fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity
Files
main
ita-latn-100mb-after-ppt-sh…/checkpoint-6000/trainer_state.json
ModelHub XC 5d2860cc28 初始化项目，由ModelHub XC社区提供模型 
Model: fpadovani/ita-latn-100mb-after-ppt-shuff-dyck-100mb-ckpt500_seed3407
Source: Original Platform
2026-08-12 16:01:20 +08:00

12057 lines
330 KiB
JSON
Raw Permalink Blame History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5954743946010321,
"eval_steps": 3000,
"global_step": 6000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"entropy": 4.8886785984039305,
"epoch": 0.0004962286621675268,
"grad_norm": 24.0,
"learning_rate": 2e-06,
"loss": 14.1768,
"mean_token_accuracy": 0.0,
"num_tokens": 11448.0,
"step": 5
},
{
"entropy": 4.918353462219239,
"epoch": 0.0009924573243350536,
"grad_norm": 27.125,
"learning_rate": 4.5e-06,
"loss": 14.1453,
"mean_token_accuracy": 9.523809421807528e-05,
"num_tokens": 22646.0,
"step": 10
},
{
"entropy": 5.069689464569092,
"epoch": 0.0014886859865025804,
"grad_norm": 36.75,
"learning_rate": 7e-06,
"loss": 13.8023,
"mean_token_accuracy": 0.0,
"num_tokens": 34011.0,
"step": 15
},
{
"entropy": 5.5015480518341064,
"epoch": 0.001984914648670107,
"grad_norm": 42.75,
"learning_rate": 9.5e-06,
"loss": 13.2381,
"mean_token_accuracy": 0.0,
"num_tokens": 44566.0,
"step": 20
},
{
"entropy": 7.525242757797241,
"epoch": 0.002481143310837634,
"grad_norm": 18.875,
"learning_rate": 1.2e-05,
"loss": 11.959,
"mean_token_accuracy": 0.0,
"num_tokens": 56274.0,
"step": 25
},
{
"entropy": 9.429654407501221,
"epoch": 0.0029773719730051607,
"grad_norm": 5.25,
"learning_rate": 1.4500000000000002e-05,
"loss": 11.1919,
"mean_token_accuracy": 0.0,
"num_tokens": 66584.0,
"step": 30
},
{
"entropy": 10.561602783203124,
"epoch": 0.0034736006351726877,
"grad_norm": 2.40625,
"learning_rate": 1.7000000000000003e-05,
"loss": 10.7394,
"mean_token_accuracy": 0.0,
"num_tokens": 76948.0,
"step": 35
},
{
"entropy": 10.702620887756348,
"epoch": 0.003969829297340214,
"grad_norm": 2.3125,
"learning_rate": 1.95e-05,
"loss": 10.496,
"mean_token_accuracy": 0.012860012706369162,
"num_tokens": 87253.0,
"step": 40
},
{
"entropy": 10.697653579711915,
"epoch": 0.004466057959507741,
"grad_norm": 2.46875,
"learning_rate": 2.2e-05,
"loss": 10.1993,
"mean_token_accuracy": 0.04175656381994486,
"num_tokens": 98273.0,
"step": 45
},
{
"entropy": 10.658453273773194,
"epoch": 0.004962286621675268,
"grad_norm": 2.15625,
"learning_rate": 2.4500000000000003e-05,
"loss": 9.9063,
"mean_token_accuracy": 0.04210629928857088,
"num_tokens": 107788.0,
"step": 50
},
{
"entropy": 10.590508460998535,
"epoch": 0.0054585152838427945,
"grad_norm": 2.078125,
"learning_rate": 2.7e-05,
"loss": 9.7831,
"mean_token_accuracy": 0.03733267150819301,
"num_tokens": 118632.0,
"step": 55
},
{
"entropy": 10.57080307006836,
"epoch": 0.0059547439460103215,
"grad_norm": 2.046875,
"learning_rate": 2.95e-05,
"loss": 9.6152,
"mean_token_accuracy": 0.04101524651050568,
"num_tokens": 128649.0,
"step": 60
},
{
"entropy": 10.565433502197266,
"epoch": 0.0064509726081778485,
"grad_norm": 1.6953125,
"learning_rate": 3.2e-05,
"loss": 9.5678,
"mean_token_accuracy": 0.04273284543305635,
"num_tokens": 138958.0,
"step": 65
},
{
"entropy": 10.547395515441895,
"epoch": 0.0069472012703453755,
"grad_norm": 1.828125,
"learning_rate": 3.4500000000000005e-05,
"loss": 9.5162,
"mean_token_accuracy": 0.04243846274912357,
"num_tokens": 151445.0,
"step": 70
},
{
"entropy": 10.541783046722411,
"epoch": 0.007443429932512902,
"grad_norm": 1.703125,
"learning_rate": 3.7e-05,
"loss": 9.4272,
"mean_token_accuracy": 0.043674266710877416,
"num_tokens": 162239.0,
"step": 75
},
{
"entropy": 10.524649047851563,
"epoch": 0.007939658594680429,
"grad_norm": 1.84375,
"learning_rate": 3.95e-05,
"loss": 9.3667,
"mean_token_accuracy": 0.03882305920124054,
"num_tokens": 173261.0,
"step": 80
},
{
"entropy": 10.50841827392578,
"epoch": 0.008435887256847956,
"grad_norm": 1.8125,
"learning_rate": 4.2000000000000004e-05,
"loss": 9.2444,
"mean_token_accuracy": 0.037595166079699996,
"num_tokens": 183857.0,
"step": 85
},
{
"entropy": 10.483632850646973,
"epoch": 0.008932115919015483,
"grad_norm": 1.6484375,
"learning_rate": 4.45e-05,
"loss": 9.2664,
"mean_token_accuracy": 0.03864660933613777,
"num_tokens": 196635.0,
"step": 90
},
{
"entropy": 10.438802337646484,
"epoch": 0.00942834458118301,
"grad_norm": 1.6875,
"learning_rate": 4.7000000000000004e-05,
"loss": 9.079,
"mean_token_accuracy": 0.04349393919110298,
"num_tokens": 208700.0,
"step": 95
},
{
"entropy": 10.382994365692138,
"epoch": 0.009924573243350537,
"grad_norm": 1.7578125,
"learning_rate": 4.9500000000000004e-05,
"loss": 8.9171,
"mean_token_accuracy": 0.04082345124334097,
"num_tokens": 218915.0,
"step": 100
},
{
"entropy": 10.337403869628906,
"epoch": 0.010420801905518062,
"grad_norm": 1.65625,
"learning_rate": 5.2e-05,
"loss": 8.7812,
"mean_token_accuracy": 0.051039114221930505,
"num_tokens": 230211.0,
"step": 105
},
{
"entropy": 10.211695384979247,
"epoch": 0.010917030567685589,
"grad_norm": 1.6015625,
"learning_rate": 5.45e-05,
"loss": 8.6614,
"mean_token_accuracy": 0.05333258844912052,
"num_tokens": 241174.0,
"step": 110
},
{
"entropy": 10.04779806137085,
"epoch": 0.011413259229853116,
"grad_norm": 1.4140625,
"learning_rate": 5.7e-05,
"loss": 8.5665,
"mean_token_accuracy": 0.047243202850222585,
"num_tokens": 251708.0,
"step": 115
},
{
"entropy": 9.996843624114991,
"epoch": 0.011909487892020643,
"grad_norm": 1.2421875,
"learning_rate": 5.9499999999999996e-05,
"loss": 8.585,
"mean_token_accuracy": 0.04537748880684376,
"num_tokens": 263287.0,
"step": 120
},
{
"entropy": 9.867857456207275,
"epoch": 0.01240571655418817,
"grad_norm": 1.1953125,
"learning_rate": 6.2e-05,
"loss": 8.3763,
"mean_token_accuracy": 0.0451455045491457,
"num_tokens": 275179.0,
"step": 125
},
{
"entropy": 9.665265655517578,
"epoch": 0.012901945216355697,
"grad_norm": 1.2421875,
"learning_rate": 6.450000000000001e-05,
"loss": 8.1667,
"mean_token_accuracy": 0.04741326570510864,
"num_tokens": 287223.0,
"step": 130
},
{
"entropy": 9.438035774230958,
"epoch": 0.013398173878523224,
"grad_norm": 1.1015625,
"learning_rate": 6.7e-05,
"loss": 8.0475,
"mean_token_accuracy": 0.04950449950993061,
"num_tokens": 297868.0,
"step": 135
},
{
"entropy": 9.164513778686523,
"epoch": 0.013894402540690751,
"grad_norm": 0.83984375,
"learning_rate": 6.950000000000001e-05,
"loss": 8.0005,
"mean_token_accuracy": 0.05104476250708103,
"num_tokens": 309702.0,
"step": 140
},
{
"entropy": 8.994746875762939,
"epoch": 0.014390631202858276,
"grad_norm": 0.87890625,
"learning_rate": 7.2e-05,
"loss": 8.0216,
"mean_token_accuracy": 0.048004313558340075,
"num_tokens": 321613.0,
"step": 145
},
{
"entropy": 8.776610565185546,
"epoch": 0.014886859865025803,
"grad_norm": 0.9375,
"learning_rate": 7.45e-05,
"loss": 7.8727,
"mean_token_accuracy": 0.051788559183478355,
"num_tokens": 332668.0,
"step": 150
},
{
"entropy": 8.660905170440675,
"epoch": 0.01538308852719333,
"grad_norm": 0.8828125,
"learning_rate": 7.7e-05,
"loss": 7.8823,
"mean_token_accuracy": 0.049248311296105386,
"num_tokens": 342579.0,
"step": 155
},
{
"entropy": 8.499152851104736,
"epoch": 0.015879317189360857,
"grad_norm": 0.98046875,
"learning_rate": 7.950000000000001e-05,
"loss": 7.759,
"mean_token_accuracy": 0.0496104896068573,
"num_tokens": 352271.0,
"step": 160
},
{
"entropy": 8.3534499168396,
"epoch": 0.016375545851528384,
"grad_norm": 1.0078125,
"learning_rate": 8.2e-05,
"loss": 7.7938,
"mean_token_accuracy": 0.061972716078162196,
"num_tokens": 364033.0,
"step": 165
},
{
"entropy": 8.28783130645752,
"epoch": 0.01687177451369591,
"grad_norm": 1.0078125,
"learning_rate": 8.450000000000001e-05,
"loss": 7.7301,
"mean_token_accuracy": 0.05461362563073635,
"num_tokens": 373788.0,
"step": 170
},
{
"entropy": 8.24440793991089,
"epoch": 0.01736800317586344,
"grad_norm": 1.1875,
"learning_rate": 8.7e-05,
"loss": 7.7277,
"mean_token_accuracy": 0.057403456419706345,
"num_tokens": 384883.0,
"step": 175
},
{
"entropy": 8.245327854156494,
"epoch": 0.017864231838030965,
"grad_norm": 1.0625,
"learning_rate": 8.95e-05,
"loss": 7.6288,
"mean_token_accuracy": 0.06058619767427444,
"num_tokens": 396044.0,
"step": 180
},
{
"entropy": 8.175503873825074,
"epoch": 0.018360460500198492,
"grad_norm": 1.359375,
"learning_rate": 9.2e-05,
"loss": 7.733,
"mean_token_accuracy": 0.06297681033611298,
"num_tokens": 406995.0,
"step": 185
},
{
"entropy": 8.033704376220703,
"epoch": 0.01885668916236602,
"grad_norm": 0.953125,
"learning_rate": 9.45e-05,
"loss": 7.5683,
"mean_token_accuracy": 0.0688964381814003,
"num_tokens": 417238.0,
"step": 190
},
{
"entropy": 8.177022457122803,
"epoch": 0.019352917824533546,
"grad_norm": 1.109375,
"learning_rate": 9.7e-05,
"loss": 7.645,
"mean_token_accuracy": 0.06311573572456837,
"num_tokens": 427263.0,
"step": 195
},
{
"entropy": 8.155741834640503,
"epoch": 0.019849146486701073,
"grad_norm": 1.015625,
"learning_rate": 9.95e-05,
"loss": 7.5372,
"mean_token_accuracy": 0.0640363771468401,
"num_tokens": 438357.0,
"step": 200
},
{
"entropy": 8.039666271209716,
"epoch": 0.0203453751488686,
"grad_norm": 1.25,
"learning_rate": 0.000102,
"loss": 7.5516,
"mean_token_accuracy": 0.06006042025983334,
"num_tokens": 448738.0,
"step": 205
},
{
"entropy": 7.96295280456543,
"epoch": 0.020841603811036124,
"grad_norm": 0.984375,
"learning_rate": 0.00010449999999999999,
"loss": 7.5624,
"mean_token_accuracy": 0.07530630342662334,
"num_tokens": 457873.0,
"step": 210
},
{
"entropy": 8.063592433929443,
"epoch": 0.02133783247320365,
"grad_norm": 1.046875,
"learning_rate": 0.000107,
"loss": 7.4987,
"mean_token_accuracy": 0.06970696784555912,
"num_tokens": 468640.0,
"step": 215
},
{
"entropy": 8.009668874740601,
"epoch": 0.021834061135371178,
"grad_norm": 1.4375,
"learning_rate": 0.0001095,
"loss": 7.5142,
"mean_token_accuracy": 0.07091238908469677,
"num_tokens": 479922.0,
"step": 220
},
{
"entropy": 7.893428993225098,
"epoch": 0.022330289797538705,
"grad_norm": 0.921875,
"learning_rate": 0.000112,
"loss": 7.4717,
"mean_token_accuracy": 0.06831742338836193,
"num_tokens": 491056.0,
"step": 225
},
{
"entropy": 7.929876375198364,
"epoch": 0.022826518459706232,
"grad_norm": 1.0625,
"learning_rate": 0.0001145,
"loss": 7.5665,
"mean_token_accuracy": 0.06847823970019817,
"num_tokens": 504365.0,
"step": 230
},
{
"entropy": 7.870587635040283,
"epoch": 0.02332274712187376,
"grad_norm": 1.28125,
"learning_rate": 0.00011700000000000001,
"loss": 7.437,
"mean_token_accuracy": 0.06928115114569663,
"num_tokens": 514347.0,
"step": 235
},
{
"entropy": 7.846385335922241,
"epoch": 0.023818975784041286,
"grad_norm": 1.46875,
"learning_rate": 0.00011949999999999999,
"loss": 7.4702,
"mean_token_accuracy": 0.06925889551639557,
"num_tokens": 525912.0,
"step": 240
},
{
"entropy": 7.909278726577758,
"epoch": 0.024315204446208813,
"grad_norm": 1.03125,
"learning_rate": 0.000122,
"loss": 7.3455,
"mean_token_accuracy": 0.07115834765136242,
"num_tokens": 535566.0,
"step": 245
},
{
"entropy": 7.8972766399383545,
"epoch": 0.02481143310837634,
"grad_norm": 1.015625,
"learning_rate": 0.0001245,
"loss": 7.3889,
"mean_token_accuracy": 0.07348600402474403,
"num_tokens": 546474.0,
"step": 250
},
{
"entropy": 7.799334192276001,
"epoch": 0.025307661770543867,
"grad_norm": 0.95703125,
"learning_rate": 0.000127,
"loss": 7.4794,
"mean_token_accuracy": 0.06563120521605015,
"num_tokens": 558132.0,
"step": 255
},
{
"entropy": 7.913891267776489,
"epoch": 0.025803890432711394,
"grad_norm": 1.265625,
"learning_rate": 0.0001295,
"loss": 7.4623,
"mean_token_accuracy": 0.06633393727242946,
"num_tokens": 568309.0,
"step": 260
},
{
"entropy": 7.81136999130249,
"epoch": 0.02630011909487892,
"grad_norm": 1.125,
"learning_rate": 0.000132,
"loss": 7.372,
"mean_token_accuracy": 0.06931877210736274,
"num_tokens": 579701.0,
"step": 265
},
{
"entropy": 7.7742400646209715,
"epoch": 0.026796347757046448,
"grad_norm": 0.953125,
"learning_rate": 0.00013450000000000002,
"loss": 7.33,
"mean_token_accuracy": 0.07372162342071534,
"num_tokens": 590832.0,
"step": 270
},
{
"entropy": 7.759802675247192,
"epoch": 0.027292576419213975,
"grad_norm": 0.86328125,
"learning_rate": 0.00013700000000000002,
"loss": 7.274,
"mean_token_accuracy": 0.07079541869461536,
"num_tokens": 601094.0,
"step": 275
},
{
"entropy": 7.690725088119507,
"epoch": 0.027788805081381502,
"grad_norm": 1.078125,
"learning_rate": 0.0001395,
"loss": 7.2879,
"mean_token_accuracy": 0.07795346416532993,
"num_tokens": 612027.0,
"step": 280
},
{
"entropy": 7.692393255233765,
"epoch": 0.02828503374354903,
"grad_norm": 1.1171875,
"learning_rate": 0.00014199999999999998,
"loss": 7.33,
"mean_token_accuracy": 0.07078218050301074,
"num_tokens": 623698.0,
"step": 285
},
{
"entropy": 7.724156379699707,
"epoch": 0.028781262405716553,
"grad_norm": 0.828125,
"learning_rate": 0.0001445,
"loss": 7.3473,
"mean_token_accuracy": 0.07386260889470578,
"num_tokens": 635631.0,
"step": 290
},
{
"entropy": 7.633536434173584,
"epoch": 0.02927749106788408,
"grad_norm": 0.85546875,
"learning_rate": 0.000147,
"loss": 7.3238,
"mean_token_accuracy": 0.07232154980301857,
"num_tokens": 647663.0,
"step": 295
},
{
"entropy": 7.68490948677063,
"epoch": 0.029773719730051607,
"grad_norm": 1.03125,
"learning_rate": 0.0001495,
"loss": 7.2869,
"mean_token_accuracy": 0.0713068351149559,
"num_tokens": 659352.0,
"step": 300
},
{
"entropy": 7.628432846069336,
"epoch": 0.030269948392219134,
"grad_norm": 0.8984375,
"learning_rate": 0.000152,
"loss": 7.2491,
"mean_token_accuracy": 0.07672590613365174,
"num_tokens": 669379.0,
"step": 305
},
{
"entropy": 7.652841281890869,
"epoch": 0.03076617705438666,
"grad_norm": 1.203125,
"learning_rate": 0.00015450000000000001,
"loss": 7.241,
"mean_token_accuracy": 0.07731729820370674,
"num_tokens": 680144.0,
"step": 310
},
{
"entropy": 7.682145023345948,
"epoch": 0.03126240571655419,
"grad_norm": 1.53125,
"learning_rate": 0.000157,
"loss": 7.1947,
"mean_token_accuracy": 0.06403898783028125,
"num_tokens": 691165.0,
"step": 315
},
{
"entropy": 7.572655248641968,
"epoch": 0.031758634378721715,
"grad_norm": 1.109375,
"learning_rate": 0.0001595,
"loss": 7.2853,
"mean_token_accuracy": 0.07224688902497292,
"num_tokens": 703767.0,
"step": 320
},
{
"entropy": 7.581577396392822,
"epoch": 0.032254863040889245,
"grad_norm": 1.46875,
"learning_rate": 0.000162,
"loss": 7.2793,
"mean_token_accuracy": 0.07694573625922203,
"num_tokens": 714298.0,
"step": 325
},
{
"entropy": 7.558791637420654,
"epoch": 0.03275109170305677,
"grad_norm": 1.25,
"learning_rate": 0.00016450000000000001,
"loss": 7.2543,
"mean_token_accuracy": 0.08345745950937271,
"num_tokens": 724135.0,
"step": 330
},
{
"entropy": 7.579706907272339,
"epoch": 0.03324732036522429,
"grad_norm": 0.96875,
"learning_rate": 0.00016700000000000002,
"loss": 7.3114,
"mean_token_accuracy": 0.07194749340415,
"num_tokens": 735050.0,
"step": 335
},
{
"entropy": 7.66069598197937,
"epoch": 0.03374354902739182,
"grad_norm": 0.94921875,
"learning_rate": 0.00016950000000000003,
"loss": 7.2156,
"mean_token_accuracy": 0.07318415641784667,
"num_tokens": 746073.0,
"step": 340
},
{
"entropy": 7.529136419296265,
"epoch": 0.034239777689559346,
"grad_norm": 1.078125,
"learning_rate": 0.00017199999999999998,
"loss": 7.2125,
"mean_token_accuracy": 0.0734445221722126,
"num_tokens": 758998.0,
"step": 345
},
{
"entropy": 7.582425689697265,
"epoch": 0.03473600635172688,
"grad_norm": 1.0859375,
"learning_rate": 0.00017449999999999999,
"loss": 7.1518,
"mean_token_accuracy": 0.07465869188308716,
"num_tokens": 771608.0,
"step": 350
},
{
"entropy": 7.522878408432007,
"epoch": 0.0352322350138944,
"grad_norm": 0.984375,
"learning_rate": 0.000177,
"loss": 7.2138,
"mean_token_accuracy": 0.07425806559622287,
"num_tokens": 782394.0,
"step": 355
},
{
"entropy": 7.542528247833252,
"epoch": 0.03572846367606193,
"grad_norm": 1.15625,
"learning_rate": 0.0001795,
"loss": 7.1216,
"mean_token_accuracy": 0.07903320342302322,
"num_tokens": 792389.0,
"step": 360
},
{
"entropy": 7.50439190864563,
"epoch": 0.036224692338229454,
"grad_norm": 1.59375,
"learning_rate": 0.000182,
"loss": 7.1662,
"mean_token_accuracy": 0.07648928835988045,
"num_tokens": 803219.0,
"step": 365
},
{
"entropy": 7.523057889938355,
"epoch": 0.036720921000396985,
"grad_norm": 1.078125,
"learning_rate": 0.0001845,
"loss": 7.235,
"mean_token_accuracy": 0.07408236041665077,
"num_tokens": 814249.0,
"step": 370
},
{
"entropy": 7.520695543289184,
"epoch": 0.03721714966256451,
"grad_norm": 1.09375,
"learning_rate": 0.000187,
"loss": 7.1213,
"mean_token_accuracy": 0.08493848741054535,
"num_tokens": 825138.0,
"step": 375
},
{
"entropy": 7.437196350097656,
"epoch": 0.03771337832473204,
"grad_norm": 1.125,
"learning_rate": 0.0001895,
"loss": 7.1724,
"mean_token_accuracy": 0.07618945986032485,
"num_tokens": 836024.0,
"step": 380
},
{
"entropy": 7.505164337158203,
"epoch": 0.03820960698689956,
"grad_norm": 1.1328125,
"learning_rate": 0.000192,
"loss": 7.1619,
"mean_token_accuracy": 0.07474682033061981,
"num_tokens": 847765.0,
"step": 385
},
{
"entropy": 7.485037040710449,
"epoch": 0.03870583564906709,
"grad_norm": 1.296875,
"learning_rate": 0.0001945,
"loss": 7.1428,
"mean_token_accuracy": 0.07660850211977958,
"num_tokens": 859980.0,
"step": 390
},
{
"entropy": 7.523992443084717,
"epoch": 0.039202064311234616,
"grad_norm": 1.1015625,
"learning_rate": 0.00019700000000000002,
"loss": 7.1394,
"mean_token_accuracy": 0.07903257459402084,
"num_tokens": 871514.0,
"step": 395
},
{
"entropy": 7.4822288990020756,
"epoch": 0.03969829297340215,
"grad_norm": 0.95703125,
"learning_rate": 0.00019950000000000002,
"loss": 7.1353,
"mean_token_accuracy": 0.07928794771432876,
"num_tokens": 881984.0,
"step": 400
},
{
"entropy": 7.439731025695801,
"epoch": 0.04019452163556967,
"grad_norm": 1.09375,
"learning_rate": 0.000202,
"loss": 7.098,
"mean_token_accuracy": 0.08132969811558724,
"num_tokens": 894198.0,
"step": 405
},
{
"entropy": 7.445980262756348,
"epoch": 0.0406907502977372,
"grad_norm": 1.2421875,
"learning_rate": 0.00020449999999999998,
"loss": 7.1122,
"mean_token_accuracy": 0.08235671296715737,
"num_tokens": 905394.0,
"step": 410
},
{
"entropy": 7.525100374221802,
"epoch": 0.041186978959904724,
"grad_norm": 1.1328125,
"learning_rate": 0.000207,
"loss": 7.1626,
"mean_token_accuracy": 0.07699285373091698,
"num_tokens": 917010.0,
"step": 415
},
{
"entropy": 7.316656255722046,
"epoch": 0.04168320762207225,
"grad_norm": 1.15625,
"learning_rate": 0.0002095,
"loss": 7.0005,
"mean_token_accuracy": 0.08732899129390717,
"num_tokens": 926761.0,
"step": 420
},
{
"entropy": 7.296533203125,
"epoch": 0.04217943628423978,
"grad_norm": 1.1484375,
"learning_rate": 0.000212,
"loss": 7.0074,
"mean_token_accuracy": 0.08514706641435624,
"num_tokens": 938385.0,
"step": 425
},
{
"entropy": 7.430690574645996,
"epoch": 0.0426756649464073,
"grad_norm": 0.97265625,
"learning_rate": 0.0002145,
"loss": 7.0374,
"mean_token_accuracy": 0.08442675322294235,
"num_tokens": 949709.0,
"step": 430
},
{
"entropy": 7.3450085639953615,
"epoch": 0.04317189360857483,
"grad_norm": 0.87890625,
"learning_rate": 0.00021700000000000002,
"loss": 7.0699,
"mean_token_accuracy": 0.07798392251133919,
"num_tokens": 960837.0,
"step": 435
},
{
"entropy": 7.343694639205933,
"epoch": 0.043668122270742356,
"grad_norm": 1.015625,
"learning_rate": 0.0002195,
"loss": 6.903,
"mean_token_accuracy": 0.08706051260232925,
"num_tokens": 971036.0,
"step": 440
},
{
"entropy": 7.371674299240112,
"epoch": 0.044164350932909886,
"grad_norm": 0.984375,
"learning_rate": 0.000222,
"loss": 7.0502,
"mean_token_accuracy": 0.0812618188560009,
"num_tokens": 983057.0,
"step": 445
},
{
"entropy": 7.4037517547607425,
"epoch": 0.04466057959507741,
"grad_norm": 0.98046875,
"learning_rate": 0.0002245,
"loss": 7.0069,
"mean_token_accuracy": 0.08837616667151452,
"num_tokens": 994303.0,
"step": 450
},
{
"entropy": 7.2432009220123295,
"epoch": 0.04515680825724494,
"grad_norm": 1.0390625,
"learning_rate": 0.00022700000000000002,
"loss": 7.0026,
"mean_token_accuracy": 0.0851232349872589,
"num_tokens": 1005033.0,
"step": 455
},
{
"entropy": 7.39683952331543,
"epoch": 0.045653036919412464,
"grad_norm": 0.90625,
"learning_rate": 0.00022950000000000002,
"loss": 7.0681,
"mean_token_accuracy": 0.08551012352108955,
"num_tokens": 1015940.0,
"step": 460
},
{
"entropy": 7.263912868499756,
"epoch": 0.046149265581579994,
"grad_norm": 0.92578125,
"learning_rate": 0.00023200000000000003,
"loss": 7.0144,
"mean_token_accuracy": 0.08879487365484237,
"num_tokens": 1027582.0,
"step": 465
},
{
"entropy": 7.333393192291259,
"epoch": 0.04664549424374752,
"grad_norm": 0.94140625,
"learning_rate": 0.00023449999999999998,
"loss": 6.9848,
"mean_token_accuracy": 0.08102063313126565,
"num_tokens": 1040046.0,
"step": 470
},
{
"entropy": 7.31667799949646,
"epoch": 0.04714172290591505,
"grad_norm": 1.4609375,
"learning_rate": 0.000237,
"loss": 7.0628,
"mean_token_accuracy": 0.07443273328244686,
"num_tokens": 1050636.0,
"step": 475
},
{
"entropy": 7.372964000701904,
"epoch": 0.04763795156808257,
"grad_norm": 1.1328125,
"learning_rate": 0.0002395,
"loss": 6.9388,
"mean_token_accuracy": 0.08691781386733055,
"num_tokens": 1060748.0,
"step": 480
},
{
"entropy": 7.3284331321716305,
"epoch": 0.0481341802302501,
"grad_norm": 0.98046875,
"learning_rate": 0.000242,
"loss": 6.9896,
"mean_token_accuracy": 0.07861718982458114,
"num_tokens": 1071940.0,
"step": 485
},
{
"entropy": 7.240336561203003,
"epoch": 0.048630408892417626,
"grad_norm": 1.0234375,
"learning_rate": 0.0002445,
"loss": 6.995,
"mean_token_accuracy": 0.08761729970574379,
"num_tokens": 1082962.0,
"step": 490
},
{
"entropy": 7.282444810867309,
"epoch": 0.04912663755458515,
"grad_norm": 1.125,
"learning_rate": 0.000247,
"loss": 6.9754,
"mean_token_accuracy": 0.09364678412675857,
"num_tokens": 1093082.0,
"step": 495
},
{
"entropy": 7.348704719543457,
"epoch": 0.04962286621675268,
"grad_norm": 1.0546875,
"learning_rate": 0.0002495,
"loss": 7.049,
"mean_token_accuracy": 0.08253686502575874,
"num_tokens": 1103528.0,
"step": 500
},
{
"entropy": 7.214434576034546,
"epoch": 0.0501190948789202,
"grad_norm": 1.3515625,
"learning_rate": 0.000252,
"loss": 6.9393,
"mean_token_accuracy": 0.08427478075027466,
"num_tokens": 1114813.0,
"step": 505
},
{
"entropy": 7.229425621032715,
"epoch": 0.050615323541087734,
"grad_norm": 1.1015625,
"learning_rate": 0.0002545,
"loss": 6.9222,
"mean_token_accuracy": 0.08125569894909859,
"num_tokens": 1125345.0,
"step": 510
},
{
"entropy": 7.092966032028198,
"epoch": 0.05111155220325526,
"grad_norm": 1.09375,
"learning_rate": 0.000257,
"loss": 6.9414,
"mean_token_accuracy": 0.08359554558992385,
"num_tokens": 1134985.0,
"step": 515
},
{
"entropy": 7.2688844203948975,
"epoch": 0.05160778086542279,
"grad_norm": 1.0625,
"learning_rate": 0.0002595,
"loss": 6.9457,
"mean_token_accuracy": 0.08630074635148048,
"num_tokens": 1145796.0,
"step": 520
},
{
"entropy": 7.270513343811035,
"epoch": 0.05210400952759031,
"grad_norm": 1.1953125,
"learning_rate": 0.000262,
"loss": 7.0875,
"mean_token_accuracy": 0.07355156876146793,
"num_tokens": 1156936.0,
"step": 525
},
{
"entropy": 7.258271265029907,
"epoch": 0.05260023818975784,
"grad_norm": 1.1328125,
"learning_rate": 0.00026450000000000003,
"loss": 6.96,
"mean_token_accuracy": 0.08982882723212242,
"num_tokens": 1167526.0,
"step": 530
},
{
"entropy": 7.198262119293213,
"epoch": 0.053096466851925365,
"grad_norm": 1.625,
"learning_rate": 0.00026700000000000004,
"loss": 6.9928,
"mean_token_accuracy": 0.08096232041716575,
"num_tokens": 1177647.0,
"step": 535
},
{
"entropy": 7.275991725921631,
"epoch": 0.053592695514092896,
"grad_norm": 1.15625,
"learning_rate": 0.00026950000000000005,
"loss": 6.9641,
"mean_token_accuracy": 0.08726390451192856,
"num_tokens": 1188255.0,
"step": 540
},
{
"entropy": 7.114196634292602,
"epoch": 0.05408892417626042,
"grad_norm": 1.1015625,
"learning_rate": 0.00027200000000000005,
"loss": 6.8869,
"mean_token_accuracy": 0.09180611073970794,
"num_tokens": 1199433.0,
"step": 545
},
{
"entropy": 7.082750701904297,
"epoch": 0.05458515283842795,
"grad_norm": 1.046875,
"learning_rate": 0.0002745,
"loss": 6.9654,
"mean_token_accuracy": 0.08378833159804344,
"num_tokens": 1210067.0,
"step": 550
},
{
"entropy": 7.160634613037109,
"epoch": 0.055081381500595473,
"grad_norm": 1.234375,
"learning_rate": 0.000277,
"loss": 6.8504,
"mean_token_accuracy": 0.08845426812767983,
"num_tokens": 1220639.0,
"step": 555
},
{
"entropy": 7.191676330566406,
"epoch": 0.055577610162763004,
"grad_norm": 0.9609375,
"learning_rate": 0.0002795,
"loss": 6.8845,
"mean_token_accuracy": 0.09016633853316307,
"num_tokens": 1233158.0,
"step": 560
},
{
"entropy": 7.2028326988220215,
"epoch": 0.05607383882493053,
"grad_norm": 1.0703125,
"learning_rate": 0.00028199999999999997,
"loss": 6.8726,
"mean_token_accuracy": 0.08301621749997139,
"num_tokens": 1244552.0,
"step": 565
},
{
"entropy": 7.165884399414063,
"epoch": 0.05657006748709806,
"grad_norm": 0.9609375,
"learning_rate": 0.0002845,
"loss": 6.9698,
"mean_token_accuracy": 0.0811222843825817,
"num_tokens": 1255478.0,
"step": 570
},
{
"entropy": 7.039954090118409,
"epoch": 0.05706629614926558,
"grad_norm": 0.98046875,
"learning_rate": 0.000287,
"loss": 6.7999,
"mean_token_accuracy": 0.09229947701096534,
"num_tokens": 1267507.0,
"step": 575
},
{
"entropy": 7.070255470275879,
"epoch": 0.057562524811433105,
"grad_norm": 1.0234375,
"learning_rate": 0.0002895,
"loss": 6.8057,
"mean_token_accuracy": 0.0856806755065918,
"num_tokens": 1279377.0,
"step": 580
},
{
"entropy": 7.210440063476563,
"epoch": 0.058058753473600636,
"grad_norm": 1.1328125,
"learning_rate": 0.000292,
"loss": 6.8795,
"mean_token_accuracy": 0.09145521745085716,
"num_tokens": 1290671.0,
"step": 585
},
{
"entropy": 6.9877747058868405,
"epoch": 0.05855498213576816,
"grad_norm": 1.0859375,
"learning_rate": 0.0002945,
"loss": 6.8592,
"mean_token_accuracy": 0.0912296824157238,
"num_tokens": 1302792.0,
"step": 590
},
{
"entropy": 7.064874172210693,
"epoch": 0.05905121079793569,
"grad_norm": 1.1328125,
"learning_rate": 0.000297,
"loss": 6.8074,
"mean_token_accuracy": 0.08353102207183838,
"num_tokens": 1313315.0,
"step": 595
},
{
"entropy": 7.0152363777160645,
"epoch": 0.05954743946010321,
"grad_norm": 1.015625,
"learning_rate": 0.0002995,
"loss": 6.881,
"mean_token_accuracy": 0.0832609310746193,
"num_tokens": 1326358.0,
"step": 600
},
{
"entropy": 7.107955121994019,
"epoch": 0.060043668122270744,
"grad_norm": 1.0234375,
"learning_rate": 0.000302,
"loss": 6.7556,
"mean_token_accuracy": 0.0891527384519577,
"num_tokens": 1336266.0,
"step": 605
},
{
"entropy": 6.989856767654419,
"epoch": 0.06053989678443827,
"grad_norm": 1.0703125,
"learning_rate": 0.0003045,
"loss": 6.7067,
"mean_token_accuracy": 0.09343992173671722,
"num_tokens": 1347610.0,
"step": 610
},
{
"entropy": 6.953636312484742,
"epoch": 0.0610361254466058,
"grad_norm": 1.109375,
"learning_rate": 0.000307,
"loss": 6.676,
"mean_token_accuracy": 0.09296446591615677,
"num_tokens": 1358317.0,
"step": 615
},
{
"entropy": 6.972737407684326,
"epoch": 0.06153235410877332,
"grad_norm": 1.109375,
"learning_rate": 0.0003095,
"loss": 6.7807,
"mean_token_accuracy": 0.08731090128421784,
"num_tokens": 1370126.0,
"step": 620
},
{
"entropy": 7.019057750701904,
"epoch": 0.06202858277094085,
"grad_norm": 1.0703125,
"learning_rate": 0.000312,
"loss": 6.7463,
"mean_token_accuracy": 0.09216577932238579,
"num_tokens": 1381571.0,
"step": 625
},
{
"entropy": 6.846717262268067,
"epoch": 0.06252481143310838,
"grad_norm": 1.140625,
"learning_rate": 0.0003145,
"loss": 6.773,
"mean_token_accuracy": 0.09080302193760872,
"num_tokens": 1392486.0,
"step": 630
},
{
"entropy": 7.055567216873169,
"epoch": 0.0630210400952759,
"grad_norm": 0.95703125,
"learning_rate": 0.000317,
"loss": 6.7758,
"mean_token_accuracy": 0.09265405684709549,
"num_tokens": 1403086.0,
"step": 635
},
{
"entropy": 6.926388502120972,
"epoch": 0.06351726875744343,
"grad_norm": 0.98046875,
"learning_rate": 0.0003195,
"loss": 6.7449,
"mean_token_accuracy": 0.09237656593322754,
"num_tokens": 1414537.0,
"step": 640
},
{
"entropy": 6.949403572082519,
"epoch": 0.06401349741961096,
"grad_norm": 1.125,
"learning_rate": 0.000322,
"loss": 6.7572,
"mean_token_accuracy": 0.09533382952213287,
"num_tokens": 1426998.0,
"step": 645
},
{
"entropy": 6.947071886062622,
"epoch": 0.06450972608177849,
"grad_norm": 0.98046875,
"learning_rate": 0.00032450000000000003,
"loss": 6.7806,
"mean_token_accuracy": 0.09227192178368568,
"num_tokens": 1437667.0,
"step": 650
},
{
"entropy": 6.933514261245728,
"epoch": 0.065005954743946,
"grad_norm": 0.9296875,
"learning_rate": 0.00032700000000000003,
"loss": 6.652,
"mean_token_accuracy": 0.09583644419908524,
"num_tokens": 1448280.0,
"step": 655
},
{
"entropy": 6.916765022277832,
"epoch": 0.06550218340611354,
"grad_norm": 1.2421875,
"learning_rate": 0.00032950000000000004,
"loss": 6.7201,
"mean_token_accuracy": 0.09147566556930542,
"num_tokens": 1458608.0,
"step": 660
},
{
"entropy": 6.95277361869812,
"epoch": 0.06599841206828107,
"grad_norm": 1.484375,
"learning_rate": 0.00033200000000000005,
"loss": 6.9062,
"mean_token_accuracy": 0.08369888290762902,
"num_tokens": 1470435.0,
"step": 665
},
{
"entropy": 6.9358902931213375,
"epoch": 0.06649464073044858,
"grad_norm": 1.0859375,
"learning_rate": 0.00033450000000000005,
"loss": 6.7515,
"mean_token_accuracy": 0.0901063933968544,
"num_tokens": 1481959.0,
"step": 670
},
{
"entropy": 6.907770204544067,
"epoch": 0.06699086939261611,
"grad_norm": 1.1484375,
"learning_rate": 0.000337,
"loss": 6.7936,
"mean_token_accuracy": 0.08672575987875461,
"num_tokens": 1493798.0,
"step": 675
},
{
"entropy": 6.924069547653199,
"epoch": 0.06748709805478365,
"grad_norm": 1.109375,
"learning_rate": 0.0003395,
"loss": 6.6036,
"mean_token_accuracy": 0.09497400969266892,
"num_tokens": 1502969.0,
"step": 680
},
{
"entropy": 6.881195592880249,
"epoch": 0.06798332671695118,
"grad_norm": 1.109375,
"learning_rate": 0.000342,
"loss": 6.7521,
"mean_token_accuracy": 0.09069281220436096,
"num_tokens": 1513825.0,
"step": 685
},
{
"entropy": 6.85332555770874,
"epoch": 0.06847955537911869,
"grad_norm": 0.87890625,
"learning_rate": 0.00034449999999999997,
"loss": 6.6429,
"mean_token_accuracy": 0.0923303596675396,
"num_tokens": 1525328.0,
"step": 690
},
{
"entropy": 6.871306228637695,
"epoch": 0.06897578404128622,
"grad_norm": 0.96875,
"learning_rate": 0.000347,
"loss": 6.6631,
"mean_token_accuracy": 0.09412776604294777,
"num_tokens": 1537383.0,
"step": 695
},
{
"entropy": 6.971198177337646,
"epoch": 0.06947201270345375,
"grad_norm": 1.078125,
"learning_rate": 0.0003495,
"loss": 6.7733,
"mean_token_accuracy": 0.09241362139582635,
"num_tokens": 1549671.0,
"step": 700
},
{
"entropy": 6.766777658462525,
"epoch": 0.06996824136562128,
"grad_norm": 1.046875,
"learning_rate": 0.000352,
"loss": 6.6454,
"mean_token_accuracy": 0.09111724868416786,
"num_tokens": 1559974.0,
"step": 705
},
{
"entropy": 6.925774097442627,
"epoch": 0.0704644700277888,
"grad_norm": 0.97265625,
"learning_rate": 0.0003545,
"loss": 6.7489,
"mean_token_accuracy": 0.08958815187215804,
"num_tokens": 1570890.0,
"step": 710
},
{
"entropy": 6.800432395935059,
"epoch": 0.07096069868995633,
"grad_norm": 1.140625,
"learning_rate": 0.000357,
"loss": 6.6629,
"mean_token_accuracy": 0.09856393039226533,
"num_tokens": 1581434.0,
"step": 715
},
{
"entropy": 6.8169127941131595,
"epoch": 0.07145692735212386,
"grad_norm": 1.078125,
"learning_rate": 0.0003595,
"loss": 6.7193,
"mean_token_accuracy": 0.09074361547827721,
"num_tokens": 1593710.0,
"step": 720
},
{
"entropy": 6.93365249633789,
"epoch": 0.07195315601429139,
"grad_norm": 1.0546875,
"learning_rate": 0.000362,
"loss": 6.6931,
"mean_token_accuracy": 0.09370483383536339,
"num_tokens": 1604635.0,
"step": 725
},
{
"entropy": 6.909399318695068,
"epoch": 0.07244938467645891,
"grad_norm": 0.98828125,
"learning_rate": 0.0003645,
"loss": 6.7543,
"mean_token_accuracy": 0.09272385239601136,
"num_tokens": 1615258.0,
"step": 730
},
{
"entropy": 6.855506086349488,
"epoch": 0.07294561333862644,
"grad_norm": 1.015625,
"learning_rate": 0.000367,
"loss": 6.6587,
"mean_token_accuracy": 0.0944454848766327,
"num_tokens": 1625942.0,
"step": 735
},
{
"entropy": 6.828038263320923,
"epoch": 0.07344184200079397,
"grad_norm": 1.0078125,
"learning_rate": 0.0003695,
"loss": 6.6292,
"mean_token_accuracy": 0.10174066349864005,
"num_tokens": 1636270.0,
"step": 740
},
{
"entropy": 6.729400539398194,
"epoch": 0.07393807066296149,
"grad_norm": 1.0625,
"learning_rate": 0.000372,
"loss": 6.6152,
"mean_token_accuracy": 0.0955189272761345,
"num_tokens": 1647029.0,
"step": 745
},
{
"entropy": 6.830958700180053,
"epoch": 0.07443429932512902,
"grad_norm": 1.109375,
"learning_rate": 0.0003745,
"loss": 6.7986,
"mean_token_accuracy": 0.0909816287457943,
"num_tokens": 1659483.0,
"step": 750
},
{
"entropy": 6.791053581237793,
"epoch": 0.07493052798729655,
"grad_norm": 0.88671875,
"learning_rate": 0.000377,
"loss": 6.638,
"mean_token_accuracy": 0.09132714346051216,
"num_tokens": 1669425.0,
"step": 755
},
{
"entropy": 6.777761650085449,
"epoch": 0.07542675664946408,
"grad_norm": 0.9375,
"learning_rate": 0.0003795,
"loss": 6.6407,
"mean_token_accuracy": 0.09279207661747932,
"num_tokens": 1680426.0,
"step": 760
},
{
"entropy": 6.760220146179199,
"epoch": 0.0759229853116316,
"grad_norm": 0.96484375,
"learning_rate": 0.000382,
"loss": 6.6565,
"mean_token_accuracy": 0.09227121248841286,
"num_tokens": 1692803.0,
"step": 765
},
{
"entropy": 6.808647537231446,
"epoch": 0.07641921397379912,
"grad_norm": 0.85546875,
"learning_rate": 0.0003845,
"loss": 6.6722,
"mean_token_accuracy": 0.09433563277125359,
"num_tokens": 1703919.0,
"step": 770
},
{
"entropy": 6.796703290939331,
"epoch": 0.07691544263596665,
"grad_norm": 0.90625,
"learning_rate": 0.00038700000000000003,
"loss": 6.6536,
"mean_token_accuracy": 0.10081230103969574,
"num_tokens": 1714487.0,
"step": 775
},
{
"entropy": 6.826533603668213,
"epoch": 0.07741167129813419,
"grad_norm": 1.2890625,
"learning_rate": 0.00038950000000000003,
"loss": 6.6979,
"mean_token_accuracy": 0.09503390416502952,
"num_tokens": 1726601.0,
"step": 780
},
{
"entropy": 6.7289519786834715,
"epoch": 0.0779078999603017,
"grad_norm": 0.98828125,
"learning_rate": 0.00039200000000000004,
"loss": 6.5999,
"mean_token_accuracy": 0.09352485463023186,
"num_tokens": 1738743.0,
"step": 785
},
{
"entropy": 6.731641817092895,
"epoch": 0.07840412862246923,
"grad_norm": 0.98046875,
"learning_rate": 0.00039450000000000005,
"loss": 6.6684,
"mean_token_accuracy": 0.09545368328690529,
"num_tokens": 1748286.0,
"step": 790
},
{
"entropy": 6.755449342727661,
"epoch": 0.07890035728463676,
"grad_norm": 0.95703125,
"learning_rate": 0.00039700000000000005,
"loss": 6.7012,
"mean_token_accuracy": 0.10006179735064506,
"num_tokens": 1758917.0,
"step": 795
},
{
"entropy": 6.707924747467041,
"epoch": 0.0793965859468043,
"grad_norm": 0.96484375,
"learning_rate": 0.0003995,
"loss": 6.5299,
"mean_token_accuracy": 0.09730249717831611,
"num_tokens": 1769179.0,
"step": 800
},
{
"entropy": 6.734688758850098,
"epoch": 0.07989281460897181,
"grad_norm": 0.91015625,
"learning_rate": 0.000402,
"loss": 6.6672,
"mean_token_accuracy": 0.09266628846526145,
"num_tokens": 1782369.0,
"step": 805
},
{
"entropy": 6.783080816268921,
"epoch": 0.08038904327113934,
"grad_norm": 1.0625,
"learning_rate": 0.0004045,
"loss": 6.6539,
"mean_token_accuracy": 0.1000369630753994,
"num_tokens": 1795992.0,
"step": 810
},
{
"entropy": 6.690567588806152,
"epoch": 0.08088527193330687,
"grad_norm": 0.9296875,
"learning_rate": 0.00040699999999999997,
"loss": 6.5647,
"mean_token_accuracy": 0.10151613354682923,
"num_tokens": 1806938.0,
"step": 815
},
{
"entropy": 6.62420883178711,
"epoch": 0.0813815005954744,
"grad_norm": 0.99609375,
"learning_rate": 0.0004095,
"loss": 6.489,
"mean_token_accuracy": 0.10247653797268867,
"num_tokens": 1817663.0,
"step": 820
},
{
"entropy": 6.7051942348480225,
"epoch": 0.08187772925764192,
"grad_norm": 0.98046875,
"learning_rate": 0.000412,
"loss": 6.6431,
"mean_token_accuracy": 0.09830244034528732,
"num_tokens": 1828254.0,
"step": 825
},
{
"entropy": 6.740552711486816,
"epoch": 0.08237395791980945,
"grad_norm": 1.15625,
"learning_rate": 0.0004145,
"loss": 6.5703,
"mean_token_accuracy": 0.09990052878856659,
"num_tokens": 1839001.0,
"step": 830
},
{
"entropy": 6.675515031814575,
"epoch": 0.08287018658197698,
"grad_norm": 0.94921875,
"learning_rate": 0.000417,
"loss": 6.6802,
"mean_token_accuracy": 0.09807795733213424,
"num_tokens": 1849906.0,
"step": 835
},
{
"entropy": 6.5840569019317625,
"epoch": 0.0833664152441445,
"grad_norm": 1.03125,
"learning_rate": 0.0004195,
"loss": 6.5699,
"mean_token_accuracy": 0.09285617843270302,
"num_tokens": 1860982.0,
"step": 840
},
{
"entropy": 6.794158124923706,
"epoch": 0.08386264390631203,
"grad_norm": 0.9140625,
"learning_rate": 0.000422,
"loss": 6.5139,
"mean_token_accuracy": 0.10143777877092361,
"num_tokens": 1872177.0,
"step": 845
},
{
"entropy": 6.596243000030517,
"epoch": 0.08435887256847956,
"grad_norm": 0.9609375,
"learning_rate": 0.0004245,
"loss": 6.4961,
"mean_token_accuracy": 0.09911343604326248,
"num_tokens": 1882783.0,
"step": 850
},
{
"entropy": 6.771530103683472,
"epoch": 0.08485510123064709,
"grad_norm": 1.0625,
"learning_rate": 0.000427,
"loss": 6.6692,
"mean_token_accuracy": 0.09356241896748543,
"num_tokens": 1893706.0,
"step": 855
},
{
"entropy": 6.628360414505005,
"epoch": 0.0853513298928146,
"grad_norm": 1.03125,
"learning_rate": 0.0004295,
"loss": 6.545,
"mean_token_accuracy": 0.09712332859635353,
"num_tokens": 1903852.0,
"step": 860
},
{
"entropy": 6.61391053199768,
"epoch": 0.08584755855498213,
"grad_norm": 1.0546875,
"learning_rate": 0.000432,
"loss": 6.5803,
"mean_token_accuracy": 0.10353554785251617,
"num_tokens": 1914157.0,
"step": 865
},
{
"entropy": 6.660257244110108,
"epoch": 0.08634378721714966,
"grad_norm": 0.90625,
"learning_rate": 0.0004345,
"loss": 6.507,
"mean_token_accuracy": 0.09838124290108681,
"num_tokens": 1924684.0,
"step": 870
},
{
"entropy": 6.690595579147339,
"epoch": 0.0868400158793172,
"grad_norm": 0.91796875,
"learning_rate": 0.000437,
"loss": 6.6703,
"mean_token_accuracy": 0.09410161599516868,
"num_tokens": 1935657.0,
"step": 875
},
{
"entropy": 6.611545896530151,
"epoch": 0.08733624454148471,
"grad_norm": 0.89453125,
"learning_rate": 0.0004395,
"loss": 6.5409,
"mean_token_accuracy": 0.09920520260930062,
"num_tokens": 1946735.0,
"step": 880
},
{
"entropy": 6.714548015594483,
"epoch": 0.08783247320365224,
"grad_norm": 1.1015625,
"learning_rate": 0.000442,
"loss": 6.7169,
"mean_token_accuracy": 0.09320614784955979,
"num_tokens": 1957871.0,
"step": 885
},
{
"entropy": 6.615920686721802,
"epoch": 0.08832870186581977,
"grad_norm": 1.1953125,
"learning_rate": 0.0004445,
"loss": 6.5327,
"mean_token_accuracy": 0.09827413335442543,
"num_tokens": 1969032.0,
"step": 890
},
{
"entropy": 6.704651021957398,
"epoch": 0.0888249305279873,
"grad_norm": 1.0078125,
"learning_rate": 0.000447,
"loss": 6.5488,
"mean_token_accuracy": 0.09838233441114426,
"num_tokens": 1980212.0,
"step": 895
},
{
"entropy": 6.558094882965088,
"epoch": 0.08932115919015482,
"grad_norm": 0.94921875,
"learning_rate": 0.00044950000000000003,
"loss": 6.4855,
"mean_token_accuracy": 0.09865308403968812,
"num_tokens": 1990333.0,
"step": 900
},
{
"entropy": 6.647093868255615,
"epoch": 0.08981738785232235,
"grad_norm": 0.984375,
"learning_rate": 0.00045200000000000004,
"loss": 6.5654,
"mean_token_accuracy": 0.10231334716081619,
"num_tokens": 2001486.0,
"step": 905
},
{
"entropy": 6.630073118209839,
"epoch": 0.09031361651448988,
"grad_norm": 1.15625,
"learning_rate": 0.00045450000000000004,
"loss": 6.5384,
"mean_token_accuracy": 0.10137662813067436,
"num_tokens": 2012535.0,
"step": 910
},
{
"entropy": 6.546298170089722,
"epoch": 0.0908098451766574,
"grad_norm": 0.98828125,
"learning_rate": 0.00045700000000000005,
"loss": 6.4936,
"mean_token_accuracy": 0.10514995977282524,
"num_tokens": 2024116.0,
"step": 915
},
{
"entropy": 6.470860958099365,
"epoch": 0.09130607383882493,
"grad_norm": 0.83203125,
"learning_rate": 0.00045950000000000006,
"loss": 6.5403,
"mean_token_accuracy": 0.09997001364827156,
"num_tokens": 2036341.0,
"step": 920
},
{
"entropy": 6.682841682434082,
"epoch": 0.09180230250099246,
"grad_norm": 0.94140625,
"learning_rate": 0.000462,
"loss": 6.5173,
"mean_token_accuracy": 0.09831630811095238,
"num_tokens": 2046768.0,
"step": 925
},
{
"entropy": 6.5218424797058105,
"epoch": 0.09229853116315999,
"grad_norm": 0.95703125,
"learning_rate": 0.0004645,
"loss": 6.485,
"mean_token_accuracy": 0.10012486949563026,
"num_tokens": 2057315.0,
"step": 930
},
{
"entropy": 6.539970731735229,
"epoch": 0.0927947598253275,
"grad_norm": 1.0,
"learning_rate": 0.000467,
"loss": 6.4551,
"mean_token_accuracy": 0.10452077984809875,
"num_tokens": 2068267.0,
"step": 935
},
{
"entropy": 6.538546848297119,
"epoch": 0.09329098848749504,
"grad_norm": 0.9921875,
"learning_rate": 0.0004695,
"loss": 6.5244,
"mean_token_accuracy": 0.10441852286458016,
"num_tokens": 2078271.0,
"step": 940
},
{
"entropy": 6.543840551376343,
"epoch": 0.09378721714966257,
"grad_norm": 1.0078125,
"learning_rate": 0.000472,
"loss": 6.4694,
"mean_token_accuracy": 0.09577009156346321,
"num_tokens": 2088381.0,
"step": 945
},
{
"entropy": 6.56317892074585,
"epoch": 0.0942834458118301,
"grad_norm": 1.0234375,
"learning_rate": 0.0004745,
"loss": 6.4148,
"mean_token_accuracy": 0.09971964061260223,
"num_tokens": 2098899.0,
"step": 950
},
{
"entropy": 6.514262676239014,
"epoch": 0.09477967447399761,
"grad_norm": 1.0234375,
"learning_rate": 0.000477,
"loss": 6.5697,
"mean_token_accuracy": 0.09980119243264199,
"num_tokens": 2110057.0,
"step": 955
},
{
"entropy": 6.484142780303955,
"epoch": 0.09527590313616514,
"grad_norm": 0.96484375,
"learning_rate": 0.0004795,
"loss": 6.4877,
"mean_token_accuracy": 0.10196142494678498,
"num_tokens": 2121758.0,
"step": 960
},
{
"entropy": 6.572415971755982,
"epoch": 0.09577213179833267,
"grad_norm": 0.921875,
"learning_rate": 0.000482,
"loss": 6.4947,
"mean_token_accuracy": 0.10148481652140617,
"num_tokens": 2131840.0,
"step": 965
},
{
"entropy": 6.479370641708374,
"epoch": 0.0962683604605002,
"grad_norm": 1.0625,
"learning_rate": 0.0004845,
"loss": 6.4686,
"mean_token_accuracy": 0.09988197311758995,
"num_tokens": 2142084.0,
"step": 970
},
{
"entropy": 6.543777370452881,
"epoch": 0.09676458912266772,
"grad_norm": 0.91796875,
"learning_rate": 0.000487,
"loss": 6.5201,
"mean_token_accuracy": 0.10057111382484436,
"num_tokens": 2152785.0,
"step": 975
},
{
"entropy": 6.58097357749939,
"epoch": 0.09726081778483525,
"grad_norm": 0.97265625,
"learning_rate": 0.0004895,
"loss": 6.5173,
"mean_token_accuracy": 0.09812606722116471,
"num_tokens": 2163774.0,
"step": 980
},
{
"entropy": 6.429722023010254,
"epoch": 0.09775704644700278,
"grad_norm": 0.859375,
"learning_rate": 0.000492,
"loss": 6.4215,
"mean_token_accuracy": 0.10499785169959068,
"num_tokens": 2175378.0,
"step": 985
},
{
"entropy": 6.654616022109986,
"epoch": 0.0982532751091703,
"grad_norm": 0.91015625,
"learning_rate": 0.0004945,
"loss": 6.6198,
"mean_token_accuracy": 0.10243403911590576,
"num_tokens": 2186510.0,
"step": 990
},
{
"entropy": 6.519903516769409,
"epoch": 0.09874950377133783,
"grad_norm": 0.87109375,
"learning_rate": 0.000497,
"loss": 6.4311,
"mean_token_accuracy": 0.10723361372947693,
"num_tokens": 2197482.0,
"step": 995
},
{
"entropy": 6.421766948699951,
"epoch": 0.09924573243350536,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995,
"loss": 6.415,
"mean_token_accuracy": 0.10275844857096672,
"num_tokens": 2208329.0,
"step": 1000
},
{
"entropy": 6.386524295806884,
"epoch": 0.09974196109567289,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999999982145551,
"loss": 6.328,
"mean_token_accuracy": 0.1056441530585289,
"num_tokens": 2218632.0,
"step": 1005
},
{
"entropy": 6.481380987167358,
"epoch": 0.1002381897578404,
"grad_norm": 0.80078125,
"learning_rate": 0.0004999999909611855,
"loss": 6.4444,
"mean_token_accuracy": 0.1021676316857338,
"num_tokens": 2231335.0,
"step": 1010
},
{
"entropy": 6.500467729568482,
"epoch": 0.10073441842000794,
"grad_norm": 1.046875,
"learning_rate": 0.0004999999781283008,
"loss": 6.4356,
"mean_token_accuracy": 0.10310698971152306,
"num_tokens": 2241704.0,
"step": 1015
},
{
"entropy": 6.412548065185547,
"epoch": 0.10123064708217547,
"grad_norm": 0.96875,
"learning_rate": 0.0004999999597159017,
"loss": 6.5081,
"mean_token_accuracy": 0.09907391145825387,
"num_tokens": 2252310.0,
"step": 1020
},
{
"entropy": 6.580806827545166,
"epoch": 0.101726875744343,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999999357239882,
"loss": 6.4046,
"mean_token_accuracy": 0.10047375708818436,
"num_tokens": 2262270.0,
"step": 1025
},
{
"entropy": 6.458594131469726,
"epoch": 0.10222310440651051,
"grad_norm": 0.984375,
"learning_rate": 0.0004999999061525612,
"loss": 6.5467,
"mean_token_accuracy": 0.09957677125930786,
"num_tokens": 2273986.0,
"step": 1030
},
{
"entropy": 6.567448663711548,
"epoch": 0.10271933306867805,
"grad_norm": 1.046875,
"learning_rate": 0.0004999998710016214,
"loss": 6.4326,
"mean_token_accuracy": 0.10295439511537552,
"num_tokens": 2283430.0,
"step": 1035
},
{
"entropy": 6.519222831726074,
"epoch": 0.10321556173084558,
"grad_norm": 0.984375,
"learning_rate": 0.0004999998302711696,
"loss": 6.446,
"mean_token_accuracy": 0.09964360818266868,
"num_tokens": 2294201.0,
"step": 1040
},
{
"entropy": 6.324606132507324,
"epoch": 0.1037117903930131,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999997839612069,
"loss": 6.3525,
"mean_token_accuracy": 0.10616270750761032,
"num_tokens": 2303876.0,
"step": 1045
},
{
"entropy": 6.41707067489624,
"epoch": 0.10420801905518062,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999997320717342,
"loss": 6.3981,
"mean_token_accuracy": 0.10809943452477455,
"num_tokens": 2313713.0,
"step": 1050
},
{
"entropy": 6.40936689376831,
"epoch": 0.10470424771734815,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999996746027531,
"loss": 6.4,
"mean_token_accuracy": 0.10011846721172332,
"num_tokens": 2324862.0,
"step": 1055
},
{
"entropy": 6.488155174255371,
"epoch": 0.10520047637951568,
"grad_norm": 1.0625,
"learning_rate": 0.000499999611554265,
"loss": 6.4697,
"mean_token_accuracy": 0.10407731533050538,
"num_tokens": 2334974.0,
"step": 1060
},
{
"entropy": 6.491655540466309,
"epoch": 0.10569670504168321,
"grad_norm": 0.921875,
"learning_rate": 0.0004999995429262712,
"loss": 6.5141,
"mean_token_accuracy": 0.10263812616467476,
"num_tokens": 2345945.0,
"step": 1065
},
{
"entropy": 6.451990365982056,
"epoch": 0.10619293370385073,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999994687187737,
"loss": 6.424,
"mean_token_accuracy": 0.10567846298217773,
"num_tokens": 2356749.0,
"step": 1070
},
{
"entropy": 6.4968970775604244,
"epoch": 0.10668916236601826,
"grad_norm": 0.890625,
"learning_rate": 0.0004999993889317742,
"loss": 6.5025,
"mean_token_accuracy": 0.09787843525409698,
"num_tokens": 2368181.0,
"step": 1075
},
{
"entropy": 6.48430438041687,
"epoch": 0.10718539102818579,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999993035652746,
"loss": 6.4122,
"mean_token_accuracy": 0.10403527170419694,
"num_tokens": 2377737.0,
"step": 1080
},
{
"entropy": 6.356083869934082,
"epoch": 0.10768161969035331,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999992126192771,
"loss": 6.3292,
"mean_token_accuracy": 0.10865161493420601,
"num_tokens": 2388944.0,
"step": 1085
},
{
"entropy": 6.479707813262939,
"epoch": 0.10817784835252084,
"grad_norm": 1.0546875,
"learning_rate": 0.000499999116093784,
"loss": 6.4133,
"mean_token_accuracy": 0.10108591020107269,
"num_tokens": 2400758.0,
"step": 1090
},
{
"entropy": 6.363013744354248,
"epoch": 0.10867407701468837,
"grad_norm": 0.890625,
"learning_rate": 0.0004999990139887976,
"loss": 6.4044,
"mean_token_accuracy": 0.10706018880009652,
"num_tokens": 2411916.0,
"step": 1095
},
{
"entropy": 6.606807518005371,
"epoch": 0.1091703056768559,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999989063043206,
"loss": 6.4428,
"mean_token_accuracy": 0.10081739947199822,
"num_tokens": 2423310.0,
"step": 1100
},
{
"entropy": 6.337709760665893,
"epoch": 0.10966653433902342,
"grad_norm": 1.09375,
"learning_rate": 0.0004999987930403555,
"loss": 6.3665,
"mean_token_accuracy": 0.10392144694924355,
"num_tokens": 2432100.0,
"step": 1105
},
{
"entropy": 6.414704656600952,
"epoch": 0.11016276300119095,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999986741969051,
"loss": 6.4419,
"mean_token_accuracy": 0.106266438215971,
"num_tokens": 2443563.0,
"step": 1110
},
{
"entropy": 6.537985706329346,
"epoch": 0.11065899166335848,
"grad_norm": 1.015625,
"learning_rate": 0.0004999985497739724,
"loss": 6.4454,
"mean_token_accuracy": 0.10320524126291275,
"num_tokens": 2455435.0,
"step": 1115
},
{
"entropy": 6.321274328231811,
"epoch": 0.11115522032552601,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999984197715606,
"loss": 6.3804,
"mean_token_accuracy": 0.10974763482809066,
"num_tokens": 2466173.0,
"step": 1120
},
{
"entropy": 6.357246780395508,
"epoch": 0.11165144898769352,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999982841896728,
"loss": 6.3293,
"mean_token_accuracy": 0.10533915609121322,
"num_tokens": 2476349.0,
"step": 1125
},
{
"entropy": 6.488060379028321,
"epoch": 0.11214767764986106,
"grad_norm": 0.875,
"learning_rate": 0.0004999981430283123,
"loss": 6.4613,
"mean_token_accuracy": 0.10952543765306473,
"num_tokens": 2487864.0,
"step": 1130
},
{
"entropy": 6.366047811508179,
"epoch": 0.11264390631202859,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999979962874827,
"loss": 6.3898,
"mean_token_accuracy": 0.10598978772759438,
"num_tokens": 2499188.0,
"step": 1135
},
{
"entropy": 6.48383674621582,
"epoch": 0.11314013497419612,
"grad_norm": 0.953125,
"learning_rate": 0.0004999978439671877,
"loss": 6.3782,
"mean_token_accuracy": 0.10689363330602646,
"num_tokens": 2509813.0,
"step": 1140
},
{
"entropy": 6.394137144088745,
"epoch": 0.11363636363636363,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999976860674309,
"loss": 6.4071,
"mean_token_accuracy": 0.10305101722478867,
"num_tokens": 2520813.0,
"step": 1145
},
{
"entropy": 6.387377500534058,
"epoch": 0.11413259229853116,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999975225882164,
"loss": 6.3967,
"mean_token_accuracy": 0.10229461118578911,
"num_tokens": 2531802.0,
"step": 1150
},
{
"entropy": 6.323769569396973,
"epoch": 0.1146288209606987,
"grad_norm": 1.0078125,
"learning_rate": 0.000499997353529548,
"loss": 6.2047,
"mean_token_accuracy": 0.12049669027328491,
"num_tokens": 2541935.0,
"step": 1155
},
{
"entropy": 6.31162691116333,
"epoch": 0.11512504962286621,
"grad_norm": 1.046875,
"learning_rate": 0.0004999971788914303,
"loss": 6.3047,
"mean_token_accuracy": 0.1079749420285225,
"num_tokens": 2553224.0,
"step": 1160
},
{
"entropy": 6.3400167465209964,
"epoch": 0.11562127828503374,
"grad_norm": 0.83203125,
"learning_rate": 0.0004999969986738672,
"loss": 6.3378,
"mean_token_accuracy": 0.10177241489291192,
"num_tokens": 2565355.0,
"step": 1165
},
{
"entropy": 6.436748456954956,
"epoch": 0.11611750694720127,
"grad_norm": 1.0703125,
"learning_rate": 0.0004999968128768634,
"loss": 6.4301,
"mean_token_accuracy": 0.1053300179541111,
"num_tokens": 2575962.0,
"step": 1170
},
{
"entropy": 6.377365303039551,
"epoch": 0.1166137356093688,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999966215004236,
"loss": 6.4384,
"mean_token_accuracy": 0.11097546443343162,
"num_tokens": 2588136.0,
"step": 1175
},
{
"entropy": 6.446149015426636,
"epoch": 0.11710996427153632,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999964245445522,
"loss": 6.3112,
"mean_token_accuracy": 0.10942729488015175,
"num_tokens": 2597785.0,
"step": 1180
},
{
"entropy": 6.339339637756348,
"epoch": 0.11760619293370385,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999962220092545,
"loss": 6.3702,
"mean_token_accuracy": 0.11220366284251213,
"num_tokens": 2609738.0,
"step": 1185
},
{
"entropy": 6.273840665817261,
"epoch": 0.11810242159587138,
"grad_norm": 0.921875,
"learning_rate": 0.0004999960138945351,
"loss": 6.2457,
"mean_token_accuracy": 0.11884041801095009,
"num_tokens": 2620179.0,
"step": 1190
},
{
"entropy": 6.364072942733765,
"epoch": 0.11859865025803891,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999958002003996,
"loss": 6.2789,
"mean_token_accuracy": 0.10949316993355751,
"num_tokens": 2631550.0,
"step": 1195
},
{
"entropy": 6.356981039047241,
"epoch": 0.11909487892020643,
"grad_norm": 0.875,
"learning_rate": 0.0004999955809268529,
"loss": 6.33,
"mean_token_accuracy": 0.10574184879660606,
"num_tokens": 2642467.0,
"step": 1200
},
{
"entropy": 6.3026384830474855,
"epoch": 0.11959110758237396,
"grad_norm": 1.125,
"learning_rate": 0.0004999953560739008,
"loss": 6.3069,
"mean_token_accuracy": 0.10857978910207748,
"num_tokens": 2652132.0,
"step": 1205
},
{
"entropy": 6.351331615447998,
"epoch": 0.12008733624454149,
"grad_norm": 0.859375,
"learning_rate": 0.0004999951256415485,
"loss": 6.3573,
"mean_token_accuracy": 0.10829652771353722,
"num_tokens": 2663336.0,
"step": 1210
},
{
"entropy": 6.465686893463134,
"epoch": 0.12058356490670902,
"grad_norm": 1.0390625,
"learning_rate": 0.000499994889629802,
"loss": 6.4458,
"mean_token_accuracy": 0.09753695875406265,
"num_tokens": 2674201.0,
"step": 1215
},
{
"entropy": 6.390116167068482,
"epoch": 0.12107979356887653,
"grad_norm": 0.98046875,
"learning_rate": 0.000499994648038667,
"loss": 6.3737,
"mean_token_accuracy": 0.11085178926587105,
"num_tokens": 2685073.0,
"step": 1220
},
{
"entropy": 6.4054559707641605,
"epoch": 0.12157602223104406,
"grad_norm": 0.94921875,
"learning_rate": 0.0004999944008681496,
"loss": 6.3104,
"mean_token_accuracy": 0.10459840670228004,
"num_tokens": 2695024.0,
"step": 1225
},
{
"entropy": 6.346751165390015,
"epoch": 0.1220722508932116,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999941481182559,
"loss": 6.3983,
"mean_token_accuracy": 0.10288520082831383,
"num_tokens": 2706232.0,
"step": 1230
},
{
"entropy": 6.386320161819458,
"epoch": 0.12256847955537911,
"grad_norm": 1.046875,
"learning_rate": 0.0004999938897889921,
"loss": 6.3744,
"mean_token_accuracy": 0.11175505593419074,
"num_tokens": 2715874.0,
"step": 1235
},
{
"entropy": 6.317949056625366,
"epoch": 0.12306470821754664,
"grad_norm": 0.8125,
"learning_rate": 0.0004999936258803647,
"loss": 6.3121,
"mean_token_accuracy": 0.11142380386590958,
"num_tokens": 2727409.0,
"step": 1240
},
{
"entropy": 6.273382472991943,
"epoch": 0.12356093687971417,
"grad_norm": 1.046875,
"learning_rate": 0.0004999933563923802,
"loss": 6.2154,
"mean_token_accuracy": 0.1114815816283226,
"num_tokens": 2737259.0,
"step": 1245
},
{
"entropy": 6.341649723052979,
"epoch": 0.1240571655418817,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999930813250451,
"loss": 6.3557,
"mean_token_accuracy": 0.10958174094557763,
"num_tokens": 2748416.0,
"step": 1250
},
{
"entropy": 6.3774512767791744,
"epoch": 0.12455339420404922,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999928006783667,
"loss": 6.3204,
"mean_token_accuracy": 0.11053804978728295,
"num_tokens": 2760470.0,
"step": 1255
},
{
"entropy": 6.337265825271606,
"epoch": 0.12504962286621676,
"grad_norm": 1.015625,
"learning_rate": 0.0004999925144523515,
"loss": 6.3438,
"mean_token_accuracy": 0.104658143222332,
"num_tokens": 2770222.0,
"step": 1260
},
{
"entropy": 6.334999036788941,
"epoch": 0.12554585152838427,
"grad_norm": 1.015625,
"learning_rate": 0.0004999922226470066,
"loss": 6.2795,
"mean_token_accuracy": 0.10962152928113937,
"num_tokens": 2780212.0,
"step": 1265
},
{
"entropy": 6.251826810836792,
"epoch": 0.1260420801905518,
"grad_norm": 1.03125,
"learning_rate": 0.0004999919252623397,
"loss": 6.2525,
"mean_token_accuracy": 0.11011324599385261,
"num_tokens": 2790239.0,
"step": 1270
},
{
"entropy": 6.402917814254761,
"epoch": 0.12653830885271933,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999916222983577,
"loss": 6.2774,
"mean_token_accuracy": 0.1086140252649784,
"num_tokens": 2801902.0,
"step": 1275
},
{
"entropy": 6.204992580413818,
"epoch": 0.12703453751488686,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999913137550682,
"loss": 6.2721,
"mean_token_accuracy": 0.10906686112284661,
"num_tokens": 2813268.0,
"step": 1280
},
{
"entropy": 6.3671046733856205,
"epoch": 0.1275307661770544,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999909996324791,
"loss": 6.3598,
"mean_token_accuracy": 0.10677680447697639,
"num_tokens": 2824506.0,
"step": 1285
},
{
"entropy": 6.3029297351837155,
"epoch": 0.12802699483922192,
"grad_norm": 0.8984375,
"learning_rate": 0.000499990679930598,
"loss": 6.2556,
"mean_token_accuracy": 0.11228155717253685,
"num_tokens": 2835144.0,
"step": 1290
},
{
"entropy": 6.21924409866333,
"epoch": 0.12852322350138945,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999903546494327,
"loss": 6.2655,
"mean_token_accuracy": 0.10710735470056534,
"num_tokens": 2845382.0,
"step": 1295
},
{
"entropy": 6.328607368469238,
"epoch": 0.12901945216355698,
"grad_norm": 1.09375,
"learning_rate": 0.0004999900237889916,
"loss": 6.2964,
"mean_token_accuracy": 0.1106351412832737,
"num_tokens": 2854697.0,
"step": 1300
},
{
"entropy": 6.344019556045533,
"epoch": 0.12951568082572448,
"grad_norm": 0.984375,
"learning_rate": 0.0004999896873492827,
"loss": 6.2941,
"mean_token_accuracy": 0.10326201841235161,
"num_tokens": 2866216.0,
"step": 1305
},
{
"entropy": 6.258399963378906,
"epoch": 0.130011909487892,
"grad_norm": 0.921875,
"learning_rate": 0.0004999893453303143,
"loss": 6.2558,
"mean_token_accuracy": 0.11178297996520996,
"num_tokens": 2877734.0,
"step": 1310
},
{
"entropy": 6.344228315353393,
"epoch": 0.13050813815005954,
"grad_norm": 0.953125,
"learning_rate": 0.0004999889977320949,
"loss": 6.3055,
"mean_token_accuracy": 0.11218280494213104,
"num_tokens": 2889445.0,
"step": 1315
},
{
"entropy": 6.236292695999145,
"epoch": 0.13100436681222707,
"grad_norm": 0.82421875,
"learning_rate": 0.0004999886445546332,
"loss": 6.2455,
"mean_token_accuracy": 0.11572465375065803,
"num_tokens": 2900485.0,
"step": 1320
},
{
"entropy": 6.3338706493377686,
"epoch": 0.1315005954743946,
"grad_norm": 0.88671875,
"learning_rate": 0.000499988285797938,
"loss": 6.2335,
"mean_token_accuracy": 0.11599084883928298,
"num_tokens": 2909795.0,
"step": 1325
},
{
"entropy": 6.189287090301514,
"epoch": 0.13199682413656214,
"grad_norm": 0.92578125,
"learning_rate": 0.000499987921462018,
"loss": 6.2602,
"mean_token_accuracy": 0.11903260201215744,
"num_tokens": 2920153.0,
"step": 1330
},
{
"entropy": 6.2777505874633786,
"epoch": 0.13249305279872967,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999875515468825,
"loss": 6.2512,
"mean_token_accuracy": 0.1119612142443657,
"num_tokens": 2930457.0,
"step": 1335
},
{
"entropy": 6.312501382827759,
"epoch": 0.13298928146089717,
"grad_norm": 0.953125,
"learning_rate": 0.0004999871760525403,
"loss": 6.3165,
"mean_token_accuracy": 0.10896633416414261,
"num_tokens": 2941305.0,
"step": 1340
},
{
"entropy": 6.297947740554809,
"epoch": 0.1334855101230647,
"grad_norm": 1.03125,
"learning_rate": 0.0004999867949790011,
"loss": 6.2476,
"mean_token_accuracy": 0.11246784254908562,
"num_tokens": 2951798.0,
"step": 1345
},
{
"entropy": 6.243113613128662,
"epoch": 0.13398173878523223,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999864083262743,
"loss": 6.162,
"mean_token_accuracy": 0.11877489164471626,
"num_tokens": 2962486.0,
"step": 1350
},
{
"entropy": 6.228711986541748,
"epoch": 0.13447796744739976,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999860160943691,
"loss": 6.2658,
"mean_token_accuracy": 0.11053448170423508,
"num_tokens": 2974317.0,
"step": 1355
},
{
"entropy": 6.215711355209351,
"epoch": 0.1349741961095673,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999856182832958,
"loss": 6.1817,
"mean_token_accuracy": 0.11494843661785126,
"num_tokens": 2985440.0,
"step": 1360
},
{
"entropy": 6.291191291809082,
"epoch": 0.13547042477173482,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999852148930637,
"loss": 6.2546,
"mean_token_accuracy": 0.11355492547154426,
"num_tokens": 2997044.0,
"step": 1365
},
{
"entropy": 6.23797197341919,
"epoch": 0.13596665343390235,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999848059236832,
"loss": 6.262,
"mean_token_accuracy": 0.1076361782848835,
"num_tokens": 3005965.0,
"step": 1370
},
{
"entropy": 6.239419984817505,
"epoch": 0.13646288209606988,
"grad_norm": 0.890625,
"learning_rate": 0.0004999843913751643,
"loss": 6.1175,
"mean_token_accuracy": 0.11854560673236847,
"num_tokens": 3015031.0,
"step": 1375
},
{
"entropy": 6.195667743682861,
"epoch": 0.13695911075823738,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999839712475174,
"loss": 6.1909,
"mean_token_accuracy": 0.11440791487693787,
"num_tokens": 3026275.0,
"step": 1380
},
{
"entropy": 6.2530198097229,
"epoch": 0.13745533942040492,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999835455407526,
"loss": 6.2354,
"mean_token_accuracy": 0.1116865485906601,
"num_tokens": 3036778.0,
"step": 1385
},
{
"entropy": 6.314652347564698,
"epoch": 0.13795156808257245,
"grad_norm": 0.99609375,
"learning_rate": 0.0004999831142548808,
"loss": 6.3486,
"mean_token_accuracy": 0.11005391925573349,
"num_tokens": 3048149.0,
"step": 1390
},
{
"entropy": 6.2538410186767575,
"epoch": 0.13844779674473998,
"grad_norm": 1.0390625,
"learning_rate": 0.0004999826773899124,
"loss": 6.214,
"mean_token_accuracy": 0.11115290969610214,
"num_tokens": 3059149.0,
"step": 1395
},
{
"entropy": 6.2944536209106445,
"epoch": 0.1389440254069075,
"grad_norm": 0.7890625,
"learning_rate": 0.0004999822349458586,
"loss": 6.226,
"mean_token_accuracy": 0.11450750976800919,
"num_tokens": 3071613.0,
"step": 1400
},
{
"entropy": 6.322410726547242,
"epoch": 0.13944025406907504,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999817869227301,
"loss": 6.3583,
"mean_token_accuracy": 0.10745183452963829,
"num_tokens": 3084742.0,
"step": 1405
},
{
"entropy": 6.289934778213501,
"epoch": 0.13993648273124257,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999813333205381,
"loss": 6.2773,
"mean_token_accuracy": 0.11367975026369095,
"num_tokens": 3095356.0,
"step": 1410
},
{
"entropy": 6.247778844833374,
"epoch": 0.14043271139341007,
"grad_norm": 0.92578125,
"learning_rate": 0.0004999808741392937,
"loss": 6.2522,
"mean_token_accuracy": 0.10875962376594543,
"num_tokens": 3107717.0,
"step": 1415
},
{
"entropy": 6.226774549484253,
"epoch": 0.1409289400555776,
"grad_norm": 0.85546875,
"learning_rate": 0.0004999804093790086,
"loss": 6.251,
"mean_token_accuracy": 0.10959596037864686,
"num_tokens": 3119205.0,
"step": 1420
},
{
"entropy": 6.309685230255127,
"epoch": 0.14142516871774513,
"grad_norm": 0.9296875,
"learning_rate": 0.000499979939039694,
"loss": 6.2812,
"mean_token_accuracy": 0.11436648070812225,
"num_tokens": 3130564.0,
"step": 1425
},
{
"entropy": 6.275003099441529,
"epoch": 0.14192139737991266,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999794631213618,
"loss": 6.22,
"mean_token_accuracy": 0.11114049553871155,
"num_tokens": 3141883.0,
"step": 1430
},
{
"entropy": 6.281518220901489,
"epoch": 0.1424176260420802,
"grad_norm": 0.90625,
"learning_rate": 0.0004999789816240236,
"loss": 6.2302,
"mean_token_accuracy": 0.11397273391485215,
"num_tokens": 3152817.0,
"step": 1435
},
{
"entropy": 6.17591028213501,
"epoch": 0.14291385470424772,
"grad_norm": 1.03125,
"learning_rate": 0.0004999784945476915,
"loss": 6.1486,
"mean_token_accuracy": 0.11776702702045441,
"num_tokens": 3163546.0,
"step": 1440
},
{
"entropy": 6.216448545455933,
"epoch": 0.14341008336641525,
"grad_norm": 1.0,
"learning_rate": 0.0004999780018923775,
"loss": 6.1949,
"mean_token_accuracy": 0.10940220728516578,
"num_tokens": 3173899.0,
"step": 1445
},
{
"entropy": 6.15954270362854,
"epoch": 0.14390631202858278,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999775036580939,
"loss": 6.0969,
"mean_token_accuracy": 0.11880587413907051,
"num_tokens": 3184433.0,
"step": 1450
},
{
"entropy": 6.242514944076538,
"epoch": 0.1444025406907503,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999769998448529,
"loss": 6.2142,
"mean_token_accuracy": 0.10818950086832047,
"num_tokens": 3196077.0,
"step": 1455
},
{
"entropy": 6.138770914077758,
"epoch": 0.14489876935291782,
"grad_norm": 1.015625,
"learning_rate": 0.0004999764904526672,
"loss": 6.1403,
"mean_token_accuracy": 0.11515245363116264,
"num_tokens": 3206645.0,
"step": 1460
},
{
"entropy": 6.3275079250335695,
"epoch": 0.14539499801508535,
"grad_norm": 0.93359375,
"learning_rate": 0.0004999759754815492,
"loss": 6.2513,
"mean_token_accuracy": 0.11938706040382385,
"num_tokens": 3216786.0,
"step": 1465
},
{
"entropy": 6.199605226516724,
"epoch": 0.14589122667725288,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999754549315118,
"loss": 6.1823,
"mean_token_accuracy": 0.11293254643678666,
"num_tokens": 3227384.0,
"step": 1470
},
{
"entropy": 6.108070659637451,
"epoch": 0.1463874553394204,
"grad_norm": 0.828125,
"learning_rate": 0.000499974928802568,
"loss": 6.0334,
"mean_token_accuracy": 0.11813693642616271,
"num_tokens": 3238800.0,
"step": 1475
},
{
"entropy": 6.180989933013916,
"epoch": 0.14688368400158794,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999743970947306,
"loss": 6.2034,
"mean_token_accuracy": 0.11198758184909821,
"num_tokens": 3248627.0,
"step": 1480
},
{
"entropy": 6.197469997406006,
"epoch": 0.14737991266375547,
"grad_norm": 0.8984375,
"learning_rate": 0.000499973859808013,
"loss": 6.2012,
"mean_token_accuracy": 0.11548065319657326,
"num_tokens": 3258989.0,
"step": 1485
},
{
"entropy": 6.222706890106201,
"epoch": 0.14787614132592297,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999733169424284,
"loss": 6.1897,
"mean_token_accuracy": 0.11579868942499161,
"num_tokens": 3269474.0,
"step": 1490
},
{
"entropy": 6.1594078540802,
"epoch": 0.1483723699880905,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999727684979904,
"loss": 6.1833,
"mean_token_accuracy": 0.11907703801989555,
"num_tokens": 3280453.0,
"step": 1495
},
{
"entropy": 6.195897006988526,
"epoch": 0.14886859865025803,
"grad_norm": 0.8828125,
"learning_rate": 0.0004999722144747125,
"loss": 6.1237,
"mean_token_accuracy": 0.11352067068219185,
"num_tokens": 3290234.0,
"step": 1500
},
{
"entropy": 6.077924489974976,
"epoch": 0.14936482731242556,
"grad_norm": 1.0,
"learning_rate": 0.0004999716548726085,
"loss": 6.1053,
"mean_token_accuracy": 0.1176950380206108,
"num_tokens": 3300943.0,
"step": 1505
},
{
"entropy": 6.199636650085449,
"epoch": 0.1498610559745931,
"grad_norm": 1.03125,
"learning_rate": 0.0004999710896916921,
"loss": 6.0607,
"mean_token_accuracy": 0.11784961298108101,
"num_tokens": 3311296.0,
"step": 1510
},
{
"entropy": 6.159174585342408,
"epoch": 0.15035728463676062,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999705189319774,
"loss": 6.2248,
"mean_token_accuracy": 0.10928900018334389,
"num_tokens": 3322310.0,
"step": 1515
},
{
"entropy": 6.187262296676636,
"epoch": 0.15085351329892815,
"grad_norm": 1.0,
"learning_rate": 0.0004999699425934788,
"loss": 6.2243,
"mean_token_accuracy": 0.11323094964027405,
"num_tokens": 3334176.0,
"step": 1520
},
{
"entropy": 6.178784799575806,
"epoch": 0.15134974196109569,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999693606762103,
"loss": 6.2073,
"mean_token_accuracy": 0.10905725955963134,
"num_tokens": 3344839.0,
"step": 1525
},
{
"entropy": 6.270957374572754,
"epoch": 0.1518459706232632,
"grad_norm": 0.95703125,
"learning_rate": 0.0004999687731801865,
"loss": 6.1825,
"mean_token_accuracy": 0.11094732880592346,
"num_tokens": 3357046.0,
"step": 1530
},
{
"entropy": 6.150786876678467,
"epoch": 0.15234219928543072,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999681801054217,
"loss": 6.2501,
"mean_token_accuracy": 0.11335301697254181,
"num_tokens": 3367783.0,
"step": 1535
},
{
"entropy": 6.206290578842163,
"epoch": 0.15283842794759825,
"grad_norm": 0.9609375,
"learning_rate": 0.000499967581451931,
"loss": 6.1134,
"mean_token_accuracy": 0.11155367642641068,
"num_tokens": 3378529.0,
"step": 1540
},
{
"entropy": 6.179344654083252,
"epoch": 0.15333465660976578,
"grad_norm": 1.03125,
"learning_rate": 0.000499966977219729,
"loss": 6.3094,
"mean_token_accuracy": 0.11146700233221055,
"num_tokens": 3389318.0,
"step": 1545
},
{
"entropy": 6.294518518447876,
"epoch": 0.1538308852719333,
"grad_norm": 0.9453125,
"learning_rate": 0.0004999663674088306,
"loss": 6.313,
"mean_token_accuracy": 0.10722307339310647,
"num_tokens": 3400603.0,
"step": 1550
},
{
"entropy": 6.270757818222046,
"epoch": 0.15432711393410084,
"grad_norm": 1.078125,
"learning_rate": 0.0004999657520192511,
"loss": 6.1938,
"mean_token_accuracy": 0.11140423566102982,
"num_tokens": 3410316.0,
"step": 1555
},
{
"entropy": 6.242702484130859,
"epoch": 0.15482334259626837,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999651310510058,
"loss": 6.1446,
"mean_token_accuracy": 0.11679755821824074,
"num_tokens": 3422155.0,
"step": 1560
},
{
"entropy": 6.17017388343811,
"epoch": 0.15531957125843587,
"grad_norm": 0.921875,
"learning_rate": 0.0004999645045041099,
"loss": 6.0721,
"mean_token_accuracy": 0.11942542716860771,
"num_tokens": 3432892.0,
"step": 1565
},
{
"entropy": 6.174464702606201,
"epoch": 0.1558157999206034,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999638723785791,
"loss": 6.2059,
"mean_token_accuracy": 0.11832942962646484,
"num_tokens": 3443967.0,
"step": 1570
},
{
"entropy": 6.144176197052002,
"epoch": 0.15631202858277093,
"grad_norm": 0.90234375,
"learning_rate": 0.000499963234674429,
"loss": 6.0957,
"mean_token_accuracy": 0.12020640000700951,
"num_tokens": 3454354.0,
"step": 1575
},
{
"entropy": 6.038722705841065,
"epoch": 0.15680825724493846,
"grad_norm": 0.91796875,
"learning_rate": 0.0004999625913916754,
"loss": 6.1151,
"mean_token_accuracy": 0.11729008257389069,
"num_tokens": 3465232.0,
"step": 1580
},
{
"entropy": 6.125705909729004,
"epoch": 0.157304485907106,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999619425303343,
"loss": 6.0879,
"mean_token_accuracy": 0.12173460498452186,
"num_tokens": 3475443.0,
"step": 1585
},
{
"entropy": 6.21633563041687,
"epoch": 0.15780071456927353,
"grad_norm": 0.875,
"learning_rate": 0.0004999612880904219,
"loss": 6.1283,
"mean_token_accuracy": 0.11958795338869095,
"num_tokens": 3486643.0,
"step": 1590
},
{
"entropy": 6.145956325531006,
"epoch": 0.15829694323144106,
"grad_norm": 0.9375,
"learning_rate": 0.0004999606280719542,
"loss": 6.173,
"mean_token_accuracy": 0.11473753675818443,
"num_tokens": 3497557.0,
"step": 1595
},
{
"entropy": 6.26501202583313,
"epoch": 0.1587931718936086,
"grad_norm": 1.09375,
"learning_rate": 0.0004999599624749475,
"loss": 6.1633,
"mean_token_accuracy": 0.1159546658396721,
"num_tokens": 3509790.0,
"step": 1600
},
{
"entropy": 6.065554237365722,
"epoch": 0.1592894005557761,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999592912994187,
"loss": 6.0811,
"mean_token_accuracy": 0.11343559995293617,
"num_tokens": 3521577.0,
"step": 1605
},
{
"entropy": 6.207467794418335,
"epoch": 0.15978562921794362,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999586145453842,
"loss": 6.1295,
"mean_token_accuracy": 0.10831035524606705,
"num_tokens": 3532126.0,
"step": 1610
},
{
"entropy": 5.990396785736084,
"epoch": 0.16028185788011115,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999579322128608,
"loss": 5.9618,
"mean_token_accuracy": 0.11795824691653252,
"num_tokens": 3542560.0,
"step": 1615
},
{
"entropy": 6.117848062515259,
"epoch": 0.16077808654227868,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999572443018654,
"loss": 6.137,
"mean_token_accuracy": 0.11492020189762116,
"num_tokens": 3553863.0,
"step": 1620
},
{
"entropy": 6.154970407485962,
"epoch": 0.1612743152044462,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999565508124151,
"loss": 6.1508,
"mean_token_accuracy": 0.11818645223975181,
"num_tokens": 3564953.0,
"step": 1625
},
{
"entropy": 6.148259401321411,
"epoch": 0.16177054386661374,
"grad_norm": 0.84375,
"learning_rate": 0.000499955851744527,
"loss": 6.1469,
"mean_token_accuracy": 0.11931650564074517,
"num_tokens": 3576247.0,
"step": 1630
},
{
"entropy": 6.276931715011597,
"epoch": 0.16226677252878127,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999551470982185,
"loss": 6.1795,
"mean_token_accuracy": 0.11549673676490783,
"num_tokens": 3587600.0,
"step": 1635
},
{
"entropy": 6.206128883361816,
"epoch": 0.1627630011909488,
"grad_norm": 1.015625,
"learning_rate": 0.0004999544368735073,
"loss": 6.1665,
"mean_token_accuracy": 0.11787421181797982,
"num_tokens": 3597727.0,
"step": 1640
},
{
"entropy": 6.091890716552735,
"epoch": 0.1632592298531163,
"grad_norm": 0.9765625,
"learning_rate": 0.0004999537210704106,
"loss": 6.1341,
"mean_token_accuracy": 0.1190246470272541,
"num_tokens": 3608066.0,
"step": 1645
},
{
"entropy": 6.20475926399231,
"epoch": 0.16375545851528384,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999529996889463,
"loss": 6.1987,
"mean_token_accuracy": 0.11030768677592277,
"num_tokens": 3618263.0,
"step": 1650
},
{
"entropy": 6.184187507629394,
"epoch": 0.16425168717745137,
"grad_norm": 1.0,
"learning_rate": 0.0004999522727291325,
"loss": 6.1268,
"mean_token_accuracy": 0.11970260366797447,
"num_tokens": 3629465.0,
"step": 1655
},
{
"entropy": 6.179236936569214,
"epoch": 0.1647479158396189,
"grad_norm": 0.91015625,
"learning_rate": 0.000499951540190987,
"loss": 6.154,
"mean_token_accuracy": 0.11628529205918312,
"num_tokens": 3639657.0,
"step": 1660
},
{
"entropy": 6.174856472015381,
"epoch": 0.16524414450178643,
"grad_norm": 0.8984375,
"learning_rate": 0.0004999508020745281,
"loss": 6.1086,
"mean_token_accuracy": 0.1154894195497036,
"num_tokens": 3652374.0,
"step": 1665
},
{
"entropy": 6.104015922546386,
"epoch": 0.16574037316395396,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999500583797739,
"loss": 6.0969,
"mean_token_accuracy": 0.11629031077027321,
"num_tokens": 3664083.0,
"step": 1670
},
{
"entropy": 6.143244981765747,
"epoch": 0.1662366018261215,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999493091067431,
"loss": 6.1398,
"mean_token_accuracy": 0.1213216483592987,
"num_tokens": 3675314.0,
"step": 1675
},
{
"entropy": 6.131026697158814,
"epoch": 0.166732830488289,
"grad_norm": 0.8359375,
"learning_rate": 0.0004999485542554541,
"loss": 6.0897,
"mean_token_accuracy": 0.10728424638509751,
"num_tokens": 3686019.0,
"step": 1680
},
{
"entropy": 6.1852058410644535,
"epoch": 0.16722905915045652,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999477938259256,
"loss": 6.196,
"mean_token_accuracy": 0.11261390075087548,
"num_tokens": 3698170.0,
"step": 1685
},
{
"entropy": 6.133717155456543,
"epoch": 0.16772528781262405,
"grad_norm": 0.8203125,
"learning_rate": 0.0004999470278181766,
"loss": 6.186,
"mean_token_accuracy": 0.11304959654808044,
"num_tokens": 3709243.0,
"step": 1690
},
{
"entropy": 6.20488805770874,
"epoch": 0.16822151647479158,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999462562322261,
"loss": 6.1447,
"mean_token_accuracy": 0.11698395982384682,
"num_tokens": 3720196.0,
"step": 1695
},
{
"entropy": 6.172057008743286,
"epoch": 0.1687177451369591,
"grad_norm": 0.92578125,
"learning_rate": 0.000499945479068093,
"loss": 6.1913,
"mean_token_accuracy": 0.11760392040014267,
"num_tokens": 3732714.0,
"step": 1700
},
{
"entropy": 6.189049911499024,
"epoch": 0.16921397379912664,
"grad_norm": 0.765625,
"learning_rate": 0.0004999446963257969,
"loss": 6.1714,
"mean_token_accuracy": 0.11059175580739974,
"num_tokens": 3746474.0,
"step": 1705
},
{
"entropy": 6.084878253936767,
"epoch": 0.16971020246129417,
"grad_norm": 0.9140625,
"learning_rate": 0.0004999439080053571,
"loss": 6.0414,
"mean_token_accuracy": 0.12301717177033425,
"num_tokens": 3757528.0,
"step": 1710
},
{
"entropy": 6.24871335029602,
"epoch": 0.1702064311234617,
"grad_norm": 0.8671875,
"learning_rate": 0.0004999431141067928,
"loss": 6.1689,
"mean_token_accuracy": 0.11376198828220367,
"num_tokens": 3768360.0,
"step": 1715
},
{
"entropy": 6.100267028808593,
"epoch": 0.1707026597856292,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999423146301242,
"loss": 6.0366,
"mean_token_accuracy": 0.11526194885373116,
"num_tokens": 3779809.0,
"step": 1720
},
{
"entropy": 6.038203620910645,
"epoch": 0.17119888844779674,
"grad_norm": 0.90234375,
"learning_rate": 0.0004999415095753709,
"loss": 6.03,
"mean_token_accuracy": 0.11550426036119461,
"num_tokens": 3791455.0,
"step": 1725
},
{
"entropy": 6.056922388076782,
"epoch": 0.17169511710996427,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999406989425527,
"loss": 6.0831,
"mean_token_accuracy": 0.11582357957959175,
"num_tokens": 3803209.0,
"step": 1730
},
{
"entropy": 6.087919235229492,
"epoch": 0.1721913457721318,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999398827316901,
"loss": 6.0977,
"mean_token_accuracy": 0.11691041514277459,
"num_tokens": 3813553.0,
"step": 1735
},
{
"entropy": 6.157799959182739,
"epoch": 0.17268757443429933,
"grad_norm": 0.91796875,
"learning_rate": 0.000499939060942803,
"loss": 6.1439,
"mean_token_accuracy": 0.10938044339418411,
"num_tokens": 3824856.0,
"step": 1740
},
{
"entropy": 6.220856618881226,
"epoch": 0.17318380309646686,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999382335759119,
"loss": 6.2309,
"mean_token_accuracy": 0.11000438928604125,
"num_tokens": 3835489.0,
"step": 1745
},
{
"entropy": 6.134221601486206,
"epoch": 0.1736800317586344,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999374006310372,
"loss": 6.0638,
"mean_token_accuracy": 0.11413465365767479,
"num_tokens": 3844838.0,
"step": 1750
},
{
"entropy": 6.067860126495361,
"epoch": 0.1741762604208019,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999365621081999,
"loss": 6.0341,
"mean_token_accuracy": 0.11537155881524086,
"num_tokens": 3855288.0,
"step": 1755
},
{
"entropy": 6.108763790130615,
"epoch": 0.17467248908296942,
"grad_norm": 1.0546875,
"learning_rate": 0.0004999357180074204,
"loss": 6.0853,
"mean_token_accuracy": 0.12003114819526672,
"num_tokens": 3865585.0,
"step": 1760
},
{
"entropy": 6.129130601882935,
"epoch": 0.17516871774513695,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999348683287199,
"loss": 6.0621,
"mean_token_accuracy": 0.11469487771391869,
"num_tokens": 3877694.0,
"step": 1765
},
{
"entropy": 6.094071865081787,
"epoch": 0.17566494640730448,
"grad_norm": 0.84765625,
"learning_rate": 0.0004999340130721192,
"loss": 6.0386,
"mean_token_accuracy": 0.12670635432004929,
"num_tokens": 3888802.0,
"step": 1770
},
{
"entropy": 6.04920973777771,
"epoch": 0.17616117506947201,
"grad_norm": 0.90625,
"learning_rate": 0.0004999331522376396,
"loss": 6.0047,
"mean_token_accuracy": 0.11465661451220513,
"num_tokens": 3900294.0,
"step": 1775
},
{
"entropy": 6.030979776382447,
"epoch": 0.17665740373163955,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999322858253027,
"loss": 6.0339,
"mean_token_accuracy": 0.12006412371993065,
"num_tokens": 3910865.0,
"step": 1780
},
{
"entropy": 6.070213460922242,
"epoch": 0.17715363239380708,
"grad_norm": 0.9296875,
"learning_rate": 0.0004999314138351297,
"loss": 6.1034,
"mean_token_accuracy": 0.1195277787744999,
"num_tokens": 3921960.0,
"step": 1785
},
{
"entropy": 6.161261892318725,
"epoch": 0.1776498610559746,
"grad_norm": 0.98828125,
"learning_rate": 0.0004999305362671424,
"loss": 6.1101,
"mean_token_accuracy": 0.11471805498003959,
"num_tokens": 3934085.0,
"step": 1790
},
{
"entropy": 6.052437019348145,
"epoch": 0.1781460897181421,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999296531213624,
"loss": 5.9506,
"mean_token_accuracy": 0.1281500868499279,
"num_tokens": 3944532.0,
"step": 1795
},
{
"entropy": 6.051175928115844,
"epoch": 0.17864231838030964,
"grad_norm": 0.98046875,
"learning_rate": 0.0004999287643978117,
"loss": 5.9968,
"mean_token_accuracy": 0.12282953560352325,
"num_tokens": 3954995.0,
"step": 1800
},
{
"entropy": 5.990139293670654,
"epoch": 0.17913854704247717,
"grad_norm": 0.8515625,
"learning_rate": 0.0004999278700965124,
"loss": 6.0559,
"mean_token_accuracy": 0.12489112317562104,
"num_tokens": 3966621.0,
"step": 1805
},
{
"entropy": 6.166021299362183,
"epoch": 0.1796347757046447,
"grad_norm": 0.87890625,
"learning_rate": 0.0004999269702174866,
"loss": 6.119,
"mean_token_accuracy": 0.11739438995718957,
"num_tokens": 3977568.0,
"step": 1810
},
{
"entropy": 6.073067665100098,
"epoch": 0.18013100436681223,
"grad_norm": 1.0234375,
"learning_rate": 0.0004999260647607565,
"loss": 6.1373,
"mean_token_accuracy": 0.11110868230462075,
"num_tokens": 3988571.0,
"step": 1815
},
{
"entropy": 6.138613891601563,
"epoch": 0.18062723302897976,
"grad_norm": 1.046875,
"learning_rate": 0.0004999251537263448,
"loss": 5.9477,
"mean_token_accuracy": 0.1258134886622429,
"num_tokens": 3998380.0,
"step": 1820
},
{
"entropy": 6.0303796291351315,
"epoch": 0.1811234616911473,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999242371142739,
"loss": 6.1287,
"mean_token_accuracy": 0.12175759598612786,
"num_tokens": 4010017.0,
"step": 1825
},
{
"entropy": 6.189897441864014,
"epoch": 0.1816196903533148,
"grad_norm": 0.859375,
"learning_rate": 0.0004999233149245665,
"loss": 6.0939,
"mean_token_accuracy": 0.1167963296175003,
"num_tokens": 4020922.0,
"step": 1830
},
{
"entropy": 6.123414802551269,
"epoch": 0.18211591901548232,
"grad_norm": 0.91015625,
"learning_rate": 0.0004999223871572456,
"loss": 6.0808,
"mean_token_accuracy": 0.11856624335050583,
"num_tokens": 4030927.0,
"step": 1835
},
{
"entropy": 6.089998006820679,
"epoch": 0.18261214767764986,
"grad_norm": 0.9609375,
"learning_rate": 0.0004999214538123342,
"loss": 6.1432,
"mean_token_accuracy": 0.11269862428307534,
"num_tokens": 4041375.0,
"step": 1840
},
{
"entropy": 6.0862884521484375,
"epoch": 0.18310837633981739,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999205148898554,
"loss": 6.1411,
"mean_token_accuracy": 0.11809576898813248,
"num_tokens": 4053213.0,
"step": 1845
},
{
"entropy": 6.142365455627441,
"epoch": 0.18360460500198492,
"grad_norm": 1.046875,
"learning_rate": 0.0004999195703898325,
"loss": 5.9824,
"mean_token_accuracy": 0.12277794778347015,
"num_tokens": 4064536.0,
"step": 1850
},
{
"entropy": 6.081491136550904,
"epoch": 0.18410083366415245,
"grad_norm": 1.015625,
"learning_rate": 0.0004999186203122889,
"loss": 6.0625,
"mean_token_accuracy": 0.12416149452328681,
"num_tokens": 4075282.0,
"step": 1855
},
{
"entropy": 6.013858938217163,
"epoch": 0.18459706232631998,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999176646572482,
"loss": 6.0563,
"mean_token_accuracy": 0.1242367833852768,
"num_tokens": 4085501.0,
"step": 1860
},
{
"entropy": 6.091145706176758,
"epoch": 0.1850932909884875,
"grad_norm": 0.921875,
"learning_rate": 0.000499916703424734,
"loss": 6.0064,
"mean_token_accuracy": 0.11573861911892891,
"num_tokens": 4095697.0,
"step": 1865
},
{
"entropy": 6.072324752807617,
"epoch": 0.185589519650655,
"grad_norm": 0.86328125,
"learning_rate": 0.0004999157366147702,
"loss": 6.0599,
"mean_token_accuracy": 0.1197610542178154,
"num_tokens": 4106459.0,
"step": 1870
},
{
"entropy": 6.090449953079224,
"epoch": 0.18608574831282254,
"grad_norm": 0.87109375,
"learning_rate": 0.0004999147642273809,
"loss": 5.998,
"mean_token_accuracy": 0.11759251281619072,
"num_tokens": 4118320.0,
"step": 1875
},
{
"entropy": 5.991074848175049,
"epoch": 0.18658197697499007,
"grad_norm": 0.92578125,
"learning_rate": 0.00049991378626259,
"loss": 6.0245,
"mean_token_accuracy": 0.12806328162550926,
"num_tokens": 4131249.0,
"step": 1880
},
{
"entropy": 6.093191146850586,
"epoch": 0.1870782056371576,
"grad_norm": 1.0078125,
"learning_rate": 0.0004999128027204219,
"loss": 6.096,
"mean_token_accuracy": 0.11801933944225311,
"num_tokens": 4141480.0,
"step": 1885
},
{
"entropy": 6.126098394393921,
"epoch": 0.18757443429932513,
"grad_norm": 0.96484375,
"learning_rate": 0.000499911813600901,
"loss": 6.1558,
"mean_token_accuracy": 0.11598219275474549,
"num_tokens": 4152344.0,
"step": 1890
},
{
"entropy": 6.173122406005859,
"epoch": 0.18807066296149266,
"grad_norm": 0.96484375,
"learning_rate": 0.0004999108189040516,
"loss": 6.1105,
"mean_token_accuracy": 0.1155496172606945,
"num_tokens": 4164772.0,
"step": 1895
},
{
"entropy": 6.0153892040252686,
"epoch": 0.1885668916236602,
"grad_norm": 0.97265625,
"learning_rate": 0.0004999098186298986,
"loss": 6.0434,
"mean_token_accuracy": 0.11878785714507104,
"num_tokens": 4175511.0,
"step": 1900
},
{
"entropy": 6.039416551589966,
"epoch": 0.1890631202858277,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999088127784667,
"loss": 6.0148,
"mean_token_accuracy": 0.11949604898691177,
"num_tokens": 4187445.0,
"step": 1905
},
{
"entropy": 6.069828367233276,
"epoch": 0.18955934894799523,
"grad_norm": 0.8515625,
"learning_rate": 0.000499907801349781,
"loss": 6.0138,
"mean_token_accuracy": 0.11969439908862115,
"num_tokens": 4198124.0,
"step": 1910
},
{
"entropy": 6.048256969451904,
"epoch": 0.19005557761016276,
"grad_norm": 0.96875,
"learning_rate": 0.0004999067843438664,
"loss": 6.0505,
"mean_token_accuracy": 0.11893324404954911,
"num_tokens": 4208954.0,
"step": 1915
},
{
"entropy": 6.1492753505706785,
"epoch": 0.1905518062723303,
"grad_norm": 0.984375,
"learning_rate": 0.0004999057617607482,
"loss": 6.1068,
"mean_token_accuracy": 0.11971905007958412,
"num_tokens": 4218896.0,
"step": 1920
},
{
"entropy": 6.014862298965454,
"epoch": 0.19104803493449782,
"grad_norm": 0.94140625,
"learning_rate": 0.0004999047336004517,
"loss": 6.0095,
"mean_token_accuracy": 0.11468009054660797,
"num_tokens": 4228851.0,
"step": 1925
},
{
"entropy": 5.982360315322876,
"epoch": 0.19154426359666535,
"grad_norm": 0.9921875,
"learning_rate": 0.0004999036998630024,
"loss": 5.9659,
"mean_token_accuracy": 0.12004356682300568,
"num_tokens": 4240377.0,
"step": 1930
},
{
"entropy": 5.99826283454895,
"epoch": 0.19204049225883288,
"grad_norm": 0.92578125,
"learning_rate": 0.000499902660548426,
"loss": 5.9635,
"mean_token_accuracy": 0.1215371236205101,
"num_tokens": 4251341.0,
"step": 1935
},
{
"entropy": 6.100321102142334,
"epoch": 0.1925367209210004,
"grad_norm": 1.0,
"learning_rate": 0.0004999016156567482,
"loss": 6.0659,
"mean_token_accuracy": 0.12046532705426216,
"num_tokens": 4261206.0,
"step": 1940
},
{
"entropy": 6.080335712432861,
"epoch": 0.1930329495831679,
"grad_norm": 0.875,
"learning_rate": 0.000499900565187995,
"loss": 6.0723,
"mean_token_accuracy": 0.11344361081719398,
"num_tokens": 4272662.0,
"step": 1945
},
{
"entropy": 6.123919200897217,
"epoch": 0.19352917824533544,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998995091421924,
"loss": 6.0541,
"mean_token_accuracy": 0.11754847764968872,
"num_tokens": 4283302.0,
"step": 1950
},
{
"entropy": 5.979746294021607,
"epoch": 0.19402540690750297,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998984475193665,
"loss": 5.9978,
"mean_token_accuracy": 0.12176699414849282,
"num_tokens": 4296087.0,
"step": 1955
},
{
"entropy": 6.052849388122558,
"epoch": 0.1945216355696705,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998973803195439,
"loss": 6.0093,
"mean_token_accuracy": 0.12155119329690933,
"num_tokens": 4305918.0,
"step": 1960
},
{
"entropy": 6.0047338008880615,
"epoch": 0.19501786423183803,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998963075427508,
"loss": 6.0725,
"mean_token_accuracy": 0.11939249560236931,
"num_tokens": 4316507.0,
"step": 1965
},
{
"entropy": 6.079337644577026,
"epoch": 0.19551409289400556,
"grad_norm": 0.796875,
"learning_rate": 0.0004998952291890138,
"loss": 5.9977,
"mean_token_accuracy": 0.1253109961748123,
"num_tokens": 4329091.0,
"step": 1970
},
{
"entropy": 6.064958190917968,
"epoch": 0.1960103215561731,
"grad_norm": 1.03125,
"learning_rate": 0.0004998941452583599,
"loss": 6.0762,
"mean_token_accuracy": 0.12064750641584396,
"num_tokens": 4339266.0,
"step": 1975
},
{
"entropy": 5.988303995132446,
"epoch": 0.1965065502183406,
"grad_norm": 0.96875,
"learning_rate": 0.0004998930557508156,
"loss": 5.9826,
"mean_token_accuracy": 0.12658628076314926,
"num_tokens": 4349300.0,
"step": 1980
},
{
"entropy": 5.995124006271363,
"epoch": 0.19700277888050813,
"grad_norm": 0.93359375,
"learning_rate": 0.0004998919606664082,
"loss": 5.9143,
"mean_token_accuracy": 0.13212018981575965,
"num_tokens": 4360314.0,
"step": 1985
},
{
"entropy": 6.041628313064575,
"epoch": 0.19749900754267566,
"grad_norm": 0.953125,
"learning_rate": 0.0004998908600051648,
"loss": 5.9705,
"mean_token_accuracy": 0.12074373364448547,
"num_tokens": 4370077.0,
"step": 1990
},
{
"entropy": 5.971292448043823,
"epoch": 0.1979952362048432,
"grad_norm": 0.9453125,
"learning_rate": 0.0004998897537671126,
"loss": 5.9705,
"mean_token_accuracy": 0.12351988479495049,
"num_tokens": 4381000.0,
"step": 1995
},
{
"entropy": 6.094802951812744,
"epoch": 0.19849146486701072,
"grad_norm": 0.9921875,
"learning_rate": 0.0004998886419522792,
"loss": 6.1871,
"mean_token_accuracy": 0.11414133310317993,
"num_tokens": 4393485.0,
"step": 2000
},
{
"entropy": 6.160151147842408,
"epoch": 0.19898769352917825,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998875245606919,
"loss": 6.0252,
"mean_token_accuracy": 0.12122476175427437,
"num_tokens": 4404354.0,
"step": 2005
},
{
"entropy": 6.004584503173828,
"epoch": 0.19948392219134578,
"grad_norm": 0.984375,
"learning_rate": 0.0004998864015923788,
"loss": 6.1679,
"mean_token_accuracy": 0.12271130830049515,
"num_tokens": 4414818.0,
"step": 2010
},
{
"entropy": 6.125269937515259,
"epoch": 0.1999801508535133,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998852730473674,
"loss": 6.0353,
"mean_token_accuracy": 0.11513861268758774,
"num_tokens": 4426271.0,
"step": 2015
},
{
"entropy": 6.017019605636596,
"epoch": 0.2004763795156808,
"grad_norm": 1.0,
"learning_rate": 0.0004998841389256859,
"loss": 6.1004,
"mean_token_accuracy": 0.11922633796930313,
"num_tokens": 4436981.0,
"step": 2020
},
{
"entropy": 6.022520351409912,
"epoch": 0.20097260817784834,
"grad_norm": 1.0546875,
"learning_rate": 0.0004998829992273622,
"loss": 5.9114,
"mean_token_accuracy": 0.12870077788829803,
"num_tokens": 4446494.0,
"step": 2025
},
{
"entropy": 5.996196699142456,
"epoch": 0.20146883684001587,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998818539524248,
"loss": 6.0074,
"mean_token_accuracy": 0.1251174710690975,
"num_tokens": 4458535.0,
"step": 2030
},
{
"entropy": 6.087194585800171,
"epoch": 0.2019650655021834,
"grad_norm": 0.9609375,
"learning_rate": 0.000499880703100902,
"loss": 6.0014,
"mean_token_accuracy": 0.12391166165471076,
"num_tokens": 4469025.0,
"step": 2035
},
{
"entropy": 6.010956525802612,
"epoch": 0.20246129416435094,
"grad_norm": 0.984375,
"learning_rate": 0.0004998795466728223,
"loss": 5.986,
"mean_token_accuracy": 0.12361643686890603,
"num_tokens": 4478431.0,
"step": 2040
},
{
"entropy": 5.994862222671509,
"epoch": 0.20295752282651847,
"grad_norm": 1.046875,
"learning_rate": 0.0004998783846682144,
"loss": 6.1286,
"mean_token_accuracy": 0.11792842224240303,
"num_tokens": 4490215.0,
"step": 2045
},
{
"entropy": 6.091261959075927,
"epoch": 0.203453751488686,
"grad_norm": 0.80859375,
"learning_rate": 0.0004998772170871072,
"loss": 5.9872,
"mean_token_accuracy": 0.1232263281941414,
"num_tokens": 4502895.0,
"step": 2050
},
{
"entropy": 6.012570142745972,
"epoch": 0.20394998015085353,
"grad_norm": 0.87109375,
"learning_rate": 0.0004998760439295295,
"loss": 6.0185,
"mean_token_accuracy": 0.12015356048941613,
"num_tokens": 4514468.0,
"step": 2055
},
{
"entropy": 6.082999658584595,
"epoch": 0.20444620881302103,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998748651955105,
"loss": 6.0319,
"mean_token_accuracy": 0.12087535560131073,
"num_tokens": 4525787.0,
"step": 2060
},
{
"entropy": 6.01540241241455,
"epoch": 0.20494243747518856,
"grad_norm": 0.828125,
"learning_rate": 0.0004998736808850795,
"loss": 6.1173,
"mean_token_accuracy": 0.11445825323462486,
"num_tokens": 4537536.0,
"step": 2065
},
{
"entropy": 6.1401674270629885,
"epoch": 0.2054386661373561,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998724909982656,
"loss": 6.0429,
"mean_token_accuracy": 0.12223877683281899,
"num_tokens": 4549619.0,
"step": 2070
},
{
"entropy": 5.962413740158081,
"epoch": 0.20593489479952362,
"grad_norm": 0.875,
"learning_rate": 0.0004998712955350986,
"loss": 5.9808,
"mean_token_accuracy": 0.12422219887375832,
"num_tokens": 4560248.0,
"step": 2075
},
{
"entropy": 6.061628007888794,
"epoch": 0.20643112346169115,
"grad_norm": 0.8359375,
"learning_rate": 0.0004998700944956081,
"loss": 6.0998,
"mean_token_accuracy": 0.12046362236142158,
"num_tokens": 4571888.0,
"step": 2080
},
{
"entropy": 6.062523126602173,
"epoch": 0.20692735212385868,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998688878798237,
"loss": 5.9366,
"mean_token_accuracy": 0.13008833229541777,
"num_tokens": 4583234.0,
"step": 2085
},
{
"entropy": 6.068667030334472,
"epoch": 0.2074235807860262,
"grad_norm": 0.890625,
"learning_rate": 0.0004998676756877753,
"loss": 6.1538,
"mean_token_accuracy": 0.11486675590276718,
"num_tokens": 4595252.0,
"step": 2090
},
{
"entropy": 6.0270226955413815,
"epoch": 0.20791980944819372,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998664579194934,
"loss": 5.9196,
"mean_token_accuracy": 0.12394649386405945,
"num_tokens": 4607091.0,
"step": 2095
},
{
"entropy": 6.029794692993164,
"epoch": 0.20841603811036125,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998652345750077,
"loss": 5.8893,
"mean_token_accuracy": 0.12757945880293847,
"num_tokens": 4618003.0,
"step": 2100
},
{
"entropy": 5.918219375610351,
"epoch": 0.20891226677252878,
"grad_norm": 0.8984375,
"learning_rate": 0.0004998640056543487,
"loss": 5.8908,
"mean_token_accuracy": 0.1205370806157589,
"num_tokens": 4628429.0,
"step": 2105
},
{
"entropy": 6.010912561416626,
"epoch": 0.2094084954346963,
"grad_norm": 0.9609375,
"learning_rate": 0.0004998627711575469,
"loss": 5.9631,
"mean_token_accuracy": 0.12769097983837127,
"num_tokens": 4639414.0,
"step": 2110
},
{
"entropy": 5.969498491287231,
"epoch": 0.20990472409686384,
"grad_norm": 0.7890625,
"learning_rate": 0.000499861531084633,
"loss": 6.0658,
"mean_token_accuracy": 0.12155803292989731,
"num_tokens": 4652594.0,
"step": 2115
},
{
"entropy": 5.99975962638855,
"epoch": 0.21040095275903137,
"grad_norm": 0.921875,
"learning_rate": 0.0004998602854356376,
"loss": 5.8884,
"mean_token_accuracy": 0.12490163445472717,
"num_tokens": 4662981.0,
"step": 2120
},
{
"entropy": 5.956492328643799,
"epoch": 0.2108971814211989,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998590342105917,
"loss": 5.9854,
"mean_token_accuracy": 0.12499078437685966,
"num_tokens": 4672469.0,
"step": 2125
},
{
"entropy": 6.015379524230957,
"epoch": 0.21139341008336643,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998577774095262,
"loss": 5.9799,
"mean_token_accuracy": 0.11904595419764519,
"num_tokens": 4683640.0,
"step": 2130
},
{
"entropy": 5.978434896469116,
"epoch": 0.21188963874553393,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998565150324724,
"loss": 5.9373,
"mean_token_accuracy": 0.12495681792497634,
"num_tokens": 4694032.0,
"step": 2135
},
{
"entropy": 5.911871337890625,
"epoch": 0.21238586740770146,
"grad_norm": 0.921875,
"learning_rate": 0.0004998552470794616,
"loss": 5.823,
"mean_token_accuracy": 0.12693770080804825,
"num_tokens": 4704253.0,
"step": 2140
},
{
"entropy": 5.94983696937561,
"epoch": 0.212882096069869,
"grad_norm": 0.88671875,
"learning_rate": 0.0004998539735505251,
"loss": 5.9686,
"mean_token_accuracy": 0.11972972378134727,
"num_tokens": 4715327.0,
"step": 2145
},
{
"entropy": 6.121762084960937,
"epoch": 0.21337832473203652,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998526944456946,
"loss": 5.9843,
"mean_token_accuracy": 0.11618738770484924,
"num_tokens": 4726450.0,
"step": 2150
},
{
"entropy": 5.960577487945557,
"epoch": 0.21387455339420405,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998514097650018,
"loss": 6.0075,
"mean_token_accuracy": 0.11687625348567962,
"num_tokens": 4736630.0,
"step": 2155
},
{
"entropy": 6.02658486366272,
"epoch": 0.21437078205637158,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998501195084786,
"loss": 5.8837,
"mean_token_accuracy": 0.12827948555350305,
"num_tokens": 4746592.0,
"step": 2160
},
{
"entropy": 5.982325458526612,
"epoch": 0.21486701071853911,
"grad_norm": 0.94140625,
"learning_rate": 0.0004998488236761569,
"loss": 6.0762,
"mean_token_accuracy": 0.1284046731889248,
"num_tokens": 4757824.0,
"step": 2165
},
{
"entropy": 6.027835941314697,
"epoch": 0.21536323938070662,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998475222680688,
"loss": 6.0,
"mean_token_accuracy": 0.12244343236088753,
"num_tokens": 4768453.0,
"step": 2170
},
{
"entropy": 5.9532242774963375,
"epoch": 0.21585946804287415,
"grad_norm": 0.9375,
"learning_rate": 0.0004998462152842468,
"loss": 5.9826,
"mean_token_accuracy": 0.11837821900844574,
"num_tokens": 4779664.0,
"step": 2175
},
{
"entropy": 6.096589851379394,
"epoch": 0.21635569670504168,
"grad_norm": 0.88671875,
"learning_rate": 0.000499844902724723,
"loss": 6.0019,
"mean_token_accuracy": 0.1223648227751255,
"num_tokens": 4791012.0,
"step": 2180
},
{
"entropy": 5.902175521850586,
"epoch": 0.2168519253672092,
"grad_norm": 0.91015625,
"learning_rate": 0.0004998435845895302,
"loss": 5.8794,
"mean_token_accuracy": 0.12457833737134934,
"num_tokens": 4800578.0,
"step": 2185
},
{
"entropy": 5.982421445846557,
"epoch": 0.21734815402937674,
"grad_norm": 0.9609375,
"learning_rate": 0.000499842260878701,
"loss": 5.8903,
"mean_token_accuracy": 0.12047785967588424,
"num_tokens": 4810940.0,
"step": 2190
},
{
"entropy": 5.987110233306884,
"epoch": 0.21784438269154427,
"grad_norm": 0.77734375,
"learning_rate": 0.0004998409315922683,
"loss": 6.0421,
"mean_token_accuracy": 0.12021683305501937,
"num_tokens": 4822696.0,
"step": 2195
},
{
"entropy": 6.000821590423584,
"epoch": 0.2183406113537118,
"grad_norm": 0.95703125,
"learning_rate": 0.0004998395967302649,
"loss": 5.896,
"mean_token_accuracy": 0.1261095680296421,
"num_tokens": 4833669.0,
"step": 2200
},
{
"entropy": 6.080050802230835,
"epoch": 0.21883684001587933,
"grad_norm": 0.91015625,
"learning_rate": 0.000499838256292724,
"loss": 6.0807,
"mean_token_accuracy": 0.12102781087160111,
"num_tokens": 4844021.0,
"step": 2205
},
{
"entropy": 5.95439338684082,
"epoch": 0.21933306867804683,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998369102796787,
"loss": 5.9517,
"mean_token_accuracy": 0.12605141550302507,
"num_tokens": 4855336.0,
"step": 2210
},
{
"entropy": 6.0016449928283695,
"epoch": 0.21982929734021436,
"grad_norm": 0.77734375,
"learning_rate": 0.0004998355586911627,
"loss": 5.9464,
"mean_token_accuracy": 0.12409858405590057,
"num_tokens": 4867713.0,
"step": 2215
},
{
"entropy": 5.9480170726776125,
"epoch": 0.2203255260023819,
"grad_norm": 0.8515625,
"learning_rate": 0.0004998342015272093,
"loss": 5.9325,
"mean_token_accuracy": 0.12322600036859513,
"num_tokens": 4879134.0,
"step": 2220
},
{
"entropy": 6.047569274902344,
"epoch": 0.22082175466454942,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998328387878521,
"loss": 6.0989,
"mean_token_accuracy": 0.1215038277208805,
"num_tokens": 4891755.0,
"step": 2225
},
{
"entropy": 6.152658557891845,
"epoch": 0.22131798332671695,
"grad_norm": 0.953125,
"learning_rate": 0.0004998314704731249,
"loss": 6.0836,
"mean_token_accuracy": 0.11992574110627174,
"num_tokens": 4903264.0,
"step": 2230
},
{
"entropy": 5.933616685867309,
"epoch": 0.22181421198888449,
"grad_norm": 1.0234375,
"learning_rate": 0.0004998300965830618,
"loss": 5.9148,
"mean_token_accuracy": 0.12403179481625556,
"num_tokens": 4912533.0,
"step": 2235
},
{
"entropy": 5.878796815872192,
"epoch": 0.22231044065105202,
"grad_norm": 0.90625,
"learning_rate": 0.0004998287171176968,
"loss": 5.9132,
"mean_token_accuracy": 0.12680847868323325,
"num_tokens": 4923288.0,
"step": 2240
},
{
"entropy": 6.013169240951538,
"epoch": 0.22280666931321952,
"grad_norm": 0.9296875,
"learning_rate": 0.0004998273320770641,
"loss": 5.9415,
"mean_token_accuracy": 0.12333439588546753,
"num_tokens": 4934759.0,
"step": 2245
},
{
"entropy": 5.932636260986328,
"epoch": 0.22330289797538705,
"grad_norm": 1.0234375,
"learning_rate": 0.000499825941461198,
"loss": 5.966,
"mean_token_accuracy": 0.12490196079015732,
"num_tokens": 4947024.0,
"step": 2250
},
{
"entropy": 5.971539068222046,
"epoch": 0.22379912663755458,
"grad_norm": 0.86328125,
"learning_rate": 0.0004998245452701329,
"loss": 5.9222,
"mean_token_accuracy": 0.12477421313524246,
"num_tokens": 4959281.0,
"step": 2255
},
{
"entropy": 6.050222492218017,
"epoch": 0.2242953552997221,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998231435039036,
"loss": 5.9793,
"mean_token_accuracy": 0.11924687549471855,
"num_tokens": 4970466.0,
"step": 2260
},
{
"entropy": 5.949010181427002,
"epoch": 0.22479158396188964,
"grad_norm": 0.90234375,
"learning_rate": 0.0004998217361625449,
"loss": 5.9377,
"mean_token_accuracy": 0.12520918771624565,
"num_tokens": 4981135.0,
"step": 2265
},
{
"entropy": 5.9944313049316404,
"epoch": 0.22528781262405717,
"grad_norm": 0.92578125,
"learning_rate": 0.0004998203232460915,
"loss": 5.9933,
"mean_token_accuracy": 0.12928032726049424,
"num_tokens": 4991292.0,
"step": 2270
},
{
"entropy": 5.963805294036865,
"epoch": 0.2257840412862247,
"grad_norm": 0.97265625,
"learning_rate": 0.0004998189047545785,
"loss": 5.9373,
"mean_token_accuracy": 0.12618958726525306,
"num_tokens": 5002806.0,
"step": 2275
},
{
"entropy": 5.945759534835815,
"epoch": 0.22628026994839223,
"grad_norm": 0.98046875,
"learning_rate": 0.0004998174806880411,
"loss": 5.8989,
"mean_token_accuracy": 0.13040453866124152,
"num_tokens": 5013474.0,
"step": 2280
},
{
"entropy": 5.951687908172607,
"epoch": 0.22677649861055973,
"grad_norm": 0.91796875,
"learning_rate": 0.0004998160510465146,
"loss": 5.9352,
"mean_token_accuracy": 0.13000500723719596,
"num_tokens": 5025151.0,
"step": 2285
},
{
"entropy": 5.926647806167603,
"epoch": 0.22727272727272727,
"grad_norm": 0.87890625,
"learning_rate": 0.0004998146158300345,
"loss": 5.9359,
"mean_token_accuracy": 0.1301646463572979,
"num_tokens": 5037166.0,
"step": 2290
},
{
"entropy": 6.0174994468688965,
"epoch": 0.2277689559348948,
"grad_norm": 0.8671875,
"learning_rate": 0.0004998131750386364,
"loss": 5.9849,
"mean_token_accuracy": 0.12377216964960099,
"num_tokens": 5047452.0,
"step": 2295
},
{
"entropy": 5.9414409637451175,
"epoch": 0.22826518459706233,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998117286723559,
"loss": 5.9811,
"mean_token_accuracy": 0.12003129869699478,
"num_tokens": 5058501.0,
"step": 2300
},
{
"entropy": 6.017748069763184,
"epoch": 0.22876141325922986,
"grad_norm": 0.96875,
"learning_rate": 0.000499810276731229,
"loss": 5.9855,
"mean_token_accuracy": 0.12389934659004212,
"num_tokens": 5069716.0,
"step": 2305
},
{
"entropy": 6.036419868469238,
"epoch": 0.2292576419213974,
"grad_norm": 0.99609375,
"learning_rate": 0.0004998088192152918,
"loss": 5.9543,
"mean_token_accuracy": 0.12356158047914505,
"num_tokens": 5078509.0,
"step": 2310
},
{
"entropy": 5.953884792327881,
"epoch": 0.22975387058356492,
"grad_norm": 0.96875,
"learning_rate": 0.0004998073561245801,
"loss": 6.0043,
"mean_token_accuracy": 0.12320954129099845,
"num_tokens": 5090355.0,
"step": 2315
},
{
"entropy": 6.02992992401123,
"epoch": 0.23025009924573242,
"grad_norm": 0.8828125,
"learning_rate": 0.0004998058874591305,
"loss": 5.8885,
"mean_token_accuracy": 0.12766256108880042,
"num_tokens": 5100582.0,
"step": 2320
},
{
"entropy": 5.984819793701172,
"epoch": 0.23074632790789995,
"grad_norm": 0.79296875,
"learning_rate": 0.0004998044132189793,
"loss": 5.9563,
"mean_token_accuracy": 0.12425923943519593,
"num_tokens": 5112176.0,
"step": 2325
},
{
"entropy": 5.964567947387695,
"epoch": 0.23124255657006748,
"grad_norm": 1.09375,
"learning_rate": 0.000499802933404163,
"loss": 5.8697,
"mean_token_accuracy": 0.12405415698885917,
"num_tokens": 5122840.0,
"step": 2330
},
{
"entropy": 5.942762327194214,
"epoch": 0.231738785232235,
"grad_norm": 0.96484375,
"learning_rate": 0.0004998014480147183,
"loss": 5.8782,
"mean_token_accuracy": 0.12970717400312423,
"num_tokens": 5132926.0,
"step": 2335
},
{
"entropy": 5.883530950546264,
"epoch": 0.23223501389440254,
"grad_norm": 1.0,
"learning_rate": 0.0004997999570506823,
"loss": 5.8536,
"mean_token_accuracy": 0.12876308858394622,
"num_tokens": 5143394.0,
"step": 2340
},
{
"entropy": 5.885425662994384,
"epoch": 0.23273124255657007,
"grad_norm": 0.84375,
"learning_rate": 0.0004997984605120917,
"loss": 5.8876,
"mean_token_accuracy": 0.12427316978573799,
"num_tokens": 5155710.0,
"step": 2345
},
{
"entropy": 5.837533855438233,
"epoch": 0.2332274712187376,
"grad_norm": 0.98046875,
"learning_rate": 0.0004997969583989835,
"loss": 5.8163,
"mean_token_accuracy": 0.12433635219931602,
"num_tokens": 5167065.0,
"step": 2350
},
{
"entropy": 5.938689565658569,
"epoch": 0.23372369988090513,
"grad_norm": 0.9765625,
"learning_rate": 0.0004997954507113952,
"loss": 5.9198,
"mean_token_accuracy": 0.1236913837492466,
"num_tokens": 5177535.0,
"step": 2355
},
{
"entropy": 5.9044514179229735,
"epoch": 0.23421992854307264,
"grad_norm": 0.9140625,
"learning_rate": 0.0004997939374493642,
"loss": 5.8341,
"mean_token_accuracy": 0.12915939837694168,
"num_tokens": 5188324.0,
"step": 2360
},
{
"entropy": 5.918756198883057,
"epoch": 0.23471615720524017,
"grad_norm": 0.91015625,
"learning_rate": 0.000499792418612928,
"loss": 5.9219,
"mean_token_accuracy": 0.12620971575379372,
"num_tokens": 5199223.0,
"step": 2365
},
{
"entropy": 5.933010911941528,
"epoch": 0.2352123858674077,
"grad_norm": 0.9296875,
"learning_rate": 0.0004997908942021241,
"loss": 5.9016,
"mean_token_accuracy": 0.12857112437486648,
"num_tokens": 5210080.0,
"step": 2370
},
{
"entropy": 6.007422018051147,
"epoch": 0.23570861452957523,
"grad_norm": 0.90625,
"learning_rate": 0.0004997893642169904,
"loss": 6.0153,
"mean_token_accuracy": 0.12386548593640327,
"num_tokens": 5220867.0,
"step": 2375
},
{
"entropy": 5.9750877857208256,
"epoch": 0.23620484319174276,
"grad_norm": 1.015625,
"learning_rate": 0.0004997878286575648,
"loss": 5.8596,
"mean_token_accuracy": 0.12206832617521286,
"num_tokens": 5231451.0,
"step": 2380
},
{
"entropy": 5.932880973815918,
"epoch": 0.2367010718539103,
"grad_norm": 0.953125,
"learning_rate": 0.0004997862875238855,
"loss": 5.9662,
"mean_token_accuracy": 0.12139840200543403,
"num_tokens": 5244643.0,
"step": 2385
},
{
"entropy": 5.986804580688476,
"epoch": 0.23719730051607782,
"grad_norm": 1.046875,
"learning_rate": 0.0004997847408159907,
"loss": 5.9781,
"mean_token_accuracy": 0.11682926490902901,
"num_tokens": 5256279.0,
"step": 2390
},
{
"entropy": 5.988444232940674,
"epoch": 0.23769352917824532,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997831885339186,
"loss": 5.9486,
"mean_token_accuracy": 0.12903917878866195,
"num_tokens": 5268498.0,
"step": 2395
},
{
"entropy": 5.876736688613891,
"epoch": 0.23818975784041285,
"grad_norm": 0.8984375,
"learning_rate": 0.0004997816306777077,
"loss": 5.9269,
"mean_token_accuracy": 0.1273012116551399,
"num_tokens": 5280869.0,
"step": 2400
},
{
"entropy": 6.107938194274903,
"epoch": 0.23868598650258038,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997800672473969,
"loss": 6.024,
"mean_token_accuracy": 0.12057831212878227,
"num_tokens": 5293082.0,
"step": 2405
},
{
"entropy": 5.8743205070495605,
"epoch": 0.2391822151647479,
"grad_norm": 1.0625,
"learning_rate": 0.0004997784982430247,
"loss": 5.8092,
"mean_token_accuracy": 0.1349871814250946,
"num_tokens": 5304039.0,
"step": 2410
},
{
"entropy": 5.888135862350464,
"epoch": 0.23967844382691544,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997769236646301,
"loss": 5.8874,
"mean_token_accuracy": 0.12667458057403563,
"num_tokens": 5315016.0,
"step": 2415
},
{
"entropy": 6.046527481079101,
"epoch": 0.24017467248908297,
"grad_norm": 0.953125,
"learning_rate": 0.0004997753435122522,
"loss": 6.0304,
"mean_token_accuracy": 0.1209819495677948,
"num_tokens": 5326090.0,
"step": 2420
},
{
"entropy": 5.8429787158966064,
"epoch": 0.2406709011512505,
"grad_norm": 1.0,
"learning_rate": 0.00049977375778593,
"loss": 5.8061,
"mean_token_accuracy": 0.128286737203598,
"num_tokens": 5336423.0,
"step": 2425
},
{
"entropy": 5.994162750244141,
"epoch": 0.24116712981341804,
"grad_norm": 0.91796875,
"learning_rate": 0.0004997721664857032,
"loss": 5.9764,
"mean_token_accuracy": 0.1251453071832657,
"num_tokens": 5346982.0,
"step": 2430
},
{
"entropy": 5.9075380802154545,
"epoch": 0.24166335847558554,
"grad_norm": 0.94921875,
"learning_rate": 0.0004997705696116108,
"loss": 5.8802,
"mean_token_accuracy": 0.13289263024926184,
"num_tokens": 5359210.0,
"step": 2435
},
{
"entropy": 5.939221620559692,
"epoch": 0.24215958713775307,
"grad_norm": 0.86328125,
"learning_rate": 0.0004997689671636926,
"loss": 5.8914,
"mean_token_accuracy": 0.11953317150473594,
"num_tokens": 5369891.0,
"step": 2440
},
{
"entropy": 5.891269207000732,
"epoch": 0.2426558157999206,
"grad_norm": 0.85546875,
"learning_rate": 0.0004997673591419884,
"loss": 5.8777,
"mean_token_accuracy": 0.12496770992875099,
"num_tokens": 5380850.0,
"step": 2445
},
{
"entropy": 5.939934349060058,
"epoch": 0.24315204446208813,
"grad_norm": 0.82421875,
"learning_rate": 0.0004997657455465381,
"loss": 5.9339,
"mean_token_accuracy": 0.12732330337166786,
"num_tokens": 5391903.0,
"step": 2450
},
{
"entropy": 5.993845796585083,
"epoch": 0.24364827312425566,
"grad_norm": 0.93359375,
"learning_rate": 0.0004997641263773815,
"loss": 5.9966,
"mean_token_accuracy": 0.12384477332234382,
"num_tokens": 5402517.0,
"step": 2455
},
{
"entropy": 5.8536583423614506,
"epoch": 0.2441445017864232,
"grad_norm": 0.94921875,
"learning_rate": 0.000499762501634559,
"loss": 5.9017,
"mean_token_accuracy": 0.1264289692044258,
"num_tokens": 5414110.0,
"step": 2460
},
{
"entropy": 6.013503313064575,
"epoch": 0.24464073044859072,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997608713181108,
"loss": 5.9143,
"mean_token_accuracy": 0.12651659920811653,
"num_tokens": 5424466.0,
"step": 2465
},
{
"entropy": 5.9579092502594,
"epoch": 0.24513695911075822,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997592354280772,
"loss": 5.9171,
"mean_token_accuracy": 0.13171297907829285,
"num_tokens": 5434461.0,
"step": 2470
},
{
"entropy": 5.863575887680054,
"epoch": 0.24563318777292575,
"grad_norm": 0.94140625,
"learning_rate": 0.000499757593964499,
"loss": 5.8918,
"mean_token_accuracy": 0.1223128654062748,
"num_tokens": 5445921.0,
"step": 2475
},
{
"entropy": 5.945754766464233,
"epoch": 0.24612941643509328,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997559469274166,
"loss": 5.8846,
"mean_token_accuracy": 0.12594258412718773,
"num_tokens": 5456283.0,
"step": 2480
},
{
"entropy": 5.98380069732666,
"epoch": 0.24662564509726081,
"grad_norm": 0.91015625,
"learning_rate": 0.0004997542943168712,
"loss": 5.9281,
"mean_token_accuracy": 0.1247189499437809,
"num_tokens": 5467616.0,
"step": 2485
},
{
"entropy": 5.927638578414917,
"epoch": 0.24712187375942835,
"grad_norm": 1.03125,
"learning_rate": 0.0004997526361329035,
"loss": 5.8764,
"mean_token_accuracy": 0.13195810765028,
"num_tokens": 5478050.0,
"step": 2490
},
{
"entropy": 5.97781229019165,
"epoch": 0.24761810242159588,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997509723755547,
"loss": 5.8846,
"mean_token_accuracy": 0.12755061611533164,
"num_tokens": 5487185.0,
"step": 2495
},
{
"entropy": 5.910194063186646,
"epoch": 0.2481143310837634,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997493030448662,
"loss": 5.9018,
"mean_token_accuracy": 0.12408641576766968,
"num_tokens": 5498802.0,
"step": 2500
},
{
"entropy": 5.906022214889527,
"epoch": 0.24861055974593094,
"grad_norm": 0.88671875,
"learning_rate": 0.0004997476281408793,
"loss": 5.9155,
"mean_token_accuracy": 0.12289765030145645,
"num_tokens": 5510285.0,
"step": 2505
},
{
"entropy": 5.956116676330566,
"epoch": 0.24910678840809844,
"grad_norm": 0.9609375,
"learning_rate": 0.0004997459476636354,
"loss": 5.8077,
"mean_token_accuracy": 0.13291431069374085,
"num_tokens": 5520028.0,
"step": 2510
},
{
"entropy": 5.836493730545044,
"epoch": 0.24960301707026597,
"grad_norm": 1.0703125,
"learning_rate": 0.0004997442616131763,
"loss": 5.9187,
"mean_token_accuracy": 0.12413758859038353,
"num_tokens": 5529271.0,
"step": 2515
},
{
"entropy": 5.923549509048462,
"epoch": 0.25009924573243353,
"grad_norm": 1.0859375,
"learning_rate": 0.0004997425699895438,
"loss": 5.8478,
"mean_token_accuracy": 0.12018645405769349,
"num_tokens": 5539396.0,
"step": 2520
},
{
"entropy": 5.942753076553345,
"epoch": 0.25059547439460106,
"grad_norm": 0.83203125,
"learning_rate": 0.0004997408727927798,
"loss": 5.9803,
"mean_token_accuracy": 0.12607812359929085,
"num_tokens": 5551674.0,
"step": 2525
},
{
"entropy": 5.946086692810058,
"epoch": 0.25109170305676853,
"grad_norm": 0.9921875,
"learning_rate": 0.0004997391700229264,
"loss": 5.9052,
"mean_token_accuracy": 0.1281418062746525,
"num_tokens": 5562749.0,
"step": 2530
},
{
"entropy": 5.937380933761597,
"epoch": 0.25158793171893606,
"grad_norm": 0.9296875,
"learning_rate": 0.000499737461680026,
"loss": 5.9167,
"mean_token_accuracy": 0.12441470995545387,
"num_tokens": 5573478.0,
"step": 2535
},
{
"entropy": 5.853802585601807,
"epoch": 0.2520841603811036,
"grad_norm": 0.98828125,
"learning_rate": 0.0004997357477641207,
"loss": 5.9051,
"mean_token_accuracy": 0.12962415516376496,
"num_tokens": 5584546.0,
"step": 2540
},
{
"entropy": 5.911726713180542,
"epoch": 0.2525803890432711,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997340282752532,
"loss": 5.7971,
"mean_token_accuracy": 0.13067309483885764,
"num_tokens": 5594004.0,
"step": 2545
},
{
"entropy": 5.941144037246704,
"epoch": 0.25307661770543866,
"grad_norm": 0.9140625,
"learning_rate": 0.000499732303213466,
"loss": 5.8453,
"mean_token_accuracy": 0.1209590882062912,
"num_tokens": 5606095.0,
"step": 2550
},
{
"entropy": 5.907631063461304,
"epoch": 0.2535728463676062,
"grad_norm": 1.109375,
"learning_rate": 0.0004997305725788019,
"loss": 5.8957,
"mean_token_accuracy": 0.11941977143287659,
"num_tokens": 5615596.0,
"step": 2555
},
{
"entropy": 5.896251630783081,
"epoch": 0.2540690750297737,
"grad_norm": 0.88671875,
"learning_rate": 0.000499728836371304,
"loss": 5.827,
"mean_token_accuracy": 0.12794101238250732,
"num_tokens": 5626753.0,
"step": 2560
},
{
"entropy": 5.832917833328247,
"epoch": 0.25456530369194125,
"grad_norm": 0.96875,
"learning_rate": 0.0004997270945910151,
"loss": 5.8598,
"mean_token_accuracy": 0.12301537916064262,
"num_tokens": 5636135.0,
"step": 2565
},
{
"entropy": 5.945082187652588,
"epoch": 0.2550615323541088,
"grad_norm": 0.99609375,
"learning_rate": 0.0004997253472379785,
"loss": 5.9181,
"mean_token_accuracy": 0.1255082905292511,
"num_tokens": 5647074.0,
"step": 2570
},
{
"entropy": 5.991348171234131,
"epoch": 0.2555577610162763,
"grad_norm": 0.79296875,
"learning_rate": 0.0004997235943122375,
"loss": 5.9454,
"mean_token_accuracy": 0.12038756832480431,
"num_tokens": 5658600.0,
"step": 2575
},
{
"entropy": 5.845171403884888,
"epoch": 0.25605398967844384,
"grad_norm": 0.90234375,
"learning_rate": 0.0004997218358138356,
"loss": 5.8206,
"mean_token_accuracy": 0.12867578938603402,
"num_tokens": 5668453.0,
"step": 2580
},
{
"entropy": 5.950637292861939,
"epoch": 0.25655021834061137,
"grad_norm": 1.078125,
"learning_rate": 0.0004997200717428165,
"loss": 5.9086,
"mean_token_accuracy": 0.1230459950864315,
"num_tokens": 5678855.0,
"step": 2585
},
{
"entropy": 5.937513780593872,
"epoch": 0.2570464470027789,
"grad_norm": 0.92578125,
"learning_rate": 0.0004997183020992237,
"loss": 5.9609,
"mean_token_accuracy": 0.12079612538218498,
"num_tokens": 5690063.0,
"step": 2590
},
{
"entropy": 5.882065010070801,
"epoch": 0.25754267566494643,
"grad_norm": 0.95703125,
"learning_rate": 0.0004997165268831014,
"loss": 5.8518,
"mean_token_accuracy": 0.12909654155373573,
"num_tokens": 5701487.0,
"step": 2595
},
{
"entropy": 5.876619815826416,
"epoch": 0.25803890432711396,
"grad_norm": 0.87890625,
"learning_rate": 0.0004997147460944933,
"loss": 5.8497,
"mean_token_accuracy": 0.12846047356724738,
"num_tokens": 5713034.0,
"step": 2600
},
{
"entropy": 5.968547630310058,
"epoch": 0.25853513298928144,
"grad_norm": 0.84375,
"learning_rate": 0.0004997129597334438,
"loss": 5.8875,
"mean_token_accuracy": 0.1237348847091198,
"num_tokens": 5725469.0,
"step": 2605
},
{
"entropy": 5.886534070968628,
"epoch": 0.25903136165144897,
"grad_norm": 0.84375,
"learning_rate": 0.0004997111677999971,
"loss": 5.8853,
"mean_token_accuracy": 0.12469863295555114,
"num_tokens": 5737099.0,
"step": 2610
},
{
"entropy": 5.934344339370727,
"epoch": 0.2595275903136165,
"grad_norm": 0.89453125,
"learning_rate": 0.0004997093702941976,
"loss": 5.8837,
"mean_token_accuracy": 0.12119583413004875,
"num_tokens": 5748756.0,
"step": 2615
},
{
"entropy": 5.952655696868897,
"epoch": 0.260023818975784,
"grad_norm": 1.0078125,
"learning_rate": 0.00049970756721609,
"loss": 5.8767,
"mean_token_accuracy": 0.12939024344086647,
"num_tokens": 5759598.0,
"step": 2620
},
{
"entropy": 5.8704029560089115,
"epoch": 0.26052004763795156,
"grad_norm": 0.84765625,
"learning_rate": 0.000499705758565719,
"loss": 5.9176,
"mean_token_accuracy": 0.12523015439510346,
"num_tokens": 5770397.0,
"step": 2625
},
{
"entropy": 5.990549516677857,
"epoch": 0.2610162763001191,
"grad_norm": 0.890625,
"learning_rate": 0.0004997039443431292,
"loss": 6.0049,
"mean_token_accuracy": 0.1170310489833355,
"num_tokens": 5780935.0,
"step": 2630
},
{
"entropy": 5.979841566085815,
"epoch": 0.2615125049622866,
"grad_norm": 1.0234375,
"learning_rate": 0.0004997021245483659,
"loss": 5.8621,
"mean_token_accuracy": 0.12379876151680946,
"num_tokens": 5791557.0,
"step": 2635
},
{
"entropy": 5.852493476867676,
"epoch": 0.26200873362445415,
"grad_norm": 0.92578125,
"learning_rate": 0.000499700299181474,
"loss": 5.8745,
"mean_token_accuracy": 0.12755739986896514,
"num_tokens": 5802233.0,
"step": 2640
},
{
"entropy": 5.857036972045899,
"epoch": 0.2625049622866217,
"grad_norm": 0.94140625,
"learning_rate": 0.0004996984682424989,
"loss": 5.8121,
"mean_token_accuracy": 0.13497790470719337,
"num_tokens": 5813424.0,
"step": 2645
},
{
"entropy": 5.838797950744629,
"epoch": 0.2630011909487892,
"grad_norm": 0.7890625,
"learning_rate": 0.000499696631731486,
"loss": 5.7849,
"mean_token_accuracy": 0.12850171998143195,
"num_tokens": 5825652.0,
"step": 2650
},
{
"entropy": 5.9219934940338135,
"epoch": 0.26349741961095674,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996947896484809,
"loss": 5.882,
"mean_token_accuracy": 0.12964971140027046,
"num_tokens": 5836603.0,
"step": 2655
},
{
"entropy": 5.882232713699341,
"epoch": 0.26399364827312427,
"grad_norm": 0.96484375,
"learning_rate": 0.000499692941993529,
"loss": 5.9252,
"mean_token_accuracy": 0.12897176370024682,
"num_tokens": 5849365.0,
"step": 2660
},
{
"entropy": 5.976453876495361,
"epoch": 0.2644898769352918,
"grad_norm": 0.87890625,
"learning_rate": 0.0004996910887666765,
"loss": 5.8506,
"mean_token_accuracy": 0.130551964789629,
"num_tokens": 5860305.0,
"step": 2665
},
{
"entropy": 5.838088703155518,
"epoch": 0.26498610559745933,
"grad_norm": 0.87109375,
"learning_rate": 0.000499689229967969,
"loss": 5.7878,
"mean_token_accuracy": 0.1278667666018009,
"num_tokens": 5872272.0,
"step": 2670
},
{
"entropy": 5.881339263916016,
"epoch": 0.26548233425962686,
"grad_norm": 1.0390625,
"learning_rate": 0.0004996873655974529,
"loss": 5.8187,
"mean_token_accuracy": 0.12291400730609894,
"num_tokens": 5884107.0,
"step": 2675
},
{
"entropy": 5.816775846481323,
"epoch": 0.26597856292179434,
"grad_norm": 0.9765625,
"learning_rate": 0.0004996854956551743,
"loss": 5.8165,
"mean_token_accuracy": 0.13490741476416587,
"num_tokens": 5894060.0,
"step": 2680
},
{
"entropy": 5.898659563064575,
"epoch": 0.26647479158396187,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996836201411794,
"loss": 5.8378,
"mean_token_accuracy": 0.1263569936156273,
"num_tokens": 5904804.0,
"step": 2685
},
{
"entropy": 5.890391683578491,
"epoch": 0.2669710202461294,
"grad_norm": 1.046875,
"learning_rate": 0.000499681739055515,
"loss": 5.8976,
"mean_token_accuracy": 0.12535513564944267,
"num_tokens": 5916496.0,
"step": 2690
},
{
"entropy": 5.904550838470459,
"epoch": 0.26746724890829693,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996798523982276,
"loss": 5.8488,
"mean_token_accuracy": 0.12368987798690796,
"num_tokens": 5926920.0,
"step": 2695
},
{
"entropy": 6.00138578414917,
"epoch": 0.26796347757046446,
"grad_norm": 0.9375,
"learning_rate": 0.000499677960169364,
"loss": 5.9644,
"mean_token_accuracy": 0.12470580637454987,
"num_tokens": 5940098.0,
"step": 2700
},
{
"entropy": 5.859459924697876,
"epoch": 0.268459706232632,
"grad_norm": 0.96875,
"learning_rate": 0.0004996760623689711,
"loss": 5.8033,
"mean_token_accuracy": 0.12954767867922784,
"num_tokens": 5949687.0,
"step": 2705
},
{
"entropy": 5.925769805908203,
"epoch": 0.2689559348947995,
"grad_norm": 0.890625,
"learning_rate": 0.000499674158997096,
"loss": 5.9077,
"mean_token_accuracy": 0.12583340257406234,
"num_tokens": 5960777.0,
"step": 2710
},
{
"entropy": 5.945502376556396,
"epoch": 0.26945216355696705,
"grad_norm": 0.96875,
"learning_rate": 0.0004996722500537859,
"loss": 5.8254,
"mean_token_accuracy": 0.12575791031122208,
"num_tokens": 5972298.0,
"step": 2715
},
{
"entropy": 5.855357360839844,
"epoch": 0.2699483922191346,
"grad_norm": 0.8984375,
"learning_rate": 0.0004996703355390881,
"loss": 5.9376,
"mean_token_accuracy": 0.12119691222906112,
"num_tokens": 5985613.0,
"step": 2720
},
{
"entropy": 5.991959714889527,
"epoch": 0.2704446208813021,
"grad_norm": 0.9140625,
"learning_rate": 0.00049966841545305,
"loss": 5.9844,
"mean_token_accuracy": 0.1248553104698658,
"num_tokens": 5997530.0,
"step": 2725
},
{
"entropy": 5.943222093582153,
"epoch": 0.27094084954346964,
"grad_norm": 0.890625,
"learning_rate": 0.0004996664897957195,
"loss": 5.8614,
"mean_token_accuracy": 0.12187350690364837,
"num_tokens": 6008123.0,
"step": 2730
},
{
"entropy": 5.869727277755738,
"epoch": 0.2714370782056372,
"grad_norm": 0.890625,
"learning_rate": 0.0004996645585671441,
"loss": 5.8702,
"mean_token_accuracy": 0.1277486801147461,
"num_tokens": 6020578.0,
"step": 2735
},
{
"entropy": 5.862465906143188,
"epoch": 0.2719333068678047,
"grad_norm": 0.9375,
"learning_rate": 0.0004996626217673717,
"loss": 5.7927,
"mean_token_accuracy": 0.13384850472211837,
"num_tokens": 6031625.0,
"step": 2740
},
{
"entropy": 5.851309490203858,
"epoch": 0.27242953552997223,
"grad_norm": 0.859375,
"learning_rate": 0.0004996606793964504,
"loss": 5.8372,
"mean_token_accuracy": 0.13454477563500405,
"num_tokens": 6041320.0,
"step": 2745
},
{
"entropy": 5.875042676925659,
"epoch": 0.27292576419213976,
"grad_norm": 0.8828125,
"learning_rate": 0.0004996587314544284,
"loss": 5.8325,
"mean_token_accuracy": 0.1278717115521431,
"num_tokens": 6052627.0,
"step": 2750
},
{
"entropy": 5.901455354690552,
"epoch": 0.27342199285430724,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996567779413539,
"loss": 5.8164,
"mean_token_accuracy": 0.13307981044054032,
"num_tokens": 6063468.0,
"step": 2755
},
{
"entropy": 5.850373125076294,
"epoch": 0.27391822151647477,
"grad_norm": 0.94921875,
"learning_rate": 0.0004996548188572755,
"loss": 5.7943,
"mean_token_accuracy": 0.134738889336586,
"num_tokens": 6074208.0,
"step": 2760
},
{
"entropy": 5.893016290664673,
"epoch": 0.2744144501786423,
"grad_norm": 0.93359375,
"learning_rate": 0.0004996528542022416,
"loss": 5.8448,
"mean_token_accuracy": 0.1275886908173561,
"num_tokens": 6085068.0,
"step": 2765
},
{
"entropy": 5.887360763549805,
"epoch": 0.27491067884080983,
"grad_norm": 0.875,
"learning_rate": 0.000499650883976301,
"loss": 5.8585,
"mean_token_accuracy": 0.12763815000653267,
"num_tokens": 6096445.0,
"step": 2770
},
{
"entropy": 5.812702941894531,
"epoch": 0.27540690750297736,
"grad_norm": 0.96484375,
"learning_rate": 0.0004996489081795026,
"loss": 5.7673,
"mean_token_accuracy": 0.12923398315906526,
"num_tokens": 6106527.0,
"step": 2775
},
{
"entropy": 5.8115157127380375,
"epoch": 0.2759031361651449,
"grad_norm": 1.0546875,
"learning_rate": 0.0004996469268118953,
"loss": 5.7841,
"mean_token_accuracy": 0.13443052023649216,
"num_tokens": 6115923.0,
"step": 2780
},
{
"entropy": 5.800508975982666,
"epoch": 0.2763993648273124,
"grad_norm": 0.921875,
"learning_rate": 0.0004996449398735283,
"loss": 5.6704,
"mean_token_accuracy": 0.13257666900753975,
"num_tokens": 6125861.0,
"step": 2785
},
{
"entropy": 5.881771564483643,
"epoch": 0.27689559348947995,
"grad_norm": 0.86328125,
"learning_rate": 0.0004996429473644509,
"loss": 5.8686,
"mean_token_accuracy": 0.1287565641105175,
"num_tokens": 6137189.0,
"step": 2790
},
{
"entropy": 5.870003747940063,
"epoch": 0.2773918221516475,
"grad_norm": 0.84375,
"learning_rate": 0.0004996409492847124,
"loss": 5.7946,
"mean_token_accuracy": 0.129666505753994,
"num_tokens": 6147920.0,
"step": 2795
},
{
"entropy": 5.8141978740692135,
"epoch": 0.277888050813815,
"grad_norm": 0.89453125,
"learning_rate": 0.0004996389456343624,
"loss": 5.7874,
"mean_token_accuracy": 0.13501562774181367,
"num_tokens": 6159371.0,
"step": 2800
},
{
"entropy": 5.945824146270752,
"epoch": 0.27838427947598254,
"grad_norm": 0.828125,
"learning_rate": 0.0004996369364134507,
"loss": 5.9822,
"mean_token_accuracy": 0.12169614806771278,
"num_tokens": 6172679.0,
"step": 2805
},
{
"entropy": 5.854594326019287,
"epoch": 0.2788805081381501,
"grad_norm": 0.8828125,
"learning_rate": 0.000499634921622027,
"loss": 5.8441,
"mean_token_accuracy": 0.12827418893575668,
"num_tokens": 6184244.0,
"step": 2810
},
{
"entropy": 5.838413953781128,
"epoch": 0.2793767368003176,
"grad_norm": 0.9140625,
"learning_rate": 0.0004996329012601413,
"loss": 5.7281,
"mean_token_accuracy": 0.12748083770275115,
"num_tokens": 6194826.0,
"step": 2815
},
{
"entropy": 5.906148958206177,
"epoch": 0.27987296546248513,
"grad_norm": 0.96875,
"learning_rate": 0.0004996308753278437,
"loss": 5.8572,
"mean_token_accuracy": 0.12726689577102662,
"num_tokens": 6206808.0,
"step": 2820
},
{
"entropy": 5.886915063858032,
"epoch": 0.28036919412465267,
"grad_norm": 0.95703125,
"learning_rate": 0.0004996288438251843,
"loss": 5.8789,
"mean_token_accuracy": 0.12301005497574806,
"num_tokens": 6217303.0,
"step": 2825
},
{
"entropy": 5.944884872436523,
"epoch": 0.28086542278682014,
"grad_norm": 0.9921875,
"learning_rate": 0.0004996268067522137,
"loss": 5.8766,
"mean_token_accuracy": 0.12609019502997398,
"num_tokens": 6227965.0,
"step": 2830
},
{
"entropy": 5.942210054397583,
"epoch": 0.28136165144898767,
"grad_norm": 1.046875,
"learning_rate": 0.0004996247641089823,
"loss": 5.9284,
"mean_token_accuracy": 0.12285050153732299,
"num_tokens": 6239011.0,
"step": 2835
},
{
"entropy": 5.796023035049439,
"epoch": 0.2818578801111552,
"grad_norm": 1.046875,
"learning_rate": 0.0004996227158955409,
"loss": 5.7826,
"mean_token_accuracy": 0.13204773068428038,
"num_tokens": 6249084.0,
"step": 2840
},
{
"entropy": 5.8865186214447025,
"epoch": 0.28235410877332273,
"grad_norm": 0.9296875,
"learning_rate": 0.00049962066211194,
"loss": 5.8446,
"mean_token_accuracy": 0.12603942677378654,
"num_tokens": 6260706.0,
"step": 2845
},
{
"entropy": 5.841519212722778,
"epoch": 0.28285033743549026,
"grad_norm": 1.0078125,
"learning_rate": 0.0004996186027582308,
"loss": 5.7821,
"mean_token_accuracy": 0.13092539235949516,
"num_tokens": 6272848.0,
"step": 2850
},
{
"entropy": 5.836033535003662,
"epoch": 0.2833465660976578,
"grad_norm": 0.98828125,
"learning_rate": 0.0004996165378344642,
"loss": 5.7843,
"mean_token_accuracy": 0.13042048811912538,
"num_tokens": 6283820.0,
"step": 2855
},
{
"entropy": 5.854663515090943,
"epoch": 0.2838427947598253,
"grad_norm": 0.98046875,
"learning_rate": 0.0004996144673406914,
"loss": 5.9457,
"mean_token_accuracy": 0.12040838748216628,
"num_tokens": 6294193.0,
"step": 2860
},
{
"entropy": 5.911272716522217,
"epoch": 0.28433902342199285,
"grad_norm": 0.9453125,
"learning_rate": 0.0004996123912769639,
"loss": 5.7933,
"mean_token_accuracy": 0.13677143752574922,
"num_tokens": 6304407.0,
"step": 2865
},
{
"entropy": 5.796115159988403,
"epoch": 0.2848352520841604,
"grad_norm": 0.88671875,
"learning_rate": 0.0004996103096433332,
"loss": 5.867,
"mean_token_accuracy": 0.12861735597252846,
"num_tokens": 6316311.0,
"step": 2870
},
{
"entropy": 5.917466163635254,
"epoch": 0.2853314807463279,
"grad_norm": 0.890625,
"learning_rate": 0.0004996082224398506,
"loss": 5.8353,
"mean_token_accuracy": 0.13029164150357248,
"num_tokens": 6327248.0,
"step": 2875
},
{
"entropy": 5.865206480026245,
"epoch": 0.28582770940849545,
"grad_norm": 0.9296875,
"learning_rate": 0.0004996061296665682,
"loss": 5.8516,
"mean_token_accuracy": 0.13090819716453553,
"num_tokens": 6338494.0,
"step": 2880
},
{
"entropy": 5.866081285476684,
"epoch": 0.286323938070663,
"grad_norm": 0.99609375,
"learning_rate": 0.0004996040313235377,
"loss": 5.7413,
"mean_token_accuracy": 0.1322721965610981,
"num_tokens": 6349038.0,
"step": 2885
},
{
"entropy": 5.829685258865356,
"epoch": 0.2868201667328305,
"grad_norm": 0.8671875,
"learning_rate": 0.0004996019274108112,
"loss": 5.847,
"mean_token_accuracy": 0.12434035912156105,
"num_tokens": 6360180.0,
"step": 2890
},
{
"entropy": 5.818691205978394,
"epoch": 0.28731639539499804,
"grad_norm": 0.97265625,
"learning_rate": 0.000499599817928441,
"loss": 5.7882,
"mean_token_accuracy": 0.1332053415477276,
"num_tokens": 6370931.0,
"step": 2895
},
{
"entropy": 5.768586444854736,
"epoch": 0.28781262405716557,
"grad_norm": 0.88671875,
"learning_rate": 0.0004995977028764792,
"loss": 5.7551,
"mean_token_accuracy": 0.13535102233290672,
"num_tokens": 6381810.0,
"step": 2900
},
{
"entropy": 5.846061563491821,
"epoch": 0.28830885271933304,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995955822549783,
"loss": 5.7766,
"mean_token_accuracy": 0.13201335817575455,
"num_tokens": 6392268.0,
"step": 2905
},
{
"entropy": 5.875686454772949,
"epoch": 0.2888050813815006,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995934560639909,
"loss": 5.8363,
"mean_token_accuracy": 0.12547914758324624,
"num_tokens": 6403650.0,
"step": 2910
},
{
"entropy": 5.90367488861084,
"epoch": 0.2893013100436681,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995913243035699,
"loss": 5.804,
"mean_token_accuracy": 0.12914050742983818,
"num_tokens": 6414672.0,
"step": 2915
},
{
"entropy": 5.841298294067383,
"epoch": 0.28979753870583563,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995891869737678,
"loss": 5.8772,
"mean_token_accuracy": 0.13006574884057046,
"num_tokens": 6424916.0,
"step": 2920
},
{
"entropy": 5.897180557250977,
"epoch": 0.29029376736800316,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995870440746379,
"loss": 5.8448,
"mean_token_accuracy": 0.13426648080348969,
"num_tokens": 6435950.0,
"step": 2925
},
{
"entropy": 5.772582006454468,
"epoch": 0.2907899960301707,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995848956062333,
"loss": 5.7987,
"mean_token_accuracy": 0.1304109625518322,
"num_tokens": 6446336.0,
"step": 2930
},
{
"entropy": 5.867930555343628,
"epoch": 0.2912862246923382,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995827415686071,
"loss": 5.8551,
"mean_token_accuracy": 0.12857231721282006,
"num_tokens": 6457655.0,
"step": 2935
},
{
"entropy": 5.871946573257446,
"epoch": 0.29178245335450576,
"grad_norm": 1.0,
"learning_rate": 0.000499580581961813,
"loss": 5.7452,
"mean_token_accuracy": 0.13132286220788955,
"num_tokens": 6469627.0,
"step": 2940
},
{
"entropy": 5.811709070205689,
"epoch": 0.2922786820166733,
"grad_norm": 0.87890625,
"learning_rate": 0.0004995784167859043,
"loss": 5.854,
"mean_token_accuracy": 0.12665076926350594,
"num_tokens": 6480351.0,
"step": 2945
},
{
"entropy": 5.942422008514404,
"epoch": 0.2927749106788408,
"grad_norm": 0.953125,
"learning_rate": 0.0004995762460409349,
"loss": 5.8867,
"mean_token_accuracy": 0.13129101172089577,
"num_tokens": 6491651.0,
"step": 2950
},
{
"entropy": 5.803939962387085,
"epoch": 0.29327113934100835,
"grad_norm": 0.89453125,
"learning_rate": 0.0004995740697269584,
"loss": 5.7101,
"mean_token_accuracy": 0.13372641056776047,
"num_tokens": 6502531.0,
"step": 2955
},
{
"entropy": 5.742104434967041,
"epoch": 0.2937673680031759,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995718878440289,
"loss": 5.7394,
"mean_token_accuracy": 0.13551529124379158,
"num_tokens": 6511938.0,
"step": 2960
},
{
"entropy": 5.916677808761596,
"epoch": 0.2942635966653434,
"grad_norm": 1.09375,
"learning_rate": 0.0004995697003922006,
"loss": 5.881,
"mean_token_accuracy": 0.12271229401230813,
"num_tokens": 6522775.0,
"step": 2965
},
{
"entropy": 5.853166818618774,
"epoch": 0.29475982532751094,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995675073715274,
"loss": 5.7017,
"mean_token_accuracy": 0.12792166620492934,
"num_tokens": 6534660.0,
"step": 2970
},
{
"entropy": 5.804327535629272,
"epoch": 0.29525605398967847,
"grad_norm": 0.92578125,
"learning_rate": 0.0004995653087820641,
"loss": 5.8612,
"mean_token_accuracy": 0.12866377979516982,
"num_tokens": 6545673.0,
"step": 2975
},
{
"entropy": 5.902544212341309,
"epoch": 0.29575228265184594,
"grad_norm": 1.0078125,
"learning_rate": 0.000499563104623865,
"loss": 5.7985,
"mean_token_accuracy": 0.13151154071092605,
"num_tokens": 6556912.0,
"step": 2980
},
{
"entropy": 5.905428886413574,
"epoch": 0.2962485113140135,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995608948969849,
"loss": 5.9056,
"mean_token_accuracy": 0.12208881825208664,
"num_tokens": 6567532.0,
"step": 2985
},
{
"entropy": 5.880066728591919,
"epoch": 0.296744739976181,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995586796014784,
"loss": 5.8723,
"mean_token_accuracy": 0.12799570932984353,
"num_tokens": 6577830.0,
"step": 2990
},
{
"entropy": 5.8459772109985355,
"epoch": 0.29724096863834853,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995564587374005,
"loss": 5.7618,
"mean_token_accuracy": 0.1300479270517826,
"num_tokens": 6587254.0,
"step": 2995
},
{
"entropy": 5.901417541503906,
"epoch": 0.29773719730051607,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995542323048063,
"loss": 5.9346,
"mean_token_accuracy": 0.12555868849158286,
"num_tokens": 6598294.0,
"step": 3000
},
{
"epoch": 0.29773719730051607,
"eval_entropy": 5.737077051513012,
"eval_loss": 5.8997344970703125,
"eval_mean_token_accuracy": 0.1301897406935635,
"eval_num_tokens": 6598294.0,
"eval_runtime": 26.6497,
"eval_samples_per_second": 1263.389,
"eval_steps_per_second": 157.938,
"step": 3000
},
{
"entropy": 5.908546400070191,
"epoch": 0.2982334259626836,
"grad_norm": 0.9453125,
"learning_rate": 0.000499552000303751,
"loss": 5.8254,
"mean_token_accuracy": 0.13424069806933403,
"num_tokens": 6608772.0,
"step": 3005
},
{
"entropy": 5.863250255584717,
"epoch": 0.2987296546248511,
"grad_norm": 0.90625,
"learning_rate": 0.00049954976273429,
"loss": 5.8225,
"mean_token_accuracy": 0.13384647592902182,
"num_tokens": 6619973.0,
"step": 3010
},
{
"entropy": 5.848355102539062,
"epoch": 0.29922588328701866,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995475195964787,
"loss": 5.856,
"mean_token_accuracy": 0.1277636930346489,
"num_tokens": 6631172.0,
"step": 3015
},
{
"entropy": 5.830797290802002,
"epoch": 0.2997221119491862,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995452708903727,
"loss": 5.8381,
"mean_token_accuracy": 0.1252981200814247,
"num_tokens": 6641343.0,
"step": 3020
},
{
"entropy": 5.853375816345215,
"epoch": 0.3002183406113537,
"grad_norm": 0.890625,
"learning_rate": 0.000499543016616028,
"loss": 5.7881,
"mean_token_accuracy": 0.12906054332852362,
"num_tokens": 6652892.0,
"step": 3025
},
{
"entropy": 5.813239049911499,
"epoch": 0.30071456927352125,
"grad_norm": 0.86328125,
"learning_rate": 0.0004995407567735002,
"loss": 5.8103,
"mean_token_accuracy": 0.1291997916996479,
"num_tokens": 6664177.0,
"step": 3030
},
{
"entropy": 5.8448244571685795,
"epoch": 0.3012107979356888,
"grad_norm": 1.0390625,
"learning_rate": 0.0004995384913628456,
"loss": 5.7806,
"mean_token_accuracy": 0.12814074456691743,
"num_tokens": 6674402.0,
"step": 3035
},
{
"entropy": 5.886237621307373,
"epoch": 0.3017070265978563,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995362203841202,
"loss": 5.9118,
"mean_token_accuracy": 0.12575332671403885,
"num_tokens": 6684886.0,
"step": 3040
},
{
"entropy": 5.884566354751587,
"epoch": 0.30220325526002384,
"grad_norm": 0.94921875,
"learning_rate": 0.0004995339438373804,
"loss": 5.7383,
"mean_token_accuracy": 0.1311859056353569,
"num_tokens": 6694774.0,
"step": 3045
},
{
"entropy": 5.778323936462402,
"epoch": 0.30269948392219137,
"grad_norm": 0.98046875,
"learning_rate": 0.0004995316617226825,
"loss": 5.8032,
"mean_token_accuracy": 0.13333578556776046,
"num_tokens": 6705346.0,
"step": 3050
},
{
"entropy": 5.849145555496216,
"epoch": 0.30319571258435885,
"grad_norm": 0.91015625,
"learning_rate": 0.0004995293740400833,
"loss": 5.7601,
"mean_token_accuracy": 0.1354569710791111,
"num_tokens": 6715579.0,
"step": 3055
},
{
"entropy": 5.777801465988159,
"epoch": 0.3036919412465264,
"grad_norm": 0.9296875,
"learning_rate": 0.0004995270807896395,
"loss": 5.7179,
"mean_token_accuracy": 0.1327461652457714,
"num_tokens": 6726382.0,
"step": 3060
},
{
"entropy": 5.833434057235718,
"epoch": 0.3041881699086939,
"grad_norm": 1.0234375,
"learning_rate": 0.0004995247819714079,
"loss": 5.7975,
"mean_token_accuracy": 0.12714954614639282,
"num_tokens": 6737245.0,
"step": 3065
},
{
"entropy": 5.905564641952514,
"epoch": 0.30468439857086144,
"grad_norm": 0.9765625,
"learning_rate": 0.0004995224775854454,
"loss": 6.0002,
"mean_token_accuracy": 0.12385166138410568,
"num_tokens": 6749113.0,
"step": 3070
},
{
"entropy": 5.978738498687744,
"epoch": 0.30518062723302897,
"grad_norm": 0.8828125,
"learning_rate": 0.0004995201676318093,
"loss": 5.8518,
"mean_token_accuracy": 0.13237554132938384,
"num_tokens": 6761987.0,
"step": 3075
},
{
"entropy": 5.816679763793945,
"epoch": 0.3056768558951965,
"grad_norm": 0.96484375,
"learning_rate": 0.0004995178521105568,
"loss": 5.7908,
"mean_token_accuracy": 0.1237344652414322,
"num_tokens": 6771279.0,
"step": 3080
},
{
"entropy": 5.852257347106933,
"epoch": 0.30617308455736403,
"grad_norm": 1.0078125,
"learning_rate": 0.0004995155310217454,
"loss": 5.7286,
"mean_token_accuracy": 0.13410176187753678,
"num_tokens": 6781038.0,
"step": 3085
},
{
"entropy": 5.75889139175415,
"epoch": 0.30666931321953156,
"grad_norm": 0.93359375,
"learning_rate": 0.0004995132043654327,
"loss": 5.8619,
"mean_token_accuracy": 0.12660413309931756,
"num_tokens": 6792686.0,
"step": 3090
},
{
"entropy": 5.84913330078125,
"epoch": 0.3071655418816991,
"grad_norm": 0.88671875,
"learning_rate": 0.0004995108721416763,
"loss": 5.8304,
"mean_token_accuracy": 0.1372841201722622,
"num_tokens": 6804414.0,
"step": 3095
},
{
"entropy": 5.846239280700684,
"epoch": 0.3076617705438666,
"grad_norm": 0.90234375,
"learning_rate": 0.0004995085343505339,
"loss": 5.7936,
"mean_token_accuracy": 0.13219277635216714,
"num_tokens": 6816180.0,
"step": 3100
},
{
"entropy": 5.857756471633911,
"epoch": 0.30815799920603415,
"grad_norm": 0.95703125,
"learning_rate": 0.0004995061909920638,
"loss": 5.7985,
"mean_token_accuracy": 0.12769369333982467,
"num_tokens": 6827114.0,
"step": 3105
},
{
"entropy": 5.881041431427002,
"epoch": 0.3086542278682017,
"grad_norm": 0.96875,
"learning_rate": 0.0004995038420663238,
"loss": 5.8539,
"mean_token_accuracy": 0.13202394172549248,
"num_tokens": 6836953.0,
"step": 3110
},
{
"entropy": 5.872609329223633,
"epoch": 0.3091504565303692,
"grad_norm": 1.046875,
"learning_rate": 0.0004995014875733724,
"loss": 5.9082,
"mean_token_accuracy": 0.12458430081605912,
"num_tokens": 6847963.0,
"step": 3115
},
{
"entropy": 5.93118109703064,
"epoch": 0.30964668519253674,
"grad_norm": 0.99609375,
"learning_rate": 0.0004994991275132678,
"loss": 5.9155,
"mean_token_accuracy": 0.12210882604122161,
"num_tokens": 6858264.0,
"step": 3120
},
{
"entropy": 5.832281446456909,
"epoch": 0.31014291385470427,
"grad_norm": 0.96484375,
"learning_rate": 0.0004994967618860685,
"loss": 5.7974,
"mean_token_accuracy": 0.13236498534679414,
"num_tokens": 6868861.0,
"step": 3125
},
{
"entropy": 5.914390993118286,
"epoch": 0.31063914251687175,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994943906918333,
"loss": 5.8243,
"mean_token_accuracy": 0.13056366965174676,
"num_tokens": 6879797.0,
"step": 3130
},
{
"entropy": 5.8452612400054935,
"epoch": 0.3111353711790393,
"grad_norm": 0.921875,
"learning_rate": 0.0004994920139306211,
"loss": 5.7794,
"mean_token_accuracy": 0.13094358667731285,
"num_tokens": 6891323.0,
"step": 3135
},
{
"entropy": 5.814275884628296,
"epoch": 0.3116315998412068,
"grad_norm": 1.0625,
"learning_rate": 0.0004994896316024906,
"loss": 5.8629,
"mean_token_accuracy": 0.12925291806459427,
"num_tokens": 6903345.0,
"step": 3140
},
{
"entropy": 5.829950189590454,
"epoch": 0.31212782850337434,
"grad_norm": 1.0,
"learning_rate": 0.000499487243707501,
"loss": 5.787,
"mean_token_accuracy": 0.12979979440569878,
"num_tokens": 6914342.0,
"step": 3145
},
{
"entropy": 5.8646656513214115,
"epoch": 0.31262405716554187,
"grad_norm": 0.86328125,
"learning_rate": 0.0004994848502457114,
"loss": 5.8328,
"mean_token_accuracy": 0.12842394262552262,
"num_tokens": 6926691.0,
"step": 3150
},
{
"entropy": 5.829729843139648,
"epoch": 0.3131202858277094,
"grad_norm": 0.9375,
"learning_rate": 0.0004994824512171812,
"loss": 5.7117,
"mean_token_accuracy": 0.13195387348532678,
"num_tokens": 6937841.0,
"step": 3155
},
{
"entropy": 5.766626453399658,
"epoch": 0.31361651448987693,
"grad_norm": 1.015625,
"learning_rate": 0.0004994800466219701,
"loss": 5.8114,
"mean_token_accuracy": 0.12932611480355263,
"num_tokens": 6948601.0,
"step": 3160
},
{
"entropy": 5.855544185638427,
"epoch": 0.31411274315204446,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994776364601374,
"loss": 5.7358,
"mean_token_accuracy": 0.13258148804306985,
"num_tokens": 6959775.0,
"step": 3165
},
{
"entropy": 5.78338475227356,
"epoch": 0.314608971814212,
"grad_norm": 0.98828125,
"learning_rate": 0.0004994752207317431,
"loss": 5.7707,
"mean_token_accuracy": 0.13972898945212364,
"num_tokens": 6969202.0,
"step": 3170
},
{
"entropy": 5.805487060546875,
"epoch": 0.3151052004763795,
"grad_norm": 0.95703125,
"learning_rate": 0.000499472799436847,
"loss": 5.7517,
"mean_token_accuracy": 0.13298066705465317,
"num_tokens": 6978633.0,
"step": 3175
},
{
"entropy": 5.894915008544922,
"epoch": 0.31560142913854705,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994703725755092,
"loss": 5.8186,
"mean_token_accuracy": 0.12706844583153726,
"num_tokens": 6989638.0,
"step": 3180
},
{
"entropy": 5.812424945831299,
"epoch": 0.3160976578007146,
"grad_norm": 1.046875,
"learning_rate": 0.0004994679401477899,
"loss": 5.7826,
"mean_token_accuracy": 0.12963953241705894,
"num_tokens": 7001606.0,
"step": 3185
},
{
"entropy": 5.813755512237549,
"epoch": 0.3165938864628821,
"grad_norm": 1.0234375,
"learning_rate": 0.0004994655021537492,
"loss": 5.7604,
"mean_token_accuracy": 0.1284653961658478,
"num_tokens": 7012915.0,
"step": 3190
},
{
"entropy": 5.80635085105896,
"epoch": 0.31709011512504964,
"grad_norm": 0.9453125,
"learning_rate": 0.0004994630585934478,
"loss": 5.8141,
"mean_token_accuracy": 0.12510425895452498,
"num_tokens": 7023191.0,
"step": 3195
},
{
"entropy": 5.844677400588989,
"epoch": 0.3175863437872172,
"grad_norm": 0.9609375,
"learning_rate": 0.0004994606094669463,
"loss": 5.708,
"mean_token_accuracy": 0.13605407029390335,
"num_tokens": 7032506.0,
"step": 3200
},
{
"entropy": 5.819374847412109,
"epoch": 0.3180825724493847,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994581547743053,
"loss": 5.8462,
"mean_token_accuracy": 0.12448615580797195,
"num_tokens": 7046098.0,
"step": 3205
},
{
"entropy": 5.952435874938965,
"epoch": 0.3185788011115522,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994556945155857,
"loss": 5.8499,
"mean_token_accuracy": 0.13355428129434585,
"num_tokens": 7057434.0,
"step": 3210
},
{
"entropy": 5.792888545989991,
"epoch": 0.3190750297737197,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994532286908486,
"loss": 5.7786,
"mean_token_accuracy": 0.12661978080868722,
"num_tokens": 7067674.0,
"step": 3215
},
{
"entropy": 5.934251117706299,
"epoch": 0.31957125843588724,
"grad_norm": 1.1171875,
"learning_rate": 0.000499450757300155,
"loss": 5.8491,
"mean_token_accuracy": 0.12803201824426652,
"num_tokens": 7077898.0,
"step": 3220
},
{
"entropy": 5.873095369338989,
"epoch": 0.32006748709805477,
"grad_norm": 1.0703125,
"learning_rate": 0.0004994482803435664,
"loss": 5.8114,
"mean_token_accuracy": 0.12614069283008575,
"num_tokens": 7088566.0,
"step": 3225
},
{
"entropy": 5.777619647979736,
"epoch": 0.3205637157602223,
"grad_norm": 0.90234375,
"learning_rate": 0.000499445797821144,
"loss": 5.8038,
"mean_token_accuracy": 0.13038980588316917,
"num_tokens": 7099533.0,
"step": 3230
},
{
"entropy": 5.901455020904541,
"epoch": 0.32105994442238983,
"grad_norm": 0.94140625,
"learning_rate": 0.0004994433097329494,
"loss": 5.8054,
"mean_token_accuracy": 0.12909889370203018,
"num_tokens": 7112254.0,
"step": 3235
},
{
"entropy": 5.893978786468506,
"epoch": 0.32155617308455736,
"grad_norm": 0.890625,
"learning_rate": 0.0004994408160790444,
"loss": 5.8276,
"mean_token_accuracy": 0.12560412287712097,
"num_tokens": 7124704.0,
"step": 3240
},
{
"entropy": 5.970049476623535,
"epoch": 0.3220524017467249,
"grad_norm": 0.87109375,
"learning_rate": 0.0004994383168594907,
"loss": 5.9963,
"mean_token_accuracy": 0.11985118389129638,
"num_tokens": 7137741.0,
"step": 3245
},
{
"entropy": 5.848871469497681,
"epoch": 0.3225486304088924,
"grad_norm": 1.015625,
"learning_rate": 0.0004994358120743506,
"loss": 5.7898,
"mean_token_accuracy": 0.12539106085896493,
"num_tokens": 7147888.0,
"step": 3250
},
{
"entropy": 5.752130508422852,
"epoch": 0.32304485907105995,
"grad_norm": 0.93359375,
"learning_rate": 0.0004994333017236858,
"loss": 5.7357,
"mean_token_accuracy": 0.13781259804964066,
"num_tokens": 7158950.0,
"step": 3255
},
{
"entropy": 5.861024618148804,
"epoch": 0.3235410877332275,
"grad_norm": 0.9765625,
"learning_rate": 0.0004994307858075587,
"loss": 5.8815,
"mean_token_accuracy": 0.12370717674493789,
"num_tokens": 7170156.0,
"step": 3260
},
{
"entropy": 5.8257866382598875,
"epoch": 0.324037316395395,
"grad_norm": 1.0078125,
"learning_rate": 0.0004994282643260319,
"loss": 5.7285,
"mean_token_accuracy": 0.1357402391731739,
"num_tokens": 7180576.0,
"step": 3265
},
{
"entropy": 5.768695497512818,
"epoch": 0.32453354505756254,
"grad_norm": 0.93359375,
"learning_rate": 0.0004994257372791676,
"loss": 5.7863,
"mean_token_accuracy": 0.13116331100463868,
"num_tokens": 7192380.0,
"step": 3270
},
{
"entropy": 5.890905237197876,
"epoch": 0.3250297737197301,
"grad_norm": 0.875,
"learning_rate": 0.0004994232046670286,
"loss": 5.8419,
"mean_token_accuracy": 0.1313177950680256,
"num_tokens": 7203898.0,
"step": 3275
},
{
"entropy": 5.896629285812378,
"epoch": 0.3255260023818976,
"grad_norm": 0.9375,
"learning_rate": 0.0004994206664896778,
"loss": 5.8155,
"mean_token_accuracy": 0.1287803702056408,
"num_tokens": 7215241.0,
"step": 3280
},
{
"entropy": 5.85573182106018,
"epoch": 0.3260222310440651,
"grad_norm": 0.98046875,
"learning_rate": 0.0004994181227471779,
"loss": 5.7653,
"mean_token_accuracy": 0.1314986191689968,
"num_tokens": 7226039.0,
"step": 3285
},
{
"entropy": 5.7266041278839115,
"epoch": 0.3265184597062326,
"grad_norm": 0.94921875,
"learning_rate": 0.0004994155734395923,
"loss": 5.79,
"mean_token_accuracy": 0.13011762425303458,
"num_tokens": 7237148.0,
"step": 3290
},
{
"entropy": 5.825259494781494,
"epoch": 0.32701468836840014,
"grad_norm": 0.9765625,
"learning_rate": 0.000499413018566984,
"loss": 5.7883,
"mean_token_accuracy": 0.1279461018741131,
"num_tokens": 7249284.0,
"step": 3295
},
{
"entropy": 5.842443895339966,
"epoch": 0.32751091703056767,
"grad_norm": 0.9375,
"learning_rate": 0.0004994104581294163,
"loss": 5.7961,
"mean_token_accuracy": 0.12790419161319733,
"num_tokens": 7259967.0,
"step": 3300
},
{
"entropy": 5.762500810623169,
"epoch": 0.3280071456927352,
"grad_norm": 0.97265625,
"learning_rate": 0.0004994078921269529,
"loss": 5.7547,
"mean_token_accuracy": 0.13474935069680213,
"num_tokens": 7271145.0,
"step": 3305
},
{
"entropy": 5.918145751953125,
"epoch": 0.32850337435490273,
"grad_norm": 0.91796875,
"learning_rate": 0.0004994053205596573,
"loss": 5.7698,
"mean_token_accuracy": 0.1367741271853447,
"num_tokens": 7281872.0,
"step": 3310
},
{
"entropy": 5.750212907791138,
"epoch": 0.32899960301707026,
"grad_norm": 0.8828125,
"learning_rate": 0.0004994027434275934,
"loss": 5.788,
"mean_token_accuracy": 0.12981380969285966,
"num_tokens": 7293330.0,
"step": 3315
},
{
"entropy": 5.967539882659912,
"epoch": 0.3294958316792378,
"grad_norm": 0.828125,
"learning_rate": 0.0004994001607308249,
"loss": 5.8512,
"mean_token_accuracy": 0.12412105202674865,
"num_tokens": 7304187.0,
"step": 3320
},
{
"entropy": 5.885866308212281,
"epoch": 0.3299920603414053,
"grad_norm": 0.86328125,
"learning_rate": 0.000499397572469416,
"loss": 5.7521,
"mean_token_accuracy": 0.13047175034880637,
"num_tokens": 7314756.0,
"step": 3325
},
{
"entropy": 5.751908969879151,
"epoch": 0.33048828900357285,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993949786434308,
"loss": 5.7817,
"mean_token_accuracy": 0.1319439634680748,
"num_tokens": 7325711.0,
"step": 3330
},
{
"entropy": 5.782564353942871,
"epoch": 0.3309845176657404,
"grad_norm": 1.0,
"learning_rate": 0.0004993923792529337,
"loss": 5.7167,
"mean_token_accuracy": 0.13527547493577002,
"num_tokens": 7335639.0,
"step": 3335
},
{
"entropy": 5.883091735839844,
"epoch": 0.3314807463279079,
"grad_norm": 0.921875,
"learning_rate": 0.0004993897742979892,
"loss": 5.8349,
"mean_token_accuracy": 0.1300416111946106,
"num_tokens": 7346621.0,
"step": 3340
},
{
"entropy": 5.872514009475708,
"epoch": 0.33197697499007545,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993871637786616,
"loss": 5.7901,
"mean_token_accuracy": 0.12466496452689171,
"num_tokens": 7358950.0,
"step": 3345
},
{
"entropy": 5.732246828079224,
"epoch": 0.332473203652243,
"grad_norm": 0.94140625,
"learning_rate": 0.000499384547695016,
"loss": 5.6516,
"mean_token_accuracy": 0.13917058929800988,
"num_tokens": 7369629.0,
"step": 3350
},
{
"entropy": 5.782614231109619,
"epoch": 0.3329694323144105,
"grad_norm": 1.0078125,
"learning_rate": 0.000499381926047117,
"loss": 5.7491,
"mean_token_accuracy": 0.12965478897094726,
"num_tokens": 7379853.0,
"step": 3355
},
{
"entropy": 5.828644323348999,
"epoch": 0.333465660976578,
"grad_norm": 0.98828125,
"learning_rate": 0.00049937929883503,
"loss": 5.7902,
"mean_token_accuracy": 0.12588309049606322,
"num_tokens": 7391323.0,
"step": 3360
},
{
"entropy": 5.824267625808716,
"epoch": 0.3339618896387455,
"grad_norm": 1.1953125,
"learning_rate": 0.0004993766660588196,
"loss": 5.7624,
"mean_token_accuracy": 0.1319116823375225,
"num_tokens": 7402201.0,
"step": 3365
},
{
"entropy": 5.922031736373901,
"epoch": 0.33445811830091304,
"grad_norm": 0.953125,
"learning_rate": 0.0004993740277185515,
"loss": 5.9897,
"mean_token_accuracy": 0.12059160396456718,
"num_tokens": 7413783.0,
"step": 3370
},
{
"entropy": 5.8004532814025875,
"epoch": 0.3349543469630806,
"grad_norm": 0.953125,
"learning_rate": 0.000499371383814291,
"loss": 5.7574,
"mean_token_accuracy": 0.13145501986145974,
"num_tokens": 7425909.0,
"step": 3375
},
{
"entropy": 5.759319877624511,
"epoch": 0.3354505756252481,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993687343461036,
"loss": 5.6652,
"mean_token_accuracy": 0.13739446327090263,
"num_tokens": 7436346.0,
"step": 3380
},
{
"entropy": 5.760404396057129,
"epoch": 0.33594680428741563,
"grad_norm": 1.0,
"learning_rate": 0.0004993660793140551,
"loss": 5.7156,
"mean_token_accuracy": 0.13563819900155066,
"num_tokens": 7446596.0,
"step": 3385
},
{
"entropy": 5.784324502944946,
"epoch": 0.33644303294958317,
"grad_norm": 1.046875,
"learning_rate": 0.0004993634187182113,
"loss": 5.806,
"mean_token_accuracy": 0.1257621392607689,
"num_tokens": 7457697.0,
"step": 3390
},
{
"entropy": 5.917649269104004,
"epoch": 0.3369392616117507,
"grad_norm": 0.9140625,
"learning_rate": 0.0004993607525586381,
"loss": 5.9285,
"mean_token_accuracy": 0.1271599940955639,
"num_tokens": 7468541.0,
"step": 3395
},
{
"entropy": 5.8661158084869385,
"epoch": 0.3374354902739182,
"grad_norm": 0.91015625,
"learning_rate": 0.0004993580808354018,
"loss": 5.8033,
"mean_token_accuracy": 0.13219190388917923,
"num_tokens": 7479781.0,
"step": 3400
},
{
"entropy": 5.800511312484741,
"epoch": 0.33793171893608576,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993554035485685,
"loss": 5.7042,
"mean_token_accuracy": 0.12884424477815629,
"num_tokens": 7490722.0,
"step": 3405
},
{
"entropy": 5.8494329929351805,
"epoch": 0.3384279475982533,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993527206982047,
"loss": 5.8264,
"mean_token_accuracy": 0.12879022359848022,
"num_tokens": 7502293.0,
"step": 3410
},
{
"entropy": 5.8656847953796385,
"epoch": 0.3389241762604208,
"grad_norm": 0.84375,
"learning_rate": 0.0004993500322843767,
"loss": 5.7507,
"mean_token_accuracy": 0.13226759135723115,
"num_tokens": 7513846.0,
"step": 3415
},
{
"entropy": 5.711007070541382,
"epoch": 0.33942040492258835,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993473383071514,
"loss": 5.7865,
"mean_token_accuracy": 0.13022084534168243,
"num_tokens": 7524062.0,
"step": 3420
},
{
"entropy": 5.862089157104492,
"epoch": 0.3399166335847559,
"grad_norm": 0.90625,
"learning_rate": 0.0004993446387665956,
"loss": 5.8329,
"mean_token_accuracy": 0.13513429313898087,
"num_tokens": 7535533.0,
"step": 3425
},
{
"entropy": 5.865432071685791,
"epoch": 0.3404128622469234,
"grad_norm": 0.9296875,
"learning_rate": 0.0004993419336627761,
"loss": 5.7147,
"mean_token_accuracy": 0.13362642973661423,
"num_tokens": 7545951.0,
"step": 3430
},
{
"entropy": 5.711605072021484,
"epoch": 0.3409090909090909,
"grad_norm": 0.99609375,
"learning_rate": 0.0004993392229957602,
"loss": 5.671,
"mean_token_accuracy": 0.1315312497317791,
"num_tokens": 7557320.0,
"step": 3435
},
{
"entropy": 5.690997362136841,
"epoch": 0.3414053195712584,
"grad_norm": 1.0,
"learning_rate": 0.0004993365067656148,
"loss": 5.7777,
"mean_token_accuracy": 0.1280876100063324,
"num_tokens": 7568078.0,
"step": 3440
},
{
"entropy": 5.824246025085449,
"epoch": 0.34190154823342594,
"grad_norm": 0.890625,
"learning_rate": 0.0004993337849724076,
"loss": 5.7856,
"mean_token_accuracy": 0.1305572994053364,
"num_tokens": 7578940.0,
"step": 3445
},
{
"entropy": 5.8472373485565186,
"epoch": 0.3423977768955935,
"grad_norm": 1.0546875,
"learning_rate": 0.0004993310576162059,
"loss": 5.7803,
"mean_token_accuracy": 0.13081290274858476,
"num_tokens": 7589864.0,
"step": 3450
},
{
"entropy": 5.871004867553711,
"epoch": 0.342894005557761,
"grad_norm": 1.0390625,
"learning_rate": 0.0004993283246970772,
"loss": 5.725,
"mean_token_accuracy": 0.12843016013503075,
"num_tokens": 7599795.0,
"step": 3455
},
{
"entropy": 5.8094720363616945,
"epoch": 0.34339023421992854,
"grad_norm": 0.96484375,
"learning_rate": 0.0004993255862150896,
"loss": 5.7121,
"mean_token_accuracy": 0.12965178340673447,
"num_tokens": 7610056.0,
"step": 3460
},
{
"entropy": 5.721919441223145,
"epoch": 0.34388646288209607,
"grad_norm": 0.9609375,
"learning_rate": 0.0004993228421703108,
"loss": 5.7053,
"mean_token_accuracy": 0.1275747649371624,
"num_tokens": 7620218.0,
"step": 3465
},
{
"entropy": 5.75278491973877,
"epoch": 0.3443826915442636,
"grad_norm": 0.91015625,
"learning_rate": 0.0004993200925628088,
"loss": 5.8105,
"mean_token_accuracy": 0.1322753168642521,
"num_tokens": 7632781.0,
"step": 3470
},
{
"entropy": 5.820105791091919,
"epoch": 0.3448789202064311,
"grad_norm": 0.98046875,
"learning_rate": 0.000499317337392652,
"loss": 5.7817,
"mean_token_accuracy": 0.12995077893137932,
"num_tokens": 7643845.0,
"step": 3475
},
{
"entropy": 5.8669881343841555,
"epoch": 0.34537514886859866,
"grad_norm": 0.9375,
"learning_rate": 0.0004993145766599085,
"loss": 5.8008,
"mean_token_accuracy": 0.13089126273989676,
"num_tokens": 7654583.0,
"step": 3480
},
{
"entropy": 5.7667046070098875,
"epoch": 0.3458713775307662,
"grad_norm": 1.0078125,
"learning_rate": 0.0004993118103646469,
"loss": 5.6611,
"mean_token_accuracy": 0.1325427420437336,
"num_tokens": 7665714.0,
"step": 3485
},
{
"entropy": 5.769314384460449,
"epoch": 0.3463676061929337,
"grad_norm": 0.95703125,
"learning_rate": 0.0004993090385069357,
"loss": 5.7067,
"mean_token_accuracy": 0.1339074306190014,
"num_tokens": 7676011.0,
"step": 3490
},
{
"entropy": 5.743018531799317,
"epoch": 0.34686383485510125,
"grad_norm": 0.91796875,
"learning_rate": 0.0004993062610868437,
"loss": 5.763,
"mean_token_accuracy": 0.1294798120856285,
"num_tokens": 7686927.0,
"step": 3495
},
{
"entropy": 5.764451122283935,
"epoch": 0.3473600635172688,
"grad_norm": 0.88671875,
"learning_rate": 0.0004993034781044398,
"loss": 5.7245,
"mean_token_accuracy": 0.1329188510775566,
"num_tokens": 7698419.0,
"step": 3500
},
{
"entropy": 5.829875087738037,
"epoch": 0.3478562921794363,
"grad_norm": 0.94921875,
"learning_rate": 0.000499300689559793,
"loss": 5.6837,
"mean_token_accuracy": 0.13597684502601623,
"num_tokens": 7709695.0,
"step": 3505
},
{
"entropy": 5.726664352416992,
"epoch": 0.3483525208416038,
"grad_norm": 1.4921875,
"learning_rate": 0.0004992978954529724,
"loss": 5.6869,
"mean_token_accuracy": 0.136630592495203,
"num_tokens": 7721309.0,
"step": 3510
},
{
"entropy": 5.803768014907837,
"epoch": 0.3488487495037713,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992950957840473,
"loss": 5.7554,
"mean_token_accuracy": 0.132266054302454,
"num_tokens": 7732074.0,
"step": 3515
},
{
"entropy": 5.816020488739014,
"epoch": 0.34934497816593885,
"grad_norm": 1.0390625,
"learning_rate": 0.000499292290553087,
"loss": 5.8194,
"mean_token_accuracy": 0.13066235557198524,
"num_tokens": 7742869.0,
"step": 3520
},
{
"entropy": 5.7433083057403564,
"epoch": 0.3498412068281064,
"grad_norm": 0.953125,
"learning_rate": 0.0004992894797601614,
"loss": 5.7274,
"mean_token_accuracy": 0.13449874967336656,
"num_tokens": 7754895.0,
"step": 3525
},
{
"entropy": 5.830230379104615,
"epoch": 0.3503374354902739,
"grad_norm": 0.96484375,
"learning_rate": 0.00049928666340534,
"loss": 5.7012,
"mean_token_accuracy": 0.12928271517157555,
"num_tokens": 7765262.0,
"step": 3530
},
{
"entropy": 5.747873067855835,
"epoch": 0.35083366415244144,
"grad_norm": 1.0,
"learning_rate": 0.0004992838414886925,
"loss": 5.7814,
"mean_token_accuracy": 0.12928831875324248,
"num_tokens": 7777348.0,
"step": 3535
},
{
"entropy": 5.870401191711426,
"epoch": 0.35132989281460897,
"grad_norm": 0.93359375,
"learning_rate": 0.0004992810140102891,
"loss": 5.7629,
"mean_token_accuracy": 0.13162485882639885,
"num_tokens": 7788161.0,
"step": 3540
},
{
"entropy": 5.795786046981812,
"epoch": 0.3518261214767765,
"grad_norm": 0.953125,
"learning_rate": 0.0004992781809701999,
"loss": 5.7618,
"mean_token_accuracy": 0.12922549918293952,
"num_tokens": 7798629.0,
"step": 3545
},
{
"entropy": 5.7486223697662355,
"epoch": 0.35232235013894403,
"grad_norm": 0.92578125,
"learning_rate": 0.0004992753423684951,
"loss": 5.7428,
"mean_token_accuracy": 0.13399974927306174,
"num_tokens": 7810148.0,
"step": 3550
},
{
"entropy": 5.806520700454712,
"epoch": 0.35281857880111156,
"grad_norm": 1.0390625,
"learning_rate": 0.0004992724982052449,
"loss": 5.732,
"mean_token_accuracy": 0.13214099556207656,
"num_tokens": 7821667.0,
"step": 3555
},
{
"entropy": 5.731984186172485,
"epoch": 0.3533148074632791,
"grad_norm": 0.890625,
"learning_rate": 0.00049926964848052,
"loss": 5.7207,
"mean_token_accuracy": 0.13429917991161347,
"num_tokens": 7832184.0,
"step": 3560
},
{
"entropy": 5.8297947406768795,
"epoch": 0.3538110361254466,
"grad_norm": 0.94921875,
"learning_rate": 0.0004992667931943913,
"loss": 5.7387,
"mean_token_accuracy": 0.130974131077528,
"num_tokens": 7842983.0,
"step": 3565
},
{
"entropy": 5.766924142837524,
"epoch": 0.35430726478761415,
"grad_norm": 0.890625,
"learning_rate": 0.0004992639323469293,
"loss": 5.6723,
"mean_token_accuracy": 0.1407732456922531,
"num_tokens": 7854783.0,
"step": 3570
},
{
"entropy": 5.724614095687866,
"epoch": 0.3548034934497817,
"grad_norm": 0.98046875,
"learning_rate": 0.000499261065938205,
"loss": 5.695,
"mean_token_accuracy": 0.1421818196773529,
"num_tokens": 7864684.0,
"step": 3575
},
{
"entropy": 5.8322038650512695,
"epoch": 0.3552997221119492,
"grad_norm": 1.015625,
"learning_rate": 0.0004992581939682895,
"loss": 5.8576,
"mean_token_accuracy": 0.12228314802050591,
"num_tokens": 7876846.0,
"step": 3580
},
{
"entropy": 5.905322122573852,
"epoch": 0.3557959507741167,
"grad_norm": 0.9765625,
"learning_rate": 0.000499255316437254,
"loss": 5.7116,
"mean_token_accuracy": 0.1306779131293297,
"num_tokens": 7886826.0,
"step": 3585
},
{
"entropy": 5.699662113189698,
"epoch": 0.3562921794362842,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992524333451699,
"loss": 5.7923,
"mean_token_accuracy": 0.1284032307565212,
"num_tokens": 7898941.0,
"step": 3590
},
{
"entropy": 5.777550268173218,
"epoch": 0.35678840809845175,
"grad_norm": 1.03125,
"learning_rate": 0.0004992495446921087,
"loss": 5.6991,
"mean_token_accuracy": 0.1283227153122425,
"num_tokens": 7909916.0,
"step": 3595
},
{
"entropy": 5.718798351287842,
"epoch": 0.3572846367606193,
"grad_norm": 0.875,
"learning_rate": 0.0004992466504781419,
"loss": 5.723,
"mean_token_accuracy": 0.13592231124639512,
"num_tokens": 7922137.0,
"step": 3600
},
{
"entropy": 5.725904083251953,
"epoch": 0.3577808654227868,
"grad_norm": 0.9921875,
"learning_rate": 0.0004992437507033415,
"loss": 5.6544,
"mean_token_accuracy": 0.13926874473690987,
"num_tokens": 7932940.0,
"step": 3605
},
{
"entropy": 5.8571995258331295,
"epoch": 0.35827709408495434,
"grad_norm": 0.89453125,
"learning_rate": 0.0004992408453677792,
"loss": 5.8249,
"mean_token_accuracy": 0.12425437346100807,
"num_tokens": 7943735.0,
"step": 3610
},
{
"entropy": 5.761858224868774,
"epoch": 0.35877332274712187,
"grad_norm": 1.015625,
"learning_rate": 0.0004992379344715273,
"loss": 5.7084,
"mean_token_accuracy": 0.13300363793969155,
"num_tokens": 7955323.0,
"step": 3615
},
{
"entropy": 5.747606563568115,
"epoch": 0.3592695514092894,
"grad_norm": 0.98046875,
"learning_rate": 0.0004992350180146577,
"loss": 5.656,
"mean_token_accuracy": 0.13175439611077308,
"num_tokens": 7966073.0,
"step": 3620
},
{
"entropy": 5.769119215011597,
"epoch": 0.35976578007145693,
"grad_norm": 1.03125,
"learning_rate": 0.0004992320959972428,
"loss": 5.8377,
"mean_token_accuracy": 0.12840200886130332,
"num_tokens": 7977027.0,
"step": 3625
},
{
"entropy": 5.802391815185547,
"epoch": 0.36026200873362446,
"grad_norm": 0.953125,
"learning_rate": 0.0004992291684193551,
"loss": 5.6963,
"mean_token_accuracy": 0.1378728047013283,
"num_tokens": 7987368.0,
"step": 3630
},
{
"entropy": 5.810581016540527,
"epoch": 0.360758237395792,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992262352810672,
"loss": 5.7589,
"mean_token_accuracy": 0.13187605291604995,
"num_tokens": 8000060.0,
"step": 3635
},
{
"entropy": 5.778606843948364,
"epoch": 0.3612544660579595,
"grad_norm": 0.9453125,
"learning_rate": 0.0004992232965824518,
"loss": 5.8151,
"mean_token_accuracy": 0.1257768027484417,
"num_tokens": 8011388.0,
"step": 3640
},
{
"entropy": 5.7404289722442625,
"epoch": 0.36175069472012705,
"grad_norm": 0.96484375,
"learning_rate": 0.000499220352323582,
"loss": 5.6808,
"mean_token_accuracy": 0.13884039595723152,
"num_tokens": 8021682.0,
"step": 3645
},
{
"entropy": 5.798848342895508,
"epoch": 0.3622469233822946,
"grad_norm": 0.94140625,
"learning_rate": 0.0004992174025045304,
"loss": 5.6941,
"mean_token_accuracy": 0.13487540781497956,
"num_tokens": 8032799.0,
"step": 3650
},
{
"entropy": 5.706443929672242,
"epoch": 0.3627431520444621,
"grad_norm": 1.046875,
"learning_rate": 0.0004992144471253704,
"loss": 5.5725,
"mean_token_accuracy": 0.14214823096990586,
"num_tokens": 8042742.0,
"step": 3655
},
{
"entropy": 5.763505363464356,
"epoch": 0.3632393807066296,
"grad_norm": 0.8984375,
"learning_rate": 0.0004992114861861753,
"loss": 5.7613,
"mean_token_accuracy": 0.13013699501752854,
"num_tokens": 8054068.0,
"step": 3660
},
{
"entropy": 5.813911485671997,
"epoch": 0.3637356093687971,
"grad_norm": 0.90625,
"learning_rate": 0.0004992085196870184,
"loss": 5.7419,
"mean_token_accuracy": 0.1301536478102207,
"num_tokens": 8065286.0,
"step": 3665
},
{
"entropy": 5.763180732727051,
"epoch": 0.36423183803096465,
"grad_norm": 0.9609375,
"learning_rate": 0.0004992055476279734,
"loss": 5.7993,
"mean_token_accuracy": 0.12314197868108749,
"num_tokens": 8077122.0,
"step": 3670
},
{
"entropy": 5.856541395187378,
"epoch": 0.3647280666931322,
"grad_norm": 0.96484375,
"learning_rate": 0.0004992025700091139,
"loss": 5.7467,
"mean_token_accuracy": 0.1376276656985283,
"num_tokens": 8087036.0,
"step": 3675
},
{
"entropy": 5.808696556091308,
"epoch": 0.3652242953552997,
"grad_norm": 0.89453125,
"learning_rate": 0.0004991995868305138,
"loss": 5.7756,
"mean_token_accuracy": 0.128545381128788,
"num_tokens": 8098517.0,
"step": 3680
},
{
"entropy": 5.738351106643677,
"epoch": 0.36572052401746724,
"grad_norm": 0.9765625,
"learning_rate": 0.0004991965980922471,
"loss": 5.6491,
"mean_token_accuracy": 0.13984912484884263,
"num_tokens": 8107475.0,
"step": 3685
},
{
"entropy": 5.764809942245483,
"epoch": 0.36621675267963477,
"grad_norm": 0.88671875,
"learning_rate": 0.0004991936037943878,
"loss": 5.7499,
"mean_token_accuracy": 0.13070235177874565,
"num_tokens": 8118529.0,
"step": 3690
},
{
"entropy": 5.828148555755615,
"epoch": 0.3667129813418023,
"grad_norm": 0.98046875,
"learning_rate": 0.0004991906039370104,
"loss": 5.6623,
"mean_token_accuracy": 0.13350174129009246,
"num_tokens": 8128951.0,
"step": 3695
},
{
"entropy": 5.638001918792725,
"epoch": 0.36720921000396983,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991875985201889,
"loss": 5.6795,
"mean_token_accuracy": 0.13438779413700103,
"num_tokens": 8139385.0,
"step": 3700
},
{
"entropy": 5.794062376022339,
"epoch": 0.36770543866613736,
"grad_norm": 0.9296875,
"learning_rate": 0.000499184587543998,
"loss": 5.7161,
"mean_token_accuracy": 0.13666265308856965,
"num_tokens": 8149376.0,
"step": 3705
},
{
"entropy": 5.7886322975158695,
"epoch": 0.3682016673283049,
"grad_norm": 0.87109375,
"learning_rate": 0.0004991815710085126,
"loss": 5.7389,
"mean_token_accuracy": 0.13325911313295363,
"num_tokens": 8160687.0,
"step": 3710
},
{
"entropy": 5.797859144210816,
"epoch": 0.3686978959904724,
"grad_norm": 0.8828125,
"learning_rate": 0.0004991785489138072,
"loss": 5.7453,
"mean_token_accuracy": 0.12990956529974937,
"num_tokens": 8171762.0,
"step": 3715
},
{
"entropy": 5.650119733810425,
"epoch": 0.36919412465263995,
"grad_norm": 0.99609375,
"learning_rate": 0.0004991755212599569,
"loss": 5.5183,
"mean_token_accuracy": 0.1405628427863121,
"num_tokens": 8181643.0,
"step": 3720
},
{
"entropy": 5.618019437789917,
"epoch": 0.3696903533148075,
"grad_norm": 0.9375,
"learning_rate": 0.0004991724880470366,
"loss": 5.7692,
"mean_token_accuracy": 0.13439096063375472,
"num_tokens": 8192692.0,
"step": 3725
},
{
"entropy": 5.8731317043304445,
"epoch": 0.370186581976975,
"grad_norm": 1.0078125,
"learning_rate": 0.0004991694492751217,
"loss": 5.7165,
"mean_token_accuracy": 0.1306888848543167,
"num_tokens": 8202941.0,
"step": 3730
},
{
"entropy": 5.741935539245605,
"epoch": 0.3706828106391425,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991664049442876,
"loss": 5.721,
"mean_token_accuracy": 0.13041145205497742,
"num_tokens": 8214206.0,
"step": 3735
},
{
"entropy": 5.719397354125976,
"epoch": 0.37117903930131,
"grad_norm": 1.15625,
"learning_rate": 0.0004991633550546097,
"loss": 5.5478,
"mean_token_accuracy": 0.14364924728870393,
"num_tokens": 8223672.0,
"step": 3740
},
{
"entropy": 5.712414312362671,
"epoch": 0.37167526796347755,
"grad_norm": 0.9453125,
"learning_rate": 0.0004991602996061635,
"loss": 5.6767,
"mean_token_accuracy": 0.13784719556570052,
"num_tokens": 8234189.0,
"step": 3745
},
{
"entropy": 5.789827442169189,
"epoch": 0.3721714966256451,
"grad_norm": 0.9140625,
"learning_rate": 0.0004991572385990249,
"loss": 5.7125,
"mean_token_accuracy": 0.13273593932390212,
"num_tokens": 8245382.0,
"step": 3750
},
{
"entropy": 5.801346778869629,
"epoch": 0.3726677252878126,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991541720332698,
"loss": 5.6675,
"mean_token_accuracy": 0.13517806306481361,
"num_tokens": 8254530.0,
"step": 3755
},
{
"entropy": 5.70973916053772,
"epoch": 0.37316395394998014,
"grad_norm": 0.99609375,
"learning_rate": 0.0004991510999089743,
"loss": 5.7233,
"mean_token_accuracy": 0.13753556311130524,
"num_tokens": 8264030.0,
"step": 3760
},
{
"entropy": 5.7114022731781,
"epoch": 0.3736601826121477,
"grad_norm": 0.9375,
"learning_rate": 0.0004991480222262145,
"loss": 5.6906,
"mean_token_accuracy": 0.13835713118314744,
"num_tokens": 8273537.0,
"step": 3765
},
{
"entropy": 5.805447435379028,
"epoch": 0.3741564112743152,
"grad_norm": 0.79296875,
"learning_rate": 0.0004991449389850668,
"loss": 5.7144,
"mean_token_accuracy": 0.131557959318161,
"num_tokens": 8286123.0,
"step": 3770
},
{
"entropy": 5.782719993591309,
"epoch": 0.37465263993648273,
"grad_norm": 0.984375,
"learning_rate": 0.0004991418501856076,
"loss": 5.7359,
"mean_token_accuracy": 0.1322738565504551,
"num_tokens": 8298115.0,
"step": 3775
},
{
"entropy": 5.8190759181976315,
"epoch": 0.37514886859865026,
"grad_norm": 1.0625,
"learning_rate": 0.0004991387558279134,
"loss": 5.7701,
"mean_token_accuracy": 0.13059676066040993,
"num_tokens": 8309376.0,
"step": 3780
},
{
"entropy": 5.72836856842041,
"epoch": 0.3756450972608178,
"grad_norm": 0.98828125,
"learning_rate": 0.0004991356559120612,
"loss": 5.7162,
"mean_token_accuracy": 0.13034249395132064,
"num_tokens": 8319272.0,
"step": 3785
},
{
"entropy": 5.763256311416626,
"epoch": 0.3761413259229853,
"grad_norm": 1.03125,
"learning_rate": 0.0004991325504381275,
"loss": 5.7034,
"mean_token_accuracy": 0.1289884179830551,
"num_tokens": 8330042.0,
"step": 3790
},
{
"entropy": 5.748146390914917,
"epoch": 0.37663755458515286,
"grad_norm": 0.9453125,
"learning_rate": 0.0004991294394061896,
"loss": 5.6656,
"mean_token_accuracy": 0.13286871761083602,
"num_tokens": 8340394.0,
"step": 3795
},
{
"entropy": 5.679359817504883,
"epoch": 0.3771337832473204,
"grad_norm": 0.94140625,
"learning_rate": 0.0004991263228163246,
"loss": 5.7017,
"mean_token_accuracy": 0.1317191243171692,
"num_tokens": 8351103.0,
"step": 3800
},
{
"entropy": 5.786353254318238,
"epoch": 0.3776300119094879,
"grad_norm": 0.87890625,
"learning_rate": 0.0004991232006686096,
"loss": 5.7306,
"mean_token_accuracy": 0.13431305363774299,
"num_tokens": 8363215.0,
"step": 3805
},
{
"entropy": 5.844559383392334,
"epoch": 0.3781262405716554,
"grad_norm": 0.921875,
"learning_rate": 0.0004991200729631223,
"loss": 5.7776,
"mean_token_accuracy": 0.12665583416819573,
"num_tokens": 8374602.0,
"step": 3810
},
{
"entropy": 5.70382399559021,
"epoch": 0.3786224692338229,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991169396999401,
"loss": 5.7011,
"mean_token_accuracy": 0.1290666401386261,
"num_tokens": 8386470.0,
"step": 3815
},
{
"entropy": 5.795304489135742,
"epoch": 0.37911869789599045,
"grad_norm": 1.0859375,
"learning_rate": 0.0004991138008791408,
"loss": 5.7216,
"mean_token_accuracy": 0.1339471235871315,
"num_tokens": 8396284.0,
"step": 3820
},
{
"entropy": 5.835215473175049,
"epoch": 0.379614926558158,
"grad_norm": 0.88671875,
"learning_rate": 0.0004991106565008021,
"loss": 5.7438,
"mean_token_accuracy": 0.12994968369603158,
"num_tokens": 8407856.0,
"step": 3825
},
{
"entropy": 5.737123155593872,
"epoch": 0.3801111552203255,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991075065650019,
"loss": 5.6464,
"mean_token_accuracy": 0.1383263237774372,
"num_tokens": 8419560.0,
"step": 3830
},
{
"entropy": 5.79041862487793,
"epoch": 0.38060738388249304,
"grad_norm": 0.9921875,
"learning_rate": 0.0004991043510718186,
"loss": 5.6539,
"mean_token_accuracy": 0.13635289072990417,
"num_tokens": 8430594.0,
"step": 3835
},
{
"entropy": 5.666716289520264,
"epoch": 0.3811036125446606,
"grad_norm": 0.92578125,
"learning_rate": 0.0004991011900213303,
"loss": 5.6231,
"mean_token_accuracy": 0.13809854835271834,
"num_tokens": 8440549.0,
"step": 3840
},
{
"entropy": 5.663881063461304,
"epoch": 0.3815998412068281,
"grad_norm": 0.953125,
"learning_rate": 0.0004990980234136152,
"loss": 5.6503,
"mean_token_accuracy": 0.1358008362352848,
"num_tokens": 8451553.0,
"step": 3845
},
{
"entropy": 5.762479019165039,
"epoch": 0.38209606986899564,
"grad_norm": 0.9375,
"learning_rate": 0.0004990948512487522,
"loss": 5.6673,
"mean_token_accuracy": 0.13858909234404565,
"num_tokens": 8462899.0,
"step": 3850
},
{
"entropy": 5.748224115371704,
"epoch": 0.38259229853116317,
"grad_norm": 1.015625,
"learning_rate": 0.0004990916735268198,
"loss": 5.7106,
"mean_token_accuracy": 0.13554939329624177,
"num_tokens": 8474566.0,
"step": 3855
},
{
"entropy": 5.790412521362304,
"epoch": 0.3830885271933307,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990884902478966,
"loss": 5.7276,
"mean_token_accuracy": 0.13737209066748618,
"num_tokens": 8484786.0,
"step": 3860
},
{
"entropy": 5.723254299163818,
"epoch": 0.3835847558554982,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990853014120619,
"loss": 5.6614,
"mean_token_accuracy": 0.13724090456962584,
"num_tokens": 8496416.0,
"step": 3865
},
{
"entropy": 5.744536972045898,
"epoch": 0.38408098451766576,
"grad_norm": 0.97265625,
"learning_rate": 0.0004990821070193944,
"loss": 5.6919,
"mean_token_accuracy": 0.13323881477117538,
"num_tokens": 8507287.0,
"step": 3870
},
{
"entropy": 5.806073141098023,
"epoch": 0.3845772131798333,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990789070699736,
"loss": 5.7355,
"mean_token_accuracy": 0.1301907204091549,
"num_tokens": 8517922.0,
"step": 3875
},
{
"entropy": 5.694337511062622,
"epoch": 0.3850734418420008,
"grad_norm": 1.03125,
"learning_rate": 0.0004990757015638787,
"loss": 5.6752,
"mean_token_accuracy": 0.1399615302681923,
"num_tokens": 8529532.0,
"step": 3880
},
{
"entropy": 5.744428539276123,
"epoch": 0.3855696705041683,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990724905011894,
"loss": 5.6327,
"mean_token_accuracy": 0.13914796561002732,
"num_tokens": 8540207.0,
"step": 3885
},
{
"entropy": 5.743565320968628,
"epoch": 0.3860658991663358,
"grad_norm": 1.03125,
"learning_rate": 0.000499069273881985,
"loss": 5.6818,
"mean_token_accuracy": 0.13373377323150634,
"num_tokens": 8551634.0,
"step": 3890
},
{
"entropy": 5.73831787109375,
"epoch": 0.38656212782850335,
"grad_norm": 0.88671875,
"learning_rate": 0.0004990660517063455,
"loss": 5.7359,
"mean_token_accuracy": 0.12965402603149415,
"num_tokens": 8563256.0,
"step": 3895
},
{
"entropy": 5.75383768081665,
"epoch": 0.3870583564906709,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990628239743508,
"loss": 5.7733,
"mean_token_accuracy": 0.13520312905311585,
"num_tokens": 8573833.0,
"step": 3900
},
{
"entropy": 5.775561237335205,
"epoch": 0.3875545851528384,
"grad_norm": 1.015625,
"learning_rate": 0.0004990595906860807,
"loss": 5.6575,
"mean_token_accuracy": 0.13614343851804733,
"num_tokens": 8584283.0,
"step": 3905
},
{
"entropy": 5.725812149047852,
"epoch": 0.38805081381500595,
"grad_norm": 1.0703125,
"learning_rate": 0.0004990563518416157,
"loss": 5.715,
"mean_token_accuracy": 0.13555167317390443,
"num_tokens": 8595344.0,
"step": 3910
},
{
"entropy": 5.760308265686035,
"epoch": 0.3885470424771735,
"grad_norm": 1.125,
"learning_rate": 0.0004990531074410358,
"loss": 5.6478,
"mean_token_accuracy": 0.13378319814801215,
"num_tokens": 8605625.0,
"step": 3915
},
{
"entropy": 5.705371856689453,
"epoch": 0.389043271139341,
"grad_norm": 0.9296875,
"learning_rate": 0.0004990498574844218,
"loss": 5.6268,
"mean_token_accuracy": 0.13527522459626198,
"num_tokens": 8616259.0,
"step": 3920
},
{
"entropy": 5.714511823654175,
"epoch": 0.38953949980150854,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990466019718542,
"loss": 5.6418,
"mean_token_accuracy": 0.13701364398002625,
"num_tokens": 8627179.0,
"step": 3925
},
{
"entropy": 5.776921558380127,
"epoch": 0.39003572846367607,
"grad_norm": 0.9609375,
"learning_rate": 0.0004990433409034134,
"loss": 5.7392,
"mean_token_accuracy": 0.13890910968184472,
"num_tokens": 8637448.0,
"step": 3930
},
{
"entropy": 5.656614494323731,
"epoch": 0.3905319571258436,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990400742791806,
"loss": 5.6588,
"mean_token_accuracy": 0.13583290949463844,
"num_tokens": 8649168.0,
"step": 3935
},
{
"entropy": 5.842466402053833,
"epoch": 0.39102818578801113,
"grad_norm": 0.91796875,
"learning_rate": 0.0004990368020992368,
"loss": 5.77,
"mean_token_accuracy": 0.13096642941236497,
"num_tokens": 8660653.0,
"step": 3940
},
{
"entropy": 5.779029989242554,
"epoch": 0.39152441445017866,
"grad_norm": 0.93359375,
"learning_rate": 0.000499033524363663,
"loss": 5.7419,
"mean_token_accuracy": 0.13649297654628753,
"num_tokens": 8672000.0,
"step": 3945
},
{
"entropy": 5.689775228500366,
"epoch": 0.3920206431123462,
"grad_norm": 1.0078125,
"learning_rate": 0.0004990302410725405,
"loss": 5.6659,
"mean_token_accuracy": 0.14118805006146432,
"num_tokens": 8682455.0,
"step": 3950
},
{
"entropy": 5.721955442428589,
"epoch": 0.3925168717745137,
"grad_norm": 0.921875,
"learning_rate": 0.0004990269522259509,
"loss": 5.6392,
"mean_token_accuracy": 0.13590310513973236,
"num_tokens": 8692975.0,
"step": 3955
},
{
"entropy": 5.674862480163574,
"epoch": 0.3930131004366812,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990236578239755,
"loss": 5.6253,
"mean_token_accuracy": 0.1390941359102726,
"num_tokens": 8703569.0,
"step": 3960
},
{
"entropy": 5.759145498275757,
"epoch": 0.3935093290988487,
"grad_norm": 0.9765625,
"learning_rate": 0.0004990203578666962,
"loss": 5.7267,
"mean_token_accuracy": 0.13206208124756813,
"num_tokens": 8713734.0,
"step": 3965
},
{
"entropy": 5.743315029144287,
"epoch": 0.39400555776101626,
"grad_norm": 1.0234375,
"learning_rate": 0.0004990170523541948,
"loss": 5.6232,
"mean_token_accuracy": 0.13998670354485512,
"num_tokens": 8723852.0,
"step": 3970
},
{
"entropy": 5.716074228286743,
"epoch": 0.3945017864231838,
"grad_norm": 0.93359375,
"learning_rate": 0.0004990137412865531,
"loss": 5.7699,
"mean_token_accuracy": 0.13368171229958534,
"num_tokens": 8734300.0,
"step": 3975
},
{
"entropy": 5.844010591506958,
"epoch": 0.3949980150853513,
"grad_norm": 1.1015625,
"learning_rate": 0.0004990104246638533,
"loss": 5.7851,
"mean_token_accuracy": 0.12670395150780678,
"num_tokens": 8745022.0,
"step": 3980
},
{
"entropy": 5.770476388931274,
"epoch": 0.39549424374751885,
"grad_norm": 0.9453125,
"learning_rate": 0.0004990071024861778,
"loss": 5.6707,
"mean_token_accuracy": 0.13448970913887023,
"num_tokens": 8754870.0,
"step": 3985
},
{
"entropy": 5.8409770965576175,
"epoch": 0.3959904724096864,
"grad_norm": 0.92578125,
"learning_rate": 0.0004990037747536087,
"loss": 5.7111,
"mean_token_accuracy": 0.13520667031407357,
"num_tokens": 8765536.0,
"step": 3990
},
{
"entropy": 5.734442949295044,
"epoch": 0.3964867010718539,
"grad_norm": 1.0859375,
"learning_rate": 0.0004990004414662288,
"loss": 5.6826,
"mean_token_accuracy": 0.13069720268249513,
"num_tokens": 8776898.0,
"step": 3995
},
{
"entropy": 5.730175352096557,
"epoch": 0.39698292973402144,
"grad_norm": 0.90234375,
"learning_rate": 0.0004989971026241205,
"loss": 5.7419,
"mean_token_accuracy": 0.13170261383056642,
"num_tokens": 8788176.0,
"step": 4000
},
{
"entropy": 5.692081499099731,
"epoch": 0.39747915839618897,
"grad_norm": 1.0546875,
"learning_rate": 0.0004989937582273668,
"loss": 5.6738,
"mean_token_accuracy": 0.1344978965818882,
"num_tokens": 8798939.0,
"step": 4005
},
{
"entropy": 5.766072082519531,
"epoch": 0.3979753870583565,
"grad_norm": 0.9921875,
"learning_rate": 0.0004989904082760506,
"loss": 5.7043,
"mean_token_accuracy": 0.13863254114985465,
"num_tokens": 8809092.0,
"step": 4010
},
{
"entropy": 5.7443944931030275,
"epoch": 0.39847161572052403,
"grad_norm": 0.9140625,
"learning_rate": 0.000498987052770255,
"loss": 5.6311,
"mean_token_accuracy": 0.14151344299316407,
"num_tokens": 8819352.0,
"step": 4015
},
{
"entropy": 5.7538347244262695,
"epoch": 0.39896784438269156,
"grad_norm": 0.99609375,
"learning_rate": 0.000498983691710063,
"loss": 5.7579,
"mean_token_accuracy": 0.13163795024156572,
"num_tokens": 8829878.0,
"step": 4020
},
{
"entropy": 5.75619592666626,
"epoch": 0.3994640730448591,
"grad_norm": 0.97265625,
"learning_rate": 0.000498980325095558,
"loss": 5.6265,
"mean_token_accuracy": 0.1341935031116009,
"num_tokens": 8840364.0,
"step": 4025
},
{
"entropy": 5.720586156845092,
"epoch": 0.3999603017070266,
"grad_norm": 1.046875,
"learning_rate": 0.0004989769529268238,
"loss": 5.7439,
"mean_token_accuracy": 0.12914118841290473,
"num_tokens": 8851340.0,
"step": 4030
},
{
"entropy": 5.715707445144654,
"epoch": 0.4004565303691941,
"grad_norm": 0.953125,
"learning_rate": 0.0004989735752039436,
"loss": 5.6986,
"mean_token_accuracy": 0.13207997828722,
"num_tokens": 8862571.0,
"step": 4035
},
{
"entropy": 5.753764533996582,
"epoch": 0.4009527590313616,
"grad_norm": 0.96875,
"learning_rate": 0.0004989701919270015,
"loss": 5.7026,
"mean_token_accuracy": 0.13718848526477814,
"num_tokens": 8873927.0,
"step": 4040
},
{
"entropy": 5.802546691894531,
"epoch": 0.40144898769352916,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989668030960812,
"loss": 5.8891,
"mean_token_accuracy": 0.12460704743862153,
"num_tokens": 8886542.0,
"step": 4045
},
{
"entropy": 5.795607662200927,
"epoch": 0.4019452163556967,
"grad_norm": 1.0078125,
"learning_rate": 0.0004989634087112667,
"loss": 5.6369,
"mean_token_accuracy": 0.13462657555937768,
"num_tokens": 8896114.0,
"step": 4050
},
{
"entropy": 5.739918994903564,
"epoch": 0.4024414450178642,
"grad_norm": 1.046875,
"learning_rate": 0.0004989600087726425,
"loss": 5.7415,
"mean_token_accuracy": 0.12707119584083557,
"num_tokens": 8906303.0,
"step": 4055
},
{
"entropy": 5.770398187637329,
"epoch": 0.40293767368003175,
"grad_norm": 0.98046875,
"learning_rate": 0.0004989566032802924,
"loss": 5.7568,
"mean_token_accuracy": 0.12863401621580123,
"num_tokens": 8916961.0,
"step": 4060
},
{
"entropy": 5.899053716659546,
"epoch": 0.4034339023421993,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989531922343013,
"loss": 5.8522,
"mean_token_accuracy": 0.1265175573527813,
"num_tokens": 8927700.0,
"step": 4065
},
{
"entropy": 5.775335073471069,
"epoch": 0.4039301310043668,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989497756347534,
"loss": 5.6768,
"mean_token_accuracy": 0.1359332799911499,
"num_tokens": 8939655.0,
"step": 4070
},
{
"entropy": 5.701972103118896,
"epoch": 0.40442635966653434,
"grad_norm": 0.94140625,
"learning_rate": 0.0004989463534817339,
"loss": 5.6153,
"mean_token_accuracy": 0.13829075172543526,
"num_tokens": 8951238.0,
"step": 4075
},
{
"entropy": 5.765563440322876,
"epoch": 0.40492258832870187,
"grad_norm": 0.921875,
"learning_rate": 0.0004989429257753273,
"loss": 5.69,
"mean_token_accuracy": 0.14064479097723961,
"num_tokens": 8962019.0,
"step": 4080
},
{
"entropy": 5.6876383304595945,
"epoch": 0.4054188169908694,
"grad_norm": 1.0390625,
"learning_rate": 0.0004989394925156187,
"loss": 5.6758,
"mean_token_accuracy": 0.13742703646421434,
"num_tokens": 8973861.0,
"step": 4085
},
{
"entropy": 5.685515975952148,
"epoch": 0.40591504565303693,
"grad_norm": 0.9296875,
"learning_rate": 0.0004989360537026932,
"loss": 5.6754,
"mean_token_accuracy": 0.13646822348237037,
"num_tokens": 8985981.0,
"step": 4090
},
{
"entropy": 5.736924839019776,
"epoch": 0.40641127431520446,
"grad_norm": 0.953125,
"learning_rate": 0.000498932609336636,
"loss": 5.5817,
"mean_token_accuracy": 0.14304670542478562,
"num_tokens": 8995318.0,
"step": 4095
},
{
"entropy": 5.731611108779907,
"epoch": 0.406907502977372,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989291594175327,
"loss": 5.6133,
"mean_token_accuracy": 0.13422033861279486,
"num_tokens": 9004679.0,
"step": 4100
},
{
"entropy": 5.830239486694336,
"epoch": 0.4074037316395395,
"grad_norm": 0.9375,
"learning_rate": 0.0004989257039454689,
"loss": 5.738,
"mean_token_accuracy": 0.133600851893425,
"num_tokens": 9016422.0,
"step": 4105
},
{
"entropy": 5.641452980041504,
"epoch": 0.40789996030170705,
"grad_norm": 0.9453125,
"learning_rate": 0.00049892224292053,
"loss": 5.6599,
"mean_token_accuracy": 0.13614288121461868,
"num_tokens": 9027914.0,
"step": 4110
},
{
"entropy": 5.735976791381836,
"epoch": 0.40839618896387453,
"grad_norm": 0.92578125,
"learning_rate": 0.0004989187763428021,
"loss": 5.6392,
"mean_token_accuracy": 0.1383671998977661,
"num_tokens": 9038701.0,
"step": 4115
},
{
"entropy": 5.70672926902771,
"epoch": 0.40889241762604206,
"grad_norm": 0.9765625,
"learning_rate": 0.000498915304212371,
"loss": 5.6956,
"mean_token_accuracy": 0.13438485115766524,
"num_tokens": 9050620.0,
"step": 4120
},
{
"entropy": 5.784137392044068,
"epoch": 0.4093886462882096,
"grad_norm": 0.93359375,
"learning_rate": 0.0004989118265293228,
"loss": 5.7157,
"mean_token_accuracy": 0.13783123940229416,
"num_tokens": 9060900.0,
"step": 4125
},
{
"entropy": 5.692770051956177,
"epoch": 0.4098848749503771,
"grad_norm": 0.9453125,
"learning_rate": 0.0004989083432937439,
"loss": 5.6358,
"mean_token_accuracy": 0.13960246965289116,
"num_tokens": 9072122.0,
"step": 4130
},
{
"entropy": 5.695958614349365,
"epoch": 0.41038110361254465,
"grad_norm": 0.94921875,
"learning_rate": 0.0004989048545057204,
"loss": 5.6621,
"mean_token_accuracy": 0.13656162321567536,
"num_tokens": 9082780.0,
"step": 4135
},
{
"entropy": 5.6983644485473635,
"epoch": 0.4108773322747122,
"grad_norm": 1.03125,
"learning_rate": 0.0004989013601653391,
"loss": 5.6915,
"mean_token_accuracy": 0.13537346571683884,
"num_tokens": 9093518.0,
"step": 4140
},
{
"entropy": 5.7246016502380375,
"epoch": 0.4113735609368797,
"grad_norm": 0.890625,
"learning_rate": 0.0004988978602726866,
"loss": 5.716,
"mean_token_accuracy": 0.14137836471199988,
"num_tokens": 9105035.0,
"step": 4145
},
{
"entropy": 5.734498691558838,
"epoch": 0.41186978959904724,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988943548278494,
"loss": 5.6336,
"mean_token_accuracy": 0.13698274716734887,
"num_tokens": 9117154.0,
"step": 4150
},
{
"entropy": 5.706672430038452,
"epoch": 0.4123660182612148,
"grad_norm": 0.96875,
"learning_rate": 0.0004988908438309149,
"loss": 5.7051,
"mean_token_accuracy": 0.1334255114197731,
"num_tokens": 9128818.0,
"step": 4155
},
{
"entropy": 5.768472909927368,
"epoch": 0.4128622469233823,
"grad_norm": 0.953125,
"learning_rate": 0.0004988873272819697,
"loss": 5.7179,
"mean_token_accuracy": 0.1378393739461899,
"num_tokens": 9140188.0,
"step": 4160
},
{
"entropy": 5.915740299224853,
"epoch": 0.41335847558554983,
"grad_norm": 1.0703125,
"learning_rate": 0.0004988838051811014,
"loss": 5.8077,
"mean_token_accuracy": 0.12831991240382196,
"num_tokens": 9150565.0,
"step": 4165
},
{
"entropy": 5.691146659851074,
"epoch": 0.41385470424771736,
"grad_norm": 0.9921875,
"learning_rate": 0.000498880277528397,
"loss": 5.659,
"mean_token_accuracy": 0.13782408088445663,
"num_tokens": 9160130.0,
"step": 4170
},
{
"entropy": 5.652090454101563,
"epoch": 0.4143509329098849,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988767443239443,
"loss": 5.652,
"mean_token_accuracy": 0.14071165323257445,
"num_tokens": 9171740.0,
"step": 4175
},
{
"entropy": 5.747271871566772,
"epoch": 0.4148471615720524,
"grad_norm": 1.0234375,
"learning_rate": 0.0004988732055678307,
"loss": 5.609,
"mean_token_accuracy": 0.15193828493356704,
"num_tokens": 9182459.0,
"step": 4180
},
{
"entropy": 5.772048234939575,
"epoch": 0.41534339023421996,
"grad_norm": 0.953125,
"learning_rate": 0.000498869661260144,
"loss": 5.6562,
"mean_token_accuracy": 0.13707233741879463,
"num_tokens": 9192284.0,
"step": 4185
},
{
"entropy": 5.678713798522949,
"epoch": 0.41583961889638743,
"grad_norm": 1.0234375,
"learning_rate": 0.000498866111400972,
"loss": 5.6578,
"mean_token_accuracy": 0.13831765949726105,
"num_tokens": 9202700.0,
"step": 4190
},
{
"entropy": 5.702182912826538,
"epoch": 0.41633584755855496,
"grad_norm": 0.99609375,
"learning_rate": 0.0004988625559904029,
"loss": 5.6639,
"mean_token_accuracy": 0.1302245832979679,
"num_tokens": 9213013.0,
"step": 4195
},
{
"entropy": 5.792655801773071,
"epoch": 0.4168320762207225,
"grad_norm": 0.8203125,
"learning_rate": 0.0004988589950285248,
"loss": 5.7366,
"mean_token_accuracy": 0.13581578209996223,
"num_tokens": 9224568.0,
"step": 4200
},
{
"entropy": 5.735432529449463,
"epoch": 0.41732830488289,
"grad_norm": 0.93359375,
"learning_rate": 0.0004988554285154259,
"loss": 5.6132,
"mean_token_accuracy": 0.13576259464025497,
"num_tokens": 9233791.0,
"step": 4205
},
{
"entropy": 5.714122867584228,
"epoch": 0.41782453354505755,
"grad_norm": 0.9140625,
"learning_rate": 0.0004988518564511948,
"loss": 5.6566,
"mean_token_accuracy": 0.13220425993204116,
"num_tokens": 9245048.0,
"step": 4210
},
{
"entropy": 5.798601531982422,
"epoch": 0.4183207622072251,
"grad_norm": 0.91796875,
"learning_rate": 0.0004988482788359202,
"loss": 5.6919,
"mean_token_accuracy": 0.14111457020044327,
"num_tokens": 9256508.0,
"step": 4215
},
{
"entropy": 5.750736999511719,
"epoch": 0.4188169908693926,
"grad_norm": 0.99609375,
"learning_rate": 0.0004988446956696904,
"loss": 5.7043,
"mean_token_accuracy": 0.13517057597637178,
"num_tokens": 9267871.0,
"step": 4220
},
{
"entropy": 5.788787364959717,
"epoch": 0.41931321953156014,
"grad_norm": 0.94140625,
"learning_rate": 0.0004988411069525947,
"loss": 5.7662,
"mean_token_accuracy": 0.13160478845238685,
"num_tokens": 9279287.0,
"step": 4225
},
{
"entropy": 5.814012050628662,
"epoch": 0.4198094481937277,
"grad_norm": 0.9921875,
"learning_rate": 0.0004988375126847216,
"loss": 5.7846,
"mean_token_accuracy": 0.12892357259988785,
"num_tokens": 9291257.0,
"step": 4230
},
{
"entropy": 5.751958894729614,
"epoch": 0.4203056768558952,
"grad_norm": 1.1015625,
"learning_rate": 0.0004988339128661607,
"loss": 5.7387,
"mean_token_accuracy": 0.13436317667365075,
"num_tokens": 9302590.0,
"step": 4235
},
{
"entropy": 5.624151992797851,
"epoch": 0.42080190551806274,
"grad_norm": 0.92578125,
"learning_rate": 0.000498830307497001,
"loss": 5.5986,
"mean_token_accuracy": 0.13606150150299073,
"num_tokens": 9313174.0,
"step": 4240
},
{
"entropy": 5.751293659210205,
"epoch": 0.42129813418023027,
"grad_norm": 1.0078125,
"learning_rate": 0.000498826696577332,
"loss": 5.6813,
"mean_token_accuracy": 0.13548489436507224,
"num_tokens": 9325034.0,
"step": 4245
},
{
"entropy": 5.739410495758056,
"epoch": 0.4217943628423978,
"grad_norm": 0.92578125,
"learning_rate": 0.0004988230801072432,
"loss": 5.6622,
"mean_token_accuracy": 0.13756145015358925,
"num_tokens": 9336834.0,
"step": 4250
},
{
"entropy": 5.642874622344971,
"epoch": 0.4222905915045653,
"grad_norm": 0.9609375,
"learning_rate": 0.0004988194580868244,
"loss": 5.6777,
"mean_token_accuracy": 0.13649028465151786,
"num_tokens": 9347747.0,
"step": 4255
},
{
"entropy": 5.70621657371521,
"epoch": 0.42278682016673286,
"grad_norm": 0.90234375,
"learning_rate": 0.0004988158305161654,
"loss": 5.69,
"mean_token_accuracy": 0.13723429888486863,
"num_tokens": 9359186.0,
"step": 4260
},
{
"entropy": 5.729804754257202,
"epoch": 0.42328304882890033,
"grad_norm": 0.91796875,
"learning_rate": 0.000498812197395356,
"loss": 5.6134,
"mean_token_accuracy": 0.1391861580312252,
"num_tokens": 9369485.0,
"step": 4265
},
{
"entropy": 5.728609323501587,
"epoch": 0.42377927749106786,
"grad_norm": 1.1171875,
"learning_rate": 0.0004988085587244863,
"loss": 5.6961,
"mean_token_accuracy": 0.13410992473363875,
"num_tokens": 9382658.0,
"step": 4270
},
{
"entropy": 5.681890535354614,
"epoch": 0.4242755061532354,
"grad_norm": 0.90625,
"learning_rate": 0.0004988049145036466,
"loss": 5.674,
"mean_token_accuracy": 0.13061357513070107,
"num_tokens": 9394151.0,
"step": 4275
},
{
"entropy": 5.812398242950439,
"epoch": 0.4247717348154029,
"grad_norm": 0.9453125,
"learning_rate": 0.0004988012647329273,
"loss": 5.7482,
"mean_token_accuracy": 0.13043404445052148,
"num_tokens": 9405956.0,
"step": 4280
},
{
"entropy": 5.757098770141601,
"epoch": 0.42526796347757045,
"grad_norm": 0.98046875,
"learning_rate": 0.000498797609412419,
"loss": 5.6348,
"mean_token_accuracy": 0.1390029162168503,
"num_tokens": 9416889.0,
"step": 4285
},
{
"entropy": 5.752142858505249,
"epoch": 0.425764192139738,
"grad_norm": 0.9375,
"learning_rate": 0.0004987939485422122,
"loss": 5.7588,
"mean_token_accuracy": 0.12929404750466347,
"num_tokens": 9427602.0,
"step": 4290
},
{
"entropy": 5.780097198486328,
"epoch": 0.4262604208019055,
"grad_norm": 0.86328125,
"learning_rate": 0.0004987902821223976,
"loss": 5.6649,
"mean_token_accuracy": 0.13806519359350206,
"num_tokens": 9438557.0,
"step": 4295
},
{
"entropy": 5.725652647018433,
"epoch": 0.42675664946407305,
"grad_norm": 0.921875,
"learning_rate": 0.0004987866101530663,
"loss": 5.7168,
"mean_token_accuracy": 0.13426719680428506,
"num_tokens": 9449598.0,
"step": 4300
},
{
"entropy": 5.707887935638428,
"epoch": 0.4272528781262406,
"grad_norm": 0.96875,
"learning_rate": 0.0004987829326343092,
"loss": 5.6417,
"mean_token_accuracy": 0.1364934429526329,
"num_tokens": 9459740.0,
"step": 4305
},
{
"entropy": 5.700517320632935,
"epoch": 0.4277491067884081,
"grad_norm": 0.984375,
"learning_rate": 0.0004987792495662177,
"loss": 5.659,
"mean_token_accuracy": 0.13137739896774292,
"num_tokens": 9469835.0,
"step": 4310
},
{
"entropy": 5.794715166091919,
"epoch": 0.42824533545057564,
"grad_norm": 1.0390625,
"learning_rate": 0.000498775560948883,
"loss": 5.741,
"mean_token_accuracy": 0.13609717413783073,
"num_tokens": 9480735.0,
"step": 4315
},
{
"entropy": 5.718795871734619,
"epoch": 0.42874156411274317,
"grad_norm": 0.9765625,
"learning_rate": 0.0004987718667823965,
"loss": 5.6311,
"mean_token_accuracy": 0.14181154519319533,
"num_tokens": 9491321.0,
"step": 4320
},
{
"entropy": 5.699181270599365,
"epoch": 0.4292377927749107,
"grad_norm": 0.87109375,
"learning_rate": 0.00049876816706685,
"loss": 5.7662,
"mean_token_accuracy": 0.12978544905781747,
"num_tokens": 9503507.0,
"step": 4325
},
{
"entropy": 5.800131225585938,
"epoch": 0.42973402143707823,
"grad_norm": 1.8984375,
"learning_rate": 0.000498764461802335,
"loss": 5.7704,
"mean_token_accuracy": 0.13711531162261964,
"num_tokens": 9514404.0,
"step": 4330
},
{
"entropy": 5.757583141326904,
"epoch": 0.43023025009924576,
"grad_norm": 0.99609375,
"learning_rate": 0.0004987607509889437,
"loss": 5.631,
"mean_token_accuracy": 0.1362014576792717,
"num_tokens": 9525555.0,
"step": 4335
},
{
"entropy": 5.753689956665039,
"epoch": 0.43072647876141323,
"grad_norm": 0.98046875,
"learning_rate": 0.0004987570346267677,
"loss": 5.6269,
"mean_token_accuracy": 0.1425912782549858,
"num_tokens": 9535084.0,
"step": 4340
},
{
"entropy": 5.658543395996094,
"epoch": 0.43122270742358076,
"grad_norm": 0.91796875,
"learning_rate": 0.0004987533127158996,
"loss": 5.6504,
"mean_token_accuracy": 0.1396719478070736,
"num_tokens": 9546170.0,
"step": 4345
},
{
"entropy": 5.718197584152222,
"epoch": 0.4317189360857483,
"grad_norm": 0.9296875,
"learning_rate": 0.0004987495852564314,
"loss": 5.6516,
"mean_token_accuracy": 0.1370712161064148,
"num_tokens": 9556793.0,
"step": 4350
},
{
"entropy": 5.743592357635498,
"epoch": 0.4322151647479158,
"grad_norm": 1.046875,
"learning_rate": 0.0004987458522484556,
"loss": 5.6103,
"mean_token_accuracy": 0.1390673503279686,
"num_tokens": 9568114.0,
"step": 4355
},
{
"entropy": 5.67595329284668,
"epoch": 0.43271139341008336,
"grad_norm": 1.078125,
"learning_rate": 0.0004987421136920648,
"loss": 5.607,
"mean_token_accuracy": 0.13263363316655158,
"num_tokens": 9579119.0,
"step": 4360
},
{
"entropy": 5.631229734420776,
"epoch": 0.4332076220722509,
"grad_norm": 1.0625,
"learning_rate": 0.0004987383695873517,
"loss": 5.6654,
"mean_token_accuracy": 0.13817091807723045,
"num_tokens": 9590259.0,
"step": 4365
},
{
"entropy": 5.799467992782593,
"epoch": 0.4337038507344184,
"grad_norm": 0.984375,
"learning_rate": 0.0004987346199344092,
"loss": 5.7005,
"mean_token_accuracy": 0.12761994451284409,
"num_tokens": 9600431.0,
"step": 4370
},
{
"entropy": 5.686568689346314,
"epoch": 0.43420007939658595,
"grad_norm": 0.87109375,
"learning_rate": 0.0004987308647333303,
"loss": 5.675,
"mean_token_accuracy": 0.1352570503950119,
"num_tokens": 9610286.0,
"step": 4375
},
{
"entropy": 5.780293607711792,
"epoch": 0.4346963080587535,
"grad_norm": 1.0625,
"learning_rate": 0.0004987271039842079,
"loss": 5.691,
"mean_token_accuracy": 0.13502518460154533,
"num_tokens": 9621148.0,
"step": 4380
},
{
"entropy": 5.705330896377563,
"epoch": 0.435192536720921,
"grad_norm": 0.9375,
"learning_rate": 0.0004987233376871355,
"loss": 5.577,
"mean_token_accuracy": 0.13657931089401246,
"num_tokens": 9631300.0,
"step": 4385
},
{
"entropy": 5.670144701004029,
"epoch": 0.43568876538308854,
"grad_norm": 0.88671875,
"learning_rate": 0.0004987195658422063,
"loss": 5.6862,
"mean_token_accuracy": 0.13548445776104928,
"num_tokens": 9641986.0,
"step": 4390
},
{
"entropy": 5.737906885147095,
"epoch": 0.43618499404525607,
"grad_norm": 0.875,
"learning_rate": 0.0004987157884495141,
"loss": 5.6967,
"mean_token_accuracy": 0.1371404081583023,
"num_tokens": 9652962.0,
"step": 4395
},
{
"entropy": 5.726242446899414,
"epoch": 0.4366812227074236,
"grad_norm": 0.90625,
"learning_rate": 0.0004987120055091523,
"loss": 5.7136,
"mean_token_accuracy": 0.13460650369524957,
"num_tokens": 9664084.0,
"step": 4400
},
{
"entropy": 5.708842039108276,
"epoch": 0.43717745136959113,
"grad_norm": 0.8984375,
"learning_rate": 0.0004987082170212149,
"loss": 5.6156,
"mean_token_accuracy": 0.13606675416231157,
"num_tokens": 9675821.0,
"step": 4405
},
{
"entropy": 5.75541615486145,
"epoch": 0.43767368003175866,
"grad_norm": 1.0546875,
"learning_rate": 0.0004987044229857957,
"loss": 5.7358,
"mean_token_accuracy": 0.13673339784145355,
"num_tokens": 9687964.0,
"step": 4410
},
{
"entropy": 5.705683469772339,
"epoch": 0.43816990869392614,
"grad_norm": 0.87109375,
"learning_rate": 0.0004987006234029889,
"loss": 5.7125,
"mean_token_accuracy": 0.1395598903298378,
"num_tokens": 9699653.0,
"step": 4415
},
{
"entropy": 5.797198009490967,
"epoch": 0.43866613735609367,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986968182728885,
"loss": 5.7274,
"mean_token_accuracy": 0.13438003659248351,
"num_tokens": 9709872.0,
"step": 4420
},
{
"entropy": 5.767149305343628,
"epoch": 0.4391623660182612,
"grad_norm": 1.1484375,
"learning_rate": 0.0004986930075955892,
"loss": 5.6454,
"mean_token_accuracy": 0.13367924019694327,
"num_tokens": 9720589.0,
"step": 4425
},
{
"entropy": 5.612972831726074,
"epoch": 0.4396585946804287,
"grad_norm": 0.94921875,
"learning_rate": 0.0004986891913711853,
"loss": 5.5894,
"mean_token_accuracy": 0.14131948202848435,
"num_tokens": 9731862.0,
"step": 4430
},
{
"entropy": 5.685967874526978,
"epoch": 0.44015482334259626,
"grad_norm": 0.9296875,
"learning_rate": 0.0004986853695997715,
"loss": 5.7326,
"mean_token_accuracy": 0.1408631294965744,
"num_tokens": 9743710.0,
"step": 4435
},
{
"entropy": 5.753010702133179,
"epoch": 0.4406510520047638,
"grad_norm": 0.9921875,
"learning_rate": 0.0004986815422814425,
"loss": 5.6725,
"mean_token_accuracy": 0.12475050538778305,
"num_tokens": 9753855.0,
"step": 4440
},
{
"entropy": 5.8120142936706545,
"epoch": 0.4411472806669313,
"grad_norm": 1.015625,
"learning_rate": 0.0004986777094162932,
"loss": 5.6878,
"mean_token_accuracy": 0.14027968794107437,
"num_tokens": 9763883.0,
"step": 4445
},
{
"entropy": 5.672481060028076,
"epoch": 0.44164350932909885,
"grad_norm": 1.0,
"learning_rate": 0.0004986738710044187,
"loss": 5.6472,
"mean_token_accuracy": 0.13763445019721984,
"num_tokens": 9774890.0,
"step": 4450
},
{
"entropy": 5.662131595611572,
"epoch": 0.4421397379912664,
"grad_norm": 0.9453125,
"learning_rate": 0.0004986700270459142,
"loss": 5.635,
"mean_token_accuracy": 0.13798378705978392,
"num_tokens": 9785249.0,
"step": 4455
},
{
"entropy": 5.721683168411255,
"epoch": 0.4426359666534339,
"grad_norm": 1.0390625,
"learning_rate": 0.0004986661775408749,
"loss": 5.6417,
"mean_token_accuracy": 0.13845691978931426,
"num_tokens": 9795723.0,
"step": 4460
},
{
"entropy": 5.741473865509033,
"epoch": 0.44313219531560144,
"grad_norm": 0.90234375,
"learning_rate": 0.0004986623224893965,
"loss": 5.5845,
"mean_token_accuracy": 0.1356796771287918,
"num_tokens": 9807637.0,
"step": 4465
},
{
"entropy": 5.690484952926636,
"epoch": 0.44362842397776897,
"grad_norm": 0.95703125,
"learning_rate": 0.0004986584618915744,
"loss": 5.6982,
"mean_token_accuracy": 0.13647759109735488,
"num_tokens": 9820298.0,
"step": 4470
},
{
"entropy": 5.721742868423462,
"epoch": 0.4441246526399365,
"grad_norm": 1.09375,
"learning_rate": 0.0004986545957475045,
"loss": 5.5906,
"mean_token_accuracy": 0.14316966831684114,
"num_tokens": 9831251.0,
"step": 4475
},
{
"entropy": 5.652020597457886,
"epoch": 0.44462088130210403,
"grad_norm": 0.93359375,
"learning_rate": 0.0004986507240572825,
"loss": 5.6525,
"mean_token_accuracy": 0.13798210993409157,
"num_tokens": 9842884.0,
"step": 4480
},
{
"entropy": 5.637630605697632,
"epoch": 0.44511710996427156,
"grad_norm": 0.9765625,
"learning_rate": 0.0004986468468210042,
"loss": 5.5504,
"mean_token_accuracy": 0.13658547177910804,
"num_tokens": 9854413.0,
"step": 4485
},
{
"entropy": 5.747673511505127,
"epoch": 0.44561333862643904,
"grad_norm": 0.98046875,
"learning_rate": 0.0004986429640387663,
"loss": 5.6892,
"mean_token_accuracy": 0.13593099191784858,
"num_tokens": 9865771.0,
"step": 4490
},
{
"entropy": 5.705465602874756,
"epoch": 0.44610956728860657,
"grad_norm": 0.84765625,
"learning_rate": 0.0004986390757106648,
"loss": 5.6164,
"mean_token_accuracy": 0.14181874468922614,
"num_tokens": 9876717.0,
"step": 4495
},
{
"entropy": 5.68349666595459,
"epoch": 0.4466057959507741,
"grad_norm": 0.87890625,
"learning_rate": 0.0004986351818367959,
"loss": 5.69,
"mean_token_accuracy": 0.1398135282099247,
"num_tokens": 9887616.0,
"step": 4500
},
{
"entropy": 5.7810155868530275,
"epoch": 0.44710202461294163,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986312824172566,
"loss": 5.661,
"mean_token_accuracy": 0.13780627772212029,
"num_tokens": 9898979.0,
"step": 4505
},
{
"entropy": 5.6977684020996096,
"epoch": 0.44759825327510916,
"grad_norm": 0.89453125,
"learning_rate": 0.0004986273774521433,
"loss": 5.6925,
"mean_token_accuracy": 0.13116807863116264,
"num_tokens": 9912158.0,
"step": 4510
},
{
"entropy": 5.686586380004883,
"epoch": 0.4480944819372767,
"grad_norm": 1.0078125,
"learning_rate": 0.0004986234669415529,
"loss": 5.6842,
"mean_token_accuracy": 0.13569653257727624,
"num_tokens": 9923886.0,
"step": 4515
},
{
"entropy": 5.7598377704620365,
"epoch": 0.4485907105994442,
"grad_norm": 0.9375,
"learning_rate": 0.0004986195508855824,
"loss": 5.6603,
"mean_token_accuracy": 0.13824379518628122,
"num_tokens": 9934979.0,
"step": 4520
},
{
"entropy": 5.7572063446044925,
"epoch": 0.44908693926161175,
"grad_norm": 1.0078125,
"learning_rate": 0.000498615629284329,
"loss": 5.7586,
"mean_token_accuracy": 0.128438288718462,
"num_tokens": 9944715.0,
"step": 4525
},
{
"entropy": 5.716220664978027,
"epoch": 0.4495831679237793,
"grad_norm": 0.890625,
"learning_rate": 0.0004986117021378896,
"loss": 5.6142,
"mean_token_accuracy": 0.13333934247493745,
"num_tokens": 9954736.0,
"step": 4530
},
{
"entropy": 5.768646383285523,
"epoch": 0.4500793965859468,
"grad_norm": 0.8671875,
"learning_rate": 0.000498607769446362,
"loss": 5.7149,
"mean_token_accuracy": 0.13724260032176971,
"num_tokens": 9965952.0,
"step": 4535
},
{
"entropy": 5.718695592880249,
"epoch": 0.45057562524811434,
"grad_norm": 1.0390625,
"learning_rate": 0.0004986038312098435,
"loss": 5.6581,
"mean_token_accuracy": 0.14347179606556892,
"num_tokens": 9976428.0,
"step": 4540
},
{
"entropy": 5.752839088439941,
"epoch": 0.4510718539102819,
"grad_norm": 0.98828125,
"learning_rate": 0.0004985998874284319,
"loss": 5.5975,
"mean_token_accuracy": 0.1452907532453537,
"num_tokens": 9986100.0,
"step": 4545
},
{
"entropy": 5.5815338611602785,
"epoch": 0.4515680825724494,
"grad_norm": 1.0390625,
"learning_rate": 0.0004985959381022248,
"loss": 5.5761,
"mean_token_accuracy": 0.1382984571158886,
"num_tokens": 9996584.0,
"step": 4550
},
{
"entropy": 5.693500804901123,
"epoch": 0.45206431123461693,
"grad_norm": 1.078125,
"learning_rate": 0.0004985919832313203,
"loss": 5.6048,
"mean_token_accuracy": 0.13820491433143617,
"num_tokens": 10007792.0,
"step": 4555
},
{
"entropy": 5.809909439086914,
"epoch": 0.45256053989678446,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985880228158164,
"loss": 5.7928,
"mean_token_accuracy": 0.12971990257501603,
"num_tokens": 10019615.0,
"step": 4560
},
{
"entropy": 5.679464054107666,
"epoch": 0.45305676855895194,
"grad_norm": 1.03125,
"learning_rate": 0.0004985840568558112,
"loss": 5.635,
"mean_token_accuracy": 0.1400697186589241,
"num_tokens": 10030223.0,
"step": 4565
},
{
"entropy": 5.715747213363647,
"epoch": 0.45355299722111947,
"grad_norm": 0.9296875,
"learning_rate": 0.0004985800853514033,
"loss": 5.6386,
"mean_token_accuracy": 0.13305401057004929,
"num_tokens": 10041522.0,
"step": 4570
},
{
"entropy": 5.730625104904175,
"epoch": 0.454049225883287,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985761083026911,
"loss": 5.5584,
"mean_token_accuracy": 0.1452631801366806,
"num_tokens": 10052847.0,
"step": 4575
},
{
"entropy": 5.649207019805909,
"epoch": 0.45454545454545453,
"grad_norm": 0.9921875,
"learning_rate": 0.0004985721257097731,
"loss": 5.5915,
"mean_token_accuracy": 0.13804994523525238,
"num_tokens": 10062259.0,
"step": 4580
},
{
"entropy": 5.785450220108032,
"epoch": 0.45504168320762206,
"grad_norm": 0.99609375,
"learning_rate": 0.0004985681375727482,
"loss": 5.7556,
"mean_token_accuracy": 0.13155426308512688,
"num_tokens": 10073341.0,
"step": 4585
},
{
"entropy": 5.724898719787598,
"epoch": 0.4555379118697896,
"grad_norm": 0.859375,
"learning_rate": 0.0004985641438917151,
"loss": 5.6249,
"mean_token_accuracy": 0.14122139066457748,
"num_tokens": 10083992.0,
"step": 4590
},
{
"entropy": 5.7234375,
"epoch": 0.4560341405319571,
"grad_norm": 0.97265625,
"learning_rate": 0.0004985601446667731,
"loss": 5.6081,
"mean_token_accuracy": 0.13897195234894752,
"num_tokens": 10094245.0,
"step": 4595
},
{
"entropy": 5.732874202728271,
"epoch": 0.45653036919412465,
"grad_norm": 0.96484375,
"learning_rate": 0.0004985561398980213,
"loss": 5.5905,
"mean_token_accuracy": 0.14286764189600945,
"num_tokens": 10105976.0,
"step": 4600
},
{
"entropy": 5.620663595199585,
"epoch": 0.4570265978562922,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985521295855588,
"loss": 5.6699,
"mean_token_accuracy": 0.13586745038628578,
"num_tokens": 10117373.0,
"step": 4605
},
{
"entropy": 5.703548860549927,
"epoch": 0.4575228265184597,
"grad_norm": 0.90234375,
"learning_rate": 0.0004985481137294853,
"loss": 5.612,
"mean_token_accuracy": 0.13380024060606957,
"num_tokens": 10128470.0,
"step": 4610
},
{
"entropy": 5.61048436164856,
"epoch": 0.45801905518062724,
"grad_norm": 0.90625,
"learning_rate": 0.0004985440923299002,
"loss": 5.5932,
"mean_token_accuracy": 0.14190664738416672,
"num_tokens": 10140217.0,
"step": 4615
},
{
"entropy": 5.656254386901855,
"epoch": 0.4585152838427948,
"grad_norm": 1.0234375,
"learning_rate": 0.0004985400653869034,
"loss": 5.6962,
"mean_token_accuracy": 0.13559751883149146,
"num_tokens": 10152308.0,
"step": 4620
},
{
"entropy": 5.842973804473877,
"epoch": 0.4590115125049623,
"grad_norm": 0.984375,
"learning_rate": 0.0004985360329005945,
"loss": 5.7106,
"mean_token_accuracy": 0.13268141373991965,
"num_tokens": 10164234.0,
"step": 4625
},
{
"entropy": 5.825473308563232,
"epoch": 0.45950774116712984,
"grad_norm": 0.95703125,
"learning_rate": 0.0004985319948710738,
"loss": 5.7591,
"mean_token_accuracy": 0.13398344442248344,
"num_tokens": 10175287.0,
"step": 4630
},
{
"entropy": 5.714378023147583,
"epoch": 0.46000396982929737,
"grad_norm": 1.109375,
"learning_rate": 0.0004985279512984413,
"loss": 5.648,
"mean_token_accuracy": 0.14291325584053993,
"num_tokens": 10184658.0,
"step": 4635
},
{
"entropy": 5.6979296684265135,
"epoch": 0.46050019849146484,
"grad_norm": 1.0390625,
"learning_rate": 0.0004985239021827972,
"loss": 5.6358,
"mean_token_accuracy": 0.13907306492328644,
"num_tokens": 10194593.0,
"step": 4640
},
{
"entropy": 5.647254276275635,
"epoch": 0.46099642715363237,
"grad_norm": 0.8984375,
"learning_rate": 0.000498519847524242,
"loss": 5.5415,
"mean_token_accuracy": 0.13873762339353563,
"num_tokens": 10205200.0,
"step": 4645
},
{
"entropy": 5.642165088653565,
"epoch": 0.4614926558157999,
"grad_norm": 0.94921875,
"learning_rate": 0.0004985157873228762,
"loss": 5.58,
"mean_token_accuracy": 0.13976380601525307,
"num_tokens": 10216848.0,
"step": 4650
},
{
"entropy": 5.7167640209198,
"epoch": 0.46198888447796743,
"grad_norm": 0.89453125,
"learning_rate": 0.0004985117215788005,
"loss": 5.6333,
"mean_token_accuracy": 0.13801254853606224,
"num_tokens": 10228188.0,
"step": 4655
},
{
"entropy": 5.695525503158569,
"epoch": 0.46248511314013496,
"grad_norm": 1.015625,
"learning_rate": 0.0004985076502921158,
"loss": 5.7119,
"mean_token_accuracy": 0.13519898056983948,
"num_tokens": 10238253.0,
"step": 4660
},
{
"entropy": 5.7912256717681885,
"epoch": 0.4629813418023025,
"grad_norm": 1.0703125,
"learning_rate": 0.000498503573462923,
"loss": 5.6643,
"mean_token_accuracy": 0.13310108035802842,
"num_tokens": 10248169.0,
"step": 4665
},
{
"entropy": 5.646713399887085,
"epoch": 0.46347757046447,
"grad_norm": 0.89453125,
"learning_rate": 0.0004984994910913232,
"loss": 5.5364,
"mean_token_accuracy": 0.13920091986656188,
"num_tokens": 10258962.0,
"step": 4670
},
{
"entropy": 5.681157016754151,
"epoch": 0.46397379912663755,
"grad_norm": 0.8984375,
"learning_rate": 0.0004984954031774175,
"loss": 5.6515,
"mean_token_accuracy": 0.14098910093307496,
"num_tokens": 10270303.0,
"step": 4675
},
{
"entropy": 5.650723314285278,
"epoch": 0.4644700277888051,
"grad_norm": 0.890625,
"learning_rate": 0.0004984913097213075,
"loss": 5.5564,
"mean_token_accuracy": 0.1381896175444126,
"num_tokens": 10281782.0,
"step": 4680
},
{
"entropy": 5.697708034515381,
"epoch": 0.4649662564509726,
"grad_norm": 1.0546875,
"learning_rate": 0.0004984872107230945,
"loss": 5.5888,
"mean_token_accuracy": 0.14147323966026307,
"num_tokens": 10292937.0,
"step": 4685
},
{
"entropy": 5.6948168754577635,
"epoch": 0.46546248511314015,
"grad_norm": 0.9375,
"learning_rate": 0.0004984831061828803,
"loss": 5.6582,
"mean_token_accuracy": 0.13611662909388542,
"num_tokens": 10305443.0,
"step": 4690
},
{
"entropy": 5.675160598754883,
"epoch": 0.4659587137753077,
"grad_norm": 0.9296875,
"learning_rate": 0.0004984789961007667,
"loss": 5.6069,
"mean_token_accuracy": 0.1387884348630905,
"num_tokens": 10316489.0,
"step": 4695
},
{
"entropy": 5.629416275024414,
"epoch": 0.4664549424374752,
"grad_norm": 1.0390625,
"learning_rate": 0.0004984748804768554,
"loss": 5.6583,
"mean_token_accuracy": 0.1354706920683384,
"num_tokens": 10326690.0,
"step": 4700
},
{
"entropy": 5.758411979675293,
"epoch": 0.46695117109964274,
"grad_norm": 1.0390625,
"learning_rate": 0.0004984707593112487,
"loss": 5.6436,
"mean_token_accuracy": 0.13553577437996864,
"num_tokens": 10337059.0,
"step": 4705
},
{
"entropy": 5.687545251846314,
"epoch": 0.46744739976181027,
"grad_norm": 0.921875,
"learning_rate": 0.0004984666326040487,
"loss": 5.5146,
"mean_token_accuracy": 0.14385205060243605,
"num_tokens": 10347035.0,
"step": 4710
},
{
"entropy": 5.659956359863282,
"epoch": 0.46794362842397774,
"grad_norm": 1.0,
"learning_rate": 0.0004984625003553578,
"loss": 5.6523,
"mean_token_accuracy": 0.13844538107514381,
"num_tokens": 10359492.0,
"step": 4715
},
{
"entropy": 5.71567964553833,
"epoch": 0.4684398570861453,
"grad_norm": 0.9765625,
"learning_rate": 0.0004984583625652783,
"loss": 5.6339,
"mean_token_accuracy": 0.13988834768533706,
"num_tokens": 10370128.0,
"step": 4720
},
{
"entropy": 5.68728232383728,
"epoch": 0.4689360857483128,
"grad_norm": 1.015625,
"learning_rate": 0.000498454219233913,
"loss": 5.6302,
"mean_token_accuracy": 0.13606520891189575,
"num_tokens": 10380276.0,
"step": 4725
},
{
"entropy": 5.617489004135132,
"epoch": 0.46943231441048033,
"grad_norm": 1.0,
"learning_rate": 0.0004984500703613645,
"loss": 5.621,
"mean_token_accuracy": 0.1403240218758583,
"num_tokens": 10392489.0,
"step": 4730
},
{
"entropy": 5.8123973369598385,
"epoch": 0.46992854307264786,
"grad_norm": 0.84375,
"learning_rate": 0.0004984459159477358,
"loss": 5.699,
"mean_token_accuracy": 0.13760418444871902,
"num_tokens": 10404259.0,
"step": 4735
},
{
"entropy": 5.617903470993042,
"epoch": 0.4704247717348154,
"grad_norm": 0.9765625,
"learning_rate": 0.0004984417559931298,
"loss": 5.5535,
"mean_token_accuracy": 0.14490504711866378,
"num_tokens": 10415000.0,
"step": 4740
},
{
"entropy": 5.712674140930176,
"epoch": 0.4709210003969829,
"grad_norm": 1.0078125,
"learning_rate": 0.0004984375904976498,
"loss": 5.6848,
"mean_token_accuracy": 0.1319558687508106,
"num_tokens": 10426078.0,
"step": 4745
},
{
"entropy": 5.728394031524658,
"epoch": 0.47141722905915046,
"grad_norm": 0.890625,
"learning_rate": 0.0004984334194613989,
"loss": 5.6254,
"mean_token_accuracy": 0.13398135229945182,
"num_tokens": 10436838.0,
"step": 4750
},
{
"entropy": 5.687630033493042,
"epoch": 0.471913457721318,
"grad_norm": 1.0703125,
"learning_rate": 0.0004984292428844809,
"loss": 5.5996,
"mean_token_accuracy": 0.13922261223196983,
"num_tokens": 10448544.0,
"step": 4755
},
{
"entropy": 5.7294745445251465,
"epoch": 0.4724096863834855,
"grad_norm": 0.94140625,
"learning_rate": 0.0004984250607669989,
"loss": 5.6673,
"mean_token_accuracy": 0.13299350515007974,
"num_tokens": 10459730.0,
"step": 4760
},
{
"entropy": 5.798430013656616,
"epoch": 0.47290591504565305,
"grad_norm": 0.9609375,
"learning_rate": 0.0004984208731090569,
"loss": 5.7226,
"mean_token_accuracy": 0.13423294350504875,
"num_tokens": 10470257.0,
"step": 4765
},
{
"entropy": 5.7634024143219,
"epoch": 0.4734021437078206,
"grad_norm": 0.9453125,
"learning_rate": 0.0004984166799107587,
"loss": 5.6724,
"mean_token_accuracy": 0.13788133040070533,
"num_tokens": 10480835.0,
"step": 4770
},
{
"entropy": 5.634775018692016,
"epoch": 0.4738983723699881,
"grad_norm": 1.140625,
"learning_rate": 0.0004984124811722081,
"loss": 5.707,
"mean_token_accuracy": 0.1415470227599144,
"num_tokens": 10491720.0,
"step": 4775
},
{
"entropy": 5.792123174667358,
"epoch": 0.47439460103215564,
"grad_norm": 0.9375,
"learning_rate": 0.0004984082768935096,
"loss": 5.7232,
"mean_token_accuracy": 0.13873425871133804,
"num_tokens": 10502313.0,
"step": 4780
},
{
"entropy": 5.727884674072266,
"epoch": 0.47489082969432317,
"grad_norm": 0.98046875,
"learning_rate": 0.0004984040670747672,
"loss": 5.602,
"mean_token_accuracy": 0.14081369638442992,
"num_tokens": 10513333.0,
"step": 4785
},
{
"entropy": 5.630955696105957,
"epoch": 0.47538705835649064,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983998517160852,
"loss": 5.5825,
"mean_token_accuracy": 0.14495161324739456,
"num_tokens": 10522766.0,
"step": 4790
},
{
"entropy": 5.686036777496338,
"epoch": 0.4758832870186582,
"grad_norm": 0.9296875,
"learning_rate": 0.0004983956308175684,
"loss": 5.6204,
"mean_token_accuracy": 0.138974317163229,
"num_tokens": 10534362.0,
"step": 4795
},
{
"entropy": 5.655626821517944,
"epoch": 0.4763795156808257,
"grad_norm": 0.96875,
"learning_rate": 0.0004983914043793212,
"loss": 5.6309,
"mean_token_accuracy": 0.1388360910117626,
"num_tokens": 10545072.0,
"step": 4800
},
{
"entropy": 5.630531120300293,
"epoch": 0.47687574434299324,
"grad_norm": 0.93359375,
"learning_rate": 0.0004983871724014487,
"loss": 5.5675,
"mean_token_accuracy": 0.1337438069283962,
"num_tokens": 10555953.0,
"step": 4805
},
{
"entropy": 5.7464807510375975,
"epoch": 0.47737197300516077,
"grad_norm": 0.890625,
"learning_rate": 0.0004983829348840556,
"loss": 5.6218,
"mean_token_accuracy": 0.14246708303689956,
"num_tokens": 10567757.0,
"step": 4810
},
{
"entropy": 5.569893312454224,
"epoch": 0.4778682016673283,
"grad_norm": 0.94140625,
"learning_rate": 0.000498378691827247,
"loss": 5.516,
"mean_token_accuracy": 0.14553011506795882,
"num_tokens": 10579390.0,
"step": 4815
},
{
"entropy": 5.7074120998382565,
"epoch": 0.4783644303294958,
"grad_norm": 1.046875,
"learning_rate": 0.0004983744432311282,
"loss": 5.7099,
"mean_token_accuracy": 0.13435145616531372,
"num_tokens": 10589624.0,
"step": 4820
},
{
"entropy": 5.779653644561767,
"epoch": 0.47886065899166336,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983701890958046,
"loss": 5.686,
"mean_token_accuracy": 0.13085080310702324,
"num_tokens": 10601927.0,
"step": 4825
},
{
"entropy": 5.600057601928711,
"epoch": 0.4793568876538309,
"grad_norm": 0.92578125,
"learning_rate": 0.0004983659294213815,
"loss": 5.5707,
"mean_token_accuracy": 0.1411229468882084,
"num_tokens": 10614073.0,
"step": 4830
},
{
"entropy": 5.6393006324768065,
"epoch": 0.4798531163159984,
"grad_norm": 1.015625,
"learning_rate": 0.0004983616642079649,
"loss": 5.6096,
"mean_token_accuracy": 0.1381899058818817,
"num_tokens": 10625535.0,
"step": 4835
},
{
"entropy": 5.789355182647705,
"epoch": 0.48034934497816595,
"grad_norm": 1.0390625,
"learning_rate": 0.0004983573934556601,
"loss": 5.5666,
"mean_token_accuracy": 0.1391058512032032,
"num_tokens": 10637160.0,
"step": 4840
},
{
"entropy": 5.716371202468872,
"epoch": 0.4808455736403335,
"grad_norm": 1.015625,
"learning_rate": 0.0004983531171645734,
"loss": 5.6816,
"mean_token_accuracy": 0.1327929750084877,
"num_tokens": 10649061.0,
"step": 4845
},
{
"entropy": 5.577041244506836,
"epoch": 0.481341802302501,
"grad_norm": 0.98046875,
"learning_rate": 0.0004983488353348106,
"loss": 5.549,
"mean_token_accuracy": 0.13940927758812904,
"num_tokens": 10661130.0,
"step": 4850
},
{
"entropy": 5.676693344116211,
"epoch": 0.48183803096466854,
"grad_norm": 0.96875,
"learning_rate": 0.0004983445479664779,
"loss": 5.603,
"mean_token_accuracy": 0.1405676446855068,
"num_tokens": 10673017.0,
"step": 4855
},
{
"entropy": 5.712483978271484,
"epoch": 0.48233425962683607,
"grad_norm": 1.0,
"learning_rate": 0.0004983402550596817,
"loss": 5.6683,
"mean_token_accuracy": 0.13832890316843988,
"num_tokens": 10683268.0,
"step": 4860
},
{
"entropy": 5.790274238586425,
"epoch": 0.48283048828900355,
"grad_norm": 1.0859375,
"learning_rate": 0.0004983359566145285,
"loss": 5.6758,
"mean_token_accuracy": 0.13833464533090592,
"num_tokens": 10694220.0,
"step": 4865
},
{
"entropy": 5.6954034805297855,
"epoch": 0.4833267169511711,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983316526311248,
"loss": 5.5939,
"mean_token_accuracy": 0.1338975928723812,
"num_tokens": 10704829.0,
"step": 4870
},
{
"entropy": 5.71824460029602,
"epoch": 0.4838229456133386,
"grad_norm": 1.078125,
"learning_rate": 0.0004983273431095773,
"loss": 5.7732,
"mean_token_accuracy": 0.127245507389307,
"num_tokens": 10717160.0,
"step": 4875
},
{
"entropy": 5.743475103378296,
"epoch": 0.48431917427550614,
"grad_norm": 1.0078125,
"learning_rate": 0.0004983230280499929,
"loss": 5.5908,
"mean_token_accuracy": 0.14171051159501075,
"num_tokens": 10728638.0,
"step": 4880
},
{
"entropy": 5.668865203857422,
"epoch": 0.48481540293767367,
"grad_norm": 0.96875,
"learning_rate": 0.0004983187074524787,
"loss": 5.6492,
"mean_token_accuracy": 0.13725975155830383,
"num_tokens": 10740122.0,
"step": 4885
},
{
"entropy": 5.668728685379028,
"epoch": 0.4853116315998412,
"grad_norm": 0.984375,
"learning_rate": 0.0004983143813171418,
"loss": 5.5982,
"mean_token_accuracy": 0.1439150556921959,
"num_tokens": 10751103.0,
"step": 4890
},
{
"entropy": 5.711651372909546,
"epoch": 0.48580786026200873,
"grad_norm": 0.94140625,
"learning_rate": 0.0004983100496440895,
"loss": 5.5642,
"mean_token_accuracy": 0.14204839766025543,
"num_tokens": 10761937.0,
"step": 4895
},
{
"entropy": 5.758747482299805,
"epoch": 0.48630408892417626,
"grad_norm": 0.96875,
"learning_rate": 0.000498305712433429,
"loss": 5.7783,
"mean_token_accuracy": 0.13372551202774047,
"num_tokens": 10773527.0,
"step": 4900
},
{
"entropy": 5.714013719558716,
"epoch": 0.4868003175863438,
"grad_norm": 0.9609375,
"learning_rate": 0.000498301369685268,
"loss": 5.6167,
"mean_token_accuracy": 0.1352584406733513,
"num_tokens": 10783873.0,
"step": 4905
},
{
"entropy": 5.625506639480591,
"epoch": 0.4872965462485113,
"grad_norm": 0.96875,
"learning_rate": 0.0004982970213997142,
"loss": 5.4722,
"mean_token_accuracy": 0.14466914534568787,
"num_tokens": 10796124.0,
"step": 4910
},
{
"entropy": 5.614594602584839,
"epoch": 0.48779277491067885,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982926675768756,
"loss": 5.5741,
"mean_token_accuracy": 0.14142126515507697,
"num_tokens": 10806468.0,
"step": 4915
},
{
"entropy": 5.673980665206909,
"epoch": 0.4882890035728464,
"grad_norm": 0.953125,
"learning_rate": 0.0004982883082168599,
"loss": 5.6164,
"mean_token_accuracy": 0.1384105958044529,
"num_tokens": 10817126.0,
"step": 4920
},
{
"entropy": 5.736818456649781,
"epoch": 0.4887852322350139,
"grad_norm": 1.0,
"learning_rate": 0.0004982839433197752,
"loss": 5.6523,
"mean_token_accuracy": 0.13175676092505456,
"num_tokens": 10828985.0,
"step": 4925
},
{
"entropy": 5.645940256118775,
"epoch": 0.48928146089718144,
"grad_norm": 0.890625,
"learning_rate": 0.00049827957288573,
"loss": 5.5669,
"mean_token_accuracy": 0.13916460573673248,
"num_tokens": 10840528.0,
"step": 4930
},
{
"entropy": 5.6961616516113285,
"epoch": 0.48977768955934897,
"grad_norm": 1.0,
"learning_rate": 0.0004982751969148324,
"loss": 5.5796,
"mean_token_accuracy": 0.13913155645132064,
"num_tokens": 10851312.0,
"step": 4935
},
{
"entropy": 5.611951398849487,
"epoch": 0.49027391822151645,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982708154071912,
"loss": 5.5415,
"mean_token_accuracy": 0.14414673000574113,
"num_tokens": 10861882.0,
"step": 4940
},
{
"entropy": 5.652145767211914,
"epoch": 0.490770146883684,
"grad_norm": 0.97265625,
"learning_rate": 0.0004982664283629147,
"loss": 5.6924,
"mean_token_accuracy": 0.13200924098491668,
"num_tokens": 10873187.0,
"step": 4945
},
{
"entropy": 5.704933023452758,
"epoch": 0.4912663755458515,
"grad_norm": 0.98046875,
"learning_rate": 0.0004982620357821118,
"loss": 5.5722,
"mean_token_accuracy": 0.14038059413433074,
"num_tokens": 10884254.0,
"step": 4950
},
{
"entropy": 5.631897354125977,
"epoch": 0.49176260420801904,
"grad_norm": 0.96484375,
"learning_rate": 0.0004982576376648918,
"loss": 5.5608,
"mean_token_accuracy": 0.14028386399149895,
"num_tokens": 10895770.0,
"step": 4955
},
{
"entropy": 5.675318670272827,
"epoch": 0.49225883287018657,
"grad_norm": 0.86328125,
"learning_rate": 0.0004982532340113633,
"loss": 5.5861,
"mean_token_accuracy": 0.1447230599820614,
"num_tokens": 10906436.0,
"step": 4960
},
{
"entropy": 5.669725561141968,
"epoch": 0.4927550615323541,
"grad_norm": 0.91796875,
"learning_rate": 0.0004982488248216357,
"loss": 5.547,
"mean_token_accuracy": 0.14086605086922646,
"num_tokens": 10917655.0,
"step": 4965
},
{
"entropy": 5.621703243255615,
"epoch": 0.49325129019452163,
"grad_norm": 0.9609375,
"learning_rate": 0.0004982444100958182,
"loss": 5.5673,
"mean_token_accuracy": 0.1466407895088196,
"num_tokens": 10928359.0,
"step": 4970
},
{
"entropy": 5.687765741348267,
"epoch": 0.49374751885668916,
"grad_norm": 0.859375,
"learning_rate": 0.0004982399898340205,
"loss": 5.7322,
"mean_token_accuracy": 0.13373062685132026,
"num_tokens": 10940494.0,
"step": 4975
},
{
"entropy": 5.796570491790772,
"epoch": 0.4942437475188567,
"grad_norm": 0.8984375,
"learning_rate": 0.000498235564036352,
"loss": 5.7147,
"mean_token_accuracy": 0.13257310763001443,
"num_tokens": 10951298.0,
"step": 4980
},
{
"entropy": 5.7094841480255125,
"epoch": 0.4947399761810242,
"grad_norm": 1.0546875,
"learning_rate": 0.0004982311327029226,
"loss": 5.6223,
"mean_token_accuracy": 0.14185833930969238,
"num_tokens": 10961680.0,
"step": 4985
},
{
"entropy": 5.760863256454468,
"epoch": 0.49523620484319175,
"grad_norm": 1.3046875,
"learning_rate": 0.0004982266958338421,
"loss": 5.7317,
"mean_token_accuracy": 0.1363532744348049,
"num_tokens": 10974968.0,
"step": 4990
},
{
"entropy": 5.669859409332275,
"epoch": 0.4957324335053593,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982222534292205,
"loss": 5.6744,
"mean_token_accuracy": 0.1317122034728527,
"num_tokens": 10986383.0,
"step": 4995
},
{
"entropy": 5.73351149559021,
"epoch": 0.4962286621675268,
"grad_norm": 0.96875,
"learning_rate": 0.0004982178054891681,
"loss": 5.7072,
"mean_token_accuracy": 0.13707910105586052,
"num_tokens": 10998384.0,
"step": 5000
},
{
"entropy": 5.655369520187378,
"epoch": 0.49672489082969434,
"grad_norm": 0.9765625,
"learning_rate": 0.0004982133520137951,
"loss": 5.4711,
"mean_token_accuracy": 0.14474554359912872,
"num_tokens": 11009420.0,
"step": 5005
},
{
"entropy": 5.682604455947876,
"epoch": 0.4972211194918619,
"grad_norm": 0.921875,
"learning_rate": 0.0004982088930032119,
"loss": 5.6522,
"mean_token_accuracy": 0.13571663498878478,
"num_tokens": 11019601.0,
"step": 5010
},
{
"entropy": 5.638244915008545,
"epoch": 0.4977173481540294,
"grad_norm": 0.9453125,
"learning_rate": 0.0004982044284575293,
"loss": 5.5649,
"mean_token_accuracy": 0.14048463478684425,
"num_tokens": 11031138.0,
"step": 5015
},
{
"entropy": 5.659575700759888,
"epoch": 0.4982135768161969,
"grad_norm": 0.9609375,
"learning_rate": 0.0004981999583768576,
"loss": 5.5857,
"mean_token_accuracy": 0.1401713401079178,
"num_tokens": 11041643.0,
"step": 5020
},
{
"entropy": 5.65928955078125,
"epoch": 0.4987098054783644,
"grad_norm": 1.0390625,
"learning_rate": 0.000498195482761308,
"loss": 5.5992,
"mean_token_accuracy": 0.1333266206085682,
"num_tokens": 11052116.0,
"step": 5025
},
{
"entropy": 5.857709455490112,
"epoch": 0.49920603414053194,
"grad_norm": 0.921875,
"learning_rate": 0.0004981910016109914,
"loss": 5.7914,
"mean_token_accuracy": 0.13015363439917566,
"num_tokens": 11064796.0,
"step": 5030
},
{
"entropy": 5.7740106105804445,
"epoch": 0.49970226280269947,
"grad_norm": 0.96875,
"learning_rate": 0.0004981865149260189,
"loss": 5.6277,
"mean_token_accuracy": 0.1381896987557411,
"num_tokens": 11075029.0,
"step": 5035
},
{
"entropy": 5.702889013290405,
"epoch": 0.5001984914648671,
"grad_norm": 0.8984375,
"learning_rate": 0.0004981820227065016,
"loss": 5.6478,
"mean_token_accuracy": 0.13852666169404984,
"num_tokens": 11086702.0,
"step": 5040
},
{
"entropy": 5.650403547286987,
"epoch": 0.5006947201270345,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981775249525513,
"loss": 5.6376,
"mean_token_accuracy": 0.13566814363002777,
"num_tokens": 11097417.0,
"step": 5045
},
{
"entropy": 5.620085144042969,
"epoch": 0.5011909487892021,
"grad_norm": 1.0,
"learning_rate": 0.0004981730216642791,
"loss": 5.5452,
"mean_token_accuracy": 0.1435072675347328,
"num_tokens": 11108458.0,
"step": 5050
},
{
"entropy": 5.6926917552948,
"epoch": 0.5016871774513696,
"grad_norm": 0.90625,
"learning_rate": 0.0004981685128417969,
"loss": 5.7132,
"mean_token_accuracy": 0.13271809369325638,
"num_tokens": 11121466.0,
"step": 5055
},
{
"entropy": 5.681808233261108,
"epoch": 0.5021834061135371,
"grad_norm": 0.94921875,
"learning_rate": 0.0004981639984852165,
"loss": 5.6499,
"mean_token_accuracy": 0.13097651302814484,
"num_tokens": 11132489.0,
"step": 5060
},
{
"entropy": 5.709500074386597,
"epoch": 0.5026796347757047,
"grad_norm": 1.0703125,
"learning_rate": 0.0004981594785946498,
"loss": 5.6078,
"mean_token_accuracy": 0.13881729319691657,
"num_tokens": 11143087.0,
"step": 5065
},
{
"entropy": 5.733527803421021,
"epoch": 0.5031758634378721,
"grad_norm": 1.0,
"learning_rate": 0.0004981549531702089,
"loss": 5.6467,
"mean_token_accuracy": 0.132865297049284,
"num_tokens": 11154229.0,
"step": 5070
},
{
"entropy": 5.670539712905883,
"epoch": 0.5036720921000397,
"grad_norm": 0.890625,
"learning_rate": 0.000498150422212006,
"loss": 5.6091,
"mean_token_accuracy": 0.1388283282518387,
"num_tokens": 11165698.0,
"step": 5075
},
{
"entropy": 5.66808500289917,
"epoch": 0.5041683207622072,
"grad_norm": 0.96484375,
"learning_rate": 0.0004981458857201534,
"loss": 5.5995,
"mean_token_accuracy": 0.1360068902373314,
"num_tokens": 11177025.0,
"step": 5080
},
{
"entropy": 5.711760187149048,
"epoch": 0.5046645494243748,
"grad_norm": 1.171875,
"learning_rate": 0.0004981413436947639,
"loss": 5.6881,
"mean_token_accuracy": 0.13852130770683288,
"num_tokens": 11188766.0,
"step": 5085
},
{
"entropy": 5.712253618240356,
"epoch": 0.5051607780865423,
"grad_norm": 0.93359375,
"learning_rate": 0.0004981367961359496,
"loss": 5.6996,
"mean_token_accuracy": 0.12998327538371085,
"num_tokens": 11201626.0,
"step": 5090
},
{
"entropy": 5.721598958969116,
"epoch": 0.5056570067487098,
"grad_norm": 1.0078125,
"learning_rate": 0.0004981322430438239,
"loss": 5.6179,
"mean_token_accuracy": 0.13517760410904883,
"num_tokens": 11213975.0,
"step": 5095
},
{
"entropy": 5.614361000061035,
"epoch": 0.5061532354108773,
"grad_norm": 1.0546875,
"learning_rate": 0.0004981276844184991,
"loss": 5.5788,
"mean_token_accuracy": 0.14420121163129807,
"num_tokens": 11223636.0,
"step": 5100
},
{
"entropy": 5.614958047866821,
"epoch": 0.5066494640730449,
"grad_norm": 1.0703125,
"learning_rate": 0.0004981231202600886,
"loss": 5.5454,
"mean_token_accuracy": 0.14166136980056762,
"num_tokens": 11233549.0,
"step": 5105
},
{
"entropy": 5.635168075561523,
"epoch": 0.5071456927352124,
"grad_norm": 0.95703125,
"learning_rate": 0.0004981185505687056,
"loss": 5.5515,
"mean_token_accuracy": 0.1432757556438446,
"num_tokens": 11242915.0,
"step": 5110
},
{
"entropy": 5.779569911956787,
"epoch": 0.50764192139738,
"grad_norm": 0.9140625,
"learning_rate": 0.0004981139753444632,
"loss": 5.6672,
"mean_token_accuracy": 0.13216919153928758,
"num_tokens": 11254370.0,
"step": 5115
},
{
"entropy": 5.714865493774414,
"epoch": 0.5081381500595474,
"grad_norm": 1.09375,
"learning_rate": 0.0004981093945874752,
"loss": 5.6075,
"mean_token_accuracy": 0.14307128861546517,
"num_tokens": 11264226.0,
"step": 5120
},
{
"entropy": 5.591129541397095,
"epoch": 0.508634378721715,
"grad_norm": 1.0234375,
"learning_rate": 0.0004981048082978547,
"loss": 5.5089,
"mean_token_accuracy": 0.14743735790252685,
"num_tokens": 11275304.0,
"step": 5125
},
{
"entropy": 5.604630851745606,
"epoch": 0.5091306073838825,
"grad_norm": 0.98828125,
"learning_rate": 0.0004981002164757158,
"loss": 5.582,
"mean_token_accuracy": 0.13893767669796944,
"num_tokens": 11286179.0,
"step": 5130
},
{
"entropy": 5.739297389984131,
"epoch": 0.50962683604605,
"grad_norm": 1.0234375,
"learning_rate": 0.0004980956191211723,
"loss": 5.6016,
"mean_token_accuracy": 0.138123244792223,
"num_tokens": 11297244.0,
"step": 5135
},
{
"entropy": 5.725628232955932,
"epoch": 0.5101230647082176,
"grad_norm": 0.984375,
"learning_rate": 0.0004980910162343382,
"loss": 5.604,
"mean_token_accuracy": 0.1446719452738762,
"num_tokens": 11307251.0,
"step": 5140
},
{
"entropy": 5.683520317077637,
"epoch": 0.510619293370385,
"grad_norm": 1.0859375,
"learning_rate": 0.0004980864078153275,
"loss": 5.5841,
"mean_token_accuracy": 0.13879474326968194,
"num_tokens": 11318441.0,
"step": 5145
},
{
"entropy": 5.578943824768066,
"epoch": 0.5111155220325526,
"grad_norm": 1.0703125,
"learning_rate": 0.0004980817938642548,
"loss": 5.5705,
"mean_token_accuracy": 0.14160208776593208,
"num_tokens": 11328743.0,
"step": 5150
},
{
"entropy": 5.591680717468262,
"epoch": 0.5116117506947201,
"grad_norm": 0.95703125,
"learning_rate": 0.000498077174381234,
"loss": 5.5533,
"mean_token_accuracy": 0.14184579849243165,
"num_tokens": 11340745.0,
"step": 5155
},
{
"entropy": 5.689578866958618,
"epoch": 0.5121079793568877,
"grad_norm": 1.1015625,
"learning_rate": 0.0004980725493663802,
"loss": 5.6165,
"mean_token_accuracy": 0.14150336384773254,
"num_tokens": 11352060.0,
"step": 5160
},
{
"entropy": 5.689644479751587,
"epoch": 0.5126042080190552,
"grad_norm": 1.0390625,
"learning_rate": 0.0004980679188198078,
"loss": 5.6365,
"mean_token_accuracy": 0.13577819392085075,
"num_tokens": 11363097.0,
"step": 5165
},
{
"entropy": 5.711014699935913,
"epoch": 0.5131004366812227,
"grad_norm": 0.984375,
"learning_rate": 0.0004980632827416317,
"loss": 5.6585,
"mean_token_accuracy": 0.14745849817991258,
"num_tokens": 11373410.0,
"step": 5170
},
{
"entropy": 5.753261661529541,
"epoch": 0.5135966653433902,
"grad_norm": 0.95703125,
"learning_rate": 0.0004980586411319669,
"loss": 5.7296,
"mean_token_accuracy": 0.13415995016694068,
"num_tokens": 11385615.0,
"step": 5175
},
{
"entropy": 5.596803760528564,
"epoch": 0.5140928940055578,
"grad_norm": 0.84765625,
"learning_rate": 0.0004980539939909283,
"loss": 5.5276,
"mean_token_accuracy": 0.13587600886821746,
"num_tokens": 11396492.0,
"step": 5180
},
{
"entropy": 5.7470043182373045,
"epoch": 0.5145891226677253,
"grad_norm": 0.98828125,
"learning_rate": 0.0004980493413186314,
"loss": 5.6269,
"mean_token_accuracy": 0.1332610622048378,
"num_tokens": 11406377.0,
"step": 5185
},
{
"entropy": 5.6779357433319095,
"epoch": 0.5150853513298929,
"grad_norm": 1.046875,
"learning_rate": 0.0004980446831151914,
"loss": 5.6101,
"mean_token_accuracy": 0.14095872342586518,
"num_tokens": 11416862.0,
"step": 5190
},
{
"entropy": 5.6290970802307125,
"epoch": 0.5155815799920603,
"grad_norm": 0.984375,
"learning_rate": 0.0004980400193807239,
"loss": 5.5687,
"mean_token_accuracy": 0.14431979805231093,
"num_tokens": 11428246.0,
"step": 5195
},
{
"entropy": 5.705343961715698,
"epoch": 0.5160778086542279,
"grad_norm": 0.98046875,
"learning_rate": 0.0004980353501153446,
"loss": 5.6012,
"mean_token_accuracy": 0.13967292085289956,
"num_tokens": 11438147.0,
"step": 5200
},
{
"entropy": 5.709432315826416,
"epoch": 0.5165740373163954,
"grad_norm": 1.0078125,
"learning_rate": 0.0004980306753191691,
"loss": 5.7081,
"mean_token_accuracy": 0.1305124878883362,
"num_tokens": 11449154.0,
"step": 5205
},
{
"entropy": 5.751299238204956,
"epoch": 0.5170702659785629,
"grad_norm": 1.0234375,
"learning_rate": 0.0004980259949923135,
"loss": 5.6124,
"mean_token_accuracy": 0.13552996069192885,
"num_tokens": 11459262.0,
"step": 5210
},
{
"entropy": 5.671953725814819,
"epoch": 0.5175664946407305,
"grad_norm": 0.99609375,
"learning_rate": 0.0004980213091348938,
"loss": 5.5976,
"mean_token_accuracy": 0.13675811737775803,
"num_tokens": 11471991.0,
"step": 5215
},
{
"entropy": 5.656467247009277,
"epoch": 0.5180627233028979,
"grad_norm": 1.0859375,
"learning_rate": 0.0004980166177470262,
"loss": 5.5399,
"mean_token_accuracy": 0.13962042704224586,
"num_tokens": 11483176.0,
"step": 5220
},
{
"entropy": 5.702274036407471,
"epoch": 0.5185589519650655,
"grad_norm": 1.0625,
"learning_rate": 0.0004980119208288271,
"loss": 5.6268,
"mean_token_accuracy": 0.13966499343514444,
"num_tokens": 11494015.0,
"step": 5225
},
{
"entropy": 5.585013294219971,
"epoch": 0.519055180627233,
"grad_norm": 0.8828125,
"learning_rate": 0.0004980072183804128,
"loss": 5.5451,
"mean_token_accuracy": 0.13971148952841758,
"num_tokens": 11506365.0,
"step": 5230
},
{
"entropy": 5.696497058868408,
"epoch": 0.5195514092894006,
"grad_norm": 1.03125,
"learning_rate": 0.0004980025104019,
"loss": 5.642,
"mean_token_accuracy": 0.14106786623597145,
"num_tokens": 11517199.0,
"step": 5235
},
{
"entropy": 5.651773166656494,
"epoch": 0.520047637951568,
"grad_norm": 0.98046875,
"learning_rate": 0.0004979977968934056,
"loss": 5.4995,
"mean_token_accuracy": 0.14559734538197516,
"num_tokens": 11527591.0,
"step": 5240
},
{
"entropy": 5.688210153579712,
"epoch": 0.5205438666137356,
"grad_norm": 1.046875,
"learning_rate": 0.0004979930778550463,
"loss": 5.665,
"mean_token_accuracy": 0.1405502274632454,
"num_tokens": 11538657.0,
"step": 5245
},
{
"entropy": 5.663222169876098,
"epoch": 0.5210400952759031,
"grad_norm": 1.0390625,
"learning_rate": 0.0004979883532869392,
"loss": 5.4846,
"mean_token_accuracy": 0.15589443147182463,
"num_tokens": 11549513.0,
"step": 5250
},
{
"entropy": 5.668476867675781,
"epoch": 0.5215363239380707,
"grad_norm": 1.015625,
"learning_rate": 0.0004979836231892013,
"loss": 5.5658,
"mean_token_accuracy": 0.1353978134691715,
"num_tokens": 11560376.0,
"step": 5255
},
{
"entropy": 5.587079095840454,
"epoch": 0.5220325526002382,
"grad_norm": 1.0,
"learning_rate": 0.0004979788875619501,
"loss": 5.557,
"mean_token_accuracy": 0.13797224387526513,
"num_tokens": 11571218.0,
"step": 5260
},
{
"entropy": 5.676311302185058,
"epoch": 0.5225287812624058,
"grad_norm": 0.9453125,
"learning_rate": 0.000497974146405303,
"loss": 5.6194,
"mean_token_accuracy": 0.1385549485683441,
"num_tokens": 11581981.0,
"step": 5265
},
{
"entropy": 5.653447818756104,
"epoch": 0.5230250099245732,
"grad_norm": 0.984375,
"learning_rate": 0.0004979693997193776,
"loss": 5.5821,
"mean_token_accuracy": 0.13908568024635315,
"num_tokens": 11592561.0,
"step": 5270
},
{
"entropy": 5.684787130355835,
"epoch": 0.5235212385867408,
"grad_norm": 1.046875,
"learning_rate": 0.0004979646475042914,
"loss": 5.5489,
"mean_token_accuracy": 0.14116989001631736,
"num_tokens": 11601933.0,
"step": 5275
},
{
"entropy": 5.618759107589722,
"epoch": 0.5240174672489083,
"grad_norm": 1.0234375,
"learning_rate": 0.0004979598897601625,
"loss": 5.4843,
"mean_token_accuracy": 0.14859692752361298,
"num_tokens": 11613938.0,
"step": 5280
},
{
"entropy": 5.7467328071594235,
"epoch": 0.5245136959110758,
"grad_norm": 1.0546875,
"learning_rate": 0.0004979551264871088,
"loss": 5.6766,
"mean_token_accuracy": 0.13326044529676437,
"num_tokens": 11624890.0,
"step": 5285
},
{
"entropy": 5.669732666015625,
"epoch": 0.5250099245732434,
"grad_norm": 0.9375,
"learning_rate": 0.0004979503576852482,
"loss": 5.6504,
"mean_token_accuracy": 0.1438838467001915,
"num_tokens": 11635080.0,
"step": 5290
},
{
"entropy": 5.576575374603271,
"epoch": 0.5255061532354108,
"grad_norm": 1.0,
"learning_rate": 0.0004979455833546994,
"loss": 5.5554,
"mean_token_accuracy": 0.14168847277760505,
"num_tokens": 11645180.0,
"step": 5295
},
{
"entropy": 5.7158714771270756,
"epoch": 0.5260023818975784,
"grad_norm": 0.9765625,
"learning_rate": 0.0004979408034955804,
"loss": 5.5907,
"mean_token_accuracy": 0.13906485587358475,
"num_tokens": 11657599.0,
"step": 5300
},
{
"entropy": 5.743302679061889,
"epoch": 0.5264986105597459,
"grad_norm": 0.98046875,
"learning_rate": 0.0004979360181080099,
"loss": 5.6472,
"mean_token_accuracy": 0.1282064363360405,
"num_tokens": 11669177.0,
"step": 5305
},
{
"entropy": 5.601555204391479,
"epoch": 0.5269948392219135,
"grad_norm": 0.9765625,
"learning_rate": 0.0004979312271921065,
"loss": 5.5045,
"mean_token_accuracy": 0.13916629999876023,
"num_tokens": 11679756.0,
"step": 5310
},
{
"entropy": 5.641785907745361,
"epoch": 0.527491067884081,
"grad_norm": 1.0,
"learning_rate": 0.000497926430747989,
"loss": 5.573,
"mean_token_accuracy": 0.14440273195505143,
"num_tokens": 11689805.0,
"step": 5315
},
{
"entropy": 5.68066782951355,
"epoch": 0.5279872965462485,
"grad_norm": 0.9609375,
"learning_rate": 0.0004979216287757764,
"loss": 5.5939,
"mean_token_accuracy": 0.13892873302102088,
"num_tokens": 11701150.0,
"step": 5320
},
{
"entropy": 5.569952583312988,
"epoch": 0.528483525208416,
"grad_norm": 0.953125,
"learning_rate": 0.000497916821275588,
"loss": 5.5064,
"mean_token_accuracy": 0.14227857440710068,
"num_tokens": 11712256.0,
"step": 5325
},
{
"entropy": 5.684609746932983,
"epoch": 0.5289797538705836,
"grad_norm": 0.95703125,
"learning_rate": 0.0004979120082475424,
"loss": 5.5706,
"mean_token_accuracy": 0.14501353949308396,
"num_tokens": 11722547.0,
"step": 5330
},
{
"entropy": 5.679641437530518,
"epoch": 0.5294759825327511,
"grad_norm": 1.0078125,
"learning_rate": 0.0004979071896917595,
"loss": 5.5902,
"mean_token_accuracy": 0.14043826460838318,
"num_tokens": 11732309.0,
"step": 5335
},
{
"entropy": 5.695653581619263,
"epoch": 0.5299722111949187,
"grad_norm": 1.0,
"learning_rate": 0.0004979023656083586,
"loss": 5.6359,
"mean_token_accuracy": 0.13951552361249925,
"num_tokens": 11743982.0,
"step": 5340
},
{
"entropy": 5.662636709213257,
"epoch": 0.5304684398570861,
"grad_norm": 0.890625,
"learning_rate": 0.0004978975359974594,
"loss": 5.5874,
"mean_token_accuracy": 0.14191087186336518,
"num_tokens": 11755506.0,
"step": 5345
},
{
"entropy": 5.689707183837891,
"epoch": 0.5309646685192537,
"grad_norm": 1.015625,
"learning_rate": 0.0004978927008591816,
"loss": 5.5879,
"mean_token_accuracy": 0.1396624781191349,
"num_tokens": 11767024.0,
"step": 5350
},
{
"entropy": 5.733471298217774,
"epoch": 0.5314608971814212,
"grad_norm": 1.046875,
"learning_rate": 0.0004978878601936451,
"loss": 5.5983,
"mean_token_accuracy": 0.13833921551704406,
"num_tokens": 11777625.0,
"step": 5355
},
{
"entropy": 5.586740970611572,
"epoch": 0.5319571258435887,
"grad_norm": 0.95703125,
"learning_rate": 0.0004978830140009698,
"loss": 5.5512,
"mean_token_accuracy": 0.1478976860642433,
"num_tokens": 11788610.0,
"step": 5360
},
{
"entropy": 5.6329813480377195,
"epoch": 0.5324533545057563,
"grad_norm": 1.0,
"learning_rate": 0.0004978781622812762,
"loss": 5.665,
"mean_token_accuracy": 0.1432223603129387,
"num_tokens": 11800828.0,
"step": 5365
},
{
"entropy": 5.692148733139038,
"epoch": 0.5329495831679237,
"grad_norm": 0.921875,
"learning_rate": 0.0004978733050346844,
"loss": 5.6581,
"mean_token_accuracy": 0.1416023150086403,
"num_tokens": 11811729.0,
"step": 5370
},
{
"entropy": 5.688501024246216,
"epoch": 0.5334458118300913,
"grad_norm": 1.125,
"learning_rate": 0.0004978684422613149,
"loss": 5.6875,
"mean_token_accuracy": 0.13832212015986442,
"num_tokens": 11822081.0,
"step": 5375
},
{
"entropy": 5.696527004241943,
"epoch": 0.5339420404922588,
"grad_norm": 0.92578125,
"learning_rate": 0.0004978635739612882,
"loss": 5.5622,
"mean_token_accuracy": 0.14342110902070998,
"num_tokens": 11832984.0,
"step": 5380
},
{
"entropy": 5.6571948528289795,
"epoch": 0.5344382691544264,
"grad_norm": 0.984375,
"learning_rate": 0.0004978587001347252,
"loss": 5.5092,
"mean_token_accuracy": 0.1376336097717285,
"num_tokens": 11842918.0,
"step": 5385
},
{
"entropy": 5.60179328918457,
"epoch": 0.5349344978165939,
"grad_norm": 0.9453125,
"learning_rate": 0.0004978538207817465,
"loss": 5.5555,
"mean_token_accuracy": 0.14260684698820114,
"num_tokens": 11854364.0,
"step": 5390
},
{
"entropy": 5.65335168838501,
"epoch": 0.5354307264787614,
"grad_norm": 1.078125,
"learning_rate": 0.0004978489359024734,
"loss": 5.6382,
"mean_token_accuracy": 0.1374676614999771,
"num_tokens": 11866400.0,
"step": 5395
},
{
"entropy": 5.74446587562561,
"epoch": 0.5359269551409289,
"grad_norm": 0.94921875,
"learning_rate": 0.0004978440454970268,
"loss": 5.6181,
"mean_token_accuracy": 0.13733726665377616,
"num_tokens": 11877094.0,
"step": 5400
},
{
"entropy": 5.667227172851563,
"epoch": 0.5364231838030965,
"grad_norm": 1.0390625,
"learning_rate": 0.0004978391495655281,
"loss": 5.5841,
"mean_token_accuracy": 0.1390235625207424,
"num_tokens": 11887289.0,
"step": 5405
},
{
"entropy": 5.692342138290405,
"epoch": 0.536919412465264,
"grad_norm": 1.109375,
"learning_rate": 0.0004978342481080987,
"loss": 5.555,
"mean_token_accuracy": 0.1351594567298889,
"num_tokens": 11897700.0,
"step": 5410
},
{
"entropy": 5.5396300792694095,
"epoch": 0.5374156411274316,
"grad_norm": 1.0,
"learning_rate": 0.0004978293411248601,
"loss": 5.4873,
"mean_token_accuracy": 0.14424988850951195,
"num_tokens": 11908623.0,
"step": 5415
},
{
"entropy": 5.636839294433594,
"epoch": 0.537911869789599,
"grad_norm": 1.03125,
"learning_rate": 0.0004978244286159339,
"loss": 5.5237,
"mean_token_accuracy": 0.14568247199058532,
"num_tokens": 11917883.0,
"step": 5420
},
{
"entropy": 5.5925007343292235,
"epoch": 0.5384080984517666,
"grad_norm": 0.859375,
"learning_rate": 0.0004978195105814421,
"loss": 5.548,
"mean_token_accuracy": 0.14374667555093765,
"num_tokens": 11929145.0,
"step": 5425
},
{
"entropy": 5.692426109313965,
"epoch": 0.5389043271139341,
"grad_norm": 0.94140625,
"learning_rate": 0.0004978145870215065,
"loss": 5.5511,
"mean_token_accuracy": 0.13841745555400847,
"num_tokens": 11940073.0,
"step": 5430
},
{
"entropy": 5.681694412231446,
"epoch": 0.5394005557761016,
"grad_norm": 1.015625,
"learning_rate": 0.0004978096579362493,
"loss": 5.6635,
"mean_token_accuracy": 0.13862710148096086,
"num_tokens": 11953140.0,
"step": 5435
},
{
"entropy": 5.658891391754151,
"epoch": 0.5398967844382692,
"grad_norm": 1.1015625,
"learning_rate": 0.0004978047233257927,
"loss": 5.587,
"mean_token_accuracy": 0.14590081721544265,
"num_tokens": 11963853.0,
"step": 5440
},
{
"entropy": 5.626586198806763,
"epoch": 0.5403930131004366,
"grad_norm": 0.9296875,
"learning_rate": 0.0004977997831902591,
"loss": 5.5616,
"mean_token_accuracy": 0.1489775650203228,
"num_tokens": 11975469.0,
"step": 5445
},
{
"entropy": 5.679513597488404,
"epoch": 0.5408892417626042,
"grad_norm": 1.078125,
"learning_rate": 0.0004977948375297709,
"loss": 5.5389,
"mean_token_accuracy": 0.14958526194095612,
"num_tokens": 11985227.0,
"step": 5450
},
{
"entropy": 5.535920810699463,
"epoch": 0.5413854704247717,
"grad_norm": 0.91015625,
"learning_rate": 0.0004977898863444509,
"loss": 5.5404,
"mean_token_accuracy": 0.14351814836263657,
"num_tokens": 11995459.0,
"step": 5455
},
{
"entropy": 5.629694604873658,
"epoch": 0.5418816990869393,
"grad_norm": 0.984375,
"learning_rate": 0.0004977849296344217,
"loss": 5.5895,
"mean_token_accuracy": 0.14124198332428933,
"num_tokens": 12007705.0,
"step": 5460
},
{
"entropy": 5.60025725364685,
"epoch": 0.5423779277491068,
"grad_norm": 0.98046875,
"learning_rate": 0.0004977799673998063,
"loss": 5.4983,
"mean_token_accuracy": 0.1446823038160801,
"num_tokens": 12019729.0,
"step": 5465
},
{
"entropy": 5.651974487304687,
"epoch": 0.5428741564112743,
"grad_norm": 0.94140625,
"learning_rate": 0.0004977749996407278,
"loss": 5.5737,
"mean_token_accuracy": 0.13683584704995155,
"num_tokens": 12031009.0,
"step": 5470
},
{
"entropy": 5.686952304840088,
"epoch": 0.5433703850734418,
"grad_norm": 1.2578125,
"learning_rate": 0.0004977700263573094,
"loss": 5.534,
"mean_token_accuracy": 0.14341633021831512,
"num_tokens": 12041396.0,
"step": 5475
},
{
"entropy": 5.643152189254761,
"epoch": 0.5438666137356094,
"grad_norm": 0.96484375,
"learning_rate": 0.0004977650475496743,
"loss": 5.6958,
"mean_token_accuracy": 0.1346547156572342,
"num_tokens": 12053112.0,
"step": 5480
},
{
"entropy": 5.705150461196899,
"epoch": 0.5443628423977769,
"grad_norm": 0.89453125,
"learning_rate": 0.0004977600632179459,
"loss": 5.5545,
"mean_token_accuracy": 0.13762412071228028,
"num_tokens": 12064823.0,
"step": 5485
},
{
"entropy": 5.654553079605103,
"epoch": 0.5448590710599445,
"grad_norm": 0.96484375,
"learning_rate": 0.0004977550733622482,
"loss": 5.6758,
"mean_token_accuracy": 0.13756117820739747,
"num_tokens": 12076656.0,
"step": 5490
},
{
"entropy": 5.667914390563965,
"epoch": 0.5453552997221119,
"grad_norm": 1.0390625,
"learning_rate": 0.0004977500779827045,
"loss": 5.4727,
"mean_token_accuracy": 0.14255305528640747,
"num_tokens": 12086324.0,
"step": 5495
},
{
"entropy": 5.517617225646973,
"epoch": 0.5458515283842795,
"grad_norm": 0.90625,
"learning_rate": 0.0004977450770794389,
"loss": 5.511,
"mean_token_accuracy": 0.1395778737962246,
"num_tokens": 12098076.0,
"step": 5500
},
{
"entropy": 5.608063220977783,
"epoch": 0.546347757046447,
"grad_norm": 0.859375,
"learning_rate": 0.0004977400706525755,
"loss": 5.5807,
"mean_token_accuracy": 0.1360511936247349,
"num_tokens": 12111282.0,
"step": 5505
},
{
"entropy": 5.660810661315918,
"epoch": 0.5468439857086145,
"grad_norm": 0.95703125,
"learning_rate": 0.0004977350587022381,
"loss": 5.5235,
"mean_token_accuracy": 0.14769414961338043,
"num_tokens": 12121932.0,
"step": 5510
},
{
"entropy": 5.603894424438477,
"epoch": 0.5473402143707821,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977300412285513,
"loss": 5.5111,
"mean_token_accuracy": 0.14588091373443604,
"num_tokens": 12133040.0,
"step": 5515
},
{
"entropy": 5.646938419342041,
"epoch": 0.5478364430329495,
"grad_norm": 0.96875,
"learning_rate": 0.0004977250182316394,
"loss": 5.5897,
"mean_token_accuracy": 0.14238848984241487,
"num_tokens": 12142577.0,
"step": 5520
},
{
"entropy": 5.619527673721313,
"epoch": 0.5483326716951171,
"grad_norm": 1.0078125,
"learning_rate": 0.000497719989711627,
"loss": 5.5633,
"mean_token_accuracy": 0.1430242106318474,
"num_tokens": 12153941.0,
"step": 5525
},
{
"entropy": 5.63915753364563,
"epoch": 0.5488289003572846,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977149556686387,
"loss": 5.621,
"mean_token_accuracy": 0.13971951976418495,
"num_tokens": 12165347.0,
"step": 5530
},
{
"entropy": 5.595114469528198,
"epoch": 0.5493251290194522,
"grad_norm": 1.0234375,
"learning_rate": 0.0004977099161027996,
"loss": 5.4943,
"mean_token_accuracy": 0.14236187785863877,
"num_tokens": 12175705.0,
"step": 5535
},
{
"entropy": 5.639246988296509,
"epoch": 0.5498213576816197,
"grad_norm": 1.0546875,
"learning_rate": 0.0004977048710142342,
"loss": 5.5436,
"mean_token_accuracy": 0.14288176596164703,
"num_tokens": 12186781.0,
"step": 5540
},
{
"entropy": 5.679701423645019,
"epoch": 0.5503175863437872,
"grad_norm": 1.0625,
"learning_rate": 0.0004976998204030681,
"loss": 5.5818,
"mean_token_accuracy": 0.13985614106059074,
"num_tokens": 12197145.0,
"step": 5545
},
{
"entropy": 5.598316383361817,
"epoch": 0.5508138150059547,
"grad_norm": 1.1171875,
"learning_rate": 0.0004976947642694262,
"loss": 5.5746,
"mean_token_accuracy": 0.143681750446558,
"num_tokens": 12207637.0,
"step": 5550
},
{
"entropy": 5.597862482070923,
"epoch": 0.5513100436681223,
"grad_norm": 1.0703125,
"learning_rate": 0.0004976897026134341,
"loss": 5.5142,
"mean_token_accuracy": 0.1443020895123482,
"num_tokens": 12218156.0,
"step": 5555
},
{
"entropy": 5.6665167808532715,
"epoch": 0.5518062723302898,
"grad_norm": 1.09375,
"learning_rate": 0.0004976846354352172,
"loss": 5.5992,
"mean_token_accuracy": 0.1401934579014778,
"num_tokens": 12229987.0,
"step": 5560
},
{
"entropy": 5.672375059127807,
"epoch": 0.5523025009924574,
"grad_norm": 1.0,
"learning_rate": 0.0004976795627349012,
"loss": 5.569,
"mean_token_accuracy": 0.13795140534639358,
"num_tokens": 12240508.0,
"step": 5565
},
{
"entropy": 5.598812580108643,
"epoch": 0.5527987296546248,
"grad_norm": 0.9921875,
"learning_rate": 0.0004976744845126118,
"loss": 5.5086,
"mean_token_accuracy": 0.14796727448701857,
"num_tokens": 12250317.0,
"step": 5570
},
{
"entropy": 5.604941272735596,
"epoch": 0.5532949583167924,
"grad_norm": 0.94921875,
"learning_rate": 0.0004976694007684751,
"loss": 5.5522,
"mean_token_accuracy": 0.14205480813980104,
"num_tokens": 12262104.0,
"step": 5575
},
{
"entropy": 5.717854738235474,
"epoch": 0.5537911869789599,
"grad_norm": 0.97265625,
"learning_rate": 0.000497664311502617,
"loss": 5.5855,
"mean_token_accuracy": 0.1443232163786888,
"num_tokens": 12273259.0,
"step": 5580
},
{
"entropy": 5.621715688705445,
"epoch": 0.5542874156411274,
"grad_norm": 1.0625,
"learning_rate": 0.0004976592167151638,
"loss": 5.6283,
"mean_token_accuracy": 0.1344245620071888,
"num_tokens": 12285504.0,
"step": 5585
},
{
"entropy": 5.691664218902588,
"epoch": 0.554783644303295,
"grad_norm": 1.0625,
"learning_rate": 0.0004976541164062418,
"loss": 5.5621,
"mean_token_accuracy": 0.1360063761472702,
"num_tokens": 12296512.0,
"step": 5590
},
{
"entropy": 5.55618257522583,
"epoch": 0.5552798729654624,
"grad_norm": 0.984375,
"learning_rate": 0.0004976490105759774,
"loss": 5.422,
"mean_token_accuracy": 0.14680247455835344,
"num_tokens": 12307948.0,
"step": 5595
},
{
"entropy": 5.599129724502563,
"epoch": 0.55577610162763,
"grad_norm": 1.078125,
"learning_rate": 0.0004976438992244976,
"loss": 5.5904,
"mean_token_accuracy": 0.1344291388988495,
"num_tokens": 12320020.0,
"step": 5600
},
{
"entropy": 5.646237850189209,
"epoch": 0.5562723302897975,
"grad_norm": 0.97265625,
"learning_rate": 0.0004976387823519286,
"loss": 5.4605,
"mean_token_accuracy": 0.14551302939653396,
"num_tokens": 12331091.0,
"step": 5605
},
{
"entropy": 5.672171211242675,
"epoch": 0.5567685589519651,
"grad_norm": 1.0390625,
"learning_rate": 0.0004976336599583975,
"loss": 5.5695,
"mean_token_accuracy": 0.13842309042811393,
"num_tokens": 12342621.0,
"step": 5610
},
{
"entropy": 5.710894536972046,
"epoch": 0.5572647876141326,
"grad_norm": 1.5625,
"learning_rate": 0.0004976285320440315,
"loss": 5.8307,
"mean_token_accuracy": 0.13015428707003593,
"num_tokens": 12353330.0,
"step": 5615
},
{
"entropy": 5.714410448074341,
"epoch": 0.5577610162763001,
"grad_norm": 0.91015625,
"learning_rate": 0.0004976233986089577,
"loss": 5.5702,
"mean_token_accuracy": 0.1422278732061386,
"num_tokens": 12364418.0,
"step": 5620
},
{
"entropy": 5.62553653717041,
"epoch": 0.5582572449384676,
"grad_norm": 0.9609375,
"learning_rate": 0.0004976182596533031,
"loss": 5.5475,
"mean_token_accuracy": 0.14112602323293685,
"num_tokens": 12377022.0,
"step": 5625
},
{
"entropy": 5.702981090545654,
"epoch": 0.5587534736006352,
"grad_norm": 1.0390625,
"learning_rate": 0.0004976131151771955,
"loss": 5.6238,
"mean_token_accuracy": 0.13408835902810096,
"num_tokens": 12387799.0,
"step": 5630
},
{
"entropy": 5.595159196853638,
"epoch": 0.5592497022628027,
"grad_norm": 0.97265625,
"learning_rate": 0.0004976079651807623,
"loss": 5.5573,
"mean_token_accuracy": 0.13506573513150216,
"num_tokens": 12397387.0,
"step": 5635
},
{
"entropy": 5.600511312484741,
"epoch": 0.5597459309249703,
"grad_norm": 0.9453125,
"learning_rate": 0.0004976028096641313,
"loss": 5.4878,
"mean_token_accuracy": 0.14869325160980223,
"num_tokens": 12407994.0,
"step": 5640
},
{
"entropy": 5.74589524269104,
"epoch": 0.5602421595871377,
"grad_norm": 0.96875,
"learning_rate": 0.0004975976486274302,
"loss": 5.6628,
"mean_token_accuracy": 0.13694743812084198,
"num_tokens": 12419124.0,
"step": 5645
},
{
"entropy": 5.674187421798706,
"epoch": 0.5607383882493053,
"grad_norm": 1.015625,
"learning_rate": 0.0004975924820707872,
"loss": 5.5547,
"mean_token_accuracy": 0.14174269288778304,
"num_tokens": 12428699.0,
"step": 5650
},
{
"entropy": 5.653844833374023,
"epoch": 0.5612346169114728,
"grad_norm": 1.1015625,
"learning_rate": 0.0004975873099943302,
"loss": 5.6174,
"mean_token_accuracy": 0.13702422231435776,
"num_tokens": 12439025.0,
"step": 5655
},
{
"entropy": 5.675407028198242,
"epoch": 0.5617308455736403,
"grad_norm": 0.8984375,
"learning_rate": 0.0004975821323981875,
"loss": 5.5836,
"mean_token_accuracy": 0.14168438836932182,
"num_tokens": 12449439.0,
"step": 5660
},
{
"entropy": 5.67562084197998,
"epoch": 0.5622270742358079,
"grad_norm": 1.0,
"learning_rate": 0.0004975769492824877,
"loss": 5.5282,
"mean_token_accuracy": 0.14081865772604943,
"num_tokens": 12460066.0,
"step": 5665
},
{
"entropy": 5.611133432388305,
"epoch": 0.5627233028979753,
"grad_norm": 1.0,
"learning_rate": 0.0004975717606473589,
"loss": 5.5553,
"mean_token_accuracy": 0.14057415798306466,
"num_tokens": 12471093.0,
"step": 5670
},
{
"entropy": 5.6840332508087155,
"epoch": 0.5632195315601429,
"grad_norm": 1.046875,
"learning_rate": 0.0004975665664929302,
"loss": 5.5663,
"mean_token_accuracy": 0.13668758645653725,
"num_tokens": 12481181.0,
"step": 5675
},
{
"entropy": 5.649426221847534,
"epoch": 0.5637157602223104,
"grad_norm": 1.09375,
"learning_rate": 0.0004975613668193303,
"loss": 5.5932,
"mean_token_accuracy": 0.1397934749722481,
"num_tokens": 12493068.0,
"step": 5680
},
{
"entropy": 5.638387966156006,
"epoch": 0.564211988884478,
"grad_norm": 1.03125,
"learning_rate": 0.000497556161626688,
"loss": 5.5406,
"mean_token_accuracy": 0.14199285358190536,
"num_tokens": 12504025.0,
"step": 5685
},
{
"entropy": 5.657924365997315,
"epoch": 0.5647082175466455,
"grad_norm": 0.9453125,
"learning_rate": 0.0004975509509151325,
"loss": 5.554,
"mean_token_accuracy": 0.1422862321138382,
"num_tokens": 12514567.0,
"step": 5690
},
{
"entropy": 5.7168656349182125,
"epoch": 0.565204446208813,
"grad_norm": 1.015625,
"learning_rate": 0.0004975457346847929,
"loss": 5.6092,
"mean_token_accuracy": 0.13971372693777084,
"num_tokens": 12525088.0,
"step": 5695
},
{
"entropy": 5.613651514053345,
"epoch": 0.5657006748709805,
"grad_norm": 0.9140625,
"learning_rate": 0.0004975405129357986,
"loss": 5.5527,
"mean_token_accuracy": 0.14063025638461113,
"num_tokens": 12536002.0,
"step": 5700
},
{
"entropy": 5.65606837272644,
"epoch": 0.5661969035331481,
"grad_norm": 0.98828125,
"learning_rate": 0.0004975352856682792,
"loss": 5.5765,
"mean_token_accuracy": 0.14051345139741897,
"num_tokens": 12547468.0,
"step": 5705
},
{
"entropy": 5.7039741516113285,
"epoch": 0.5666931321953156,
"grad_norm": 0.9921875,
"learning_rate": 0.0004975300528823642,
"loss": 5.5173,
"mean_token_accuracy": 0.14239700585603715,
"num_tokens": 12557670.0,
"step": 5710
},
{
"entropy": 5.606233406066894,
"epoch": 0.5671893608574832,
"grad_norm": 1.0703125,
"learning_rate": 0.0004975248145781834,
"loss": 5.5891,
"mean_token_accuracy": 0.1419870972633362,
"num_tokens": 12568425.0,
"step": 5715
},
{
"entropy": 5.56580696105957,
"epoch": 0.5676855895196506,
"grad_norm": 0.99609375,
"learning_rate": 0.0004975195707558667,
"loss": 5.5525,
"mean_token_accuracy": 0.14227935597300528,
"num_tokens": 12579138.0,
"step": 5720
},
{
"entropy": 5.643411302566529,
"epoch": 0.5681818181818182,
"grad_norm": 1.0,
"learning_rate": 0.0004975143214155443,
"loss": 5.565,
"mean_token_accuracy": 0.14773009121418,
"num_tokens": 12589743.0,
"step": 5725
},
{
"entropy": 5.697768497467041,
"epoch": 0.5686780468439857,
"grad_norm": 1.109375,
"learning_rate": 0.0004975090665573459,
"loss": 5.5554,
"mean_token_accuracy": 0.1414801999926567,
"num_tokens": 12600648.0,
"step": 5730
},
{
"entropy": 5.628736925125122,
"epoch": 0.5691742755061532,
"grad_norm": 0.984375,
"learning_rate": 0.0004975038061814023,
"loss": 5.5563,
"mean_token_accuracy": 0.14277652502059937,
"num_tokens": 12611369.0,
"step": 5735
},
{
"entropy": 5.631605100631714,
"epoch": 0.5696705041683208,
"grad_norm": 0.91796875,
"learning_rate": 0.0004974985402878437,
"loss": 5.5256,
"mean_token_accuracy": 0.1400889627635479,
"num_tokens": 12621992.0,
"step": 5740
},
{
"entropy": 5.6745518207550045,
"epoch": 0.5701667328304882,
"grad_norm": 1.0546875,
"learning_rate": 0.0004974932688768007,
"loss": 5.6084,
"mean_token_accuracy": 0.13641005828976632,
"num_tokens": 12632463.0,
"step": 5745
},
{
"entropy": 5.578773403167725,
"epoch": 0.5706629614926558,
"grad_norm": 1.0546875,
"learning_rate": 0.0004974879919484042,
"loss": 5.4683,
"mean_token_accuracy": 0.14433844685554503,
"num_tokens": 12642282.0,
"step": 5750
},
{
"entropy": 5.600848770141601,
"epoch": 0.5711591901548233,
"grad_norm": 1.0859375,
"learning_rate": 0.0004974827095027848,
"loss": 5.5055,
"mean_token_accuracy": 0.14577008932828903,
"num_tokens": 12653495.0,
"step": 5755
},
{
"entropy": 5.607797050476075,
"epoch": 0.5716554188169909,
"grad_norm": 0.9375,
"learning_rate": 0.0004974774215400736,
"loss": 5.4378,
"mean_token_accuracy": 0.14466027393937111,
"num_tokens": 12664322.0,
"step": 5760
},
{
"entropy": 5.465302228927612,
"epoch": 0.5721516474791584,
"grad_norm": 1.078125,
"learning_rate": 0.0004974721280604019,
"loss": 5.5297,
"mean_token_accuracy": 0.14000416174530983,
"num_tokens": 12673981.0,
"step": 5765
},
{
"entropy": 5.594118261337281,
"epoch": 0.572647876141326,
"grad_norm": 1.1171875,
"learning_rate": 0.0004974668290639007,
"loss": 5.5568,
"mean_token_accuracy": 0.14178637713193892,
"num_tokens": 12684778.0,
"step": 5770
},
{
"entropy": 5.7291247844696045,
"epoch": 0.5731441048034934,
"grad_norm": 1.03125,
"learning_rate": 0.0004974615245507014,
"loss": 5.5139,
"mean_token_accuracy": 0.14449861347675325,
"num_tokens": 12695287.0,
"step": 5775
},
{
"entropy": 5.5744297981262205,
"epoch": 0.573640333465661,
"grad_norm": 1.078125,
"learning_rate": 0.0004974562145209359,
"loss": 5.5849,
"mean_token_accuracy": 0.1379687674343586,
"num_tokens": 12705773.0,
"step": 5780
},
{
"entropy": 5.574149894714355,
"epoch": 0.5741365621278285,
"grad_norm": 1.078125,
"learning_rate": 0.0004974508989747354,
"loss": 5.5093,
"mean_token_accuracy": 0.1443384602665901,
"num_tokens": 12716783.0,
"step": 5785
},
{
"entropy": 5.629298830032349,
"epoch": 0.5746327907899961,
"grad_norm": 0.984375,
"learning_rate": 0.0004974455779122321,
"loss": 5.5771,
"mean_token_accuracy": 0.14027536660432816,
"num_tokens": 12728146.0,
"step": 5790
},
{
"entropy": 5.6705690860748295,
"epoch": 0.5751290194521635,
"grad_norm": 1.0390625,
"learning_rate": 0.0004974402513335577,
"loss": 5.579,
"mean_token_accuracy": 0.1366913639008999,
"num_tokens": 12739683.0,
"step": 5795
},
{
"entropy": 5.693634462356568,
"epoch": 0.5756252481143311,
"grad_norm": 1.1953125,
"learning_rate": 0.0004974349192388444,
"loss": 5.5689,
"mean_token_accuracy": 0.13982002437114716,
"num_tokens": 12750257.0,
"step": 5800
},
{
"entropy": 5.732922649383545,
"epoch": 0.5761214767764986,
"grad_norm": 0.9765625,
"learning_rate": 0.0004974295816282245,
"loss": 5.6395,
"mean_token_accuracy": 0.13885747790336608,
"num_tokens": 12761055.0,
"step": 5805
},
{
"entropy": 5.630460023880005,
"epoch": 0.5766177054386661,
"grad_norm": 1.078125,
"learning_rate": 0.0004974242385018303,
"loss": 5.5709,
"mean_token_accuracy": 0.13992611467838287,
"num_tokens": 12771014.0,
"step": 5810
},
{
"entropy": 5.592522716522216,
"epoch": 0.5771139341008337,
"grad_norm": 0.984375,
"learning_rate": 0.000497418889859794,
"loss": 5.4865,
"mean_token_accuracy": 0.14401526898145675,
"num_tokens": 12782623.0,
"step": 5815
},
{
"entropy": 5.690291833877564,
"epoch": 0.5776101627630011,
"grad_norm": 1.09375,
"learning_rate": 0.0004974135357022488,
"loss": 5.6016,
"mean_token_accuracy": 0.1452617257833481,
"num_tokens": 12793533.0,
"step": 5820
},
{
"entropy": 5.548589134216309,
"epoch": 0.5781063914251687,
"grad_norm": 1.015625,
"learning_rate": 0.0004974081760293269,
"loss": 5.4656,
"mean_token_accuracy": 0.14004177302122117,
"num_tokens": 12804389.0,
"step": 5825
},
{
"entropy": 5.63265175819397,
"epoch": 0.5786026200873362,
"grad_norm": 0.9609375,
"learning_rate": 0.0004974028108411617,
"loss": 5.5731,
"mean_token_accuracy": 0.14461976215243338,
"num_tokens": 12815672.0,
"step": 5830
},
{
"entropy": 5.738226652145386,
"epoch": 0.5790988487495038,
"grad_norm": 1.0234375,
"learning_rate": 0.0004973974401378859,
"loss": 5.579,
"mean_token_accuracy": 0.14301157742738724,
"num_tokens": 12826113.0,
"step": 5835
},
{
"entropy": 5.558589267730713,
"epoch": 0.5795950774116713,
"grad_norm": 1.0546875,
"learning_rate": 0.0004973920639196329,
"loss": 5.5078,
"mean_token_accuracy": 0.14471369311213494,
"num_tokens": 12835941.0,
"step": 5840
},
{
"entropy": 5.6044713973999025,
"epoch": 0.5800913060738389,
"grad_norm": 0.97265625,
"learning_rate": 0.0004973866821865358,
"loss": 5.5725,
"mean_token_accuracy": 0.13366698920726777,
"num_tokens": 12846186.0,
"step": 5845
},
{
"entropy": 5.69937424659729,
"epoch": 0.5805875347360063,
"grad_norm": 0.97265625,
"learning_rate": 0.0004973812949387283,
"loss": 5.5559,
"mean_token_accuracy": 0.14240086674690247,
"num_tokens": 12856380.0,
"step": 5850
},
{
"entropy": 5.54889440536499,
"epoch": 0.5810837633981739,
"grad_norm": 0.98828125,
"learning_rate": 0.0004973759021763439,
"loss": 5.5947,
"mean_token_accuracy": 0.13854385763406754,
"num_tokens": 12865999.0,
"step": 5855
},
{
"entropy": 5.616566944122314,
"epoch": 0.5815799920603414,
"grad_norm": 1.0,
"learning_rate": 0.0004973705038995162,
"loss": 5.4763,
"mean_token_accuracy": 0.14360741078853606,
"num_tokens": 12876388.0,
"step": 5860
},
{
"entropy": 5.634979963302612,
"epoch": 0.582076220722509,
"grad_norm": 1.0078125,
"learning_rate": 0.0004973651001083792,
"loss": 5.4929,
"mean_token_accuracy": 0.14220962598919867,
"num_tokens": 12885862.0,
"step": 5865
},
{
"entropy": 5.571282958984375,
"epoch": 0.5825724493846764,
"grad_norm": 1.0234375,
"learning_rate": 0.0004973596908030669,
"loss": 5.561,
"mean_token_accuracy": 0.13786889910697936,
"num_tokens": 12897489.0,
"step": 5870
},
{
"entropy": 5.679914665222168,
"epoch": 0.583068678046844,
"grad_norm": 1.046875,
"learning_rate": 0.0004973542759837135,
"loss": 5.5908,
"mean_token_accuracy": 0.13352037519216536,
"num_tokens": 12907395.0,
"step": 5875
},
{
"entropy": 5.6124382495880125,
"epoch": 0.5835649067090115,
"grad_norm": 0.98046875,
"learning_rate": 0.000497348855650453,
"loss": 5.532,
"mean_token_accuracy": 0.14500339403748513,
"num_tokens": 12919439.0,
"step": 5880
},
{
"entropy": 5.6644408226013185,
"epoch": 0.584061135371179,
"grad_norm": 0.9921875,
"learning_rate": 0.0004973434298034201,
"loss": 5.6109,
"mean_token_accuracy": 0.138861233741045,
"num_tokens": 12931892.0,
"step": 5885
},
{
"entropy": 5.736296224594116,
"epoch": 0.5845573640333466,
"grad_norm": 0.984375,
"learning_rate": 0.0004973379984427492,
"loss": 5.6699,
"mean_token_accuracy": 0.1363135479390621,
"num_tokens": 12943189.0,
"step": 5890
},
{
"entropy": 5.682193660736084,
"epoch": 0.585053592695514,
"grad_norm": 1.03125,
"learning_rate": 0.000497332561568575,
"loss": 5.6078,
"mean_token_accuracy": 0.13762683048844337,
"num_tokens": 12954662.0,
"step": 5895
},
{
"entropy": 5.6221672058105465,
"epoch": 0.5855498213576816,
"grad_norm": 0.98046875,
"learning_rate": 0.0004973271191810325,
"loss": 5.5561,
"mean_token_accuracy": 0.14156940579414368,
"num_tokens": 12964924.0,
"step": 5900
},
{
"entropy": 5.638089656829834,
"epoch": 0.5860460500198491,
"grad_norm": 0.94921875,
"learning_rate": 0.0004973216712802564,
"loss": 5.6244,
"mean_token_accuracy": 0.13616342917084695,
"num_tokens": 12976483.0,
"step": 5905
},
{
"entropy": 5.591444826126098,
"epoch": 0.5865422786820167,
"grad_norm": 0.94140625,
"learning_rate": 0.000497316217866382,
"loss": 5.537,
"mean_token_accuracy": 0.1453517943620682,
"num_tokens": 12988054.0,
"step": 5910
},
{
"entropy": 5.677403211593628,
"epoch": 0.5870385073441842,
"grad_norm": 0.9609375,
"learning_rate": 0.0004973107589395444,
"loss": 5.5777,
"mean_token_accuracy": 0.14270464181900025,
"num_tokens": 13000596.0,
"step": 5915
},
{
"entropy": 5.562046909332276,
"epoch": 0.5875347360063518,
"grad_norm": 0.92578125,
"learning_rate": 0.0004973052944998791,
"loss": 5.4187,
"mean_token_accuracy": 0.14594281539320947,
"num_tokens": 13012360.0,
"step": 5920
},
{
"entropy": 5.601492261886596,
"epoch": 0.5880309646685192,
"grad_norm": 1.0546875,
"learning_rate": 0.0004972998245475214,
"loss": 5.576,
"mean_token_accuracy": 0.13968132436275482,
"num_tokens": 13023806.0,
"step": 5925
},
{
"entropy": 5.61679368019104,
"epoch": 0.5885271933306868,
"grad_norm": 1.2265625,
"learning_rate": 0.000497294349082607,
"loss": 5.5009,
"mean_token_accuracy": 0.14438683837652205,
"num_tokens": 13034049.0,
"step": 5930
},
{
"entropy": 5.655051136016846,
"epoch": 0.5890234219928543,
"grad_norm": 0.9609375,
"learning_rate": 0.0004972888681052719,
"loss": 5.5951,
"mean_token_accuracy": 0.1382374234497547,
"num_tokens": 13045006.0,
"step": 5935
},
{
"entropy": 5.7110090255737305,
"epoch": 0.5895196506550219,
"grad_norm": 1.0234375,
"learning_rate": 0.0004972833816156518,
"loss": 5.512,
"mean_token_accuracy": 0.14079046621918678,
"num_tokens": 13055337.0,
"step": 5940
},
{
"entropy": 5.549333000183106,
"epoch": 0.5900158793171894,
"grad_norm": 1.1171875,
"learning_rate": 0.0004972778896138828,
"loss": 5.506,
"mean_token_accuracy": 0.1479167029261589,
"num_tokens": 13065267.0,
"step": 5945
},
{
"entropy": 5.594224071502685,
"epoch": 0.5905121079793569,
"grad_norm": 1.0859375,
"learning_rate": 0.0004972723921001011,
"loss": 5.5886,
"mean_token_accuracy": 0.13957096710801126,
"num_tokens": 13075173.0,
"step": 5950
},
{
"entropy": 5.692317104339599,
"epoch": 0.5910083366415244,
"grad_norm": 0.96484375,
"learning_rate": 0.0004972668890744431,
"loss": 5.5502,
"mean_token_accuracy": 0.13957106247544288,
"num_tokens": 13086848.0,
"step": 5955
},
{
"entropy": 5.640107107162476,
"epoch": 0.5915045653036919,
"grad_norm": 0.94140625,
"learning_rate": 0.0004972613805370452,
"loss": 5.5422,
"mean_token_accuracy": 0.1432167962193489,
"num_tokens": 13096865.0,
"step": 5960
},
{
"entropy": 5.574533891677857,
"epoch": 0.5920007939658595,
"grad_norm": 1.0390625,
"learning_rate": 0.0004972558664880439,
"loss": 5.5043,
"mean_token_accuracy": 0.14752691388130187,
"num_tokens": 13107907.0,
"step": 5965
},
{
"entropy": 5.533886098861695,
"epoch": 0.592497022628027,
"grad_norm": 1.0859375,
"learning_rate": 0.000497250346927576,
"loss": 5.4469,
"mean_token_accuracy": 0.14774303063750266,
"num_tokens": 13118171.0,
"step": 5970
},
{
"entropy": 5.584347295761108,
"epoch": 0.5929932512901945,
"grad_norm": 1.1171875,
"learning_rate": 0.0004972448218557785,
"loss": 5.4152,
"mean_token_accuracy": 0.1491492673754692,
"num_tokens": 13127042.0,
"step": 5975
},
{
"entropy": 5.696653079986572,
"epoch": 0.593489479952362,
"grad_norm": 0.9921875,
"learning_rate": 0.0004972392912727882,
"loss": 5.6837,
"mean_token_accuracy": 0.14060692191123964,
"num_tokens": 13137026.0,
"step": 5980
},
{
"entropy": 5.656273984909058,
"epoch": 0.5939857086145296,
"grad_norm": 0.95703125,
"learning_rate": 0.0004972337551787425,
"loss": 5.5281,
"mean_token_accuracy": 0.13935455083847045,
"num_tokens": 13146947.0,
"step": 5985
},
{
"entropy": 5.662847995758057,
"epoch": 0.5944819372766971,
"grad_norm": 1.0,
"learning_rate": 0.0004972282135737785,
"loss": 5.5781,
"mean_token_accuracy": 0.14084555357694625,
"num_tokens": 13158714.0,
"step": 5990
},
{
"entropy": 5.671388864517212,
"epoch": 0.5949781659388647,
"grad_norm": 1.0,
"learning_rate": 0.0004972226664580337,
"loss": 5.5832,
"mean_token_accuracy": 0.1438387528061867,
"num_tokens": 13170156.0,
"step": 5995
},
{
"entropy": 5.593986845016479,
"epoch": 0.5954743946010321,
"grad_norm": 0.98828125,
"learning_rate": 0.0004972171138316456,
"loss": 5.4974,
"mean_token_accuracy": 0.14140935316681863,
"num_tokens": 13179919.0,
"step": 6000
},
{
"epoch": 0.5954743946010321,
"eval_entropy": 5.511412963133069,
"eval_loss": 5.631018161773682,
"eval_mean_token_accuracy": 0.14324472136627356,
"eval_num_tokens": 13179919.0,
"eval_runtime": 26.7243,
"eval_samples_per_second": 1259.863,
"eval_steps_per_second": 157.497,
"step": 6000
}
],
"logging_steps": 5,
"max_steps": 100750,
"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": 2.2080572983296e+16,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}
Reference in New Issue View Git Blame Copy Permalink