 5.4062,
"mean_token_accuracy": 0.18373901546001434,
"num_tokens": 17417694.0,
"step": 7800
},
{
"entropy": 5.58258056640625,
"epoch": 0.8766215533217274,
"grad_norm": 1.4296875,
"learning_rate": 0.0004933993828352898,
"loss": 5.6499,
"mean_token_accuracy": 0.17717401087284088,
"num_tokens": 17428731.0,
"step": 7805
},
{
"entropy": 5.5873980045318605,
"epoch": 0.877183130229685,
"grad_norm": 1.3203125,
"learning_rate": 0.0004933897259875551,
"loss": 5.4018,
"mean_token_accuracy": 0.18061320036649703,
"num_tokens": 17439242.0,
"step": 7810
},
{
"entropy": 5.63393006324768,
"epoch": 0.8777447071376425,
"grad_norm": 1.2890625,
"learning_rate": 0.0004933800621862069,
"loss": 5.5174,
"mean_token_accuracy": 0.17823126912117004,
"num_tokens": 17449980.0,
"step": 7815
},
{
"entropy": 5.709502840042115,
"epoch": 0.8783062840456001,
"grad_norm": 1.390625,
"learning_rate": 0.0004933703914315527,
"loss": 5.6121,
"mean_token_accuracy": 0.1679139956831932,
"num_tokens": 17464140.0,
"step": 7820
},
{
"entropy": 5.7078413486480715,
"epoch": 0.8788678609535576,
"grad_norm": 1.5703125,
"learning_rate": 0.0004933607137239006,
"loss": 5.627,
"mean_token_accuracy": 0.17199864089488984,
"num_tokens": 17476246.0,
"step": 7825
},
{
"entropy": 5.667858457565307,
"epoch": 0.8794294378615152,
"grad_norm": 1.359375,
"learning_rate": 0.0004933510290635587,
"loss": 5.6288,
"mean_token_accuracy": 0.1642494574189186,
"num_tokens": 17488169.0,
"step": 7830
},
{
"entropy": 5.613598489761353,
"epoch": 0.8799910147694727,
"grad_norm": 1.6640625,
"learning_rate": 0.0004933413374508353,
"loss": 5.4535,
"mean_token_accuracy": 0.1825757086277008,
"num_tokens": 17498465.0,
"step": 7835
},
{
"entropy": 5.598409128189087,
"epoch": 0.8805525916774303,
"grad_norm": 1.40625,
"learning_rate": 0.0004933316388860391,
"loss": 5.4515,
"mean_token_accuracy": 0.18811351358890532,
"num_tokens": 17509217.0,
"step": 7840
},
{
"entropy": 5.5887054920196535,
"epoch": 0.8811141685853877,
"grad_norm": 1.4296875,
"learning_rate": 0.0004933219333694787,
"loss": 5.4835,
"mean_token_accuracy": 0.1845836415886879,
"num_tokens": 17519851.0,
"step": 7845
},
{
"entropy": 5.6024613857269285,
"epoch": 0.8816757454933453,
"grad_norm": 1.3515625,
"learning_rate": 0.0004933122209014634,
"loss": 5.5613,
"mean_token_accuracy": 0.17829443216323854,
"num_tokens": 17532144.0,
"step": 7850
},
{
"entropy": 5.545286226272583,
"epoch": 0.8822373224013028,
"grad_norm": 1.3515625,
"learning_rate": 0.0004933025014823022,
"loss": 5.4638,
"mean_token_accuracy": 0.18536658734083175,
"num_tokens": 17542929.0,
"step": 7855
},
{
"entropy": 5.696783208847046,
"epoch": 0.8827988993092604,
"grad_norm": 1.3515625,
"learning_rate": 0.0004932927751123048,
"loss": 5.6418,
"mean_token_accuracy": 0.1682240843772888,
"num_tokens": 17554931.0,
"step": 7860
},
{
"entropy": 5.614430141448975,
"epoch": 0.8833604762172179,
"grad_norm": 1.2890625,
"learning_rate": 0.0004932830417917808,
"loss": 5.4358,
"mean_token_accuracy": 0.17923623323440552,
"num_tokens": 17566026.0,
"step": 7865
},
{
"entropy": 5.579472923278809,
"epoch": 0.8839220531251755,
"grad_norm": 1.3125,
"learning_rate": 0.00049327330152104,
"loss": 5.5445,
"mean_token_accuracy": 0.17528396099805832,
"num_tokens": 17578396.0,
"step": 7870
},
{
"entropy": 5.683245325088501,
"epoch": 0.884483630033133,
"grad_norm": 1.4140625,
"learning_rate": 0.0004932635543003927,
"loss": 5.5328,
"mean_token_accuracy": 0.17948647439479828,
"num_tokens": 17590559.0,
"step": 7875
},
{
"entropy": 5.639725875854492,
"epoch": 0.8850452069410906,
"grad_norm": 1.3984375,
"learning_rate": 0.0004932538001301491,
"loss": 5.5084,
"mean_token_accuracy": 0.1739957883954048,
"num_tokens": 17600833.0,
"step": 7880
},
{
"entropy": 5.563118839263916,
"epoch": 0.8856067838490481,
"grad_norm": 1.3203125,
"learning_rate": 0.0004932440390106199,
"loss": 5.5107,
"mean_token_accuracy": 0.17993877083063126,
"num_tokens": 17612166.0,
"step": 7885
},
{
"entropy": 5.629817819595337,
"epoch": 0.8861683607570057,
"grad_norm": 1.578125,
"learning_rate": 0.0004932342709421159,
"loss": 5.6097,
"mean_token_accuracy": 0.17699260264635086,
"num_tokens": 17623070.0,
"step": 7890
},
{
"entropy": 5.711673831939697,
"epoch": 0.8867299376649632,
"grad_norm": 1.3828125,
"learning_rate": 0.000493224495924948,
"loss": 5.6642,
"mean_token_accuracy": 0.1751502826809883,
"num_tokens": 17633319.0,
"step": 7895
},
{
"entropy": 5.621243238449097,
"epoch": 0.8872915145729208,
"grad_norm": 1.84375,
"learning_rate": 0.0004932147139594276,
"loss": 5.4792,
"mean_token_accuracy": 0.1780954748392105,
"num_tokens": 17643818.0,
"step": 7900
},
{
"entropy": 5.584720182418823,
"epoch": 0.8878530914808783,
"grad_norm": 1.5078125,
"learning_rate": 0.0004932049250458659,
"loss": 5.4764,
"mean_token_accuracy": 0.17586888521909713,
"num_tokens": 17655665.0,
"step": 7905
},
{
"entropy": 5.57195897102356,
"epoch": 0.8884146683888359,
"grad_norm": 1.796875,
"learning_rate": 0.0004931951291845749,
"loss": 5.4033,
"mean_token_accuracy": 0.18850912749767304,
"num_tokens": 17666200.0,
"step": 7910
},
{
"entropy": 5.5873377323150635,
"epoch": 0.8889762452967934,
"grad_norm": 1.3203125,
"learning_rate": 0.0004931853263758663,
"loss": 5.5917,
"mean_token_accuracy": 0.1713733896613121,
"num_tokens": 17677299.0,
"step": 7915
},
{
"entropy": 5.623600816726684,
"epoch": 0.889537822204751,
"grad_norm": 1.2734375,
"learning_rate": 0.0004931755166200523,
"loss": 5.447,
"mean_token_accuracy": 0.18628597259521484,
"num_tokens": 17687870.0,
"step": 7920
},
{
"entropy": 5.617124795913696,
"epoch": 0.8900993991127085,
"grad_norm": 1.2578125,
"learning_rate": 0.0004931656999174452,
"loss": 5.4821,
"mean_token_accuracy": 0.1771150842308998,
"num_tokens": 17698853.0,
"step": 7925
},
{
"entropy": 5.584203290939331,
"epoch": 0.8906609760206661,
"grad_norm": 1.5,
"learning_rate": 0.0004931558762683577,
"loss": 5.5337,
"mean_token_accuracy": 0.18087521344423294,
"num_tokens": 17709897.0,
"step": 7930
},
{
"entropy": 5.599167251586914,
"epoch": 0.8912225529286236,
"grad_norm": 1.515625,
"learning_rate": 0.0004931460456731024,
"loss": 5.5389,
"mean_token_accuracy": 0.17602657079696654,
"num_tokens": 17720974.0,
"step": 7935
},
{
"entropy": 5.811507320404052,
"epoch": 0.8917841298365812,
"grad_norm": 1.421875,
"learning_rate": 0.0004931362081319925,
"loss": 5.6923,
"mean_token_accuracy": 0.17019293010234832,
"num_tokens": 17734449.0,
"step": 7940
},
{
"entropy": 5.525227355957031,
"epoch": 0.8923457067445386,
"grad_norm": 1.7578125,
"learning_rate": 0.0004931263636453409,
"loss": 5.4196,
"mean_token_accuracy": 0.18271054327487946,
"num_tokens": 17744735.0,
"step": 7945
},
{
"entropy": 5.512601280212403,
"epoch": 0.8929072836524962,
"grad_norm": 1.2578125,
"learning_rate": 0.0004931165122134616,
"loss": 5.4592,
"mean_token_accuracy": 0.1792737439274788,
"num_tokens": 17755641.0,
"step": 7950
},
{
"entropy": 5.646299123764038,
"epoch": 0.8934688605604537,
"grad_norm": 1.1953125,
"learning_rate": 0.0004931066538366677,
"loss": 5.558,
"mean_token_accuracy": 0.1742606371641159,
"num_tokens": 17767621.0,
"step": 7955
},
{
"entropy": 5.780982971191406,
"epoch": 0.8940304374684113,
"grad_norm": 1.4609375,
"learning_rate": 0.0004930967885152734,
"loss": 5.6773,
"mean_token_accuracy": 0.17321367263793946,
"num_tokens": 17779507.0,
"step": 7960
},
{
"entropy": 5.6253196716308596,
"epoch": 0.8945920143763688,
"grad_norm": 1.1640625,
"learning_rate": 0.0004930869162495929,
"loss": 5.498,
"mean_token_accuracy": 0.17570409774780274,
"num_tokens": 17790371.0,
"step": 7965
},
{
"entropy": 5.60542311668396,
"epoch": 0.8951535912843264,
"grad_norm": 1.28125,
"learning_rate": 0.0004930770370399403,
"loss": 5.5254,
"mean_token_accuracy": 0.18335310220718384,
"num_tokens": 17800903.0,
"step": 7970
},
{
"entropy": 5.690413856506348,
"epoch": 0.8957151681922839,
"grad_norm": 1.40625,
"learning_rate": 0.0004930671508866303,
"loss": 5.6967,
"mean_token_accuracy": 0.1646069772541523,
"num_tokens": 17812053.0,
"step": 7975
},
{
"entropy": 5.607906246185303,
"epoch": 0.8962767451002415,
"grad_norm": 1.4609375,
"learning_rate": 0.0004930572577899776,
"loss": 5.4354,
"mean_token_accuracy": 0.1805379405617714,
"num_tokens": 17822113.0,
"step": 7980
},
{
"entropy": 5.564527559280395,
"epoch": 0.896838322008199,
"grad_norm": 1.9453125,
"learning_rate": 0.0004930473577502972,
"loss": 5.4785,
"mean_token_accuracy": 0.17969129532575606,
"num_tokens": 17832741.0,
"step": 7985
},
{
"entropy": 5.442588520050049,
"epoch": 0.8973998989161566,
"grad_norm": 1.421875,
"learning_rate": 0.0004930374507679044,
"loss": 5.4005,
"mean_token_accuracy": 0.17895758152008057,
"num_tokens": 17842968.0,
"step": 7990
},
{
"entropy": 5.706737661361695,
"epoch": 0.8979614758241141,
"grad_norm": 1.4453125,
"learning_rate": 0.0004930275368431146,
"loss": 5.6759,
"mean_token_accuracy": 0.16634028255939484,
"num_tokens": 17854295.0,
"step": 7995
},
{
"entropy": 5.707269096374512,
"epoch": 0.8985230527320717,
"grad_norm": 1.3125,
"learning_rate": 0.0004930176159762435,
"loss": 5.5899,
"mean_token_accuracy": 0.18050725758075714,
"num_tokens": 17865597.0,
"step": 8000
},
{
"entropy": 5.453447532653809,
"epoch": 0.8990846296400292,
"grad_norm": 1.40625,
"learning_rate": 0.0004930076881676069,
"loss": 5.364,
"mean_token_accuracy": 0.1827785074710846,
"num_tokens": 17877405.0,
"step": 8005
},
{
"entropy": 5.70424222946167,
"epoch": 0.8996462065479868,
"grad_norm": 1.890625,
"learning_rate": 0.0004929977534175211,
"loss": 5.5679,
"mean_token_accuracy": 0.17531423419713973,
"num_tokens": 17889273.0,
"step": 8010
},
{
"entropy": 5.589845323562622,
"epoch": 0.9002077834559443,
"grad_norm": 1.28125,
"learning_rate": 0.0004929878117263021,
"loss": 5.4607,
"mean_token_accuracy": 0.1855960011482239,
"num_tokens": 17899761.0,
"step": 8015
},
{
"entropy": 5.525432872772217,
"epoch": 0.9007693603639019,
"grad_norm": 1.1953125,
"learning_rate": 0.0004929778630942668,
"loss": 5.4732,
"mean_token_accuracy": 0.17647491991519929,
"num_tokens": 17910607.0,
"step": 8020
},
{
"entropy": 5.6552492618560795,
"epoch": 0.9013309372718594,
"grad_norm": 1.671875,
"learning_rate": 0.0004929679075217317,
"loss": 5.5412,
"mean_token_accuracy": 0.17555252462625504,
"num_tokens": 17922538.0,
"step": 8025
},
{
"entropy": 5.672417497634887,
"epoch": 0.901892514179817,
"grad_norm": 1.34375,
"learning_rate": 0.0004929579450090139,
"loss": 5.5553,
"mean_token_accuracy": 0.17781907618045806,
"num_tokens": 17933313.0,
"step": 8030
},
{
"entropy": 5.647804689407349,
"epoch": 0.9024540910877745,
"grad_norm": 1.59375,
"learning_rate": 0.0004929479755564305,
"loss": 5.6576,
"mean_token_accuracy": 0.17102947384119033,
"num_tokens": 17944893.0,
"step": 8035
},
{
"entropy": 5.640274715423584,
"epoch": 0.903015667995732,
"grad_norm": 1.4765625,
"learning_rate": 0.0004929379991642991,
"loss": 5.5376,
"mean_token_accuracy": 0.17819006145000457,
"num_tokens": 17955019.0,
"step": 8040
},
{
"entropy": 5.7109626770019535,
"epoch": 0.9035772449036895,
"grad_norm": 1.3671875,
"learning_rate": 0.0004929280158329373,
"loss": 5.5609,
"mean_token_accuracy": 0.17464952319860458,
"num_tokens": 17965793.0,
"step": 8045
},
{
"entropy": 5.557077836990357,
"epoch": 0.9041388218116471,
"grad_norm": 1.421875,
"learning_rate": 0.000492918025562663,
"loss": 5.4021,
"mean_token_accuracy": 0.17762136310338975,
"num_tokens": 17975709.0,
"step": 8050
},
{
"entropy": 5.665270137786865,
"epoch": 0.9047003987196046,
"grad_norm": 1.3515625,
"learning_rate": 0.0004929080283537942,
"loss": 5.6261,
"mean_token_accuracy": 0.17274135500192642,
"num_tokens": 17985405.0,
"step": 8055
},
{
"entropy": 5.707554960250855,
"epoch": 0.9052619756275622,
"grad_norm": 1.3046875,
"learning_rate": 0.0004928980242066492,
"loss": 5.6475,
"mean_token_accuracy": 0.1707550197839737,
"num_tokens": 17997105.0,
"step": 8060
},
{
"entropy": 5.625894451141358,
"epoch": 0.9058235525355197,
"grad_norm": 1.53125,
"learning_rate": 0.0004928880131215467,
"loss": 5.5461,
"mean_token_accuracy": 0.18265556246042253,
"num_tokens": 18007914.0,
"step": 8065
},
{
"entropy": 5.568912935256958,
"epoch": 0.9063851294434773,
"grad_norm": 1.359375,
"learning_rate": 0.0004928779950988053,
"loss": 5.5392,
"mean_token_accuracy": 0.1795172333717346,
"num_tokens": 18018328.0,
"step": 8070
},
{
"entropy": 5.669607162475586,
"epoch": 0.9069467063514348,
"grad_norm": 1.1953125,
"learning_rate": 0.000492867970138744,
"loss": 5.5011,
"mean_token_accuracy": 0.17769642770290375,
"num_tokens": 18028568.0,
"step": 8075
},
{
"entropy": 5.6619072437286375,
"epoch": 0.9075082832593924,
"grad_norm": 1.6015625,
"learning_rate": 0.0004928579382416821,
"loss": 5.5106,
"mean_token_accuracy": 0.17568157464265824,
"num_tokens": 18038655.0,
"step": 8080
},
{
"entropy": 5.538381576538086,
"epoch": 0.9080698601673499,
"grad_norm": 1.4375,
"learning_rate": 0.000492847899407939,
"loss": 5.4484,
"mean_token_accuracy": 0.17773688584566116,
"num_tokens": 18049716.0,
"step": 8085
},
{
"entropy": 5.514719867706299,
"epoch": 0.9086314370753075,
"grad_norm": 1.640625,
"learning_rate": 0.0004928378536378341,
"loss": 5.4517,
"mean_token_accuracy": 0.18262287229299545,
"num_tokens": 18062530.0,
"step": 8090
},
{
"entropy": 5.555636405944824,
"epoch": 0.909193013983265,
"grad_norm": 1.34375,
"learning_rate": 0.0004928278009316877,
"loss": 5.5461,
"mean_token_accuracy": 0.1714445248246193,
"num_tokens": 18073076.0,
"step": 8095
},
{
"entropy": 5.674294662475586,
"epoch": 0.9097545908912226,
"grad_norm": 1.46875,
"learning_rate": 0.0004928177412898195,
"loss": 5.4974,
"mean_token_accuracy": 0.17785727828741074,
"num_tokens": 18083759.0,
"step": 8100
},
{
"entropy": 5.553316640853882,
"epoch": 0.9103161677991801,
"grad_norm": 1.3359375,
"learning_rate": 0.0004928076747125499,
"loss": 5.4432,
"mean_token_accuracy": 0.18112020641565324,
"num_tokens": 18095980.0,
"step": 8105
},
{
"entropy": 5.620355939865112,
"epoch": 0.9108777447071377,
"grad_norm": 1.6640625,
"learning_rate": 0.0004927976012001995,
"loss": 5.5237,
"mean_token_accuracy": 0.18174582123756408,
"num_tokens": 18106672.0,
"step": 8110
},
{
"entropy": 5.602421522140503,
"epoch": 0.9114393216150952,
"grad_norm": 1.3515625,
"learning_rate": 0.0004927875207530889,
"loss": 5.5352,
"mean_token_accuracy": 0.17589570879936217,
"num_tokens": 18117481.0,
"step": 8115
},
{
"entropy": 5.587666273117065,
"epoch": 0.9120008985230528,
"grad_norm": 1.375,
"learning_rate": 0.0004927774333715394,
"loss": 5.4752,
"mean_token_accuracy": 0.1781352072954178,
"num_tokens": 18127602.0,
"step": 8120
},
{
"entropy": 5.659744215011597,
"epoch": 0.9125624754310103,
"grad_norm": 1.34375,
"learning_rate": 0.0004927673390558718,
"loss": 5.5534,
"mean_token_accuracy": 0.17465065717697142,
"num_tokens": 18140109.0,
"step": 8125
},
{
"entropy": 5.601297378540039,
"epoch": 0.9131240523389679,
"grad_norm": 1.578125,
"learning_rate": 0.0004927572378064077,
"loss": 5.4315,
"mean_token_accuracy": 0.18548547625541686,
"num_tokens": 18150271.0,
"step": 8130
},
{
"entropy": 5.6010276794433596,
"epoch": 0.9136856292469253,
"grad_norm": 1.7265625,
"learning_rate": 0.0004927471296234687,
"loss": 5.547,
"mean_token_accuracy": 0.18272665441036223,
"num_tokens": 18160746.0,
"step": 8135
},
{
"entropy": 5.667695474624634,
"epoch": 0.9142472061548829,
"grad_norm": 1.65625,
"learning_rate": 0.0004927370145073767,
"loss": 5.5463,
"mean_token_accuracy": 0.17292810827493668,
"num_tokens": 18172271.0,
"step": 8140
},
{
"entropy": 5.585734510421753,
"epoch": 0.9148087830628404,
"grad_norm": 1.4921875,
"learning_rate": 0.0004927268924584536,
"loss": 5.4676,
"mean_token_accuracy": 0.18177113831043243,
"num_tokens": 18183116.0,
"step": 8145
},
{
"entropy": 5.53406457901001,
"epoch": 0.915370359970798,
"grad_norm": 1.484375,
"learning_rate": 0.0004927167634770219,
"loss": 5.4368,
"mean_token_accuracy": 0.18690151274204253,
"num_tokens": 18195147.0,
"step": 8150
},
{
"entropy": 5.602677059173584,
"epoch": 0.9159319368787555,
"grad_norm": 1.9375,
"learning_rate": 0.000492706627563404,
"loss": 5.5105,
"mean_token_accuracy": 0.178358194231987,
"num_tokens": 18206217.0,
"step": 8155
},
{
"entropy": 5.642651128768921,
"epoch": 0.9164935137867131,
"grad_norm": 1.453125,
"learning_rate": 0.0004926964847179225,
"loss": 5.5325,
"mean_token_accuracy": 0.17337826192378997,
"num_tokens": 18217434.0,
"step": 8160
},
{
"entropy": 5.662899017333984,
"epoch": 0.9170550906946706,
"grad_norm": 1.9140625,
"learning_rate": 0.0004926863349409006,
"loss": 5.5186,
"mean_token_accuracy": 0.1751028224825859,
"num_tokens": 18229277.0,
"step": 8165
},
{
"entropy": 5.53931770324707,
"epoch": 0.9176166676026282,
"grad_norm": 1.5234375,
"learning_rate": 0.0004926761782326613,
"loss": 5.3734,
"mean_token_accuracy": 0.18082480281591415,
"num_tokens": 18240320.0,
"step": 8170
},
{
"entropy": 5.523886585235596,
"epoch": 0.9181782445105857,
"grad_norm": 2.109375,
"learning_rate": 0.0004926660145935282,
"loss": 5.4145,
"mean_token_accuracy": 0.187174254655838,
"num_tokens": 18250380.0,
"step": 8175
},
{
"entropy": 5.667395401000976,
"epoch": 0.9187398214185433,
"grad_norm": 1.5078125,
"learning_rate": 0.0004926558440238247,
"loss": 5.603,
"mean_token_accuracy": 0.17506964653730392,
"num_tokens": 18261703.0,
"step": 8180
},
{
"entropy": 5.624309778213501,
"epoch": 0.9193013983265008,
"grad_norm": 1.4765625,
"learning_rate": 0.0004926456665238746,
"loss": 5.525,
"mean_token_accuracy": 0.17211228907108306,
"num_tokens": 18272472.0,
"step": 8185
},
{
"entropy": 5.7036622047424315,
"epoch": 0.9198629752344584,
"grad_norm": 1.7734375,
"learning_rate": 0.0004926354820940023,
"loss": 5.6061,
"mean_token_accuracy": 0.17764961421489717,
"num_tokens": 18282168.0,
"step": 8190
},
{
"entropy": 5.730700492858887,
"epoch": 0.9204245521424159,
"grad_norm": 1.9921875,
"learning_rate": 0.0004926252907345317,
"loss": 5.6116,
"mean_token_accuracy": 0.16850340217351914,
"num_tokens": 18293750.0,
"step": 8195
},
{
"entropy": 5.595494985580444,
"epoch": 0.9209861290503735,
"grad_norm": 1.703125,
"learning_rate": 0.0004926150924457874,
"loss": 5.5203,
"mean_token_accuracy": 0.18100045323371888,
"num_tokens": 18304568.0,
"step": 8200
},
{
"entropy": 5.582277059555054,
"epoch": 0.921547705958331,
"grad_norm": 1.3203125,
"learning_rate": 0.0004926048872280943,
"loss": 5.4496,
"mean_token_accuracy": 0.18073858618736266,
"num_tokens": 18315129.0,
"step": 8205
},
{
"entropy": 5.515413236618042,
"epoch": 0.9221092828662886,
"grad_norm": 1.6875,
"learning_rate": 0.000492594675081777,
"loss": 5.4058,
"mean_token_accuracy": 0.1768901064991951,
"num_tokens": 18324628.0,
"step": 8210
},
{
"entropy": 5.543982601165771,
"epoch": 0.9226708597742461,
"grad_norm": 1.3203125,
"learning_rate": 0.000492584456007161,
"loss": 5.4199,
"mean_token_accuracy": 0.18777298629283906,
"num_tokens": 18334899.0,
"step": 8215
},
{
"entropy": 5.608406925201416,
"epoch": 0.9232324366822037,
"grad_norm": 1.515625,
"learning_rate": 0.0004925742300045716,
"loss": 5.437,
"mean_token_accuracy": 0.18415032029151918,
"num_tokens": 18345719.0,
"step": 8220
},
{
"entropy": 5.591996860504151,
"epoch": 0.9237940135901612,
"grad_norm": 1.7578125,
"learning_rate": 0.0004925639970743341,
"loss": 5.5229,
"mean_token_accuracy": 0.18159339874982833,
"num_tokens": 18355814.0,
"step": 8225
},
{
"entropy": 5.624569225311279,
"epoch": 0.9243555904981187,
"grad_norm": 1.5859375,
"learning_rate": 0.0004925537572167746,
"loss": 5.6014,
"mean_token_accuracy": 0.1739961490035057,
"num_tokens": 18366746.0,
"step": 8230
},
{
"entropy": 5.603530931472778,
"epoch": 0.9249171674060762,
"grad_norm": 1.5234375,
"learning_rate": 0.0004925435104322192,
"loss": 5.4027,
"mean_token_accuracy": 0.18583776354789733,
"num_tokens": 18375846.0,
"step": 8235
},
{
"entropy": 5.523084449768066,
"epoch": 0.9254787443140338,
"grad_norm": 1.328125,
"learning_rate": 0.000492533256720994,
"loss": 5.5645,
"mean_token_accuracy": 0.17348302602767945,
"num_tokens": 18386850.0,
"step": 8240
},
{
"entropy": 5.581809234619141,
"epoch": 0.9260403212219913,
"grad_norm": 1.390625,
"learning_rate": 0.0004925229960834256,
"loss": 5.4328,
"mean_token_accuracy": 0.18448738604784012,
"num_tokens": 18396347.0,
"step": 8245
},
{
"entropy": 5.617673635482788,
"epoch": 0.9266018981299489,
"grad_norm": 1.4375,
"learning_rate": 0.0004925127285198407,
"loss": 5.4742,
"mean_token_accuracy": 0.18064238131046295,
"num_tokens": 18406583.0,
"step": 8250
},
{
"entropy": 5.671968841552735,
"epoch": 0.9271634750379064,
"grad_norm": 1.8203125,
"learning_rate": 0.000492502454030566,
"loss": 5.5948,
"mean_token_accuracy": 0.17629928290843963,
"num_tokens": 18418586.0,
"step": 8255
},
{
"entropy": 5.630609846115112,
"epoch": 0.927725051945864,
"grad_norm": 1.5390625,
"learning_rate": 0.0004924921726159288,
"loss": 5.435,
"mean_token_accuracy": 0.1769808605313301,
"num_tokens": 18429752.0,
"step": 8260
},
{
"entropy": 5.625945520401001,
"epoch": 0.9282866288538215,
"grad_norm": 2.25,
"learning_rate": 0.0004924818842762565,
"loss": 5.5318,
"mean_token_accuracy": 0.17295176684856414,
"num_tokens": 18440663.0,
"step": 8265
},
{
"entropy": 5.5104950904846195,
"epoch": 0.9288482057617791,
"grad_norm": 1.6640625,
"learning_rate": 0.0004924715890118768,
"loss": 5.4761,
"mean_token_accuracy": 0.18121709376573564,
"num_tokens": 18451497.0,
"step": 8270
},
{
"entropy": 5.563171768188477,
"epoch": 0.9294097826697366,
"grad_norm": 1.5,
"learning_rate": 0.0004924612868231173,
"loss": 5.465,
"mean_token_accuracy": 0.17596760988235474,
"num_tokens": 18462794.0,
"step": 8275
},
{
"entropy": 5.6284508228302,
"epoch": 0.9299713595776942,
"grad_norm": 1.6171875,
"learning_rate": 0.0004924509777103059,
"loss": 5.4347,
"mean_token_accuracy": 0.18390606492757797,
"num_tokens": 18472973.0,
"step": 8280
},
{
"entropy": 5.636531925201416,
"epoch": 0.9305329364856517,
"grad_norm": 1.7578125,
"learning_rate": 0.0004924406616737712,
"loss": 5.6303,
"mean_token_accuracy": 0.175663585960865,
"num_tokens": 18484451.0,
"step": 8285
},
{
"entropy": 5.633027505874634,
"epoch": 0.9310945133936093,
"grad_norm": 1.703125,
"learning_rate": 0.0004924303387138415,
"loss": 5.4765,
"mean_token_accuracy": 0.1841089352965355,
"num_tokens": 18496412.0,
"step": 8290
},
{
"entropy": 5.642653465270996,
"epoch": 0.9316560903015668,
"grad_norm": 1.3515625,
"learning_rate": 0.0004924200088308455,
"loss": 5.5051,
"mean_token_accuracy": 0.18072894364595413,
"num_tokens": 18508547.0,
"step": 8295
},
{
"entropy": 5.54476842880249,
"epoch": 0.9322176672095244,
"grad_norm": 1.375,
"learning_rate": 0.000492409672025112,
"loss": 5.5494,
"mean_token_accuracy": 0.1738213375210762,
"num_tokens": 18518954.0,
"step": 8300
},
{
"entropy": 5.554588794708252,
"epoch": 0.9327792441174819,
"grad_norm": 1.5078125,
"learning_rate": 0.0004923993282969703,
"loss": 5.3958,
"mean_token_accuracy": 0.1817837253212929,
"num_tokens": 18530444.0,
"step": 8305
},
{
"entropy": 5.612212324142456,
"epoch": 0.9333408210254395,
"grad_norm": 1.453125,
"learning_rate": 0.0004923889776467495,
"loss": 5.4589,
"mean_token_accuracy": 0.178005214035511,
"num_tokens": 18541612.0,
"step": 8310
},
{
"entropy": 5.6077526092529295,
"epoch": 0.933902397933397,
"grad_norm": 1.6171875,
"learning_rate": 0.0004923786200747795,
"loss": 5.6307,
"mean_token_accuracy": 0.1739485114812851,
"num_tokens": 18553316.0,
"step": 8315
},
{
"entropy": 5.603684616088867,
"epoch": 0.9344639748413546,
"grad_norm": 1.5546875,
"learning_rate": 0.00049236825558139,
"loss": 5.4891,
"mean_token_accuracy": 0.18104055970907212,
"num_tokens": 18564468.0,
"step": 8320
},
{
"entropy": 5.562555170059204,
"epoch": 0.9350255517493121,
"grad_norm": 1.796875,
"learning_rate": 0.0004923578841669108,
"loss": 5.4325,
"mean_token_accuracy": 0.18535131514072417,
"num_tokens": 18575536.0,
"step": 8325
},
{
"entropy": 5.574767017364502,
"epoch": 0.9355871286572696,
"grad_norm": 1.890625,
"learning_rate": 0.0004923475058316723,
"loss": 5.5441,
"mean_token_accuracy": 0.1676850900053978,
"num_tokens": 18586906.0,
"step": 8330
},
{
"entropy": 5.6700334548950195,
"epoch": 0.9361487055652271,
"grad_norm": 1.5859375,
"learning_rate": 0.0004923371205760049,
"loss": 5.5413,
"mean_token_accuracy": 0.178540375828743,
"num_tokens": 18599064.0,
"step": 8335
},
{
"entropy": 5.645085000991822,
"epoch": 0.9367102824731847,
"grad_norm": 1.40625,
"learning_rate": 0.0004923267284002394,
"loss": 5.5614,
"mean_token_accuracy": 0.17388303577899933,
"num_tokens": 18610057.0,
"step": 8340
},
{
"entropy": 5.50980396270752,
"epoch": 0.9372718593811422,
"grad_norm": 2.015625,
"learning_rate": 0.0004923163293047065,
"loss": 5.3869,
"mean_token_accuracy": 0.1858271449804306,
"num_tokens": 18620008.0,
"step": 8345
},
{
"entropy": 5.488223457336426,
"epoch": 0.9378334362890998,
"grad_norm": 1.3203125,
"learning_rate": 0.0004923059232897374,
"loss": 5.3856,
"mean_token_accuracy": 0.18158367723226548,
"num_tokens": 18630599.0,
"step": 8350
},
{
"entropy": 5.595425701141357,
"epoch": 0.9383950131970573,
"grad_norm": 1.546875,
"learning_rate": 0.0004922955103556633,
"loss": 5.5807,
"mean_token_accuracy": 0.1751416489481926,
"num_tokens": 18643898.0,
"step": 8355
},
{
"entropy": 5.642851734161377,
"epoch": 0.9389565901050149,
"grad_norm": 1.484375,
"learning_rate": 0.000492285090502816,
"loss": 5.5449,
"mean_token_accuracy": 0.17289358079433442,
"num_tokens": 18654760.0,
"step": 8360
},
{
"entropy": 5.659446096420288,
"epoch": 0.9395181670129724,
"grad_norm": 1.2421875,
"learning_rate": 0.0004922746637315272,
"loss": 5.5041,
"mean_token_accuracy": 0.17735429108142853,
"num_tokens": 18666036.0,
"step": 8365
},
{
"entropy": 5.664956855773926,
"epoch": 0.94007974392093,
"grad_norm": 1.4375,
"learning_rate": 0.0004922642300421288,
"loss": 5.5601,
"mean_token_accuracy": 0.1765066161751747,
"num_tokens": 18677806.0,
"step": 8370
},
{
"entropy": 5.629373168945312,
"epoch": 0.9406413208288875,
"grad_norm": 1.328125,
"learning_rate": 0.000492253789434953,
"loss": 5.6758,
"mean_token_accuracy": 0.17487102895975112,
"num_tokens": 18688971.0,
"step": 8375
},
{
"entropy": 5.593070936203003,
"epoch": 0.9412028977368451,
"grad_norm": 1.484375,
"learning_rate": 0.0004922433419103323,
"loss": 5.5522,
"mean_token_accuracy": 0.17334904670715331,
"num_tokens": 18700153.0,
"step": 8380
},
{
"entropy": 5.7966571807861325,
"epoch": 0.9417644746448026,
"grad_norm": 1.390625,
"learning_rate": 0.0004922328874685993,
"loss": 5.6141,
"mean_token_accuracy": 0.17159850597381593,
"num_tokens": 18711935.0,
"step": 8385
},
{
"entropy": 5.5403656482696535,
"epoch": 0.9423260515527602,
"grad_norm": 1.328125,
"learning_rate": 0.0004922224261100869,
"loss": 5.4273,
"mean_token_accuracy": 0.17966998219490052,
"num_tokens": 18722694.0,
"step": 8390
},
{
"entropy": 5.590994548797608,
"epoch": 0.9428876284607177,
"grad_norm": 1.8828125,
"learning_rate": 0.0004922119578351283,
"loss": 5.5656,
"mean_token_accuracy": 0.1821168154478073,
"num_tokens": 18733443.0,
"step": 8395
},
{
"entropy": 5.655079126358032,
"epoch": 0.9434492053686753,
"grad_norm": 1.46875,
"learning_rate": 0.0004922014826440566,
"loss": 5.4872,
"mean_token_accuracy": 0.18000147938728334,
"num_tokens": 18743965.0,
"step": 8400
},
{
"entropy": 5.625576639175415,
"epoch": 0.9440107822766328,
"grad_norm": 1.4296875,
"learning_rate": 0.0004921910005372056,
"loss": 5.5779,
"mean_token_accuracy": 0.1763213112950325,
"num_tokens": 18755712.0,
"step": 8405
},
{
"entropy": 5.609548139572143,
"epoch": 0.9445723591845904,
"grad_norm": 1.6171875,
"learning_rate": 0.0004921805115149088,
"loss": 5.4723,
"mean_token_accuracy": 0.18412862867116928,
"num_tokens": 18766756.0,
"step": 8410
},
{
"entropy": 5.637346601486206,
"epoch": 0.9451339360925479,
"grad_norm": 1.6640625,
"learning_rate": 0.0004921700155775002,
"loss": 5.5916,
"mean_token_accuracy": 0.17035791128873826,
"num_tokens": 18776932.0,
"step": 8415
},
{
"entropy": 5.694090223312378,
"epoch": 0.9456955130005055,
"grad_norm": 1.5546875,
"learning_rate": 0.0004921595127253142,
"loss": 5.4927,
"mean_token_accuracy": 0.17952510267496108,
"num_tokens": 18787197.0,
"step": 8420
},
{
"entropy": 5.575382041931152,
"epoch": 0.9462570899084629,
"grad_norm": 1.375,
"learning_rate": 0.0004921490029586852,
"loss": 5.4721,
"mean_token_accuracy": 0.18035454750061036,
"num_tokens": 18797861.0,
"step": 8425
},
{
"entropy": 5.579099893569946,
"epoch": 0.9468186668164205,
"grad_norm": 2.890625,
"learning_rate": 0.0004921384862779475,
"loss": 5.4606,
"mean_token_accuracy": 0.17814618200063706,
"num_tokens": 18809111.0,
"step": 8430
},
{
"entropy": 5.586229658126831,
"epoch": 0.947380243724378,
"grad_norm": 1.5078125,
"learning_rate": 0.0004921279626834363,
"loss": 5.4737,
"mean_token_accuracy": 0.18840118795633315,
"num_tokens": 18820236.0,
"step": 8435
},
{
"entropy": 5.532220554351807,
"epoch": 0.9479418206323356,
"grad_norm": 1.3828125,
"learning_rate": 0.0004921174321754865,
"loss": 5.4854,
"mean_token_accuracy": 0.17972348481416703,
"num_tokens": 18831942.0,
"step": 8440
},
{
"entropy": 5.609068870544434,
"epoch": 0.9485033975402931,
"grad_norm": 1.34375,
"learning_rate": 0.0004921068947544335,
"loss": 5.5117,
"mean_token_accuracy": 0.17793938964605333,
"num_tokens": 18843271.0,
"step": 8445
},
{
"entropy": 5.594819259643555,
"epoch": 0.9490649744482507,
"grad_norm": 1.3515625,
"learning_rate": 0.0004920963504206128,
"loss": 5.5577,
"mean_token_accuracy": 0.1800818532705307,
"num_tokens": 18856285.0,
"step": 8450
},
{
"entropy": 5.683270978927612,
"epoch": 0.9496265513562082,
"grad_norm": 1.375,
"learning_rate": 0.00049208579917436,
"loss": 5.5811,
"mean_token_accuracy": 0.1756241112947464,
"num_tokens": 18867203.0,
"step": 8455
},
{
"entropy": 5.6580253601074215,
"epoch": 0.9501881282641658,
"grad_norm": 1.609375,
"learning_rate": 0.0004920752410160113,
"loss": 5.6136,
"mean_token_accuracy": 0.17068895548582078,
"num_tokens": 18878598.0,
"step": 8460
},
{
"entropy": 5.674833869934082,
"epoch": 0.9507497051721233,
"grad_norm": 1.484375,
"learning_rate": 0.0004920646759459026,
"loss": 5.5171,
"mean_token_accuracy": 0.17998390197753905,
"num_tokens": 18890233.0,
"step": 8465
},
{
"entropy": 5.683036375045776,
"epoch": 0.9513112820800809,
"grad_norm": 1.59375,
"learning_rate": 0.0004920541039643705,
"loss": 5.6194,
"mean_token_accuracy": 0.17345030903816222,
"num_tokens": 18900362.0,
"step": 8470
},
{
"entropy": 5.6739569187164305,
"epoch": 0.9518728589880384,
"grad_norm": 1.6328125,
"learning_rate": 0.0004920435250717515,
"loss": 5.5368,
"mean_token_accuracy": 0.18129113167524338,
"num_tokens": 18911924.0,
"step": 8475
},
{
"entropy": 5.6541563987731935,
"epoch": 0.952434435895996,
"grad_norm": 1.5,
"learning_rate": 0.0004920329392683826,
"loss": 5.5561,
"mean_token_accuracy": 0.17487876415252684,
"num_tokens": 18923931.0,
"step": 8480
},
{
"entropy": 5.57630500793457,
"epoch": 0.9529960128039535,
"grad_norm": 1.5234375,
"learning_rate": 0.0004920223465546006,
"loss": 5.4559,
"mean_token_accuracy": 0.17989639490842818,
"num_tokens": 18935859.0,
"step": 8485
},
{
"entropy": 5.631529188156128,
"epoch": 0.9535575897119111,
"grad_norm": 1.484375,
"learning_rate": 0.0004920117469307429,
"loss": 5.5916,
"mean_token_accuracy": 0.1710250437259674,
"num_tokens": 18946939.0,
"step": 8490
},
{
"entropy": 5.654216194152832,
"epoch": 0.9541191666198686,
"grad_norm": 1.7421875,
"learning_rate": 0.0004920011403971471,
"loss": 5.5519,
"mean_token_accuracy": 0.1704327180981636,
"num_tokens": 18958067.0,
"step": 8495
},
{
"entropy": 5.604382085800171,
"epoch": 0.9546807435278262,
"grad_norm": 2.078125,
"learning_rate": 0.0004919905269541508,
"loss": 5.4813,
"mean_token_accuracy": 0.1850312188267708,
"num_tokens": 18967920.0,
"step": 8500
},
{
"entropy": 5.629701709747314,
"epoch": 0.9552423204357837,
"grad_norm": 1.8671875,
"learning_rate": 0.000491979906602092,
"loss": 5.4963,
"mean_token_accuracy": 0.18280581682920455,
"num_tokens": 18978244.0,
"step": 8505
},
{
"entropy": 5.638136291503907,
"epoch": 0.9558038973437413,
"grad_norm": 1.578125,
"learning_rate": 0.0004919692793413089,
"loss": 5.4639,
"mean_token_accuracy": 0.17874942868947982,
"num_tokens": 18988464.0,
"step": 8510
},
{
"entropy": 5.515000438690185,
"epoch": 0.9563654742516988,
"grad_norm": 1.890625,
"learning_rate": 0.0004919586451721396,
"loss": 5.4093,
"mean_token_accuracy": 0.1843804895877838,
"num_tokens": 18998318.0,
"step": 8515
},
{
"entropy": 5.461893558502197,
"epoch": 0.9569270511596563,
"grad_norm": 2.0625,
"learning_rate": 0.0004919480040949229,
"loss": 5.4039,
"mean_token_accuracy": 0.18554521799087526,
"num_tokens": 19009802.0,
"step": 8520
},
{
"entropy": 5.656184339523316,
"epoch": 0.9574886280676138,
"grad_norm": 1.6875,
"learning_rate": 0.0004919373561099977,
"loss": 5.614,
"mean_token_accuracy": 0.176939395070076,
"num_tokens": 19020174.0,
"step": 8525
},
{
"entropy": 5.651794147491455,
"epoch": 0.9580502049755714,
"grad_norm": 1.4140625,
"learning_rate": 0.000491926701217703,
"loss": 5.5486,
"mean_token_accuracy": 0.17270150780677795,
"num_tokens": 19031505.0,
"step": 8530
},
{
"entropy": 5.582490539550781,
"epoch": 0.9586117818835289,
"grad_norm": 1.703125,
"learning_rate": 0.0004919160394183779,
"loss": 5.4457,
"mean_token_accuracy": 0.18328225016593933,
"num_tokens": 19042052.0,
"step": 8535
},
{
"entropy": 5.795409345626831,
"epoch": 0.9591733587914865,
"grad_norm": 1.453125,
"learning_rate": 0.000491905370712362,
"loss": 5.6475,
"mean_token_accuracy": 0.1725166603922844,
"num_tokens": 19052050.0,
"step": 8540
},
{
"entropy": 5.677691125869751,
"epoch": 0.959734935699444,
"grad_norm": 1.3515625,
"learning_rate": 0.0004918946950999949,
"loss": 5.5423,
"mean_token_accuracy": 0.1743814915418625,
"num_tokens": 19062053.0,
"step": 8545
},
{
"entropy": 5.537538623809814,
"epoch": 0.9602965126074016,
"grad_norm": 1.7578125,
"learning_rate": 0.0004918840125816167,
"loss": 5.4915,
"mean_token_accuracy": 0.1790384292602539,
"num_tokens": 19073095.0,
"step": 8550
},
{
"entropy": 5.599767732620239,
"epoch": 0.9608580895153591,
"grad_norm": 1.5625,
"learning_rate": 0.0004918733231575673,
"loss": 5.5627,
"mean_token_accuracy": 0.18295877426862717,
"num_tokens": 19084521.0,
"step": 8555
},
{
"entropy": 5.611762762069702,
"epoch": 0.9614196664233167,
"grad_norm": 1.421875,
"learning_rate": 0.0004918626268281874,
"loss": 5.5612,
"mean_token_accuracy": 0.17635897248983384,
"num_tokens": 19096372.0,
"step": 8560
},
{
"entropy": 5.741964197158813,
"epoch": 0.9619812433312742,
"grad_norm": 1.5,
"learning_rate": 0.0004918519235938171,
"loss": 5.6813,
"mean_token_accuracy": 0.16932807564735414,
"num_tokens": 19108535.0,
"step": 8565
},
{
"entropy": 5.73540415763855,
"epoch": 0.9625428202392318,
"grad_norm": 1.5546875,
"learning_rate": 0.0004918412134547976,
"loss": 5.5135,
"mean_token_accuracy": 0.18378160893917084,
"num_tokens": 19119480.0,
"step": 8570
},
{
"entropy": 5.5963503360748295,
"epoch": 0.9631043971471893,
"grad_norm": 1.40625,
"learning_rate": 0.0004918304964114696,
"loss": 5.5208,
"mean_token_accuracy": 0.1820914030075073,
"num_tokens": 19131168.0,
"step": 8575
},
{
"entropy": 5.660545873641968,
"epoch": 0.9636659740551469,
"grad_norm": 1.5703125,
"learning_rate": 0.0004918197724641747,
"loss": 5.6548,
"mean_token_accuracy": 0.17506289929151536,
"num_tokens": 19142454.0,
"step": 8580
},
{
"entropy": 5.6714612483978275,
"epoch": 0.9642275509631044,
"grad_norm": 1.4453125,
"learning_rate": 0.000491809041613254,
"loss": 5.5238,
"mean_token_accuracy": 0.18075565248727798,
"num_tokens": 19154682.0,
"step": 8585
},
{
"entropy": 5.611203002929687,
"epoch": 0.964789127871062,
"grad_norm": 1.5,
"learning_rate": 0.0004917983038590495,
"loss": 5.4611,
"mean_token_accuracy": 0.17857588678598404,
"num_tokens": 19165190.0,
"step": 8590
},
{
"entropy": 5.630858659744263,
"epoch": 0.9653507047790195,
"grad_norm": 2.1875,
"learning_rate": 0.0004917875592019028,
"loss": 5.5794,
"mean_token_accuracy": 0.18112334609031677,
"num_tokens": 19176075.0,
"step": 8595
},
{
"entropy": 5.677709007263184,
"epoch": 0.9659122816869771,
"grad_norm": 1.4765625,
"learning_rate": 0.0004917768076421561,
"loss": 5.5837,
"mean_token_accuracy": 0.1755153328180313,
"num_tokens": 19188450.0,
"step": 8600
},
{
"entropy": 5.596295738220215,
"epoch": 0.9664738585949346,
"grad_norm": 1.4453125,
"learning_rate": 0.0004917660491801518,
"loss": 5.4402,
"mean_token_accuracy": 0.186106838285923,
"num_tokens": 19201254.0,
"step": 8605
},
{
"entropy": 5.584579753875732,
"epoch": 0.9670354355028922,
"grad_norm": 1.6328125,
"learning_rate": 0.0004917552838162324,
"loss": 5.5275,
"mean_token_accuracy": 0.17696987092494965,
"num_tokens": 19214062.0,
"step": 8610
},
{
"entropy": 5.587477159500122,
"epoch": 0.9675970124108496,
"grad_norm": 1.40625,
"learning_rate": 0.0004917445115507408,
"loss": 5.4383,
"mean_token_accuracy": 0.18441233038902283,
"num_tokens": 19224844.0,
"step": 8615
},
{
"entropy": 5.623504972457885,
"epoch": 0.9681585893188072,
"grad_norm": 1.65625,
"learning_rate": 0.0004917337323840198,
"loss": 5.6162,
"mean_token_accuracy": 0.1746974691748619,
"num_tokens": 19237347.0,
"step": 8620
},
{
"entropy": 5.684951734542847,
"epoch": 0.9687201662267647,
"grad_norm": 1.5,
"learning_rate": 0.0004917229463164128,
"loss": 5.482,
"mean_token_accuracy": 0.1786902904510498,
"num_tokens": 19248743.0,
"step": 8625
},
{
"entropy": 5.5819010734558105,
"epoch": 0.9692817431347223,
"grad_norm": 1.4921875,
"learning_rate": 0.0004917121533482631,
"loss": 5.4752,
"mean_token_accuracy": 0.18344754427671434,
"num_tokens": 19259904.0,
"step": 8630
},
{
"entropy": 5.546244192123413,
"epoch": 0.9698433200426798,
"grad_norm": 1.609375,
"learning_rate": 0.0004917013534799144,
"loss": 5.4092,
"mean_token_accuracy": 0.18333079069852828,
"num_tokens": 19271388.0,
"step": 8635
},
{
"entropy": 5.543841552734375,
"epoch": 0.9704048969506374,
"grad_norm": 1.421875,
"learning_rate": 0.0004916905467117104,
"loss": 5.5508,
"mean_token_accuracy": 0.17571745067834854,
"num_tokens": 19282790.0,
"step": 8640
},
{
"entropy": 5.62225866317749,
"epoch": 0.9709664738585949,
"grad_norm": 1.484375,
"learning_rate": 0.0004916797330439955,
"loss": 5.4496,
"mean_token_accuracy": 0.18349325507879258,
"num_tokens": 19293389.0,
"step": 8645
},
{
"entropy": 5.61131796836853,
"epoch": 0.9715280507665525,
"grad_norm": 1.5234375,
"learning_rate": 0.0004916689124771138,
"loss": 5.5204,
"mean_token_accuracy": 0.17595339864492415,
"num_tokens": 19304593.0,
"step": 8650
},
{
"entropy": 5.610396957397461,
"epoch": 0.97208962767451,
"grad_norm": 1.3984375,
"learning_rate": 0.0004916580850114099,
"loss": 5.4531,
"mean_token_accuracy": 0.1782217487692833,
"num_tokens": 19317388.0,
"step": 8655
},
{
"entropy": 5.636252164840698,
"epoch": 0.9726512045824676,
"grad_norm": 1.5546875,
"learning_rate": 0.0004916472506472286,
"loss": 5.5075,
"mean_token_accuracy": 0.17837999612092972,
"num_tokens": 19329228.0,
"step": 8660
},
{
"entropy": 5.614098024368286,
"epoch": 0.9732127814904251,
"grad_norm": 1.5,
"learning_rate": 0.0004916364093849147,
"loss": 5.4799,
"mean_token_accuracy": 0.1822504386305809,
"num_tokens": 19339421.0,
"step": 8665
},
{
"entropy": 5.445722770690918,
"epoch": 0.9737743583983827,
"grad_norm": 1.4765625,
"learning_rate": 0.0004916255612248136,
"loss": 5.3835,
"mean_token_accuracy": 0.18440828174352647,
"num_tokens": 19349618.0,
"step": 8670
},
{
"entropy": 5.49763560295105,
"epoch": 0.9743359353063402,
"grad_norm": 1.921875,
"learning_rate": 0.0004916147061672706,
"loss": 5.4728,
"mean_token_accuracy": 0.18142055571079255,
"num_tokens": 19361324.0,
"step": 8675
},
{
"entropy": 5.624416255950928,
"epoch": 0.9748975122142978,
"grad_norm": 1.5078125,
"learning_rate": 0.0004916038442126314,
"loss": 5.5066,
"mean_token_accuracy": 0.17429686337709427,
"num_tokens": 19373631.0,
"step": 8680
},
{
"entropy": 5.562902307510376,
"epoch": 0.9754590891222553,
"grad_norm": 2.09375,
"learning_rate": 0.0004915929753612416,
"loss": 5.4534,
"mean_token_accuracy": 0.17012406140565872,
"num_tokens": 19384682.0,
"step": 8685
},
{
"entropy": 5.596801662445069,
"epoch": 0.9760206660302129,
"grad_norm": 1.4296875,
"learning_rate": 0.0004915820996134477,
"loss": 5.5707,
"mean_token_accuracy": 0.17153397798538209,
"num_tokens": 19396344.0,
"step": 8690
},
{
"entropy": 5.583506011962891,
"epoch": 0.9765822429381704,
"grad_norm": 1.390625,
"learning_rate": 0.0004915712169695956,
"loss": 5.4249,
"mean_token_accuracy": 0.1793520987033844,
"num_tokens": 19406921.0,
"step": 8695
},
{
"entropy": 5.591095924377441,
"epoch": 0.977143819846128,
"grad_norm": 1.296875,
"learning_rate": 0.000491560327430032,
"loss": 5.5272,
"mean_token_accuracy": 0.17503703236579896,
"num_tokens": 19416851.0,
"step": 8700
},
{
"entropy": 5.603719997406006,
"epoch": 0.9777053967540855,
"grad_norm": 1.6796875,
"learning_rate": 0.0004915494309951035,
"loss": 5.5099,
"mean_token_accuracy": 0.17532387673854827,
"num_tokens": 19428180.0,
"step": 8705
},
{
"entropy": 5.703129768371582,
"epoch": 0.9782669736620431,
"grad_norm": 2.3125,
"learning_rate": 0.0004915385276651573,
"loss": 5.571,
"mean_token_accuracy": 0.17758786529302598,
"num_tokens": 19439575.0,
"step": 8710
},
{
"entropy": 5.652683734893799,
"epoch": 0.9788285505700005,
"grad_norm": 1.6796875,
"learning_rate": 0.0004915276174405403,
"loss": 5.532,
"mean_token_accuracy": 0.1791491389274597,
"num_tokens": 19452074.0,
"step": 8715
},
{
"entropy": 5.535808801651001,
"epoch": 0.9793901274779581,
"grad_norm": 1.3671875,
"learning_rate": 0.0004915167003215999,
"loss": 5.4529,
"mean_token_accuracy": 0.1823381319642067,
"num_tokens": 19463085.0,
"step": 8720
},
{
"entropy": 5.735648345947266,
"epoch": 0.9799517043859156,
"grad_norm": 1.5078125,
"learning_rate": 0.0004915057763086838,
"loss": 5.618,
"mean_token_accuracy": 0.16804364919662476,
"num_tokens": 19472726.0,
"step": 8725
},
{
"entropy": 5.588753509521484,
"epoch": 0.9805132812938732,
"grad_norm": 1.46875,
"learning_rate": 0.00049149484540214,
"loss": 5.4411,
"mean_token_accuracy": 0.17308186143636703,
"num_tokens": 19482275.0,
"step": 8730
},
{
"entropy": 5.58228497505188,
"epoch": 0.9810748582018307,
"grad_norm": 1.8046875,
"learning_rate": 0.0004914839076023161,
"loss": 5.5428,
"mean_token_accuracy": 0.1803490549325943,
"num_tokens": 19492782.0,
"step": 8735
},
{
"entropy": 5.515206623077392,
"epoch": 0.9816364351097883,
"grad_norm": 1.359375,
"learning_rate": 0.0004914729629095609,
"loss": 5.4404,
"mean_token_accuracy": 0.17642996162176133,
"num_tokens": 19505709.0,
"step": 8740
},
{
"entropy": 5.615311050415039,
"epoch": 0.9821980120177458,
"grad_norm": 1.859375,
"learning_rate": 0.0004914620113242224,
"loss": 5.5054,
"mean_token_accuracy": 0.1781864732503891,
"num_tokens": 19516269.0,
"step": 8745
},
{
"entropy": 5.609285163879394,
"epoch": 0.9827595889257034,
"grad_norm": 1.6171875,
"learning_rate": 0.0004914510528466495,
"loss": 5.4946,
"mean_token_accuracy": 0.17529860883951187,
"num_tokens": 19526786.0,
"step": 8750
},
{
"entropy": 5.624034786224366,
"epoch": 0.9833211658336609,
"grad_norm": 1.4453125,
"learning_rate": 0.0004914400874771913,
"loss": 5.534,
"mean_token_accuracy": 0.17428242564201354,
"num_tokens": 19537865.0,
"step": 8755
},
{
"entropy": 5.530050611495971,
"epoch": 0.9838827427416185,
"grad_norm": 1.734375,
"learning_rate": 0.0004914291152161967,
"loss": 5.4499,
"mean_token_accuracy": 0.18495927155017852,
"num_tokens": 19548320.0,
"step": 8760
},
{
"entropy": 5.584422206878662,
"epoch": 0.984444319649576,
"grad_norm": 1.6484375,
"learning_rate": 0.000491418136064015,
"loss": 5.4559,
"mean_token_accuracy": 0.18140354454517366,
"num_tokens": 19558856.0,
"step": 8765
},
{
"entropy": 5.628677558898926,
"epoch": 0.9850058965575336,
"grad_norm": 2.265625,
"learning_rate": 0.000491407150020996,
"loss": 5.5771,
"mean_token_accuracy": 0.17783913463354112,
"num_tokens": 19570792.0,
"step": 8770
},
{
"entropy": 5.606200504302978,
"epoch": 0.9855674734654911,
"grad_norm": 1.5,
"learning_rate": 0.0004913961570874895,
"loss": 5.4772,
"mean_token_accuracy": 0.18223483562469484,
"num_tokens": 19581982.0,
"step": 8775
},
{
"entropy": 5.543896007537842,
"epoch": 0.9861290503734487,
"grad_norm": 1.8671875,
"learning_rate": 0.0004913851572638453,
"loss": 5.4264,
"mean_token_accuracy": 0.1824976310133934,
"num_tokens": 19592882.0,
"step": 8780
},
{
"entropy": 5.670445775985717,
"epoch": 0.9866906272814062,
"grad_norm": 1.859375,
"learning_rate": 0.0004913741505504138,
"loss": 5.6407,
"mean_token_accuracy": 0.17869670689105988,
"num_tokens": 19605967.0,
"step": 8785
},
{
"entropy": 5.670494270324707,
"epoch": 0.9872522041893638,
"grad_norm": 1.765625,
"learning_rate": 0.0004913631369475454,
"loss": 5.4842,
"mean_token_accuracy": 0.18213990330696106,
"num_tokens": 19616494.0,
"step": 8790
},
{
"entropy": 5.6691545963287355,
"epoch": 0.9878137810973213,
"grad_norm": 1.890625,
"learning_rate": 0.0004913521164555909,
"loss": 5.5951,
"mean_token_accuracy": 0.18049925267696382,
"num_tokens": 19627013.0,
"step": 8795
},
{
"entropy": 5.616021585464478,
"epoch": 0.9883753580052789,
"grad_norm": 2.734375,
"learning_rate": 0.000491341089074901,
"loss": 5.4885,
"mean_token_accuracy": 0.18730600774288178,
"num_tokens": 19639099.0,
"step": 8800
},
{
"entropy": 5.521750783920288,
"epoch": 0.9889369349132364,
"grad_norm": 1.9296875,
"learning_rate": 0.000491330054805827,
"loss": 5.4709,
"mean_token_accuracy": 0.18193579018115996,
"num_tokens": 19650650.0,
"step": 8805
},
{
"entropy": 5.658022880554199,
"epoch": 0.9894985118211939,
"grad_norm": 1.8359375,
"learning_rate": 0.0004913190136487201,
"loss": 5.5169,
"mean_token_accuracy": 0.17676839530467986,
"num_tokens": 19662869.0,
"step": 8810
},
{
"entropy": 5.6313587665557865,
"epoch": 0.9900600887291514,
"grad_norm": 1.640625,
"learning_rate": 0.0004913079656039318,
"loss": 5.5147,
"mean_token_accuracy": 0.1779134064912796,
"num_tokens": 19673232.0,
"step": 8815
},
{
"entropy": 5.623973941802978,
"epoch": 0.990621665637109,
"grad_norm": 1.6484375,
"learning_rate": 0.0004912969106718142,
"loss": 5.5265,
"mean_token_accuracy": 0.17837637960910796,
"num_tokens": 19685069.0,
"step": 8820
},
{
"entropy": 5.623715114593506,
"epoch": 0.9911832425450665,
"grad_norm": 1.9296875,
"learning_rate": 0.000491285848852719,
"loss": 5.6024,
"mean_token_accuracy": 0.1681622862815857,
"num_tokens": 19697018.0,
"step": 8825
},
{
"entropy": 5.636829423904419,
"epoch": 0.9917448194530241,
"grad_norm": 1.5078125,
"learning_rate": 0.0004912747801469984,
"loss": 5.4772,
"mean_token_accuracy": 0.18188306987285613,
"num_tokens": 19707922.0,
"step": 8830
},
{
"entropy": 5.520493841171264,
"epoch": 0.9923063963609816,
"grad_norm": 1.4140625,
"learning_rate": 0.000491263704555005,
"loss": 5.4086,
"mean_token_accuracy": 0.18108968585729598,
"num_tokens": 19718693.0,
"step": 8835
},
{
"entropy": 5.6003697395324705,
"epoch": 0.9928679732689392,
"grad_norm": 1.9921875,
"learning_rate": 0.0004912526220770913,
"loss": 5.4895,
"mean_token_accuracy": 0.17914043813943864,
"num_tokens": 19730846.0,
"step": 8840
},
{
"entropy": 5.606626510620117,
"epoch": 0.9934295501768967,
"grad_norm": 1.5,
"learning_rate": 0.0004912415327136103,
"loss": 5.4938,
"mean_token_accuracy": 0.18238769322633744,
"num_tokens": 19742099.0,
"step": 8845
},
{
"entropy": 5.617462921142578,
"epoch": 0.9939911270848543,
"grad_norm": 1.8046875,
"learning_rate": 0.000491230436464915,
"loss": 5.5105,
"mean_token_accuracy": 0.17904304414987565,
"num_tokens": 19753216.0,
"step": 8850
},
{
"entropy": 5.592471361160278,
"epoch": 0.9945527039928118,
"grad_norm": 1.640625,
"learning_rate": 0.0004912193333313589,
"loss": 5.4399,
"mean_token_accuracy": 0.1775902733206749,
"num_tokens": 19765316.0,
"step": 8855
},
{
"entropy": 5.604142141342163,
"epoch": 0.9951142809007694,
"grad_norm": 1.4609375,
"learning_rate": 0.0004912082233132951,
"loss": 5.4697,
"mean_token_accuracy": 0.18025630861520767,
"num_tokens": 19776208.0,
"step": 8860
},
{
"entropy": 5.65906400680542,
"epoch": 0.9956758578087269,
"grad_norm": 2.140625,
"learning_rate": 0.0004911971064110778,
"loss": 5.516,
"mean_token_accuracy": 0.18256006240844727,
"num_tokens": 19788065.0,
"step": 8865
},
{
"entropy": 5.610693168640137,
"epoch": 0.9962374347166845,
"grad_norm": 1.4765625,
"learning_rate": 0.0004911859826250607,
"loss": 5.45,
"mean_token_accuracy": 0.17921019494533538,
"num_tokens": 19799388.0,
"step": 8870
},
{
"entropy": 5.604173183441162,
"epoch": 0.996799011624642,
"grad_norm": 1.8203125,
"learning_rate": 0.0004911748519555981,
"loss": 5.5239,
"mean_token_accuracy": 0.17349327206611634,
"num_tokens": 19809933.0,
"step": 8875
},
{
"entropy": 5.742883348464966,
"epoch": 0.9973605885325996,
"grad_norm": 1.546875,
"learning_rate": 0.0004911637144030444,
"loss": 5.6232,
"mean_token_accuracy": 0.17247010320425032,
"num_tokens": 19821566.0,
"step": 8880
},
{
"entropy": 5.730854606628418,
"epoch": 0.9979221654405571,
"grad_norm": 1.921875,
"learning_rate": 0.0004911525699677541,
"loss": 5.5842,
"mean_token_accuracy": 0.1794293627142906,
"num_tokens": 19833727.0,
"step": 8885
},
{
"entropy": 5.679917335510254,
"epoch": 0.9984837423485147,
"grad_norm": 1.6015625,
"learning_rate": 0.0004911414186500822,
"loss": 5.5871,
"mean_token_accuracy": 0.1722554475069046,
"num_tokens": 19845176.0,
"step": 8890
},
{
"entropy": 5.59209976196289,
"epoch": 0.9990453192564722,
"grad_norm": 1.6796875,
"learning_rate": 0.0004911302604503837,
"loss": 5.5041,
"mean_token_accuracy": 0.17181590646505357,
"num_tokens": 19857528.0,
"step": 8895
},
{
"entropy": 5.6015464782714846,
"epoch": 0.9996068961644298,
"grad_norm": 1.359375,
"learning_rate": 0.0004911190953690139,
"loss": 5.4147,
"mean_token_accuracy": 0.18054603338241576,
"num_tokens": 19868232.0,
"step": 8900
},
{
"entropy": 5.562246057722303,
"epoch": 1.0001123153815914,
"grad_norm": 1.6171875,
"learning_rate": 0.0004911079234063284,
"loss": 5.3609,
"mean_token_accuracy": 0.18293264011542001,
"num_tokens": 19877881.0,
"step": 8905
},
{
"entropy": 5.539458322525024,
"epoch": 1.000673892289549,
"grad_norm": 1.859375,
"learning_rate": 0.0004910967445626826,
"loss": 5.4465,
"mean_token_accuracy": 0.18633491098880767,
"num_tokens": 19888398.0,
"step": 8910
},
{
"entropy": 5.642461967468262,
"epoch": 1.0012354691975065,
"grad_norm": 1.5859375,
"learning_rate": 0.0004910855588384327,
"loss": 5.4312,
"mean_token_accuracy": 0.18025522381067277,
"num_tokens": 19899661.0,
"step": 8915
},
{
"entropy": 5.645243978500366,
"epoch": 1.0017970461054642,
"grad_norm": 1.4921875,
"learning_rate": 0.0004910743662339348,
"loss": 5.4736,
"mean_token_accuracy": 0.18382737189531326,
"num_tokens": 19911073.0,
"step": 8920
},
{
"entropy": 5.556240701675415,
"epoch": 1.0023586230134216,
"grad_norm": 1.7421875,
"learning_rate": 0.0004910631667495452,
"loss": 5.4091,
"mean_token_accuracy": 0.18406222015619278,
"num_tokens": 19922567.0,
"step": 8925
},
{
"entropy": 5.616884708404541,
"epoch": 1.0029201999213793,
"grad_norm": 2.84375,
"learning_rate": 0.0004910519603856207,
"loss": 5.5136,
"mean_token_accuracy": 0.1795503541827202,
"num_tokens": 19934547.0,
"step": 8930
},
{
"entropy": 5.61356029510498,
"epoch": 1.0034817768293367,
"grad_norm": 1.6171875,
"learning_rate": 0.0004910407471425179,
"loss": 5.5106,
"mean_token_accuracy": 0.1801147550344467,
"num_tokens": 19947659.0,
"step": 8935
},
{
"entropy": 5.629051876068115,
"epoch": 1.0040433537372944,
"grad_norm": 1.359375,
"learning_rate": 0.000491029527020594,
"loss": 5.3941,
"mean_token_accuracy": 0.17544277608394623,
"num_tokens": 19958150.0,
"step": 8940
},
{
"entropy": 5.533365726470947,
"epoch": 1.0046049306452518,
"grad_norm": 1.4921875,
"learning_rate": 0.0004910183000202061,
"loss": 5.3806,
"mean_token_accuracy": 0.18481847792863845,
"num_tokens": 19968173.0,
"step": 8945
},
{
"entropy": 5.540070486068726,
"epoch": 1.0051665075532095,
"grad_norm": 2.015625,
"learning_rate": 0.0004910070661417119,
"loss": 5.4432,
"mean_token_accuracy": 0.17907491475343704,
"num_tokens": 19978543.0,
"step": 8950
},
{
"entropy": 5.5845966815948485,
"epoch": 1.005728084461167,
"grad_norm": 1.40625,
"learning_rate": 0.0004909958253854687,
"loss": 5.4215,
"mean_token_accuracy": 0.1752791002392769,
"num_tokens": 19989586.0,
"step": 8955
},
{
"entropy": 5.585644245147705,
"epoch": 1.0062896613691246,
"grad_norm": 1.375,
"learning_rate": 0.0004909845777518347,
"loss": 5.4254,
"mean_token_accuracy": 0.18111957609653473,
"num_tokens": 20000524.0,
"step": 8960
},
{
"entropy": 5.660670804977417,
"epoch": 1.006851238277082,
"grad_norm": 1.40625,
"learning_rate": 0.0004909733232411681,
"loss": 5.5361,
"mean_token_accuracy": 0.17346834987401963,
"num_tokens": 20012935.0,
"step": 8965
},
{
"entropy": 5.6432962894439695,
"epoch": 1.0074128151850397,
"grad_norm": 1.390625,
"learning_rate": 0.0004909620618538271,
"loss": 5.4947,
"mean_token_accuracy": 0.17685724496841432,
"num_tokens": 20024435.0,
"step": 8970
},
{
"entropy": 5.63218126296997,
"epoch": 1.007974392092997,
"grad_norm": 2.0625,
"learning_rate": 0.0004909507935901702,
"loss": 5.5035,
"mean_token_accuracy": 0.17428297847509383,
"num_tokens": 20036017.0,
"step": 8975
},
{
"entropy": 5.571589469909668,
"epoch": 1.0085359690009548,
"grad_norm": 1.3984375,
"learning_rate": 0.0004909395184505564,
"loss": 5.4946,
"mean_token_accuracy": 0.17588852047920228,
"num_tokens": 20047216.0,
"step": 8980
},
{
"entropy": 5.5961521625518795,
"epoch": 1.0090975459089122,
"grad_norm": 1.8671875,
"learning_rate": 0.0004909282364353445,
"loss": 5.3567,
"mean_token_accuracy": 0.18210920989513396,
"num_tokens": 20057439.0,
"step": 8985
},
{
"entropy": 5.619443702697754,
"epoch": 1.0096591228168699,
"grad_norm": 1.734375,
"learning_rate": 0.0004909169475448938,
"loss": 5.4823,
"mean_token_accuracy": 0.17967794239521026,
"num_tokens": 20069602.0,
"step": 8990
},
{
"entropy": 5.5129979133605955,
"epoch": 1.0102206997248273,
"grad_norm": 1.5546875,
"learning_rate": 0.0004909056517795638,
"loss": 5.4573,
"mean_token_accuracy": 0.18127780854701997,
"num_tokens": 20081366.0,
"step": 8995
},
{
"entropy": 5.6062075138092045,
"epoch": 1.010782276632785,
"grad_norm": 1.5234375,
"learning_rate": 0.0004908943491397141,
"loss": 5.4481,
"mean_token_accuracy": 0.1833415746688843,
"num_tokens": 20091384.0,
"step": 9000
},
{
"epoch": 1.010782276632785,
"eval_entropy": 5.480559121157233,
"eval_loss": 5.537463665008545,
"eval_mean_token_accuracy": 0.18390065402384423,
"eval_num_tokens": 20091384.0,
"eval_runtime": 23.7803,
"eval_samples_per_second": 1304.147,
"eval_steps_per_second": 163.034,
"step": 9000
},
{
"entropy": 5.691113615036011,
"epoch": 1.0113438535407424,
"grad_norm": 1.4921875,
"learning_rate": 0.0004908830396257045,
"loss": 5.4677,
"mean_token_accuracy": 0.1781607523560524,
"num_tokens": 20102953.0,
"step": 9005
},
{
"entropy": 5.603990077972412,
"epoch": 1.0119054304486998,
"grad_norm": 1.453125,
"learning_rate": 0.0004908717232378953,
"loss": 5.4063,
"mean_token_accuracy": 0.18225425183773042,
"num_tokens": 20113672.0,
"step": 9010
},
{
"entropy": 5.6040881156921385,
"epoch": 1.0124670073566575,
"grad_norm": 1.4609375,
"learning_rate": 0.0004908603999766467,
"loss": 5.4497,
"mean_token_accuracy": 0.18056603372097016,
"num_tokens": 20125143.0,
"step": 9015
},
{
"entropy": 5.5918341159820555,
"epoch": 1.013028584264615,
"grad_norm": 1.2890625,
"learning_rate": 0.0004908490698423191,
"loss": 5.4492,
"mean_token_accuracy": 0.18057763576507568,
"num_tokens": 20135489.0,
"step": 9020
},
{
"entropy": 5.535129261016846,
"epoch": 1.0135901611725726,
"grad_norm": 1.640625,
"learning_rate": 0.0004908377328352736,
"loss": 5.4407,
"mean_token_accuracy": 0.18352580964565277,
"num_tokens": 20146768.0,
"step": 9025
},
{
"entropy": 5.373632717132568,
"epoch": 1.01415173808053,
"grad_norm": 1.4453125,
"learning_rate": 0.0004908263889558709,
"loss": 5.2613,
"mean_token_accuracy": 0.18804869502782823,
"num_tokens": 20156894.0,
"step": 9030
},
{
"entropy": 5.601631212234497,
"epoch": 1.0147133149884877,
"grad_norm": 2.234375,
"learning_rate": 0.0004908150382044722,
"loss": 5.4923,
"mean_token_accuracy": 0.1849379286170006,
"num_tokens": 20168174.0,
"step": 9035
},
{
"entropy": 5.69824914932251,
"epoch": 1.0152748918964452,
"grad_norm": 1.8046875,
"learning_rate": 0.000490803680581439,
"loss": 5.5578,
"mean_token_accuracy": 0.17783243358135223,
"num_tokens": 20179962.0,
"step": 9040
},
{
"entropy": 5.585590744018555,
"epoch": 1.0158364688044028,
"grad_norm": 1.3828125,
"learning_rate": 0.000490792316087133,
"loss": 5.4571,
"mean_token_accuracy": 0.18198922425508499,
"num_tokens": 20189842.0,
"step": 9045
},
{
"entropy": 5.543523645401001,
"epoch": 1.0163980457123603,
"grad_norm": 1.4140625,
"learning_rate": 0.0004907809447219158,
"loss": 5.3633,
"mean_token_accuracy": 0.18173344135284425,
"num_tokens": 20200471.0,
"step": 9050
},
{
"entropy": 5.579350757598877,
"epoch": 1.016959622620318,
"grad_norm": 1.3046875,
"learning_rate": 0.0004907695664861498,
"loss": 5.421,
"mean_token_accuracy": 0.17939640134572982,
"num_tokens": 20212145.0,
"step": 9055
},
{
"entropy": 5.54137225151062,
"epoch": 1.0175211995282754,
"grad_norm": 1.4453125,
"learning_rate": 0.0004907581813801971,
"loss": 5.4286,
"mean_token_accuracy": 0.18690215945243835,
"num_tokens": 20222454.0,
"step": 9060
},
{
"entropy": 5.641435432434082,
"epoch": 1.018082776436233,
"grad_norm": 1.609375,
"learning_rate": 0.0004907467894044202,
"loss": 5.5542,
"mean_token_accuracy": 0.17519815713167192,
"num_tokens": 20233402.0,
"step": 9065
},
{
"entropy": 5.5757880210876465,
"epoch": 1.0186443533441905,
"grad_norm": 1.3515625,
"learning_rate": 0.0004907353905591817,
"loss": 5.353,
"mean_token_accuracy": 0.18518764525651932,
"num_tokens": 20244102.0,
"step": 9070
},
{
"entropy": 5.643955421447754,
"epoch": 1.0192059302521481,
"grad_norm": 1.4765625,
"learning_rate": 0.000490723984844845,
"loss": 5.5567,
"mean_token_accuracy": 0.1780735209584236,
"num_tokens": 20255003.0,
"step": 9075
},
{
"entropy": 5.592462253570557,
"epoch": 1.0197675071601056,
"grad_norm": 1.921875,
"learning_rate": 0.0004907125722617727,
"loss": 5.3919,
"mean_token_accuracy": 0.18512506633996964,
"num_tokens": 20265211.0,
"step": 9080
},
{
"entropy": 5.553256559371948,
"epoch": 1.0203290840680632,
"grad_norm": 1.4375,
"learning_rate": 0.0004907011528103285,
"loss": 5.3715,
"mean_token_accuracy": 0.18377617597579957,
"num_tokens": 20276552.0,
"step": 9085
},
{
"entropy": 5.614937782287598,
"epoch": 1.0208906609760207,
"grad_norm": 1.46875,
"learning_rate": 0.0004906897264908759,
"loss": 5.4936,
"mean_token_accuracy": 0.17746528685092927,
"num_tokens": 20286805.0,
"step": 9090
},
{
"entropy": 5.604516696929932,
"epoch": 1.0214522378839783,
"grad_norm": 1.2421875,
"learning_rate": 0.0004906782933037789,
"loss": 5.4633,
"mean_token_accuracy": 0.17977613657712938,
"num_tokens": 20298536.0,
"step": 9095
},
{
"entropy": 5.542336702346802,
"epoch": 1.0220138147919358,
"grad_norm": 2.109375,
"learning_rate": 0.0004906668532494012,
"loss": 5.4469,
"mean_token_accuracy": 0.17571872472763062,
"num_tokens": 20309842.0,
"step": 9100
},
{
"entropy": 5.541093826293945,
"epoch": 1.0225753916998932,
"grad_norm": 1.375,
"learning_rate": 0.0004906554063281074,
"loss": 5.3586,
"mean_token_accuracy": 0.18835304379463197,
"num_tokens": 20320413.0,
"step": 9105
},
{
"entropy": 5.60887451171875,
"epoch": 1.0231369686078509,
"grad_norm": 1.34375,
"learning_rate": 0.0004906439525402618,
"loss": 5.4304,
"mean_token_accuracy": 0.1837931513786316,
"num_tokens": 20331812.0,
"step": 9110
},
{
"entropy": 5.560403633117676,
"epoch": 1.0236985455158083,
"grad_norm": 1.4140625,
"learning_rate": 0.0004906324918862291,
"loss": 5.348,
"mean_token_accuracy": 0.191845141351223,
"num_tokens": 20342148.0,
"step": 9115
},
{
"entropy": 5.601693248748779,
"epoch": 1.024260122423766,
"grad_norm": 1.2109375,
"learning_rate": 0.0004906210243663742,
"loss": 5.4775,
"mean_token_accuracy": 0.18139202743768693,
"num_tokens": 20353843.0,
"step": 9120
},
{
"entropy": 5.5893807888031,
"epoch": 1.0248216993317234,
"grad_norm": 1.3828125,
"learning_rate": 0.0004906095499810622,
"loss": 5.4662,
"mean_token_accuracy": 0.1824822723865509,
"num_tokens": 20364846.0,
"step": 9125
},
{
"entropy": 5.624654531478882,
"epoch": 1.025383276239681,
"grad_norm": 1.3828125,
"learning_rate": 0.0004905980687306586,
"loss": 5.4121,
"mean_token_accuracy": 0.1834669217467308,
"num_tokens": 20376874.0,
"step": 9130
},
{
"entropy": 5.52737340927124,
"epoch": 1.0259448531476385,
"grad_norm": 1.71875,
"learning_rate": 0.0004905865806155288,
"loss": 5.4147,
"mean_token_accuracy": 0.1804659754037857,
"num_tokens": 20388189.0,
"step": 9135
},
{
"entropy": 5.579295015335083,
"epoch": 1.0265064300555962,
"grad_norm": 1.546875,
"learning_rate": 0.0004905750856360387,
"loss": 5.4464,
"mean_token_accuracy": 0.17909622192382812,
"num_tokens": 20399155.0,
"step": 9140
},
{
"entropy": 5.574539470672607,
"epoch": 1.0270680069635536,
"grad_norm": 2.0625,
"learning_rate": 0.0004905635837925543,
"loss": 5.3352,
"mean_token_accuracy": 0.18723117113113402,
"num_tokens": 20410569.0,
"step": 9145
},
{
"entropy": 5.487085962295533,
"epoch": 1.0276295838715113,
"grad_norm": 1.2421875,
"learning_rate": 0.0004905520750854418,
"loss": 5.3868,
"mean_token_accuracy": 0.1845887452363968,
"num_tokens": 20420438.0,
"step": 9150
},
{
"entropy": 5.608818244934082,
"epoch": 1.0281911607794687,
"grad_norm": 1.703125,
"learning_rate": 0.0004905405595150674,
"loss": 5.4525,
"mean_token_accuracy": 0.17809222787618637,
"num_tokens": 20431874.0,
"step": 9155
},
{
"entropy": 5.5994439125061035,
"epoch": 1.0287527376874264,
"grad_norm": 1.3046875,
"learning_rate": 0.0004905290370817982,
"loss": 5.4328,
"mean_token_accuracy": 0.17693003863096238,
"num_tokens": 20442792.0,
"step": 9160
},
{
"entropy": 5.572349596023559,
"epoch": 1.0293143145953838,
"grad_norm": 1.375,
"learning_rate": 0.000490517507786001,
"loss": 5.4362,
"mean_token_accuracy": 0.1795343801379204,
"num_tokens": 20454410.0,
"step": 9165
},
{
"entropy": 5.672254419326782,
"epoch": 1.0298758915033415,
"grad_norm": 1.40625,
"learning_rate": 0.0004905059716280425,
"loss": 5.5683,
"mean_token_accuracy": 0.17336199283599854,
"num_tokens": 20467072.0,
"step": 9170
},
{
"entropy": 5.609378290176392,
"epoch": 1.030437468411299,
"grad_norm": 2.328125,
"learning_rate": 0.0004904944286082904,
"loss": 5.4962,
"mean_token_accuracy": 0.17962452471256257,
"num_tokens": 20477697.0,
"step": 9175
},
{
"entropy": 5.550300884246826,
"epoch": 1.0309990453192566,
"grad_norm": 1.46875,
"learning_rate": 0.0004904828787271121,
"loss": 5.3361,
"mean_token_accuracy": 0.18696321547031403,
"num_tokens": 20489024.0,
"step": 9180
},
{
"entropy": 5.53287148475647,
"epoch": 1.031560622227214,
"grad_norm": 1.4609375,
"learning_rate": 0.0004904713219848753,
"loss": 5.4164,
"mean_token_accuracy": 0.1792910784482956,
"num_tokens": 20499928.0,
"step": 9185
},
{
"entropy": 5.53876633644104,
"epoch": 1.0321221991351717,
"grad_norm": 1.328125,
"learning_rate": 0.0004904597583819481,
"loss": 5.4654,
"mean_token_accuracy": 0.17975662052631378,
"num_tokens": 20511085.0,
"step": 9190
},
{
"entropy": 5.6245105266571045,
"epoch": 1.032683776043129,
"grad_norm": 1.609375,
"learning_rate": 0.0004904481879186987,
"loss": 5.328,
"mean_token_accuracy": 0.18221411108970642,
"num_tokens": 20522164.0,
"step": 9195
},
{
"entropy": 5.588701152801514,
"epoch": 1.0332453529510865,
"grad_norm": 1.296875,
"learning_rate": 0.0004904366105954954,
"loss": 5.5253,
"mean_token_accuracy": 0.1788549467921257,
"num_tokens": 20534187.0,
"step": 9200
},
{
"entropy": 5.654664373397827,
"epoch": 1.0338069298590442,
"grad_norm": 1.5703125,
"learning_rate": 0.0004904250264127067,
"loss": 5.4762,
"mean_token_accuracy": 0.18382192701101302,
"num_tokens": 20545819.0,
"step": 9205
},
{
"entropy": 5.6495167255401615,
"epoch": 1.0343685067670016,
"grad_norm": 1.453125,
"learning_rate": 0.0004904134353707018,
"loss": 5.4851,
"mean_token_accuracy": 0.1770500898361206,
"num_tokens": 20556803.0,
"step": 9210
},
{
"entropy": 5.597296667098999,
"epoch": 1.0349300836749593,
"grad_norm": 1.6953125,
"learning_rate": 0.0004904018374698496,
"loss": 5.4868,
"mean_token_accuracy": 0.17946896255016326,
"num_tokens": 20567366.0,
"step": 9215
},
{
"entropy": 5.5590893745422365,
"epoch": 1.0354916605829168,
"grad_norm": 1.3671875,
"learning_rate": 0.0004903902327105191,
"loss": 5.5067,
"mean_token_accuracy": 0.17471158802509307,
"num_tokens": 20579182.0,
"step": 9220
},
{
"entropy": 5.565256214141845,
"epoch": 1.0360532374908744,
"grad_norm": 1.3671875,
"learning_rate": 0.0004903786210930803,
"loss": 5.3575,
"mean_token_accuracy": 0.18208726793527602,
"num_tokens": 20590346.0,
"step": 9225
},
{
"entropy": 5.5901665687561035,
"epoch": 1.0366148143988319,
"grad_norm": 1.5078125,
"learning_rate": 0.0004903670026179025,
"loss": 5.4691,
"mean_token_accuracy": 0.18377649337053298,
"num_tokens": 20601279.0,
"step": 9230
},
{
"entropy": 5.5744115829467775,
"epoch": 1.0371763913067895,
"grad_norm": 1.53125,
"learning_rate": 0.000490355377285356,
"loss": 5.3805,
"mean_token_accuracy": 0.18020678460597991,
"num_tokens": 20612163.0,
"step": 9235
},
{
"entropy": 5.60702919960022,
"epoch": 1.037737968214747,
"grad_norm": 1.296875,
"learning_rate": 0.0004903437450958106,
"loss": 5.347,
"mean_token_accuracy": 0.18913260847330093,
"num_tokens": 20622587.0,
"step": 9240
},
{
"entropy": 5.58750729560852,
"epoch": 1.0382995451227046,
"grad_norm": 1.4375,
"learning_rate": 0.0004903321060496369,
"loss": 5.5223,
"mean_token_accuracy": 0.17142360657453537,
"num_tokens": 20633341.0,
"step": 9245
},
{
"entropy": 5.495819091796875,
"epoch": 1.038861122030662,
"grad_norm": 1.6328125,
"learning_rate": 0.0004903204601472054,
"loss": 5.4048,
"mean_token_accuracy": 0.18205715268850325,
"num_tokens": 20644582.0,
"step": 9250
},
{
"entropy": 5.603433322906494,
"epoch": 1.0394226989386197,
"grad_norm": 1.875,
"learning_rate": 0.0004903088073888869,
"loss": 5.3732,
"mean_token_accuracy": 0.19030302315950393,
"num_tokens": 20654936.0,
"step": 9255
},
{
"entropy": 5.563142204284668,
"epoch": 1.0399842758465772,
"grad_norm": 1.71875,
"learning_rate": 0.0004902971477750526,
"loss": 5.392,
"mean_token_accuracy": 0.18279054313898085,
"num_tokens": 20666348.0,
"step": 9260
},
{
"entropy": 5.637576532363892,
"epoch": 1.0405458527545348,
"grad_norm": 1.5859375,
"learning_rate": 0.0004902854813060736,
"loss": 5.5547,
"mean_token_accuracy": 0.17308980077505112,
"num_tokens": 20677482.0,
"step": 9265
},
{
"entropy": 5.641957092285156,
"epoch": 1.0411074296624923,
"grad_norm": 1.65625,
"learning_rate": 0.0004902738079823213,
"loss": 5.4744,
"mean_token_accuracy": 0.18279003202915192,
"num_tokens": 20688369.0,
"step": 9270
},
{
"entropy": 5.59079270362854,
"epoch": 1.04166900657045,
"grad_norm": 1.890625,
"learning_rate": 0.0004902621278041674,
"loss": 5.4776,
"mean_token_accuracy": 0.1730587124824524,
"num_tokens": 20698933.0,
"step": 9275
},
{
"entropy": 5.571591997146607,
"epoch": 1.0422305834784074,
"grad_norm": 1.3203125,
"learning_rate": 0.0004902504407719839,
"loss": 5.3587,
"mean_token_accuracy": 0.185298191010952,
"num_tokens": 20710192.0,
"step": 9280
},
{
"entropy": 5.5448774814605715,
"epoch": 1.042792160386365,
"grad_norm": 1.7421875,
"learning_rate": 0.0004902387468861429,
"loss": 5.4052,
"mean_token_accuracy": 0.18007587641477585,
"num_tokens": 20721408.0,
"step": 9285
},
{
"entropy": 5.49824595451355,
"epoch": 1.0433537372943225,
"grad_norm": 1.8984375,
"learning_rate": 0.0004902270461470169,
"loss": 5.3154,
"mean_token_accuracy": 0.18607982248067856,
"num_tokens": 20732155.0,
"step": 9290
},
{
"entropy": 5.546027183532715,
"epoch": 1.04391531420228,
"grad_norm": 1.3203125,
"learning_rate": 0.000490215338554978,
"loss": 5.3144,
"mean_token_accuracy": 0.1886076509952545,
"num_tokens": 20741819.0,
"step": 9295
},
{
"entropy": 5.550833320617675,
"epoch": 1.0444768911102376,
"grad_norm": 2.03125,
"learning_rate": 0.0004902036241103994,
"loss": 5.4538,
"mean_token_accuracy": 0.17962786108255385,
"num_tokens": 20754931.0,
"step": 9300
},
{
"entropy": 5.599490976333618,
"epoch": 1.045038468018195,
"grad_norm": 1.7109375,
"learning_rate": 0.0004901919028136539,
"loss": 5.5344,
"mean_token_accuracy": 0.17407327592372895,
"num_tokens": 20766533.0,
"step": 9305
},
{
"entropy": 5.590666437149048,
"epoch": 1.0456000449261527,
"grad_norm": 1.7109375,
"learning_rate": 0.0004901801746651147,
"loss": 5.3912,
"mean_token_accuracy": 0.18835558146238326,
"num_tokens": 20777055.0,
"step": 9310
},
{
"entropy": 5.483608150482178,
"epoch": 1.04616162183411,
"grad_norm": 1.4609375,
"learning_rate": 0.0004901684396651554,
"loss": 5.3269,
"mean_token_accuracy": 0.18506430685520173,
"num_tokens": 20786871.0,
"step": 9315
},
{
"entropy": 5.6322643756866455,
"epoch": 1.0467231987420678,
"grad_norm": 1.5078125,
"learning_rate": 0.0004901566978141494,
"loss": 5.4943,
"mean_token_accuracy": 0.17778564393520355,
"num_tokens": 20798546.0,
"step": 9320
},
{
"entropy": 5.616176462173462,
"epoch": 1.0472847756500252,
"grad_norm": 1.359375,
"learning_rate": 0.0004901449491124708,
"loss": 5.4207,
"mean_token_accuracy": 0.1779840424656868,
"num_tokens": 20809657.0,
"step": 9325
},
{
"entropy": 5.534231042861938,
"epoch": 1.0478463525579829,
"grad_norm": 1.59375,
"learning_rate": 0.0004901331935604936,
"loss": 5.4322,
"mean_token_accuracy": 0.17593244463205338,
"num_tokens": 20819662.0,
"step": 9330
},
{
"entropy": 5.477989149093628,
"epoch": 1.0484079294659403,
"grad_norm": 1.734375,
"learning_rate": 0.000490121431158592,
"loss": 5.3112,
"mean_token_accuracy": 0.1878904953598976,
"num_tokens": 20830554.0,
"step": 9335
},
{
"entropy": 5.5836528778076175,
"epoch": 1.048969506373898,
"grad_norm": 1.546875,
"learning_rate": 0.0004901096619071407,
"loss": 5.4813,
"mean_token_accuracy": 0.17338499724864959,
"num_tokens": 20842003.0,
"step": 9340
},
{
"entropy": 5.564203596115112,
"epoch": 1.0495310832818554,
"grad_norm": 1.5,
"learning_rate": 0.0004900978858065142,
"loss": 5.4577,
"mean_token_accuracy": 0.18443517833948136,
"num_tokens": 20853453.0,
"step": 9345
},
{
"entropy": 5.701768493652343,
"epoch": 1.050092660189813,
"grad_norm": 1.4453125,
"learning_rate": 0.0004900861028570877,
"loss": 5.5307,
"mean_token_accuracy": 0.17674930691719054,
"num_tokens": 20867296.0,
"step": 9350
},
{
"entropy": 5.510841417312622,
"epoch": 1.0506542370977705,
"grad_norm": 1.3671875,
"learning_rate": 0.0004900743130592362,
"loss": 5.4277,
"mean_token_accuracy": 0.18456286936998367,
"num_tokens": 20878951.0,
"step": 9355
},
{
"entropy": 5.653782701492309,
"epoch": 1.0512158140057282,
"grad_norm": 1.5078125,
"learning_rate": 0.0004900625164133351,
"loss": 5.4964,
"mean_token_accuracy": 0.17913446426391602,
"num_tokens": 20891103.0,
"step": 9360
},
{
"entropy": 5.65149302482605,
"epoch": 1.0517773909136856,
"grad_norm": 1.84375,
"learning_rate": 0.0004900507129197601,
"loss": 5.5754,
"mean_token_accuracy": 0.17431244999170303,
"num_tokens": 20903570.0,
"step": 9365
},
{
"entropy": 5.618903541564942,
"epoch": 1.0523389678216433,
"grad_norm": 1.5,
"learning_rate": 0.0004900389025788871,
"loss": 5.5061,
"mean_token_accuracy": 0.18127326369285585,
"num_tokens": 20914156.0,
"step": 9370
},
{
"entropy": 5.577870607376099,
"epoch": 1.0529005447296007,
"grad_norm": 1.4765625,
"learning_rate": 0.000490027085391092,
"loss": 5.4117,
"mean_token_accuracy": 0.18191829770803453,
"num_tokens": 20925192.0,
"step": 9375
},
{
"entropy": 5.585703182220459,
"epoch": 1.0534621216375584,
"grad_norm": 1.328125,
"learning_rate": 0.000490015261356751,
"loss": 5.3052,
"mean_token_accuracy": 0.19129057973623276,
"num_tokens": 20935281.0,
"step": 9380
},
{
"entropy": 5.664695072174072,
"epoch": 1.0540236985455158,
"grad_norm": 1.359375,
"learning_rate": 0.0004900034304762408,
"loss": 5.5269,
"mean_token_accuracy": 0.1798120155930519,
"num_tokens": 20947397.0,
"step": 9385
},
{
"entropy": 5.517253017425537,
"epoch": 1.0545852754534732,
"grad_norm": 1.6875,
"learning_rate": 0.000489991592749938,
"loss": 5.4103,
"mean_token_accuracy": 0.18664222508668898,
"num_tokens": 20958473.0,
"step": 9390
},
{
"entropy": 5.717044496536255,
"epoch": 1.055146852361431,
"grad_norm": 1.609375,
"learning_rate": 0.0004899797481782193,
"loss": 5.5098,
"mean_token_accuracy": 0.17320843338966369,
"num_tokens": 20970112.0,
"step": 9395
},
{
"entropy": 5.536887359619141,
"epoch": 1.0557084292693883,
"grad_norm": 1.296875,
"learning_rate": 0.0004899678967614623,
"loss": 5.4111,
"mean_token_accuracy": 0.18350721150636673,
"num_tokens": 20980762.0,
"step": 9400
},
{
"entropy": 5.546196126937867,
"epoch": 1.056270006177346,
"grad_norm": 1.390625,
"learning_rate": 0.0004899560385000439,
"loss": 5.3186,
"mean_token_accuracy": 0.18761096000671387,
"num_tokens": 20991663.0,
"step": 9405
},
{
"entropy": 5.518143081665039,
"epoch": 1.0568315830853034,
"grad_norm": 1.6796875,
"learning_rate": 0.000489944173394342,
"loss": 5.3317,
"mean_token_accuracy": 0.18784303665161134,
"num_tokens": 21002121.0,
"step": 9410
},
{
"entropy": 5.542761516571045,
"epoch": 1.0573931599932611,
"grad_norm": 1.484375,
"learning_rate": 0.0004899323014447342,
"loss": 5.4147,
"mean_token_accuracy": 0.17626750022172927,
"num_tokens": 21013791.0,
"step": 9415
},
{
"entropy": 5.550657606124878,
"epoch": 1.0579547369012186,
"grad_norm": 1.96875,
"learning_rate": 0.0004899204226515985,
"loss": 5.3462,
"mean_token_accuracy": 0.1866498664021492,
"num_tokens": 21022977.0,
"step": 9420
},
{
"entropy": 5.573056221008301,
"epoch": 1.0585163138091762,
"grad_norm": 1.9296875,
"learning_rate": 0.0004899085370153133,
"loss": 5.5053,
"mean_token_accuracy": 0.18175559490919113,
"num_tokens": 21036372.0,
"step": 9425
},
{
"entropy": 5.691931056976318,
"epoch": 1.0590778907171337,
"grad_norm": 1.5,
"learning_rate": 0.0004898966445362569,
"loss": 5.5844,
"mean_token_accuracy": 0.1683250144124031,
"num_tokens": 21047984.0,
"step": 9430
},
{
"entropy": 5.670980310440063,
"epoch": 1.0596394676250913,
"grad_norm": 1.3515625,
"learning_rate": 0.0004898847452148079,
"loss": 5.4507,
"mean_token_accuracy": 0.1792193114757538,
"num_tokens": 21060127.0,
"step": 9435
},
{
"entropy": 5.593993377685547,
"epoch": 1.0602010445330488,
"grad_norm": 1.5234375,
"learning_rate": 0.0004898728390513454,
"loss": 5.4223,
"mean_token_accuracy": 0.17926856875419617,
"num_tokens": 21070659.0,
"step": 9440
},
{
"entropy": 5.565860843658447,
"epoch": 1.0607626214410064,
"grad_norm": 1.40625,
"learning_rate": 0.0004898609260462483,
"loss": 5.3858,
"mean_token_accuracy": 0.18973694741725922,
"num_tokens": 21081667.0,
"step": 9445
},
{
"entropy": 5.589852952957154,
"epoch": 1.0613241983489639,
"grad_norm": 1.4296875,
"learning_rate": 0.0004898490061998961,
"loss": 5.3988,
"mean_token_accuracy": 0.18442354649305343,
"num_tokens": 21094275.0,
"step": 9450
},
{
"entropy": 5.5819648742675785,
"epoch": 1.0618857752569215,
"grad_norm": 1.46875,
"learning_rate": 0.0004898370795126681,
"loss": 5.4607,
"mean_token_accuracy": 0.17688476741313935,
"num_tokens": 21106360.0,
"step": 9455
},
{
"entropy": 5.612158203125,
"epoch": 1.062447352164879,
"grad_norm": 1.296875,
"learning_rate": 0.0004898251459849441,
"loss": 5.4207,
"mean_token_accuracy": 0.17969983965158462,
"num_tokens": 21117647.0,
"step": 9460
},
{
"entropy": 5.523744916915893,
"epoch": 1.0630089290728366,
"grad_norm": 2.671875,
"learning_rate": 0.0004898132056171041,
"loss": 5.2776,
"mean_token_accuracy": 0.1911955028772354,
"num_tokens": 21129599.0,
"step": 9465
},
{
"entropy": 5.53771333694458,
"epoch": 1.063570505980794,
"grad_norm": 1.578125,
"learning_rate": 0.0004898012584095285,
"loss": 5.3936,
"mean_token_accuracy": 0.18242655992507933,
"num_tokens": 21140573.0,
"step": 9470
},
{
"entropy": 5.550946378707886,
"epoch": 1.0641320828887517,
"grad_norm": 1.4765625,
"learning_rate": 0.0004897893043625975,
"loss": 5.484,
"mean_token_accuracy": 0.18450344651937484,
"num_tokens": 21151899.0,
"step": 9475
},
{
"entropy": 5.662395000457764,
"epoch": 1.0646936597967092,
"grad_norm": 1.46875,
"learning_rate": 0.0004897773434766917,
"loss": 5.5427,
"mean_token_accuracy": 0.1746914952993393,
"num_tokens": 21162748.0,
"step": 9480
},
{
"entropy": 5.640436840057373,
"epoch": 1.0652552367046666,
"grad_norm": 1.484375,
"learning_rate": 0.000489765375752192,
"loss": 5.4205,
"mean_token_accuracy": 0.1824389725923538,
"num_tokens": 21172962.0,
"step": 9485
},
{
"entropy": 5.578457355499268,
"epoch": 1.0658168136126243,
"grad_norm": 1.3203125,
"learning_rate": 0.0004897534011894793,
"loss": 5.4295,
"mean_token_accuracy": 0.18313320875167846,
"num_tokens": 21183805.0,
"step": 9490
},
{
"entropy": 5.5050335884094235,
"epoch": 1.0663783905205817,
"grad_norm": 1.4375,
"learning_rate": 0.0004897414197889352,
"loss": 5.3376,
"mean_token_accuracy": 0.1862918108701706,
"num_tokens": 21195129.0,
"step": 9495
},
{
"entropy": 5.514494466781616,
"epoch": 1.0669399674285394,
"grad_norm": 1.6875,
"learning_rate": 0.000489729431550941,
"loss": 5.4572,
"mean_token_accuracy": 0.17399467676877975,
"num_tokens": 21206300.0,
"step": 9500
},
{
"entropy": 5.5885984897613525,
"epoch": 1.0675015443364968,
"grad_norm": 1.4375,
"learning_rate": 0.0004897174364758783,
"loss": 5.4012,
"mean_token_accuracy": 0.1795855462551117,
"num_tokens": 21217137.0,
"step": 9505
},
{
"entropy": 5.585361433029175,
"epoch": 1.0680631212444545,
"grad_norm": 1.5,
"learning_rate": 0.0004897054345641291,
"loss": 5.3814,
"mean_token_accuracy": 0.18651138097047806,
"num_tokens": 21228399.0,
"step": 9510
},
{
"entropy": 5.591248798370361,
"epoch": 1.068624698152412,
"grad_norm": 1.4609375,
"learning_rate": 0.0004896934258160757,
"loss": 5.4908,
"mean_token_accuracy": 0.1792047217488289,
"num_tokens": 21239396.0,
"step": 9515
},
{
"entropy": 5.542936086654663,
"epoch": 1.0691862750603696,
"grad_norm": 1.6640625,
"learning_rate": 0.0004896814102321004,
"loss": 5.3795,
"mean_token_accuracy": 0.18302502185106279,
"num_tokens": 21249966.0,
"step": 9520
},
{
"entropy": 5.578098154067993,
"epoch": 1.069747851968327,
"grad_norm": 1.4140625,
"learning_rate": 0.0004896693878125857,
"loss": 5.4004,
"mean_token_accuracy": 0.18019044697284697,
"num_tokens": 21260157.0,
"step": 9525
},
{
"entropy": 5.556897497177124,
"epoch": 1.0703094288762847,
"grad_norm": 1.46875,
"learning_rate": 0.0004896573585579144,
"loss": 5.468,
"mean_token_accuracy": 0.1864493280649185,
"num_tokens": 21270738.0,
"step": 9530
},
{
"entropy": 5.54735255241394,
"epoch": 1.070871005784242,
"grad_norm": 1.734375,
"learning_rate": 0.0004896453224684695,
"loss": 5.4107,
"mean_token_accuracy": 0.18065569996833802,
"num_tokens": 21282469.0,
"step": 9535
},
{
"entropy": 5.5747642993927,
"epoch": 1.0714325826921998,
"grad_norm": 1.5078125,
"learning_rate": 0.0004896332795446345,
"loss": 5.3681,
"mean_token_accuracy": 0.18048644065856934,
"num_tokens": 21293474.0,
"step": 9540
},
{
"entropy": 5.59945912361145,
"epoch": 1.0719941596001572,
"grad_norm": 2.140625,
"learning_rate": 0.0004896212297867924,
"loss": 5.4915,
"mean_token_accuracy": 0.1727932184934616,
"num_tokens": 21304904.0,
"step": 9545
},
{
"entropy": 5.654782056808472,
"epoch": 1.0725557365081149,
"grad_norm": 1.5625,
"learning_rate": 0.0004896091731953271,
"loss": 5.4232,
"mean_token_accuracy": 0.17975277155637742,
"num_tokens": 21315407.0,
"step": 9550
},
{
"entropy": 5.5493122100830075,
"epoch": 1.0731173134160723,
"grad_norm": 2.421875,
"learning_rate": 0.0004895971097706227,
"loss": 5.3632,
"mean_token_accuracy": 0.18576829135417938,
"num_tokens": 21326496.0,
"step": 9555
},
{
"entropy": 5.543086862564087,
"epoch": 1.07367889032403,
"grad_norm": 1.40625,
"learning_rate": 0.0004895850395130629,
"loss": 5.455,
"mean_token_accuracy": 0.17998800724744796,
"num_tokens": 21339449.0,
"step": 9560
},
{
"entropy": 5.541301727294922,
"epoch": 1.0742404672319874,
"grad_norm": 1.6484375,
"learning_rate": 0.0004895729624230323,
"loss": 5.519,
"mean_token_accuracy": 0.17460471093654634,
"num_tokens": 21350431.0,
"step": 9565
},
{
"entropy": 5.565596199035644,
"epoch": 1.074802044139945,
"grad_norm": 1.59375,
"learning_rate": 0.0004895608785009153,
"loss": 5.407,
"mean_token_accuracy": 0.18352611511945724,
"num_tokens": 21361168.0,
"step": 9570
},
{
"entropy": 5.59822564125061,
"epoch": 1.0753636210479025,
"grad_norm": 1.5625,
"learning_rate": 0.0004895487877470969,
"loss": 5.3979,
"mean_token_accuracy": 0.1866555020213127,
"num_tokens": 21372635.0,
"step": 9575
},
{
"entropy": 5.468242645263672,
"epoch": 1.07592519795586,
"grad_norm": 1.828125,
"learning_rate": 0.0004895366901619617,
"loss": 5.3891,
"mean_token_accuracy": 0.17935798466205596,
"num_tokens": 21384051.0,
"step": 9580
},
{
"entropy": 5.569312000274659,
"epoch": 1.0764867748638176,
"grad_norm": 1.46875,
"learning_rate": 0.0004895245857458951,
"loss": 5.4206,
"mean_token_accuracy": 0.1748897612094879,
"num_tokens": 21397496.0,
"step": 9585
},
{
"entropy": 5.565934467315674,
"epoch": 1.077048351771775,
"grad_norm": 1.3671875,
"learning_rate": 0.0004895124744992825,
"loss": 5.4898,
"mean_token_accuracy": 0.1753292426466942,
"num_tokens": 21409349.0,
"step": 9590
},
{
"entropy": 5.582330703735352,
"epoch": 1.0776099286797327,
"grad_norm": 1.46875,
"learning_rate": 0.0004895003564225095,
"loss": 5.4631,
"mean_token_accuracy": 0.18393794894218446,
"num_tokens": 21421843.0,
"step": 9595
},
{
"entropy": 5.5389892578125,
"epoch": 1.0781715055876901,
"grad_norm": 1.640625,
"learning_rate": 0.0004894882315159621,
"loss": 5.3112,
"mean_token_accuracy": 0.1861307591199875,
"num_tokens": 21433230.0,
"step": 9600
},
{
"entropy": 5.589205980300903,
"epoch": 1.0787330824956478,
"grad_norm": 1.671875,
"learning_rate": 0.000489476099780026,
"loss": 5.4496,
"mean_token_accuracy": 0.1781141757965088,
"num_tokens": 21444295.0,
"step": 9605
},
{
"entropy": 5.607203674316406,
"epoch": 1.0792946594036052,
"grad_norm": 1.421875,
"learning_rate": 0.0004894639612150879,
"loss": 5.3914,
"mean_token_accuracy": 0.17885364592075348,
"num_tokens": 21455721.0,
"step": 9610
},
{
"entropy": 5.572823572158813,
"epoch": 1.079856236311563,
"grad_norm": 1.4375,
"learning_rate": 0.0004894518158215339,
"loss": 5.417,
"mean_token_accuracy": 0.18138462007045747,
"num_tokens": 21466514.0,
"step": 9615
},
{
"entropy": 5.550282049179077,
"epoch": 1.0804178132195204,
"grad_norm": 1.984375,
"learning_rate": 0.000489439663599751,
"loss": 5.4382,
"mean_token_accuracy": 0.17771633267402648,
"num_tokens": 21478782.0,
"step": 9620
},
{
"entropy": 5.597379207611084,
"epoch": 1.080979390127478,
"grad_norm": 2.765625,
"learning_rate": 0.000489427504550126,
"loss": 5.4343,
"mean_token_accuracy": 0.18803733289241792,
"num_tokens": 21490428.0,
"step": 9625
},
{
"entropy": 5.434914636611938,
"epoch": 1.0815409670354355,
"grad_norm": 1.6015625,
"learning_rate": 0.0004894153386730462,
"loss": 5.2931,
"mean_token_accuracy": 0.18199682086706162,
"num_tokens": 21501530.0,
"step": 9630
},
{
"entropy": 5.531746816635132,
"epoch": 1.0821025439433931,
"grad_norm": 2.0,
"learning_rate": 0.0004894031659688988,
"loss": 5.3693,
"mean_token_accuracy": 0.18740325123071672,
"num_tokens": 21512122.0,
"step": 9635
},
{
"entropy": 5.512206077575684,
"epoch": 1.0826641208513506,
"grad_norm": 1.4765625,
"learning_rate": 0.0004893909864380714,
"loss": 5.2966,
"mean_token_accuracy": 0.1824502244591713,
"num_tokens": 21521896.0,
"step": 9640
},
{
"entropy": 5.563923120498657,
"epoch": 1.0832256977593082,
"grad_norm": 1.4609375,
"learning_rate": 0.000489378800080952,
"loss": 5.4808,
"mean_token_accuracy": 0.18009958118200303,
"num_tokens": 21533584.0,
"step": 9645
},
{
"entropy": 5.519917678833008,
"epoch": 1.0837872746672657,
"grad_norm": 1.5546875,
"learning_rate": 0.0004893666068979283,
"loss": 5.4307,
"mean_token_accuracy": 0.18158730268478393,
"num_tokens": 21545432.0,
"step": 9650
},
{
"entropy": 5.574348258972168,
"epoch": 1.0843488515752233,
"grad_norm": 1.6953125,
"learning_rate": 0.0004893544068893887,
"loss": 5.4372,
"mean_token_accuracy": 0.1786968931555748,
"num_tokens": 21556547.0,
"step": 9655
},
{
"entropy": 5.554831790924072,
"epoch": 1.0849104284831808,
"grad_norm": 1.3671875,
"learning_rate": 0.0004893422000557217,
"loss": 5.4916,
"mean_token_accuracy": 0.1782551109790802,
"num_tokens": 21567354.0,
"step": 9660
},
{
"entropy": 5.596163082122803,
"epoch": 1.0854720053911384,
"grad_norm": 1.515625,
"learning_rate": 0.0004893299863973159,
"loss": 5.3728,
"mean_token_accuracy": 0.18017571568489074,
"num_tokens": 21578694.0,
"step": 9665
},
{
"entropy": 5.505556535720825,
"epoch": 1.0860335822990959,
"grad_norm": 1.40625,
"learning_rate": 0.0004893177659145601,
"loss": 5.3177,
"mean_token_accuracy": 0.19020393937826158,
"num_tokens": 21589365.0,
"step": 9670
},
{
"entropy": 5.54207854270935,
"epoch": 1.0865951592070533,
"grad_norm": 1.8046875,
"learning_rate": 0.0004893055386078438,
"loss": 5.3774,
"mean_token_accuracy": 0.1809840142726898,
"num_tokens": 21600766.0,
"step": 9675
},
{
"entropy": 5.614822673797607,
"epoch": 1.087156736115011,
"grad_norm": 1.6796875,
"learning_rate": 0.0004892933044775558,
"loss": 5.584,
"mean_token_accuracy": 0.17393954694271088,
"num_tokens": 21613969.0,
"step": 9680
},
{
"entropy": 5.558384132385254,
"epoch": 1.0877183130229684,
"grad_norm": 1.609375,
"learning_rate": 0.0004892810635240859,
"loss": 5.4117,
"mean_token_accuracy": 0.18534248769283296,
"num_tokens": 21626302.0,
"step": 9685
},
{
"entropy": 5.591034650802612,
"epoch": 1.088279889930926,
"grad_norm": 1.484375,
"learning_rate": 0.0004892688157478238,
"loss": 5.4689,
"mean_token_accuracy": 0.18126342445611954,
"num_tokens": 21637058.0,
"step": 9690
},
{
"entropy": 5.602719736099243,
"epoch": 1.0888414668388835,
"grad_norm": 1.7890625,
"learning_rate": 0.0004892565611491597,
"loss": 5.491,
"mean_token_accuracy": 0.17440066039562224,
"num_tokens": 21648729.0,
"step": 9695
},
{
"entropy": 5.500168991088867,
"epoch": 1.0894030437468412,
"grad_norm": 1.6484375,
"learning_rate": 0.0004892442997284834,
"loss": 5.2885,
"mean_token_accuracy": 0.18886132836341857,
"num_tokens": 21659961.0,
"step": 9700
},
{
"entropy": 5.561039829254151,
"epoch": 1.0899646206547986,
"grad_norm": 2.046875,
"learning_rate": 0.0004892320314861855,
"loss": 5.3975,
"mean_token_accuracy": 0.1876556694507599,
"num_tokens": 21670577.0,
"step": 9705
},
{
"entropy": 5.520331525802613,
"epoch": 1.0905261975627563,
"grad_norm": 2.203125,
"learning_rate": 0.0004892197564226568,
"loss": 5.3512,
"mean_token_accuracy": 0.19123491048812866,
"num_tokens": 21679882.0,
"step": 9710
},
{
"entropy": 5.5742918968200685,
"epoch": 1.0910877744707137,
"grad_norm": 1.7734375,
"learning_rate": 0.0004892074745382878,
"loss": 5.4671,
"mean_token_accuracy": 0.18306453973054887,
"num_tokens": 21690595.0,
"step": 9715
},
{
"entropy": 5.552133989334107,
"epoch": 1.0916493513786714,
"grad_norm": 1.65625,
"learning_rate": 0.0004891951858334698,
"loss": 5.341,
"mean_token_accuracy": 0.18278181701898574,
"num_tokens": 21701251.0,
"step": 9720
},
{
"entropy": 5.628836297988892,
"epoch": 1.0922109282866288,
"grad_norm": 2.1875,
"learning_rate": 0.0004891828903085939,
"loss": 5.5265,
"mean_token_accuracy": 0.17879033833742142,
"num_tokens": 21713286.0,
"step": 9725
},
{
"entropy": 5.638147830963135,
"epoch": 1.0927725051945865,
"grad_norm": 1.6640625,
"learning_rate": 0.0004891705879640518,
"loss": 5.5545,
"mean_token_accuracy": 0.17670438587665557,
"num_tokens": 21725508.0,
"step": 9730
},
{
"entropy": 5.621787881851196,
"epoch": 1.093334082102544,
"grad_norm": 1.7734375,
"learning_rate": 0.0004891582788002351,
"loss": 5.3617,
"mean_token_accuracy": 0.18192603141069413,
"num_tokens": 21736467.0,
"step": 9735
},
{
"entropy": 5.675878572463989,
"epoch": 1.0938956590105016,
"grad_norm": 1.6796875,
"learning_rate": 0.0004891459628175358,
"loss": 5.4562,
"mean_token_accuracy": 0.1849345237016678,
"num_tokens": 21746114.0,
"step": 9740
},
{
"entropy": 5.521394109725952,
"epoch": 1.094457235918459,
"grad_norm": 2.171875,
"learning_rate": 0.000489133640016346,
"loss": 5.4623,
"mean_token_accuracy": 0.1824353888630867,
"num_tokens": 21757145.0,
"step": 9745
},
{
"entropy": 5.5815990447998045,
"epoch": 1.0950188128264167,
"grad_norm": 1.46875,
"learning_rate": 0.0004891213103970579,
"loss": 5.4467,
"mean_token_accuracy": 0.17666808068752288,
"num_tokens": 21768294.0,
"step": 9750
},
{
"entropy": 5.635635757446289,
"epoch": 1.095580389734374,
"grad_norm": 2.21875,
"learning_rate": 0.0004891089739600643,
"loss": 5.4621,
"mean_token_accuracy": 0.1817360684275627,
"num_tokens": 21779667.0,
"step": 9755
},
{
"entropy": 5.491716384887695,
"epoch": 1.0961419666423318,
"grad_norm": 1.59375,
"learning_rate": 0.0004890966307057581,
"loss": 5.3061,
"mean_token_accuracy": 0.1869865745306015,
"num_tokens": 21789635.0,
"step": 9760
},
{
"entropy": 5.446166086196899,
"epoch": 1.0967035435502892,
"grad_norm": 4.84375,
"learning_rate": 0.000489084280634532,
"loss": 5.2964,
"mean_token_accuracy": 0.19281961023807526,
"num_tokens": 21799462.0,
"step": 9765
},
{
"entropy": 5.469609546661377,
"epoch": 1.0972651204582466,
"grad_norm": 1.421875,
"learning_rate": 0.0004890719237467794,
"loss": 5.3578,
"mean_token_accuracy": 0.186594557762146,
"num_tokens": 21811520.0,
"step": 9770
},
{
"entropy": 5.71375298500061,
"epoch": 1.0978266973662043,
"grad_norm": 1.5,
"learning_rate": 0.0004890595600428937,
"loss": 5.4264,
"mean_token_accuracy": 0.1815893143415451,
"num_tokens": 21821997.0,
"step": 9775
},
{
"entropy": 5.603670978546143,
"epoch": 1.0983882742741617,
"grad_norm": 3.078125,
"learning_rate": 0.0004890471895232684,
"loss": 5.4854,
"mean_token_accuracy": 0.17578408271074294,
"num_tokens": 21833353.0,
"step": 9780
},
{
"entropy": 5.467174339294433,
"epoch": 1.0989498511821194,
"grad_norm": 1.9765625,
"learning_rate": 0.0004890348121882978,
"loss": 5.4127,
"mean_token_accuracy": 0.18557709753513335,
"num_tokens": 21843691.0,
"step": 9785
},
{
"entropy": 5.504107522964477,
"epoch": 1.0995114280900768,
"grad_norm": 1.7578125,
"learning_rate": 0.0004890224280383758,
"loss": 5.3735,
"mean_token_accuracy": 0.1838335409760475,
"num_tokens": 21853771.0,
"step": 9790
},
{
"entropy": 5.450034236907959,
"epoch": 1.1000730049980345,
"grad_norm": 1.8984375,
"learning_rate": 0.0004890100370738966,
"loss": 5.2654,
"mean_token_accuracy": 0.1906389370560646,
"num_tokens": 21863424.0,
"step": 9795
},
{
"entropy": 5.546683502197266,
"epoch": 1.100634581905992,
"grad_norm": 1.515625,
"learning_rate": 0.0004889976392952548,
"loss": 5.3924,
"mean_token_accuracy": 0.184905107319355,
"num_tokens": 21875407.0,
"step": 9800
},
{
"entropy": 5.5834568500518795,
"epoch": 1.1011961588139496,
"grad_norm": 1.75,
"learning_rate": 0.0004889852347028452,
"loss": 5.4814,
"mean_token_accuracy": 0.1803658202290535,
"num_tokens": 21887081.0,
"step": 9805
},
{
"entropy": 5.516337871551514,
"epoch": 1.101757735721907,
"grad_norm": 1.6796875,
"learning_rate": 0.0004889728232970626,
"loss": 5.3741,
"mean_token_accuracy": 0.18136187195777892,
"num_tokens": 21897858.0,
"step": 9810
},
{
"entropy": 5.532627153396606,
"epoch": 1.1023193126298647,
"grad_norm": 1.546875,
"learning_rate": 0.0004889604050783025,
"loss": 5.4023,
"mean_token_accuracy": 0.18114989697933198,
"num_tokens": 21907985.0,
"step": 9815
},
{
"entropy": 5.544833374023438,
"epoch": 1.1028808895378222,
"grad_norm": 1.4609375,
"learning_rate": 0.00048894798004696,
"loss": 5.4484,
"mean_token_accuracy": 0.1773737668991089,
"num_tokens": 21918726.0,
"step": 9820
},
{
"entropy": 5.545290565490722,
"epoch": 1.1034424664457798,
"grad_norm": 1.484375,
"learning_rate": 0.0004889355482034309,
"loss": 5.3736,
"mean_token_accuracy": 0.17921681851148605,
"num_tokens": 21929027.0,
"step": 9825
},
{
"entropy": 5.522619104385376,
"epoch": 1.1040040433537373,
"grad_norm": 1.5390625,
"learning_rate": 0.0004889231095481111,
"loss": 5.286,
"mean_token_accuracy": 0.194375142455101,
"num_tokens": 21938356.0,
"step": 9830
},
{
"entropy": 5.50746865272522,
"epoch": 1.104565620261695,
"grad_norm": 1.734375,
"learning_rate": 0.0004889106640813964,
"loss": 5.345,
"mean_token_accuracy": 0.18472387194633483,
"num_tokens": 21947889.0,
"step": 9835
},
{
"entropy": 5.5559900283813475,
"epoch": 1.1051271971696524,
"grad_norm": 1.5078125,
"learning_rate": 0.0004888982118036831,
"loss": 5.4323,
"mean_token_accuracy": 0.18415676951408386,
"num_tokens": 21958409.0,
"step": 9840
},
{
"entropy": 5.516073417663574,
"epoch": 1.10568877407761,
"grad_norm": 1.9375,
"learning_rate": 0.0004888857527153679,
"loss": 5.4153,
"mean_token_accuracy": 0.1805197849869728,
"num_tokens": 21970120.0,
"step": 9845
},
{
"entropy": 5.548484039306641,
"epoch": 1.1062503509855675,
"grad_norm": 1.6796875,
"learning_rate": 0.0004888732868168473,
"loss": 5.4058,
"mean_token_accuracy": 0.1855633169412613,
"num_tokens": 21981499.0,
"step": 9850
},
{
"entropy": 5.4856383323669435,
"epoch": 1.1068119278935251,
"grad_norm": 2.671875,
"learning_rate": 0.0004888608141085183,
"loss": 5.2943,
"mean_token_accuracy": 0.1931164503097534,
"num_tokens": 21991263.0,
"step": 9855
},
{
"entropy": 5.502034711837768,
"epoch": 1.1073735048014826,
"grad_norm": 1.4921875,
"learning_rate": 0.0004888483345907781,
"loss": 5.2812,
"mean_token_accuracy": 0.18006295412778855,
"num_tokens": 22001582.0,
"step": 9860
},
{
"entropy": 5.533217477798462,
"epoch": 1.10793508170944,
"grad_norm": 2.0,
"learning_rate": 0.000488835848264024,
"loss": 5.3762,
"mean_token_accuracy": 0.18841667622327804,
"num_tokens": 22012447.0,
"step": 9865
},
{
"entropy": 5.56195034980774,
"epoch": 1.1084966586173977,
"grad_norm": 3.109375,
"learning_rate": 0.0004888233551286534,
"loss": 5.4659,
"mean_token_accuracy": 0.18337970823049546,
"num_tokens": 22024865.0,
"step": 9870
},
{
"entropy": 5.522898006439209,
"epoch": 1.109058235525355,
"grad_norm": 1.7421875,
"learning_rate": 0.0004888108551850644,
"loss": 5.379,
"mean_token_accuracy": 0.18430568873882294,
"num_tokens": 22036298.0,
"step": 9875
},
{
"entropy": 5.606694984436035,
"epoch": 1.1096198124333128,
"grad_norm": 1.71875,
"learning_rate": 0.0004887983484336548,
"loss": 5.3814,
"mean_token_accuracy": 0.1878729984164238,
"num_tokens": 22047499.0,
"step": 9880
},
{
"entropy": 5.555080795288086,
"epoch": 1.1101813893412702,
"grad_norm": 1.5703125,
"learning_rate": 0.0004887858348748229,
"loss": 5.4249,
"mean_token_accuracy": 0.1805736765265465,
"num_tokens": 22057612.0,
"step": 9885
},
{
"entropy": 5.5653626918792725,
"epoch": 1.1107429662492279,
"grad_norm": 1.828125,
"learning_rate": 0.000488773314508967,
"loss": 5.4504,
"mean_token_accuracy": 0.18037348687648774,
"num_tokens": 22068566.0,
"step": 9890
},
{
"entropy": 5.5397467613220215,
"epoch": 1.1113045431571853,
"grad_norm": 1.7734375,
"learning_rate": 0.0004887607873364859,
"loss": 5.3317,
"mean_token_accuracy": 0.1811254158616066,
"num_tokens": 22079821.0,
"step": 9895
},
{
"entropy": 5.541785717010498,
"epoch": 1.111866120065143,
"grad_norm": 1.6015625,
"learning_rate": 0.0004887482533577786,
"loss": 5.3924,
"mean_token_accuracy": 0.18046450912952422,
"num_tokens": 22090143.0,
"step": 9900
},
{
"entropy": 5.471843194961548,
"epoch": 1.1124276969731004,
"grad_norm": 2.109375,
"learning_rate": 0.0004887357125732438,
"loss": 5.435,
"mean_token_accuracy": 0.1844059243798256,
"num_tokens": 22101601.0,
"step": 9905
},
{
"entropy": 5.537988519668579,
"epoch": 1.112989273881058,
"grad_norm": 2.375,
"learning_rate": 0.000488723164983281,
"loss": 5.3426,
"mean_token_accuracy": 0.19047263860702515,
"num_tokens": 22113102.0,
"step": 9910
},
{
"entropy": 5.642801189422608,
"epoch": 1.1135508507890155,
"grad_norm": 1.4296875,
"learning_rate": 0.0004887106105882898,
"loss": 5.5048,
"mean_token_accuracy": 0.17974539697170258,
"num_tokens": 22124650.0,
"step": 9915
},
{
"entropy": 5.567287397384644,
"epoch": 1.1141124276969732,
"grad_norm": 1.5625,
"learning_rate": 0.0004886980493886699,
"loss": 5.4383,
"mean_token_accuracy": 0.18499141782522202,
"num_tokens": 22136637.0,
"step": 9920
},
{
"entropy": 5.544352722167969,
"epoch": 1.1146740046049306,
"grad_norm": 2.3125,
"learning_rate": 0.0004886854813848212,
"loss": 5.3459,
"mean_token_accuracy": 0.18668277859687804,
"num_tokens": 22149318.0,
"step": 9925
},
{
"entropy": 5.486715412139892,
"epoch": 1.1152355815128883,
"grad_norm": 1.5625,
"learning_rate": 0.0004886729065771439,
"loss": 5.237,
"mean_token_accuracy": 0.18983374834060668,
"num_tokens": 22159410.0,
"step": 9930
},
{
"entropy": 5.475853776931762,
"epoch": 1.1157971584208457,
"grad_norm": 1.3359375,
"learning_rate": 0.0004886603249660384,
"loss": 5.3269,
"mean_token_accuracy": 0.18352052867412566,
"num_tokens": 22170017.0,
"step": 9935
},
{
"entropy": 5.4734269142150875,
"epoch": 1.1163587353288034,
"grad_norm": 1.625,
"learning_rate": 0.0004886477365519052,
"loss": 5.3446,
"mean_token_accuracy": 0.18532877564430236,
"num_tokens": 22179961.0,
"step": 9940
},
{
"entropy": 5.600820922851563,
"epoch": 1.1169203122367608,
"grad_norm": 2.046875,
"learning_rate": 0.0004886351413351453,
"loss": 5.4747,
"mean_token_accuracy": 0.17250280678272248,
"num_tokens": 22191415.0,
"step": 9945
},
{
"entropy": 5.640557813644409,
"epoch": 1.1174818891447185,
"grad_norm": 1.921875,
"learning_rate": 0.0004886225393161596,
"loss": 5.4397,
"mean_token_accuracy": 0.19490916877985,
"num_tokens": 22205354.0,
"step": 9950
},
{
"entropy": 5.683057022094727,
"epoch": 1.118043466052676,
"grad_norm": 2.203125,
"learning_rate": 0.0004886099304953495,
"loss": 5.4675,
"mean_token_accuracy": 0.1827710524201393,
"num_tokens": 22216989.0,
"step": 9955
},
{
"entropy": 5.61019287109375,
"epoch": 1.1186050429606333,
"grad_norm": 2.078125,
"learning_rate": 0.0004885973148731162,
"loss": 5.4679,
"mean_token_accuracy": 0.17870722562074662,
"num_tokens": 22228421.0,
"step": 9960
},
{
"entropy": 5.540753507614136,
"epoch": 1.119166619868591,
"grad_norm": 1.828125,
"learning_rate": 0.0004885846924498617,
"loss": 5.3349,
"mean_token_accuracy": 0.18364361822605133,
"num_tokens": 22240621.0,
"step": 9965
},
{
"entropy": 5.533217191696167,
"epoch": 1.1197281967765484,
"grad_norm": 1.8359375,
"learning_rate": 0.0004885720632259876,
"loss": 5.3275,
"mean_token_accuracy": 0.18726191073656082,
"num_tokens": 22250936.0,
"step": 9970
},
{
"entropy": 5.530289983749389,
"epoch": 1.120289773684506,
"grad_norm": 2.0625,
"learning_rate": 0.0004885594272018963,
"loss": 5.4357,
"mean_token_accuracy": 0.18193468153476716,
"num_tokens": 22261766.0,
"step": 9975
},
{
"entropy": 5.512259244918823,
"epoch": 1.1208513505924635,
"grad_norm": 1.9453125,
"learning_rate": 0.00048854678437799,
"loss": 5.3763,
"mean_token_accuracy": 0.1885952264070511,
"num_tokens": 22271831.0,
"step": 9980
},
{
"entropy": 5.471192646026611,
"epoch": 1.1214129275004212,
"grad_norm": 1.703125,
"learning_rate": 0.0004885341347546713,
"loss": 5.4048,
"mean_token_accuracy": 0.18108141124248506,
"num_tokens": 22283756.0,
"step": 9985
},
{
"entropy": 5.457755899429321,
"epoch": 1.1219745044083786,
"grad_norm": 1.71875,
"learning_rate": 0.0004885214783323428,
"loss": 5.3537,
"mean_token_accuracy": 0.1867587924003601,
"num_tokens": 22294576.0,
"step": 9990
},
{
"entropy": 5.53228349685669,
"epoch": 1.1225360813163363,
"grad_norm": 2.3125,
"learning_rate": 0.0004885088151114076,
"loss": 5.4136,
"mean_token_accuracy": 0.18535175323486328,
"num_tokens": 22305003.0,
"step": 9995
},
{
"entropy": 5.517486572265625,
"epoch": 1.1230976582242937,
"grad_norm": 1.7734375,
"learning_rate": 0.0004884961450922691,
"loss": 5.3841,
"mean_token_accuracy": 0.18587131351232528,
"num_tokens": 22316505.0,
"step": 10000
},
{
"entropy": 5.5659466743469235,
"epoch": 1.1236592351322514,
"grad_norm": 1.40625,
"learning_rate": 0.0004884834682753305,
"loss": 5.4279,
"mean_token_accuracy": 0.18523892611265183,
"num_tokens": 22327853.0,
"step": 10005
},
{
"entropy": 5.587063503265381,
"epoch": 1.1242208120402089,
"grad_norm": 1.6328125,
"learning_rate": 0.0004884707846609953,
"loss": 5.3257,
"mean_token_accuracy": 0.1895351827144623,
"num_tokens": 22337386.0,
"step": 10010
},
{
"entropy": 5.557631874084473,
"epoch": 1.1247823889481665,
"grad_norm": 1.9140625,
"learning_rate": 0.0004884580942496677,
"loss": 5.5339,
"mean_token_accuracy": 0.17486243098974227,
"num_tokens": 22348742.0,
"step": 10015
},
{
"entropy": 5.66574993133545,
"epoch": 1.125343965856124,
"grad_norm": 1.5546875,
"learning_rate": 0.0004884453970417515,
"loss": 5.4563,
"mean_token_accuracy": 0.18302006721496583,
"num_tokens": 22360083.0,
"step": 10020
},
{
"entropy": 5.593824005126953,
"epoch": 1.1259055427640816,
"grad_norm": 1.4453125,
"learning_rate": 0.000488432693037651,
"loss": 5.4059,
"mean_token_accuracy": 0.17249902039766313,
"num_tokens": 22370982.0,
"step": 10025
},
{
"entropy": 5.49117431640625,
"epoch": 1.126467119672039,
"grad_norm": 1.53125,
"learning_rate": 0.000488419982237771,
"loss": 5.3673,
"mean_token_accuracy": 0.1939145654439926,
"num_tokens": 22382383.0,
"step": 10030
},
{
"entropy": 5.546765899658203,
"epoch": 1.1270286965799967,
"grad_norm": 1.5234375,
"learning_rate": 0.0004884072646425157,
"loss": 5.3851,
"mean_token_accuracy": 0.17847073674201966,
"num_tokens": 22393994.0,
"step": 10035
},
{
"entropy": 5.4657097339630125,
"epoch": 1.1275902734879542,
"grad_norm": 1.6796875,
"learning_rate": 0.0004883945402522907,
"loss": 5.2868,
"mean_token_accuracy": 0.19209165275096893,
"num_tokens": 22405354.0,
"step": 10040
},
{
"entropy": 5.461618375778198,
"epoch": 1.1281518503959118,
"grad_norm": 1.8515625,
"learning_rate": 0.0004883818090675005,
"loss": 5.3483,
"mean_token_accuracy": 0.18297628313302994,
"num_tokens": 22416708.0,
"step": 10045
},
{
"entropy": 5.5068932056427,
"epoch": 1.1287134273038693,
"grad_norm": 1.4921875,
"learning_rate": 0.0004883690710885507,
"loss": 5.3395,
"mean_token_accuracy": 0.1844644859433174,
"num_tokens": 22427724.0,
"step": 10050
},
{
"entropy": 5.6016395568847654,
"epoch": 1.1292750042118267,
"grad_norm": 1.4765625,
"learning_rate": 0.0004883563263158471,
"loss": 5.4281,
"mean_token_accuracy": 0.18196694999933244,
"num_tokens": 22437823.0,
"step": 10055
},
{
"entropy": 5.559715366363525,
"epoch": 1.1298365811197844,
"grad_norm": 1.34375,
"learning_rate": 0.0004883435747497952,
"loss": 5.4634,
"mean_token_accuracy": 0.18190222829580308,
"num_tokens": 22450155.0,
"step": 10060
},
{
"entropy": 5.5579976558685305,
"epoch": 1.130398158027742,
"grad_norm": 1.5390625,
"learning_rate": 0.0004883308163908012,
"loss": 5.3721,
"mean_token_accuracy": 0.18461428582668304,
"num_tokens": 22460484.0,
"step": 10065
},
{
"entropy": 5.473817014694214,
"epoch": 1.1309597349356995,
"grad_norm": 1.5078125,
"learning_rate": 0.0004883180512392714,
"loss": 5.4124,
"mean_token_accuracy": 0.18000868260860442,
"num_tokens": 22472063.0,
"step": 10070
},
{
"entropy": 5.49380784034729,
"epoch": 1.131521311843657,
"grad_norm": 1.453125,
"learning_rate": 0.0004883052792956119,
"loss": 5.3984,
"mean_token_accuracy": 0.18874550759792327,
"num_tokens": 22483669.0,
"step": 10075
},
{
"entropy": 5.638711643218994,
"epoch": 1.1320828887516146,
"grad_norm": 1.625,
"learning_rate": 0.0004882925005602296,
"loss": 5.4623,
"mean_token_accuracy": 0.18453362882137297,
"num_tokens": 22495505.0,
"step": 10080
},
{
"entropy": 5.526845788955688,
"epoch": 1.132644465659572,
"grad_norm": 1.6640625,
"learning_rate": 0.0004882797150335314,
"loss": 5.41,
"mean_token_accuracy": 0.18623802810907364,
"num_tokens": 22505260.0,
"step": 10085
},
{
"entropy": 5.525846385955811,
"epoch": 1.1332060425675297,
"grad_norm": 1.7578125,
"learning_rate": 0.00048826692271592434,
"loss": 5.4727,
"mean_token_accuracy": 0.18053144514560698,
"num_tokens": 22516362.0,
"step": 10090
},
{
"entropy": 5.612450122833252,
"epoch": 1.133767619475487,
"grad_norm": 1.53125,
"learning_rate": 0.0004882541236078158,
"loss": 5.4307,
"mean_token_accuracy": 0.17897112667560577,
"num_tokens": 22527760.0,
"step": 10095
},
{
"entropy": 5.753468751907349,
"epoch": 1.1343291963834448,
"grad_norm": 1.546875,
"learning_rate": 0.00048824131770961316,
"loss": 5.5351,
"mean_token_accuracy": 0.1731613278388977,
"num_tokens": 22540601.0,
"step": 10100
},
{
"entropy": 5.555566501617432,
"epoch": 1.1348907732914022,
"grad_norm": 1.6484375,
"learning_rate": 0.00048822850502172437,
"loss": 5.4363,
"mean_token_accuracy": 0.1790024608373642,
"num_tokens": 22552469.0,
"step": 10105
},
{
"entropy": 5.479272031784058,
"epoch": 1.1354523501993599,
"grad_norm": 1.8125,
"learning_rate": 0.0004882156855445571,
"loss": 5.3767,
"mean_token_accuracy": 0.18386278450489044,
"num_tokens": 22563552.0,
"step": 10110
},
{
"entropy": 5.55500750541687,
"epoch": 1.1360139271073173,
"grad_norm": 1.65625,
"learning_rate": 0.0004882028592785199,
"loss": 5.3535,
"mean_token_accuracy": 0.1852160021662712,
"num_tokens": 22575619.0,
"step": 10115
},
{
"entropy": 5.504810905456543,
"epoch": 1.136575504015275,
"grad_norm": 1.5,
"learning_rate": 0.00048819002622402095,
"loss": 5.3291,
"mean_token_accuracy": 0.18821215778589248,
"num_tokens": 22585854.0,
"step": 10120
},
{
"entropy": 5.5628915309906,
"epoch": 1.1371370809232324,
"grad_norm": 1.7578125,
"learning_rate": 0.0004881771863814688,
"loss": 5.3538,
"mean_token_accuracy": 0.18873200714588165,
"num_tokens": 22595899.0,
"step": 10125
},
{
"entropy": 5.531556415557861,
"epoch": 1.13769865783119,
"grad_norm": 1.5390625,
"learning_rate": 0.00048816433975127244,
"loss": 5.342,
"mean_token_accuracy": 0.18686713576316832,
"num_tokens": 22607517.0,
"step": 10130
},
{
"entropy": 5.560656261444092,
"epoch": 1.1382602347391475,
"grad_norm": 1.734375,
"learning_rate": 0.0004881514863338407,
"loss": 5.4519,
"mean_token_accuracy": 0.18017259687185289,
"num_tokens": 22618531.0,
"step": 10135
},
{
"entropy": 5.434736108779907,
"epoch": 1.1388218116471052,
"grad_norm": 1.765625,
"learning_rate": 0.00048813862612958307,
"loss": 5.291,
"mean_token_accuracy": 0.18618203103542327,
"num_tokens": 22629193.0,
"step": 10140
},
{
"entropy": 5.588760328292847,
"epoch": 1.1393833885550626,
"grad_norm": 1.5,
"learning_rate": 0.0004881257591389089,
"loss": 5.4145,
"mean_token_accuracy": 0.1879473477602005,
"num_tokens": 22639794.0,
"step": 10145
},
{
"entropy": 5.57397780418396,
"epoch": 1.13994496546302,
"grad_norm": 1.625,
"learning_rate": 0.00048811288536222794,
"loss": 5.3496,
"mean_token_accuracy": 0.19252934455871581,
"num_tokens": 22650061.0,
"step": 10150
},
{
"entropy": 5.516801214218139,
"epoch": 1.1405065423709777,
"grad_norm": 1.9375,
"learning_rate": 0.00048810000479995,
"loss": 5.4531,
"mean_token_accuracy": 0.1800699770450592,
"num_tokens": 22661399.0,
"step": 10155
},
{
"entropy": 5.6263528823852536,
"epoch": 1.1410681192789354,
"grad_norm": 1.59375,
"learning_rate": 0.00048808711745248524,
"loss": 5.5274,
"mean_token_accuracy": 0.17378983944654464,
"num_tokens": 22672445.0,
"step": 10160
},
{
"entropy": 5.571680116653442,
"epoch": 1.1416296961868928,
"grad_norm": 1.875,
"learning_rate": 0.00048807422332024404,
"loss": 5.3516,
"mean_token_accuracy": 0.18933580219745635,
"num_tokens": 22682545.0,
"step": 10165
},
{
"entropy": 5.568514394760132,
"epoch": 1.1421912730948502,
"grad_norm": 1.609375,
"learning_rate": 0.0004880613224036369,
"loss": 5.4084,
"mean_token_accuracy": 0.18127067238092423,
"num_tokens": 22693885.0,
"step": 10170
},
{
"entropy": 5.504605484008789,
"epoch": 1.142752850002808,
"grad_norm": 2.25,
"learning_rate": 0.00048804841470307466,
"loss": 5.3398,
"mean_token_accuracy": 0.1840388536453247,
"num_tokens": 22703841.0,
"step": 10175
},
{
"entropy": 5.564083290100098,
"epoch": 1.1433144269107653,
"grad_norm": 1.5859375,
"learning_rate": 0.00048803550021896824,
"loss": 5.4906,
"mean_token_accuracy": 0.177508445084095,
"num_tokens": 22714142.0,
"step": 10180
},
{
"entropy": 5.54605655670166,
"epoch": 1.143876003818723,
"grad_norm": 1.6484375,
"learning_rate": 0.0004880225789517289,
"loss": 5.3677,
"mean_token_accuracy": 0.17855244129896164,
"num_tokens": 22725056.0,
"step": 10185
},
{
"entropy": 5.49986515045166,
"epoch": 1.1444375807266804,
"grad_norm": 2.09375,
"learning_rate": 0.000488009650901768,
"loss": 5.4081,
"mean_token_accuracy": 0.19012307822704316,
"num_tokens": 22735664.0,
"step": 10190
},
{
"entropy": 5.529146480560303,
"epoch": 1.144999157634638,
"grad_norm": 1.75,
"learning_rate": 0.00048799671606949724,
"loss": 5.4006,
"mean_token_accuracy": 0.1848233625292778,
"num_tokens": 22748071.0,
"step": 10195
},
{
"entropy": 5.642807197570801,
"epoch": 1.1455607345425955,
"grad_norm": 1.5234375,
"learning_rate": 0.00048798377445532837,
"loss": 5.4244,
"mean_token_accuracy": 0.1818073496222496,
"num_tokens": 22759477.0,
"step": 10200
},
{
"entropy": 5.474043178558349,
"epoch": 1.1461223114505532,
"grad_norm": 1.7734375,
"learning_rate": 0.0004879708260596736,
"loss": 5.3003,
"mean_token_accuracy": 0.19191243052482604,
"num_tokens": 22769955.0,
"step": 10205
},
{
"entropy": 5.468821096420288,
"epoch": 1.1466838883585107,
"grad_norm": 1.6953125,
"learning_rate": 0.0004879578708829451,
"loss": 5.3958,
"mean_token_accuracy": 0.17963045537471772,
"num_tokens": 22781216.0,
"step": 10210
},
{
"entropy": 5.5151043891906735,
"epoch": 1.1472454652664683,
"grad_norm": 1.3828125,
"learning_rate": 0.0004879449089255553,
"loss": 5.3411,
"mean_token_accuracy": 0.1846957817673683,
"num_tokens": 22792581.0,
"step": 10215
},
{
"entropy": 5.50584044456482,
"epoch": 1.1478070421744258,
"grad_norm": 2.03125,
"learning_rate": 0.0004879319401879171,
"loss": 5.2997,
"mean_token_accuracy": 0.18768150210380555,
"num_tokens": 22803305.0,
"step": 10220
},
{
"entropy": 5.559268569946289,
"epoch": 1.1483686190823834,
"grad_norm": 1.4140625,
"learning_rate": 0.00048791896467044335,
"loss": 5.44,
"mean_token_accuracy": 0.1822998359799385,
"num_tokens": 22815558.0,
"step": 10225
},
{
"entropy": 5.511287355422974,
"epoch": 1.1489301959903409,
"grad_norm": 2.046875,
"learning_rate": 0.00048790598237354716,
"loss": 5.3411,
"mean_token_accuracy": 0.18940191119909286,
"num_tokens": 22825428.0,
"step": 10230
},
{
"entropy": 5.58812575340271,
"epoch": 1.1494917728982985,
"grad_norm": 1.3359375,
"learning_rate": 0.0004878929932976419,
"loss": 5.4051,
"mean_token_accuracy": 0.18259328454732895,
"num_tokens": 22836294.0,
"step": 10235
},
{
"entropy": 5.573909568786621,
"epoch": 1.150053349806256,
"grad_norm": 1.765625,
"learning_rate": 0.0004878799974431412,
"loss": 5.4918,
"mean_token_accuracy": 0.1832794114947319,
"num_tokens": 22847835.0,
"step": 10240
},
{
"entropy": 5.5758397579193115,
"epoch": 1.1506149267142134,
"grad_norm": 1.8125,
"learning_rate": 0.0004878669948104588,
"loss": 5.4278,
"mean_token_accuracy": 0.17983417361974716,
"num_tokens": 22859726.0,
"step": 10245
},
{
"entropy": 5.519074010848999,
"epoch": 1.151176503622171,
"grad_norm": 2.25,
"learning_rate": 0.00048785398540000867,
"loss": 5.3759,
"mean_token_accuracy": 0.1819480836391449,
"num_tokens": 22871250.0,
"step": 10250
},
{
"entropy": 5.583474159240723,
"epoch": 1.1517380805301287,
"grad_norm": 1.9921875,
"learning_rate": 0.00048784096921220514,
"loss": 5.4386,
"mean_token_accuracy": 0.17781830579042435,
"num_tokens": 22882034.0,
"step": 10255
},
{
"entropy": 5.651230001449585,
"epoch": 1.1522996574380862,
"grad_norm": 1.3515625,
"learning_rate": 0.0004878279462474627,
"loss": 5.5374,
"mean_token_accuracy": 0.1767759457230568,
"num_tokens": 22894606.0,
"step": 10260
},
{
"entropy": 5.51959376335144,
"epoch": 1.1528612343460436,
"grad_norm": 1.5625,
"learning_rate": 0.00048781491650619574,
"loss": 5.3617,
"mean_token_accuracy": 0.1838405191898346,
"num_tokens": 22905246.0,
"step": 10265
},
{
"entropy": 5.52998046875,
"epoch": 1.1534228112540013,
"grad_norm": 1.4453125,
"learning_rate": 0.00048780187998881944,
"loss": 5.3146,
"mean_token_accuracy": 0.18105529844760895,
"num_tokens": 22916202.0,
"step": 10270
},
{
"entropy": 5.5451719760894775,
"epoch": 1.1539843881619587,
"grad_norm": 1.9453125,
"learning_rate": 0.0004877888366957486,
"loss": 5.292,
"mean_token_accuracy": 0.1853562369942665,
"num_tokens": 22926451.0,
"step": 10275
},
{
"entropy": 5.546468830108642,
"epoch": 1.1545459650699164,
"grad_norm": 1.390625,
"learning_rate": 0.0004877757866273988,
"loss": 5.3888,
"mean_token_accuracy": 0.18690523654222488,
"num_tokens": 22936618.0,
"step": 10280
},
{
"entropy": 5.582164907455445,
"epoch": 1.1551075419778738,
"grad_norm": 1.71875,
"learning_rate": 0.0004877627297841853,
"loss": 5.4255,
"mean_token_accuracy": 0.17415512651205062,
"num_tokens": 22948278.0,
"step": 10285
},
{
"entropy": 5.508408832550049,
"epoch": 1.1556691188858315,
"grad_norm": 1.375,
"learning_rate": 0.0004877496661665241,
"loss": 5.3349,
"mean_token_accuracy": 0.18515396118164062,
"num_tokens": 22959928.0,
"step": 10290
},
{
"entropy": 5.487191104888916,
"epoch": 1.156230695793789,
"grad_norm": 1.3984375,
"learning_rate": 0.000487736595774831,
"loss": 5.3879,
"mean_token_accuracy": 0.18180178105831146,
"num_tokens": 22971050.0,
"step": 10295
},
{
"entropy": 5.4649248123168945,
"epoch": 1.1567922727017466,
"grad_norm": 1.7734375,
"learning_rate": 0.00048772351860952215,
"loss": 5.2915,
"mean_token_accuracy": 0.18241112381219865,
"num_tokens": 22981471.0,
"step": 10300
},
{
"entropy": 5.625771188735962,
"epoch": 1.157353849609704,
"grad_norm": 1.421875,
"learning_rate": 0.0004877104346710139,
"loss": 5.5228,
"mean_token_accuracy": 0.1869986429810524,
"num_tokens": 22992526.0,
"step": 10305
},
{
"entropy": 5.554246520996093,
"epoch": 1.1579154265176617,
"grad_norm": 1.1953125,
"learning_rate": 0.000487697343959723,
"loss": 5.3941,
"mean_token_accuracy": 0.1876555636525154,
"num_tokens": 23003408.0,
"step": 10310
},
{
"entropy": 5.525618267059326,
"epoch": 1.158477003425619,
"grad_norm": 1.484375,
"learning_rate": 0.0004876842464760661,
"loss": 5.2942,
"mean_token_accuracy": 0.1880863904953003,
"num_tokens": 23013406.0,
"step": 10315
},
{
"entropy": 5.48646183013916,
"epoch": 1.1590385803335768,
"grad_norm": 1.609375,
"learning_rate": 0.00048767114222046036,
"loss": 5.3822,
"mean_token_accuracy": 0.18457602560520173,
"num_tokens": 23024960.0,
"step": 10320
},
{
"entropy": 5.643769216537476,
"epoch": 1.1596001572415342,
"grad_norm": 1.703125,
"learning_rate": 0.000487658031193323,
"loss": 5.4033,
"mean_token_accuracy": 0.17984264940023423,
"num_tokens": 23037514.0,
"step": 10325
},
{
"entropy": 5.640411520004273,
"epoch": 1.1601617341494919,
"grad_norm": 1.4453125,
"learning_rate": 0.0004876449133950714,
"loss": 5.5009,
"mean_token_accuracy": 0.1778225228190422,
"num_tokens": 23048460.0,
"step": 10330
},
{
"entropy": 5.5605621337890625,
"epoch": 1.1607233110574493,
"grad_norm": 1.4765625,
"learning_rate": 0.0004876317888261233,
"loss": 5.3477,
"mean_token_accuracy": 0.18494987934827806,
"num_tokens": 23058866.0,
"step": 10335
},
{
"entropy": 5.472915935516357,
"epoch": 1.1612848879654067,
"grad_norm": 1.421875,
"learning_rate": 0.00048761865748689657,
"loss": 5.3886,
"mean_token_accuracy": 0.17650222778320312,
"num_tokens": 23069097.0,
"step": 10340
},
{
"entropy": 5.625676202774048,
"epoch": 1.1618464648733644,
"grad_norm": 1.3125,
"learning_rate": 0.00048760551937780927,
"loss": 5.3773,
"mean_token_accuracy": 0.17985702753067018,
"num_tokens": 23079553.0,
"step": 10345
},
{
"entropy": 5.523149013519287,
"epoch": 1.162408041781322,
"grad_norm": 2.140625,
"learning_rate": 0.00048759237449927986,
"loss": 5.3802,
"mean_token_accuracy": 0.17997240871191025,
"num_tokens": 23090554.0,
"step": 10350
},
{
"entropy": 5.453018045425415,
"epoch": 1.1629696186892795,
"grad_norm": 1.4296875,
"learning_rate": 0.00048757922285172674,
"loss": 5.2945,
"mean_token_accuracy": 0.1890457645058632,
"num_tokens": 23101740.0,
"step": 10355
},
{
"entropy": 5.594650745391846,
"epoch": 1.163531195597237,
"grad_norm": 1.453125,
"learning_rate": 0.00048756606443556874,
"loss": 5.3827,
"mean_token_accuracy": 0.18782007098197936,
"num_tokens": 23112842.0,
"step": 10360
},
{
"entropy": 5.485249614715576,
"epoch": 1.1640927725051946,
"grad_norm": 1.21875,
"learning_rate": 0.00048755289925122474,
"loss": 5.3785,
"mean_token_accuracy": 0.18219865560531617,
"num_tokens": 23123822.0,
"step": 10365
},
{
"entropy": 5.564452314376831,
"epoch": 1.164654349413152,
"grad_norm": 2.140625,
"learning_rate": 0.0004875397272991141,
"loss": 5.3952,
"mean_token_accuracy": 0.17662669867277145,
"num_tokens": 23134045.0,
"step": 10370
},
{
"entropy": 5.515670347213745,
"epoch": 1.1652159263211097,
"grad_norm": 1.609375,
"learning_rate": 0.000487526548579656,
"loss": 5.4265,
"mean_token_accuracy": 0.17346135824918746,
"num_tokens": 23144200.0,
"step": 10375
},
{
"entropy": 5.5541013240814205,
"epoch": 1.1657775032290671,
"grad_norm": 1.375,
"learning_rate": 0.00048751336309327026,
"loss": 5.3633,
"mean_token_accuracy": 0.18630352914333342,
"num_tokens": 23155671.0,
"step": 10380
},
{
"entropy": 5.6557896614074705,
"epoch": 1.1663390801370248,
"grad_norm": 1.515625,
"learning_rate": 0.0004875001708403766,
"loss": 5.5055,
"mean_token_accuracy": 0.1809536561369896,
"num_tokens": 23166644.0,
"step": 10385
},
{
"entropy": 5.539685440063477,
"epoch": 1.1669006570449822,
"grad_norm": 1.2578125,
"learning_rate": 0.00048748697182139503,
"loss": 5.3663,
"mean_token_accuracy": 0.1850558638572693,
"num_tokens": 23177074.0,
"step": 10390
},
{
"entropy": 5.574979686737061,
"epoch": 1.16746223395294,
"grad_norm": 1.640625,
"learning_rate": 0.0004874737660367459,
"loss": 5.5029,
"mean_token_accuracy": 0.17499616444110871,
"num_tokens": 23188724.0,
"step": 10395
},
{
"entropy": 5.500261211395264,
"epoch": 1.1680238108608973,
"grad_norm": 1.2578125,
"learning_rate": 0.00048746055348684967,
"loss": 5.4077,
"mean_token_accuracy": 0.18461918383836745,
"num_tokens": 23198459.0,
"step": 10400
},
{
"entropy": 5.62602801322937,
"epoch": 1.168585387768855,
"grad_norm": 1.4375,
"learning_rate": 0.000487447334172127,
"loss": 5.4224,
"mean_token_accuracy": 0.17996439784765245,
"num_tokens": 23208991.0,
"step": 10405
},
{
"entropy": 5.506118202209473,
"epoch": 1.1691469646768125,
"grad_norm": 1.40625,
"learning_rate": 0.00048743410809299877,
"loss": 5.3498,
"mean_token_accuracy": 0.18172885626554489,
"num_tokens": 23219309.0,
"step": 10410
},
{
"entropy": 5.504933452606201,
"epoch": 1.1697085415847701,
"grad_norm": 1.5,
"learning_rate": 0.0004874208752498862,
"loss": 5.4649,
"mean_token_accuracy": 0.17699553221464157,
"num_tokens": 23230551.0,
"step": 10415
},
{
"entropy": 5.649904394149781,
"epoch": 1.1702701184927276,
"grad_norm": 1.3828125,
"learning_rate": 0.0004874076356432106,
"loss": 5.5288,
"mean_token_accuracy": 0.17727755159139633,
"num_tokens": 23242799.0,
"step": 10420
},
{
"entropy": 5.687228488922119,
"epoch": 1.1708316954006852,
"grad_norm": 1.5,
"learning_rate": 0.00048739438927339346,
"loss": 5.51,
"mean_token_accuracy": 0.1780794993042946,
"num_tokens": 23255112.0,
"step": 10425
},
{
"entropy": 5.518121862411499,
"epoch": 1.1713932723086427,
"grad_norm": 1.28125,
"learning_rate": 0.00048738113614085656,
"loss": 5.3902,
"mean_token_accuracy": 0.18921437561511995,
"num_tokens": 23264800.0,
"step": 10430
},
{
"entropy": 5.480174922943116,
"epoch": 1.1719548492166,
"grad_norm": 1.3984375,
"learning_rate": 0.000487367876246022,
"loss": 5.3398,
"mean_token_accuracy": 0.18184347897768022,
"num_tokens": 23276672.0,
"step": 10435
},
{
"entropy": 5.547278547286988,
"epoch": 1.1725164261245578,
"grad_norm": 1.3984375,
"learning_rate": 0.0004873546095893118,
"loss": 5.3776,
"mean_token_accuracy": 0.18913681507110597,
"num_tokens": 23288282.0,
"step": 10440
},
{
"entropy": 5.566451549530029,
"epoch": 1.1730780030325154,
"grad_norm": 1.3828125,
"learning_rate": 0.0004873413361711485,
"loss": 5.3972,
"mean_token_accuracy": 0.1814500406384468,
"num_tokens": 23299078.0,
"step": 10445
},
{
"entropy": 5.57575101852417,
"epoch": 1.1736395799404729,
"grad_norm": 1.5078125,
"learning_rate": 0.00048732805599195465,
"loss": 5.3851,
"mean_token_accuracy": 0.18296760469675064,
"num_tokens": 23310036.0,
"step": 10450
},
{
"entropy": 5.537910652160645,
"epoch": 1.1742011568484303,
"grad_norm": 1.40625,
"learning_rate": 0.00048731476905215326,
"loss": 5.3323,
"mean_token_accuracy": 0.1914816156029701,
"num_tokens": 23321029.0,
"step": 10455
},
{
"entropy": 5.535822772979737,
"epoch": 1.174762733756388,
"grad_norm": 1.3515625,
"learning_rate": 0.0004873014753521673,
"loss": 5.433,
"mean_token_accuracy": 0.1838971644639969,
"num_tokens": 23331584.0,
"step": 10460
},
{
"entropy": 5.698354768753052,
"epoch": 1.1753243106643454,
"grad_norm": 1.703125,
"learning_rate": 0.0004872881748924199,
"loss": 5.5246,
"mean_token_accuracy": 0.17070927619934081,
"num_tokens": 23343171.0,
"step": 10465
},
{
"entropy": 5.671799373626709,
"epoch": 1.175885887572303,
"grad_norm": 1.28125,
"learning_rate": 0.0004872748676733347,
"loss": 5.4167,
"mean_token_accuracy": 0.18251493722200393,
"num_tokens": 23353047.0,
"step": 10470
},
{
"entropy": 5.52349157333374,
"epoch": 1.1764474644802605,
"grad_norm": 1.59375,
"learning_rate": 0.00048726155369533547,
"loss": 5.4151,
"mean_token_accuracy": 0.1785216212272644,
"num_tokens": 23364447.0,
"step": 10475
},
{
"entropy": 5.562612390518188,
"epoch": 1.1770090413882182,
"grad_norm": 1.40625,
"learning_rate": 0.000487248232958846,
"loss": 5.4461,
"mean_token_accuracy": 0.17994829565286635,
"num_tokens": 23375616.0,
"step": 10480
},
{
"entropy": 5.636350965499878,
"epoch": 1.1775706182961756,
"grad_norm": 1.359375,
"learning_rate": 0.0004872349054642904,
"loss": 5.5429,
"mean_token_accuracy": 0.17499152421951295,
"num_tokens": 23388176.0,
"step": 10485
},
{
"entropy": 5.631733179092407,
"epoch": 1.1781321952041333,
"grad_norm": 1.671875,
"learning_rate": 0.0004872215712120932,
"loss": 5.4744,
"mean_token_accuracy": 0.18584634810686113,
"num_tokens": 23400563.0,
"step": 10490
},
{
"entropy": 5.568887710571289,
"epoch": 1.1786937721120907,
"grad_norm": 1.3125,
"learning_rate": 0.0004872082302026788,
"loss": 5.4117,
"mean_token_accuracy": 0.17976100891828536,
"num_tokens": 23410740.0,
"step": 10495
},
{
"entropy": 5.53800597190857,
"epoch": 1.1792553490200484,
"grad_norm": 1.5234375,
"learning_rate": 0.0004871948824364721,
"loss": 5.5024,
"mean_token_accuracy": 0.17933253198862076,
"num_tokens": 23423086.0,
"step": 10500
},
{
"entropy": 5.653190851211548,
"epoch": 1.1798169259280058,
"grad_norm": 1.5859375,
"learning_rate": 0.000487181527913898,
"loss": 5.5136,
"mean_token_accuracy": 0.1760350927710533,
"num_tokens": 23435869.0,
"step": 10505
},
{
"entropy": 5.604878997802734,
"epoch": 1.1803785028359635,
"grad_norm": 1.484375,
"learning_rate": 0.00048716816663538177,
"loss": 5.3534,
"mean_token_accuracy": 0.19044994711875915,
"num_tokens": 23446635.0,
"step": 10510
},
{
"entropy": 5.475135326385498,
"epoch": 1.180940079743921,
"grad_norm": 1.7578125,
"learning_rate": 0.0004871547986013488,
"loss": 5.3777,
"mean_token_accuracy": 0.18240721374750138,
"num_tokens": 23458214.0,
"step": 10515
},
{
"entropy": 5.534139919281006,
"epoch": 1.1815016566518786,
"grad_norm": 1.203125,
"learning_rate": 0.0004871414238122248,
"loss": 5.4009,
"mean_token_accuracy": 0.1886911153793335,
"num_tokens": 23469383.0,
"step": 10520
},
{
"entropy": 5.5301612377166744,
"epoch": 1.182063233559836,
"grad_norm": 1.625,
"learning_rate": 0.0004871280422684355,
"loss": 5.373,
"mean_token_accuracy": 0.18497077524662017,
"num_tokens": 23480841.0,
"step": 10525
},
{
"entropy": 5.533479261398315,
"epoch": 1.1826248104677934,
"grad_norm": 1.390625,
"learning_rate": 0.00048711465397040726,
"loss": 5.3236,
"mean_token_accuracy": 0.19169697612524034,
"num_tokens": 23490796.0,
"step": 10530
},
{
"entropy": 5.526725435256958,
"epoch": 1.183186387375751,
"grad_norm": 1.6875,
"learning_rate": 0.00048710125891856604,
"loss": 5.3612,
"mean_token_accuracy": 0.18012424409389496,
"num_tokens": 23501759.0,
"step": 10535
},
{
"entropy": 5.565225553512573,
"epoch": 1.1837479642837088,
"grad_norm": 1.484375,
"learning_rate": 0.00048708785711333845,
"loss": 5.4042,
"mean_token_accuracy": 0.1877269357442856,
"num_tokens": 23512209.0,
"step": 10540
},
{
"entropy": 5.584983777999878,
"epoch": 1.1843095411916662,
"grad_norm": 1.6328125,
"learning_rate": 0.00048707444855515124,
"loss": 5.489,
"mean_token_accuracy": 0.17574080228805541,
"num_tokens": 23524316.0,
"step": 10545
},
{
"entropy": 5.539467144012451,
"epoch": 1.1848711180996236,
"grad_norm": 1.703125,
"learning_rate": 0.0004870610332444314,
"loss": 5.463,
"mean_token_accuracy": 0.1843484714627266,
"num_tokens": 23536199.0,
"step": 10550
},
{
"entropy": 5.499478530883789,
"epoch": 1.1854326950075813,
"grad_norm": 1.28125,
"learning_rate": 0.0004870476111816059,
"loss": 5.4534,
"mean_token_accuracy": 0.18032947033643723,
"num_tokens": 23548034.0,
"step": 10555
},
{
"entropy": 5.595110321044922,
"epoch": 1.1859942719155387,
"grad_norm": 1.90625,
"learning_rate": 0.00048703418236710234,
"loss": 5.4083,
"mean_token_accuracy": 0.18150479197502137,
"num_tokens": 23559280.0,
"step": 10560
},
{
"entropy": 5.575775098800659,
"epoch": 1.1865558488234964,
"grad_norm": 1.296875,
"learning_rate": 0.0004870207468013481,
"loss": 5.3748,
"mean_token_accuracy": 0.18385760635137557,
"num_tokens": 23569163.0,
"step": 10565
},
{
"entropy": 5.538456678390503,
"epoch": 1.1871174257314538,
"grad_norm": 1.5078125,
"learning_rate": 0.0004870073044847711,
"loss": 5.4468,
"mean_token_accuracy": 0.18220896422863006,
"num_tokens": 23580526.0,
"step": 10570
},
{
"entropy": 5.460603857040406,
"epoch": 1.1876790026394115,
"grad_norm": 1.4296875,
"learning_rate": 0.0004869938554177993,
"loss": 5.3237,
"mean_token_accuracy": 0.18026399314403535,
"num_tokens": 23590552.0,
"step": 10575
},
{
"entropy": 5.445132064819336,
"epoch": 1.188240579547369,
"grad_norm": 1.484375,
"learning_rate": 0.00048698039960086084,
"loss": 5.2257,
"mean_token_accuracy": 0.19017021209001542,
"num_tokens": 23601889.0,
"step": 10580
},
{
"entropy": 5.593432188034058,
"epoch": 1.1888021564553266,
"grad_norm": 1.390625,
"learning_rate": 0.0004869669370343843,
"loss": 5.4955,
"mean_token_accuracy": 0.18662466406822203,
"num_tokens": 23613833.0,
"step": 10585
},
{
"entropy": 5.577199935913086,
"epoch": 1.189363733363284,
"grad_norm": 1.28125,
"learning_rate": 0.0004869534677187983,
"loss": 5.4397,
"mean_token_accuracy": 0.1868535280227661,
"num_tokens": 23625404.0,
"step": 10590
},
{
"entropy": 5.4918896675109865,
"epoch": 1.1899253102712417,
"grad_norm": 1.3203125,
"learning_rate": 0.0004869399916545317,
"loss": 5.3212,
"mean_token_accuracy": 0.18868361711502074,
"num_tokens": 23635670.0,
"step": 10595
},
{
"entropy": 5.480903911590576,
"epoch": 1.1904868871791991,
"grad_norm": 1.34375,
"learning_rate": 0.0004869265088420135,
"loss": 5.3242,
"mean_token_accuracy": 0.18785989433526992,
"num_tokens": 23646893.0,
"step": 10600
},
{
"entropy": 5.456339502334595,
"epoch": 1.1910484640871568,
"grad_norm": 1.515625,
"learning_rate": 0.0004869130192816731,
"loss": 5.4368,
"mean_token_accuracy": 0.1841205343604088,
"num_tokens": 23657733.0,
"step": 10605
},
{
"entropy": 5.478459024429322,
"epoch": 1.1916100409951143,
"grad_norm": 1.7109375,
"learning_rate": 0.00048689952297394,
"loss": 5.2801,
"mean_token_accuracy": 0.1917136237025261,
"num_tokens": 23668100.0,
"step": 10610
},
{
"entropy": 5.6521234035491945,
"epoch": 1.192171617903072,
"grad_norm": 1.2890625,
"learning_rate": 0.0004868860199192438,
"loss": 5.4466,
"mean_token_accuracy": 0.1784281462430954,
"num_tokens": 23678295.0,
"step": 10615
},
{
"entropy": 5.496122264862061,
"epoch": 1.1927331948110294,
"grad_norm": 1.453125,
"learning_rate": 0.0004868725101180146,
"loss": 5.2846,
"mean_token_accuracy": 0.18780205696821212,
"num_tokens": 23689310.0,
"step": 10620
},
{
"entropy": 5.449780464172363,
"epoch": 1.1932947717189868,
"grad_norm": 1.4609375,
"learning_rate": 0.0004868589935706826,
"loss": 5.4365,
"mean_token_accuracy": 0.18089919090270995,
"num_tokens": 23699475.0,
"step": 10625
},
{
"entropy": 5.622437334060669,
"epoch": 1.1938563486269445,
"grad_norm": 1.5703125,
"learning_rate": 0.00048684547027767793,
"loss": 5.4304,
"mean_token_accuracy": 0.18654530495405197,
"num_tokens": 23710186.0,
"step": 10630
},
{
"entropy": 5.491124629974365,
"epoch": 1.1944179255349021,
"grad_norm": 1.421875,
"learning_rate": 0.0004868319402394314,
"loss": 5.2445,
"mean_token_accuracy": 0.18947111517190934,
"num_tokens": 23720449.0,
"step": 10635
},
{
"entropy": 5.566216516494751,
"epoch": 1.1949795024428596,
"grad_norm": 1.5390625,
"learning_rate": 0.00048681840345637375,
"loss": 5.5177,
"mean_token_accuracy": 0.17807645350694656,
"num_tokens": 23732295.0,
"step": 10640
},
{
"entropy": 5.508579587936401,
"epoch": 1.195541079350817,
"grad_norm": 1.4296875,
"learning_rate": 0.00048680485992893597,
"loss": 5.4047,
"mean_token_accuracy": 0.17882717549800872,
"num_tokens": 23743616.0,
"step": 10645
},
{
"entropy": 5.593965530395508,
"epoch": 1.1961026562587747,
"grad_norm": 1.6015625,
"learning_rate": 0.00048679130965754926,
"loss": 5.4388,
"mean_token_accuracy": 0.17683009654283524,
"num_tokens": 23755288.0,
"step": 10650
},
{
"entropy": 5.548993110656738,
"epoch": 1.196664233166732,
"grad_norm": 1.4375,
"learning_rate": 0.0004867777526426451,
"loss": 5.3122,
"mean_token_accuracy": 0.18785137087106704,
"num_tokens": 23765916.0,
"step": 10655
},
{
"entropy": 5.436146879196167,
"epoch": 1.1972258100746898,
"grad_norm": 1.7734375,
"learning_rate": 0.00048676418888465525,
"loss": 5.2743,
"mean_token_accuracy": 0.1874302476644516,
"num_tokens": 23778891.0,
"step": 10660
},
{
"entropy": 5.542696475982666,
"epoch": 1.1977873869826472,
"grad_norm": 1.75,
"learning_rate": 0.0004867506183840114,
"loss": 5.4036,
"mean_token_accuracy": 0.18881048113107682,
"num_tokens": 23790402.0,
"step": 10665
},
{
"entropy": 5.529537010192871,
"epoch": 1.1983489638906049,
"grad_norm": 1.7734375,
"learning_rate": 0.0004867370411411459,
"loss": 5.3666,
"mean_token_accuracy": 0.19051992893218994,
"num_tokens": 23801928.0,
"step": 10670
},
{
"entropy": 5.551019525527954,
"epoch": 1.1989105407985623,
"grad_norm": 1.3828125,
"learning_rate": 0.00048672345715649066,
"loss": 5.4378,
"mean_token_accuracy": 0.17207740098237992,
"num_tokens": 23812302.0,
"step": 10675
},
{
"entropy": 5.61078667640686,
"epoch": 1.19947211770652,
"grad_norm": 1.375,
"learning_rate": 0.0004867098664304785,
"loss": 5.4047,
"mean_token_accuracy": 0.19216869473457338,
"num_tokens": 23824033.0,
"step": 10680
},
{
"entropy": 5.506804609298706,
"epoch": 1.2000336946144774,
"grad_norm": 1.296875,
"learning_rate": 0.0004866962689635422,
"loss": 5.3649,
"mean_token_accuracy": 0.1889210745692253,
"num_tokens": 23834907.0,
"step": 10685
},
{
"entropy": 5.551431941986084,
"epoch": 1.200595271522435,
"grad_norm": 1.6015625,
"learning_rate": 0.00048668266475611447,
"loss": 5.5548,
"mean_token_accuracy": 0.18483441919088364,
"num_tokens": 23846418.0,
"step": 10690
},
{
"entropy": 5.686411952972412,
"epoch": 1.2011568484303925,
"grad_norm": 2.28125,
"learning_rate": 0.0004866690538086286,
"loss": 5.476,
"mean_token_accuracy": 0.17966153621673583,
"num_tokens": 23858423.0,
"step": 10695
},
{
"entropy": 5.584620475769043,
"epoch": 1.2017184253383502,
"grad_norm": 1.3203125,
"learning_rate": 0.00048665543612151803,
"loss": 5.3748,
"mean_token_accuracy": 0.18819677382707595,
"num_tokens": 23868340.0,
"step": 10700
},
{
"entropy": 5.465174055099487,
"epoch": 1.2022800022463076,
"grad_norm": 1.5703125,
"learning_rate": 0.00048664181169521625,
"loss": 5.3152,
"mean_token_accuracy": 0.18985991328954696,
"num_tokens": 23878752.0,
"step": 10705
},
{
"entropy": 5.558954381942749,
"epoch": 1.2028415791542653,
"grad_norm": 1.625,
"learning_rate": 0.0004866281805301571,
"loss": 5.364,
"mean_token_accuracy": 0.18771320432424546,
"num_tokens": 23889374.0,
"step": 10710
},
{
"entropy": 5.61924147605896,
"epoch": 1.2034031560622227,
"grad_norm": 1.5859375,
"learning_rate": 0.0004866145426267746,
"loss": 5.4328,
"mean_token_accuracy": 0.1817406177520752,
"num_tokens": 23901651.0,
"step": 10715
},
{
"entropy": 5.467511081695557,
"epoch": 1.2039647329701804,
"grad_norm": 1.3828125,
"learning_rate": 0.000486600897985503,
"loss": 5.349,
"mean_token_accuracy": 0.1878811702132225,
"num_tokens": 23911818.0,
"step": 10720
},
{
"entropy": 5.572879886627197,
"epoch": 1.2045263098781378,
"grad_norm": 1.5,
"learning_rate": 0.00048658724660677674,
"loss": 5.3874,
"mean_token_accuracy": 0.18842100352048874,
"num_tokens": 23923840.0,
"step": 10725
},
{
"entropy": 5.503609943389892,
"epoch": 1.2050878867860955,
"grad_norm": 1.7109375,
"learning_rate": 0.0004865735884910305,
"loss": 5.2929,
"mean_token_accuracy": 0.19194971174001693,
"num_tokens": 23935406.0,
"step": 10730
},
{
"entropy": 5.5736291885375975,
"epoch": 1.205649463694053,
"grad_norm": 1.421875,
"learning_rate": 0.0004865599236386991,
"loss": 5.3712,
"mean_token_accuracy": 0.18695309311151503,
"num_tokens": 23946350.0,
"step": 10735
},
{
"entropy": 5.441629934310913,
"epoch": 1.2062110406020103,
"grad_norm": 1.6640625,
"learning_rate": 0.0004865462520502177,
"loss": 5.3226,
"mean_token_accuracy": 0.18511815518140792,
"num_tokens": 23956315.0,
"step": 10740
},
{
"entropy": 5.474482727050781,
"epoch": 1.206772617509968,
"grad_norm": 1.6171875,
"learning_rate": 0.0004865325737260216,
"loss": 5.2626,
"mean_token_accuracy": 0.18442393392324447,
"num_tokens": 23966649.0,
"step": 10745
},
{
"entropy": 5.612815856933594,
"epoch": 1.2073341944179254,
"grad_norm": 1.578125,
"learning_rate": 0.00048651888866654624,
"loss": 5.5242,
"mean_token_accuracy": 0.17632318437099456,
"num_tokens": 23977674.0,
"step": 10750
},
{
"entropy": 5.5524626731872555,
"epoch": 1.207895771325883,
"grad_norm": 1.734375,
"learning_rate": 0.0004865051968722275,
"loss": 5.3997,
"mean_token_accuracy": 0.18081315606832504,
"num_tokens": 23987985.0,
"step": 10755
},
{
"entropy": 5.556069183349609,
"epoch": 1.2084573482338405,
"grad_norm": 1.5859375,
"learning_rate": 0.00048649149834350126,
"loss": 5.5131,
"mean_token_accuracy": 0.1744318813085556,
"num_tokens": 23998730.0,
"step": 10760
},
{
"entropy": 5.487083148956299,
"epoch": 1.2090189251417982,
"grad_norm": 1.4296875,
"learning_rate": 0.0004864777930808037,
"loss": 5.3319,
"mean_token_accuracy": 0.17671567648649217,
"num_tokens": 24008743.0,
"step": 10765
},
{
"entropy": 5.540325689315796,
"epoch": 1.2095805020497556,
"grad_norm": 1.3046875,
"learning_rate": 0.0004864640810845711,
"loss": 5.3834,
"mean_token_accuracy": 0.18542118966579438,
"num_tokens": 24019805.0,
"step": 10770
},
{
"entropy": 5.601198768615722,
"epoch": 1.2101420789577133,
"grad_norm": 1.59375,
"learning_rate": 0.0004864503623552402,
"loss": 5.3792,
"mean_token_accuracy": 0.1825328588485718,
"num_tokens": 24031437.0,
"step": 10775
},
{
"entropy": 5.610545492172241,
"epoch": 1.2107036558656707,
"grad_norm": 1.78125,
"learning_rate": 0.0004864366368932478,
"loss": 5.5177,
"mean_token_accuracy": 0.17343033999204635,
"num_tokens": 24043922.0,
"step": 10780
},
{
"entropy": 5.526181936264038,
"epoch": 1.2112652327736284,
"grad_norm": 1.3828125,
"learning_rate": 0.00048642290469903083,
"loss": 5.3081,
"mean_token_accuracy": 0.19014033824205398,
"num_tokens": 24054755.0,
"step": 10785
},
{
"entropy": 5.526547384262085,
"epoch": 1.2118268096815858,
"grad_norm": 1.34375,
"learning_rate": 0.00048640916577302664,
"loss": 5.3497,
"mean_token_accuracy": 0.1742018148303032,
"num_tokens": 24067347.0,
"step": 10790
},
{
"entropy": 5.556707811355591,
"epoch": 1.2123883865895435,
"grad_norm": 1.3828125,
"learning_rate": 0.00048639542011567243,
"loss": 5.393,
"mean_token_accuracy": 0.18730001747608185,
"num_tokens": 24078938.0,
"step": 10795
},
{
"entropy": 5.544803810119629,
"epoch": 1.212949963497501,
"grad_norm": 1.1953125,
"learning_rate": 0.0004863816677274062,
"loss": 5.4039,
"mean_token_accuracy": 0.18177982270717621,
"num_tokens": 24090275.0,
"step": 10800
},
{
"entropy": 5.484041452407837,
"epoch": 1.2135115404054586,
"grad_norm": 1.3984375,
"learning_rate": 0.00048636790860866576,
"loss": 5.4311,
"mean_token_accuracy": 0.1850898191332817,
"num_tokens": 24101518.0,
"step": 10805
},
{
"entropy": 5.576614189147949,
"epoch": 1.214073117313416,
"grad_norm": 1.578125,
"learning_rate": 0.0004863541427598889,
"loss": 5.503,
"mean_token_accuracy": 0.17933304756879806,
"num_tokens": 24113320.0,
"step": 10810
},
{
"entropy": 5.526159954071045,
"epoch": 1.2146346942213737,
"grad_norm": 1.3359375,
"learning_rate": 0.0004863403701815144,
"loss": 5.316,
"mean_token_accuracy": 0.1886845663189888,
"num_tokens": 24124842.0,
"step": 10815
},
{
"entropy": 5.565841579437256,
"epoch": 1.2151962711293312,
"grad_norm": 1.6796875,
"learning_rate": 0.0004863265908739805,
"loss": 5.4691,
"mean_token_accuracy": 0.1836393803358078,
"num_tokens": 24136391.0,
"step": 10820
},
{
"entropy": 5.5986581325531,
"epoch": 1.2157578480372888,
"grad_norm": 1.59375,
"learning_rate": 0.0004863128048377259,
"loss": 5.4351,
"mean_token_accuracy": 0.18641193956136703,
"num_tokens": 24147194.0,
"step": 10825
},
{
"entropy": 5.582269477844238,
"epoch": 1.2163194249452463,
"grad_norm": 1.5234375,
"learning_rate": 0.0004862990120731896,
"loss": 5.4982,
"mean_token_accuracy": 0.17690225094556808,
"num_tokens": 24158170.0,
"step": 10830
},
{
"entropy": 5.57523398399353,
"epoch": 1.2168810018532037,
"grad_norm": 1.6171875,
"learning_rate": 0.00048628521258081096,
"loss": 5.3389,
"mean_token_accuracy": 0.1854415312409401,
"num_tokens": 24168671.0,
"step": 10835
},
{
"entropy": 5.528084230422974,
"epoch": 1.2174425787611614,
"grad_norm": 1.6640625,
"learning_rate": 0.00048627140636102906,
"loss": 5.3721,
"mean_token_accuracy": 0.1879003405570984,
"num_tokens": 24179898.0,
"step": 10840
},
{
"entropy": 5.473521327972412,
"epoch": 1.2180041556691188,
"grad_norm": 1.5546875,
"learning_rate": 0.0004862575934142837,
"loss": 5.3632,
"mean_token_accuracy": 0.18821609169244766,
"num_tokens": 24190622.0,
"step": 10845
},
{
"entropy": 5.702591371536255,
"epoch": 1.2185657325770765,
"grad_norm": 1.7109375,
"learning_rate": 0.0004862437737410147,
"loss": 5.5295,
"mean_token_accuracy": 0.17465068995952607,
"num_tokens": 24202526.0,
"step": 10850
},
{
"entropy": 5.511393880844116,
"epoch": 1.219127309485034,
"grad_norm": 1.390625,
"learning_rate": 0.0004862299473416619,
"loss": 5.3578,
"mean_token_accuracy": 0.1836084470152855,
"num_tokens": 24213186.0,
"step": 10855
},
{
"entropy": 5.476003789901734,
"epoch": 1.2196888863929916,
"grad_norm": 1.765625,
"learning_rate": 0.00048621611421666573,
"loss": 5.3763,
"mean_token_accuracy": 0.1823277324438095,
"num_tokens": 24225490.0,
"step": 10860
},
{
"entropy": 5.59591326713562,
"epoch": 1.220250463300949,
"grad_norm": 1.671875,
"learning_rate": 0.0004862022743664665,
"loss": 5.3683,
"mean_token_accuracy": 0.1869743973016739,
"num_tokens": 24236653.0,
"step": 10865
},
{
"entropy": 5.528941822052002,
"epoch": 1.2208120402089067,
"grad_norm": 1.5234375,
"learning_rate": 0.00048618842779150494,
"loss": 5.3245,
"mean_token_accuracy": 0.19164512753486634,
"num_tokens": 24247843.0,
"step": 10870
},
{
"entropy": 5.4394721508026125,
"epoch": 1.221373617116864,
"grad_norm": 1.71875,
"learning_rate": 0.00048617457449222205,
"loss": 5.3916,
"mean_token_accuracy": 0.18765619844198228,
"num_tokens": 24258332.0,
"step": 10875
},
{
"entropy": 5.594397068023682,
"epoch": 1.2219351940248218,
"grad_norm": 1.4375,
"learning_rate": 0.0004861607144690587,
"loss": 5.3851,
"mean_token_accuracy": 0.18180985301733016,
"num_tokens": 24268685.0,
"step": 10880
},
{
"entropy": 5.592393112182617,
"epoch": 1.2224967709327792,
"grad_norm": 1.328125,
"learning_rate": 0.0004861468477224563,
"loss": 5.4277,
"mean_token_accuracy": 0.17993273884057998,
"num_tokens": 24280660.0,
"step": 10885
},
{
"entropy": 5.530588817596436,
"epoch": 1.2230583478407369,
"grad_norm": 1.3515625,
"learning_rate": 0.00048613297425285635,
"loss": 5.3614,
"mean_token_accuracy": 0.18627060800790787,
"num_tokens": 24291649.0,
"step": 10890
},
{
"entropy": 5.519206476211548,
"epoch": 1.2236199247486943,
"grad_norm": 1.2109375,
"learning_rate": 0.0004861190940607007,
"loss": 5.3737,
"mean_token_accuracy": 0.18917765468358994,
"num_tokens": 24303108.0,
"step": 10895
},
{
"entropy": 5.500387191772461,
"epoch": 1.224181501656652,
"grad_norm": 1.484375,
"learning_rate": 0.0004861052071464312,
"loss": 5.4078,
"mean_token_accuracy": 0.1736799031496048,
"num_tokens": 24315939.0,
"step": 10900
},
{
"entropy": 5.516234683990478,
"epoch": 1.2247430785646094,
"grad_norm": 1.359375,
"learning_rate": 0.00048609131351048994,
"loss": 5.3741,
"mean_token_accuracy": 0.18524931669235228,
"num_tokens": 24326916.0,
"step": 10905
},
{
"entropy": 5.545850133895874,
"epoch": 1.225304655472567,
"grad_norm": 1.5625,
"learning_rate": 0.0004860774131533195,
"loss": 5.4314,
"mean_token_accuracy": 0.18763466477394103,
"num_tokens": 24337790.0,
"step": 10910
},
{
"entropy": 5.43498272895813,
"epoch": 1.2258662323805245,
"grad_norm": 1.7265625,
"learning_rate": 0.00048606350607536235,
"loss": 5.2317,
"mean_token_accuracy": 0.19259089678525926,
"num_tokens": 24347930.0,
"step": 10915
},
{
"entropy": 5.524466276168823,
"epoch": 1.2264278092884822,
"grad_norm": 1.59375,
"learning_rate": 0.0004860495922770612,
"loss": 5.3503,
"mean_token_accuracy": 0.1838722288608551,
"num_tokens": 24359213.0,
"step": 10920
},
{
"entropy": 5.510011482238769,
"epoch": 1.2269893861964396,
"grad_norm": 1.515625,
"learning_rate": 0.00048603567175885926,
"loss": 5.306,
"mean_token_accuracy": 0.18401524871587754,
"num_tokens": 24371000.0,
"step": 10925
},
{
"entropy": 5.462797927856445,
"epoch": 1.227550963104397,
"grad_norm": 1.3203125,
"learning_rate": 0.0004860217445211997,
"loss": 5.2571,
"mean_token_accuracy": 0.1897013396024704,
"num_tokens": 24381301.0,
"step": 10930
},
{
"entropy": 5.551956415176392,
"epoch": 1.2281125400123547,
"grad_norm": 1.40625,
"learning_rate": 0.0004860078105645259,
"loss": 5.4957,
"mean_token_accuracy": 0.17296673804521562,
"num_tokens": 24391808.0,
"step": 10935
},
{
"entropy": 5.632581472396851,
"epoch": 1.2286741169203121,
"grad_norm": 1.328125,
"learning_rate": 0.0004859938698892816,
"loss": 5.6079,
"mean_token_accuracy": 0.17544290125370027,
"num_tokens": 24404220.0,
"step": 10940
},
{
"entropy": 5.598527240753174,
"epoch": 1.2292356938282698,
"grad_norm": 1.3671875,
"learning_rate": 0.00048597992249591065,
"loss": 5.4001,
"mean_token_accuracy": 0.18651440143585205,
"num_tokens": 24414906.0,
"step": 10945
},
{
"entropy": 5.521800994873047,
"epoch": 1.2297972707362272,
"grad_norm": 1.4375,
"learning_rate": 0.00048596596838485704,
"loss": 5.3944,
"mean_token_accuracy": 0.1784210979938507,
"num_tokens": 24426469.0,
"step": 10950
},
{
"entropy": 5.487932920455933,
"epoch": 1.230358847644185,
"grad_norm": 1.453125,
"learning_rate": 0.0004859520075565654,
"loss": 5.3604,
"mean_token_accuracy": 0.18698634654283525,
"num_tokens": 24437683.0,
"step": 10955
},
{
"entropy": 5.4326096534729,
"epoch": 1.2309204245521423,
"grad_norm": 1.3671875,
"learning_rate": 0.00048593804001147986,
"loss": 5.2604,
"mean_token_accuracy": 0.18916900008916854,
"num_tokens": 24449038.0,
"step": 10960
},
{
"entropy": 5.56870403289795,
"epoch": 1.2314820014601,
"grad_norm": 1.4609375,
"learning_rate": 0.0004859240657500453,
"loss": 5.3902,
"mean_token_accuracy": 0.1866658717393875,
"num_tokens": 24460228.0,
"step": 10965
},
{
"entropy": 5.54244647026062,
"epoch": 1.2320435783680574,
"grad_norm": 1.375,
"learning_rate": 0.00048591008477270675,
"loss": 5.452,
"mean_token_accuracy": 0.1829054042696953,
"num_tokens": 24471222.0,
"step": 10970
},
{
"entropy": 5.4597314357757565,
"epoch": 1.232605155276015,
"grad_norm": 1.859375,
"learning_rate": 0.00048589609707990925,
"loss": 5.3876,
"mean_token_accuracy": 0.1801070064306259,
"num_tokens": 24482127.0,
"step": 10975
},
{
"entropy": 5.596265506744385,
"epoch": 1.2331667321839725,
"grad_norm": 1.6328125,
"learning_rate": 0.00048588210267209823,
"loss": 5.3634,
"mean_token_accuracy": 0.1806269481778145,
"num_tokens": 24492963.0,
"step": 10980
},
{
"entropy": 5.636275148391723,
"epoch": 1.2337283090919302,
"grad_norm": 1.5234375,
"learning_rate": 0.00048586810154971935,
"loss": 5.4027,
"mean_token_accuracy": 0.18529999256134033,
"num_tokens": 24502885.0,
"step": 10985
},
{
"entropy": 5.577983617782593,
"epoch": 1.2342898859998876,
"grad_norm": 1.59375,
"learning_rate": 0.00048585409371321826,
"loss": 5.5075,
"mean_token_accuracy": 0.18308964669704436,
"num_tokens": 24514990.0,
"step": 10990
},
{
"entropy": 5.5980823040008545,
"epoch": 1.2348514629078453,
"grad_norm": 1.5078125,
"learning_rate": 0.00048584007916304106,
"loss": 5.3985,
"mean_token_accuracy": 0.1800481677055359,
"num_tokens": 24525555.0,
"step": 10995
},
{
"entropy": 5.588136816024781,
"epoch": 1.2354130398158027,
"grad_norm": 1.4453125,
"learning_rate": 0.00048582605789963395,
"loss": 5.4489,
"mean_token_accuracy": 0.1760024070739746,
"num_tokens": 24536350.0,
"step": 11000
},
{
"entropy": 5.645097064971924,
"epoch": 1.2359746167237604,
"grad_norm": 1.5,
"learning_rate": 0.0004858120299234434,
"loss": 5.4628,
"mean_token_accuracy": 0.17649738043546676,
"num_tokens": 24547756.0,
"step": 11005
},
{
"entropy": 5.528808212280273,
"epoch": 1.2365361936317179,
"grad_norm": 1.2578125,
"learning_rate": 0.00048579799523491607,
"loss": 5.3388,
"mean_token_accuracy": 0.18447330594062805,
"num_tokens": 24559897.0,
"step": 11010
},
{
"entropy": 5.520873928070069,
"epoch": 1.2370977705396755,
"grad_norm": 1.9765625,
"learning_rate": 0.0004857839538344988,
"loss": 5.3593,
"mean_token_accuracy": 0.17841513305902482,
"num_tokens": 24570168.0,
"step": 11015
},
{
"entropy": 5.546609354019165,
"epoch": 1.237659347447633,
"grad_norm": 1.703125,
"learning_rate": 0.00048576990572263875,
"loss": 5.3118,
"mean_token_accuracy": 0.1912241041660309,
"num_tokens": 24580918.0,
"step": 11020
},
{
"entropy": 5.5412928581237795,
"epoch": 1.2382209243555904,
"grad_norm": 1.625,
"learning_rate": 0.0004857558508997831,
"loss": 5.3396,
"mean_token_accuracy": 0.1820438638329506,
"num_tokens": 24592378.0,
"step": 11025
},
{
"entropy": 5.585161304473877,
"epoch": 1.238782501263548,
"grad_norm": 1.5859375,
"learning_rate": 0.0004857417893663794,
"loss": 5.451,
"mean_token_accuracy": 0.1831345170736313,
"num_tokens": 24603666.0,
"step": 11030
},
{
"entropy": 5.490879392623901,
"epoch": 1.2393440781715055,
"grad_norm": 1.9765625,
"learning_rate": 0.0004857277211228754,
"loss": 5.3965,
"mean_token_accuracy": 0.18626144230365754,
"num_tokens": 24615299.0,
"step": 11035
},
{
"entropy": 5.523722887039185,
"epoch": 1.2399056550794632,
"grad_norm": 1.6796875,
"learning_rate": 0.0004857136461697191,
"loss": 5.3665,
"mean_token_accuracy": 0.18449280112981797,
"num_tokens": 24626138.0,
"step": 11040
},
{
"entropy": 5.527344226837158,
"epoch": 1.2404672319874206,
"grad_norm": 2.203125,
"learning_rate": 0.00048569956450735853,
"loss": 5.3489,
"mean_token_accuracy": 0.1806508094072342,
"num_tokens": 24636720.0,
"step": 11045
},
{
"entropy": 5.565697383880615,
"epoch": 1.2410288088953783,
"grad_norm": 1.7265625,
"learning_rate": 0.00048568547613624206,
"loss": 5.3685,
"mean_token_accuracy": 0.1792961373925209,
"num_tokens": 24648147.0,
"step": 11050
},
{
"entropy": 5.525727605819702,
"epoch": 1.2415903858033357,
"grad_norm": 1.6171875,
"learning_rate": 0.00048567138105681833,
"loss": 5.4299,
"mean_token_accuracy": 0.1852990508079529,
"num_tokens": 24659825.0,
"step": 11055
},
{
"entropy": 5.542980194091797,
"epoch": 1.2421519627112934,
"grad_norm": 1.5,
"learning_rate": 0.00048565727926953617,
"loss": 5.3681,
"mean_token_accuracy": 0.18868448287248613,
"num_tokens": 24671583.0,
"step": 11060
},
{
"entropy": 5.503182983398437,
"epoch": 1.2427135396192508,
"grad_norm": 1.5625,
"learning_rate": 0.0004856431707748445,
"loss": 5.2989,
"mean_token_accuracy": 0.1851896122097969,
"num_tokens": 24682034.0,
"step": 11065
},
{
"entropy": 5.538720893859863,
"epoch": 1.2432751165272085,
"grad_norm": 1.859375,
"learning_rate": 0.0004856290555731926,
"loss": 5.3925,
"mean_token_accuracy": 0.18266519755125046,
"num_tokens": 24693624.0,
"step": 11070
},
{
"entropy": 5.612336730957031,
"epoch": 1.243836693435166,
"grad_norm": 1.734375,
"learning_rate": 0.0004856149336650299,
"loss": 5.4247,
"mean_token_accuracy": 0.1814174622297287,
"num_tokens": 24705340.0,
"step": 11075
},
{
"entropy": 5.599944067001343,
"epoch": 1.2443982703431236,
"grad_norm": 1.328125,
"learning_rate": 0.0004856008050508059,
"loss": 5.4811,
"mean_token_accuracy": 0.17706038504838945,
"num_tokens": 24717115.0,
"step": 11080
},
{
"entropy": 5.521806907653809,
"epoch": 1.244959847251081,
"grad_norm": 1.5234375,
"learning_rate": 0.0004855866697309707,
"loss": 5.2976,
"mean_token_accuracy": 0.19211477637290955,
"num_tokens": 24727768.0,
"step": 11085
},
{
"entropy": 5.573913049697876,
"epoch": 1.2455214241590387,
"grad_norm": 1.6796875,
"learning_rate": 0.0004855725277059742,
"loss": 5.4781,
"mean_token_accuracy": 0.17264461368322373,
"num_tokens": 24740054.0,
"step": 11090
},
{
"entropy": 5.5935383319854735,
"epoch": 1.246083001066996,
"grad_norm": 1.6015625,
"learning_rate": 0.0004855583789762668,
"loss": 5.3534,
"mean_token_accuracy": 0.17889434248209,
"num_tokens": 24750658.0,
"step": 11095
},
{
"entropy": 5.65900297164917,
"epoch": 1.2466445779749538,
"grad_norm": 1.3125,
"learning_rate": 0.000485544223542299,
"loss": 5.5375,
"mean_token_accuracy": 0.17957090884447097,
"num_tokens": 24762730.0,
"step": 11100
},
{
"entropy": 5.48320631980896,
"epoch": 1.2472061548829112,
"grad_norm": 1.453125,
"learning_rate": 0.0004855300614045214,
"loss": 5.2921,
"mean_token_accuracy": 0.1923157498240471,
"num_tokens": 24773865.0,
"step": 11105
},
{
"entropy": 5.457174348831177,
"epoch": 1.2477677317908689,
"grad_norm": 1.5390625,
"learning_rate": 0.00048551589256338497,
"loss": 5.3806,
"mean_token_accuracy": 0.18625938296318054,
"num_tokens": 24783030.0,
"step": 11110
},
{
"entropy": 5.499928426742554,
"epoch": 1.2483293086988263,
"grad_norm": 1.4296875,
"learning_rate": 0.00048550171701934087,
"loss": 5.3228,
"mean_token_accuracy": 0.18795024901628493,
"num_tokens": 24793177.0,
"step": 11115
},
{
"entropy": 5.657392835617065,
"epoch": 1.2488908856067837,
"grad_norm": 1.5,
"learning_rate": 0.00048548753477284053,
"loss": 5.4296,
"mean_token_accuracy": 0.18393117934465408,
"num_tokens": 24804796.0,
"step": 11120
},
{
"entropy": 5.611145496368408,
"epoch": 1.2494524625147414,
"grad_norm": 1.7265625,
"learning_rate": 0.0004854733458243354,
"loss": 5.4286,
"mean_token_accuracy": 0.18265710920095443,
"num_tokens": 24816806.0,
"step": 11125
},
{
"entropy": 5.494162368774414,
"epoch": 1.250014039422699,
"grad_norm": 1.796875,
"learning_rate": 0.0004854591501742773,
"loss": 5.385,
"mean_token_accuracy": 0.1911620482802391,
"num_tokens": 24826888.0,
"step": 11130
},
{
"entropy": 5.570735883712769,
"epoch": 1.2505756163306565,
"grad_norm": 1.421875,
"learning_rate": 0.00048544494782311825,
"loss": 5.3331,
"mean_token_accuracy": 0.18903171718120576,
"num_tokens": 24837731.0,
"step": 11135
},
{
"entropy": 5.493582153320313,
"epoch": 1.251137193238614,
"grad_norm": 1.3359375,
"learning_rate": 0.00048543073877131037,
"loss": 5.3452,
"mean_token_accuracy": 0.1867301270365715,
"num_tokens": 24847900.0,
"step": 11140
},
{
"entropy": 5.510366678237915,
"epoch": 1.2516987701465716,
"grad_norm": 1.421875,
"learning_rate": 0.00048541652301930627,
"loss": 5.4158,
"mean_token_accuracy": 0.18399305045604705,
"num_tokens": 24858878.0,
"step": 11145
},
{
"entropy": 5.532511138916016,
"epoch": 1.252260347054529,
"grad_norm": 1.3359375,
"learning_rate": 0.0004854023005675585,
"loss": 5.3851,
"mean_token_accuracy": 0.18545827120542527,
"num_tokens": 24870051.0,
"step": 11150
},
{
"entropy": 5.565869760513306,
"epoch": 1.2528219239624867,
"grad_norm": 1.5234375,
"learning_rate": 0.0004853880714165197,
"loss": 5.4576,
"mean_token_accuracy": 0.1822757229208946,
"num_tokens": 24881671.0,
"step": 11155
},
{
"entropy": 5.590207195281982,
"epoch": 1.2533835008704441,
"grad_norm": 2.0,
"learning_rate": 0.00048537383556664315,
"loss": 5.3193,
"mean_token_accuracy": 0.17835842519998552,
"num_tokens": 24892944.0,
"step": 11160
},
{
"entropy": 5.554069185256958,
"epoch": 1.2539450777784018,
"grad_norm": 1.546875,
"learning_rate": 0.0004853595930183822,
"loss": 5.4265,
"mean_token_accuracy": 0.17911816984415055,
"num_tokens": 24903346.0,
"step": 11165
},
{
"entropy": 5.517368221282959,
"epoch": 1.2545066546863592,
"grad_norm": 1.390625,
"learning_rate": 0.0004853453437721901,
"loss": 5.3855,
"mean_token_accuracy": 0.1830117017030716,
"num_tokens": 24913904.0,
"step": 11170
},
{
"entropy": 5.5398530006408695,
"epoch": 1.255068231594317,
"grad_norm": 1.5625,
"learning_rate": 0.0004853310878285208,
"loss": 5.3344,
"mean_token_accuracy": 0.18951363265514373,
"num_tokens": 24924635.0,
"step": 11175
},
{
"entropy": 5.5834269523620605,
"epoch": 1.2556298085022743,
"grad_norm": 1.6640625,
"learning_rate": 0.00048531682518782797,
"loss": 5.5285,
"mean_token_accuracy": 0.17625329047441482,
"num_tokens": 24935466.0,
"step": 11180
},
{
"entropy": 5.616320610046387,
"epoch": 1.256191385410232,
"grad_norm": 1.984375,
"learning_rate": 0.00048530255585056594,
"loss": 5.5005,
"mean_token_accuracy": 0.17889171093702316,
"num_tokens": 24948348.0,
"step": 11185
},
{
"entropy": 5.557467460632324,
"epoch": 1.2567529623181894,
"grad_norm": 1.5078125,
"learning_rate": 0.0004852882798171889,
"loss": 5.3282,
"mean_token_accuracy": 0.18867427259683608,
"num_tokens": 24959242.0,
"step": 11190
},
{
"entropy": 5.477604389190674,
"epoch": 1.2573145392261469,
"grad_norm": 1.5234375,
"learning_rate": 0.0004852739970881515,
"loss": 5.2975,
"mean_token_accuracy": 0.1895180121064186,
"num_tokens": 24969855.0,
"step": 11195
},
{
"entropy": 5.489260339736939,
"epoch": 1.2578761161341045,
"grad_norm": 1.296875,
"learning_rate": 0.00048525970766390847,
"loss": 5.3323,
"mean_token_accuracy": 0.18774235248565674,
"num_tokens": 24980378.0,
"step": 11200
},
{
"entropy": 5.622737693786621,
"epoch": 1.2584376930420622,
"grad_norm": 2.171875,
"learning_rate": 0.0004852454115449148,
"loss": 5.4009,
"mean_token_accuracy": 0.1819795548915863,
"num_tokens": 24990646.0,
"step": 11205
},
{
"entropy": 5.678129863739014,
"epoch": 1.2589992699500197,
"grad_norm": 1.3984375,
"learning_rate": 0.0004852311087316257,
"loss": 5.5155,
"mean_token_accuracy": 0.17269572764635086,
"num_tokens": 25002644.0,
"step": 11210
},
{
"entropy": 5.52016019821167,
"epoch": 1.259560846857977,
"grad_norm": 1.265625,
"learning_rate": 0.0004852167992244966,
"loss": 5.4083,
"mean_token_accuracy": 0.18929736018180848,
"num_tokens": 25012635.0,
"step": 11215
},
{
"entropy": 5.605111694335937,
"epoch": 1.2601224237659348,
"grad_norm": 1.8984375,
"learning_rate": 0.000485202483023983,
"loss": 5.5114,
"mean_token_accuracy": 0.17794987559318542,
"num_tokens": 25024289.0,
"step": 11220
},
{
"entropy": 5.579466772079468,
"epoch": 1.2606840006738924,
"grad_norm": 1.2421875,
"learning_rate": 0.00048518816013054083,
"loss": 5.2855,
"mean_token_accuracy": 0.18669219315052032,
"num_tokens": 25035316.0,
"step": 11225
},
{
"entropy": 5.480184412002563,
"epoch": 1.2612455775818499,
"grad_norm": 1.53125,
"learning_rate": 0.00048517383054462617,
"loss": 5.3394,
"mean_token_accuracy": 0.1875536784529686,
"num_tokens": 25046011.0,
"step": 11230
},
{
"entropy": 5.56017050743103,
"epoch": 1.2618071544898073,
"grad_norm": 1.5234375,
"learning_rate": 0.0004851594942666952,
"loss": 5.3461,
"mean_token_accuracy": 0.18598045855760575,
"num_tokens": 25057402.0,
"step": 11235
},
{
"entropy": 5.498341083526611,
"epoch": 1.262368731397765,
"grad_norm": 1.421875,
"learning_rate": 0.00048514515129720445,
"loss": 5.3397,
"mean_token_accuracy": 0.18736465126276017,
"num_tokens": 25068287.0,
"step": 11240
},
{
"entropy": 5.465812349319458,
"epoch": 1.2629303083057224,
"grad_norm": 1.8125,
"learning_rate": 0.0004851308016366105,
"loss": 5.3174,
"mean_token_accuracy": 0.18423836529254914,
"num_tokens": 25079534.0,
"step": 11245
},
{
"entropy": 5.446156024932861,
"epoch": 1.26349188521368,
"grad_norm": 1.546875,
"learning_rate": 0.00048511644528537043,
"loss": 5.2952,
"mean_token_accuracy": 0.18630336225032806,
"num_tokens": 25090825.0,
"step": 11250
},
{
"entropy": 5.561888027191162,
"epoch": 1.2640534621216375,
"grad_norm": 1.53125,
"learning_rate": 0.0004851020822439412,
"loss": 5.4059,
"mean_token_accuracy": 0.18145754039287568,
"num_tokens": 25101859.0,
"step": 11255
},
{
"entropy": 5.458061361312867,
"epoch": 1.2646150390295952,
"grad_norm": 1.7109375,
"learning_rate": 0.00048508771251278015,
"loss": 5.3019,
"mean_token_accuracy": 0.19039927273988724,
"num_tokens": 25111416.0,
"step": 11260
},
{
"entropy": 5.544156122207641,
"epoch": 1.2651766159375526,
"grad_norm": 1.5625,
"learning_rate": 0.0004850733360923449,
"loss": 5.4135,
"mean_token_accuracy": 0.18203299790620803,
"num_tokens": 25122129.0,
"step": 11265
},
{
"entropy": 5.537650680541992,
"epoch": 1.2657381928455103,
"grad_norm": 1.671875,
"learning_rate": 0.0004850589529830931,
"loss": 5.4059,
"mean_token_accuracy": 0.1873589739203453,
"num_tokens": 25132429.0,
"step": 11270
},
{
"entropy": 5.480603504180908,
"epoch": 1.2662997697534677,
"grad_norm": 1.4609375,
"learning_rate": 0.00048504456318548285,
"loss": 5.43,
"mean_token_accuracy": 0.18622178435325623,
"num_tokens": 25144242.0,
"step": 11275
},
{
"entropy": 5.66269006729126,
"epoch": 1.2668613466614254,
"grad_norm": 1.7890625,
"learning_rate": 0.0004850301666999722,
"loss": 5.4552,
"mean_token_accuracy": 0.1849849745631218,
"num_tokens": 25155332.0,
"step": 11280
},
{
"entropy": 5.616597080230713,
"epoch": 1.2674229235693828,
"grad_norm": 1.8671875,
"learning_rate": 0.0004850157635270196,
"loss": 5.3407,
"mean_token_accuracy": 0.1864331230521202,
"num_tokens": 25166024.0,
"step": 11285
},
{
"entropy": 5.45111083984375,
"epoch": 1.2679845004773402,
"grad_norm": 1.484375,
"learning_rate": 0.0004850013536670836,
"loss": 5.3015,
"mean_token_accuracy": 0.18800063282251359,
"num_tokens": 25176697.0,
"step": 11290
},
{
"entropy": 5.514375114440918,
"epoch": 1.268546077385298,
"grad_norm": 1.453125,
"learning_rate": 0.0004849869371206231,
"loss": 5.4099,
"mean_token_accuracy": 0.1873180702328682,
"num_tokens": 25187421.0,
"step": 11295
},
{
"entropy": 5.570020055770874,
"epoch": 1.2691076542932556,
"grad_norm": 1.9453125,
"learning_rate": 0.00048497251388809697,
"loss": 5.4205,
"mean_token_accuracy": 0.1843590721487999,
"num_tokens": 25198851.0,
"step": 11300
},
{
"entropy": 5.59329400062561,
"epoch": 1.269669231201213,
"grad_norm": 1.265625,
"learning_rate": 0.0004849580839699647,
"loss": 5.4779,
"mean_token_accuracy": 0.18099639266729356,
"num_tokens": 25209984.0,
"step": 11305
},
{
"entropy": 5.5673970699310305,
"epoch": 1.2702308081091704,
"grad_norm": 1.3515625,
"learning_rate": 0.00048494364736668554,
"loss": 5.3447,
"mean_token_accuracy": 0.18685221821069717,
"num_tokens": 25220672.0,
"step": 11310
},
{
"entropy": 5.570596647262573,
"epoch": 1.270792385017128,
"grad_norm": 1.3515625,
"learning_rate": 0.0004849292040787192,
"loss": 5.3764,
"mean_token_accuracy": 0.18671778887510299,
"num_tokens": 25231906.0,
"step": 11315
},
{
"entropy": 5.5479958057403564,
"epoch": 1.2713539619250858,
"grad_norm": 1.3359375,
"learning_rate": 0.0004849147541065256,
"loss": 5.398,
"mean_token_accuracy": 0.1802460730075836,
"num_tokens": 25242902.0,
"step": 11320
},
{
"entropy": 5.533203411102295,
"epoch": 1.2719155388330432,
"grad_norm": 1.328125,
"learning_rate": 0.0004849002974505649,
"loss": 5.4524,
"mean_token_accuracy": 0.18079664260149003,
"num_tokens": 25254363.0,
"step": 11325
},
{
"entropy": 5.515308284759522,
"epoch": 1.2724771157410006,
"grad_norm": 1.375,
"learning_rate": 0.00048488583411129723,
"loss": 5.4012,
"mean_token_accuracy": 0.18078778684139252,
"num_tokens": 25267108.0,
"step": 11330
},
{
"entropy": 5.502285194396973,
"epoch": 1.2730386926489583,
"grad_norm": 1.46875,
"learning_rate": 0.0004848713640891832,
"loss": 5.3524,
"mean_token_accuracy": 0.1893145740032196,
"num_tokens": 25278583.0,
"step": 11335
},
{
"entropy": 5.625028800964356,
"epoch": 1.2736002695569157,
"grad_norm": 1.390625,
"learning_rate": 0.0004848568873846835,
"loss": 5.4381,
"mean_token_accuracy": 0.18827813118696213,
"num_tokens": 25289813.0,
"step": 11340
},
{
"entropy": 5.5319176197052,
"epoch": 1.2741618464648734,
"grad_norm": 1.359375,
"learning_rate": 0.00048484240399825925,
"loss": 5.3381,
"mean_token_accuracy": 0.1844951793551445,
"num_tokens": 25301003.0,
"step": 11345
},
{
"entropy": 5.5084202766418455,
"epoch": 1.2747234233728308,
"grad_norm": 1.6484375,
"learning_rate": 0.00048482791393037134,
"loss": 5.3353,
"mean_token_accuracy": 0.19017457813024521,
"num_tokens": 25312535.0,
"step": 11350
},
{
"entropy": 5.570276975631714,
"epoch": 1.2752850002807885,
"grad_norm": 1.5703125,
"learning_rate": 0.00048481341718148137,
"loss": 5.3613,
"mean_token_accuracy": 0.1851690262556076,
"num_tokens": 25323333.0,
"step": 11355
},
{
"entropy": 5.616423273086548,
"epoch": 1.275846577188746,
"grad_norm": 1.3359375,
"learning_rate": 0.00048479891375205074,
"loss": 5.4858,
"mean_token_accuracy": 0.18305027931928636,
"num_tokens": 25334865.0,
"step": 11360
},
{
"entropy": 5.590129423141479,
"epoch": 1.2764081540967036,
"grad_norm": 1.328125,
"learning_rate": 0.00048478440364254136,
"loss": 5.4448,
"mean_token_accuracy": 0.18169519305229187,
"num_tokens": 25346672.0,
"step": 11365
},
{
"entropy": 5.503706979751587,
"epoch": 1.276969731004661,
"grad_norm": 1.25,
"learning_rate": 0.0004847698868534152,
"loss": 5.4429,
"mean_token_accuracy": 0.1864679977297783,
"num_tokens": 25358702.0,
"step": 11370
},
{
"entropy": 5.58003077507019,
"epoch": 1.2775313079126187,
"grad_norm": 1.2890625,
"learning_rate": 0.0004847553633851345,
"loss": 5.3701,
"mean_token_accuracy": 0.18267892450094222,
"num_tokens": 25369813.0,
"step": 11375
},
{
"entropy": 5.589847373962402,
"epoch": 1.2780928848205761,
"grad_norm": 1.625,
"learning_rate": 0.0004847408332381616,
"loss": 5.4202,
"mean_token_accuracy": 0.17997910231351852,
"num_tokens": 25381042.0,
"step": 11380
},
{
"entropy": 5.527003765106201,
"epoch": 1.2786544617285336,
"grad_norm": 1.3046875,
"learning_rate": 0.00048472629641295933,
"loss": 5.3977,
"mean_token_accuracy": 0.1788851201534271,
"num_tokens": 25391286.0,
"step": 11385
},
{
"entropy": 5.515652847290039,
"epoch": 1.2792160386364912,
"grad_norm": 1.4765625,
"learning_rate": 0.0004847117529099904,
"loss": 5.3388,
"mean_token_accuracy": 0.1879303202033043,
"num_tokens": 25402335.0,
"step": 11390
},
{
"entropy": 5.475042819976807,
"epoch": 1.279777615544449,
"grad_norm": 1.375,
"learning_rate": 0.00048469720272971797,
"loss": 5.2812,
"mean_token_accuracy": 0.19062529057264327,
"num_tokens": 25413484.0,
"step": 11395
},
{
"entropy": 5.586601734161377,
"epoch": 1.2803391924524063,
"grad_norm": 1.4375,
"learning_rate": 0.0004846826458726052,
"loss": 5.4061,
"mean_token_accuracy": 0.18350547552108765,
"num_tokens": 25424276.0,
"step": 11400
},
{
"entropy": 5.424251127243042,
"epoch": 1.2809007693603638,
"grad_norm": 1.625,
"learning_rate": 0.00048466808233911567,
"loss": 5.2862,
"mean_token_accuracy": 0.1907456785440445,
"num_tokens": 25435178.0,
"step": 11405
},
{
"entropy": 5.415072298049926,
"epoch": 1.2814623462683215,
"grad_norm": 1.515625,
"learning_rate": 0.00048465351212971313,
"loss": 5.2432,
"mean_token_accuracy": 0.18663379400968552,
"num_tokens": 25445792.0,
"step": 11410
},
{
"entropy": 5.482068681716919,
"epoch": 1.2820239231762791,
"grad_norm": 1.390625,
"learning_rate": 0.0004846389352448614,
"loss": 5.3653,
"mean_token_accuracy": 0.17930356413125992,
"num_tokens": 25456598.0,
"step": 11415
},
{
"entropy": 5.516743421554565,
"epoch": 1.2825855000842366,
"grad_norm": 1.3515625,
"learning_rate": 0.0004846243516850247,
"loss": 5.4225,
"mean_token_accuracy": 0.18563919216394426,
"num_tokens": 25468311.0,
"step": 11420
},
{
"entropy": 5.636369371414185,
"epoch": 1.283147076992194,
"grad_norm": 1.5078125,
"learning_rate": 0.0004846097614506673,
"loss": 5.4329,
"mean_token_accuracy": 0.18258702605962754,
"num_tokens": 25480680.0,
"step": 11425
},
{
"entropy": 5.62813925743103,
"epoch": 1.2837086539001517,
"grad_norm": 1.5390625,
"learning_rate": 0.00048459516454225384,
"loss": 5.4043,
"mean_token_accuracy": 0.1892827644944191,
"num_tokens": 25492255.0,
"step": 11430
},
{
"entropy": 5.5068169116973875,
"epoch": 1.284270230808109,
"grad_norm": 1.546875,
"learning_rate": 0.000484580560960249,
"loss": 5.442,
"mean_token_accuracy": 0.18167399764060974,
"num_tokens": 25504530.0,
"step": 11435
},
{
"entropy": 5.481035470962524,
"epoch": 1.2848318077160668,
"grad_norm": 1.71875,
"learning_rate": 0.0004845659507051178,
"loss": 5.4019,
"mean_token_accuracy": 0.1859348937869072,
"num_tokens": 25515273.0,
"step": 11440
},
{
"entropy": 5.557617950439453,
"epoch": 1.2853933846240242,
"grad_norm": 1.3515625,
"learning_rate": 0.0004845513337773255,
"loss": 5.4997,
"mean_token_accuracy": 0.179445381462574,
"num_tokens": 25526050.0,
"step": 11445
},
{
"entropy": 5.624632930755615,
"epoch": 1.2859549615319819,
"grad_norm": 1.5078125,
"learning_rate": 0.00048453671017733755,
"loss": 5.4057,
"mean_token_accuracy": 0.17732493877410888,
"num_tokens": 25538372.0,
"step": 11450
},
{
"entropy": 5.637783575057983,
"epoch": 1.2865165384399393,
"grad_norm": 1.4453125,
"learning_rate": 0.0004845220799056194,
"loss": 5.5086,
"mean_token_accuracy": 0.18480895459651947,
"num_tokens": 25549636.0,
"step": 11455
},
{
"entropy": 5.54378924369812,
"epoch": 1.287078115347897,
"grad_norm": 1.5390625,
"learning_rate": 0.00048450744296263685,
"loss": 5.4242,
"mean_token_accuracy": 0.18077969402074814,
"num_tokens": 25560850.0,
"step": 11460
},
{
"entropy": 5.531536817550659,
"epoch": 1.2876396922558544,
"grad_norm": 1.4296875,
"learning_rate": 0.00048449279934885625,
"loss": 5.3632,
"mean_token_accuracy": 0.18788646757602692,
"num_tokens": 25572433.0,
"step": 11465
},
{
"entropy": 5.622464942932129,
"epoch": 1.288201269163812,
"grad_norm": 1.5546875,
"learning_rate": 0.0004844781490647435,
"loss": 5.4431,
"mean_token_accuracy": 0.18224688172340392,
"num_tokens": 25582372.0,
"step": 11470
},
{
"entropy": 5.511342620849609,
"epoch": 1.2887628460717695,
"grad_norm": 1.421875,
"learning_rate": 0.0004844634921107654,
"loss": 5.2633,
"mean_token_accuracy": 0.18831834346055984,
"num_tokens": 25592570.0,
"step": 11475
},
{
"entropy": 5.526323223114014,
"epoch": 1.289324422979727,
"grad_norm": 1.34375,
"learning_rate": 0.00048444882848738835,
"loss": 5.4006,
"mean_token_accuracy": 0.1773192510008812,
"num_tokens": 25603573.0,
"step": 11480
},
{
"entropy": 5.606195068359375,
"epoch": 1.2898859998876846,
"grad_norm": 1.2578125,
"learning_rate": 0.00048443415819507944,
"loss": 5.4187,
"mean_token_accuracy": 0.18668091744184495,
"num_tokens": 25614233.0,
"step": 11485
},
{
"entropy": 5.529943943023682,
"epoch": 1.2904475767956423,
"grad_norm": 1.421875,
"learning_rate": 0.0004844194812343057,
"loss": 5.3708,
"mean_token_accuracy": 0.18135939985513688,
"num_tokens": 25625077.0,
"step": 11490
},
{
"entropy": 5.437040853500366,
"epoch": 1.2910091537035997,
"grad_norm": 1.3046875,
"learning_rate": 0.0004844047976055344,
"loss": 5.4078,
"mean_token_accuracy": 0.1813334822654724,
"num_tokens": 25636693.0,
"step": 11495
},
{
"entropy": 5.443415784835816,
"epoch": 1.2915707306115571,
"grad_norm": 1.3671875,
"learning_rate": 0.00048439010730923313,
"loss": 5.2647,
"mean_token_accuracy": 0.18576423078775406,
"num_tokens": 25647644.0,
"step": 11500
},
{
"entropy": 5.528429746627808,
"epoch": 1.2921323075195148,
"grad_norm": 1.5859375,
"learning_rate": 0.00048437541034586964,
"loss": 5.3072,
"mean_token_accuracy": 0.19314247518777847,
"num_tokens": 25659373.0,
"step": 11505
},
{
"entropy": 5.558218193054199,
"epoch": 1.2926938844274725,
"grad_norm": 1.515625,
"learning_rate": 0.00048436070671591195,
"loss": 5.4365,
"mean_token_accuracy": 0.1921704426407814,
"num_tokens": 25670479.0,
"step": 11510
},
{
"entropy": 5.565453147888183,
"epoch": 1.29325546133543,
"grad_norm": 1.828125,
"learning_rate": 0.00048434599641982806,
"loss": 5.4512,
"mean_token_accuracy": 0.18551308661699295,
"num_tokens": 25682748.0,
"step": 11515
},
{
"entropy": 5.477804899215698,
"epoch": 1.2938170382433873,
"grad_norm": 1.6171875,
"learning_rate": 0.00048433127945808645,
"loss": 5.2604,
"mean_token_accuracy": 0.19417942613363265,
"num_tokens": 25693170.0,
"step": 11520
},
{
"entropy": 5.502341604232788,
"epoch": 1.294378615151345,
"grad_norm": 2.25,
"learning_rate": 0.0004843165558311558,
"loss": 5.4199,
"mean_token_accuracy": 0.18010897934436798,
"num_tokens": 25706009.0,
"step": 11525
},
{
"entropy": 5.608281230926513,
"epoch": 1.2949401920593024,
"grad_norm": 2.28125,
"learning_rate": 0.0004843018255395048,
"loss": 5.4345,
"mean_token_accuracy": 0.18155258893966675,
"num_tokens": 25716036.0,
"step": 11530
},
{
"entropy": 5.521420240402222,
"epoch": 1.29550176896726,
"grad_norm": 1.546875,
"learning_rate": 0.00048428708858360244,
"loss": 5.234,
"mean_token_accuracy": 0.18922073394060135,
"num_tokens": 25726516.0,
"step": 11535
},
{
"entropy": 5.485184192657471,
"epoch": 1.2960633458752175,
"grad_norm": 1.7578125,
"learning_rate": 0.00048427234496391816,
"loss": 5.4031,
"mean_token_accuracy": 0.17972793877124787,
"num_tokens": 25736340.0,
"step": 11540
},
{
"entropy": 5.407545709609986,
"epoch": 1.2966249227831752,
"grad_norm": 1.53125,
"learning_rate": 0.00048425759468092113,
"loss": 5.3503,
"mean_token_accuracy": 0.18365047127008438,
"num_tokens": 25746439.0,
"step": 11545
},
{
"entropy": 5.508850526809693,
"epoch": 1.2971864996911326,
"grad_norm": 1.84375,
"learning_rate": 0.0004842428377350811,
"loss": 5.2843,
"mean_token_accuracy": 0.18573250621557236,
"num_tokens": 25757065.0,
"step": 11550
},
{
"entropy": 5.478228139877319,
"epoch": 1.2977480765990903,
"grad_norm": 1.5390625,
"learning_rate": 0.00048422807412686796,
"loss": 5.3346,
"mean_token_accuracy": 0.1797130286693573,
"num_tokens": 25768021.0,
"step": 11555
},
{
"entropy": 5.541759300231933,
"epoch": 1.2983096535070477,
"grad_norm": 1.3671875,
"learning_rate": 0.0004842133038567519,
"loss": 5.4442,
"mean_token_accuracy": 0.18450896590948104,
"num_tokens": 25780197.0,
"step": 11560
},
{
"entropy": 5.564715099334717,
"epoch": 1.2988712304150054,
"grad_norm": 1.65625,
"learning_rate": 0.000484198526925203,
"loss": 5.3554,
"mean_token_accuracy": 0.18042299896478653,
"num_tokens": 25790536.0,
"step": 11565
},
{
"entropy": 5.497778463363647,
"epoch": 1.2994328073229628,
"grad_norm": 1.671875,
"learning_rate": 0.0004841837433326919,
"loss": 5.3268,
"mean_token_accuracy": 0.18720477521419526,
"num_tokens": 25802397.0,
"step": 11570
},
{
"entropy": 5.507333898544312,
"epoch": 1.2999943842309205,
"grad_norm": 1.5078125,
"learning_rate": 0.0004841689530796893,
"loss": 5.435,
"mean_token_accuracy": 0.18581217527389526,
"num_tokens": 25813961.0,
"step": 11575
},
{
"entropy": 5.600154256820678,
"epoch": 1.300555961138878,
"grad_norm": 1.9765625,
"learning_rate": 0.0004841541561666661,
"loss": 5.4788,
"mean_token_accuracy": 0.17767149806022645,
"num_tokens": 25825610.0,
"step": 11580
},
{
"entropy": 5.540296459197998,
"epoch": 1.3011175380468356,
"grad_norm": 1.46875,
"learning_rate": 0.00048413935259409343,
"loss": 5.2714,
"mean_token_accuracy": 0.19181762039661407,
"num_tokens": 25836076.0,
"step": 11585
},
{
"entropy": 5.4756081104278564,
"epoch": 1.301679114954793,
"grad_norm": 1.59375,
"learning_rate": 0.00048412454236244267,
"loss": 5.4008,
"mean_token_accuracy": 0.19011079221963884,
"num_tokens": 25847341.0,
"step": 11590
},
{
"entropy": 5.433564901351929,
"epoch": 1.3022406918627505,
"grad_norm": 1.5234375,
"learning_rate": 0.00048410972547218545,
"loss": 5.2915,
"mean_token_accuracy": 0.18946561217308044,
"num_tokens": 25856912.0,
"step": 11595
},
{
"entropy": 5.54172649383545,
"epoch": 1.3028022687707081,
"grad_norm": 1.3359375,
"learning_rate": 0.00048409490192379337,
"loss": 5.3231,
"mean_token_accuracy": 0.17947252094745636,
"num_tokens": 25868370.0,
"step": 11600
},
{
"entropy": 5.5106137752532955,
"epoch": 1.3033638456786658,
"grad_norm": 1.3828125,
"learning_rate": 0.0004840800717177386,
"loss": 5.3456,
"mean_token_accuracy": 0.18653806895017624,
"num_tokens": 25879104.0,
"step": 11605
},
{
"entropy": 5.544076919555664,
"epoch": 1.3039254225866233,
"grad_norm": 2.109375,
"learning_rate": 0.0004840652348544931,
"loss": 5.4085,
"mean_token_accuracy": 0.17772976458072662,
"num_tokens": 25889744.0,
"step": 11610
},
{
"entropy": 5.610517454147339,
"epoch": 1.3044869994945807,
"grad_norm": 1.375,
"learning_rate": 0.0004840503913345296,
"loss": 5.5274,
"mean_token_accuracy": 0.17647068947553635,
"num_tokens": 25901615.0,
"step": 11615
},
{
"entropy": 5.594880771636963,
"epoch": 1.3050485764025384,
"grad_norm": 1.515625,
"learning_rate": 0.0004840355411583205,
"loss": 5.3706,
"mean_token_accuracy": 0.18056469559669494,
"num_tokens": 25911418.0,
"step": 11620
},
{
"entropy": 5.579060363769531,
"epoch": 1.3056101533104958,
"grad_norm": 1.2578125,
"learning_rate": 0.00048402068432633864,
"loss": 5.4749,
"mean_token_accuracy": 0.18650885969400405,
"num_tokens": 25922141.0,
"step": 11625
},
{
"entropy": 5.5719592571258545,
"epoch": 1.3061717302184535,
"grad_norm": 1.484375,
"learning_rate": 0.0004840058208390572,
"loss": 5.4661,
"mean_token_accuracy": 0.18023855686187745,
"num_tokens": 25933897.0,
"step": 11630
},
{
"entropy": 5.60835452079773,
"epoch": 1.306733307126411,
"grad_norm": 1.4453125,
"learning_rate": 0.0004839909506969494,
"loss": 5.3831,
"mean_token_accuracy": 0.18573159277439116,
"num_tokens": 25943960.0,
"step": 11635
},
{
"entropy": 5.61869249343872,
"epoch": 1.3072948840343686,
"grad_norm": 1.34375,
"learning_rate": 0.0004839760739004886,
"loss": 5.4561,
"mean_token_accuracy": 0.18403375893831253,
"num_tokens": 25955104.0,
"step": 11640
},
{
"entropy": 5.544709253311157,
"epoch": 1.307856460942326,
"grad_norm": 1.546875,
"learning_rate": 0.00048396119045014853,
"loss": 5.3972,
"mean_token_accuracy": 0.18485745042562485,
"num_tokens": 25965180.0,
"step": 11645
},
{
"entropy": 5.54657621383667,
"epoch": 1.3084180378502837,
"grad_norm": 1.4296875,
"learning_rate": 0.0004839463003464032,
"loss": 5.3883,
"mean_token_accuracy": 0.18459032475948334,
"num_tokens": 25976600.0,
"step": 11650
},
{
"entropy": 5.479190397262573,
"epoch": 1.308979614758241,
"grad_norm": 1.5859375,
"learning_rate": 0.0004839314035897265,
"loss": 5.3605,
"mean_token_accuracy": 0.1780159667134285,
"num_tokens": 25987509.0,
"step": 11655
},
{
"entropy": 5.535476970672607,
"epoch": 1.3095411916661988,
"grad_norm": 1.625,
"learning_rate": 0.000483916500180593,
"loss": 5.3075,
"mean_token_accuracy": 0.19037485867738724,
"num_tokens": 25997962.0,
"step": 11660
},
{
"entropy": 5.537170934677124,
"epoch": 1.3101027685741562,
"grad_norm": 1.3984375,
"learning_rate": 0.00048390159011947706,
"loss": 5.3623,
"mean_token_accuracy": 0.18489459455013274,
"num_tokens": 26008932.0,
"step": 11665
},
{
"entropy": 5.52904691696167,
"epoch": 1.3106643454821139,
"grad_norm": 1.625,
"learning_rate": 0.0004838866734068535,
"loss": 5.358,
"mean_token_accuracy": 0.18140730857849122,
"num_tokens": 26020513.0,
"step": 11670
},
{
"entropy": 5.585319471359253,
"epoch": 1.3112259223900713,
"grad_norm": 1.7578125,
"learning_rate": 0.00048387175004319723,
"loss": 5.4397,
"mean_token_accuracy": 0.18062340021133422,
"num_tokens": 26032526.0,
"step": 11675
},
{
"entropy": 5.558652591705322,
"epoch": 1.311787499298029,
"grad_norm": 1.5078125,
"learning_rate": 0.0004838568200289834,
"loss": 5.3627,
"mean_token_accuracy": 0.18629241585731507,
"num_tokens": 26044216.0,
"step": 11680
},
{
"entropy": 5.475536251068116,
"epoch": 1.3123490762059864,
"grad_norm": 1.7109375,
"learning_rate": 0.0004838418833646875,
"loss": 5.2774,
"mean_token_accuracy": 0.18576706498861312,
"num_tokens": 26054012.0,
"step": 11685
},
{
"entropy": 5.512315034866333,
"epoch": 1.3129106531139438,
"grad_norm": 1.34375,
"learning_rate": 0.00048382694005078485,
"loss": 5.4288,
"mean_token_accuracy": 0.1846940502524376,
"num_tokens": 26065365.0,
"step": 11690
},
{
"entropy": 5.569101142883301,
"epoch": 1.3134722300219015,
"grad_norm": 1.46875,
"learning_rate": 0.00048381199008775165,
"loss": 5.4221,
"mean_token_accuracy": 0.178159599006176,
"num_tokens": 26075964.0,
"step": 11695
},
{
"entropy": 5.524285459518433,
"epoch": 1.3140338069298592,
"grad_norm": 1.4453125,
"learning_rate": 0.0004837970334760635,
"loss": 5.3631,
"mean_token_accuracy": 0.18604728877544402,
"num_tokens": 26087572.0,
"step": 11700
},
{
"entropy": 5.485309648513794,
"epoch": 1.3145953838378166,
"grad_norm": 1.4375,
"learning_rate": 0.000483782070216197,
"loss": 5.3007,
"mean_token_accuracy": 0.18665737509727479,
"num_tokens": 26098612.0,
"step": 11705
},
{
"entropy": 5.567965412139893,
"epoch": 1.315156960745774,
"grad_norm": 1.5078125,
"learning_rate": 0.0004837671003086283,
"loss": 5.4566,
"mean_token_accuracy": 0.1815529242157936,
"num_tokens": 26109491.0,
"step": 11710
},
{
"entropy": 5.513163423538208,
"epoch": 1.3157185376537317,
"grad_norm": 1.6875,
"learning_rate": 0.0004837521237538342,
"loss": 5.3384,
"mean_token_accuracy": 0.188800385594368,
"num_tokens": 26120544.0,
"step": 11715
},
{
"entropy": 5.586646270751953,
"epoch": 1.3162801145616891,
"grad_norm": 1.6953125,
"learning_rate": 0.0004837371405522915,
"loss": 5.4445,
"mean_token_accuracy": 0.1796777680516243,
"num_tokens": 26131549.0,
"step": 11720
},
{
"entropy": 5.521964359283447,
"epoch": 1.3168416914696468,
"grad_norm": 1.984375,
"learning_rate": 0.00048372215070447734,
"loss": 5.4153,
"mean_token_accuracy": 0.18669757544994353,
"num_tokens": 26142871.0,
"step": 11725
},
{
"entropy": 5.603567409515381,
"epoch": 1.3174032683776042,
"grad_norm": 1.3515625,
"learning_rate": 0.00048370715421086884,
"loss": 5.5224,
"mean_token_accuracy": 0.1823717549443245,
"num_tokens": 26155293.0,
"step": 11730
},
{
"entropy": 5.6257116317749025,
"epoch": 1.317964845285562,
"grad_norm": 1.5078125,
"learning_rate": 0.0004836921510719438,
"loss": 5.4433,
"mean_token_accuracy": 0.18703652769327164,
"num_tokens": 26167069.0,
"step": 11735
},
{
"entropy": 5.545571231842041,
"epoch": 1.3185264221935193,
"grad_norm": 1.5,
"learning_rate": 0.0004836771412881795,
"loss": 5.4402,
"mean_token_accuracy": 0.17432665079832077,
"num_tokens": 26179142.0,
"step": 11740
},
{
"entropy": 5.564441299438476,
"epoch": 1.319087999101477,
"grad_norm": 1.640625,
"learning_rate": 0.0004836621248600542,
"loss": 5.366,
"mean_token_accuracy": 0.18550941199064255,
"num_tokens": 26190059.0,
"step": 11745
},
{
"entropy": 5.579143333435058,
"epoch": 1.3196495760094344,
"grad_norm": 1.3515625,
"learning_rate": 0.0004836471017880459,
"loss": 5.3903,
"mean_token_accuracy": 0.18530660420656203,
"num_tokens": 26199859.0,
"step": 11750
},
{
"entropy": 5.5864002227783205,
"epoch": 1.320211152917392,
"grad_norm": 1.515625,
"learning_rate": 0.0004836320720726329,
"loss": 5.4449,
"mean_token_accuracy": 0.1793688178062439,
"num_tokens": 26211920.0,
"step": 11755
},
{
"entropy": 5.540060997009277,
"epoch": 1.3207727298253495,
"grad_norm": 1.765625,
"learning_rate": 0.0004836170357142939,
"loss": 5.3744,
"mean_token_accuracy": 0.18224322199821472,
"num_tokens": 26222875.0,
"step": 11760
},
{
"entropy": 5.574612808227539,
"epoch": 1.3213343067333072,
"grad_norm": 1.375,
"learning_rate": 0.0004836019927135076,
"loss": 5.4034,
"mean_token_accuracy": 0.18436541259288788,
"num_tokens": 26233549.0,
"step": 11765
},
{
"entropy": 5.520120668411255,
"epoch": 1.3218958836412646,
"grad_norm": 3.046875,
"learning_rate": 0.0004835869430707529,
"loss": 5.3098,
"mean_token_accuracy": 0.19505088478326799,
"num_tokens": 26243886.0,
"step": 11770
},
{
"entropy": 5.475279235839844,
"epoch": 1.3224574605492223,
"grad_norm": 1.2578125,
"learning_rate": 0.00048357188678650905,
"loss": 5.3268,
"mean_token_accuracy": 0.19218631684780121,
"num_tokens": 26254475.0,
"step": 11775
},
{
"entropy": 5.592524242401123,
"epoch": 1.3230190374571797,
"grad_norm": 1.375,
"learning_rate": 0.0004835568238612553,
"loss": 5.5042,
"mean_token_accuracy": 0.17934693098068238,
"num_tokens": 26265412.0,
"step": 11780
},
{
"entropy": 5.648862266540528,
"epoch": 1.3235806143651372,
"grad_norm": 1.5703125,
"learning_rate": 0.00048354175429547155,
"loss": 5.4351,
"mean_token_accuracy": 0.18346074372529983,
"num_tokens": 26276942.0,
"step": 11785
},
{
"entropy": 5.557818794250489,
"epoch": 1.3241421912730948,
"grad_norm": 1.6015625,
"learning_rate": 0.0004835266780896374,
"loss": 5.4089,
"mean_token_accuracy": 0.1820205733180046,
"num_tokens": 26288009.0,
"step": 11790
},
{
"entropy": 5.58009705543518,
"epoch": 1.3247037681810525,
"grad_norm": 1.46875,
"learning_rate": 0.0004835115952442329,
"loss": 5.547,
"mean_token_accuracy": 0.1757894366979599,
"num_tokens": 26301583.0,
"step": 11795
},
{
"entropy": 5.621467113494873,
"epoch": 1.32526534508901,
"grad_norm": 1.578125,
"learning_rate": 0.00048349650575973843,
"loss": 5.4468,
"mean_token_accuracy": 0.18118268996477127,
"num_tokens": 26312922.0,
"step": 11800
},
{
"entropy": 5.592855548858642,
"epoch": 1.3258269219969674,
"grad_norm": 1.59375,
"learning_rate": 0.00048348140963663436,
"loss": 5.4065,
"mean_token_accuracy": 0.1871054247021675,
"num_tokens": 26324108.0,
"step": 11805
},
{
"entropy": 5.597504615783691,
"epoch": 1.326388498904925,
"grad_norm": 1.453125,
"learning_rate": 0.0004834663068754012,
"loss": 5.4342,
"mean_token_accuracy": 0.18054017201066017,
"num_tokens": 26335669.0,
"step": 11810
},
{
"entropy": 5.4381874084472654,
"epoch": 1.3269500758128827,
"grad_norm": 1.40625,
"learning_rate": 0.0004834511974765202,
"loss": 5.3297,
"mean_token_accuracy": 0.18750093281269073,
"num_tokens": 26346065.0,
"step": 11815
},
{
"entropy": 5.532921218872071,
"epoch": 1.3275116527208402,
"grad_norm": 1.5,
"learning_rate": 0.00048343608144047194,
"loss": 5.3105,
"mean_token_accuracy": 0.18406322300434114,
"num_tokens": 26357567.0,
"step": 11820
},
{
"entropy": 5.542247104644775,
"epoch": 1.3280732296287976,
"grad_norm": 1.6328125,
"learning_rate": 0.00048342095876773825,
"loss": 5.3546,
"mean_token_accuracy": 0.18369968980550766,
"num_tokens": 26367805.0,
"step": 11825
},
{
"entropy": 5.474393701553344,
"epoch": 1.3286348065367553,
"grad_norm": 1.84375,
"learning_rate": 0.0004834058294588003,
"loss": 5.2772,
"mean_token_accuracy": 0.18795597553253174,
"num_tokens": 26379343.0,
"step": 11830
},
{
"entropy": 5.581504678726196,
"epoch": 1.3291963834447127,
"grad_norm": 1.5859375,
"learning_rate": 0.00048339069351413993,
"loss": 5.5046,
"mean_token_accuracy": 0.17808228135108947,
"num_tokens": 26392039.0,
"step": 11835
},
{
"entropy": 5.475123500823974,
"epoch": 1.3297579603526704,
"grad_norm": 1.4765625,
"learning_rate": 0.000483375550934239,
"loss": 5.2611,
"mean_token_accuracy": 0.1969547837972641,
"num_tokens": 26403324.0,
"step": 11840
},
{
"entropy": 5.52301516532898,
"epoch": 1.3303195372606278,
"grad_norm": 1.484375,
"learning_rate": 0.0004833604017195798,
"loss": 5.3372,
"mean_token_accuracy": 0.19097117930650712,
"num_tokens": 26413435.0,
"step": 11845
},
{
"entropy": 5.520868492126465,
"epoch": 1.3308811141685855,
"grad_norm": 1.875,
"learning_rate": 0.0004833452458706445,
"loss": 5.453,
"mean_token_accuracy": 0.1833457753062248,
"num_tokens": 26424254.0,
"step": 11850
},
{
"entropy": 5.576958417892456,
"epoch": 1.331442691076543,
"grad_norm": 1.5234375,
"learning_rate": 0.0004833300833879159,
"loss": 5.3837,
"mean_token_accuracy": 0.18457611799240112,
"num_tokens": 26435658.0,
"step": 11855
},
{
"entropy": 5.604117393493652,
"epoch": 1.3320042679845006,
"grad_norm": 1.421875,
"learning_rate": 0.0004833149142718765,
"loss": 5.4327,
"mean_token_accuracy": 0.17981316149234772,
"num_tokens": 26447174.0,
"step": 11860
},
{
"entropy": 5.557124423980713,
"epoch": 1.332565844892458,
"grad_norm": 1.5234375,
"learning_rate": 0.0004832997385230096,
"loss": 5.5234,
"mean_token_accuracy": 0.17711610347032547,
"num_tokens": 26459303.0,
"step": 11865
},
{
"entropy": 5.566341400146484,
"epoch": 1.3331274218004157,
"grad_norm": 1.5390625,
"learning_rate": 0.0004832845561417981,
"loss": 5.3758,
"mean_token_accuracy": 0.18285171091556549,
"num_tokens": 26469981.0,
"step": 11870
},
{
"entropy": 5.527902889251709,
"epoch": 1.333688998708373,
"grad_norm": 1.40625,
"learning_rate": 0.0004832693671287256,
"loss": 5.3879,
"mean_token_accuracy": 0.18742477744817734,
"num_tokens": 26481176.0,
"step": 11875
},
{
"entropy": 5.484760522842407,
"epoch": 1.3342505756163305,
"grad_norm": 1.46875,
"learning_rate": 0.0004832541714842757,
"loss": 5.3441,
"mean_token_accuracy": 0.18890009820461273,
"num_tokens": 26492795.0,
"step": 11880
},
{
"entropy": 5.596258354187012,
"epoch": 1.3348121525242882,
"grad_norm": 1.3125,
"learning_rate": 0.0004832389692089321,
"loss": 5.4828,
"mean_token_accuracy": 0.1816885456442833,
"num_tokens": 26504731.0,
"step": 11885
},
{
"entropy": 5.5984110832214355,
"epoch": 1.3353737294322459,
"grad_norm": 1.3671875,
"learning_rate": 0.00048322376030317905,
"loss": 5.3686,
"mean_token_accuracy": 0.18662869334220886,
"num_tokens": 26516816.0,
"step": 11890
},
{
"entropy": 5.575429677963257,
"epoch": 1.3359353063402033,
"grad_norm": 1.484375,
"learning_rate": 0.0004832085447675007,
"loss": 5.4691,
"mean_token_accuracy": 0.18194129019975663,
"num_tokens": 26528861.0,
"step": 11895
},
{
"entropy": 5.6260394096374515,
"epoch": 1.3364968832481607,
"grad_norm": 1.390625,
"learning_rate": 0.00048319332260238153,
"loss": 5.4364,
"mean_token_accuracy": 0.182245472073555,
"num_tokens": 26538816.0,
"step": 11900
},
{
"entropy": 5.668881893157959,
"epoch": 1.3370584601561184,
"grad_norm": 1.4765625,
"learning_rate": 0.00048317809380830624,
"loss": 5.4961,
"mean_token_accuracy": 0.18144724071025847,
"num_tokens": 26548959.0,
"step": 11905
},
{
"entropy": 5.445233726501465,
"epoch": 1.337620037064076,
"grad_norm": 1.8515625,
"learning_rate": 0.0004831628583857597,
"loss": 5.3144,
"mean_token_accuracy": 0.18857230246067047,
"num_tokens": 26559189.0,
"step": 11910
},
{
"entropy": 5.413756036758423,
"epoch": 1.3381816139720335,
"grad_norm": 1.6171875,
"learning_rate": 0.00048314761633522696,
"loss": 5.3043,
"mean_token_accuracy": 0.18844056725502015,
"num_tokens": 26570066.0,
"step": 11915
},
{
"entropy": 5.549174070358276,
"epoch": 1.338743190879991,
"grad_norm": 1.8984375,
"learning_rate": 0.0004831323676571934,
"loss": 5.4773,
"mean_token_accuracy": 0.1837492346763611,
"num_tokens": 26581489.0,
"step": 11920
},
{
"entropy": 5.602809953689575,
"epoch": 1.3393047677879486,
"grad_norm": 1.609375,
"learning_rate": 0.0004831171123521446,
"loss": 5.3858,
"mean_token_accuracy": 0.18264483511447907,
"num_tokens": 26592051.0,
"step": 11925
},
{
"entropy": 5.610823392868042,
"epoch": 1.339866344695906,
"grad_norm": 1.4765625,
"learning_rate": 0.0004831018504205662,
"loss": 5.4028,
"mean_token_accuracy": 0.18691159188747405,
"num_tokens": 26602296.0,
"step": 11930
},
{
"entropy": 5.538307571411133,
"epoch": 1.3404279216038637,
"grad_norm": 1.34375,
"learning_rate": 0.00048308658186294414,
"loss": 5.3676,
"mean_token_accuracy": 0.1819119155406952,
"num_tokens": 26613837.0,
"step": 11935
},
{
"entropy": 5.509485054016113,
"epoch": 1.3409894985118211,
"grad_norm": 1.375,
"learning_rate": 0.0004830713066797646,
"loss": 5.4264,
"mean_token_accuracy": 0.1852059096097946,
"num_tokens": 26624898.0,
"step": 11940
},
{
"entropy": 5.5755088329315186,
"epoch": 1.3415510754197788,
"grad_norm": 1.265625,
"learning_rate": 0.00048305602487151396,
"loss": 5.4113,
"mean_token_accuracy": 0.18815446197986602,
"num_tokens": 26636449.0,
"step": 11945
},
{
"entropy": 5.671694946289063,
"epoch": 1.3421126523277362,
"grad_norm": 1.359375,
"learning_rate": 0.0004830407364386789,
"loss": 5.5193,
"mean_token_accuracy": 0.18132117837667466,
"num_tokens": 26649192.0,
"step": 11950
},
{
"entropy": 5.483750677108764,
"epoch": 1.342674229235694,
"grad_norm": 1.609375,
"learning_rate": 0.000483025441381746,
"loss": 5.3163,
"mean_token_accuracy": 0.1901243507862091,
"num_tokens": 26660005.0,
"step": 11955
},
{
"entropy": 5.616004753112793,
"epoch": 1.3432358061436513,
"grad_norm": 1.6171875,
"learning_rate": 0.00048301013970120235,
"loss": 5.4727,
"mean_token_accuracy": 0.17908573597669603,
"num_tokens": 26672165.0,
"step": 11960
},
{
"entropy": 5.516784906387329,
"epoch": 1.343797383051609,
"grad_norm": 1.2421875,
"learning_rate": 0.00048299483139753517,
"loss": 5.398,
"mean_token_accuracy": 0.18836424350738526,
"num_tokens": 26683135.0,
"step": 11965
},
{
"entropy": 5.527017068862915,
"epoch": 1.3443589599595664,
"grad_norm": 1.4453125,
"learning_rate": 0.0004829795164712319,
"loss": 5.2884,
"mean_token_accuracy": 0.19392993599176406,
"num_tokens": 26693724.0,
"step": 11970
},
{
"entropy": 5.5618674755096436,
"epoch": 1.3449205368675239,
"grad_norm": 1.4765625,
"learning_rate": 0.00048296419492278023,
"loss": 5.4864,
"mean_token_accuracy": 0.18246716856956482,
"num_tokens": 26705880.0,
"step": 11975
},
{
"entropy": 5.567191076278687,
"epoch": 1.3454821137754815,
"grad_norm": 1.7265625,
"learning_rate": 0.0004829488667526679,
"loss": 5.4731,
"mean_token_accuracy": 0.1866876259446144,
"num_tokens": 26716978.0,
"step": 11980
},
{
"entropy": 5.613245725631714,
"epoch": 1.3460436906834392,
"grad_norm": 1.3515625,
"learning_rate": 0.00048293353196138294,
"loss": 5.404,
"mean_token_accuracy": 0.19643960446119307,
"num_tokens": 26728380.0,
"step": 11985
},
{
"entropy": 5.493567848205567,
"epoch": 1.3466052675913966,
"grad_norm": 1.453125,
"learning_rate": 0.00048291819054941375,
"loss": 5.3661,
"mean_token_accuracy": 0.18693873286247253,
"num_tokens": 26739034.0,
"step": 11990
},
{
"entropy": 5.366666221618653,
"epoch": 1.347166844499354,
"grad_norm": 1.5,
"learning_rate": 0.0004829028425172487,
"loss": 5.1749,
"mean_token_accuracy": 0.19657352417707444,
"num_tokens": 26750123.0,
"step": 11995
},
{
"entropy": 5.472678327560425,
"epoch": 1.3477284214073117,
"grad_norm": 1.65625,
"learning_rate": 0.00048288748786537656,
"loss": 5.3465,
"mean_token_accuracy": 0.19450501650571822,
"num_tokens": 26760411.0,
"step": 12000
},
{
"epoch": 1.3477284214073117,
"eval_entropy": 5.362528239140229,
"eval_loss": 5.47659158706665,
"eval_mean_token_accuracy": 0.18887238511573048,
"eval_num_tokens": 26760411.0,
"eval_runtime": 23.7703,
"eval_samples_per_second": 1304.694,
"eval_steps_per_second": 163.103,
"step": 12000
}
],
"logging_steps": 5,
"max_steps": 89030,
"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": 4.1848956444672e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink