fpadovani/hin-deva-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed455
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
hin-deva-100mb-after-ppt-sh…/checkpoint-3000/trainer_state.json
ModelHub XC bc63150e88 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/hin-deva-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed455
Source: Original Platform
2026-09-07 05:28:19 +08:00

6046 lines
164 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.2881844380403458,
"eval_steps": 3000,
"global_step": 3000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.909158945083618,
"epoch": 0.0004803073967339097,
"grad_norm": 18.0,
"learning_rate": 2e-06,
"loss": 14.0493,
"mean_token_accuracy": 0.0,
"num_tokens": 10821.0,
"step": 5
},
{
"entropy": 4.897940301895142,
"epoch": 0.0009606147934678194,
"grad_norm": 23.0,
"learning_rate": 4.5e-06,
"loss": 14.0706,
"mean_token_accuracy": 0.0,
"num_tokens": 21838.0,
"step": 10
},
{
"entropy": 5.05525016784668,
"epoch": 0.001440922190201729,
"grad_norm": 26.25,
"learning_rate": 7e-06,
"loss": 13.7388,
"mean_token_accuracy": 0.0,
"num_tokens": 33483.0,
"step": 15
},
{
"entropy": 5.374046516418457,
"epoch": 0.0019212295869356388,
"grad_norm": 36.0,
"learning_rate": 9.5e-06,
"loss": 13.2353,
"mean_token_accuracy": 0.0,
"num_tokens": 45150.0,
"step": 20
},
{
"entropy": 6.746120595932007,
"epoch": 0.0024015369836695487,
"grad_norm": 26.25,
"learning_rate": 1.2e-05,
"loss": 12.2263,
"mean_token_accuracy": 0.0,
"num_tokens": 55862.0,
"step": 25
},
{
"entropy": 9.457833766937256,
"epoch": 0.002881844380403458,
"grad_norm": 4.59375,
"learning_rate": 1.4500000000000002e-05,
"loss": 11.1521,
"mean_token_accuracy": 0.0,
"num_tokens": 66766.0,
"step": 30
},
{
"entropy": 10.62428560256958,
"epoch": 0.0033621517771373678,
"grad_norm": 2.71875,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.7428,
"mean_token_accuracy": 0.0,
"num_tokens": 77012.0,
"step": 35
},
{
"entropy": 10.70099697113037,
"epoch": 0.0038424591738712775,
"grad_norm": 2.46875,
"learning_rate": 1.95e-05,
"loss": 10.5653,
"mean_token_accuracy": 0.0068274421093519775,
"num_tokens": 89034.0,
"step": 40
},
{
"entropy": 10.700504493713378,
"epoch": 0.004322766570605188,
"grad_norm": 2.484375,
"learning_rate": 2.2e-05,
"loss": 10.264,
"mean_token_accuracy": 0.03058484271168709,
"num_tokens": 100341.0,
"step": 45
},
{
"entropy": 10.682637310028076,
"epoch": 0.004803073967339097,
"grad_norm": 2.28125,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9901,
"mean_token_accuracy": 0.028485930524766445,
"num_tokens": 111997.0,
"step": 50
},
{
"entropy": 10.652040386199952,
"epoch": 0.005283381364073006,
"grad_norm": 2.1875,
"learning_rate": 2.7e-05,
"loss": 9.8134,
"mean_token_accuracy": 0.030870671570301055,
"num_tokens": 123705.0,
"step": 55
},
{
"entropy": 10.639647006988525,
"epoch": 0.005763688760806916,
"grad_norm": 2.125,
"learning_rate": 2.95e-05,
"loss": 9.6274,
"mean_token_accuracy": 0.029013283923268317,
"num_tokens": 134091.0,
"step": 60
},
{
"entropy": 10.625792407989502,
"epoch": 0.006243996157540826,
"grad_norm": 1.953125,
"learning_rate": 3.2e-05,
"loss": 9.5799,
"mean_token_accuracy": 0.035698648542165756,
"num_tokens": 145731.0,
"step": 65
},
{
"entropy": 10.611747932434081,
"epoch": 0.0067243035542747355,
"grad_norm": 2.109375,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.4796,
"mean_token_accuracy": 0.029607368633151054,
"num_tokens": 156979.0,
"step": 70
},
{
"entropy": 10.60178451538086,
"epoch": 0.007204610951008645,
"grad_norm": 2.015625,
"learning_rate": 3.7e-05,
"loss": 9.391,
"mean_token_accuracy": 0.03300900887697935,
"num_tokens": 167021.0,
"step": 75
},
{
"entropy": 10.588104724884033,
"epoch": 0.007684918347742555,
"grad_norm": 2.015625,
"learning_rate": 3.95e-05,
"loss": 9.3298,
"mean_token_accuracy": 0.03809518590569496,
"num_tokens": 179269.0,
"step": 80
},
{
"entropy": 10.573075771331787,
"epoch": 0.008165225744476465,
"grad_norm": 1.8828125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.2226,
"mean_token_accuracy": 0.034396625123918056,
"num_tokens": 190328.0,
"step": 85
},
{
"entropy": 10.551105213165282,
"epoch": 0.008645533141210375,
"grad_norm": 2.03125,
"learning_rate": 4.45e-05,
"loss": 9.0564,
"mean_token_accuracy": 0.032392379455268384,
"num_tokens": 201522.0,
"step": 90
},
{
"entropy": 10.511863613128662,
"epoch": 0.009125840537944284,
"grad_norm": 1.8671875,
"learning_rate": 4.7000000000000004e-05,
"loss": 8.995,
"mean_token_accuracy": 0.036259174533188346,
"num_tokens": 212963.0,
"step": 95
},
{
"entropy": 10.454308891296387,
"epoch": 0.009606147934678195,
"grad_norm": 1.578125,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.9701,
"mean_token_accuracy": 0.0323725325986743,
"num_tokens": 224446.0,
"step": 100
},
{
"entropy": 10.384486865997314,
"epoch": 0.010086455331412104,
"grad_norm": 1.765625,
"learning_rate": 5.2e-05,
"loss": 8.7698,
"mean_token_accuracy": 0.03222609367221594,
"num_tokens": 235334.0,
"step": 105
},
{
"entropy": 10.287021160125732,
"epoch": 0.010566762728146013,
"grad_norm": 1.5546875,
"learning_rate": 5.45e-05,
"loss": 8.6568,
"mean_token_accuracy": 0.033952686190605166,
"num_tokens": 247024.0,
"step": 110
},
{
"entropy": 10.163359928131104,
"epoch": 0.011047070124879923,
"grad_norm": 1.578125,
"learning_rate": 5.7e-05,
"loss": 8.4531,
"mean_token_accuracy": 0.03682863060384989,
"num_tokens": 258354.0,
"step": 115
},
{
"entropy": 10.03564281463623,
"epoch": 0.011527377521613832,
"grad_norm": 1.4296875,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.3556,
"mean_token_accuracy": 0.039085553586483,
"num_tokens": 271202.0,
"step": 120
},
{
"entropy": 9.844954490661621,
"epoch": 0.012007684918347743,
"grad_norm": 1.2890625,
"learning_rate": 6.2e-05,
"loss": 8.2264,
"mean_token_accuracy": 0.03886361531913281,
"num_tokens": 282405.0,
"step": 125
},
{
"entropy": 9.633767795562743,
"epoch": 0.012487992315081652,
"grad_norm": 1.203125,
"learning_rate": 6.450000000000001e-05,
"loss": 8.0588,
"mean_token_accuracy": 0.04458533097058535,
"num_tokens": 294211.0,
"step": 130
},
{
"entropy": 9.427376461029052,
"epoch": 0.012968299711815562,
"grad_norm": 1.1015625,
"learning_rate": 6.7e-05,
"loss": 8.0067,
"mean_token_accuracy": 0.044877220317721365,
"num_tokens": 305420.0,
"step": 135
},
{
"entropy": 9.113702964782714,
"epoch": 0.013448607108549471,
"grad_norm": 0.96875,
"learning_rate": 6.950000000000001e-05,
"loss": 7.8472,
"mean_token_accuracy": 0.04761340096592903,
"num_tokens": 316430.0,
"step": 140
},
{
"entropy": 8.837570667266846,
"epoch": 0.013928914505283382,
"grad_norm": 1.03125,
"learning_rate": 7.2e-05,
"loss": 7.7606,
"mean_token_accuracy": 0.05185309983789921,
"num_tokens": 328406.0,
"step": 145
},
{
"entropy": 8.593095207214356,
"epoch": 0.01440922190201729,
"grad_norm": 1.0,
"learning_rate": 7.45e-05,
"loss": 7.6877,
"mean_token_accuracy": 0.05381656400859356,
"num_tokens": 339051.0,
"step": 150
},
{
"entropy": 8.415284156799316,
"epoch": 0.014889529298751201,
"grad_norm": 1.0625,
"learning_rate": 7.7e-05,
"loss": 7.6235,
"mean_token_accuracy": 0.055242300406098364,
"num_tokens": 350308.0,
"step": 155
},
{
"entropy": 8.24738597869873,
"epoch": 0.01536983669548511,
"grad_norm": 0.80078125,
"learning_rate": 7.950000000000001e-05,
"loss": 7.5935,
"mean_token_accuracy": 0.054396551474928855,
"num_tokens": 362159.0,
"step": 160
},
{
"entropy": 8.126520538330078,
"epoch": 0.01585014409221902,
"grad_norm": 1.015625,
"learning_rate": 8.2e-05,
"loss": 7.5837,
"mean_token_accuracy": 0.05628910064697266,
"num_tokens": 375157.0,
"step": 165
},
{
"entropy": 8.08217649459839,
"epoch": 0.01633045148895293,
"grad_norm": 0.84375,
"learning_rate": 8.450000000000001e-05,
"loss": 7.4497,
"mean_token_accuracy": 0.06717152446508408,
"num_tokens": 387318.0,
"step": 170
},
{
"entropy": 8.051177406311036,
"epoch": 0.01681075888568684,
"grad_norm": 1.1796875,
"learning_rate": 8.7e-05,
"loss": 7.4622,
"mean_token_accuracy": 0.07725611329078674,
"num_tokens": 398933.0,
"step": 175
},
{
"entropy": 7.939669942855835,
"epoch": 0.01729106628242075,
"grad_norm": 0.9296875,
"learning_rate": 8.95e-05,
"loss": 7.4158,
"mean_token_accuracy": 0.0700771912932396,
"num_tokens": 410460.0,
"step": 180
},
{
"entropy": 7.9294243335723875,
"epoch": 0.01777137367915466,
"grad_norm": 1.1171875,
"learning_rate": 9.2e-05,
"loss": 7.39,
"mean_token_accuracy": 0.07831323072314263,
"num_tokens": 421894.0,
"step": 185
},
{
"entropy": 7.895280361175537,
"epoch": 0.01825168107588857,
"grad_norm": 0.953125,
"learning_rate": 9.45e-05,
"loss": 7.3806,
"mean_token_accuracy": 0.07451360076665878,
"num_tokens": 432976.0,
"step": 190
},
{
"entropy": 7.967076063156128,
"epoch": 0.018731988472622477,
"grad_norm": 1.015625,
"learning_rate": 9.7e-05,
"loss": 7.4475,
"mean_token_accuracy": 0.07888153940439224,
"num_tokens": 444859.0,
"step": 195
},
{
"entropy": 7.939164543151856,
"epoch": 0.01921229586935639,
"grad_norm": 1.390625,
"learning_rate": 9.95e-05,
"loss": 7.3302,
"mean_token_accuracy": 0.08403046280145646,
"num_tokens": 456508.0,
"step": 200
},
{
"entropy": 7.868430519104004,
"epoch": 0.0196926032660903,
"grad_norm": 1.125,
"learning_rate": 0.000102,
"loss": 7.3456,
"mean_token_accuracy": 0.07998873889446259,
"num_tokens": 469064.0,
"step": 205
},
{
"entropy": 7.868760108947754,
"epoch": 0.020172910662824207,
"grad_norm": 1.828125,
"learning_rate": 0.00010449999999999999,
"loss": 7.2899,
"mean_token_accuracy": 0.08207204826176166,
"num_tokens": 480743.0,
"step": 210
},
{
"entropy": 7.7555779933929445,
"epoch": 0.020653218059558116,
"grad_norm": 1.265625,
"learning_rate": 0.000107,
"loss": 7.3069,
"mean_token_accuracy": 0.08280007690191268,
"num_tokens": 493378.0,
"step": 215
},
{
"entropy": 7.839309644699097,
"epoch": 0.021133525456292025,
"grad_norm": 1.1171875,
"learning_rate": 0.0001095,
"loss": 7.2462,
"mean_token_accuracy": 0.0859088484197855,
"num_tokens": 505641.0,
"step": 220
},
{
"entropy": 7.805751466751099,
"epoch": 0.021613832853025938,
"grad_norm": 1.15625,
"learning_rate": 0.000112,
"loss": 7.3079,
"mean_token_accuracy": 0.0868067942559719,
"num_tokens": 517783.0,
"step": 225
},
{
"entropy": 7.756006574630737,
"epoch": 0.022094140249759846,
"grad_norm": 1.8828125,
"learning_rate": 0.0001145,
"loss": 7.1612,
"mean_token_accuracy": 0.09106182381510734,
"num_tokens": 528407.0,
"step": 230
},
{
"entropy": 7.80887017250061,
"epoch": 0.022574447646493755,
"grad_norm": 1.0859375,
"learning_rate": 0.00011700000000000001,
"loss": 7.3031,
"mean_token_accuracy": 0.08784078434109688,
"num_tokens": 539974.0,
"step": 235
},
{
"entropy": 7.641842365264893,
"epoch": 0.023054755043227664,
"grad_norm": 1.359375,
"learning_rate": 0.00011949999999999999,
"loss": 7.1391,
"mean_token_accuracy": 0.08666003718972207,
"num_tokens": 551127.0,
"step": 240
},
{
"entropy": 7.679393720626831,
"epoch": 0.023535062439961577,
"grad_norm": 1.2109375,
"learning_rate": 0.000122,
"loss": 7.1554,
"mean_token_accuracy": 0.09684930071234703,
"num_tokens": 561980.0,
"step": 245
},
{
"entropy": 7.723794889450073,
"epoch": 0.024015369836695485,
"grad_norm": 1.1640625,
"learning_rate": 0.0001245,
"loss": 7.1898,
"mean_token_accuracy": 0.0910400390625,
"num_tokens": 573840.0,
"step": 250
},
{
"entropy": 7.562839365005493,
"epoch": 0.024495677233429394,
"grad_norm": 1.3359375,
"learning_rate": 0.000127,
"loss": 7.0923,
"mean_token_accuracy": 0.09587239101529121,
"num_tokens": 585110.0,
"step": 255
},
{
"entropy": 7.556887006759643,
"epoch": 0.024975984630163303,
"grad_norm": 1.046875,
"learning_rate": 0.0001295,
"loss": 6.9932,
"mean_token_accuracy": 0.0963286928832531,
"num_tokens": 596902.0,
"step": 260
},
{
"entropy": 7.55956130027771,
"epoch": 0.025456292026897216,
"grad_norm": 1.265625,
"learning_rate": 0.000132,
"loss": 7.1566,
"mean_token_accuracy": 0.0950677141547203,
"num_tokens": 608738.0,
"step": 265
},
{
"entropy": 7.5788225650787355,
"epoch": 0.025936599423631124,
"grad_norm": 1.3984375,
"learning_rate": 0.00013450000000000002,
"loss": 7.0778,
"mean_token_accuracy": 0.0930832102894783,
"num_tokens": 621078.0,
"step": 270
},
{
"entropy": 7.6280433177948,
"epoch": 0.026416906820365033,
"grad_norm": 1.140625,
"learning_rate": 0.00013700000000000002,
"loss": 7.1451,
"mean_token_accuracy": 0.09341271966695786,
"num_tokens": 632333.0,
"step": 275
},
{
"entropy": 7.559076023101807,
"epoch": 0.026897214217098942,
"grad_norm": 1.578125,
"learning_rate": 0.0001395,
"loss": 6.9556,
"mean_token_accuracy": 0.10151683166623116,
"num_tokens": 644474.0,
"step": 280
},
{
"entropy": 7.4703763961792,
"epoch": 0.027377521613832854,
"grad_norm": 1.171875,
"learning_rate": 0.00014199999999999998,
"loss": 7.0352,
"mean_token_accuracy": 0.10359716191887855,
"num_tokens": 655608.0,
"step": 285
},
{
"entropy": 7.540377807617188,
"epoch": 0.027857829010566763,
"grad_norm": 1.15625,
"learning_rate": 0.0001445,
"loss": 7.0476,
"mean_token_accuracy": 0.09810357466340065,
"num_tokens": 667371.0,
"step": 290
},
{
"entropy": 7.448433876037598,
"epoch": 0.028338136407300672,
"grad_norm": 1.5625,
"learning_rate": 0.000147,
"loss": 7.0591,
"mean_token_accuracy": 0.09780933484435081,
"num_tokens": 679152.0,
"step": 295
},
{
"entropy": 7.4532647132873535,
"epoch": 0.02881844380403458,
"grad_norm": 1.2890625,
"learning_rate": 0.0001495,
"loss": 6.9632,
"mean_token_accuracy": 0.10561580210924149,
"num_tokens": 690359.0,
"step": 300
},
{
"entropy": 7.422236824035645,
"epoch": 0.029298751200768493,
"grad_norm": 1.3984375,
"learning_rate": 0.000152,
"loss": 6.9359,
"mean_token_accuracy": 0.10709809437394142,
"num_tokens": 701819.0,
"step": 305
},
{
"entropy": 7.465102815628052,
"epoch": 0.029779058597502402,
"grad_norm": 1.3046875,
"learning_rate": 0.00015450000000000001,
"loss": 6.9961,
"mean_token_accuracy": 0.10205374658107758,
"num_tokens": 711048.0,
"step": 310
},
{
"entropy": 7.381311082839966,
"epoch": 0.03025936599423631,
"grad_norm": 1.421875,
"learning_rate": 0.000157,
"loss": 6.8809,
"mean_token_accuracy": 0.10968201458454133,
"num_tokens": 721399.0,
"step": 315
},
{
"entropy": 7.40810341835022,
"epoch": 0.03073967339097022,
"grad_norm": 1.171875,
"learning_rate": 0.0001595,
"loss": 6.891,
"mean_token_accuracy": 0.10995450168848038,
"num_tokens": 732144.0,
"step": 320
},
{
"entropy": 7.357445764541626,
"epoch": 0.03121998078770413,
"grad_norm": 1.5859375,
"learning_rate": 0.000162,
"loss": 6.8626,
"mean_token_accuracy": 0.10403823405504227,
"num_tokens": 743962.0,
"step": 325
},
{
"entropy": 7.388841199874878,
"epoch": 0.03170028818443804,
"grad_norm": 1.640625,
"learning_rate": 0.00016450000000000001,
"loss": 7.0076,
"mean_token_accuracy": 0.10685212463140488,
"num_tokens": 755641.0,
"step": 330
},
{
"entropy": 7.286140584945679,
"epoch": 0.03218059558117195,
"grad_norm": 1.3359375,
"learning_rate": 0.00016700000000000002,
"loss": 6.8385,
"mean_token_accuracy": 0.1151984341442585,
"num_tokens": 766798.0,
"step": 335
},
{
"entropy": 7.339591550827026,
"epoch": 0.03266090297790586,
"grad_norm": 1.265625,
"learning_rate": 0.00016950000000000003,
"loss": 6.9311,
"mean_token_accuracy": 0.10881740972399712,
"num_tokens": 778097.0,
"step": 340
},
{
"entropy": 7.252063226699829,
"epoch": 0.03314121037463977,
"grad_norm": 1.21875,
"learning_rate": 0.00017199999999999998,
"loss": 6.8496,
"mean_token_accuracy": 0.11310645863413811,
"num_tokens": 789775.0,
"step": 345
},
{
"entropy": 7.39683985710144,
"epoch": 0.03362151777137368,
"grad_norm": 1.96875,
"learning_rate": 0.00017449999999999999,
"loss": 6.9062,
"mean_token_accuracy": 0.10776558816432953,
"num_tokens": 800155.0,
"step": 350
},
{
"entropy": 7.379228162765503,
"epoch": 0.034101825168107586,
"grad_norm": 1.9765625,
"learning_rate": 0.000177,
"loss": 6.9239,
"mean_token_accuracy": 0.11197539716959,
"num_tokens": 811213.0,
"step": 355
},
{
"entropy": 7.30075511932373,
"epoch": 0.0345821325648415,
"grad_norm": 1.1953125,
"learning_rate": 0.0001795,
"loss": 6.903,
"mean_token_accuracy": 0.10408640876412392,
"num_tokens": 822832.0,
"step": 360
},
{
"entropy": 7.287055206298828,
"epoch": 0.03506243996157541,
"grad_norm": 1.5625,
"learning_rate": 0.000182,
"loss": 6.8727,
"mean_token_accuracy": 0.11483326479792595,
"num_tokens": 834824.0,
"step": 365
},
{
"entropy": 7.292969608306885,
"epoch": 0.03554274735830932,
"grad_norm": 1.2578125,
"learning_rate": 0.0001845,
"loss": 6.9053,
"mean_token_accuracy": 0.10474938601255417,
"num_tokens": 846969.0,
"step": 370
},
{
"entropy": 7.357876491546631,
"epoch": 0.03602305475504323,
"grad_norm": 1.2578125,
"learning_rate": 0.000187,
"loss": 6.9773,
"mean_token_accuracy": 0.10199039503931999,
"num_tokens": 859271.0,
"step": 375
},
{
"entropy": 7.247849559783935,
"epoch": 0.03650336215177714,
"grad_norm": 1.265625,
"learning_rate": 0.0001895,
"loss": 6.8723,
"mean_token_accuracy": 0.10891828313469887,
"num_tokens": 869754.0,
"step": 380
},
{
"entropy": 7.2068620204925535,
"epoch": 0.036983669548511046,
"grad_norm": 1.6484375,
"learning_rate": 0.000192,
"loss": 6.819,
"mean_token_accuracy": 0.11546091064810753,
"num_tokens": 880519.0,
"step": 385
},
{
"entropy": 7.166879415512085,
"epoch": 0.037463976945244955,
"grad_norm": 1.4140625,
"learning_rate": 0.0001945,
"loss": 6.775,
"mean_token_accuracy": 0.10799357295036316,
"num_tokens": 890765.0,
"step": 390
},
{
"entropy": 7.240225791931152,
"epoch": 0.037944284341978864,
"grad_norm": 1.4453125,
"learning_rate": 0.00019700000000000002,
"loss": 6.7757,
"mean_token_accuracy": 0.10960216820240021,
"num_tokens": 901717.0,
"step": 395
},
{
"entropy": 7.299621725082398,
"epoch": 0.03842459173871278,
"grad_norm": 1.4375,
"learning_rate": 0.00019950000000000002,
"loss": 6.8428,
"mean_token_accuracy": 0.11034838110208511,
"num_tokens": 913334.0,
"step": 400
},
{
"entropy": 7.204387092590332,
"epoch": 0.03890489913544669,
"grad_norm": 1.5234375,
"learning_rate": 0.000202,
"loss": 6.8557,
"mean_token_accuracy": 0.11134258806705474,
"num_tokens": 924753.0,
"step": 405
},
{
"entropy": 7.204525709152222,
"epoch": 0.0393852065321806,
"grad_norm": 1.421875,
"learning_rate": 0.00020449999999999998,
"loss": 6.7857,
"mean_token_accuracy": 0.11340023577213287,
"num_tokens": 935692.0,
"step": 410
},
{
"entropy": 7.154750633239746,
"epoch": 0.039865513928914506,
"grad_norm": 1.2265625,
"learning_rate": 0.000207,
"loss": 6.8747,
"mean_token_accuracy": 0.11008304730057716,
"num_tokens": 947902.0,
"step": 415
},
{
"entropy": 7.122686529159546,
"epoch": 0.040345821325648415,
"grad_norm": 1.3515625,
"learning_rate": 0.0002095,
"loss": 6.8128,
"mean_token_accuracy": 0.11032308787107467,
"num_tokens": 959437.0,
"step": 420
},
{
"entropy": 7.040357637405395,
"epoch": 0.040826128722382324,
"grad_norm": 1.65625,
"learning_rate": 0.000212,
"loss": 6.6189,
"mean_token_accuracy": 0.12019914388656616,
"num_tokens": 970191.0,
"step": 425
},
{
"entropy": 7.0764172077178955,
"epoch": 0.04130643611911623,
"grad_norm": 1.1875,
"learning_rate": 0.0002145,
"loss": 6.7458,
"mean_token_accuracy": 0.11545630767941475,
"num_tokens": 981811.0,
"step": 430
},
{
"entropy": 7.159797048568725,
"epoch": 0.04178674351585014,
"grad_norm": 1.125,
"learning_rate": 0.00021700000000000002,
"loss": 6.8529,
"mean_token_accuracy": 0.10769774615764618,
"num_tokens": 994034.0,
"step": 435
},
{
"entropy": 7.207771253585816,
"epoch": 0.04226705091258405,
"grad_norm": 1.5546875,
"learning_rate": 0.0002195,
"loss": 6.7989,
"mean_token_accuracy": 0.11341427490115166,
"num_tokens": 1005506.0,
"step": 440
},
{
"entropy": 7.045540857315063,
"epoch": 0.042747358309317966,
"grad_norm": 1.7421875,
"learning_rate": 0.000222,
"loss": 6.7647,
"mean_token_accuracy": 0.10924506187438965,
"num_tokens": 1017103.0,
"step": 445
},
{
"entropy": 7.124048948287964,
"epoch": 0.043227665706051875,
"grad_norm": 1.390625,
"learning_rate": 0.0002245,
"loss": 6.8523,
"mean_token_accuracy": 0.10999714508652687,
"num_tokens": 1029606.0,
"step": 450
},
{
"entropy": 7.184269189834595,
"epoch": 0.043707973102785784,
"grad_norm": 1.8046875,
"learning_rate": 0.00022700000000000002,
"loss": 6.7961,
"mean_token_accuracy": 0.11004141569137574,
"num_tokens": 1041899.0,
"step": 455
},
{
"entropy": 7.053015661239624,
"epoch": 0.04418828049951969,
"grad_norm": 1.3359375,
"learning_rate": 0.00022950000000000002,
"loss": 6.7001,
"mean_token_accuracy": 0.11952895596623421,
"num_tokens": 1053577.0,
"step": 460
},
{
"entropy": 7.075217962265015,
"epoch": 0.0446685878962536,
"grad_norm": 1.1875,
"learning_rate": 0.00023200000000000003,
"loss": 6.7525,
"mean_token_accuracy": 0.12150197252631187,
"num_tokens": 1065541.0,
"step": 465
},
{
"entropy": 7.141125965118408,
"epoch": 0.04514889529298751,
"grad_norm": 1.7421875,
"learning_rate": 0.00023449999999999998,
"loss": 6.7007,
"mean_token_accuracy": 0.11925816610455513,
"num_tokens": 1076138.0,
"step": 470
},
{
"entropy": 7.022414493560791,
"epoch": 0.04562920268972142,
"grad_norm": 1.203125,
"learning_rate": 0.000237,
"loss": 6.7675,
"mean_token_accuracy": 0.11674837097525596,
"num_tokens": 1087620.0,
"step": 475
},
{
"entropy": 7.041283512115479,
"epoch": 0.04610951008645533,
"grad_norm": 1.1796875,
"learning_rate": 0.0002395,
"loss": 6.6396,
"mean_token_accuracy": 0.12078191563487053,
"num_tokens": 1099493.0,
"step": 480
},
{
"entropy": 7.031822586059571,
"epoch": 0.046589817483189244,
"grad_norm": 1.359375,
"learning_rate": 0.000242,
"loss": 6.7,
"mean_token_accuracy": 0.11970095857977867,
"num_tokens": 1110559.0,
"step": 485
},
{
"entropy": 6.978209018707275,
"epoch": 0.04707012487992315,
"grad_norm": 1.21875,
"learning_rate": 0.0002445,
"loss": 6.6632,
"mean_token_accuracy": 0.12008017003536224,
"num_tokens": 1121051.0,
"step": 490
},
{
"entropy": 6.980980300903321,
"epoch": 0.04755043227665706,
"grad_norm": 1.203125,
"learning_rate": 0.000247,
"loss": 6.6395,
"mean_token_accuracy": 0.12107920199632645,
"num_tokens": 1132305.0,
"step": 495
},
{
"entropy": 7.038042116165161,
"epoch": 0.04803073967339097,
"grad_norm": 1.2890625,
"learning_rate": 0.0002495,
"loss": 6.6505,
"mean_token_accuracy": 0.1215340994298458,
"num_tokens": 1142979.0,
"step": 500
},
{
"entropy": 6.921855020523071,
"epoch": 0.04851104707012488,
"grad_norm": 1.5859375,
"learning_rate": 0.000252,
"loss": 6.646,
"mean_token_accuracy": 0.12661107257008553,
"num_tokens": 1153889.0,
"step": 505
},
{
"entropy": 6.98356146812439,
"epoch": 0.04899135446685879,
"grad_norm": 1.265625,
"learning_rate": 0.0002545,
"loss": 6.6417,
"mean_token_accuracy": 0.1176763840019703,
"num_tokens": 1166030.0,
"step": 510
},
{
"entropy": 7.046093416213989,
"epoch": 0.0494716618635927,
"grad_norm": 1.328125,
"learning_rate": 0.000257,
"loss": 6.8173,
"mean_token_accuracy": 0.11064628511667252,
"num_tokens": 1177545.0,
"step": 515
},
{
"entropy": 7.001708602905273,
"epoch": 0.049951969260326606,
"grad_norm": 1.34375,
"learning_rate": 0.0002595,
"loss": 6.6353,
"mean_token_accuracy": 0.12478243634104728,
"num_tokens": 1188944.0,
"step": 520
},
{
"entropy": 6.840310907363891,
"epoch": 0.05043227665706052,
"grad_norm": 1.2421875,
"learning_rate": 0.000262,
"loss": 6.696,
"mean_token_accuracy": 0.1172410435974598,
"num_tokens": 1199517.0,
"step": 525
},
{
"entropy": 7.022114849090576,
"epoch": 0.05091258405379443,
"grad_norm": 1.4921875,
"learning_rate": 0.00026450000000000003,
"loss": 6.6212,
"mean_token_accuracy": 0.11661999747157097,
"num_tokens": 1211536.0,
"step": 530
},
{
"entropy": 6.830185031890869,
"epoch": 0.05139289145052834,
"grad_norm": 1.3125,
"learning_rate": 0.00026700000000000004,
"loss": 6.5521,
"mean_token_accuracy": 0.12921451553702354,
"num_tokens": 1224159.0,
"step": 535
},
{
"entropy": 6.94548978805542,
"epoch": 0.05187319884726225,
"grad_norm": 1.2578125,
"learning_rate": 0.00026950000000000005,
"loss": 6.614,
"mean_token_accuracy": 0.1301414117217064,
"num_tokens": 1235226.0,
"step": 540
},
{
"entropy": 6.777507877349853,
"epoch": 0.05235350624399616,
"grad_norm": 1.296875,
"learning_rate": 0.00027200000000000005,
"loss": 6.5321,
"mean_token_accuracy": 0.13609066978096962,
"num_tokens": 1245862.0,
"step": 545
},
{
"entropy": 6.856529331207275,
"epoch": 0.052833813640730067,
"grad_norm": 1.2421875,
"learning_rate": 0.0002745,
"loss": 6.5508,
"mean_token_accuracy": 0.1289893075823784,
"num_tokens": 1257512.0,
"step": 550
},
{
"entropy": 6.894668006896973,
"epoch": 0.053314121037463975,
"grad_norm": 1.2265625,
"learning_rate": 0.000277,
"loss": 6.6539,
"mean_token_accuracy": 0.1206201434135437,
"num_tokens": 1269317.0,
"step": 555
},
{
"entropy": 6.811044359207154,
"epoch": 0.053794428434197884,
"grad_norm": 1.4765625,
"learning_rate": 0.0002795,
"loss": 6.551,
"mean_token_accuracy": 0.11646598875522614,
"num_tokens": 1281878.0,
"step": 560
},
{
"entropy": 6.738573932647705,
"epoch": 0.05427473583093179,
"grad_norm": 1.1953125,
"learning_rate": 0.00028199999999999997,
"loss": 6.5612,
"mean_token_accuracy": 0.12958591282367707,
"num_tokens": 1293178.0,
"step": 565
},
{
"entropy": 6.851106452941894,
"epoch": 0.05475504322766571,
"grad_norm": 1.3359375,
"learning_rate": 0.0002845,
"loss": 6.5421,
"mean_token_accuracy": 0.12348497658967972,
"num_tokens": 1304969.0,
"step": 570
},
{
"entropy": 6.937162446975708,
"epoch": 0.05523535062439962,
"grad_norm": 1.34375,
"learning_rate": 0.000287,
"loss": 6.5765,
"mean_token_accuracy": 0.12681149393320085,
"num_tokens": 1317301.0,
"step": 575
},
{
"entropy": 6.686863088607788,
"epoch": 0.05571565802113353,
"grad_norm": 1.1875,
"learning_rate": 0.0002895,
"loss": 6.5505,
"mean_token_accuracy": 0.12601301521062852,
"num_tokens": 1327688.0,
"step": 580
},
{
"entropy": 6.901493978500366,
"epoch": 0.056195965417867436,
"grad_norm": 1.5703125,
"learning_rate": 0.000292,
"loss": 6.4949,
"mean_token_accuracy": 0.1258933037519455,
"num_tokens": 1338750.0,
"step": 585
},
{
"entropy": 6.832601118087768,
"epoch": 0.056676272814601344,
"grad_norm": 1.1328125,
"learning_rate": 0.0002945,
"loss": 6.6508,
"mean_token_accuracy": 0.1182264193892479,
"num_tokens": 1350899.0,
"step": 590
},
{
"entropy": 6.818350124359131,
"epoch": 0.05715658021133525,
"grad_norm": 1.7890625,
"learning_rate": 0.000297,
"loss": 6.5006,
"mean_token_accuracy": 0.1271736852824688,
"num_tokens": 1363184.0,
"step": 595
},
{
"entropy": 6.7492002010345455,
"epoch": 0.05763688760806916,
"grad_norm": 1.1875,
"learning_rate": 0.0002995,
"loss": 6.4969,
"mean_token_accuracy": 0.12888728752732276,
"num_tokens": 1374192.0,
"step": 600
},
{
"entropy": 6.7368873119354244,
"epoch": 0.05811719500480307,
"grad_norm": 1.3671875,
"learning_rate": 0.000302,
"loss": 6.5138,
"mean_token_accuracy": 0.12501192465424538,
"num_tokens": 1384888.0,
"step": 605
},
{
"entropy": 6.81034517288208,
"epoch": 0.05859750240153699,
"grad_norm": 1.578125,
"learning_rate": 0.0003045,
"loss": 6.6139,
"mean_token_accuracy": 0.12904802039265634,
"num_tokens": 1395139.0,
"step": 610
},
{
"entropy": 6.729338312149048,
"epoch": 0.059077809798270896,
"grad_norm": 1.3515625,
"learning_rate": 0.000307,
"loss": 6.4134,
"mean_token_accuracy": 0.12848665565252304,
"num_tokens": 1407219.0,
"step": 615
},
{
"entropy": 6.779170131683349,
"epoch": 0.059558117195004805,
"grad_norm": 1.125,
"learning_rate": 0.0003095,
"loss": 6.5328,
"mean_token_accuracy": 0.12432346194982528,
"num_tokens": 1418917.0,
"step": 620
},
{
"entropy": 6.744690942764282,
"epoch": 0.060038424591738714,
"grad_norm": 1.6015625,
"learning_rate": 0.000312,
"loss": 6.4838,
"mean_token_accuracy": 0.12512711808085442,
"num_tokens": 1430792.0,
"step": 625
},
{
"entropy": 6.667728471755981,
"epoch": 0.06051873198847262,
"grad_norm": 1.1875,
"learning_rate": 0.0003145,
"loss": 6.504,
"mean_token_accuracy": 0.13070958107709885,
"num_tokens": 1442825.0,
"step": 630
},
{
"entropy": 6.764918041229248,
"epoch": 0.06099903938520653,
"grad_norm": 1.2109375,
"learning_rate": 0.000317,
"loss": 6.5837,
"mean_token_accuracy": 0.1336795650422573,
"num_tokens": 1454078.0,
"step": 635
},
{
"entropy": 6.783257913589478,
"epoch": 0.06147934678194044,
"grad_norm": 1.453125,
"learning_rate": 0.0003195,
"loss": 6.5293,
"mean_token_accuracy": 0.12106153219938279,
"num_tokens": 1465164.0,
"step": 640
},
{
"entropy": 6.80000376701355,
"epoch": 0.06195965417867435,
"grad_norm": 1.09375,
"learning_rate": 0.000322,
"loss": 6.5039,
"mean_token_accuracy": 0.13234740868210793,
"num_tokens": 1476813.0,
"step": 645
},
{
"entropy": 6.68659987449646,
"epoch": 0.06243996157540826,
"grad_norm": 1.1796875,
"learning_rate": 0.00032450000000000003,
"loss": 6.5251,
"mean_token_accuracy": 0.1282544143497944,
"num_tokens": 1488547.0,
"step": 650
},
{
"entropy": 6.814279413223266,
"epoch": 0.06292026897214217,
"grad_norm": 1.109375,
"learning_rate": 0.00032700000000000003,
"loss": 6.531,
"mean_token_accuracy": 0.12601352855563164,
"num_tokens": 1500714.0,
"step": 655
},
{
"entropy": 6.687133073806763,
"epoch": 0.06340057636887608,
"grad_norm": 1.1328125,
"learning_rate": 0.00032950000000000004,
"loss": 6.5182,
"mean_token_accuracy": 0.13100821301341056,
"num_tokens": 1512212.0,
"step": 660
},
{
"entropy": 6.712434768676758,
"epoch": 0.06388088376560999,
"grad_norm": 1.203125,
"learning_rate": 0.00033200000000000005,
"loss": 6.4878,
"mean_token_accuracy": 0.13093362599611283,
"num_tokens": 1523610.0,
"step": 665
},
{
"entropy": 6.596017265319825,
"epoch": 0.0643611911623439,
"grad_norm": 1.125,
"learning_rate": 0.00033450000000000005,
"loss": 6.4492,
"mean_token_accuracy": 0.12683645486831666,
"num_tokens": 1535543.0,
"step": 670
},
{
"entropy": 6.848142671585083,
"epoch": 0.06484149855907781,
"grad_norm": 1.203125,
"learning_rate": 0.000337,
"loss": 6.6626,
"mean_token_accuracy": 0.12445207461714744,
"num_tokens": 1547183.0,
"step": 675
},
{
"entropy": 6.626632308959961,
"epoch": 0.06532180595581172,
"grad_norm": 1.2578125,
"learning_rate": 0.0003395,
"loss": 6.4179,
"mean_token_accuracy": 0.136246507614851,
"num_tokens": 1559265.0,
"step": 680
},
{
"entropy": 6.692934179306031,
"epoch": 0.06580211335254563,
"grad_norm": 1.1875,
"learning_rate": 0.000342,
"loss": 6.4384,
"mean_token_accuracy": 0.1327574633061886,
"num_tokens": 1569774.0,
"step": 685
},
{
"entropy": 6.530060815811157,
"epoch": 0.06628242074927954,
"grad_norm": 1.25,
"learning_rate": 0.00034449999999999997,
"loss": 6.3719,
"mean_token_accuracy": 0.1348055936396122,
"num_tokens": 1579408.0,
"step": 690
},
{
"entropy": 6.638763380050659,
"epoch": 0.06676272814601344,
"grad_norm": 1.2109375,
"learning_rate": 0.000347,
"loss": 6.4333,
"mean_token_accuracy": 0.1340454787015915,
"num_tokens": 1591849.0,
"step": 695
},
{
"entropy": 6.610886335372925,
"epoch": 0.06724303554274735,
"grad_norm": 1.1015625,
"learning_rate": 0.0003495,
"loss": 6.5007,
"mean_token_accuracy": 0.12831750214099885,
"num_tokens": 1603269.0,
"step": 700
},
{
"entropy": 6.77165641784668,
"epoch": 0.06772334293948126,
"grad_norm": 1.2421875,
"learning_rate": 0.000352,
"loss": 6.5134,
"mean_token_accuracy": 0.1261325843632221,
"num_tokens": 1613639.0,
"step": 705
},
{
"entropy": 6.6583836555480955,
"epoch": 0.06820365033621517,
"grad_norm": 1.375,
"learning_rate": 0.0003545,
"loss": 6.542,
"mean_token_accuracy": 0.12910353168845176,
"num_tokens": 1625393.0,
"step": 710
},
{
"entropy": 6.6097071170806885,
"epoch": 0.0686839577329491,
"grad_norm": 1.265625,
"learning_rate": 0.000357,
"loss": 6.3672,
"mean_token_accuracy": 0.13228188902139665,
"num_tokens": 1636355.0,
"step": 715
},
{
"entropy": 6.604225015640258,
"epoch": 0.069164265129683,
"grad_norm": 1.171875,
"learning_rate": 0.0003595,
"loss": 6.409,
"mean_token_accuracy": 0.1348489813506603,
"num_tokens": 1646157.0,
"step": 720
},
{
"entropy": 6.58693528175354,
"epoch": 0.06964457252641691,
"grad_norm": 1.28125,
"learning_rate": 0.000362,
"loss": 6.4101,
"mean_token_accuracy": 0.12884832546114922,
"num_tokens": 1657827.0,
"step": 725
},
{
"entropy": 6.573770189285279,
"epoch": 0.07012487992315082,
"grad_norm": 1.5859375,
"learning_rate": 0.0003645,
"loss": 6.4863,
"mean_token_accuracy": 0.13203563466668128,
"num_tokens": 1669023.0,
"step": 730
},
{
"entropy": 6.614386749267578,
"epoch": 0.07060518731988473,
"grad_norm": 1.2265625,
"learning_rate": 0.000367,
"loss": 6.4313,
"mean_token_accuracy": 0.1316665753722191,
"num_tokens": 1680048.0,
"step": 735
},
{
"entropy": 6.710096597671509,
"epoch": 0.07108549471661864,
"grad_norm": 1.1953125,
"learning_rate": 0.0003695,
"loss": 6.6213,
"mean_token_accuracy": 0.11867887079715729,
"num_tokens": 1690907.0,
"step": 740
},
{
"entropy": 6.723827362060547,
"epoch": 0.07156580211335255,
"grad_norm": 1.1875,
"learning_rate": 0.000372,
"loss": 6.4584,
"mean_token_accuracy": 0.13038890957832336,
"num_tokens": 1702337.0,
"step": 745
},
{
"entropy": 6.607548427581787,
"epoch": 0.07204610951008646,
"grad_norm": 1.3671875,
"learning_rate": 0.0003745,
"loss": 6.3811,
"mean_token_accuracy": 0.1292431376874447,
"num_tokens": 1713242.0,
"step": 750
},
{
"entropy": 6.5345042705535885,
"epoch": 0.07252641690682037,
"grad_norm": 1.4765625,
"learning_rate": 0.000377,
"loss": 6.3688,
"mean_token_accuracy": 0.13297409936785698,
"num_tokens": 1724939.0,
"step": 755
},
{
"entropy": 6.485815048217773,
"epoch": 0.07300672430355427,
"grad_norm": 1.1171875,
"learning_rate": 0.0003795,
"loss": 6.3247,
"mean_token_accuracy": 0.14120787829160691,
"num_tokens": 1737311.0,
"step": 760
},
{
"entropy": 6.638444805145264,
"epoch": 0.07348703170028818,
"grad_norm": 1.171875,
"learning_rate": 0.000382,
"loss": 6.4629,
"mean_token_accuracy": 0.13194502517580986,
"num_tokens": 1749528.0,
"step": 765
},
{
"entropy": 6.529104328155517,
"epoch": 0.07396733909702209,
"grad_norm": 1.1171875,
"learning_rate": 0.0003845,
"loss": 6.3293,
"mean_token_accuracy": 0.13749989122152328,
"num_tokens": 1760304.0,
"step": 770
},
{
"entropy": 6.502080202102661,
"epoch": 0.074447646493756,
"grad_norm": 1.2109375,
"learning_rate": 0.00038700000000000003,
"loss": 6.3039,
"mean_token_accuracy": 0.13969450294971467,
"num_tokens": 1772014.0,
"step": 775
},
{
"entropy": 6.4714720249176025,
"epoch": 0.07492795389048991,
"grad_norm": 1.015625,
"learning_rate": 0.00038950000000000003,
"loss": 6.3781,
"mean_token_accuracy": 0.13299143686890602,
"num_tokens": 1784585.0,
"step": 780
},
{
"entropy": 6.529697895050049,
"epoch": 0.07540826128722382,
"grad_norm": 1.1640625,
"learning_rate": 0.00039200000000000004,
"loss": 6.3155,
"mean_token_accuracy": 0.13666464909911155,
"num_tokens": 1795246.0,
"step": 785
},
{
"entropy": 6.44914927482605,
"epoch": 0.07588856868395773,
"grad_norm": 1.265625,
"learning_rate": 0.00039450000000000005,
"loss": 6.2962,
"mean_token_accuracy": 0.13686830475926398,
"num_tokens": 1806826.0,
"step": 790
},
{
"entropy": 6.512354946136474,
"epoch": 0.07636887608069164,
"grad_norm": 1.1484375,
"learning_rate": 0.00039700000000000005,
"loss": 6.4276,
"mean_token_accuracy": 0.13477274253964425,
"num_tokens": 1818766.0,
"step": 795
},
{
"entropy": 6.510392951965332,
"epoch": 0.07684918347742556,
"grad_norm": 1.21875,
"learning_rate": 0.0003995,
"loss": 6.2729,
"mean_token_accuracy": 0.14873504042625427,
"num_tokens": 1830824.0,
"step": 800
},
{
"entropy": 6.404947662353516,
"epoch": 0.07732949087415947,
"grad_norm": 1.2578125,
"learning_rate": 0.000402,
"loss": 6.3783,
"mean_token_accuracy": 0.13591941818594933,
"num_tokens": 1841792.0,
"step": 805
},
{
"entropy": 6.476539421081543,
"epoch": 0.07780979827089338,
"grad_norm": 1.1328125,
"learning_rate": 0.0004045,
"loss": 6.3281,
"mean_token_accuracy": 0.1313629515469074,
"num_tokens": 1854683.0,
"step": 810
},
{
"entropy": 6.562305116653443,
"epoch": 0.07829010566762729,
"grad_norm": 1.25,
"learning_rate": 0.00040699999999999997,
"loss": 6.4129,
"mean_token_accuracy": 0.13369727060198783,
"num_tokens": 1865826.0,
"step": 815
},
{
"entropy": 6.482418060302734,
"epoch": 0.0787704130643612,
"grad_norm": 1.1875,
"learning_rate": 0.0004095,
"loss": 6.3598,
"mean_token_accuracy": 0.14122529625892638,
"num_tokens": 1876759.0,
"step": 820
},
{
"entropy": 6.471727514266968,
"epoch": 0.0792507204610951,
"grad_norm": 1.03125,
"learning_rate": 0.000412,
"loss": 6.4061,
"mean_token_accuracy": 0.13444310426712036,
"num_tokens": 1888013.0,
"step": 825
},
{
"entropy": 6.579637718200684,
"epoch": 0.07973102785782901,
"grad_norm": 1.140625,
"learning_rate": 0.0004145,
"loss": 6.4009,
"mean_token_accuracy": 0.14101578444242477,
"num_tokens": 1898964.0,
"step": 830
},
{
"entropy": 6.426772117614746,
"epoch": 0.08021133525456292,
"grad_norm": 1.171875,
"learning_rate": 0.000417,
"loss": 6.3216,
"mean_token_accuracy": 0.14054880663752556,
"num_tokens": 1909728.0,
"step": 835
},
{
"entropy": 6.411238527297973,
"epoch": 0.08069164265129683,
"grad_norm": 1.1875,
"learning_rate": 0.0004195,
"loss": 6.2457,
"mean_token_accuracy": 0.14817485064268113,
"num_tokens": 1919895.0,
"step": 840
},
{
"entropy": 6.340784978866577,
"epoch": 0.08117195004803074,
"grad_norm": 1.2734375,
"learning_rate": 0.000422,
"loss": 6.1755,
"mean_token_accuracy": 0.1461382269859314,
"num_tokens": 1930839.0,
"step": 845
},
{
"entropy": 6.435613965988159,
"epoch": 0.08165225744476465,
"grad_norm": 1.1328125,
"learning_rate": 0.0004245,
"loss": 6.3812,
"mean_token_accuracy": 0.13786052912473679,
"num_tokens": 1942605.0,
"step": 850
},
{
"entropy": 6.454065656661987,
"epoch": 0.08213256484149856,
"grad_norm": 1.1328125,
"learning_rate": 0.000427,
"loss": 6.3534,
"mean_token_accuracy": 0.13335032984614373,
"num_tokens": 1954863.0,
"step": 855
},
{
"entropy": 6.487230014801026,
"epoch": 0.08261287223823247,
"grad_norm": 1.265625,
"learning_rate": 0.0004295,
"loss": 6.4437,
"mean_token_accuracy": 0.13347432389855385,
"num_tokens": 1966088.0,
"step": 860
},
{
"entropy": 6.391985607147217,
"epoch": 0.08309317963496637,
"grad_norm": 1.171875,
"learning_rate": 0.000432,
"loss": 6.2804,
"mean_token_accuracy": 0.13734670206904412,
"num_tokens": 1978081.0,
"step": 865
},
{
"entropy": 6.432765245437622,
"epoch": 0.08357348703170028,
"grad_norm": 1.0859375,
"learning_rate": 0.0004345,
"loss": 6.3338,
"mean_token_accuracy": 0.1310129150748253,
"num_tokens": 1991212.0,
"step": 870
},
{
"entropy": 6.510468673706055,
"epoch": 0.08405379442843419,
"grad_norm": 1.2109375,
"learning_rate": 0.000437,
"loss": 6.3222,
"mean_token_accuracy": 0.14048025831580163,
"num_tokens": 2002074.0,
"step": 875
},
{
"entropy": 6.308026313781738,
"epoch": 0.0845341018251681,
"grad_norm": 1.3046875,
"learning_rate": 0.0004395,
"loss": 6.256,
"mean_token_accuracy": 0.14530111104249954,
"num_tokens": 2013039.0,
"step": 880
},
{
"entropy": 6.424022388458252,
"epoch": 0.08501440922190202,
"grad_norm": 0.984375,
"learning_rate": 0.000442,
"loss": 6.3833,
"mean_token_accuracy": 0.13611895516514777,
"num_tokens": 2024744.0,
"step": 885
},
{
"entropy": 6.35896577835083,
"epoch": 0.08549471661863593,
"grad_norm": 1.1796875,
"learning_rate": 0.0004445,
"loss": 6.2463,
"mean_token_accuracy": 0.14010558724403382,
"num_tokens": 2035587.0,
"step": 890
},
{
"entropy": 6.4153601169586185,
"epoch": 0.08597502401536984,
"grad_norm": 1.1015625,
"learning_rate": 0.000447,
"loss": 6.3276,
"mean_token_accuracy": 0.1342465154826641,
"num_tokens": 2047924.0,
"step": 895
},
{
"entropy": 6.351702404022217,
"epoch": 0.08645533141210375,
"grad_norm": 0.984375,
"learning_rate": 0.00044950000000000003,
"loss": 6.2714,
"mean_token_accuracy": 0.14150312915444374,
"num_tokens": 2060161.0,
"step": 900
},
{
"entropy": 6.34501953125,
"epoch": 0.08693563880883766,
"grad_norm": 1.2109375,
"learning_rate": 0.00045200000000000004,
"loss": 6.2744,
"mean_token_accuracy": 0.1404205270111561,
"num_tokens": 2071674.0,
"step": 905
},
{
"entropy": 6.472716283798218,
"epoch": 0.08741594620557157,
"grad_norm": 1.0390625,
"learning_rate": 0.00045450000000000004,
"loss": 6.414,
"mean_token_accuracy": 0.13350878208875655,
"num_tokens": 2084025.0,
"step": 910
},
{
"entropy": 6.409608030319214,
"epoch": 0.08789625360230548,
"grad_norm": 1.2421875,
"learning_rate": 0.00045700000000000005,
"loss": 6.3529,
"mean_token_accuracy": 0.1360016256570816,
"num_tokens": 2094501.0,
"step": 915
},
{
"entropy": 6.451559734344483,
"epoch": 0.08837656099903939,
"grad_norm": 1.171875,
"learning_rate": 0.00045950000000000006,
"loss": 6.3329,
"mean_token_accuracy": 0.14456773400306702,
"num_tokens": 2106207.0,
"step": 920
},
{
"entropy": 6.352231740951538,
"epoch": 0.0888568683957733,
"grad_norm": 1.1640625,
"learning_rate": 0.000462,
"loss": 6.2281,
"mean_token_accuracy": 0.15309673249721528,
"num_tokens": 2116879.0,
"step": 925
},
{
"entropy": 6.339853239059448,
"epoch": 0.0893371757925072,
"grad_norm": 1.1640625,
"learning_rate": 0.0004645,
"loss": 6.2992,
"mean_token_accuracy": 0.14001021683216094,
"num_tokens": 2128884.0,
"step": 930
},
{
"entropy": 6.280442619323731,
"epoch": 0.08981748318924111,
"grad_norm": 1.0546875,
"learning_rate": 0.000467,
"loss": 6.2508,
"mean_token_accuracy": 0.1376420110464096,
"num_tokens": 2140810.0,
"step": 935
},
{
"entropy": 6.40587887763977,
"epoch": 0.09029779058597502,
"grad_norm": 1.1875,
"learning_rate": 0.0004695,
"loss": 6.2755,
"mean_token_accuracy": 0.13705328851938248,
"num_tokens": 2152312.0,
"step": 940
},
{
"entropy": 6.29297366142273,
"epoch": 0.09077809798270893,
"grad_norm": 0.9140625,
"learning_rate": 0.000472,
"loss": 6.242,
"mean_token_accuracy": 0.13348551020026206,
"num_tokens": 2164408.0,
"step": 945
},
{
"entropy": 6.320023441314698,
"epoch": 0.09125840537944284,
"grad_norm": 1.1484375,
"learning_rate": 0.0004745,
"loss": 6.2521,
"mean_token_accuracy": 0.14301476031541824,
"num_tokens": 2176050.0,
"step": 950
},
{
"entropy": 6.2552725315094,
"epoch": 0.09173871277617675,
"grad_norm": 1.0,
"learning_rate": 0.000477,
"loss": 6.1363,
"mean_token_accuracy": 0.14384374693036078,
"num_tokens": 2187354.0,
"step": 955
},
{
"entropy": 6.197836017608642,
"epoch": 0.09221902017291066,
"grad_norm": 1.0703125,
"learning_rate": 0.0004795,
"loss": 6.1652,
"mean_token_accuracy": 0.14129099994897842,
"num_tokens": 2198686.0,
"step": 960
},
{
"entropy": 6.174200105667114,
"epoch": 0.09269932756964457,
"grad_norm": 1.1953125,
"learning_rate": 0.000482,
"loss": 6.1982,
"mean_token_accuracy": 0.14176447689533234,
"num_tokens": 2209780.0,
"step": 965
},
{
"entropy": 6.389877271652222,
"epoch": 0.09317963496637849,
"grad_norm": 1.0,
"learning_rate": 0.0004845,
"loss": 6.2955,
"mean_token_accuracy": 0.14297716021537782,
"num_tokens": 2221760.0,
"step": 970
},
{
"entropy": 6.263871335983277,
"epoch": 0.0936599423631124,
"grad_norm": 1.15625,
"learning_rate": 0.000487,
"loss": 6.3112,
"mean_token_accuracy": 0.13698594346642495,
"num_tokens": 2233767.0,
"step": 975
},
{
"entropy": 6.315644836425781,
"epoch": 0.0941402497598463,
"grad_norm": 1.1171875,
"learning_rate": 0.0004895,
"loss": 6.2336,
"mean_token_accuracy": 0.14062682241201402,
"num_tokens": 2244841.0,
"step": 980
},
{
"entropy": 6.291112852096558,
"epoch": 0.09462055715658022,
"grad_norm": 1.078125,
"learning_rate": 0.000492,
"loss": 6.0876,
"mean_token_accuracy": 0.14949826523661613,
"num_tokens": 2255620.0,
"step": 985
},
{
"entropy": 6.233056163787841,
"epoch": 0.09510086455331412,
"grad_norm": 1.1171875,
"learning_rate": 0.0004945,
"loss": 6.2873,
"mean_token_accuracy": 0.14527306705713272,
"num_tokens": 2266345.0,
"step": 990
},
{
"entropy": 6.2935254096984865,
"epoch": 0.09558117195004803,
"grad_norm": 1.2890625,
"learning_rate": 0.000497,
"loss": 6.2219,
"mean_token_accuracy": 0.14342693910002707,
"num_tokens": 2277190.0,
"step": 995
},
{
"entropy": 6.163033580780029,
"epoch": 0.09606147934678194,
"grad_norm": 1.03125,
"learning_rate": 0.0004995,
"loss": 6.0935,
"mean_token_accuracy": 0.14876532852649688,
"num_tokens": 2288241.0,
"step": 1000
},
{
"entropy": 6.305257749557495,
"epoch": 0.09654178674351585,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999999983283737,
"loss": 6.2413,
"mean_token_accuracy": 0.14763403087854385,
"num_tokens": 2301092.0,
"step": 1005
},
{
"entropy": 6.181101942062378,
"epoch": 0.09702209414024976,
"grad_norm": 1.09375,
"learning_rate": 0.0004999999915373924,
"loss": 6.1601,
"mean_token_accuracy": 0.14240228086709977,
"num_tokens": 2313622.0,
"step": 1010
},
{
"entropy": 6.148915004730225,
"epoch": 0.09750240153698367,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999999795225793,
"loss": 6.1146,
"mean_token_accuracy": 0.14425086975097656,
"num_tokens": 2325699.0,
"step": 1015
},
{
"entropy": 6.282630729675293,
"epoch": 0.09798270893371758,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999999622839347,
"loss": 6.181,
"mean_token_accuracy": 0.14359124302864074,
"num_tokens": 2337816.0,
"step": 1020
},
{
"entropy": 6.371134090423584,
"epoch": 0.09846301633045149,
"grad_norm": 1.015625,
"learning_rate": 0.0004999999398214593,
"loss": 6.306,
"mean_token_accuracy": 0.13660114109516144,
"num_tokens": 2348793.0,
"step": 1025
},
{
"entropy": 6.207452297210693,
"epoch": 0.0989433237271854,
"grad_norm": 1.09375,
"learning_rate": 0.0004999999121351532,
"loss": 6.2091,
"mean_token_accuracy": 0.14510286301374437,
"num_tokens": 2360209.0,
"step": 1030
},
{
"entropy": 6.3725972175598145,
"epoch": 0.0994236311239193,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999998792250173,
"loss": 6.2188,
"mean_token_accuracy": 0.14163973331451415,
"num_tokens": 2370597.0,
"step": 1035
},
{
"entropy": 6.260314416885376,
"epoch": 0.09990393852065321,
"grad_norm": 1.3203125,
"learning_rate": 0.0004999998410910524,
"loss": 6.2691,
"mean_token_accuracy": 0.13516954630613326,
"num_tokens": 2381466.0,
"step": 1040
},
{
"entropy": 6.353367328643799,
"epoch": 0.10038424591738712,
"grad_norm": 1.2578125,
"learning_rate": 0.0004999997977332592,
"loss": 6.3668,
"mean_token_accuracy": 0.13565330430865288,
"num_tokens": 2394507.0,
"step": 1045
},
{
"entropy": 6.277764844894409,
"epoch": 0.10086455331412104,
"grad_norm": 1.140625,
"learning_rate": 0.0004999997491516389,
"loss": 6.2132,
"mean_token_accuracy": 0.14350937008857728,
"num_tokens": 2406367.0,
"step": 1050
},
{
"entropy": 6.184229898452759,
"epoch": 0.10134486071085495,
"grad_norm": 1.171875,
"learning_rate": 0.0004999996953461925,
"loss": 6.1792,
"mean_token_accuracy": 0.14410098493099213,
"num_tokens": 2416962.0,
"step": 1055
},
{
"entropy": 6.294152784347534,
"epoch": 0.10182516810758886,
"grad_norm": 1.15625,
"learning_rate": 0.0004999996363169212,
"loss": 6.2309,
"mean_token_accuracy": 0.13224372863769532,
"num_tokens": 2429441.0,
"step": 1060
},
{
"entropy": 6.178936433792114,
"epoch": 0.10230547550432277,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999995720638266,
"loss": 6.1068,
"mean_token_accuracy": 0.15178997665643693,
"num_tokens": 2441360.0,
"step": 1065
},
{
"entropy": 6.242135000228882,
"epoch": 0.10278578290105668,
"grad_norm": 1.046875,
"learning_rate": 0.00049999950258691,
"loss": 6.1192,
"mean_token_accuracy": 0.1425897553563118,
"num_tokens": 2452928.0,
"step": 1070
},
{
"entropy": 6.086752128601074,
"epoch": 0.10326609029779059,
"grad_norm": 1.015625,
"learning_rate": 0.0004999994278861731,
"loss": 6.0528,
"mean_token_accuracy": 0.15216556191444397,
"num_tokens": 2464218.0,
"step": 1075
},
{
"entropy": 6.244682741165161,
"epoch": 0.1037463976945245,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999993479616175,
"loss": 6.1495,
"mean_token_accuracy": 0.15013783127069474,
"num_tokens": 2475084.0,
"step": 1080
},
{
"entropy": 6.161567115783692,
"epoch": 0.1042267050912584,
"grad_norm": 1.2109375,
"learning_rate": 0.0004999992628132451,
"loss": 6.1502,
"mean_token_accuracy": 0.142550940066576,
"num_tokens": 2486845.0,
"step": 1085
},
{
"entropy": 6.204703760147095,
"epoch": 0.10470701248799232,
"grad_norm": 1.1328125,
"learning_rate": 0.0004999991724410582,
"loss": 6.2282,
"mean_token_accuracy": 0.13688293248414993,
"num_tokens": 2498049.0,
"step": 1090
},
{
"entropy": 6.171539258956909,
"epoch": 0.10518731988472622,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999990768450583,
"loss": 6.1512,
"mean_token_accuracy": 0.14735339730978012,
"num_tokens": 2509511.0,
"step": 1095
},
{
"entropy": 6.291477632522583,
"epoch": 0.10566762728146013,
"grad_norm": 1.0625,
"learning_rate": 0.0004999989760252482,
"loss": 6.1154,
"mean_token_accuracy": 0.15118677616119386,
"num_tokens": 2521518.0,
"step": 1100
},
{
"entropy": 5.977184057235718,
"epoch": 0.10614793467819404,
"grad_norm": 1.015625,
"learning_rate": 0.0004999988699816299,
"loss": 5.9663,
"mean_token_accuracy": 0.15758592933416365,
"num_tokens": 2533141.0,
"step": 1105
},
{
"entropy": 6.063448810577393,
"epoch": 0.10662824207492795,
"grad_norm": 1.125,
"learning_rate": 0.0004999987587142058,
"loss": 5.9728,
"mean_token_accuracy": 0.14997318163514137,
"num_tokens": 2542907.0,
"step": 1110
},
{
"entropy": 6.109343767166138,
"epoch": 0.10710854947166186,
"grad_norm": 0.9375,
"learning_rate": 0.0004999986422229789,
"loss": 6.1184,
"mean_token_accuracy": 0.1399646908044815,
"num_tokens": 2554497.0,
"step": 1115
},
{
"entropy": 6.291832542419433,
"epoch": 0.10758885686839577,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999985205079514,
"loss": 6.2366,
"mean_token_accuracy": 0.14908648729324342,
"num_tokens": 2567035.0,
"step": 1120
},
{
"entropy": 6.043786144256591,
"epoch": 0.10806916426512968,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999983935691265,
"loss": 6.007,
"mean_token_accuracy": 0.15258836150169372,
"num_tokens": 2578003.0,
"step": 1125
},
{
"entropy": 6.293131971359253,
"epoch": 0.10854947166186359,
"grad_norm": 1.1015625,
"learning_rate": 0.000499998261406507,
"loss": 6.3085,
"mean_token_accuracy": 0.14156671091914177,
"num_tokens": 2590272.0,
"step": 1130
},
{
"entropy": 6.125889730453491,
"epoch": 0.10902977905859751,
"grad_norm": 1.046875,
"learning_rate": 0.0004999981240200958,
"loss": 6.1307,
"mean_token_accuracy": 0.14405303746461867,
"num_tokens": 2602874.0,
"step": 1135
},
{
"entropy": 6.140783977508545,
"epoch": 0.10951008645533142,
"grad_norm": 1.09375,
"learning_rate": 0.0004999979814098966,
"loss": 6.0883,
"mean_token_accuracy": 0.15575037002563477,
"num_tokens": 2613835.0,
"step": 1140
},
{
"entropy": 6.244704246520996,
"epoch": 0.10999039385206533,
"grad_norm": 1.03125,
"learning_rate": 0.0004999978335759121,
"loss": 6.1962,
"mean_token_accuracy": 0.14173132926225662,
"num_tokens": 2624986.0,
"step": 1145
},
{
"entropy": 6.1988218307495115,
"epoch": 0.11047070124879924,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999976805181461,
"loss": 6.1939,
"mean_token_accuracy": 0.1403311349451542,
"num_tokens": 2637654.0,
"step": 1150
},
{
"entropy": 6.199357604980468,
"epoch": 0.11095100864553314,
"grad_norm": 1.0390625,
"learning_rate": 0.000499997522236602,
"loss": 6.1037,
"mean_token_accuracy": 0.14716718941926957,
"num_tokens": 2648947.0,
"step": 1155
},
{
"entropy": 6.130415296554565,
"epoch": 0.11143131604226705,
"grad_norm": 1.046875,
"learning_rate": 0.0004999973587312837,
"loss": 6.0963,
"mean_token_accuracy": 0.14842916876077653,
"num_tokens": 2659573.0,
"step": 1160
},
{
"entropy": 6.0729498863220215,
"epoch": 0.11191162343900096,
"grad_norm": 0.984375,
"learning_rate": 0.0004999971900021947,
"loss": 6.0778,
"mean_token_accuracy": 0.15214993208646774,
"num_tokens": 2671491.0,
"step": 1165
},
{
"entropy": 6.157928133010865,
"epoch": 0.11239193083573487,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999970160493391,
"loss": 6.0571,
"mean_token_accuracy": 0.15085275322198868,
"num_tokens": 2682705.0,
"step": 1170
},
{
"entropy": 5.988412666320801,
"epoch": 0.11287223823246878,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999968368727209,
"loss": 5.9857,
"mean_token_accuracy": 0.1512448124587536,
"num_tokens": 2693539.0,
"step": 1175
},
{
"entropy": 6.143959999084473,
"epoch": 0.11335254562920269,
"grad_norm": 1.078125,
"learning_rate": 0.0004999966524723442,
"loss": 6.0359,
"mean_token_accuracy": 0.15259774029254913,
"num_tokens": 2705403.0,
"step": 1180
},
{
"entropy": 6.179290437698365,
"epoch": 0.1138328530259366,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999964628482135,
"loss": 6.1553,
"mean_token_accuracy": 0.14431207478046418,
"num_tokens": 2717320.0,
"step": 1185
},
{
"entropy": 6.080302906036377,
"epoch": 0.1143131604226705,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999962680003328,
"loss": 6.1372,
"mean_token_accuracy": 0.1427521824836731,
"num_tokens": 2730056.0,
"step": 1190
},
{
"entropy": 6.139111661911011,
"epoch": 0.11479346781940442,
"grad_norm": 1.09375,
"learning_rate": 0.000499996067928707,
"loss": 5.9903,
"mean_token_accuracy": 0.15253911167383194,
"num_tokens": 2741399.0,
"step": 1195
},
{
"entropy": 6.1936037063598635,
"epoch": 0.11527377521613832,
"grad_norm": 1.109375,
"learning_rate": 0.0004999958626333406,
"loss": 6.2196,
"mean_token_accuracy": 0.14051759988069534,
"num_tokens": 2752277.0,
"step": 1200
},
{
"entropy": 6.179323339462281,
"epoch": 0.11575408261287223,
"grad_norm": 1.09375,
"learning_rate": 0.0004999956521142383,
"loss": 6.1245,
"mean_token_accuracy": 0.1430825725197792,
"num_tokens": 2763172.0,
"step": 1205
},
{
"entropy": 6.169216156005859,
"epoch": 0.11623439000960614,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999954363714051,
"loss": 6.1048,
"mean_token_accuracy": 0.1494622513651848,
"num_tokens": 2775096.0,
"step": 1210
},
{
"entropy": 6.178955316543579,
"epoch": 0.11671469740634005,
"grad_norm": 1.078125,
"learning_rate": 0.0004999952154048459,
"loss": 6.208,
"mean_token_accuracy": 0.14561985433101654,
"num_tokens": 2786007.0,
"step": 1215
},
{
"entropy": 6.087857627868653,
"epoch": 0.11719500480307397,
"grad_norm": 1.1171875,
"learning_rate": 0.000499994989214566,
"loss": 6.0045,
"mean_token_accuracy": 0.15556134507060052,
"num_tokens": 2797265.0,
"step": 1220
},
{
"entropy": 6.131901121139526,
"epoch": 0.11767531219980788,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999947578005705,
"loss": 6.1802,
"mean_token_accuracy": 0.14725500270724295,
"num_tokens": 2809427.0,
"step": 1225
},
{
"entropy": 6.175329542160034,
"epoch": 0.11815561959654179,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999945211628648,
"loss": 6.1007,
"mean_token_accuracy": 0.15144748464226723,
"num_tokens": 2822880.0,
"step": 1230
},
{
"entropy": 6.134575986862183,
"epoch": 0.1186359269932757,
"grad_norm": 0.9375,
"learning_rate": 0.0004999942793014544,
"loss": 6.1706,
"mean_token_accuracy": 0.14877447858452797,
"num_tokens": 2834760.0,
"step": 1235
},
{
"entropy": 6.1685347080230715,
"epoch": 0.11911623439000961,
"grad_norm": 1.0546875,
"learning_rate": 0.000499994032216345,
"loss": 6.107,
"mean_token_accuracy": 0.14922730922698973,
"num_tokens": 2846517.0,
"step": 1240
},
{
"entropy": 6.030606937408447,
"epoch": 0.11959654178674352,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999937799075422,
"loss": 5.9855,
"mean_token_accuracy": 0.16014222726225852,
"num_tokens": 2857280.0,
"step": 1245
},
{
"entropy": 6.123331689834595,
"epoch": 0.12007684918347743,
"grad_norm": 0.94140625,
"learning_rate": 0.000499993522375052,
"loss": 6.0458,
"mean_token_accuracy": 0.15252126306295394,
"num_tokens": 2869735.0,
"step": 1250
},
{
"entropy": 6.090895652770996,
"epoch": 0.12055715658021134,
"grad_norm": 1.140625,
"learning_rate": 0.0004999932596188802,
"loss": 6.0477,
"mean_token_accuracy": 0.15124670863151551,
"num_tokens": 2881681.0,
"step": 1255
},
{
"entropy": 6.028911304473877,
"epoch": 0.12103746397694524,
"grad_norm": 0.984375,
"learning_rate": 0.0004999929916390331,
"loss": 6.0556,
"mean_token_accuracy": 0.1572458490729332,
"num_tokens": 2893828.0,
"step": 1260
},
{
"entropy": 5.935788679122925,
"epoch": 0.12151777137367915,
"grad_norm": 1.0625,
"learning_rate": 0.0004999927184355169,
"loss": 6.0172,
"mean_token_accuracy": 0.14968452900648116,
"num_tokens": 2905804.0,
"step": 1265
},
{
"entropy": 6.311611938476562,
"epoch": 0.12199807877041306,
"grad_norm": 1.15625,
"learning_rate": 0.0004999924400083377,
"loss": 6.1599,
"mean_token_accuracy": 0.1448735289275646,
"num_tokens": 2918709.0,
"step": 1270
},
{
"entropy": 6.015513372421265,
"epoch": 0.12247838616714697,
"grad_norm": 1.0625,
"learning_rate": 0.0004999921563575022,
"loss": 6.079,
"mean_token_accuracy": 0.1476942054927349,
"num_tokens": 2930589.0,
"step": 1275
},
{
"entropy": 6.166260385513306,
"epoch": 0.12295869356388088,
"grad_norm": 1.0,
"learning_rate": 0.0004999918674830169,
"loss": 6.1131,
"mean_token_accuracy": 0.15547159016132356,
"num_tokens": 2940635.0,
"step": 1280
},
{
"entropy": 6.008727741241455,
"epoch": 0.12343900096061479,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999915733848886,
"loss": 6.0079,
"mean_token_accuracy": 0.15664828717708587,
"num_tokens": 2952731.0,
"step": 1285
},
{
"entropy": 6.062463045120239,
"epoch": 0.1239193083573487,
"grad_norm": 1.109375,
"learning_rate": 0.000499991274063124,
"loss": 6.0144,
"mean_token_accuracy": 0.1488832116127014,
"num_tokens": 2964544.0,
"step": 1290
},
{
"entropy": 6.022071123123169,
"epoch": 0.12439961575408261,
"grad_norm": 1.140625,
"learning_rate": 0.0004999909695177301,
"loss": 5.9402,
"mean_token_accuracy": 0.1597997650504112,
"num_tokens": 2974996.0,
"step": 1295
},
{
"entropy": 5.994995880126953,
"epoch": 0.12487992315081652,
"grad_norm": 1.0703125,
"learning_rate": 0.000499990659748714,
"loss": 5.9467,
"mean_token_accuracy": 0.15251300781965255,
"num_tokens": 2986496.0,
"step": 1300
},
{
"entropy": 6.049901390075684,
"epoch": 0.12536023054755044,
"grad_norm": 1.03125,
"learning_rate": 0.0004999903447560828,
"loss": 6.037,
"mean_token_accuracy": 0.1527161717414856,
"num_tokens": 2997693.0,
"step": 1305
},
{
"entropy": 6.033747673034668,
"epoch": 0.12584053794428435,
"grad_norm": 1.125,
"learning_rate": 0.0004999900245398439,
"loss": 5.9853,
"mean_token_accuracy": 0.15558816790580748,
"num_tokens": 3009155.0,
"step": 1310
},
{
"entropy": 5.998337697982788,
"epoch": 0.12632084534101826,
"grad_norm": 0.984375,
"learning_rate": 0.0004999896991000047,
"loss": 5.9141,
"mean_token_accuracy": 0.1596623882651329,
"num_tokens": 3020944.0,
"step": 1315
},
{
"entropy": 5.977126979827881,
"epoch": 0.12680115273775217,
"grad_norm": 1.046875,
"learning_rate": 0.0004999893684365729,
"loss": 5.977,
"mean_token_accuracy": 0.15605373829603195,
"num_tokens": 3032893.0,
"step": 1320
},
{
"entropy": 6.0546129703521725,
"epoch": 0.12728146013448607,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999890325495559,
"loss": 6.0419,
"mean_token_accuracy": 0.1505540929734707,
"num_tokens": 3044937.0,
"step": 1325
},
{
"entropy": 5.99818902015686,
"epoch": 0.12776176753121998,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999886914389617,
"loss": 5.9324,
"mean_token_accuracy": 0.15481951385736464,
"num_tokens": 3056071.0,
"step": 1330
},
{
"entropy": 6.046021556854248,
"epoch": 0.1282420749279539,
"grad_norm": 1.2890625,
"learning_rate": 0.0004999883451047981,
"loss": 6.1056,
"mean_token_accuracy": 0.15031077489256858,
"num_tokens": 3067724.0,
"step": 1335
},
{
"entropy": 6.037922430038452,
"epoch": 0.1287223823246878,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999879935470733,
"loss": 5.9961,
"mean_token_accuracy": 0.1605654314160347,
"num_tokens": 3080779.0,
"step": 1340
},
{
"entropy": 6.017280960083008,
"epoch": 0.1292026897214217,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999876367657954,
"loss": 5.9936,
"mean_token_accuracy": 0.15067695081233978,
"num_tokens": 3091991.0,
"step": 1345
},
{
"entropy": 5.997524261474609,
"epoch": 0.12968299711815562,
"grad_norm": 1.015625,
"learning_rate": 0.0004999872747609725,
"loss": 5.9778,
"mean_token_accuracy": 0.15370408892631532,
"num_tokens": 3103713.0,
"step": 1350
},
{
"entropy": 6.033483362197876,
"epoch": 0.13016330451488953,
"grad_norm": 1.015625,
"learning_rate": 0.0004999869075326132,
"loss": 6.0116,
"mean_token_accuracy": 0.15471786111593247,
"num_tokens": 3115891.0,
"step": 1355
},
{
"entropy": 5.958883380889892,
"epoch": 0.13064361191162344,
"grad_norm": 1.046875,
"learning_rate": 0.000499986535080726,
"loss": 5.8745,
"mean_token_accuracy": 0.16241799741983415,
"num_tokens": 3126223.0,
"step": 1360
},
{
"entropy": 6.039581346511841,
"epoch": 0.13112391930835735,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999861574053196,
"loss": 6.0623,
"mean_token_accuracy": 0.14615725725889206,
"num_tokens": 3137537.0,
"step": 1365
},
{
"entropy": 6.128436279296875,
"epoch": 0.13160422670509125,
"grad_norm": 1.046875,
"learning_rate": 0.0004999857745064027,
"loss": 6.0553,
"mean_token_accuracy": 0.1591821163892746,
"num_tokens": 3150126.0,
"step": 1370
},
{
"entropy": 6.011444139480591,
"epoch": 0.13208453410182516,
"grad_norm": 0.94140625,
"learning_rate": 0.000499985386383984,
"loss": 6.0267,
"mean_token_accuracy": 0.1561764433979988,
"num_tokens": 3161684.0,
"step": 1375
},
{
"entropy": 5.9896073818206785,
"epoch": 0.13256484149855907,
"grad_norm": 1.078125,
"learning_rate": 0.0004999849930380729,
"loss": 5.9039,
"mean_token_accuracy": 0.1553078919649124,
"num_tokens": 3172342.0,
"step": 1380
},
{
"entropy": 6.075509262084961,
"epoch": 0.13304514889529298,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999845944686781,
"loss": 6.0467,
"mean_token_accuracy": 0.15783883035182952,
"num_tokens": 3183681.0,
"step": 1385
},
{
"entropy": 5.887626361846924,
"epoch": 0.1335254562920269,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999841906758093,
"loss": 5.8602,
"mean_token_accuracy": 0.16416978389024733,
"num_tokens": 3194122.0,
"step": 1390
},
{
"entropy": 5.952710771560669,
"epoch": 0.1340057636887608,
"grad_norm": 0.953125,
"learning_rate": 0.0004999837816594757,
"loss": 5.9189,
"mean_token_accuracy": 0.1564559370279312,
"num_tokens": 3205719.0,
"step": 1395
},
{
"entropy": 5.924874639511108,
"epoch": 0.1344860710854947,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999833674196865,
"loss": 5.9381,
"mean_token_accuracy": 0.15794704854488373,
"num_tokens": 3216262.0,
"step": 1400
},
{
"entropy": 6.069398736953735,
"epoch": 0.13496637848222862,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999829479564518,
"loss": 6.0079,
"mean_token_accuracy": 0.14638098031282426,
"num_tokens": 3228035.0,
"step": 1405
},
{
"entropy": 5.878259229660034,
"epoch": 0.13544668587896252,
"grad_norm": 0.9609375,
"learning_rate": 0.000499982523269781,
"loss": 5.8583,
"mean_token_accuracy": 0.1624181777238846,
"num_tokens": 3239643.0,
"step": 1410
},
{
"entropy": 6.007868909835816,
"epoch": 0.13592699327569643,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999820933596842,
"loss": 5.9453,
"mean_token_accuracy": 0.15799377262592315,
"num_tokens": 3250293.0,
"step": 1415
},
{
"entropy": 6.07816219329834,
"epoch": 0.13640730067243034,
"grad_norm": 0.96875,
"learning_rate": 0.000499981658226171,
"loss": 6.0542,
"mean_token_accuracy": 0.14943305402994156,
"num_tokens": 3262448.0,
"step": 1420
},
{
"entropy": 5.908995771408081,
"epoch": 0.13688760806916425,
"grad_norm": 1.0703125,
"learning_rate": 0.000499981217869252,
"loss": 5.929,
"mean_token_accuracy": 0.15811057537794113,
"num_tokens": 3274041.0,
"step": 1425
},
{
"entropy": 6.054948139190674,
"epoch": 0.1373679154658982,
"grad_norm": 0.99609375,
"learning_rate": 0.000499980772288937,
"loss": 5.9561,
"mean_token_accuracy": 0.15184218436479568,
"num_tokens": 3285063.0,
"step": 1430
},
{
"entropy": 5.976988172531128,
"epoch": 0.1378482228626321,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999803214852367,
"loss": 5.977,
"mean_token_accuracy": 0.1544366106390953,
"num_tokens": 3297031.0,
"step": 1435
},
{
"entropy": 6.070099973678589,
"epoch": 0.138328530259366,
"grad_norm": 1.0625,
"learning_rate": 0.0004999798654581613,
"loss": 5.947,
"mean_token_accuracy": 0.15437746345996856,
"num_tokens": 3308187.0,
"step": 1440
},
{
"entropy": 5.916920280456543,
"epoch": 0.13880883765609991,
"grad_norm": 1.0625,
"learning_rate": 0.0004999794042077214,
"loss": 5.9652,
"mean_token_accuracy": 0.15356913805007935,
"num_tokens": 3319439.0,
"step": 1445
},
{
"entropy": 6.065345001220703,
"epoch": 0.13928914505283382,
"grad_norm": 1.078125,
"learning_rate": 0.0004999789377339279,
"loss": 5.9672,
"mean_token_accuracy": 0.15113933831453324,
"num_tokens": 3331715.0,
"step": 1450
},
{
"entropy": 5.8917687892913815,
"epoch": 0.13976945244956773,
"grad_norm": 1.03125,
"learning_rate": 0.0004999784660367915,
"loss": 5.9209,
"mean_token_accuracy": 0.1623266249895096,
"num_tokens": 3342281.0,
"step": 1455
},
{
"entropy": 6.142686176300049,
"epoch": 0.14024975984630164,
"grad_norm": 1.140625,
"learning_rate": 0.0004999779891163231,
"loss": 6.0476,
"mean_token_accuracy": 0.1508173242211342,
"num_tokens": 3353530.0,
"step": 1460
},
{
"entropy": 5.919660520553589,
"epoch": 0.14073006724303555,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999775069725339,
"loss": 6.0189,
"mean_token_accuracy": 0.15143757462501525,
"num_tokens": 3365083.0,
"step": 1465
},
{
"entropy": 6.090314149856567,
"epoch": 0.14121037463976946,
"grad_norm": 0.84375,
"learning_rate": 0.000499977019605435,
"loss": 5.972,
"mean_token_accuracy": 0.15416960716247557,
"num_tokens": 3378143.0,
"step": 1470
},
{
"entropy": 5.997537088394165,
"epoch": 0.14169068203650337,
"grad_norm": 1.1171875,
"learning_rate": 0.0004999765270150378,
"loss": 5.9514,
"mean_token_accuracy": 0.15883016735315322,
"num_tokens": 3388438.0,
"step": 1475
},
{
"entropy": 5.8716082096099855,
"epoch": 0.14217098943323728,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999760292013536,
"loss": 5.95,
"mean_token_accuracy": 0.15198695808649063,
"num_tokens": 3400150.0,
"step": 1480
},
{
"entropy": 5.939674329757691,
"epoch": 0.14265129682997119,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999755261643941,
"loss": 5.827,
"mean_token_accuracy": 0.16298246681690215,
"num_tokens": 3411073.0,
"step": 1485
},
{
"entropy": 5.8928299903869625,
"epoch": 0.1431316042267051,
"grad_norm": 1.0,
"learning_rate": 0.0004999750179041709,
"loss": 5.8905,
"mean_token_accuracy": 0.1664283663034439,
"num_tokens": 3422196.0,
"step": 1490
},
{
"entropy": 6.0256095886230465,
"epoch": 0.143611911623439,
"grad_norm": 1.09375,
"learning_rate": 0.0004999745044206959,
"loss": 5.9115,
"mean_token_accuracy": 0.15011470019817352,
"num_tokens": 3432568.0,
"step": 1495
},
{
"entropy": 5.954646492004395,
"epoch": 0.1440922190201729,
"grad_norm": 0.953125,
"learning_rate": 0.0004999739857139809,
"loss": 5.9423,
"mean_token_accuracy": 0.1548205330967903,
"num_tokens": 3444096.0,
"step": 1500
},
{
"entropy": 5.87907772064209,
"epoch": 0.14457252641690682,
"grad_norm": 0.95703125,
"learning_rate": 0.000499973461784038,
"loss": 5.8361,
"mean_token_accuracy": 0.16604594886302948,
"num_tokens": 3455502.0,
"step": 1505
},
{
"entropy": 5.99258599281311,
"epoch": 0.14505283381364073,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999729326308792,
"loss": 5.9705,
"mean_token_accuracy": 0.15101340860128404,
"num_tokens": 3467753.0,
"step": 1510
},
{
"entropy": 5.8902027130126955,
"epoch": 0.14553314121037464,
"grad_norm": 1.1015625,
"learning_rate": 0.000499972398254517,
"loss": 5.835,
"mean_token_accuracy": 0.16128894686698914,
"num_tokens": 3479206.0,
"step": 1515
},
{
"entropy": 5.944278573989868,
"epoch": 0.14601344860710855,
"grad_norm": 1.0546875,
"learning_rate": 0.000499971858654964,
"loss": 5.8652,
"mean_token_accuracy": 0.15937050879001619,
"num_tokens": 3490914.0,
"step": 1520
},
{
"entropy": 5.884905815124512,
"epoch": 0.14649375600384246,
"grad_norm": 0.9375,
"learning_rate": 0.0004999713138322321,
"loss": 5.9293,
"mean_token_accuracy": 0.15429736226797103,
"num_tokens": 3502932.0,
"step": 1525
},
{
"entropy": 5.872306871414184,
"epoch": 0.14697406340057637,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999707637863346,
"loss": 5.836,
"mean_token_accuracy": 0.16672665476799012,
"num_tokens": 3515071.0,
"step": 1530
},
{
"entropy": 6.054158544540405,
"epoch": 0.14745437079731027,
"grad_norm": 1.125,
"learning_rate": 0.0004999702085172838,
"loss": 5.9631,
"mean_token_accuracy": 0.15781875029206277,
"num_tokens": 3526845.0,
"step": 1535
},
{
"entropy": 5.88759765625,
"epoch": 0.14793467819404418,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999696480250929,
"loss": 5.9392,
"mean_token_accuracy": 0.15117181837558746,
"num_tokens": 3538810.0,
"step": 1540
},
{
"entropy": 5.999429655075073,
"epoch": 0.1484149855907781,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999690823097747,
"loss": 5.9088,
"mean_token_accuracy": 0.16723537892103196,
"num_tokens": 3549570.0,
"step": 1545
},
{
"entropy": 5.722354888916016,
"epoch": 0.148895292987512,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999685113713426,
"loss": 5.8273,
"mean_token_accuracy": 0.1671121299266815,
"num_tokens": 3560937.0,
"step": 1550
},
{
"entropy": 5.976478576660156,
"epoch": 0.1493756003842459,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999679352098096,
"loss": 5.8734,
"mean_token_accuracy": 0.15577627420425416,
"num_tokens": 3573603.0,
"step": 1555
},
{
"entropy": 5.879854822158814,
"epoch": 0.14985590778097982,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999673538251891,
"loss": 5.8848,
"mean_token_accuracy": 0.15590892136096954,
"num_tokens": 3586034.0,
"step": 1560
},
{
"entropy": 5.8598157405853275,
"epoch": 0.15033621517771373,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999667672174947,
"loss": 5.8176,
"mean_token_accuracy": 0.1654493823647499,
"num_tokens": 3597305.0,
"step": 1565
},
{
"entropy": 6.031460905075074,
"epoch": 0.15081652257444764,
"grad_norm": 0.9609375,
"learning_rate": 0.00049996617538674,
"loss": 5.9745,
"mean_token_accuracy": 0.1476095847785473,
"num_tokens": 3608758.0,
"step": 1570
},
{
"entropy": 5.989815330505371,
"epoch": 0.15129682997118155,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999655783329386,
"loss": 5.9659,
"mean_token_accuracy": 0.1517655596137047,
"num_tokens": 3621508.0,
"step": 1575
},
{
"entropy": 5.9538108825683596,
"epoch": 0.15177713736791545,
"grad_norm": 0.875,
"learning_rate": 0.0004999649760561046,
"loss": 5.9111,
"mean_token_accuracy": 0.15990381091833114,
"num_tokens": 3634826.0,
"step": 1580
},
{
"entropy": 5.782315063476562,
"epoch": 0.15225744476464936,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999643685562519,
"loss": 5.7586,
"mean_token_accuracy": 0.17986451387405394,
"num_tokens": 3646427.0,
"step": 1585
},
{
"entropy": 5.871938896179199,
"epoch": 0.15273775216138327,
"grad_norm": 1.015625,
"learning_rate": 0.0004999637558333945,
"loss": 5.8846,
"mean_token_accuracy": 0.15845945328474045,
"num_tokens": 3658157.0,
"step": 1590
},
{
"entropy": 5.965056943893432,
"epoch": 0.15321805955811718,
"grad_norm": 1.03125,
"learning_rate": 0.0004999631378875467,
"loss": 5.9283,
"mean_token_accuracy": 0.15535358637571334,
"num_tokens": 3670695.0,
"step": 1595
},
{
"entropy": 5.852532148361206,
"epoch": 0.15369836695485112,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999625147187228,
"loss": 5.847,
"mean_token_accuracy": 0.15907091200351714,
"num_tokens": 3683379.0,
"step": 1600
},
{
"entropy": 5.896727418899536,
"epoch": 0.15417867435158503,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999618863269373,
"loss": 5.7427,
"mean_token_accuracy": 0.1712520807981491,
"num_tokens": 3694411.0,
"step": 1605
},
{
"entropy": 5.7830735206604,
"epoch": 0.15465898174831894,
"grad_norm": 1.125,
"learning_rate": 0.0004999612527122049,
"loss": 5.7999,
"mean_token_accuracy": 0.164157934486866,
"num_tokens": 3703833.0,
"step": 1610
},
{
"entropy": 5.98201847076416,
"epoch": 0.15513928914505284,
"grad_norm": 1.015625,
"learning_rate": 0.0004999606138745402,
"loss": 5.8637,
"mean_token_accuracy": 0.17183852046728135,
"num_tokens": 3714702.0,
"step": 1615
},
{
"entropy": 5.91125226020813,
"epoch": 0.15561959654178675,
"grad_norm": 1.015625,
"learning_rate": 0.0004999599698139581,
"loss": 5.9601,
"mean_token_accuracy": 0.16254139840602874,
"num_tokens": 3726000.0,
"step": 1620
},
{
"entropy": 5.942967700958252,
"epoch": 0.15609990393852066,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999593205304734,
"loss": 5.8441,
"mean_token_accuracy": 0.15695785135030746,
"num_tokens": 3737681.0,
"step": 1625
},
{
"entropy": 5.980777454376221,
"epoch": 0.15658021133525457,
"grad_norm": 0.921875,
"learning_rate": 0.0004999586660241012,
"loss": 5.9471,
"mean_token_accuracy": 0.1590610146522522,
"num_tokens": 3749477.0,
"step": 1630
},
{
"entropy": 5.983596611022949,
"epoch": 0.15706051873198848,
"grad_norm": 1.078125,
"learning_rate": 0.0004999580062948569,
"loss": 5.9269,
"mean_token_accuracy": 0.15622665733098984,
"num_tokens": 3759946.0,
"step": 1635
},
{
"entropy": 5.82830491065979,
"epoch": 0.1575408261287224,
"grad_norm": 1.046875,
"learning_rate": 0.0004999573413427556,
"loss": 5.8301,
"mean_token_accuracy": 0.1631380781531334,
"num_tokens": 3770147.0,
"step": 1640
},
{
"entropy": 5.879538917541504,
"epoch": 0.1580211335254563,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999566711678128,
"loss": 5.789,
"mean_token_accuracy": 0.16764923334121704,
"num_tokens": 3781028.0,
"step": 1645
},
{
"entropy": 5.877831649780274,
"epoch": 0.1585014409221902,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999559957700442,
"loss": 5.8577,
"mean_token_accuracy": 0.15628671497106553,
"num_tokens": 3792529.0,
"step": 1650
},
{
"entropy": 5.9889665126800535,
"epoch": 0.15898174831892412,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999553151494653,
"loss": 5.9658,
"mean_token_accuracy": 0.16207647919654847,
"num_tokens": 3804630.0,
"step": 1655
},
{
"entropy": 5.756636714935302,
"epoch": 0.15946205571565802,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999546293060919,
"loss": 5.7305,
"mean_token_accuracy": 0.17414763569831848,
"num_tokens": 3815019.0,
"step": 1660
},
{
"entropy": 5.884789562225341,
"epoch": 0.15994236311239193,
"grad_norm": 1.0859375,
"learning_rate": 0.00049995393823994,
"loss": 5.8346,
"mean_token_accuracy": 0.16163911521434784,
"num_tokens": 3825648.0,
"step": 1665
},
{
"entropy": 5.865713882446289,
"epoch": 0.16042267050912584,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999532419510255,
"loss": 5.8267,
"mean_token_accuracy": 0.16119918525218963,
"num_tokens": 3836779.0,
"step": 1670
},
{
"entropy": 5.831444215774536,
"epoch": 0.16090297790585975,
"grad_norm": 0.890625,
"learning_rate": 0.000499952540439365,
"loss": 5.8686,
"mean_token_accuracy": 0.1589963838458061,
"num_tokens": 3848350.0,
"step": 1675
},
{
"entropy": 5.968743467330933,
"epoch": 0.16138328530259366,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999518337049743,
"loss": 5.8602,
"mean_token_accuracy": 0.16284065395593644,
"num_tokens": 3859792.0,
"step": 1680
},
{
"entropy": 5.835683107376099,
"epoch": 0.16186359269932757,
"grad_norm": 0.9765625,
"learning_rate": 0.00049995112174787,
"loss": 5.7993,
"mean_token_accuracy": 0.16204592287540437,
"num_tokens": 3871539.0,
"step": 1685
},
{
"entropy": 5.7748315811157225,
"epoch": 0.16234390009606148,
"grad_norm": 0.90625,
"learning_rate": 0.0004999504045680687,
"loss": 5.8424,
"mean_token_accuracy": 0.1638094499707222,
"num_tokens": 3885025.0,
"step": 1690
},
{
"entropy": 5.921676111221314,
"epoch": 0.1628242074927954,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999496821655869,
"loss": 5.9563,
"mean_token_accuracy": 0.1556147813796997,
"num_tokens": 3896519.0,
"step": 1695
},
{
"entropy": 5.8414802074432375,
"epoch": 0.1633045148895293,
"grad_norm": 1.046875,
"learning_rate": 0.0004999489545404414,
"loss": 5.7111,
"mean_token_accuracy": 0.16769966781139373,
"num_tokens": 3907636.0,
"step": 1700
},
{
"entropy": 5.826439046859742,
"epoch": 0.1637848222862632,
"grad_norm": 1.0859375,
"learning_rate": 0.0004999482216926493,
"loss": 5.8457,
"mean_token_accuracy": 0.1643659219145775,
"num_tokens": 3917746.0,
"step": 1705
},
{
"entropy": 5.835819673538208,
"epoch": 0.1642651296829971,
"grad_norm": 1.03125,
"learning_rate": 0.0004999474836222273,
"loss": 5.7635,
"mean_token_accuracy": 0.16848202645778657,
"num_tokens": 3928901.0,
"step": 1710
},
{
"entropy": 5.7875734806060795,
"epoch": 0.16474543707973102,
"grad_norm": 1.0625,
"learning_rate": 0.0004999467403291928,
"loss": 5.8911,
"mean_token_accuracy": 0.16683052182197572,
"num_tokens": 3940094.0,
"step": 1715
},
{
"entropy": 5.896417474746704,
"epoch": 0.16522574447646493,
"grad_norm": 0.96875,
"learning_rate": 0.0004999459918135628,
"loss": 5.7946,
"mean_token_accuracy": 0.1682782679796219,
"num_tokens": 3952714.0,
"step": 1720
},
{
"entropy": 5.809536647796631,
"epoch": 0.16570605187319884,
"grad_norm": 1.0625,
"learning_rate": 0.000499945238075355,
"loss": 5.7855,
"mean_token_accuracy": 0.1620207592844963,
"num_tokens": 3963107.0,
"step": 1725
},
{
"entropy": 5.8475425243377686,
"epoch": 0.16618635926993275,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999444791145865,
"loss": 5.7234,
"mean_token_accuracy": 0.16508166193962098,
"num_tokens": 3973454.0,
"step": 1730
},
{
"entropy": 5.773418426513672,
"epoch": 0.16666666666666666,
"grad_norm": 1.1484375,
"learning_rate": 0.0004999437149312754,
"loss": 5.838,
"mean_token_accuracy": 0.1589142844080925,
"num_tokens": 3985352.0,
"step": 1735
},
{
"entropy": 5.918348503112793,
"epoch": 0.16714697406340057,
"grad_norm": 0.99609375,
"learning_rate": 0.000499942945525439,
"loss": 5.8059,
"mean_token_accuracy": 0.16445475369691848,
"num_tokens": 3997088.0,
"step": 1740
},
{
"entropy": 5.87802152633667,
"epoch": 0.16762728146013448,
"grad_norm": 1.0625,
"learning_rate": 0.0004999421708970954,
"loss": 5.8404,
"mean_token_accuracy": 0.1627303183078766,
"num_tokens": 4008551.0,
"step": 1745
},
{
"entropy": 5.842701101303101,
"epoch": 0.16810758885686838,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999413910462625,
"loss": 5.8705,
"mean_token_accuracy": 0.1607220634818077,
"num_tokens": 4020886.0,
"step": 1750
},
{
"entropy": 5.901450777053833,
"epoch": 0.1685878962536023,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999406059729586,
"loss": 5.8082,
"mean_token_accuracy": 0.16304053068161012,
"num_tokens": 4032811.0,
"step": 1755
},
{
"entropy": 5.838007640838623,
"epoch": 0.1690682036503362,
"grad_norm": 0.984375,
"learning_rate": 0.0004999398156772016,
"loss": 5.795,
"mean_token_accuracy": 0.17031055241823195,
"num_tokens": 4044102.0,
"step": 1760
},
{
"entropy": 5.838218402862549,
"epoch": 0.16954851104707014,
"grad_norm": 1.046875,
"learning_rate": 0.00049993902015901,
"loss": 5.8305,
"mean_token_accuracy": 0.16805302500724792,
"num_tokens": 4056849.0,
"step": 1765
},
{
"entropy": 5.8025048732757565,
"epoch": 0.17002881844380405,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999382194184023,
"loss": 5.8535,
"mean_token_accuracy": 0.16151585429906845,
"num_tokens": 4068392.0,
"step": 1770
},
{
"entropy": 5.792164945602417,
"epoch": 0.17050912584053796,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999374134553972,
"loss": 5.6781,
"mean_token_accuracy": 0.1671793356537819,
"num_tokens": 4078949.0,
"step": 1775
},
{
"entropy": 5.893219375610352,
"epoch": 0.17098943323727187,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999366022700131,
"loss": 5.8373,
"mean_token_accuracy": 0.1613660380244255,
"num_tokens": 4091432.0,
"step": 1780
},
{
"entropy": 5.718272161483765,
"epoch": 0.17146974063400577,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999357858622691,
"loss": 5.6977,
"mean_token_accuracy": 0.1697030022740364,
"num_tokens": 4103001.0,
"step": 1785
},
{
"entropy": 5.845357942581177,
"epoch": 0.17195004803073968,
"grad_norm": 1.0,
"learning_rate": 0.0004999349642321842,
"loss": 5.8151,
"mean_token_accuracy": 0.16591036021709443,
"num_tokens": 4113570.0,
"step": 1790
},
{
"entropy": 5.829782104492187,
"epoch": 0.1724303554274736,
"grad_norm": 1.015625,
"learning_rate": 0.0004999341373797772,
"loss": 5.8416,
"mean_token_accuracy": 0.15971099138259887,
"num_tokens": 4125658.0,
"step": 1795
},
{
"entropy": 5.787124729156494,
"epoch": 0.1729106628242075,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999333053050675,
"loss": 5.7384,
"mean_token_accuracy": 0.16838262975215912,
"num_tokens": 4137424.0,
"step": 1800
},
{
"entropy": 5.815556621551513,
"epoch": 0.1733909702209414,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999324680080744,
"loss": 5.7858,
"mean_token_accuracy": 0.16323755234479903,
"num_tokens": 4149772.0,
"step": 1805
},
{
"entropy": 5.791418552398682,
"epoch": 0.17387127761767532,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999316254888172,
"loss": 5.7849,
"mean_token_accuracy": 0.16932481080293654,
"num_tokens": 4162246.0,
"step": 1810
},
{
"entropy": 5.744324827194214,
"epoch": 0.17435158501440923,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999307777473157,
"loss": 5.7563,
"mean_token_accuracy": 0.17080532908439636,
"num_tokens": 4173449.0,
"step": 1815
},
{
"entropy": 5.768918180465699,
"epoch": 0.17483189241114314,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999299247835893,
"loss": 5.773,
"mean_token_accuracy": 0.16689184457063674,
"num_tokens": 4185560.0,
"step": 1820
},
{
"entropy": 5.852003479003907,
"epoch": 0.17531219980787704,
"grad_norm": 0.89453125,
"learning_rate": 0.000499929066597658,
"loss": 5.8456,
"mean_token_accuracy": 0.15657656863331795,
"num_tokens": 4197843.0,
"step": 1825
},
{
"entropy": 5.869591426849365,
"epoch": 0.17579250720461095,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999282031895418,
"loss": 5.6906,
"mean_token_accuracy": 0.16604516059160232,
"num_tokens": 4210009.0,
"step": 1830
},
{
"entropy": 5.797318696975708,
"epoch": 0.17627281460134486,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999273345592604,
"loss": 5.9445,
"mean_token_accuracy": 0.15455603450536728,
"num_tokens": 4222213.0,
"step": 1835
},
{
"entropy": 5.9059834480285645,
"epoch": 0.17675312199807877,
"grad_norm": 1.0,
"learning_rate": 0.0004999264607068343,
"loss": 5.7864,
"mean_token_accuracy": 0.16164651066064833,
"num_tokens": 4234159.0,
"step": 1840
},
{
"entropy": 5.803954219818115,
"epoch": 0.17723342939481268,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999255816322837,
"loss": 5.8304,
"mean_token_accuracy": 0.1548845611512661,
"num_tokens": 4246358.0,
"step": 1845
},
{
"entropy": 5.9636155128479,
"epoch": 0.1777137367915466,
"grad_norm": 1.015625,
"learning_rate": 0.000499924697335629,
"loss": 5.9285,
"mean_token_accuracy": 0.15726198703050615,
"num_tokens": 4256698.0,
"step": 1850
},
{
"entropy": 5.774808931350708,
"epoch": 0.1781940441882805,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999238078168906,
"loss": 5.7062,
"mean_token_accuracy": 0.17923928201198577,
"num_tokens": 4268007.0,
"step": 1855
},
{
"entropy": 5.802064657211304,
"epoch": 0.1786743515850144,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999229130760894,
"loss": 5.808,
"mean_token_accuracy": 0.1669637754559517,
"num_tokens": 4278583.0,
"step": 1860
},
{
"entropy": 5.939214849472046,
"epoch": 0.17915465898174832,
"grad_norm": 0.9609375,
"learning_rate": 0.000499922013113246,
"loss": 5.8335,
"mean_token_accuracy": 0.16115633994340897,
"num_tokens": 4290267.0,
"step": 1865
},
{
"entropy": 5.753700637817383,
"epoch": 0.17963496637848222,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999211079283814,
"loss": 5.7483,
"mean_token_accuracy": 0.16866709738969804,
"num_tokens": 4302177.0,
"step": 1870
},
{
"entropy": 5.758403539657593,
"epoch": 0.18011527377521613,
"grad_norm": 1.1015625,
"learning_rate": 0.0004999201975215164,
"loss": 5.6937,
"mean_token_accuracy": 0.16991591304540635,
"num_tokens": 4312420.0,
"step": 1875
},
{
"entropy": 5.855390214920044,
"epoch": 0.18059558117195004,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999192818926725,
"loss": 5.8369,
"mean_token_accuracy": 0.16301838606595992,
"num_tokens": 4324050.0,
"step": 1880
},
{
"entropy": 5.798628568649292,
"epoch": 0.18107588856868395,
"grad_norm": 1.078125,
"learning_rate": 0.0004999183610418706,
"loss": 5.6856,
"mean_token_accuracy": 0.16928814202547074,
"num_tokens": 4335336.0,
"step": 1885
},
{
"entropy": 5.6459338665008545,
"epoch": 0.18155619596541786,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999174349691322,
"loss": 5.7078,
"mean_token_accuracy": 0.1670696869492531,
"num_tokens": 4346239.0,
"step": 1890
},
{
"entropy": 5.83158860206604,
"epoch": 0.18203650336215177,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999165036744788,
"loss": 5.7856,
"mean_token_accuracy": 0.16568428725004197,
"num_tokens": 4357195.0,
"step": 1895
},
{
"entropy": 5.748570966720581,
"epoch": 0.18251681075888568,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999155671579322,
"loss": 5.7083,
"mean_token_accuracy": 0.16901581287384032,
"num_tokens": 4368905.0,
"step": 1900
},
{
"entropy": 5.84704818725586,
"epoch": 0.1829971181556196,
"grad_norm": 0.94140625,
"learning_rate": 0.000499914625419514,
"loss": 5.7747,
"mean_token_accuracy": 0.16582313925027847,
"num_tokens": 4381708.0,
"step": 1905
},
{
"entropy": 5.680188226699829,
"epoch": 0.1834774255523535,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999136784592459,
"loss": 5.6413,
"mean_token_accuracy": 0.18005160689353944,
"num_tokens": 4393199.0,
"step": 1910
},
{
"entropy": 5.790105772018433,
"epoch": 0.1839577329490874,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999127262771502,
"loss": 5.8594,
"mean_token_accuracy": 0.1600307747721672,
"num_tokens": 4403986.0,
"step": 1915
},
{
"entropy": 5.8221275806427,
"epoch": 0.1844380403458213,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999117688732487,
"loss": 5.7076,
"mean_token_accuracy": 0.16819000989198685,
"num_tokens": 4415574.0,
"step": 1920
},
{
"entropy": 5.743313217163086,
"epoch": 0.18491834774255522,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999108062475638,
"loss": 5.7407,
"mean_token_accuracy": 0.17155899852514267,
"num_tokens": 4427579.0,
"step": 1925
},
{
"entropy": 5.785640573501587,
"epoch": 0.18539865513928913,
"grad_norm": 0.91796875,
"learning_rate": 0.000499909838400118,
"loss": 5.7394,
"mean_token_accuracy": 0.16696204692125322,
"num_tokens": 4440045.0,
"step": 1930
},
{
"entropy": 5.786698246002198,
"epoch": 0.18587896253602307,
"grad_norm": 0.984375,
"learning_rate": 0.0004999088653309334,
"loss": 5.828,
"mean_token_accuracy": 0.1624658688902855,
"num_tokens": 4450129.0,
"step": 1935
},
{
"entropy": 5.755189228057861,
"epoch": 0.18635926993275698,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999078870400329,
"loss": 5.6705,
"mean_token_accuracy": 0.1707078069448471,
"num_tokens": 4461378.0,
"step": 1940
},
{
"entropy": 5.773273372650147,
"epoch": 0.18683957732949089,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999069035274391,
"loss": 5.7133,
"mean_token_accuracy": 0.17081280052661896,
"num_tokens": 4472249.0,
"step": 1945
},
{
"entropy": 5.776774835586548,
"epoch": 0.1873198847262248,
"grad_norm": 0.921875,
"learning_rate": 0.0004999059147931747,
"loss": 5.7758,
"mean_token_accuracy": 0.16512089669704438,
"num_tokens": 4484243.0,
"step": 1950
},
{
"entropy": 5.780983924865723,
"epoch": 0.1878001921229587,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999049208372629,
"loss": 5.6782,
"mean_token_accuracy": 0.16991802006959916,
"num_tokens": 4495776.0,
"step": 1955
},
{
"entropy": 5.752791500091552,
"epoch": 0.1882804995196926,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999039216597267,
"loss": 5.7767,
"mean_token_accuracy": 0.1656827986240387,
"num_tokens": 4508581.0,
"step": 1960
},
{
"entropy": 5.813444137573242,
"epoch": 0.18876080691642652,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999029172605892,
"loss": 5.7015,
"mean_token_accuracy": 0.168874491751194,
"num_tokens": 4519753.0,
"step": 1965
},
{
"entropy": 5.7711827754974365,
"epoch": 0.18924111431316043,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999019076398738,
"loss": 5.7721,
"mean_token_accuracy": 0.1760159507393837,
"num_tokens": 4529957.0,
"step": 1970
},
{
"entropy": 5.800232076644898,
"epoch": 0.18972142170989434,
"grad_norm": 0.9765625,
"learning_rate": 0.000499900892797604,
"loss": 5.7455,
"mean_token_accuracy": 0.16272206157445906,
"num_tokens": 4541181.0,
"step": 1975
},
{
"entropy": 5.730713272094727,
"epoch": 0.19020172910662825,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998998727338031,
"loss": 5.7044,
"mean_token_accuracy": 0.16977352052927017,
"num_tokens": 4552235.0,
"step": 1980
},
{
"entropy": 5.76020998954773,
"epoch": 0.19068203650336216,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998988474484952,
"loss": 5.7425,
"mean_token_accuracy": 0.17006415724754334,
"num_tokens": 4564045.0,
"step": 1985
},
{
"entropy": 5.718191051483155,
"epoch": 0.19116234390009607,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998978169417038,
"loss": 5.6126,
"mean_token_accuracy": 0.17550907731056214,
"num_tokens": 4575141.0,
"step": 1990
},
{
"entropy": 5.641534996032715,
"epoch": 0.19164265129682997,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998967812134529,
"loss": 5.739,
"mean_token_accuracy": 0.16591082215309144,
"num_tokens": 4586491.0,
"step": 1995
},
{
"entropy": 5.800497817993164,
"epoch": 0.19212295869356388,
"grad_norm": 1.03125,
"learning_rate": 0.0004998957402637664,
"loss": 5.7131,
"mean_token_accuracy": 0.17069052755832673,
"num_tokens": 4598421.0,
"step": 2000
},
{
"entropy": 5.761542940139771,
"epoch": 0.1926032660902978,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998946940926687,
"loss": 5.634,
"mean_token_accuracy": 0.1821053981781006,
"num_tokens": 4609688.0,
"step": 2005
},
{
"entropy": 5.751686668395996,
"epoch": 0.1930835734870317,
"grad_norm": 0.95703125,
"learning_rate": 0.000499893642700184,
"loss": 5.7993,
"mean_token_accuracy": 0.16548632085323334,
"num_tokens": 4621169.0,
"step": 2010
},
{
"entropy": 5.7683569431304935,
"epoch": 0.1935638808837656,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998925860863368,
"loss": 5.733,
"mean_token_accuracy": 0.17027136385440828,
"num_tokens": 4632512.0,
"step": 2015
},
{
"entropy": 5.796252727508545,
"epoch": 0.19404418828049952,
"grad_norm": 0.921875,
"learning_rate": 0.0004998915242511516,
"loss": 5.8012,
"mean_token_accuracy": 0.16176183074712752,
"num_tokens": 4644356.0,
"step": 2020
},
{
"entropy": 5.854101514816284,
"epoch": 0.19452449567723343,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998904571946528,
"loss": 5.7745,
"mean_token_accuracy": 0.16625221967697143,
"num_tokens": 4655601.0,
"step": 2025
},
{
"entropy": 5.79165997505188,
"epoch": 0.19500480307396734,
"grad_norm": 1.0859375,
"learning_rate": 0.0004998893849168655,
"loss": 5.7786,
"mean_token_accuracy": 0.16630641520023345,
"num_tokens": 4666813.0,
"step": 2030
},
{
"entropy": 5.824037742614746,
"epoch": 0.19548511047070125,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998883074178144,
"loss": 5.7745,
"mean_token_accuracy": 0.16555878072977065,
"num_tokens": 4679123.0,
"step": 2035
},
{
"entropy": 5.764426136016846,
"epoch": 0.19596541786743515,
"grad_norm": 1.0703125,
"learning_rate": 0.0004998872246975247,
"loss": 5.6658,
"mean_token_accuracy": 0.17124608904123306,
"num_tokens": 4690485.0,
"step": 2040
},
{
"entropy": 5.614950132369995,
"epoch": 0.19644572526416906,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998861367560213,
"loss": 5.6973,
"mean_token_accuracy": 0.1706745520234108,
"num_tokens": 4702573.0,
"step": 2045
},
{
"entropy": 5.8513716697692875,
"epoch": 0.19692603266090297,
"grad_norm": 1.015625,
"learning_rate": 0.0004998850435933296,
"loss": 5.7228,
"mean_token_accuracy": 0.16488247960805893,
"num_tokens": 4715653.0,
"step": 2050
},
{
"entropy": 5.668337106704712,
"epoch": 0.19740634005763688,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998839452094749,
"loss": 5.7278,
"mean_token_accuracy": 0.1701984152197838,
"num_tokens": 4726051.0,
"step": 2055
},
{
"entropy": 5.768125200271607,
"epoch": 0.1978866474543708,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998828416044829,
"loss": 5.8616,
"mean_token_accuracy": 0.1592288315296173,
"num_tokens": 4738176.0,
"step": 2060
},
{
"entropy": 5.79635853767395,
"epoch": 0.1983669548511047,
"grad_norm": 0.95703125,
"learning_rate": 0.000499881732778379,
"loss": 5.761,
"mean_token_accuracy": 0.1653179332613945,
"num_tokens": 4749547.0,
"step": 2065
},
{
"entropy": 5.7004289627075195,
"epoch": 0.1988472622478386,
"grad_norm": 0.953125,
"learning_rate": 0.000499880618731189,
"loss": 5.673,
"mean_token_accuracy": 0.16624890714883805,
"num_tokens": 4761967.0,
"step": 2070
},
{
"entropy": 5.728426837921143,
"epoch": 0.19932756964457252,
"grad_norm": 1.03125,
"learning_rate": 0.0004998794994629388,
"loss": 5.6654,
"mean_token_accuracy": 0.17760985493659973,
"num_tokens": 4772847.0,
"step": 2075
},
{
"entropy": 5.7485308170318605,
"epoch": 0.19980787704130643,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998783749736545,
"loss": 5.7332,
"mean_token_accuracy": 0.1665830835700035,
"num_tokens": 4785049.0,
"step": 2080
},
{
"entropy": 5.779238176345825,
"epoch": 0.20028818443804033,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998772452633619,
"loss": 5.7618,
"mean_token_accuracy": 0.1638600841164589,
"num_tokens": 4797484.0,
"step": 2085
},
{
"entropy": 5.727624607086182,
"epoch": 0.20076849183477424,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998761103320876,
"loss": 5.6787,
"mean_token_accuracy": 0.16071517616510392,
"num_tokens": 4809422.0,
"step": 2090
},
{
"entropy": 5.798519659042358,
"epoch": 0.20124879923150815,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998749701798577,
"loss": 5.8628,
"mean_token_accuracy": 0.164070263504982,
"num_tokens": 4822340.0,
"step": 2095
},
{
"entropy": 5.751769924163819,
"epoch": 0.2017291066282421,
"grad_norm": 1.09375,
"learning_rate": 0.0004998738248066986,
"loss": 5.6535,
"mean_token_accuracy": 0.17146731466054915,
"num_tokens": 4833382.0,
"step": 2100
},
{
"entropy": 5.6563629627227785,
"epoch": 0.202209414024976,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998726742126372,
"loss": 5.6462,
"mean_token_accuracy": 0.1777672901749611,
"num_tokens": 4843916.0,
"step": 2105
},
{
"entropy": 5.783173704147339,
"epoch": 0.2026897214217099,
"grad_norm": 0.9375,
"learning_rate": 0.0004998715183976999,
"loss": 5.6659,
"mean_token_accuracy": 0.17276445925235748,
"num_tokens": 4855420.0,
"step": 2110
},
{
"entropy": 5.747212839126587,
"epoch": 0.20317002881844382,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998703573619137,
"loss": 5.7408,
"mean_token_accuracy": 0.16771334260702134,
"num_tokens": 4866540.0,
"step": 2115
},
{
"entropy": 5.756590795516968,
"epoch": 0.20365033621517772,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998691911053056,
"loss": 5.6757,
"mean_token_accuracy": 0.16594192534685134,
"num_tokens": 4878860.0,
"step": 2120
},
{
"entropy": 5.843172788619995,
"epoch": 0.20413064361191163,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998680196279026,
"loss": 5.8017,
"mean_token_accuracy": 0.16326622664928436,
"num_tokens": 4889692.0,
"step": 2125
},
{
"entropy": 5.814892292022705,
"epoch": 0.20461095100864554,
"grad_norm": 0.96875,
"learning_rate": 0.0004998668429297319,
"loss": 5.745,
"mean_token_accuracy": 0.1690722644329071,
"num_tokens": 4901337.0,
"step": 2130
},
{
"entropy": 5.6207257270812985,
"epoch": 0.20509125840537945,
"grad_norm": 0.9765625,
"learning_rate": 0.0004998656610108208,
"loss": 5.6647,
"mean_token_accuracy": 0.175962170958519,
"num_tokens": 4912340.0,
"step": 2135
},
{
"entropy": 5.7022956848144535,
"epoch": 0.20557156580211336,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998644738711969,
"loss": 5.633,
"mean_token_accuracy": 0.17715211808681489,
"num_tokens": 4923429.0,
"step": 2140
},
{
"entropy": 5.744864225387573,
"epoch": 0.20605187319884727,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998632815108874,
"loss": 5.7004,
"mean_token_accuracy": 0.16438920944929122,
"num_tokens": 4934826.0,
"step": 2145
},
{
"entropy": 5.779660034179687,
"epoch": 0.20653218059558118,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998620839299203,
"loss": 5.7914,
"mean_token_accuracy": 0.16467878818511963,
"num_tokens": 4945189.0,
"step": 2150
},
{
"entropy": 5.814841079711914,
"epoch": 0.2070124879923151,
"grad_norm": 1.03125,
"learning_rate": 0.0004998608811283233,
"loss": 5.6626,
"mean_token_accuracy": 0.17581747621297836,
"num_tokens": 4956060.0,
"step": 2155
},
{
"entropy": 5.692324304580689,
"epoch": 0.207492795389049,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998596731061244,
"loss": 5.6785,
"mean_token_accuracy": 0.17047660052776337,
"num_tokens": 4967710.0,
"step": 2160
},
{
"entropy": 5.760519409179688,
"epoch": 0.2079731027857829,
"grad_norm": 1.0,
"learning_rate": 0.0004998584598633516,
"loss": 5.7035,
"mean_token_accuracy": 0.16868236809968948,
"num_tokens": 4978716.0,
"step": 2165
},
{
"entropy": 5.747573614120483,
"epoch": 0.2084534101825168,
"grad_norm": 1.0390625,
"learning_rate": 0.0004998572414000329,
"loss": 5.7636,
"mean_token_accuracy": 0.16640378534793854,
"num_tokens": 4990742.0,
"step": 2170
},
{
"entropy": 5.754537057876587,
"epoch": 0.20893371757925072,
"grad_norm": 1.09375,
"learning_rate": 0.0004998560177161969,
"loss": 5.7229,
"mean_token_accuracy": 0.1728250965476036,
"num_tokens": 5001217.0,
"step": 2175
},
{
"entropy": 5.7619335651397705,
"epoch": 0.20941402497598463,
"grad_norm": 0.9140625,
"learning_rate": 0.0004998547888118718,
"loss": 5.7089,
"mean_token_accuracy": 0.16733649522066116,
"num_tokens": 5012750.0,
"step": 2180
},
{
"entropy": 5.703377151489258,
"epoch": 0.20989433237271854,
"grad_norm": 1.046875,
"learning_rate": 0.0004998535546870862,
"loss": 5.7162,
"mean_token_accuracy": 0.17622778564691544,
"num_tokens": 5023818.0,
"step": 2185
},
{
"entropy": 5.8201076030731205,
"epoch": 0.21037463976945245,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998523153418687,
"loss": 5.7395,
"mean_token_accuracy": 0.16917874813079833,
"num_tokens": 5035060.0,
"step": 2190
},
{
"entropy": 5.7587230682373045,
"epoch": 0.21085494716618636,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998510707762481,
"loss": 5.7859,
"mean_token_accuracy": 0.16076416671276092,
"num_tokens": 5046780.0,
"step": 2195
},
{
"entropy": 5.703002882003784,
"epoch": 0.21133525456292027,
"grad_norm": 1.0625,
"learning_rate": 0.0004998498209902533,
"loss": 5.7498,
"mean_token_accuracy": 0.1657275453209877,
"num_tokens": 5059152.0,
"step": 2200
},
{
"entropy": 5.767661762237549,
"epoch": 0.21181556195965417,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998485659839134,
"loss": 5.7165,
"mean_token_accuracy": 0.16900938898324966,
"num_tokens": 5069971.0,
"step": 2205
},
{
"entropy": 5.757136583328247,
"epoch": 0.21229586935638808,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998473057572575,
"loss": 5.6553,
"mean_token_accuracy": 0.17132542431354522,
"num_tokens": 5080640.0,
"step": 2210
},
{
"entropy": 5.691873693466187,
"epoch": 0.212776176753122,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998460403103146,
"loss": 5.6686,
"mean_token_accuracy": 0.17301307767629623,
"num_tokens": 5091240.0,
"step": 2215
},
{
"entropy": 5.691802358627319,
"epoch": 0.2132564841498559,
"grad_norm": 1.015625,
"learning_rate": 0.0004998447696431146,
"loss": 5.7072,
"mean_token_accuracy": 0.16933174431324005,
"num_tokens": 5102920.0,
"step": 2220
},
{
"entropy": 5.7080395221710205,
"epoch": 0.2137367915465898,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998434937556865,
"loss": 5.714,
"mean_token_accuracy": 0.16653985381126404,
"num_tokens": 5114240.0,
"step": 2225
},
{
"entropy": 5.8630828857421875,
"epoch": 0.21421709894332372,
"grad_norm": 1.0078125,
"learning_rate": 0.0004998422126480602,
"loss": 5.9345,
"mean_token_accuracy": 0.1564141757786274,
"num_tokens": 5126653.0,
"step": 2230
},
{
"entropy": 5.923197984695435,
"epoch": 0.21469740634005763,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998409263202653,
"loss": 5.7951,
"mean_token_accuracy": 0.1645052909851074,
"num_tokens": 5138778.0,
"step": 2235
},
{
"entropy": 5.697419023513794,
"epoch": 0.21517771373679154,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998396347723318,
"loss": 5.6409,
"mean_token_accuracy": 0.17507038712501527,
"num_tokens": 5148670.0,
"step": 2240
},
{
"entropy": 5.607378005981445,
"epoch": 0.21565802113352545,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998383380042895,
"loss": 5.6838,
"mean_token_accuracy": 0.1660774126648903,
"num_tokens": 5159617.0,
"step": 2245
},
{
"entropy": 5.793371343612671,
"epoch": 0.21613832853025935,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998370360161688,
"loss": 5.7412,
"mean_token_accuracy": 0.16822775602340698,
"num_tokens": 5171225.0,
"step": 2250
},
{
"entropy": 5.766514635086059,
"epoch": 0.21661863592699326,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998357288079996,
"loss": 5.7291,
"mean_token_accuracy": 0.16446396410465242,
"num_tokens": 5184586.0,
"step": 2255
},
{
"entropy": 5.653764152526856,
"epoch": 0.21709894332372717,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998344163798125,
"loss": 5.6851,
"mean_token_accuracy": 0.16855508089065552,
"num_tokens": 5196089.0,
"step": 2260
},
{
"entropy": 5.756007528305053,
"epoch": 0.21757925072046108,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998330987316379,
"loss": 5.7378,
"mean_token_accuracy": 0.16411580741405488,
"num_tokens": 5207581.0,
"step": 2265
},
{
"entropy": 5.717509746551514,
"epoch": 0.21805955811719502,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998317758635062,
"loss": 5.6875,
"mean_token_accuracy": 0.17126210331916808,
"num_tokens": 5219108.0,
"step": 2270
},
{
"entropy": 5.717676448822021,
"epoch": 0.21853986551392893,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998304477754484,
"loss": 5.6361,
"mean_token_accuracy": 0.17278024703264236,
"num_tokens": 5231137.0,
"step": 2275
},
{
"entropy": 5.830537891387939,
"epoch": 0.21902017291066284,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998291144674952,
"loss": 5.7006,
"mean_token_accuracy": 0.17262052595615388,
"num_tokens": 5243528.0,
"step": 2280
},
{
"entropy": 5.570743799209595,
"epoch": 0.21950048030739674,
"grad_norm": 1.046875,
"learning_rate": 0.0004998277759396776,
"loss": 5.6544,
"mean_token_accuracy": 0.17672596722841263,
"num_tokens": 5254869.0,
"step": 2285
},
{
"entropy": 5.6864776611328125,
"epoch": 0.21998078770413065,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998264321920265,
"loss": 5.6374,
"mean_token_accuracy": 0.17273785322904586,
"num_tokens": 5265370.0,
"step": 2290
},
{
"entropy": 5.709369754791259,
"epoch": 0.22046109510086456,
"grad_norm": 0.94921875,
"learning_rate": 0.0004998250832245734,
"loss": 5.6108,
"mean_token_accuracy": 0.17232156842947005,
"num_tokens": 5277812.0,
"step": 2295
},
{
"entropy": 5.591331195831299,
"epoch": 0.22094140249759847,
"grad_norm": 1.0625,
"learning_rate": 0.0004998237290373494,
"loss": 5.5895,
"mean_token_accuracy": 0.17628773897886277,
"num_tokens": 5289107.0,
"step": 2300
},
{
"entropy": 5.745526599884033,
"epoch": 0.22142170989433238,
"grad_norm": 1.0703125,
"learning_rate": 0.000499822369630386,
"loss": 5.659,
"mean_token_accuracy": 0.16332831233739853,
"num_tokens": 5301489.0,
"step": 2305
},
{
"entropy": 5.575994968414307,
"epoch": 0.2219020172910663,
"grad_norm": 0.921875,
"learning_rate": 0.0004998210050037148,
"loss": 5.625,
"mean_token_accuracy": 0.17718679308891297,
"num_tokens": 5313273.0,
"step": 2310
},
{
"entropy": 5.763540124893188,
"epoch": 0.2223823246878002,
"grad_norm": 0.98828125,
"learning_rate": 0.0004998196351573674,
"loss": 5.6748,
"mean_token_accuracy": 0.16899870485067367,
"num_tokens": 5324227.0,
"step": 2315
},
{
"entropy": 5.596045875549317,
"epoch": 0.2228626320845341,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998182600913757,
"loss": 5.6037,
"mean_token_accuracy": 0.1776882916688919,
"num_tokens": 5335138.0,
"step": 2320
},
{
"entropy": 5.692679834365845,
"epoch": 0.22334293948126802,
"grad_norm": 0.984375,
"learning_rate": 0.0004998168798057715,
"loss": 5.6801,
"mean_token_accuracy": 0.17184858918190002,
"num_tokens": 5345652.0,
"step": 2325
},
{
"entropy": 5.842924690246582,
"epoch": 0.22382324687800192,
"grad_norm": 0.92578125,
"learning_rate": 0.000499815494300587,
"loss": 5.7807,
"mean_token_accuracy": 0.16769531071186067,
"num_tokens": 5357999.0,
"step": 2330
},
{
"entropy": 5.695201444625854,
"epoch": 0.22430355427473583,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998141035758542,
"loss": 5.6655,
"mean_token_accuracy": 0.16747171431779861,
"num_tokens": 5369032.0,
"step": 2335
},
{
"entropy": 5.685867786407471,
"epoch": 0.22478386167146974,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998127076316054,
"loss": 5.5858,
"mean_token_accuracy": 0.17415026128292083,
"num_tokens": 5380692.0,
"step": 2340
},
{
"entropy": 5.738334608078003,
"epoch": 0.22526416906820365,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998113064678734,
"loss": 5.7279,
"mean_token_accuracy": 0.16383333057165145,
"num_tokens": 5392320.0,
"step": 2345
},
{
"entropy": 5.717440557479859,
"epoch": 0.22574447646493756,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998099000846901,
"loss": 5.7129,
"mean_token_accuracy": 0.17294058203697205,
"num_tokens": 5404374.0,
"step": 2350
},
{
"entropy": 5.743234443664551,
"epoch": 0.22622478386167147,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998084884820887,
"loss": 5.7044,
"mean_token_accuracy": 0.16908725798130037,
"num_tokens": 5415471.0,
"step": 2355
},
{
"entropy": 5.572421836853027,
"epoch": 0.22670509125840538,
"grad_norm": 1.1328125,
"learning_rate": 0.0004998070716601016,
"loss": 5.5376,
"mean_token_accuracy": 0.1757811814546585,
"num_tokens": 5425971.0,
"step": 2360
},
{
"entropy": 5.761317300796509,
"epoch": 0.2271853986551393,
"grad_norm": 1.0,
"learning_rate": 0.0004998056496187618,
"loss": 5.6628,
"mean_token_accuracy": 0.17541870921850206,
"num_tokens": 5437638.0,
"step": 2365
},
{
"entropy": 5.695319604873657,
"epoch": 0.2276657060518732,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998042223581025,
"loss": 5.688,
"mean_token_accuracy": 0.17006312757730485,
"num_tokens": 5449463.0,
"step": 2370
},
{
"entropy": 5.75418791770935,
"epoch": 0.2281460134486071,
"grad_norm": 1.1171875,
"learning_rate": 0.0004998027898781565,
"loss": 5.7672,
"mean_token_accuracy": 0.16644203364849092,
"num_tokens": 5460972.0,
"step": 2375
},
{
"entropy": 5.873421955108642,
"epoch": 0.228626320845341,
"grad_norm": 0.96875,
"learning_rate": 0.0004998013521789574,
"loss": 5.7659,
"mean_token_accuracy": 0.1616092711687088,
"num_tokens": 5471575.0,
"step": 2380
},
{
"entropy": 5.647591924667358,
"epoch": 0.22910662824207492,
"grad_norm": 1.09375,
"learning_rate": 0.0004997999092605384,
"loss": 5.6174,
"mean_token_accuracy": 0.17019406706094742,
"num_tokens": 5483041.0,
"step": 2385
},
{
"entropy": 5.656691837310791,
"epoch": 0.22958693563880883,
"grad_norm": 0.9921875,
"learning_rate": 0.000499798461122933,
"loss": 5.6548,
"mean_token_accuracy": 0.17062922716140747,
"num_tokens": 5494040.0,
"step": 2390
},
{
"entropy": 5.595247745513916,
"epoch": 0.23006724303554274,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997970077661748,
"loss": 5.5368,
"mean_token_accuracy": 0.18294250816106797,
"num_tokens": 5504217.0,
"step": 2395
},
{
"entropy": 5.761266231536865,
"epoch": 0.23054755043227665,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997955491902977,
"loss": 5.7086,
"mean_token_accuracy": 0.16774140298366547,
"num_tokens": 5517180.0,
"step": 2400
},
{
"entropy": 5.650377893447876,
"epoch": 0.23102785782901056,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997940853953354,
"loss": 5.6279,
"mean_token_accuracy": 0.1671198323369026,
"num_tokens": 5529626.0,
"step": 2405
},
{
"entropy": 5.7746195793151855,
"epoch": 0.23150816522574447,
"grad_norm": 0.953125,
"learning_rate": 0.000499792616381322,
"loss": 5.6646,
"mean_token_accuracy": 0.1768963173031807,
"num_tokens": 5540844.0,
"step": 2410
},
{
"entropy": 5.665439796447754,
"epoch": 0.23198847262247838,
"grad_norm": 1.03125,
"learning_rate": 0.0004997911421482914,
"loss": 5.5842,
"mean_token_accuracy": 0.1775585025548935,
"num_tokens": 5551732.0,
"step": 2415
},
{
"entropy": 5.7055745124816895,
"epoch": 0.23246878001921228,
"grad_norm": 1.0234375,
"learning_rate": 0.000499789662696278,
"loss": 5.7224,
"mean_token_accuracy": 0.17121947556734085,
"num_tokens": 5562692.0,
"step": 2420
},
{
"entropy": 5.830398082733154,
"epoch": 0.2329490874159462,
"grad_norm": 1.015625,
"learning_rate": 0.0004997881780253162,
"loss": 5.7108,
"mean_token_accuracy": 0.16949707716703416,
"num_tokens": 5574884.0,
"step": 2425
},
{
"entropy": 5.5266454219818115,
"epoch": 0.2334293948126801,
"grad_norm": 1.078125,
"learning_rate": 0.0004997866881354403,
"loss": 5.5626,
"mean_token_accuracy": 0.17615221440792084,
"num_tokens": 5585335.0,
"step": 2430
},
{
"entropy": 5.79819769859314,
"epoch": 0.23390970220941404,
"grad_norm": 1.0,
"learning_rate": 0.000499785193026685,
"loss": 5.7122,
"mean_token_accuracy": 0.16447871327400207,
"num_tokens": 5597621.0,
"step": 2435
},
{
"entropy": 5.555981016159057,
"epoch": 0.23439000960614795,
"grad_norm": 0.96875,
"learning_rate": 0.0004997836926990851,
"loss": 5.567,
"mean_token_accuracy": 0.1729709431529045,
"num_tokens": 5609795.0,
"step": 2440
},
{
"entropy": 5.7423711776733395,
"epoch": 0.23487031700288186,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997821871526752,
"loss": 5.6678,
"mean_token_accuracy": 0.17034737765789032,
"num_tokens": 5621473.0,
"step": 2445
},
{
"entropy": 5.803370666503906,
"epoch": 0.23535062439961577,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997806763874905,
"loss": 5.7509,
"mean_token_accuracy": 0.1693995103240013,
"num_tokens": 5635064.0,
"step": 2450
},
{
"entropy": 5.740002870559692,
"epoch": 0.23583093179634967,
"grad_norm": 1.015625,
"learning_rate": 0.0004997791604035659,
"loss": 5.7351,
"mean_token_accuracy": 0.1653935804963112,
"num_tokens": 5647382.0,
"step": 2455
},
{
"entropy": 5.6946430683135985,
"epoch": 0.23631123919308358,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997776392009366,
"loss": 5.6227,
"mean_token_accuracy": 0.17229534089565277,
"num_tokens": 5658417.0,
"step": 2460
},
{
"entropy": 5.633540248870849,
"epoch": 0.2367915465898175,
"grad_norm": 0.890625,
"learning_rate": 0.0004997761127796381,
"loss": 5.5924,
"mean_token_accuracy": 0.1780801773071289,
"num_tokens": 5672190.0,
"step": 2465
},
{
"entropy": 5.630267906188965,
"epoch": 0.2372718539865514,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997745811397056,
"loss": 5.6343,
"mean_token_accuracy": 0.17877777218818663,
"num_tokens": 5684259.0,
"step": 2470
},
{
"entropy": 5.684125900268555,
"epoch": 0.2377521613832853,
"grad_norm": 1.0,
"learning_rate": 0.0004997730442811748,
"loss": 5.6829,
"mean_token_accuracy": 0.1745256558060646,
"num_tokens": 5696505.0,
"step": 2475
},
{
"entropy": 5.632245492935181,
"epoch": 0.23823246878001922,
"grad_norm": 1.0,
"learning_rate": 0.0004997715022040814,
"loss": 5.5737,
"mean_token_accuracy": 0.17421737164258957,
"num_tokens": 5707638.0,
"step": 2480
},
{
"entropy": 5.642940044403076,
"epoch": 0.23871277617675313,
"grad_norm": 0.953125,
"learning_rate": 0.000499769954908461,
"loss": 5.5426,
"mean_token_accuracy": 0.1817597672343254,
"num_tokens": 5719098.0,
"step": 2485
},
{
"entropy": 5.824435234069824,
"epoch": 0.23919308357348704,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997684023943498,
"loss": 5.7816,
"mean_token_accuracy": 0.15434520691633224,
"num_tokens": 5731345.0,
"step": 2490
},
{
"entropy": 5.546219301223755,
"epoch": 0.23967339097022095,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997668446617837,
"loss": 5.5331,
"mean_token_accuracy": 0.17521776705980302,
"num_tokens": 5742604.0,
"step": 2495
},
{
"entropy": 5.592561054229736,
"epoch": 0.24015369836695485,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997652817107989,
"loss": 5.5647,
"mean_token_accuracy": 0.1803002640604973,
"num_tokens": 5754620.0,
"step": 2500
},
{
"entropy": 5.711943244934082,
"epoch": 0.24063400576368876,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997637135414315,
"loss": 5.6407,
"mean_token_accuracy": 0.17505031377077102,
"num_tokens": 5764764.0,
"step": 2505
},
{
"entropy": 5.5460649013519285,
"epoch": 0.24111431316042267,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997621401537183,
"loss": 5.5314,
"mean_token_accuracy": 0.1876803979277611,
"num_tokens": 5776341.0,
"step": 2510
},
{
"entropy": 5.738549900054932,
"epoch": 0.24159462055715658,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997605615476955,
"loss": 5.7606,
"mean_token_accuracy": 0.16073987782001495,
"num_tokens": 5788604.0,
"step": 2515
},
{
"entropy": 5.74634370803833,
"epoch": 0.2420749279538905,
"grad_norm": 1.125,
"learning_rate": 0.0004997589777234,
"loss": 5.6553,
"mean_token_accuracy": 0.17228216230869292,
"num_tokens": 5800194.0,
"step": 2520
},
{
"entropy": 5.6268805980682375,
"epoch": 0.2425552353506244,
"grad_norm": 0.9453125,
"learning_rate": 0.0004997573886808684,
"loss": 5.6295,
"mean_token_accuracy": 0.17930565178394317,
"num_tokens": 5812978.0,
"step": 2525
},
{
"entropy": 5.637392854690551,
"epoch": 0.2430355427473583,
"grad_norm": 0.953125,
"learning_rate": 0.0004997557944201375,
"loss": 5.5556,
"mean_token_accuracy": 0.18395199477672577,
"num_tokens": 5824591.0,
"step": 2530
},
{
"entropy": 5.614971017837524,
"epoch": 0.24351585014409222,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997541949412445,
"loss": 5.6467,
"mean_token_accuracy": 0.1709784209728241,
"num_tokens": 5835970.0,
"step": 2535
},
{
"entropy": 5.781296730041504,
"epoch": 0.24399615754082613,
"grad_norm": 0.984375,
"learning_rate": 0.0004997525902442266,
"loss": 5.6819,
"mean_token_accuracy": 0.1686740532517433,
"num_tokens": 5848687.0,
"step": 2540
},
{
"entropy": 5.643764638900757,
"epoch": 0.24447646493756003,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997509803291207,
"loss": 5.6604,
"mean_token_accuracy": 0.17258683294057847,
"num_tokens": 5860820.0,
"step": 2545
},
{
"entropy": 5.67121114730835,
"epoch": 0.24495677233429394,
"grad_norm": 1.0,
"learning_rate": 0.0004997493651959647,
"loss": 5.5907,
"mean_token_accuracy": 0.1715891554951668,
"num_tokens": 5872172.0,
"step": 2550
},
{
"entropy": 5.692111682891846,
"epoch": 0.24543707973102785,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997477448447955,
"loss": 5.6336,
"mean_token_accuracy": 0.17338863462209703,
"num_tokens": 5883574.0,
"step": 2555
},
{
"entropy": 5.726970052719116,
"epoch": 0.24591738712776176,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997461192756512,
"loss": 5.7262,
"mean_token_accuracy": 0.17206473648548126,
"num_tokens": 5895223.0,
"step": 2560
},
{
"entropy": 5.618826961517334,
"epoch": 0.24639769452449567,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997444884885694,
"loss": 5.5582,
"mean_token_accuracy": 0.17600789666175842,
"num_tokens": 5907448.0,
"step": 2565
},
{
"entropy": 5.683489036560059,
"epoch": 0.24687800192122958,
"grad_norm": 0.97265625,
"learning_rate": 0.0004997428524835879,
"loss": 5.6606,
"mean_token_accuracy": 0.17142950296401976,
"num_tokens": 5918906.0,
"step": 2570
},
{
"entropy": 5.6601954936981205,
"epoch": 0.2473583093179635,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997412112607446,
"loss": 5.5257,
"mean_token_accuracy": 0.1818102404475212,
"num_tokens": 5930619.0,
"step": 2575
},
{
"entropy": 5.708859300613403,
"epoch": 0.2478386167146974,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997395648200778,
"loss": 5.728,
"mean_token_accuracy": 0.16916778534650803,
"num_tokens": 5941554.0,
"step": 2580
},
{
"entropy": 5.779385662078857,
"epoch": 0.2483189241114313,
"grad_norm": 0.9375,
"learning_rate": 0.0004997379131616257,
"loss": 5.7036,
"mean_token_accuracy": 0.17098585814237593,
"num_tokens": 5953309.0,
"step": 2585
},
{
"entropy": 5.650564956665039,
"epoch": 0.24879923150816521,
"grad_norm": 1.1015625,
"learning_rate": 0.0004997362562854266,
"loss": 5.645,
"mean_token_accuracy": 0.17450834512710572,
"num_tokens": 5965286.0,
"step": 2590
},
{
"entropy": 5.632644748687744,
"epoch": 0.24927953890489912,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997345941915187,
"loss": 5.6166,
"mean_token_accuracy": 0.17439770251512526,
"num_tokens": 5976558.0,
"step": 2595
},
{
"entropy": 5.684119987487793,
"epoch": 0.24975984630163303,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997329268799412,
"loss": 5.662,
"mean_token_accuracy": 0.16537658944725991,
"num_tokens": 5988793.0,
"step": 2600
},
{
"entropy": 5.709760332107544,
"epoch": 0.25024015369836694,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997312543507322,
"loss": 5.6874,
"mean_token_accuracy": 0.16319903284311293,
"num_tokens": 5999825.0,
"step": 2605
},
{
"entropy": 5.706351470947266,
"epoch": 0.2507204610951009,
"grad_norm": 1.0,
"learning_rate": 0.0004997295766039309,
"loss": 5.6318,
"mean_token_accuracy": 0.170637284219265,
"num_tokens": 6010408.0,
"step": 2610
},
{
"entropy": 5.758199882507324,
"epoch": 0.25120076849183476,
"grad_norm": 1.015625,
"learning_rate": 0.0004997278936395761,
"loss": 5.6601,
"mean_token_accuracy": 0.17168794125318526,
"num_tokens": 6021161.0,
"step": 2615
},
{
"entropy": 5.557199287414551,
"epoch": 0.2516810758885687,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997262054577071,
"loss": 5.5178,
"mean_token_accuracy": 0.181975619494915,
"num_tokens": 6032646.0,
"step": 2620
},
{
"entropy": 5.698000192642212,
"epoch": 0.2521613832853026,
"grad_norm": 0.96484375,
"learning_rate": 0.0004997245120583627,
"loss": 5.6845,
"mean_token_accuracy": 0.17188096046447754,
"num_tokens": 6045019.0,
"step": 2625
},
{
"entropy": 5.527618551254273,
"epoch": 0.2526416906820365,
"grad_norm": 0.96875,
"learning_rate": 0.0004997228134415825,
"loss": 5.5628,
"mean_token_accuracy": 0.17528536915779114,
"num_tokens": 6055644.0,
"step": 2630
},
{
"entropy": 5.727271509170532,
"epoch": 0.2531219980787704,
"grad_norm": 1.0625,
"learning_rate": 0.0004997211096074059,
"loss": 5.6052,
"mean_token_accuracy": 0.17573324143886565,
"num_tokens": 6066488.0,
"step": 2635
},
{
"entropy": 5.630525588989258,
"epoch": 0.25360230547550433,
"grad_norm": 1.0390625,
"learning_rate": 0.0004997194005558722,
"loss": 5.6224,
"mean_token_accuracy": 0.17289988249540328,
"num_tokens": 6078300.0,
"step": 2640
},
{
"entropy": 5.664487886428833,
"epoch": 0.2540826128722382,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997176862870216,
"loss": 5.5989,
"mean_token_accuracy": 0.1705970972776413,
"num_tokens": 6089180.0,
"step": 2645
},
{
"entropy": 5.684300136566162,
"epoch": 0.25456292026897215,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997159668008933,
"loss": 5.6268,
"mean_token_accuracy": 0.18017285764217378,
"num_tokens": 6099093.0,
"step": 2650
},
{
"entropy": 5.627106094360352,
"epoch": 0.25504322766570603,
"grad_norm": 1.109375,
"learning_rate": 0.0004997142420975277,
"loss": 5.6645,
"mean_token_accuracy": 0.17633569836616517,
"num_tokens": 6109161.0,
"step": 2655
},
{
"entropy": 5.826577854156494,
"epoch": 0.25552353506243997,
"grad_norm": 1.0,
"learning_rate": 0.0004997125121769647,
"loss": 5.7585,
"mean_token_accuracy": 0.1654739573597908,
"num_tokens": 6120031.0,
"step": 2660
},
{
"entropy": 5.65047516822815,
"epoch": 0.25600384245917385,
"grad_norm": 0.96875,
"learning_rate": 0.0004997107770392444,
"loss": 5.6178,
"mean_token_accuracy": 0.17260300368070602,
"num_tokens": 6132398.0,
"step": 2665
},
{
"entropy": 5.605051326751709,
"epoch": 0.2564841498559078,
"grad_norm": 1.09375,
"learning_rate": 0.000499709036684407,
"loss": 5.5961,
"mean_token_accuracy": 0.1707433581352234,
"num_tokens": 6143099.0,
"step": 2670
},
{
"entropy": 5.630539417266846,
"epoch": 0.25696445725264166,
"grad_norm": 1.015625,
"learning_rate": 0.0004997072911124932,
"loss": 5.567,
"mean_token_accuracy": 0.18127228319644928,
"num_tokens": 6154815.0,
"step": 2675
},
{
"entropy": 5.673951101303101,
"epoch": 0.2574447646493756,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997055403235432,
"loss": 5.6335,
"mean_token_accuracy": 0.1791025683283806,
"num_tokens": 6165696.0,
"step": 2680
},
{
"entropy": 5.711018228530884,
"epoch": 0.2579250720461095,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997037843175978,
"loss": 5.6631,
"mean_token_accuracy": 0.1713305965065956,
"num_tokens": 6177042.0,
"step": 2685
},
{
"entropy": 5.691460752487183,
"epoch": 0.2584053794428434,
"grad_norm": 1.0078125,
"learning_rate": 0.0004997020230946978,
"loss": 5.6437,
"mean_token_accuracy": 0.17093220055103303,
"num_tokens": 6188691.0,
"step": 2690
},
{
"entropy": 5.5285240650177006,
"epoch": 0.25888568683957736,
"grad_norm": 1.078125,
"learning_rate": 0.0004997002566548841,
"loss": 5.4647,
"mean_token_accuracy": 0.18098385334014894,
"num_tokens": 6199931.0,
"step": 2695
},
{
"entropy": 5.6233972072601315,
"epoch": 0.25936599423631124,
"grad_norm": 0.984375,
"learning_rate": 0.0004996984849981976,
"loss": 5.6121,
"mean_token_accuracy": 0.17830558866262436,
"num_tokens": 6211493.0,
"step": 2700
},
{
"entropy": 5.7481287002563475,
"epoch": 0.2598463016330452,
"grad_norm": 1.0,
"learning_rate": 0.0004996967081246794,
"loss": 5.5374,
"mean_token_accuracy": 0.17606928646564485,
"num_tokens": 6221505.0,
"step": 2705
},
{
"entropy": 5.519698286056519,
"epoch": 0.26032660902977905,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996949260343711,
"loss": 5.5081,
"mean_token_accuracy": 0.18392858058214187,
"num_tokens": 6233058.0,
"step": 2710
},
{
"entropy": 5.681969499588012,
"epoch": 0.260806916426513,
"grad_norm": 0.984375,
"learning_rate": 0.0004996931387273137,
"loss": 5.6666,
"mean_token_accuracy": 0.16781300455331802,
"num_tokens": 6244713.0,
"step": 2715
},
{
"entropy": 5.6033467769622805,
"epoch": 0.2612872238232469,
"grad_norm": 0.90625,
"learning_rate": 0.0004996913462035487,
"loss": 5.6218,
"mean_token_accuracy": 0.17073221057653426,
"num_tokens": 6256624.0,
"step": 2720
},
{
"entropy": 5.610658645629883,
"epoch": 0.2617675312199808,
"grad_norm": 0.84375,
"learning_rate": 0.000499689548463118,
"loss": 5.6027,
"mean_token_accuracy": 0.17021558731794356,
"num_tokens": 6271134.0,
"step": 2725
},
{
"entropy": 5.752146673202515,
"epoch": 0.2622478386167147,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996877455060631,
"loss": 5.6993,
"mean_token_accuracy": 0.1727217897772789,
"num_tokens": 6282082.0,
"step": 2730
},
{
"entropy": 5.672946786880493,
"epoch": 0.2627281460134486,
"grad_norm": 0.953125,
"learning_rate": 0.0004996859373324259,
"loss": 5.6348,
"mean_token_accuracy": 0.17068509608507157,
"num_tokens": 6293869.0,
"step": 2735
},
{
"entropy": 5.69234881401062,
"epoch": 0.2632084534101825,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996841239422485,
"loss": 5.6785,
"mean_token_accuracy": 0.16386467665433885,
"num_tokens": 6304858.0,
"step": 2740
},
{
"entropy": 5.6755890369415285,
"epoch": 0.26368876080691644,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996823053355729,
"loss": 5.6002,
"mean_token_accuracy": 0.176376673579216,
"num_tokens": 6316738.0,
"step": 2745
},
{
"entropy": 5.579497671127319,
"epoch": 0.2641690682036503,
"grad_norm": 1.0,
"learning_rate": 0.0004996804815124413,
"loss": 5.5573,
"mean_token_accuracy": 0.17610080689191818,
"num_tokens": 6327691.0,
"step": 2750
},
{
"entropy": 5.666668939590454,
"epoch": 0.26464937560038426,
"grad_norm": 1.125,
"learning_rate": 0.0004996786524728962,
"loss": 5.5749,
"mean_token_accuracy": 0.173881895840168,
"num_tokens": 6337964.0,
"step": 2755
},
{
"entropy": 5.595135927200317,
"epoch": 0.26512968299711814,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996768182169797,
"loss": 5.6286,
"mean_token_accuracy": 0.17362290471792222,
"num_tokens": 6349764.0,
"step": 2760
},
{
"entropy": 5.6827469825744625,
"epoch": 0.2656099903938521,
"grad_norm": 1.046875,
"learning_rate": 0.0004996749787447349,
"loss": 5.5855,
"mean_token_accuracy": 0.17588018625974655,
"num_tokens": 6360538.0,
"step": 2765
},
{
"entropy": 5.644002532958984,
"epoch": 0.26609029779058596,
"grad_norm": 1.0625,
"learning_rate": 0.000499673134056204,
"loss": 5.6181,
"mean_token_accuracy": 0.17757804244756697,
"num_tokens": 6372489.0,
"step": 2770
},
{
"entropy": 5.499239540100097,
"epoch": 0.2665706051873199,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996712841514303,
"loss": 5.4985,
"mean_token_accuracy": 0.18251207172870637,
"num_tokens": 6383080.0,
"step": 2775
},
{
"entropy": 5.724687099456787,
"epoch": 0.2670509125840538,
"grad_norm": 0.984375,
"learning_rate": 0.0004996694290304563,
"loss": 5.6232,
"mean_token_accuracy": 0.17356717586517334,
"num_tokens": 6394033.0,
"step": 2780
},
{
"entropy": 5.651960182189941,
"epoch": 0.2675312199807877,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996675686933255,
"loss": 5.632,
"mean_token_accuracy": 0.17403780966997145,
"num_tokens": 6404990.0,
"step": 2785
},
{
"entropy": 5.583127403259278,
"epoch": 0.2680115273775216,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996657031400807,
"loss": 5.4925,
"mean_token_accuracy": 0.18252827525138854,
"num_tokens": 6416337.0,
"step": 2790
},
{
"entropy": 5.583510828018189,
"epoch": 0.26849183477425553,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996638323707655,
"loss": 5.6999,
"mean_token_accuracy": 0.17072573751211167,
"num_tokens": 6427892.0,
"step": 2795
},
{
"entropy": 5.663077068328858,
"epoch": 0.2689721421709894,
"grad_norm": 1.015625,
"learning_rate": 0.0004996619563854232,
"loss": 5.5987,
"mean_token_accuracy": 0.17658236026763915,
"num_tokens": 6440300.0,
"step": 2800
},
{
"entropy": 5.589122152328491,
"epoch": 0.26945244956772335,
"grad_norm": 0.91796875,
"learning_rate": 0.0004996600751840974,
"loss": 5.5071,
"mean_token_accuracy": 0.17740865498781205,
"num_tokens": 6451499.0,
"step": 2805
},
{
"entropy": 5.65373477935791,
"epoch": 0.26993275696445723,
"grad_norm": 1.0,
"learning_rate": 0.0004996581887668317,
"loss": 5.5993,
"mean_token_accuracy": 0.17863193303346633,
"num_tokens": 6462882.0,
"step": 2810
},
{
"entropy": 5.653873777389526,
"epoch": 0.27041306436119117,
"grad_norm": 1.0625,
"learning_rate": 0.00049965629713367,
"loss": 5.5799,
"mean_token_accuracy": 0.1730649381875992,
"num_tokens": 6475042.0,
"step": 2815
},
{
"entropy": 5.557312488555908,
"epoch": 0.27089337175792505,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996544002846561,
"loss": 5.5842,
"mean_token_accuracy": 0.1783270537853241,
"num_tokens": 6486245.0,
"step": 2820
},
{
"entropy": 5.599606609344482,
"epoch": 0.271373679154659,
"grad_norm": 1.0234375,
"learning_rate": 0.0004996524982198343,
"loss": 5.4857,
"mean_token_accuracy": 0.1768743082880974,
"num_tokens": 6497705.0,
"step": 2825
},
{
"entropy": 5.572207498550415,
"epoch": 0.27185398655139287,
"grad_norm": 1.09375,
"learning_rate": 0.0004996505909392485,
"loss": 5.5271,
"mean_token_accuracy": 0.17951610833406448,
"num_tokens": 6508193.0,
"step": 2830
},
{
"entropy": 5.67252025604248,
"epoch": 0.2723342939481268,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996486784429429,
"loss": 5.6391,
"mean_token_accuracy": 0.17500174790620804,
"num_tokens": 6519228.0,
"step": 2835
},
{
"entropy": 5.625978231430054,
"epoch": 0.2728146013448607,
"grad_norm": 1.1171875,
"learning_rate": 0.0004996467607309622,
"loss": 5.6012,
"mean_token_accuracy": 0.1761829063296318,
"num_tokens": 6530743.0,
"step": 2840
},
{
"entropy": 5.612495994567871,
"epoch": 0.2732949087415946,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996448378033507,
"loss": 5.4899,
"mean_token_accuracy": 0.1722333922982216,
"num_tokens": 6542588.0,
"step": 2845
},
{
"entropy": 5.707457685470581,
"epoch": 0.2737752161383285,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996429096601532,
"loss": 5.6893,
"mean_token_accuracy": 0.16250439882278442,
"num_tokens": 6554002.0,
"step": 2850
},
{
"entropy": 5.732671689987183,
"epoch": 0.27425552353506244,
"grad_norm": 1.078125,
"learning_rate": 0.0004996409763014144,
"loss": 5.71,
"mean_token_accuracy": 0.17036043256521224,
"num_tokens": 6563358.0,
"step": 2855
},
{
"entropy": 5.6183513641357425,
"epoch": 0.2747358309317964,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996390377271791,
"loss": 5.6116,
"mean_token_accuracy": 0.17085390090942382,
"num_tokens": 6574986.0,
"step": 2860
},
{
"entropy": 5.605986452102661,
"epoch": 0.27521613832853026,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996370939374924,
"loss": 5.5603,
"mean_token_accuracy": 0.1796693056821823,
"num_tokens": 6587407.0,
"step": 2865
},
{
"entropy": 5.561212491989136,
"epoch": 0.2756964457252642,
"grad_norm": 1.046875,
"learning_rate": 0.0004996351449323994,
"loss": 5.5098,
"mean_token_accuracy": 0.18226875960826874,
"num_tokens": 6598483.0,
"step": 2870
},
{
"entropy": 5.56893515586853,
"epoch": 0.2761767531219981,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996331907119455,
"loss": 5.6098,
"mean_token_accuracy": 0.1776777282357216,
"num_tokens": 6610749.0,
"step": 2875
},
{
"entropy": 5.687351655960083,
"epoch": 0.276657060518732,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996312312761758,
"loss": 5.5796,
"mean_token_accuracy": 0.16841792613267897,
"num_tokens": 6621762.0,
"step": 2880
},
{
"entropy": 5.589012336730957,
"epoch": 0.2771373679154659,
"grad_norm": 1.0234375,
"learning_rate": 0.000499629266625136,
"loss": 5.6511,
"mean_token_accuracy": 0.16858434230089187,
"num_tokens": 6632296.0,
"step": 2885
},
{
"entropy": 5.639417600631714,
"epoch": 0.27761767531219983,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996272967588715,
"loss": 5.5973,
"mean_token_accuracy": 0.17394536584615708,
"num_tokens": 6644413.0,
"step": 2890
},
{
"entropy": 5.641393375396729,
"epoch": 0.2780979827089337,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996253216774283,
"loss": 5.5458,
"mean_token_accuracy": 0.17925278544425965,
"num_tokens": 6655634.0,
"step": 2895
},
{
"entropy": 5.6084517478942875,
"epoch": 0.27857829010566765,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996233413808521,
"loss": 5.4927,
"mean_token_accuracy": 0.18065261244773864,
"num_tokens": 6665977.0,
"step": 2900
},
{
"entropy": 5.606080198287964,
"epoch": 0.27905859750240153,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996213558691889,
"loss": 5.548,
"mean_token_accuracy": 0.17679802030324937,
"num_tokens": 6677165.0,
"step": 2905
},
{
"entropy": 5.642763233184814,
"epoch": 0.27953890489913547,
"grad_norm": 1.015625,
"learning_rate": 0.0004996193651424848,
"loss": 5.6721,
"mean_token_accuracy": 0.1694797120988369,
"num_tokens": 6688182.0,
"step": 2910
},
{
"entropy": 5.626778411865234,
"epoch": 0.28001921229586935,
"grad_norm": 1.0703125,
"learning_rate": 0.000499617369200786,
"loss": 5.5704,
"mean_token_accuracy": 0.18135443776845933,
"num_tokens": 6699259.0,
"step": 2915
},
{
"entropy": 5.623982906341553,
"epoch": 0.2804995196926033,
"grad_norm": 1.0,
"learning_rate": 0.0004996153680441389,
"loss": 5.5901,
"mean_token_accuracy": 0.1713014855980873,
"num_tokens": 6710071.0,
"step": 2920
},
{
"entropy": 5.646922492980957,
"epoch": 0.28097982708933716,
"grad_norm": 1.0390625,
"learning_rate": 0.00049961336167259,
"loss": 5.6166,
"mean_token_accuracy": 0.17556966245174407,
"num_tokens": 6720267.0,
"step": 2925
},
{
"entropy": 5.64747257232666,
"epoch": 0.2814601344860711,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996113500861857,
"loss": 5.5968,
"mean_token_accuracy": 0.17084414660930633,
"num_tokens": 6730919.0,
"step": 2930
},
{
"entropy": 5.558165216445923,
"epoch": 0.281940441882805,
"grad_norm": 1.0703125,
"learning_rate": 0.0004996093332849729,
"loss": 5.5952,
"mean_token_accuracy": 0.17768187820911407,
"num_tokens": 6741480.0,
"step": 2935
},
{
"entropy": 5.70050163269043,
"epoch": 0.2824207492795389,
"grad_norm": 1.1015625,
"learning_rate": 0.0004996073112689983,
"loss": 5.5396,
"mean_token_accuracy": 0.1811237409710884,
"num_tokens": 6751816.0,
"step": 2940
},
{
"entropy": 5.51968994140625,
"epoch": 0.2829010566762728,
"grad_norm": 1.203125,
"learning_rate": 0.0004996052840383088,
"loss": 5.5537,
"mean_token_accuracy": 0.17758982628583908,
"num_tokens": 6763645.0,
"step": 2945
},
{
"entropy": 5.535739755630493,
"epoch": 0.28338136407300674,
"grad_norm": 1.046875,
"learning_rate": 0.0004996032515929516,
"loss": 5.5363,
"mean_token_accuracy": 0.17651461958885192,
"num_tokens": 6774795.0,
"step": 2950
},
{
"entropy": 5.6613250255584715,
"epoch": 0.2838616714697406,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996012139329738,
"loss": 5.4491,
"mean_token_accuracy": 0.18220943808555604,
"num_tokens": 6786833.0,
"step": 2955
},
{
"entropy": 5.492148590087891,
"epoch": 0.28434197886647455,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995991710584228,
"loss": 5.471,
"mean_token_accuracy": 0.18226066827774048,
"num_tokens": 6798134.0,
"step": 2960
},
{
"entropy": 5.634646844863892,
"epoch": 0.28482228626320844,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995971229693459,
"loss": 5.5818,
"mean_token_accuracy": 0.18246834278106688,
"num_tokens": 6807918.0,
"step": 2965
},
{
"entropy": 5.643370199203491,
"epoch": 0.28530259365994237,
"grad_norm": 0.984375,
"learning_rate": 0.0004995950696657909,
"loss": 5.6627,
"mean_token_accuracy": 0.17691846936941147,
"num_tokens": 6818689.0,
"step": 2970
},
{
"entropy": 5.596081829071045,
"epoch": 0.28578290105667625,
"grad_norm": 0.9609375,
"learning_rate": 0.0004995930111478051,
"loss": 5.5265,
"mean_token_accuracy": 0.17872194498777388,
"num_tokens": 6830871.0,
"step": 2975
},
{
"entropy": 5.5695876121521,
"epoch": 0.2862632084534102,
"grad_norm": 1.046875,
"learning_rate": 0.0004995909474154365,
"loss": 5.5218,
"mean_token_accuracy": 0.18281462639570237,
"num_tokens": 6842857.0,
"step": 2980
},
{
"entropy": 5.561906528472901,
"epoch": 0.28674351585014407,
"grad_norm": 1.15625,
"learning_rate": 0.0004995888784687331,
"loss": 5.5439,
"mean_token_accuracy": 0.17641340792179108,
"num_tokens": 6853727.0,
"step": 2985
},
{
"entropy": 5.6508629322052,
"epoch": 0.287223823246878,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995868043077428,
"loss": 5.5712,
"mean_token_accuracy": 0.1839816763997078,
"num_tokens": 6864728.0,
"step": 2990
},
{
"entropy": 5.6089515209198,
"epoch": 0.2877041306436119,
"grad_norm": 1.046875,
"learning_rate": 0.0004995847249325137,
"loss": 5.6382,
"mean_token_accuracy": 0.17240980714559556,
"num_tokens": 6875001.0,
"step": 2995
},
{
"entropy": 5.575883674621582,
"epoch": 0.2881844380403458,
"grad_norm": 0.99609375,
"learning_rate": 0.0004995826403430942,
"loss": 5.5194,
"mean_token_accuracy": 0.18421110808849334,
"num_tokens": 6887026.0,
"step": 3000
},
{
"epoch": 0.2881844380403458,
"eval_entropy": 5.505591410883806,
"eval_loss": 5.596490859985352,
"eval_mean_token_accuracy": 0.18258213421383437,
"eval_num_tokens": 6887026.0,
"eval_runtime": 26.8269,
"eval_samples_per_second": 1223.214,
"eval_steps_per_second": 152.906,
"step": 3000
}
],
"logging_steps": 5,
"max_steps": 104090,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 3000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 1.0587455152128e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink