d3bff78c52
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
14112 lines
385 KiB
JSON
14112 lines
385 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 6.958727001491795,
|
|
"eval_steps": 1000,
|
|
"global_step": 7000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.8070969581604,
|
|
"epoch": 0.004972650422675286,
|
|
"grad_norm": 24.75,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.0326,
|
|
"mean_token_accuracy": 0.00021244323579594493,
|
|
"num_tokens": 9311.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.797885847091675,
|
|
"epoch": 0.009945300845350571,
|
|
"grad_norm": 26.625,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.1582,
|
|
"mean_token_accuracy": 0.00015400313423015178,
|
|
"num_tokens": 20562.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.859349775314331,
|
|
"epoch": 0.014917951268025857,
|
|
"grad_norm": 31.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.9712,
|
|
"mean_token_accuracy": 6.825938471592963e-05,
|
|
"num_tokens": 31490.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.023061370849609,
|
|
"epoch": 0.019890601690701143,
|
|
"grad_norm": 37.0,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.4915,
|
|
"mean_token_accuracy": 0.00020137293031439186,
|
|
"num_tokens": 40741.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.819656610488892,
|
|
"epoch": 0.02486325211337643,
|
|
"grad_norm": 41.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.611,
|
|
"mean_token_accuracy": 9.191176504828036e-05,
|
|
"num_tokens": 50749.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 8.102291202545166,
|
|
"epoch": 0.029835902536051714,
|
|
"grad_norm": 16.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.4574,
|
|
"mean_token_accuracy": 6.939625018276274e-05,
|
|
"num_tokens": 60296.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.558936214447021,
|
|
"epoch": 0.034808552958727,
|
|
"grad_norm": 3.25,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.7872,
|
|
"mean_token_accuracy": 0.004098818055354059,
|
|
"num_tokens": 70797.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.733526706695557,
|
|
"epoch": 0.039781203381402286,
|
|
"grad_norm": 2.984375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.6667,
|
|
"mean_token_accuracy": 0.013002976961433887,
|
|
"num_tokens": 80815.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.728536796569824,
|
|
"epoch": 0.04475385380407757,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5279,
|
|
"mean_token_accuracy": 0.01793140098452568,
|
|
"num_tokens": 90326.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.599061870574952,
|
|
"epoch": 0.04972650422675286,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.3877,
|
|
"mean_token_accuracy": 0.01745283491909504,
|
|
"num_tokens": 99800.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.65576114654541,
|
|
"epoch": 0.05469915464942814,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2903,
|
|
"mean_token_accuracy": 0.01605473877862096,
|
|
"num_tokens": 110636.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.719483566284179,
|
|
"epoch": 0.05967180507210343,
|
|
"grad_norm": 1.9765625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1623,
|
|
"mean_token_accuracy": 0.036899705231189725,
|
|
"num_tokens": 120269.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.70643663406372,
|
|
"epoch": 0.06464445549477872,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0881,
|
|
"mean_token_accuracy": 0.04810476265847683,
|
|
"num_tokens": 130176.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.66576566696167,
|
|
"epoch": 0.069617105917454,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 10.0065,
|
|
"mean_token_accuracy": 0.042827858589589596,
|
|
"num_tokens": 140020.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.634183788299561,
|
|
"epoch": 0.07458975634012929,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 10.0029,
|
|
"mean_token_accuracy": 0.050830533541738984,
|
|
"num_tokens": 150713.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.668430137634278,
|
|
"epoch": 0.07956240676280457,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.888,
|
|
"mean_token_accuracy": 0.0580036036670208,
|
|
"num_tokens": 159828.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.654301357269286,
|
|
"epoch": 0.08453505718547986,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8437,
|
|
"mean_token_accuracy": 0.05468556806445122,
|
|
"num_tokens": 169909.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.641754341125488,
|
|
"epoch": 0.08950770760815514,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.7723,
|
|
"mean_token_accuracy": 0.052299515902996065,
|
|
"num_tokens": 179462.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.614963722229003,
|
|
"epoch": 0.09448035803083044,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.7174,
|
|
"mean_token_accuracy": 0.055024531483650205,
|
|
"num_tokens": 188965.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.590071868896484,
|
|
"epoch": 0.09945300845350571,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.7249,
|
|
"mean_token_accuracy": 0.05227088816463947,
|
|
"num_tokens": 200382.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.570422649383545,
|
|
"epoch": 0.10442565887618101,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.5234,
|
|
"mean_token_accuracy": 0.05571662969887257,
|
|
"num_tokens": 209718.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.507248878479004,
|
|
"epoch": 0.10939830929885629,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.4379,
|
|
"mean_token_accuracy": 0.05849384367465973,
|
|
"num_tokens": 219216.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.474326419830323,
|
|
"epoch": 0.11437095972153158,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.3866,
|
|
"mean_token_accuracy": 0.055352504923939706,
|
|
"num_tokens": 228219.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.39169225692749,
|
|
"epoch": 0.11934361014420686,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.3081,
|
|
"mean_token_accuracy": 0.05544135756790638,
|
|
"num_tokens": 237776.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.375651550292968,
|
|
"epoch": 0.12431626056688215,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.2258,
|
|
"mean_token_accuracy": 0.05586103238165378,
|
|
"num_tokens": 248411.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.301455688476562,
|
|
"epoch": 0.12928891098955744,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.2079,
|
|
"mean_token_accuracy": 0.056549127027392385,
|
|
"num_tokens": 258797.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.293347549438476,
|
|
"epoch": 0.13426156141223272,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 9.0945,
|
|
"mean_token_accuracy": 0.050958582386374476,
|
|
"num_tokens": 269228.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.208239078521729,
|
|
"epoch": 0.139234211834908,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 9.0355,
|
|
"mean_token_accuracy": 0.05858226753771305,
|
|
"num_tokens": 279191.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.152443504333496,
|
|
"epoch": 0.14420686225758328,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.9278,
|
|
"mean_token_accuracy": 0.05867091566324234,
|
|
"num_tokens": 289382.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 10.042474365234375,
|
|
"epoch": 0.14917951268025859,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.838,
|
|
"mean_token_accuracy": 0.059500711783766744,
|
|
"num_tokens": 299163.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.933298778533935,
|
|
"epoch": 0.15415216310293386,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.8395,
|
|
"mean_token_accuracy": 0.055494238063693045,
|
|
"num_tokens": 308551.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.883023929595947,
|
|
"epoch": 0.15912481352560914,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.7292,
|
|
"mean_token_accuracy": 0.05697325877845287,
|
|
"num_tokens": 318427.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.771326446533203,
|
|
"epoch": 0.16409746394828442,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.7275,
|
|
"mean_token_accuracy": 0.057022403553128244,
|
|
"num_tokens": 327539.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.62988338470459,
|
|
"epoch": 0.16907011437095973,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.6182,
|
|
"mean_token_accuracy": 0.05923179090023041,
|
|
"num_tokens": 336676.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.472471904754638,
|
|
"epoch": 0.174042764793635,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.5835,
|
|
"mean_token_accuracy": 0.06671808362007141,
|
|
"num_tokens": 345914.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.338830947875977,
|
|
"epoch": 0.1790154152163103,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.5587,
|
|
"mean_token_accuracy": 0.06043006777763367,
|
|
"num_tokens": 355982.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.279397869110108,
|
|
"epoch": 0.1839880656389856,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.5261,
|
|
"mean_token_accuracy": 0.05726887695491314,
|
|
"num_tokens": 365398.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 9.214585781097412,
|
|
"epoch": 0.18896071606166087,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.5059,
|
|
"mean_token_accuracy": 0.05852856226265431,
|
|
"num_tokens": 374519.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 9.065228366851807,
|
|
"epoch": 0.19393336648433615,
|
|
"grad_norm": 0.73828125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.4674,
|
|
"mean_token_accuracy": 0.059703434631228444,
|
|
"num_tokens": 383536.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.976552486419678,
|
|
"epoch": 0.19890601690701143,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.5149,
|
|
"mean_token_accuracy": 0.05934412702918053,
|
|
"num_tokens": 393703.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.97565803527832,
|
|
"epoch": 0.20387866732968674,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.5195,
|
|
"mean_token_accuracy": 0.05305514298379421,
|
|
"num_tokens": 404065.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.924800586700439,
|
|
"epoch": 0.20885131775236201,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.432,
|
|
"mean_token_accuracy": 0.05896747000515461,
|
|
"num_tokens": 413869.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.856037998199463,
|
|
"epoch": 0.2138239681750373,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.4921,
|
|
"mean_token_accuracy": 0.06129511967301369,
|
|
"num_tokens": 424251.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.833969116210938,
|
|
"epoch": 0.21879661859771257,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.4019,
|
|
"mean_token_accuracy": 0.06401792429387569,
|
|
"num_tokens": 432530.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.746544361114502,
|
|
"epoch": 0.22376926902038788,
|
|
"grad_norm": 0.6796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.3987,
|
|
"mean_token_accuracy": 0.06169661693274975,
|
|
"num_tokens": 442335.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.834150791168213,
|
|
"epoch": 0.22874191944306316,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.409,
|
|
"mean_token_accuracy": 0.06196355223655701,
|
|
"num_tokens": 452048.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.751599216461182,
|
|
"epoch": 0.23371456986573844,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.3203,
|
|
"mean_token_accuracy": 0.06898252218961716,
|
|
"num_tokens": 460840.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.737278270721436,
|
|
"epoch": 0.23868722028841372,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.4082,
|
|
"mean_token_accuracy": 0.06640021838247775,
|
|
"num_tokens": 470198.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.729250431060791,
|
|
"epoch": 0.24365987071108902,
|
|
"grad_norm": 0.6953125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.3808,
|
|
"mean_token_accuracy": 0.06511146835982799,
|
|
"num_tokens": 481030.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.744795799255371,
|
|
"epoch": 0.2486325211337643,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.3219,
|
|
"mean_token_accuracy": 0.0689955297857523,
|
|
"num_tokens": 490411.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.608490371704102,
|
|
"epoch": 0.2536051715564396,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.3785,
|
|
"mean_token_accuracy": 0.07227851301431656,
|
|
"num_tokens": 499551.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.726069831848145,
|
|
"epoch": 0.2585778219791149,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.3658,
|
|
"mean_token_accuracy": 0.06381739228963852,
|
|
"num_tokens": 508885.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.69681911468506,
|
|
"epoch": 0.26355047240179014,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.379,
|
|
"mean_token_accuracy": 0.06011901162564755,
|
|
"num_tokens": 518291.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.682203197479248,
|
|
"epoch": 0.26852312282446544,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.2906,
|
|
"mean_token_accuracy": 0.06578158549964427,
|
|
"num_tokens": 528035.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.655262184143066,
|
|
"epoch": 0.27349577324714075,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.3363,
|
|
"mean_token_accuracy": 0.0644611481577158,
|
|
"num_tokens": 538410.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.610585975646973,
|
|
"epoch": 0.278468423669816,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.32,
|
|
"mean_token_accuracy": 0.06107736974954605,
|
|
"num_tokens": 549697.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.659037113189697,
|
|
"epoch": 0.2834410740924913,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.2122,
|
|
"mean_token_accuracy": 0.06606750525534152,
|
|
"num_tokens": 558932.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.545393943786621,
|
|
"epoch": 0.28841372451516656,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.2897,
|
|
"mean_token_accuracy": 0.06308084316551685,
|
|
"num_tokens": 568441.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.5941725730896,
|
|
"epoch": 0.29338637493784187,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.3462,
|
|
"mean_token_accuracy": 0.06668097078800202,
|
|
"num_tokens": 577861.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.560972976684571,
|
|
"epoch": 0.29835902536051717,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.3253,
|
|
"mean_token_accuracy": 0.06575521379709244,
|
|
"num_tokens": 587858.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.65540361404419,
|
|
"epoch": 0.3033316757831924,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.3473,
|
|
"mean_token_accuracy": 0.06740258559584618,
|
|
"num_tokens": 597806.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.687506103515625,
|
|
"epoch": 0.30830432620586773,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.3007,
|
|
"mean_token_accuracy": 0.06727509237825871,
|
|
"num_tokens": 607408.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.586903190612793,
|
|
"epoch": 0.31327697662854304,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.2631,
|
|
"mean_token_accuracy": 0.07164324074983597,
|
|
"num_tokens": 617862.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.565391349792481,
|
|
"epoch": 0.3182496270512183,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.2488,
|
|
"mean_token_accuracy": 0.06584729813039303,
|
|
"num_tokens": 627128.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.553268527984619,
|
|
"epoch": 0.3232222774738936,
|
|
"grad_norm": 0.73046875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.244,
|
|
"mean_token_accuracy": 0.06818093396723271,
|
|
"num_tokens": 636775.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.64507360458374,
|
|
"epoch": 0.32819492789656884,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.304,
|
|
"mean_token_accuracy": 0.07159532345831394,
|
|
"num_tokens": 645906.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.530301666259765,
|
|
"epoch": 0.33316757831924415,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.2612,
|
|
"mean_token_accuracy": 0.06321266070008277,
|
|
"num_tokens": 655601.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.635565948486327,
|
|
"epoch": 0.33814022874191946,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.2708,
|
|
"mean_token_accuracy": 0.0720307134091854,
|
|
"num_tokens": 665682.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.576070308685303,
|
|
"epoch": 0.3431128791645947,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.2087,
|
|
"mean_token_accuracy": 0.0684017900377512,
|
|
"num_tokens": 675200.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.49925184249878,
|
|
"epoch": 0.34808552958727,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.2196,
|
|
"mean_token_accuracy": 0.07055319398641587,
|
|
"num_tokens": 685590.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.569801521301269,
|
|
"epoch": 0.3530581800099453,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.2447,
|
|
"mean_token_accuracy": 0.07266812808811665,
|
|
"num_tokens": 695665.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.579096031188964,
|
|
"epoch": 0.3580308304326206,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.2472,
|
|
"mean_token_accuracy": 0.0726757075637579,
|
|
"num_tokens": 705850.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.673233222961425,
|
|
"epoch": 0.3630034808552959,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.2424,
|
|
"mean_token_accuracy": 0.06867737360298634,
|
|
"num_tokens": 716010.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.486479091644288,
|
|
"epoch": 0.3679761312779712,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.1923,
|
|
"mean_token_accuracy": 0.07114895731210709,
|
|
"num_tokens": 724908.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.458244609832764,
|
|
"epoch": 0.37294878170064644,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.1184,
|
|
"mean_token_accuracy": 0.07498397640883922,
|
|
"num_tokens": 733479.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.580828189849854,
|
|
"epoch": 0.37792143212332174,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1665,
|
|
"mean_token_accuracy": 0.06988212838768959,
|
|
"num_tokens": 743365.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.575690269470215,
|
|
"epoch": 0.382894082545997,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.2256,
|
|
"mean_token_accuracy": 0.07234361991286278,
|
|
"num_tokens": 753720.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.412856292724609,
|
|
"epoch": 0.3878667329686723,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.1242,
|
|
"mean_token_accuracy": 0.07504816465079785,
|
|
"num_tokens": 762524.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.54902229309082,
|
|
"epoch": 0.3928393833913476,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.1719,
|
|
"mean_token_accuracy": 0.07244503125548363,
|
|
"num_tokens": 772549.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.45740785598755,
|
|
"epoch": 0.39781203381402286,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.1328,
|
|
"mean_token_accuracy": 0.07645072191953659,
|
|
"num_tokens": 782304.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.417216968536376,
|
|
"epoch": 0.40278468423669817,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.2433,
|
|
"mean_token_accuracy": 0.0727085243910551,
|
|
"num_tokens": 792010.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.554528999328614,
|
|
"epoch": 0.40775733465937347,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.1127,
|
|
"mean_token_accuracy": 0.07516434043645859,
|
|
"num_tokens": 801667.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.40894250869751,
|
|
"epoch": 0.4127299850820487,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.1652,
|
|
"mean_token_accuracy": 0.07748018577694893,
|
|
"num_tokens": 812385.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.419648170471191,
|
|
"epoch": 0.41770263550472403,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 8.0927,
|
|
"mean_token_accuracy": 0.0735365979373455,
|
|
"num_tokens": 822177.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.465137958526611,
|
|
"epoch": 0.4226752859273993,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.1771,
|
|
"mean_token_accuracy": 0.07697029113769531,
|
|
"num_tokens": 832576.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.430140495300293,
|
|
"epoch": 0.4276479363500746,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0578,
|
|
"mean_token_accuracy": 0.08385262787342071,
|
|
"num_tokens": 842290.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.388404369354248,
|
|
"epoch": 0.4326205867727499,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0957,
|
|
"mean_token_accuracy": 0.0767692718654871,
|
|
"num_tokens": 852500.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.455948162078858,
|
|
"epoch": 0.43759323719542514,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.0283,
|
|
"mean_token_accuracy": 0.0803538903594017,
|
|
"num_tokens": 861401.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.344406127929688,
|
|
"epoch": 0.44256588761810045,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0632,
|
|
"mean_token_accuracy": 0.08101935610175133,
|
|
"num_tokens": 871523.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.46809139251709,
|
|
"epoch": 0.44753853804077576,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 8.1088,
|
|
"mean_token_accuracy": 0.06995238214731217,
|
|
"num_tokens": 881384.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.367013931274414,
|
|
"epoch": 0.452511188463451,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 8.125,
|
|
"mean_token_accuracy": 0.07545166276395321,
|
|
"num_tokens": 890511.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.372640895843507,
|
|
"epoch": 0.4574838388861263,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0739,
|
|
"mean_token_accuracy": 0.07415159381926059,
|
|
"num_tokens": 900064.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.419807624816894,
|
|
"epoch": 0.46245648930880157,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 8.1175,
|
|
"mean_token_accuracy": 0.07342575192451477,
|
|
"num_tokens": 910892.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.448912048339844,
|
|
"epoch": 0.4674291397314769,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 8.102,
|
|
"mean_token_accuracy": 0.07626463770866394,
|
|
"num_tokens": 920566.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.311750793457032,
|
|
"epoch": 0.4724017901541522,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 8.0194,
|
|
"mean_token_accuracy": 0.07890446782112122,
|
|
"num_tokens": 930609.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.425170516967773,
|
|
"epoch": 0.47737444057682743,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 8.0451,
|
|
"mean_token_accuracy": 0.075916238874197,
|
|
"num_tokens": 940383.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.47175989151001,
|
|
"epoch": 0.48234709099950274,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 0.000242,
|
|
"loss": 8.0535,
|
|
"mean_token_accuracy": 0.07686373330652714,
|
|
"num_tokens": 951302.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.380781936645509,
|
|
"epoch": 0.48731974142217804,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 8.0043,
|
|
"mean_token_accuracy": 0.07946270480751991,
|
|
"num_tokens": 961126.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.351362991333009,
|
|
"epoch": 0.4922923918448533,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.9859,
|
|
"mean_token_accuracy": 0.07787158116698265,
|
|
"num_tokens": 970845.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.2554292678833,
|
|
"epoch": 0.4972650422675286,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9292,
|
|
"mean_token_accuracy": 0.08549246788024903,
|
|
"num_tokens": 980919.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.342602348327636,
|
|
"epoch": 0.5022376926902039,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.935,
|
|
"mean_token_accuracy": 0.08340146690607071,
|
|
"num_tokens": 990659.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.281801223754883,
|
|
"epoch": 0.5072103431128792,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9936,
|
|
"mean_token_accuracy": 0.08306360021233558,
|
|
"num_tokens": 999540.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.305860233306884,
|
|
"epoch": 0.5121829935355544,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.9735,
|
|
"mean_token_accuracy": 0.07784088477492332,
|
|
"num_tokens": 1008688.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.2393967628479,
|
|
"epoch": 0.5171556439582298,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.8912,
|
|
"mean_token_accuracy": 0.08796164914965629,
|
|
"num_tokens": 1018310.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.271358203887939,
|
|
"epoch": 0.522128294380905,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.992,
|
|
"mean_token_accuracy": 0.07776187621057033,
|
|
"num_tokens": 1027973.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.25816240310669,
|
|
"epoch": 0.5271009448035803,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.9134,
|
|
"mean_token_accuracy": 0.0823927327990532,
|
|
"num_tokens": 1038141.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.307482242584229,
|
|
"epoch": 0.5320735952262556,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.9954,
|
|
"mean_token_accuracy": 0.0746560201048851,
|
|
"num_tokens": 1047932.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.11481351852417,
|
|
"epoch": 0.5370462456489309,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.815,
|
|
"mean_token_accuracy": 0.08617698922753333,
|
|
"num_tokens": 1056826.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.254195022583009,
|
|
"epoch": 0.5420188960716061,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.9513,
|
|
"mean_token_accuracy": 0.07954822443425655,
|
|
"num_tokens": 1066025.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.241740226745605,
|
|
"epoch": 0.5469915464942815,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.9671,
|
|
"mean_token_accuracy": 0.08237918838858604,
|
|
"num_tokens": 1075452.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.266239452362061,
|
|
"epoch": 0.5519641969169568,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.9517,
|
|
"mean_token_accuracy": 0.08058798983693123,
|
|
"num_tokens": 1085300.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.197205829620362,
|
|
"epoch": 0.556936847339632,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8531,
|
|
"mean_token_accuracy": 0.08992455080151558,
|
|
"num_tokens": 1094548.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.297037601470947,
|
|
"epoch": 0.5619094977623074,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8834,
|
|
"mean_token_accuracy": 0.08289314061403275,
|
|
"num_tokens": 1104321.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.174269676208496,
|
|
"epoch": 0.5668821481849826,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.9814,
|
|
"mean_token_accuracy": 0.07991863191127777,
|
|
"num_tokens": 1115047.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.311392879486084,
|
|
"epoch": 0.5718547986076579,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.9362,
|
|
"mean_token_accuracy": 0.08397412374615669,
|
|
"num_tokens": 1124757.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.168936347961425,
|
|
"epoch": 0.5768274490303331,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.9208,
|
|
"mean_token_accuracy": 0.08150090202689171,
|
|
"num_tokens": 1134255.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 8.277891921997071,
|
|
"epoch": 0.5818000994530085,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.953,
|
|
"mean_token_accuracy": 0.08248294033110141,
|
|
"num_tokens": 1143117.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.268870639801026,
|
|
"epoch": 0.5867727498756837,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8489,
|
|
"mean_token_accuracy": 0.08494550883769988,
|
|
"num_tokens": 1153203.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.201098299026489,
|
|
"epoch": 0.591745400298359,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.8512,
|
|
"mean_token_accuracy": 0.08459036871790886,
|
|
"num_tokens": 1162839.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.038985967636108,
|
|
"epoch": 0.5967180507210343,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.8073,
|
|
"mean_token_accuracy": 0.0855549156665802,
|
|
"num_tokens": 1172580.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.333229446411133,
|
|
"epoch": 0.6016907011437096,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.7966,
|
|
"mean_token_accuracy": 0.08848728835582734,
|
|
"num_tokens": 1183715.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.207336330413819,
|
|
"epoch": 0.6066633515663848,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.9676,
|
|
"mean_token_accuracy": 0.08335011675953866,
|
|
"num_tokens": 1194217.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.153238296508789,
|
|
"epoch": 0.6116360019890602,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.8307,
|
|
"mean_token_accuracy": 0.09300757572054863,
|
|
"num_tokens": 1203297.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.189011287689208,
|
|
"epoch": 0.6166086524117355,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.9124,
|
|
"mean_token_accuracy": 0.08050687834620476,
|
|
"num_tokens": 1213394.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.147252416610717,
|
|
"epoch": 0.6215813028344107,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.8772,
|
|
"mean_token_accuracy": 0.08817379847168923,
|
|
"num_tokens": 1222771.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.183475399017334,
|
|
"epoch": 0.6265539532570861,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.8564,
|
|
"mean_token_accuracy": 0.08216950967907906,
|
|
"num_tokens": 1232171.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.14948844909668,
|
|
"epoch": 0.6315266036797613,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7892,
|
|
"mean_token_accuracy": 0.08736580237746239,
|
|
"num_tokens": 1241640.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 8.020610046386718,
|
|
"epoch": 0.6364992541024366,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.8193,
|
|
"mean_token_accuracy": 0.08613748624920844,
|
|
"num_tokens": 1251785.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 8.17393879890442,
|
|
"epoch": 0.6414719045251119,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.8003,
|
|
"mean_token_accuracy": 0.0838981918990612,
|
|
"num_tokens": 1262327.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.108467531204223,
|
|
"epoch": 0.6464445549477872,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7344,
|
|
"mean_token_accuracy": 0.0935716263949871,
|
|
"num_tokens": 1270871.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 8.041076517105102,
|
|
"epoch": 0.6514172053704624,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7863,
|
|
"mean_token_accuracy": 0.08575954586267472,
|
|
"num_tokens": 1281563.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.092488050460815,
|
|
"epoch": 0.6563898557931377,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.7953,
|
|
"mean_token_accuracy": 0.0862328752875328,
|
|
"num_tokens": 1290867.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.103744792938233,
|
|
"epoch": 0.661362506215813,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.8021,
|
|
"mean_token_accuracy": 0.08568327575922012,
|
|
"num_tokens": 1300384.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 8.164299821853637,
|
|
"epoch": 0.6663351566384883,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7204,
|
|
"mean_token_accuracy": 0.08943556696176529,
|
|
"num_tokens": 1310443.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 8.06074333190918,
|
|
"epoch": 0.6713078070611636,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.6955,
|
|
"mean_token_accuracy": 0.09299010708928108,
|
|
"num_tokens": 1320345.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 8.047345161437988,
|
|
"epoch": 0.6762804574838389,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.8477,
|
|
"mean_token_accuracy": 0.09007203951478004,
|
|
"num_tokens": 1330868.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 8.039288234710693,
|
|
"epoch": 0.6812531079065142,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.7502,
|
|
"mean_token_accuracy": 0.0904244638979435,
|
|
"num_tokens": 1340605.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 8.095660161972045,
|
|
"epoch": 0.6862257583291894,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.7783,
|
|
"mean_token_accuracy": 0.09001534283161164,
|
|
"num_tokens": 1350248.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.915007543563843,
|
|
"epoch": 0.6911984087518648,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7378,
|
|
"mean_token_accuracy": 0.09054447710514069,
|
|
"num_tokens": 1360251.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 8.069479322433471,
|
|
"epoch": 0.69617105917454,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.6944,
|
|
"mean_token_accuracy": 0.09407384619116783,
|
|
"num_tokens": 1370156.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.041491079330445,
|
|
"epoch": 0.7011437095972153,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.8648,
|
|
"mean_token_accuracy": 0.08170138709247113,
|
|
"num_tokens": 1380346.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 8.085822105407715,
|
|
"epoch": 0.7061163600198906,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7145,
|
|
"mean_token_accuracy": 0.08923521786928176,
|
|
"num_tokens": 1390958.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.925909805297851,
|
|
"epoch": 0.7110890104425659,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.7802,
|
|
"mean_token_accuracy": 0.08992756679654121,
|
|
"num_tokens": 1400324.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 8.133356475830078,
|
|
"epoch": 0.7160616608652411,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.6965,
|
|
"mean_token_accuracy": 0.09310642331838608,
|
|
"num_tokens": 1410272.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.908285140991211,
|
|
"epoch": 0.7210343112879165,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.709,
|
|
"mean_token_accuracy": 0.08933724686503411,
|
|
"num_tokens": 1420061.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.976055860519409,
|
|
"epoch": 0.7260069617105918,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6598,
|
|
"mean_token_accuracy": 0.09588965848088264,
|
|
"num_tokens": 1428771.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.941432189941406,
|
|
"epoch": 0.730979612133267,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.6496,
|
|
"mean_token_accuracy": 0.09268148764967918,
|
|
"num_tokens": 1438154.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.898643589019775,
|
|
"epoch": 0.7359522625559424,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6437,
|
|
"mean_token_accuracy": 0.09478027224540711,
|
|
"num_tokens": 1447403.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 8.055390310287475,
|
|
"epoch": 0.7409249129786176,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.7377,
|
|
"mean_token_accuracy": 0.08860993757843971,
|
|
"num_tokens": 1457443.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.929141616821289,
|
|
"epoch": 0.7458975634012929,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.7108,
|
|
"mean_token_accuracy": 0.09391758814454079,
|
|
"num_tokens": 1466783.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 8.047775173187256,
|
|
"epoch": 0.7508702138239681,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.7288,
|
|
"mean_token_accuracy": 0.09357419535517693,
|
|
"num_tokens": 1476320.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.983895683288575,
|
|
"epoch": 0.7558428642466435,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.6311,
|
|
"mean_token_accuracy": 0.09203912019729614,
|
|
"num_tokens": 1487581.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.982823610305786,
|
|
"epoch": 0.7608155146693187,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.7476,
|
|
"mean_token_accuracy": 0.08909644484519959,
|
|
"num_tokens": 1496455.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 8.00187635421753,
|
|
"epoch": 0.765788165091994,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.6811,
|
|
"mean_token_accuracy": 0.0940045453608036,
|
|
"num_tokens": 1506422.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.9452033042907715,
|
|
"epoch": 0.7707608155146694,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.6576,
|
|
"mean_token_accuracy": 0.09396952167153358,
|
|
"num_tokens": 1516593.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.914292907714843,
|
|
"epoch": 0.7757334659373446,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.6933,
|
|
"mean_token_accuracy": 0.09422306939959527,
|
|
"num_tokens": 1526011.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.94219651222229,
|
|
"epoch": 0.7807061163600199,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.6456,
|
|
"mean_token_accuracy": 0.09301053732633591,
|
|
"num_tokens": 1536018.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.867347574234008,
|
|
"epoch": 0.7856787667826952,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.563,
|
|
"mean_token_accuracy": 0.09273268505930901,
|
|
"num_tokens": 1545568.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.888858270645142,
|
|
"epoch": 0.7906514172053705,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5806,
|
|
"mean_token_accuracy": 0.09568437114357949,
|
|
"num_tokens": 1556389.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 8.077014636993407,
|
|
"epoch": 0.7956240676280457,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.6664,
|
|
"mean_token_accuracy": 0.094173014909029,
|
|
"num_tokens": 1565587.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.789323377609253,
|
|
"epoch": 0.8005967180507211,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5872,
|
|
"mean_token_accuracy": 0.09633276090025902,
|
|
"num_tokens": 1574911.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.800200891494751,
|
|
"epoch": 0.8055693684733963,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5444,
|
|
"mean_token_accuracy": 0.09973727241158485,
|
|
"num_tokens": 1584477.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.85718994140625,
|
|
"epoch": 0.8105420188960716,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.5963,
|
|
"mean_token_accuracy": 0.10033143982291222,
|
|
"num_tokens": 1594408.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.81588306427002,
|
|
"epoch": 0.8155146693187469,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5669,
|
|
"mean_token_accuracy": 0.0977571152150631,
|
|
"num_tokens": 1603494.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.876517486572266,
|
|
"epoch": 0.8204873197414222,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.6542,
|
|
"mean_token_accuracy": 0.09184657931327819,
|
|
"num_tokens": 1612994.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.925539493560791,
|
|
"epoch": 0.8254599701640974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5585,
|
|
"mean_token_accuracy": 0.09361765310168266,
|
|
"num_tokens": 1622150.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.804845714569092,
|
|
"epoch": 0.8304326205867727,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5369,
|
|
"mean_token_accuracy": 0.09732099249958992,
|
|
"num_tokens": 1631040.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.844776153564453,
|
|
"epoch": 0.8354052710094481,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.5072,
|
|
"mean_token_accuracy": 0.09882537350058555,
|
|
"num_tokens": 1641519.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.830556344985962,
|
|
"epoch": 0.8403779214321233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5872,
|
|
"mean_token_accuracy": 0.09334329888224602,
|
|
"num_tokens": 1652218.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.954732227325439,
|
|
"epoch": 0.8453505718547986,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.6107,
|
|
"mean_token_accuracy": 0.08996398076415062,
|
|
"num_tokens": 1663763.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.903445243835449,
|
|
"epoch": 0.8503232222774739,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.5868,
|
|
"mean_token_accuracy": 0.0892759084701538,
|
|
"num_tokens": 1674160.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.8312461376190186,
|
|
"epoch": 0.8552958727001492,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.6735,
|
|
"mean_token_accuracy": 0.09993282482028007,
|
|
"num_tokens": 1685138.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.915536546707154,
|
|
"epoch": 0.8602685231228244,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5964,
|
|
"mean_token_accuracy": 0.09194504842162132,
|
|
"num_tokens": 1694847.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.881135511398315,
|
|
"epoch": 0.8652411735454998,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5998,
|
|
"mean_token_accuracy": 0.09743813574314117,
|
|
"num_tokens": 1704503.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.776781034469605,
|
|
"epoch": 0.870213823968175,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.5293,
|
|
"mean_token_accuracy": 0.09748897776007652,
|
|
"num_tokens": 1714011.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.801973628997803,
|
|
"epoch": 0.8751864743908503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.4955,
|
|
"mean_token_accuracy": 0.09765939489006996,
|
|
"num_tokens": 1724254.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.837724447250366,
|
|
"epoch": 0.8801591248135257,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5681,
|
|
"mean_token_accuracy": 0.09028975144028664,
|
|
"num_tokens": 1734696.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.675987672805786,
|
|
"epoch": 0.8851317752362009,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5001,
|
|
"mean_token_accuracy": 0.10280377641320229,
|
|
"num_tokens": 1743650.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.819700384140015,
|
|
"epoch": 0.8901044256588762,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.4778,
|
|
"mean_token_accuracy": 0.10012103915214539,
|
|
"num_tokens": 1753449.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.801957607269287,
|
|
"epoch": 0.8950770760815515,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.6216,
|
|
"mean_token_accuracy": 0.09693230763077736,
|
|
"num_tokens": 1764104.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.745624685287476,
|
|
"epoch": 0.9000497265042268,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.3551,
|
|
"mean_token_accuracy": 0.11145004704594612,
|
|
"num_tokens": 1773172.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.822540807723999,
|
|
"epoch": 0.905022376926902,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4545,
|
|
"mean_token_accuracy": 0.09577107727527619,
|
|
"num_tokens": 1783268.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.680446481704712,
|
|
"epoch": 0.9099950273495774,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.4938,
|
|
"mean_token_accuracy": 0.09600771963596344,
|
|
"num_tokens": 1792169.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.807571697235107,
|
|
"epoch": 0.9149676777722526,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.5072,
|
|
"mean_token_accuracy": 0.09743363708257675,
|
|
"num_tokens": 1802280.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.771829319000244,
|
|
"epoch": 0.9199403281949279,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.49,
|
|
"mean_token_accuracy": 0.09764456748962402,
|
|
"num_tokens": 1812086.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.619400691986084,
|
|
"epoch": 0.9249129786176031,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.3995,
|
|
"mean_token_accuracy": 0.10075560957193375,
|
|
"num_tokens": 1821554.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.7458329677581785,
|
|
"epoch": 0.9298856290402785,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.5033,
|
|
"mean_token_accuracy": 0.09854165613651275,
|
|
"num_tokens": 1831088.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.713270378112793,
|
|
"epoch": 0.9348582794629537,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.4947,
|
|
"mean_token_accuracy": 0.10245847553014756,
|
|
"num_tokens": 1840606.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.618422651290894,
|
|
"epoch": 0.939830929885629,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4168,
|
|
"mean_token_accuracy": 0.10519779622554778,
|
|
"num_tokens": 1850343.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.679243659973144,
|
|
"epoch": 0.9448035803083044,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.5276,
|
|
"mean_token_accuracy": 0.09985006228089333,
|
|
"num_tokens": 1860727.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.74239935874939,
|
|
"epoch": 0.9497762307309796,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.44,
|
|
"mean_token_accuracy": 0.09804870411753655,
|
|
"num_tokens": 1870703.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.642041158676148,
|
|
"epoch": 0.9547488811536549,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.4442,
|
|
"mean_token_accuracy": 0.09999973624944687,
|
|
"num_tokens": 1880694.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.688436794281006,
|
|
"epoch": 0.9597215315763302,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.421,
|
|
"mean_token_accuracy": 0.10089910924434661,
|
|
"num_tokens": 1890195.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.7042396068573,
|
|
"epoch": 0.9646941819990055,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.4035,
|
|
"mean_token_accuracy": 0.09818647429347038,
|
|
"num_tokens": 1900221.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.550181436538696,
|
|
"epoch": 0.9696668324216807,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.4253,
|
|
"mean_token_accuracy": 0.10160460993647576,
|
|
"num_tokens": 1910122.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.721483993530273,
|
|
"epoch": 0.9746394828443561,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.4121,
|
|
"mean_token_accuracy": 0.10157517120242118,
|
|
"num_tokens": 1918814.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.638028144836426,
|
|
"epoch": 0.9796121332670313,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.4151,
|
|
"mean_token_accuracy": 0.10515621416270733,
|
|
"num_tokens": 1928446.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.612313270568848,
|
|
"epoch": 0.9845847836897066,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.4049,
|
|
"mean_token_accuracy": 0.10270627811551095,
|
|
"num_tokens": 1939060.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.744016695022583,
|
|
"epoch": 0.989557434112382,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.3817,
|
|
"mean_token_accuracy": 0.10619868710637093,
|
|
"num_tokens": 1948809.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.560209703445435,
|
|
"epoch": 0.9945300845350572,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.3954,
|
|
"mean_token_accuracy": 0.09983602836728096,
|
|
"num_tokens": 1959232.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9945300845350572,
|
|
"eval_entropy": 7.663880120774019,
|
|
"eval_loss": 7.499252796173096,
|
|
"eval_mean_token_accuracy": 0.09531869574917763,
|
|
"eval_num_tokens": 1959232.0,
|
|
"eval_runtime": 1.968,
|
|
"eval_samples_per_second": 1760.666,
|
|
"eval_steps_per_second": 220.528,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.652824592590332,
|
|
"epoch": 0.9995027349577325,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004999997830922735,
|
|
"loss": 7.3392,
|
|
"mean_token_accuracy": 0.10435211062431335,
|
|
"num_tokens": 1969117.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.638497670491536,
|
|
"epoch": 1.0039781203381402,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999989019053515,
|
|
"loss": 7.3929,
|
|
"mean_token_accuracy": 0.10093366271919674,
|
|
"num_tokens": 1978202.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.454118299484253,
|
|
"epoch": 1.0089507707608154,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0004999973428851536,
|
|
"loss": 7.1938,
|
|
"mean_token_accuracy": 0.11009480580687522,
|
|
"num_tokens": 1988085.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.602870607376099,
|
|
"epoch": 1.0139234211834909,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999951060363766,
|
|
"loss": 7.1998,
|
|
"mean_token_accuracy": 0.10971582233905793,
|
|
"num_tokens": 1996977.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.682427835464478,
|
|
"epoch": 1.0188960716061661,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999921913657592,
|
|
"loss": 7.2819,
|
|
"mean_token_accuracy": 0.1068710558116436,
|
|
"num_tokens": 2006622.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.629259252548218,
|
|
"epoch": 1.0238687220288414,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004999885988820821,
|
|
"loss": 7.339,
|
|
"mean_token_accuracy": 0.10073929280042648,
|
|
"num_tokens": 2016390.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.5728232860565186,
|
|
"epoch": 1.0288413724515166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999843285961683,
|
|
"loss": 7.1706,
|
|
"mean_token_accuracy": 0.10830774009227753,
|
|
"num_tokens": 2026439.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.491561603546143,
|
|
"epoch": 1.0338140228741919,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999793805208822,
|
|
"loss": 7.2628,
|
|
"mean_token_accuracy": 0.1112583301961422,
|
|
"num_tokens": 2035611.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.578234386444092,
|
|
"epoch": 1.0387866732968671,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999737546711305,
|
|
"loss": 7.3013,
|
|
"mean_token_accuracy": 0.09911790117621422,
|
|
"num_tokens": 2047166.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.571400356292725,
|
|
"epoch": 1.0437593237195426,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004999674510638618,
|
|
"loss": 7.2021,
|
|
"mean_token_accuracy": 0.1058596208691597,
|
|
"num_tokens": 2057216.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.565094327926635,
|
|
"epoch": 1.0487319741422179,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999604697180661,
|
|
"loss": 7.2428,
|
|
"mean_token_accuracy": 0.10463513508439064,
|
|
"num_tokens": 2067339.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.47660346031189,
|
|
"epoch": 1.053704624564893,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999528106547759,
|
|
"loss": 7.1306,
|
|
"mean_token_accuracy": 0.11121664196252823,
|
|
"num_tokens": 2077815.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.4952796459197994,
|
|
"epoch": 1.0586772749875684,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999444738970644,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.10029239431023598,
|
|
"num_tokens": 2087738.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.528722715377808,
|
|
"epoch": 1.0636499254102436,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999354594700474,
|
|
"loss": 7.1288,
|
|
"mean_token_accuracy": 0.1104028731584549,
|
|
"num_tokens": 2097246.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.395474624633789,
|
|
"epoch": 1.0686225758329189,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999257674008817,
|
|
"loss": 7.0468,
|
|
"mean_token_accuracy": 0.11465127766132355,
|
|
"num_tokens": 2106415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.435509967803955,
|
|
"epoch": 1.0735952262555943,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999153977187656,
|
|
"loss": 7.0478,
|
|
"mean_token_accuracy": 0.11568540632724762,
|
|
"num_tokens": 2115950.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.555899810791016,
|
|
"epoch": 1.0785678766782696,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499904350454939,
|
|
"loss": 7.2801,
|
|
"mean_token_accuracy": 0.10703531876206399,
|
|
"num_tokens": 2127896.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.466849851608276,
|
|
"epoch": 1.0835405271009448,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004998926256426829,
|
|
"loss": 7.2072,
|
|
"mean_token_accuracy": 0.10295388251543044,
|
|
"num_tokens": 2137238.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.4008838653564455,
|
|
"epoch": 1.08851317752362,
|
|
"grad_norm": 2.140625,
|
|
"learning_rate": 0.0004998802233173196,
|
|
"loss": 7.1327,
|
|
"mean_token_accuracy": 0.10857007130980492,
|
|
"num_tokens": 2147107.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.562657022476197,
|
|
"epoch": 1.0934858279462953,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004998671435162123,
|
|
"loss": 7.2329,
|
|
"mean_token_accuracy": 0.10983152613043785,
|
|
"num_tokens": 2157128.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.367861270904541,
|
|
"epoch": 1.0984584783689706,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998533862787657,
|
|
"loss": 7.139,
|
|
"mean_token_accuracy": 0.10776777565479279,
|
|
"num_tokens": 2167981.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.565176868438721,
|
|
"epoch": 1.1034311287916458,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998389516464244,
|
|
"loss": 7.1452,
|
|
"mean_token_accuracy": 0.11406787484884262,
|
|
"num_tokens": 2177659.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.408087825775146,
|
|
"epoch": 1.1084037792143213,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998238396626746,
|
|
"loss": 7.1367,
|
|
"mean_token_accuracy": 0.11176337599754334,
|
|
"num_tokens": 2186936.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.54493260383606,
|
|
"epoch": 1.1133764296369966,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998080503730427,
|
|
"loss": 7.1795,
|
|
"mean_token_accuracy": 0.11108661815524101,
|
|
"num_tokens": 2197126.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.387998628616333,
|
|
"epoch": 1.1183490800596718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997915838250956,
|
|
"loss": 7.2127,
|
|
"mean_token_accuracy": 0.1092200145125389,
|
|
"num_tokens": 2206337.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.425206995010376,
|
|
"epoch": 1.123321730482347,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004997744400684406,
|
|
"loss": 7.1245,
|
|
"mean_token_accuracy": 0.1112686663866043,
|
|
"num_tokens": 2217112.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.536670255661011,
|
|
"epoch": 1.1282943809050223,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997566191547251,
|
|
"loss": 7.2413,
|
|
"mean_token_accuracy": 0.0996586687862873,
|
|
"num_tokens": 2226239.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.480391931533814,
|
|
"epoch": 1.1332670313276976,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004997381211376364,
|
|
"loss": 7.2435,
|
|
"mean_token_accuracy": 0.10793717727065086,
|
|
"num_tokens": 2235702.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.498857593536377,
|
|
"epoch": 1.138239681750373,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997189460729019,
|
|
"loss": 7.1985,
|
|
"mean_token_accuracy": 0.1158548802137375,
|
|
"num_tokens": 2246005.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.36326003074646,
|
|
"epoch": 1.1432123321730483,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996990940182884,
|
|
"loss": 7.1669,
|
|
"mean_token_accuracy": 0.11077979952096939,
|
|
"num_tokens": 2255534.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.547649478912353,
|
|
"epoch": 1.1481849825957235,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996785650336024,
|
|
"loss": 7.2299,
|
|
"mean_token_accuracy": 0.10904022231698036,
|
|
"num_tokens": 2266374.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.349261856079101,
|
|
"epoch": 1.1531576330183988,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996573591806898,
|
|
"loss": 7.0576,
|
|
"mean_token_accuracy": 0.11764942482113838,
|
|
"num_tokens": 2275614.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.34725432395935,
|
|
"epoch": 1.158130283441074,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004996354765234353,
|
|
"loss": 7.13,
|
|
"mean_token_accuracy": 0.118476003408432,
|
|
"num_tokens": 2284403.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.44727668762207,
|
|
"epoch": 1.1631029338637493,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004996129171277628,
|
|
"loss": 7.1453,
|
|
"mean_token_accuracy": 0.11195796579122544,
|
|
"num_tokens": 2294153.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.328168344497681,
|
|
"epoch": 1.1680755842864246,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995896810616349,
|
|
"loss": 7.1531,
|
|
"mean_token_accuracy": 0.11211650222539901,
|
|
"num_tokens": 2302897.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.60505108833313,
|
|
"epoch": 1.1730482347091,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004995657683950528,
|
|
"loss": 7.2459,
|
|
"mean_token_accuracy": 0.10690148249268532,
|
|
"num_tokens": 2312483.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.164384555816651,
|
|
"epoch": 1.1780208851317753,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004995411792000559,
|
|
"loss": 6.9792,
|
|
"mean_token_accuracy": 0.1140922635793686,
|
|
"num_tokens": 2321911.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.523615169525146,
|
|
"epoch": 1.1829935355544505,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995159135507218,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.11030113250017166,
|
|
"num_tokens": 2331932.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.2158557891845705,
|
|
"epoch": 1.1879661859771258,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000499489971523166,
|
|
"loss": 7.0356,
|
|
"mean_token_accuracy": 0.1182777851819992,
|
|
"num_tokens": 2342264.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.4898519039154055,
|
|
"epoch": 1.192938836399801,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994633531955415,
|
|
"loss": 7.2042,
|
|
"mean_token_accuracy": 0.11488257497549056,
|
|
"num_tokens": 2353068.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.342766809463501,
|
|
"epoch": 1.1979114868224765,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004994360586480391,
|
|
"loss": 7.0076,
|
|
"mean_token_accuracy": 0.1172205351293087,
|
|
"num_tokens": 2361765.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.347688913345337,
|
|
"epoch": 1.2028841372451518,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004994080879628864,
|
|
"loss": 7.0965,
|
|
"mean_token_accuracy": 0.10909276083111763,
|
|
"num_tokens": 2372721.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.394931697845459,
|
|
"epoch": 1.207856787667827,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004993794412243481,
|
|
"loss": 7.1417,
|
|
"mean_token_accuracy": 0.11445093527436256,
|
|
"num_tokens": 2382288.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.369056034088135,
|
|
"epoch": 1.2128294380905023,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004993501185187255,
|
|
"loss": 7.1763,
|
|
"mean_token_accuracy": 0.11123459935188293,
|
|
"num_tokens": 2391953.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.437870502471924,
|
|
"epoch": 1.2178020885131775,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004993201199343565,
|
|
"loss": 7.2157,
|
|
"mean_token_accuracy": 0.1062049113214016,
|
|
"num_tokens": 2402705.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.370413541793823,
|
|
"epoch": 1.2227747389358528,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004992894455616149,
|
|
"loss": 7.1436,
|
|
"mean_token_accuracy": 0.10832376927137374,
|
|
"num_tokens": 2412654.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.446359539031983,
|
|
"epoch": 1.227747389358528,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992580954929108,
|
|
"loss": 7.2253,
|
|
"mean_token_accuracy": 0.11181128472089767,
|
|
"num_tokens": 2422958.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.285688352584839,
|
|
"epoch": 1.2327200397812033,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004992260698226894,
|
|
"loss": 7.0545,
|
|
"mean_token_accuracy": 0.11718959137797355,
|
|
"num_tokens": 2432274.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.359494256973266,
|
|
"epoch": 1.2376926902038787,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991933686474315,
|
|
"loss": 7.2482,
|
|
"mean_token_accuracy": 0.1052965871989727,
|
|
"num_tokens": 2443818.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.433595800399781,
|
|
"epoch": 1.242665340626554,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004991599920656529,
|
|
"loss": 7.1006,
|
|
"mean_token_accuracy": 0.11419800966978073,
|
|
"num_tokens": 2453474.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.299234342575073,
|
|
"epoch": 1.2476379910492292,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004991259401779042,
|
|
"loss": 7.0779,
|
|
"mean_token_accuracy": 0.11456455364823341,
|
|
"num_tokens": 2462715.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.239331007003784,
|
|
"epoch": 1.2526106414719045,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004990912130867704,
|
|
"loss": 6.9991,
|
|
"mean_token_accuracy": 0.11862109750509262,
|
|
"num_tokens": 2473256.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.414499378204345,
|
|
"epoch": 1.2575832918945797,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990558108968703,
|
|
"loss": 7.0494,
|
|
"mean_token_accuracy": 0.11045486181974411,
|
|
"num_tokens": 2482475.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.343642902374268,
|
|
"epoch": 1.2625559423172552,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004990197337148571,
|
|
"loss": 7.1048,
|
|
"mean_token_accuracy": 0.11450405716896057,
|
|
"num_tokens": 2493163.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.409062671661377,
|
|
"epoch": 1.2675285927399305,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000498982981649417,
|
|
"loss": 7.1305,
|
|
"mean_token_accuracy": 0.11062562018632889,
|
|
"num_tokens": 2502062.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.319326114654541,
|
|
"epoch": 1.2725012431626057,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004989455548112695,
|
|
"loss": 7.1171,
|
|
"mean_token_accuracy": 0.11394161209464074,
|
|
"num_tokens": 2511196.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.431995296478272,
|
|
"epoch": 1.277473893585281,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989074533131671,
|
|
"loss": 7.0306,
|
|
"mean_token_accuracy": 0.11063500270247459,
|
|
"num_tokens": 2520492.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.262810373306275,
|
|
"epoch": 1.2824465440079562,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004988686772698945,
|
|
"loss": 7.0687,
|
|
"mean_token_accuracy": 0.11465330570936202,
|
|
"num_tokens": 2530368.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.381555700302124,
|
|
"epoch": 1.2874191944306315,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004988292267982689,
|
|
"loss": 7.1119,
|
|
"mean_token_accuracy": 0.1116415798664093,
|
|
"num_tokens": 2540734.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.262171602249145,
|
|
"epoch": 1.2923918448533067,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004987891020171387,
|
|
"loss": 7.1152,
|
|
"mean_token_accuracy": 0.11137261465191842,
|
|
"num_tokens": 2550907.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.432864618301392,
|
|
"epoch": 1.297364495275982,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987483030473845,
|
|
"loss": 7.1694,
|
|
"mean_token_accuracy": 0.11338013410568237,
|
|
"num_tokens": 2560062.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.376054286956787,
|
|
"epoch": 1.3023371456986574,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004987068300119173,
|
|
"loss": 7.0761,
|
|
"mean_token_accuracy": 0.11074780225753784,
|
|
"num_tokens": 2568840.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.418955898284912,
|
|
"epoch": 1.3073097961213327,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004986646830356792,
|
|
"loss": 7.1592,
|
|
"mean_token_accuracy": 0.10980475470423698,
|
|
"num_tokens": 2578566.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.359957695007324,
|
|
"epoch": 1.312282446544008,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004986218622456423,
|
|
"loss": 7.1542,
|
|
"mean_token_accuracy": 0.11024739816784859,
|
|
"num_tokens": 2589853.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.3830647468566895,
|
|
"epoch": 1.3172550969666832,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498578367770809,
|
|
"loss": 7.1113,
|
|
"mean_token_accuracy": 0.11441166028380394,
|
|
"num_tokens": 2600848.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.276734209060669,
|
|
"epoch": 1.3222277473893587,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004985341997422111,
|
|
"loss": 7.0555,
|
|
"mean_token_accuracy": 0.11667966097593307,
|
|
"num_tokens": 2610730.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.2858161449432375,
|
|
"epoch": 1.327200397812034,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004984893582929093,
|
|
"loss": 7.1501,
|
|
"mean_token_accuracy": 0.10782944262027741,
|
|
"num_tokens": 2621129.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.40053129196167,
|
|
"epoch": 1.3321730482347092,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004984438435579933,
|
|
"loss": 7.0772,
|
|
"mean_token_accuracy": 0.11545051857829094,
|
|
"num_tokens": 2630306.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.299241733551026,
|
|
"epoch": 1.3371456986573844,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983976556745812,
|
|
"loss": 7.1693,
|
|
"mean_token_accuracy": 0.10763240307569504,
|
|
"num_tokens": 2639874.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.334661054611206,
|
|
"epoch": 1.3421183490800597,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000498350794781819,
|
|
"loss": 7.1484,
|
|
"mean_token_accuracy": 0.10904535055160522,
|
|
"num_tokens": 2650340.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.202293968200683,
|
|
"epoch": 1.347090999502735,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983032610208801,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.12317663207650184,
|
|
"num_tokens": 2659541.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.332413005828857,
|
|
"epoch": 1.3520636499254102,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004982550545349649,
|
|
"loss": 7.0959,
|
|
"mean_token_accuracy": 0.11287071779370308,
|
|
"num_tokens": 2669151.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.274061250686645,
|
|
"epoch": 1.3570363003480854,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004982061754693008,
|
|
"loss": 6.9934,
|
|
"mean_token_accuracy": 0.12214738503098488,
|
|
"num_tokens": 2678867.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.2326271533966064,
|
|
"epoch": 1.362008950770761,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004981566239711412,
|
|
"loss": 7.1043,
|
|
"mean_token_accuracy": 0.1091894619166851,
|
|
"num_tokens": 2687572.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.35388331413269,
|
|
"epoch": 1.3669816011934361,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004981064001897651,
|
|
"loss": 7.0386,
|
|
"mean_token_accuracy": 0.11529299393296241,
|
|
"num_tokens": 2696922.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.23381290435791,
|
|
"epoch": 1.3719542516161114,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004980555042764772,
|
|
"loss": 7.0833,
|
|
"mean_token_accuracy": 0.1171937845647335,
|
|
"num_tokens": 2707465.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.312092256546021,
|
|
"epoch": 1.3769269020387866,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004980039363846069,
|
|
"loss": 6.9779,
|
|
"mean_token_accuracy": 0.12314596474170685,
|
|
"num_tokens": 2718708.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.292709255218506,
|
|
"epoch": 1.381899552461462,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000497951696669508,
|
|
"loss": 7.0409,
|
|
"mean_token_accuracy": 0.11527033746242524,
|
|
"num_tokens": 2728862.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.310226726531982,
|
|
"epoch": 1.3868722028841374,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004978987852885579,
|
|
"loss": 7.0155,
|
|
"mean_token_accuracy": 0.11222517043352127,
|
|
"num_tokens": 2738546.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.264519262313843,
|
|
"epoch": 1.3918448533068126,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004978452024011582,
|
|
"loss": 6.9755,
|
|
"mean_token_accuracy": 0.11780327409505845,
|
|
"num_tokens": 2747885.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.181573963165283,
|
|
"epoch": 1.3968175037294879,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000497790948168733,
|
|
"loss": 7.001,
|
|
"mean_token_accuracy": 0.1201067291200161,
|
|
"num_tokens": 2756851.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.388041162490845,
|
|
"epoch": 1.4017901541521631,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004977360227547288,
|
|
"loss": 7.0365,
|
|
"mean_token_accuracy": 0.11356016471982003,
|
|
"num_tokens": 2766309.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.225631046295166,
|
|
"epoch": 1.4067628045748384,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004976804263246146,
|
|
"loss": 7.0066,
|
|
"mean_token_accuracy": 0.11348523944616318,
|
|
"num_tokens": 2775679.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.272976732254028,
|
|
"epoch": 1.4117354549975136,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004976241590458803,
|
|
"loss": 6.9845,
|
|
"mean_token_accuracy": 0.116749869287014,
|
|
"num_tokens": 2784610.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.236735868453979,
|
|
"epoch": 1.4167081054201889,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004975672210880372,
|
|
"loss": 7.0709,
|
|
"mean_token_accuracy": 0.11795241460204124,
|
|
"num_tokens": 2793969.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.232448244094849,
|
|
"epoch": 1.4216807558428641,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004975096126226171,
|
|
"loss": 6.9142,
|
|
"mean_token_accuracy": 0.11798690259456635,
|
|
"num_tokens": 2803096.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.143263149261474,
|
|
"epoch": 1.4266534062655396,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004974513338231715,
|
|
"loss": 6.9824,
|
|
"mean_token_accuracy": 0.1235393963754177,
|
|
"num_tokens": 2812320.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.356040811538696,
|
|
"epoch": 1.4316260566882149,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004973923848652715,
|
|
"loss": 7.0575,
|
|
"mean_token_accuracy": 0.11412188708782196,
|
|
"num_tokens": 2821393.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.213506650924683,
|
|
"epoch": 1.43659870711089,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004973327659265073,
|
|
"loss": 7.018,
|
|
"mean_token_accuracy": 0.11950061470270157,
|
|
"num_tokens": 2831182.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.284598016738892,
|
|
"epoch": 1.4415713575335654,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004972724771864874,
|
|
"loss": 6.9772,
|
|
"mean_token_accuracy": 0.11995619013905526,
|
|
"num_tokens": 2840429.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.131456136703491,
|
|
"epoch": 1.4465440079562406,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004972115188268378,
|
|
"loss": 6.9833,
|
|
"mean_token_accuracy": 0.1209810994565487,
|
|
"num_tokens": 2849443.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.217880344390869,
|
|
"epoch": 1.451516658378916,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004971498910312024,
|
|
"loss": 6.9478,
|
|
"mean_token_accuracy": 0.12011919394135476,
|
|
"num_tokens": 2858908.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.324192714691162,
|
|
"epoch": 1.4564893088015913,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004970875939852412,
|
|
"loss": 7.0378,
|
|
"mean_token_accuracy": 0.11806118115782738,
|
|
"num_tokens": 2869615.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.232141828536987,
|
|
"epoch": 1.4614619592242666,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000497024627876631,
|
|
"loss": 6.9765,
|
|
"mean_token_accuracy": 0.11535097137093545,
|
|
"num_tokens": 2879007.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.193434190750122,
|
|
"epoch": 1.4664346096469418,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004969609928950639,
|
|
"loss": 6.8729,
|
|
"mean_token_accuracy": 0.12043445333838462,
|
|
"num_tokens": 2887527.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.175161981582642,
|
|
"epoch": 1.471407260069617,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004968966892322468,
|
|
"loss": 6.9183,
|
|
"mean_token_accuracy": 0.11960679963231087,
|
|
"num_tokens": 2897137.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.205398511886597,
|
|
"epoch": 1.4763799104922923,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004968317170819018,
|
|
"loss": 7.0144,
|
|
"mean_token_accuracy": 0.11745168343186378,
|
|
"num_tokens": 2908346.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.256870937347412,
|
|
"epoch": 1.4813525609149676,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004967660766397642,
|
|
"loss": 7.0309,
|
|
"mean_token_accuracy": 0.11733275279402733,
|
|
"num_tokens": 2918438.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.29434552192688,
|
|
"epoch": 1.4863252113376428,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004966997681035829,
|
|
"loss": 7.1441,
|
|
"mean_token_accuracy": 0.11370202526450157,
|
|
"num_tokens": 2928096.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.224684619903565,
|
|
"epoch": 1.4912978617603183,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004966327916731195,
|
|
"loss": 6.9681,
|
|
"mean_token_accuracy": 0.11829992830753326,
|
|
"num_tokens": 2938150.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.148546743392944,
|
|
"epoch": 1.4962705121829936,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004965651475501476,
|
|
"loss": 6.9413,
|
|
"mean_token_accuracy": 0.12215470746159554,
|
|
"num_tokens": 2949410.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.199129438400268,
|
|
"epoch": 1.5012431626056688,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004964968359384524,
|
|
"loss": 6.9022,
|
|
"mean_token_accuracy": 0.12959329560399055,
|
|
"num_tokens": 2958564.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.108926105499267,
|
|
"epoch": 1.506215813028344,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004964278570438299,
|
|
"loss": 6.9691,
|
|
"mean_token_accuracy": 0.11823615282773972,
|
|
"num_tokens": 2968750.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.239379596710205,
|
|
"epoch": 1.5111884634510195,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004963582110740865,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.12333661541342736,
|
|
"num_tokens": 2978397.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.22562575340271,
|
|
"epoch": 1.5161611138736948,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004962878982390381,
|
|
"loss": 7.0386,
|
|
"mean_token_accuracy": 0.11300068721175194,
|
|
"num_tokens": 2987625.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.173707628250122,
|
|
"epoch": 1.52113376429637,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004962169187505097,
|
|
"loss": 6.9812,
|
|
"mean_token_accuracy": 0.12145868018269539,
|
|
"num_tokens": 2997266.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.266701555252075,
|
|
"epoch": 1.5261064147190453,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004961452728223343,
|
|
"loss": 7.0008,
|
|
"mean_token_accuracy": 0.11692268773913383,
|
|
"num_tokens": 3007474.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.255027914047242,
|
|
"epoch": 1.5310790651417205,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004960729606703535,
|
|
"loss": 6.9688,
|
|
"mean_token_accuracy": 0.11842977330088615,
|
|
"num_tokens": 3016270.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.141047811508178,
|
|
"epoch": 1.5360517155643958,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004959999825124147,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.1175768367946148,
|
|
"num_tokens": 3027397.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.254582834243775,
|
|
"epoch": 1.541024365987071,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000495926338568373,
|
|
"loss": 6.9896,
|
|
"mean_token_accuracy": 0.11926171481609345,
|
|
"num_tokens": 3037324.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.117153739929199,
|
|
"epoch": 1.5459970164097463,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004958520290600884,
|
|
"loss": 6.9562,
|
|
"mean_token_accuracy": 0.11926570013165474,
|
|
"num_tokens": 3048315.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.127697944641113,
|
|
"epoch": 1.5509696668324215,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004957770542114262,
|
|
"loss": 6.8622,
|
|
"mean_token_accuracy": 0.12573121190071107,
|
|
"num_tokens": 3057875.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.2254057884216305,
|
|
"epoch": 1.555942317255097,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004957014142482563,
|
|
"loss": 6.9557,
|
|
"mean_token_accuracy": 0.12032727524638176,
|
|
"num_tokens": 3068006.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.233620023727417,
|
|
"epoch": 1.5609149676777723,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004956251093984521,
|
|
"loss": 6.9736,
|
|
"mean_token_accuracy": 0.12568886056542397,
|
|
"num_tokens": 3079450.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.086822748184204,
|
|
"epoch": 1.5658876181004475,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004955481398918902,
|
|
"loss": 6.769,
|
|
"mean_token_accuracy": 0.1328391946852207,
|
|
"num_tokens": 3088385.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.208517169952392,
|
|
"epoch": 1.5708602685231228,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004954705059604495,
|
|
"loss": 6.911,
|
|
"mean_token_accuracy": 0.11698652803897858,
|
|
"num_tokens": 3097854.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.269832944869995,
|
|
"epoch": 1.5758329189457982,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004953922078380104,
|
|
"loss": 7.0284,
|
|
"mean_token_accuracy": 0.11298786252737045,
|
|
"num_tokens": 3107877.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.0874732494354244,
|
|
"epoch": 1.5808055693684735,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004953132457604544,
|
|
"loss": 6.9755,
|
|
"mean_token_accuracy": 0.11983944177627563,
|
|
"num_tokens": 3117967.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.101063203811646,
|
|
"epoch": 1.5857782197911487,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004952336199656632,
|
|
"loss": 6.8747,
|
|
"mean_token_accuracy": 0.12173356637358665,
|
|
"num_tokens": 3129827.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.202299356460571,
|
|
"epoch": 1.590750870213824,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004951533306935181,
|
|
"loss": 6.8696,
|
|
"mean_token_accuracy": 0.12329825907945632,
|
|
"num_tokens": 3138779.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.0769623756408695,
|
|
"epoch": 1.5957235206364992,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004950723781858991,
|
|
"loss": 6.905,
|
|
"mean_token_accuracy": 0.1304717816412449,
|
|
"num_tokens": 3148741.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.097968006134034,
|
|
"epoch": 1.6006961710591745,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004949907626866844,
|
|
"loss": 6.9459,
|
|
"mean_token_accuracy": 0.12365822419524193,
|
|
"num_tokens": 3158964.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.122892951965332,
|
|
"epoch": 1.6056688214818498,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004949084844417493,
|
|
"loss": 6.7629,
|
|
"mean_token_accuracy": 0.1259840451180935,
|
|
"num_tokens": 3168347.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.102373504638672,
|
|
"epoch": 1.610641471904525,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004948255436989657,
|
|
"loss": 6.9673,
|
|
"mean_token_accuracy": 0.11866552829742431,
|
|
"num_tokens": 3178276.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.098875951766968,
|
|
"epoch": 1.6156141223272003,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004947419407082017,
|
|
"loss": 6.9047,
|
|
"mean_token_accuracy": 0.11665757969021798,
|
|
"num_tokens": 3186975.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.10897798538208,
|
|
"epoch": 1.6205867727498757,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004946576757213202,
|
|
"loss": 6.7687,
|
|
"mean_token_accuracy": 0.13888816609978677,
|
|
"num_tokens": 3197311.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.107201290130615,
|
|
"epoch": 1.625559423172551,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004945727489921784,
|
|
"loss": 6.8984,
|
|
"mean_token_accuracy": 0.12509249597787858,
|
|
"num_tokens": 3208534.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.129283905029297,
|
|
"epoch": 1.6305320735952262,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004944871607766273,
|
|
"loss": 6.8441,
|
|
"mean_token_accuracy": 0.12468883246183396,
|
|
"num_tokens": 3217440.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.190691423416138,
|
|
"epoch": 1.6355047240179017,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004944009113325104,
|
|
"loss": 6.9785,
|
|
"mean_token_accuracy": 0.1205446295440197,
|
|
"num_tokens": 3225858.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.9474011898040775,
|
|
"epoch": 1.640477374440577,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004943140009196635,
|
|
"loss": 6.7535,
|
|
"mean_token_accuracy": 0.1390519469976425,
|
|
"num_tokens": 3235921.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.141832065582276,
|
|
"epoch": 1.6454500248632522,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004942264297999135,
|
|
"loss": 6.839,
|
|
"mean_token_accuracy": 0.1311826951801777,
|
|
"num_tokens": 3245123.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.140243911743164,
|
|
"epoch": 1.6504226752859275,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004941381982370778,
|
|
"loss": 6.8651,
|
|
"mean_token_accuracy": 0.1281386412680149,
|
|
"num_tokens": 3254246.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.141241884231567,
|
|
"epoch": 1.6553953257086027,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004940493064969633,
|
|
"loss": 6.8393,
|
|
"mean_token_accuracy": 0.12432443425059318,
|
|
"num_tokens": 3263602.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.095588350296021,
|
|
"epoch": 1.660367976131278,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000493959754847366,
|
|
"loss": 7.025,
|
|
"mean_token_accuracy": 0.1197537399828434,
|
|
"num_tokens": 3274430.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.12945523262024,
|
|
"epoch": 1.6653406265539532,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004938695435580698,
|
|
"loss": 6.7486,
|
|
"mean_token_accuracy": 0.12660369649529457,
|
|
"num_tokens": 3283587.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.054620218276978,
|
|
"epoch": 1.6703132769766285,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004937786729008459,
|
|
"loss": 6.9296,
|
|
"mean_token_accuracy": 0.11086304783821106,
|
|
"num_tokens": 3294165.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.139389705657959,
|
|
"epoch": 1.6752859273993037,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004936871431494517,
|
|
"loss": 6.8881,
|
|
"mean_token_accuracy": 0.1290462113916874,
|
|
"num_tokens": 3302756.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.076407480239868,
|
|
"epoch": 1.680258577821979,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004935949545796307,
|
|
"loss": 6.8204,
|
|
"mean_token_accuracy": 0.12808602973818778,
|
|
"num_tokens": 3313108.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.133471727371216,
|
|
"epoch": 1.6852312282446544,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004935021074691106,
|
|
"loss": 6.8982,
|
|
"mean_token_accuracy": 0.1255290202796459,
|
|
"num_tokens": 3323008.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.065728569030762,
|
|
"epoch": 1.6902038786673297,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004934086020976032,
|
|
"loss": 6.8167,
|
|
"mean_token_accuracy": 0.12538206279277803,
|
|
"num_tokens": 3332490.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.0050443649292,
|
|
"epoch": 1.695176529090005,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004933144387468037,
|
|
"loss": 6.8525,
|
|
"mean_token_accuracy": 0.12207689508795738,
|
|
"num_tokens": 3343069.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.050508975982666,
|
|
"epoch": 1.7001491795126804,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004932196177003891,
|
|
"loss": 6.9425,
|
|
"mean_token_accuracy": 0.12343786805868148,
|
|
"num_tokens": 3351834.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.244025945663452,
|
|
"epoch": 1.7051218299353557,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000493124139244018,
|
|
"loss": 6.8436,
|
|
"mean_token_accuracy": 0.12098148465156555,
|
|
"num_tokens": 3360915.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.029709720611573,
|
|
"epoch": 1.710094480358031,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004930280036653294,
|
|
"loss": 6.8307,
|
|
"mean_token_accuracy": 0.1256856717169285,
|
|
"num_tokens": 3370184.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.094144105911255,
|
|
"epoch": 1.7150671307807062,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000492931211253942,
|
|
"loss": 6.9156,
|
|
"mean_token_accuracy": 0.11987519264221191,
|
|
"num_tokens": 3379545.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.09250316619873,
|
|
"epoch": 1.7200397812033814,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004928337623014535,
|
|
"loss": 6.8638,
|
|
"mean_token_accuracy": 0.13006417453289032,
|
|
"num_tokens": 3390238.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.129364538192749,
|
|
"epoch": 1.7250124316260567,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000492735657101439,
|
|
"loss": 6.8756,
|
|
"mean_token_accuracy": 0.12177557870745659,
|
|
"num_tokens": 3400437.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.023185300827026,
|
|
"epoch": 1.729985082048732,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004926368959494513,
|
|
"loss": 6.8732,
|
|
"mean_token_accuracy": 0.12441569492220879,
|
|
"num_tokens": 3410982.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.136072969436645,
|
|
"epoch": 1.7349577324714072,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004925374791430186,
|
|
"loss": 6.9436,
|
|
"mean_token_accuracy": 0.12512961998581887,
|
|
"num_tokens": 3420898.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.067667722702026,
|
|
"epoch": 1.7399303828940824,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004924374069816448,
|
|
"loss": 6.8065,
|
|
"mean_token_accuracy": 0.1288384348154068,
|
|
"num_tokens": 3429144.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.036431932449341,
|
|
"epoch": 1.7449030333167577,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004923366797668081,
|
|
"loss": 6.8471,
|
|
"mean_token_accuracy": 0.1284836709499359,
|
|
"num_tokens": 3439427.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.134074735641479,
|
|
"epoch": 1.7498756837394331,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004922352978019597,
|
|
"loss": 6.8329,
|
|
"mean_token_accuracy": 0.1320321872830391,
|
|
"num_tokens": 3448218.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.052996969223022,
|
|
"epoch": 1.7548483341621084,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004921332613925239,
|
|
"loss": 6.8842,
|
|
"mean_token_accuracy": 0.12682262808084488,
|
|
"num_tokens": 3457656.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.1880049228668215,
|
|
"epoch": 1.7598209845847836,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004920305708458964,
|
|
"loss": 6.8518,
|
|
"mean_token_accuracy": 0.12582537904381752,
|
|
"num_tokens": 3467414.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.969467973709106,
|
|
"epoch": 1.7647936350074591,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004919272264714431,
|
|
"loss": 6.8022,
|
|
"mean_token_accuracy": 0.12401163727045059,
|
|
"num_tokens": 3477195.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.0714489936828615,
|
|
"epoch": 1.7697662854301344,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004918232285805004,
|
|
"loss": 6.8776,
|
|
"mean_token_accuracy": 0.12485396265983581,
|
|
"num_tokens": 3487497.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.128742027282715,
|
|
"epoch": 1.7747389358528096,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004917185774863727,
|
|
"loss": 6.8509,
|
|
"mean_token_accuracy": 0.12480712383985519,
|
|
"num_tokens": 3497363.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.998370409011841,
|
|
"epoch": 1.7797115862754849,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004916132735043329,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.12543805167078972,
|
|
"num_tokens": 3506969.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.106174898147583,
|
|
"epoch": 1.7846842366981601,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004915073169516204,
|
|
"loss": 6.8123,
|
|
"mean_token_accuracy": 0.12593027725815772,
|
|
"num_tokens": 3516418.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.971872282028198,
|
|
"epoch": 1.7896568871208354,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004914007081474406,
|
|
"loss": 6.7415,
|
|
"mean_token_accuracy": 0.12818810865283012,
|
|
"num_tokens": 3528023.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.079099035263061,
|
|
"epoch": 1.7946295375435106,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004912934474129641,
|
|
"loss": 6.8263,
|
|
"mean_token_accuracy": 0.12469528913497925,
|
|
"num_tokens": 3537946.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.011456918716431,
|
|
"epoch": 1.7996021879661859,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004911855350713251,
|
|
"loss": 6.8237,
|
|
"mean_token_accuracy": 0.12666564732789992,
|
|
"num_tokens": 3547299.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.08799467086792,
|
|
"epoch": 1.8045748383888611,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004910769714476214,
|
|
"loss": 6.8937,
|
|
"mean_token_accuracy": 0.12470808178186417,
|
|
"num_tokens": 3556075.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.027981472015381,
|
|
"epoch": 1.8095474888115366,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004909677568689122,
|
|
"loss": 6.8554,
|
|
"mean_token_accuracy": 0.12803274467587472,
|
|
"num_tokens": 3566314.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.068058443069458,
|
|
"epoch": 1.8145201392342118,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004908578916642183,
|
|
"loss": 6.7717,
|
|
"mean_token_accuracy": 0.13315743580460548,
|
|
"num_tokens": 3575095.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.001987552642822,
|
|
"epoch": 1.819492789656887,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004907473761645205,
|
|
"loss": 6.8469,
|
|
"mean_token_accuracy": 0.12430937737226486,
|
|
"num_tokens": 3584375.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.019008684158325,
|
|
"epoch": 1.8244654400795626,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004906362107027583,
|
|
"loss": 6.7288,
|
|
"mean_token_accuracy": 0.13371379896998406,
|
|
"num_tokens": 3594504.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.956771993637085,
|
|
"epoch": 1.8294380905022378,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004905243956138298,
|
|
"loss": 6.8666,
|
|
"mean_token_accuracy": 0.12128530889749527,
|
|
"num_tokens": 3604050.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.099835300445557,
|
|
"epoch": 1.834410740924913,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004904119312345896,
|
|
"loss": 6.7725,
|
|
"mean_token_accuracy": 0.13254232332110405,
|
|
"num_tokens": 3613388.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.027640390396118,
|
|
"epoch": 1.8393833913475883,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004902988179038492,
|
|
"loss": 6.934,
|
|
"mean_token_accuracy": 0.11459680572152138,
|
|
"num_tokens": 3623758.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.102637529373169,
|
|
"epoch": 1.8443560417702636,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000490185055962374,
|
|
"loss": 6.8327,
|
|
"mean_token_accuracy": 0.12521772757172583,
|
|
"num_tokens": 3633040.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.968631267547607,
|
|
"epoch": 1.8493286921929388,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 0.0004900706457528845,
|
|
"loss": 6.7158,
|
|
"mean_token_accuracy": 0.13046001717448236,
|
|
"num_tokens": 3642980.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.0653759956359865,
|
|
"epoch": 1.854301342615614,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004899555876200535,
|
|
"loss": 6.8785,
|
|
"mean_token_accuracy": 0.12559489011764527,
|
|
"num_tokens": 3653817.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.078700351715088,
|
|
"epoch": 1.8592739930382893,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004898398819105059,
|
|
"loss": 6.8205,
|
|
"mean_token_accuracy": 0.12691501155495644,
|
|
"num_tokens": 3664460.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 7.048341941833496,
|
|
"epoch": 1.8642466434609646,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004897235289728178,
|
|
"loss": 6.8393,
|
|
"mean_token_accuracy": 0.12090610191226006,
|
|
"num_tokens": 3674221.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.959291744232178,
|
|
"epoch": 1.8692192938836398,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004896065291575145,
|
|
"loss": 6.7412,
|
|
"mean_token_accuracy": 0.13510354980826378,
|
|
"num_tokens": 3683130.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 7.107599306106567,
|
|
"epoch": 1.8741919443063153,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004894888828170708,
|
|
"loss": 6.7053,
|
|
"mean_token_accuracy": 0.13113834261894225,
|
|
"num_tokens": 3694544.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.910497951507568,
|
|
"epoch": 1.8791645947289906,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004893705903059087,
|
|
"loss": 6.6849,
|
|
"mean_token_accuracy": 0.12996711134910582,
|
|
"num_tokens": 3704857.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.996693325042725,
|
|
"epoch": 1.8841372451516658,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004892516519803971,
|
|
"loss": 6.7635,
|
|
"mean_token_accuracy": 0.13128842264413834,
|
|
"num_tokens": 3714623.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.008214235305786,
|
|
"epoch": 1.8891098955743413,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004891320681988504,
|
|
"loss": 6.7623,
|
|
"mean_token_accuracy": 0.13260336965322495,
|
|
"num_tokens": 3723349.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.997120523452759,
|
|
"epoch": 1.8940825459970165,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004890118393215277,
|
|
"loss": 6.7289,
|
|
"mean_token_accuracy": 0.13324674889445304,
|
|
"num_tokens": 3732990.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.992683362960816,
|
|
"epoch": 1.8990551964196918,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004888909657106311,
|
|
"loss": 6.7194,
|
|
"mean_token_accuracy": 0.13112411946058272,
|
|
"num_tokens": 3742343.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.902509069442749,
|
|
"epoch": 1.904027846842367,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004887694477303055,
|
|
"loss": 6.7515,
|
|
"mean_token_accuracy": 0.12507615908980368,
|
|
"num_tokens": 3752128.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.963840103149414,
|
|
"epoch": 1.9090004972650423,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004886472857466368,
|
|
"loss": 6.6778,
|
|
"mean_token_accuracy": 0.1289484165608883,
|
|
"num_tokens": 3760431.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.976652908325195,
|
|
"epoch": 1.9139731476877175,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004885244801276509,
|
|
"loss": 6.7712,
|
|
"mean_token_accuracy": 0.12675235122442247,
|
|
"num_tokens": 3769623.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.85596399307251,
|
|
"epoch": 1.9189457981103928,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004884010312433131,
|
|
"loss": 6.7387,
|
|
"mean_token_accuracy": 0.12580233588814735,
|
|
"num_tokens": 3779383.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 7.044457960128784,
|
|
"epoch": 1.923918448533068,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048827693946552615,
|
|
"loss": 6.8649,
|
|
"mean_token_accuracy": 0.12248891815543175,
|
|
"num_tokens": 3788847.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.909450054168701,
|
|
"epoch": 1.9288910989557433,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048815220516813004,
|
|
"loss": 6.7146,
|
|
"mean_token_accuracy": 0.13217381164431571,
|
|
"num_tokens": 3798711.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.975063467025757,
|
|
"epoch": 1.9338637493784185,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004880268287269,
|
|
"loss": 6.6871,
|
|
"mean_token_accuracy": 0.13259740248322488,
|
|
"num_tokens": 3808837.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.989281034469604,
|
|
"epoch": 1.938836399801094,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048790081051954605,
|
|
"loss": 6.6711,
|
|
"mean_token_accuracy": 0.1324021689593792,
|
|
"num_tokens": 3818261.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 7.002820634841919,
|
|
"epoch": 1.9438090502237693,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004877741509257116,
|
|
"loss": 6.7457,
|
|
"mean_token_accuracy": 0.1204921394586563,
|
|
"num_tokens": 3828146.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.053721761703491,
|
|
"epoch": 1.9487817006464445,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004876468503269721,
|
|
"loss": 6.9486,
|
|
"mean_token_accuracy": 0.1165396012365818,
|
|
"num_tokens": 3838658.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 7.047277545928955,
|
|
"epoch": 1.95375435106912,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048751890910683434,
|
|
"loss": 6.7216,
|
|
"mean_token_accuracy": 0.12939262613654137,
|
|
"num_tokens": 3849097.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 7.016208505630493,
|
|
"epoch": 1.9587270014917952,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048739032765073487,
|
|
"loss": 6.8353,
|
|
"mean_token_accuracy": 0.12277894169092178,
|
|
"num_tokens": 3858645.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 7.007478046417236,
|
|
"epoch": 1.9636996519144705,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048726110634603913,
|
|
"loss": 6.7223,
|
|
"mean_token_accuracy": 0.1341798223555088,
|
|
"num_tokens": 3869232.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.993938398361206,
|
|
"epoch": 1.9686723023371457,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048713124558204013,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.12780363112688065,
|
|
"num_tokens": 3879453.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.984024095535278,
|
|
"epoch": 1.973644952759821,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004870007457499573,
|
|
"loss": 6.7455,
|
|
"mean_token_accuracy": 0.1269009605050087,
|
|
"num_tokens": 3888257.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 7.0300885200500485,
|
|
"epoch": 1.9786176031824962,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 0.00048686960724293537,
|
|
"loss": 6.7464,
|
|
"mean_token_accuracy": 0.13770631626248359,
|
|
"num_tokens": 3897980.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.86609296798706,
|
|
"epoch": 1.9835902536051715,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0004867378304560432,
|
|
"loss": 6.5755,
|
|
"mean_token_accuracy": 0.13791960403323172,
|
|
"num_tokens": 3907479.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.98457088470459,
|
|
"epoch": 1.9885629040278467,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004866054157862723,
|
|
"loss": 6.8118,
|
|
"mean_token_accuracy": 0.1273365207016468,
|
|
"num_tokens": 3917249.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.9885629040278467,
|
|
"eval_entropy": 6.826489323295206,
|
|
"eval_loss": 6.902448654174805,
|
|
"eval_mean_token_accuracy": 0.12161170143616913,
|
|
"eval_num_tokens": 3917249.0,
|
|
"eval_runtime": 1.9145,
|
|
"eval_samples_per_second": 1809.844,
|
|
"eval_steps_per_second": 226.688,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.918976974487305,
|
|
"epoch": 1.993535554450522,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004864723636325361,
|
|
"loss": 6.6902,
|
|
"mean_token_accuracy": 0.13525744825601577,
|
|
"num_tokens": 3928388.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.918926954269409,
|
|
"epoch": 1.9985082048731975,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048633867439566855,
|
|
"loss": 6.7358,
|
|
"mean_token_accuracy": 0.13128256052732468,
|
|
"num_tokens": 3937323.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.967238638136122,
|
|
"epoch": 2.002983590253605,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004862043484784227,
|
|
"loss": 6.6336,
|
|
"mean_token_accuracy": 0.133645827571551,
|
|
"num_tokens": 3946815.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 7.008509635925293,
|
|
"epoch": 2.0079562406762803,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048606938628546975,
|
|
"loss": 6.4412,
|
|
"mean_token_accuracy": 0.13506631627678872,
|
|
"num_tokens": 3956125.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.9345386028289795,
|
|
"epoch": 2.0129288910989556,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004859337882233978,
|
|
"loss": 6.4573,
|
|
"mean_token_accuracy": 0.1371847689151764,
|
|
"num_tokens": 3966543.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.941262102127075,
|
|
"epoch": 2.017901541521631,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004857975547007106,
|
|
"loss": 6.4969,
|
|
"mean_token_accuracy": 0.13126529902219772,
|
|
"num_tokens": 3978200.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.889175224304199,
|
|
"epoch": 2.0228741919443065,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048566068612782595,
|
|
"loss": 6.3546,
|
|
"mean_token_accuracy": 0.14628630578517915,
|
|
"num_tokens": 3988331.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.916131162643433,
|
|
"epoch": 2.0278468423669818,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048552318291707533,
|
|
"loss": 6.4241,
|
|
"mean_token_accuracy": 0.14123102203011512,
|
|
"num_tokens": 3998976.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.849209976196289,
|
|
"epoch": 2.032819492789657,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048538504548270177,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.1385306313633919,
|
|
"num_tokens": 4008354.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.920752763748169,
|
|
"epoch": 2.0377921432123323,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048524627424085913,
|
|
"loss": 6.38,
|
|
"mean_token_accuracy": 0.14287423342466354,
|
|
"num_tokens": 4018045.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.846613597869873,
|
|
"epoch": 2.0427647936350075,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00048510686960961057,
|
|
"loss": 6.3915,
|
|
"mean_token_accuracy": 0.14863280653953553,
|
|
"num_tokens": 4027884.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.880599689483643,
|
|
"epoch": 2.0477374440576828,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004849668320089276,
|
|
"loss": 6.4221,
|
|
"mean_token_accuracy": 0.1414875164628029,
|
|
"num_tokens": 4037866.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.914599847793579,
|
|
"epoch": 2.052710094480358,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.00048482616186068826,
|
|
"loss": 6.4402,
|
|
"mean_token_accuracy": 0.13650158569216728,
|
|
"num_tokens": 4048126.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.858435869216919,
|
|
"epoch": 2.0576827449030333,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048468485958867664,
|
|
"loss": 6.4649,
|
|
"mean_token_accuracy": 0.13671665042638778,
|
|
"num_tokens": 4058179.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.904900169372558,
|
|
"epoch": 2.0626553953257085,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048454292561858083,
|
|
"loss": 6.3527,
|
|
"mean_token_accuracy": 0.14049720838665963,
|
|
"num_tokens": 4067479.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.853045988082886,
|
|
"epoch": 2.0676280457483838,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004844003603779921,
|
|
"loss": 6.4422,
|
|
"mean_token_accuracy": 0.1422359362244606,
|
|
"num_tokens": 4076624.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.825760173797607,
|
|
"epoch": 2.072600696171059,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004842571642964036,
|
|
"loss": 6.3825,
|
|
"mean_token_accuracy": 0.14210811406373977,
|
|
"num_tokens": 4085752.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.853710508346557,
|
|
"epoch": 2.0775733465937343,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048411333780520876,
|
|
"loss": 6.4289,
|
|
"mean_token_accuracy": 0.1422540456056595,
|
|
"num_tokens": 4095026.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.847144985198975,
|
|
"epoch": 2.0825459970164095,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048396888133770034,
|
|
"loss": 6.4383,
|
|
"mean_token_accuracy": 0.1379736840724945,
|
|
"num_tokens": 4104589.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.800404119491577,
|
|
"epoch": 2.087518647439085,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048382379532906897,
|
|
"loss": 6.3934,
|
|
"mean_token_accuracy": 0.14050761386752128,
|
|
"num_tokens": 4114550.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.923476219177246,
|
|
"epoch": 2.0924912978617605,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004836780802164016,
|
|
"loss": 6.4866,
|
|
"mean_token_accuracy": 0.13951282426714898,
|
|
"num_tokens": 4125491.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.806124114990235,
|
|
"epoch": 2.0974639482844357,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048353173643868064,
|
|
"loss": 6.4056,
|
|
"mean_token_accuracy": 0.1417339526116848,
|
|
"num_tokens": 4134599.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.7399749755859375,
|
|
"epoch": 2.102436598707111,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004833847644367826,
|
|
"loss": 6.3393,
|
|
"mean_token_accuracy": 0.14568568021059036,
|
|
"num_tokens": 4143640.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.867958450317383,
|
|
"epoch": 2.107409249129786,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048323716465347615,
|
|
"loss": 6.4142,
|
|
"mean_token_accuracy": 0.14574476927518845,
|
|
"num_tokens": 4152558.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.69665937423706,
|
|
"epoch": 2.1123818995524615,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004830889375334215,
|
|
"loss": 6.2609,
|
|
"mean_token_accuracy": 0.15683163851499557,
|
|
"num_tokens": 4162238.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.846102237701416,
|
|
"epoch": 2.1173545499751367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048294008352316875,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.1427025318145752,
|
|
"num_tokens": 4171437.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.78309965133667,
|
|
"epoch": 2.122327200397812,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004827906030711564,
|
|
"loss": 6.3698,
|
|
"mean_token_accuracy": 0.14197533950209618,
|
|
"num_tokens": 4181584.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.6827184677124025,
|
|
"epoch": 2.1272998508204872,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00048264049662771053,
|
|
"loss": 6.3056,
|
|
"mean_token_accuracy": 0.15470692366361619,
|
|
"num_tokens": 4191756.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.72650556564331,
|
|
"epoch": 2.1322725012431625,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048248976464504266,
|
|
"loss": 6.3363,
|
|
"mean_token_accuracy": 0.1480225808918476,
|
|
"num_tokens": 4202864.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.984028148651123,
|
|
"epoch": 2.1372451516658377,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048233840757724926,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.13635303750634192,
|
|
"num_tokens": 4212839.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.737657785415649,
|
|
"epoch": 2.142217802088513,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048218642588030943,
|
|
"loss": 6.4263,
|
|
"mean_token_accuracy": 0.1367236152291298,
|
|
"num_tokens": 4223150.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.837790012359619,
|
|
"epoch": 2.1471904525111887,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048203382001208453,
|
|
"loss": 6.4347,
|
|
"mean_token_accuracy": 0.13784121796488763,
|
|
"num_tokens": 4232206.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.845598936080933,
|
|
"epoch": 2.152163102933864,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048188059043231585,
|
|
"loss": 6.4025,
|
|
"mean_token_accuracy": 0.13808769062161447,
|
|
"num_tokens": 4240992.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.723655557632446,
|
|
"epoch": 2.157135753356539,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048172673760262414,
|
|
"loss": 6.389,
|
|
"mean_token_accuracy": 0.13779966458678244,
|
|
"num_tokens": 4250371.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.840053558349609,
|
|
"epoch": 2.1621084037792144,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048157226198650735,
|
|
"loss": 6.4304,
|
|
"mean_token_accuracy": 0.14555971547961236,
|
|
"num_tokens": 4260606.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.790835428237915,
|
|
"epoch": 2.1670810542018897,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004814171640493398,
|
|
"loss": 6.4177,
|
|
"mean_token_accuracy": 0.14312928691506385,
|
|
"num_tokens": 4269753.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.696050214767456,
|
|
"epoch": 2.172053704624565,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004812614442583708,
|
|
"loss": 6.323,
|
|
"mean_token_accuracy": 0.1503577008843422,
|
|
"num_tokens": 4279485.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.932571935653686,
|
|
"epoch": 2.17702635504724,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004811051030827227,
|
|
"loss": 6.4322,
|
|
"mean_token_accuracy": 0.14114878922700883,
|
|
"num_tokens": 4289491.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.7436810493469235,
|
|
"epoch": 2.1819990054699154,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00048094814099339006,
|
|
"loss": 6.4635,
|
|
"mean_token_accuracy": 0.14781300351023674,
|
|
"num_tokens": 4300536.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.909609174728393,
|
|
"epoch": 2.1869716558925907,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004807905584632381,
|
|
"loss": 6.5091,
|
|
"mean_token_accuracy": 0.13384719267487527,
|
|
"num_tokens": 4311804.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.913917589187622,
|
|
"epoch": 2.191944306315266,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048063235596700087,
|
|
"loss": 6.4787,
|
|
"mean_token_accuracy": 0.13571926802396775,
|
|
"num_tokens": 4320662.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.834342050552368,
|
|
"epoch": 2.196916956737941,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048047353398128056,
|
|
"loss": 6.4875,
|
|
"mean_token_accuracy": 0.13781703412532806,
|
|
"num_tokens": 4330414.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.766165018081665,
|
|
"epoch": 2.2018896071606164,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004803140929845452,
|
|
"loss": 6.4316,
|
|
"mean_token_accuracy": 0.14176806360483168,
|
|
"num_tokens": 4340478.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.771369123458863,
|
|
"epoch": 2.2068622575832917,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048015403345712796,
|
|
"loss": 6.3847,
|
|
"mean_token_accuracy": 0.14389355182647706,
|
|
"num_tokens": 4350688.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.649575281143188,
|
|
"epoch": 2.2118349080059674,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004799933558812252,
|
|
"loss": 6.2804,
|
|
"mean_token_accuracy": 0.15400244891643525,
|
|
"num_tokens": 4360554.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.75346155166626,
|
|
"epoch": 2.2168075584286426,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047983206074089546,
|
|
"loss": 6.3331,
|
|
"mean_token_accuracy": 0.1503487393260002,
|
|
"num_tokens": 4370723.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.806484365463257,
|
|
"epoch": 2.221780208851318,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004796701485220576,
|
|
"loss": 6.4262,
|
|
"mean_token_accuracy": 0.14320623576641084,
|
|
"num_tokens": 4380383.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.796637964248657,
|
|
"epoch": 2.226752859273993,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004795076197124893,
|
|
"loss": 6.5399,
|
|
"mean_token_accuracy": 0.1261807143688202,
|
|
"num_tokens": 4390408.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.769066858291626,
|
|
"epoch": 2.2317255096966684,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047934447480182626,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.14121257290244102,
|
|
"num_tokens": 4400043.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.853194093704223,
|
|
"epoch": 2.2366981601193436,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004791807142815598,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.13817397877573967,
|
|
"num_tokens": 4409420.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.736216735839844,
|
|
"epoch": 2.241670810542019,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047901633864503614,
|
|
"loss": 6.3558,
|
|
"mean_token_accuracy": 0.14454032629728317,
|
|
"num_tokens": 4418742.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.831597375869751,
|
|
"epoch": 2.246643460964694,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047885134838745433,
|
|
"loss": 6.3803,
|
|
"mean_token_accuracy": 0.14649947956204415,
|
|
"num_tokens": 4427921.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.7241028308868405,
|
|
"epoch": 2.2516161113873694,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004786857440058654,
|
|
"loss": 6.4208,
|
|
"mean_token_accuracy": 0.14122760742902757,
|
|
"num_tokens": 4438029.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.723005104064941,
|
|
"epoch": 2.2565887618100446,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004785195259991701,
|
|
"loss": 6.3936,
|
|
"mean_token_accuracy": 0.15054373890161515,
|
|
"num_tokens": 4446967.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.781913232803345,
|
|
"epoch": 2.26156141223272,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000478352694868118,
|
|
"loss": 6.332,
|
|
"mean_token_accuracy": 0.14599733129143716,
|
|
"num_tokens": 4456469.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.754271030426025,
|
|
"epoch": 2.266534062655395,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004781852511153058,
|
|
"loss": 6.472,
|
|
"mean_token_accuracy": 0.1463216945528984,
|
|
"num_tokens": 4465949.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.789528036117554,
|
|
"epoch": 2.271506713078071,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047801719524517573,
|
|
"loss": 6.2799,
|
|
"mean_token_accuracy": 0.14413850083947183,
|
|
"num_tokens": 4475460.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.774982643127442,
|
|
"epoch": 2.276479363500746,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047784852776401403,
|
|
"loss": 6.4072,
|
|
"mean_token_accuracy": 0.14366714656352997,
|
|
"num_tokens": 4485540.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.755620384216309,
|
|
"epoch": 2.2814520139234213,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004776792491799495,
|
|
"loss": 6.3927,
|
|
"mean_token_accuracy": 0.1439718559384346,
|
|
"num_tokens": 4495545.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.744416046142578,
|
|
"epoch": 2.2864246643460966,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00047750936000295213,
|
|
"loss": 6.3396,
|
|
"mean_token_accuracy": 0.14198606535792352,
|
|
"num_tokens": 4505527.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.751213884353637,
|
|
"epoch": 2.291397314768772,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047733886074483115,
|
|
"loss": 6.345,
|
|
"mean_token_accuracy": 0.14732159897685052,
|
|
"num_tokens": 4515366.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.767435216903687,
|
|
"epoch": 2.296369965191447,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004771677519192339,
|
|
"loss": 6.436,
|
|
"mean_token_accuracy": 0.14124366119503975,
|
|
"num_tokens": 4525273.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.784068965911866,
|
|
"epoch": 2.3013426156141223,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004769960340416439,
|
|
"loss": 6.3953,
|
|
"mean_token_accuracy": 0.1413241557776928,
|
|
"num_tokens": 4533568.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.775341129302978,
|
|
"epoch": 2.3063152660367976,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047682370762937976,
|
|
"loss": 6.5538,
|
|
"mean_token_accuracy": 0.1359819784760475,
|
|
"num_tokens": 4544508.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.867922401428222,
|
|
"epoch": 2.311287916459473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047665077320159315,
|
|
"loss": 6.3257,
|
|
"mean_token_accuracy": 0.14260879829525946,
|
|
"num_tokens": 4554586.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.6684229373931885,
|
|
"epoch": 2.316260566882148,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047647723127926765,
|
|
"loss": 6.3422,
|
|
"mean_token_accuracy": 0.1473762109875679,
|
|
"num_tokens": 4563883.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.690744256973266,
|
|
"epoch": 2.3212332173048233,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004763030823852168,
|
|
"loss": 6.3685,
|
|
"mean_token_accuracy": 0.14711007624864578,
|
|
"num_tokens": 4573675.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.7738425731658936,
|
|
"epoch": 2.3262058677274986,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004761283270440829,
|
|
"loss": 6.4085,
|
|
"mean_token_accuracy": 0.13553072661161422,
|
|
"num_tokens": 4583617.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.767460775375366,
|
|
"epoch": 2.331178518150174,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004759529657823349,
|
|
"loss": 6.4187,
|
|
"mean_token_accuracy": 0.14251091927289963,
|
|
"num_tokens": 4593792.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.732200193405151,
|
|
"epoch": 2.336151168572849,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047577699912826755,
|
|
"loss": 6.4096,
|
|
"mean_token_accuracy": 0.1377829432487488,
|
|
"num_tokens": 4603324.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.7790343284606935,
|
|
"epoch": 2.341123818995525,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047560042761199915,
|
|
"loss": 6.3784,
|
|
"mean_token_accuracy": 0.14205057471990584,
|
|
"num_tokens": 4613509.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.658831548690796,
|
|
"epoch": 2.3460964694182,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004754232517654704,
|
|
"loss": 6.3322,
|
|
"mean_token_accuracy": 0.14965879768133164,
|
|
"num_tokens": 4624282.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.649105596542358,
|
|
"epoch": 2.3510691198408753,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004752454721224424,
|
|
"loss": 6.2986,
|
|
"mean_token_accuracy": 0.14808024168014527,
|
|
"num_tokens": 4633217.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.667078256607056,
|
|
"epoch": 2.3560417702635505,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047506708921849535,
|
|
"loss": 6.3018,
|
|
"mean_token_accuracy": 0.15329894721508025,
|
|
"num_tokens": 4642824.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.713304805755615,
|
|
"epoch": 2.361014420686226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004748881035910269,
|
|
"loss": 6.3601,
|
|
"mean_token_accuracy": 0.1459948495030403,
|
|
"num_tokens": 4651938.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.668780422210693,
|
|
"epoch": 2.365987071108901,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004747085157792504,
|
|
"loss": 6.3679,
|
|
"mean_token_accuracy": 0.1452711433172226,
|
|
"num_tokens": 4661695.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.821518850326538,
|
|
"epoch": 2.3709597215315763,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004745283263241933,
|
|
"loss": 6.4559,
|
|
"mean_token_accuracy": 0.1399241141974926,
|
|
"num_tokens": 4671354.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.645737886428833,
|
|
"epoch": 2.3759323719542516,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004743475357686957,
|
|
"loss": 6.2939,
|
|
"mean_token_accuracy": 0.16298234164714814,
|
|
"num_tokens": 4681795.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.708989000320434,
|
|
"epoch": 2.380905022376927,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047416614465740843,
|
|
"loss": 6.3991,
|
|
"mean_token_accuracy": 0.15196624025702477,
|
|
"num_tokens": 4691229.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.741003608703613,
|
|
"epoch": 2.385877672799602,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00047398415353679167,
|
|
"loss": 6.4708,
|
|
"mean_token_accuracy": 0.14384533688426018,
|
|
"num_tokens": 4701880.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.76146674156189,
|
|
"epoch": 2.3908503232222773,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004738015629551132,
|
|
"loss": 6.3834,
|
|
"mean_token_accuracy": 0.14161035120487214,
|
|
"num_tokens": 4712041.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.534692239761353,
|
|
"epoch": 2.395822973644953,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00047361837346244655,
|
|
"loss": 6.1709,
|
|
"mean_token_accuracy": 0.15967928618192673,
|
|
"num_tokens": 4722576.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.748846340179443,
|
|
"epoch": 2.4007956240676283,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004734345856106698,
|
|
"loss": 6.398,
|
|
"mean_token_accuracy": 0.1478155106306076,
|
|
"num_tokens": 4732569.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.767484474182129,
|
|
"epoch": 2.4057682744903035,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047325019995346355,
|
|
"loss": 6.4518,
|
|
"mean_token_accuracy": 0.1463234543800354,
|
|
"num_tokens": 4741973.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.871760272979737,
|
|
"epoch": 2.4107409249129788,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004730652170463091,
|
|
"loss": 6.4214,
|
|
"mean_token_accuracy": 0.14306398406624793,
|
|
"num_tokens": 4751550.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.715447235107422,
|
|
"epoch": 2.415713575335654,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004728796374464875,
|
|
"loss": 6.363,
|
|
"mean_token_accuracy": 0.14614096358418466,
|
|
"num_tokens": 4761620.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.676720333099365,
|
|
"epoch": 2.4206862257583293,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047269346171307716,
|
|
"loss": 6.3479,
|
|
"mean_token_accuracy": 0.14403630942106246,
|
|
"num_tokens": 4771893.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.75561933517456,
|
|
"epoch": 2.4256588761810045,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004725066904069523,
|
|
"loss": 6.3992,
|
|
"mean_token_accuracy": 0.14306726008653642,
|
|
"num_tokens": 4781094.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.6643373489379885,
|
|
"epoch": 2.4306315266036798,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004723193240907814,
|
|
"loss": 6.3534,
|
|
"mean_token_accuracy": 0.1403821013867855,
|
|
"num_tokens": 4790337.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.697565507888794,
|
|
"epoch": 2.435604177026355,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004721313633290257,
|
|
"loss": 6.4024,
|
|
"mean_token_accuracy": 0.14306891933083535,
|
|
"num_tokens": 4801758.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.7525293827056885,
|
|
"epoch": 2.4405768274490303,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000471942808687937,
|
|
"loss": 6.4715,
|
|
"mean_token_accuracy": 0.13697458431124687,
|
|
"num_tokens": 4811040.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.709061479568481,
|
|
"epoch": 2.4455494778717055,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00047175366073555654,
|
|
"loss": 6.3356,
|
|
"mean_token_accuracy": 0.14218414053320885,
|
|
"num_tokens": 4821361.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.714569187164306,
|
|
"epoch": 2.4505221282943808,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004715639200417125,
|
|
"loss": 6.3643,
|
|
"mean_token_accuracy": 0.13874393850564956,
|
|
"num_tokens": 4830919.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.607773780822754,
|
|
"epoch": 2.455494778717056,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004713735871780192,
|
|
"loss": 6.1869,
|
|
"mean_token_accuracy": 0.16048071086406707,
|
|
"num_tokens": 4840364.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.678798961639404,
|
|
"epoch": 2.4604674291397313,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004711826627178747,
|
|
"loss": 6.391,
|
|
"mean_token_accuracy": 0.144606377184391,
|
|
"num_tokens": 4849044.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.706400680541992,
|
|
"epoch": 2.4654400795624065,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047099114723645944,
|
|
"loss": 6.3627,
|
|
"mean_token_accuracy": 0.14185858219861985,
|
|
"num_tokens": 4859183.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.7633781909942625,
|
|
"epoch": 2.470412729985082,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00047079904131073434,
|
|
"loss": 6.338,
|
|
"mean_token_accuracy": 0.1467789590358734,
|
|
"num_tokens": 4869625.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.702241659164429,
|
|
"epoch": 2.4753853804077575,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004706063455194389,
|
|
"loss": 6.3436,
|
|
"mean_token_accuracy": 0.1495677724480629,
|
|
"num_tokens": 4880089.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.707668209075928,
|
|
"epoch": 2.4803580308304327,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00047041306044308986,
|
|
"loss": 6.4324,
|
|
"mean_token_accuracy": 0.14488113895058632,
|
|
"num_tokens": 4890673.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.645248460769653,
|
|
"epoch": 2.485330681253108,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004702191866639792,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.1475027784705162,
|
|
"num_tokens": 4899444.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 6.6494762897491455,
|
|
"epoch": 2.490303331675783,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004700247247661726,
|
|
"loss": 6.4167,
|
|
"mean_token_accuracy": 0.14232293143868446,
|
|
"num_tokens": 4909186.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 6.649731683731079,
|
|
"epoch": 2.4952759820984585,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004698296753355072,
|
|
"loss": 6.2165,
|
|
"mean_token_accuracy": 0.1631212130188942,
|
|
"num_tokens": 4918886.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 6.710996103286743,
|
|
"epoch": 2.5002486325211337,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004696340389595903,
|
|
"loss": 6.4195,
|
|
"mean_token_accuracy": 0.13964025154709817,
|
|
"num_tokens": 4929157.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 6.656909608840943,
|
|
"epoch": 2.505221282943809,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00046943781622779733,
|
|
"loss": 6.248,
|
|
"mean_token_accuracy": 0.15459569841623305,
|
|
"num_tokens": 4938127.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 6.699398374557495,
|
|
"epoch": 2.510193933366484,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00046924100773127055,
|
|
"loss": 6.2617,
|
|
"mean_token_accuracy": 0.1559196650981903,
|
|
"num_tokens": 4947849.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 6.613239812850952,
|
|
"epoch": 2.5151665837891595,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00046904361406291633,
|
|
"loss": 6.314,
|
|
"mean_token_accuracy": 0.14947595894336702,
|
|
"num_tokens": 4957046.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 6.58072943687439,
|
|
"epoch": 2.520139234211835,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004688456358174044,
|
|
"loss": 6.3532,
|
|
"mean_token_accuracy": 0.15170142501592637,
|
|
"num_tokens": 4967128.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 6.764265871047973,
|
|
"epoch": 2.5251118846345104,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004686470735911653,
|
|
"loss": 6.414,
|
|
"mean_token_accuracy": 0.14283805042505265,
|
|
"num_tokens": 4976817.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 6.744951820373535,
|
|
"epoch": 2.5300845350571857,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004684479279823891,
|
|
"loss": 6.4092,
|
|
"mean_token_accuracy": 0.143687604367733,
|
|
"num_tokens": 4985908.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 6.697251415252685,
|
|
"epoch": 2.535057185479861,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004682481995910232,
|
|
"loss": 6.2994,
|
|
"mean_token_accuracy": 0.14897737950086593,
|
|
"num_tokens": 4995218.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 6.630112171173096,
|
|
"epoch": 2.540029835902536,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046804788901877084,
|
|
"loss": 6.4582,
|
|
"mean_token_accuracy": 0.14477261155843735,
|
|
"num_tokens": 5004197.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 6.717876386642456,
|
|
"epoch": 2.5450024863252114,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046784699686908886,
|
|
"loss": 6.3873,
|
|
"mean_token_accuracy": 0.14052913784980775,
|
|
"num_tokens": 5015408.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 6.6725719451904295,
|
|
"epoch": 2.5499751367478867,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046764552374718656,
|
|
"loss": 6.3538,
|
|
"mean_token_accuracy": 0.14534798562526702,
|
|
"num_tokens": 5025138.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 6.695027494430542,
|
|
"epoch": 2.554947787170562,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00046744347026002315,
|
|
"loss": 6.3757,
|
|
"mean_token_accuracy": 0.14619187712669374,
|
|
"num_tokens": 5034638.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 6.70813889503479,
|
|
"epoch": 2.559920437593237,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004672408370163066,
|
|
"loss": 6.4578,
|
|
"mean_token_accuracy": 0.1436978816986084,
|
|
"num_tokens": 5044826.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 6.742210865020752,
|
|
"epoch": 2.5648930880159124,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000467037624626491,
|
|
"loss": 6.532,
|
|
"mean_token_accuracy": 0.13766753375530244,
|
|
"num_tokens": 5054914.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 6.5788435459136965,
|
|
"epoch": 2.5698657384385877,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00046683383370277554,
|
|
"loss": 6.2633,
|
|
"mean_token_accuracy": 0.16008901298046113,
|
|
"num_tokens": 5066689.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 6.725404214859009,
|
|
"epoch": 2.574838388861263,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00046662946485910226,
|
|
"loss": 6.4523,
|
|
"mean_token_accuracy": 0.1429741084575653,
|
|
"num_tokens": 5076170.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 6.6832764625549315,
|
|
"epoch": 2.579811039283938,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000466424518711154,
|
|
"loss": 6.2897,
|
|
"mean_token_accuracy": 0.15062598437070845,
|
|
"num_tokens": 5085433.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 6.6068120956420895,
|
|
"epoch": 2.5847836897066134,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004662189958763531,
|
|
"loss": 6.3042,
|
|
"mean_token_accuracy": 0.14942502081394196,
|
|
"num_tokens": 5095692.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 6.593360424041748,
|
|
"epoch": 2.5897563401292887,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004660128969738591,
|
|
"loss": 6.3517,
|
|
"mean_token_accuracy": 0.14136677831411362,
|
|
"num_tokens": 5105037.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 6.632503032684326,
|
|
"epoch": 2.594728990551964,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004658062226245668,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.1514989249408245,
|
|
"num_tokens": 5113868.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 6.8026611328125,
|
|
"epoch": 2.5997016409746396,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004655989734511052,
|
|
"loss": 6.3745,
|
|
"mean_token_accuracy": 0.14606377258896827,
|
|
"num_tokens": 5123136.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 6.640027046203613,
|
|
"epoch": 2.604674291397315,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004653911500778343,
|
|
"loss": 6.2502,
|
|
"mean_token_accuracy": 0.1535334825515747,
|
|
"num_tokens": 5133869.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 6.695024394989014,
|
|
"epoch": 2.60964694181999,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004651827531308443,
|
|
"loss": 6.3504,
|
|
"mean_token_accuracy": 0.1503686174750328,
|
|
"num_tokens": 5143119.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 6.523976039886475,
|
|
"epoch": 2.6146195922426654,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004649737832379534,
|
|
"loss": 6.1359,
|
|
"mean_token_accuracy": 0.16792642772197724,
|
|
"num_tokens": 5152167.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 6.675824546813965,
|
|
"epoch": 2.6195922426653406,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004647642410287057,
|
|
"loss": 6.4482,
|
|
"mean_token_accuracy": 0.14554495215415955,
|
|
"num_tokens": 5161962.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 6.651208877563477,
|
|
"epoch": 2.624564893088016,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004645541271343696,
|
|
"loss": 6.2756,
|
|
"mean_token_accuracy": 0.1497366800904274,
|
|
"num_tokens": 5171637.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 6.499583053588867,
|
|
"epoch": 2.629537543510691,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046434344218793566,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.15593683421611787,
|
|
"num_tokens": 5181839.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 6.644394779205323,
|
|
"epoch": 2.6345101939333664,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00046413218682411485,
|
|
"loss": 6.3335,
|
|
"mean_token_accuracy": 0.1497930809855461,
|
|
"num_tokens": 5192237.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 6.702507495880127,
|
|
"epoch": 2.6394828443560416,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004639203616793365,
|
|
"loss": 6.4807,
|
|
"mean_token_accuracy": 0.1405867174267769,
|
|
"num_tokens": 5203879.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 6.699099016189575,
|
|
"epoch": 2.6444554947787173,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00046370796739174657,
|
|
"loss": 6.3422,
|
|
"mean_token_accuracy": 0.1415301501750946,
|
|
"num_tokens": 5213456.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 6.597914934158325,
|
|
"epoch": 2.6494281452013926,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00046349500460120556,
|
|
"loss": 6.2631,
|
|
"mean_token_accuracy": 0.1502414494752884,
|
|
"num_tokens": 5222234.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 6.6378380298614506,
|
|
"epoch": 2.654400795624068,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046328147394928684,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.14921867847442627,
|
|
"num_tokens": 5230989.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 6.657117033004761,
|
|
"epoch": 2.659373446046743,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004630673760792741,
|
|
"loss": 6.3314,
|
|
"mean_token_accuracy": 0.14511406496167184,
|
|
"num_tokens": 5240607.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 6.650262641906738,
|
|
"epoch": 2.6643460964694183,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004628527116361603,
|
|
"loss": 6.4381,
|
|
"mean_token_accuracy": 0.13797329291701316,
|
|
"num_tokens": 5250837.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 6.688577651977539,
|
|
"epoch": 2.6693187468920936,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004626374812666449,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.14562924653291703,
|
|
"num_tokens": 5260542.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 6.636471557617187,
|
|
"epoch": 2.674291397314769,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00046242168561913256,
|
|
"loss": 6.2402,
|
|
"mean_token_accuracy": 0.1531580999493599,
|
|
"num_tokens": 5272327.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 6.566676950454712,
|
|
"epoch": 2.679264047737444,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046220532534373067,
|
|
"loss": 6.3566,
|
|
"mean_token_accuracy": 0.14817607775330544,
|
|
"num_tokens": 5281407.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 6.7444984912872314,
|
|
"epoch": 2.6842366981601193,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004619884010922477,
|
|
"loss": 6.3364,
|
|
"mean_token_accuracy": 0.1450783297419548,
|
|
"num_tokens": 5291617.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 6.720807075500488,
|
|
"epoch": 2.6892093485827946,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004617709135181913,
|
|
"loss": 6.391,
|
|
"mean_token_accuracy": 0.14487197175621985,
|
|
"num_tokens": 5301174.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 6.627471494674682,
|
|
"epoch": 2.69418199900547,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00046155286327676594,
|
|
"loss": 6.3562,
|
|
"mean_token_accuracy": 0.14076724499464036,
|
|
"num_tokens": 5311211.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 6.596842861175537,
|
|
"epoch": 2.699154649428145,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046133425102487134,
|
|
"loss": 6.3611,
|
|
"mean_token_accuracy": 0.1456509128212929,
|
|
"num_tokens": 5321343.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 6.6324012756347654,
|
|
"epoch": 2.7041272998508203,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046111507742110035,
|
|
"loss": 6.3136,
|
|
"mean_token_accuracy": 0.15493008494377136,
|
|
"num_tokens": 5330849.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 6.662680006027221,
|
|
"epoch": 2.7090999502734956,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000460895343125737,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.14177593141794204,
|
|
"num_tokens": 5340836.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 6.671760082244873,
|
|
"epoch": 2.714072600696171,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00046067504880075425,
|
|
"loss": 6.2947,
|
|
"mean_token_accuracy": 0.15082252472639085,
|
|
"num_tokens": 5351093.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 6.577940464019775,
|
|
"epoch": 2.719045251118846,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004604541951098122,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.14810417145490645,
|
|
"num_tokens": 5360859.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 6.581160068511963,
|
|
"epoch": 2.724017901541522,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004602327827182565,
|
|
"loss": 6.1098,
|
|
"mean_token_accuracy": 0.16276506781578065,
|
|
"num_tokens": 5370591.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 6.587263441085815,
|
|
"epoch": 2.728990551964197,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004600108122931156,
|
|
"loss": 6.2278,
|
|
"mean_token_accuracy": 0.1518135279417038,
|
|
"num_tokens": 5380019.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 6.589297914505005,
|
|
"epoch": 2.7339632023868723,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045978828450309907,
|
|
"loss": 6.3256,
|
|
"mean_token_accuracy": 0.14697625041007994,
|
|
"num_tokens": 5389558.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 6.637690591812134,
|
|
"epoch": 2.7389358528095475,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004595652000185958,
|
|
"loss": 6.3364,
|
|
"mean_token_accuracy": 0.1535283610224724,
|
|
"num_tokens": 5398305.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 6.687961626052856,
|
|
"epoch": 2.743908503232223,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00045934155951167163,
|
|
"loss": 6.4303,
|
|
"mean_token_accuracy": 0.14095418080687522,
|
|
"num_tokens": 5406886.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 6.594694900512695,
|
|
"epoch": 2.748881153654898,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004591173636560675,
|
|
"loss": 6.4133,
|
|
"mean_token_accuracy": 0.14678808599710463,
|
|
"num_tokens": 5416404.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 6.695959281921387,
|
|
"epoch": 2.7538538040775733,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004588926131271974,
|
|
"loss": 6.2795,
|
|
"mean_token_accuracy": 0.1515798196196556,
|
|
"num_tokens": 5425746.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 6.649276971817017,
|
|
"epoch": 2.7588264545002485,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045866730860214655,
|
|
"loss": 6.3783,
|
|
"mean_token_accuracy": 0.14690650776028633,
|
|
"num_tokens": 5436089.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 6.562987565994263,
|
|
"epoch": 2.763799104922924,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004584414507596687,
|
|
"loss": 6.3003,
|
|
"mean_token_accuracy": 0.15092788338661195,
|
|
"num_tokens": 5446270.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 6.647756719589234,
|
|
"epoch": 2.768771755345599,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00045821504028018494,
|
|
"loss": 6.3155,
|
|
"mean_token_accuracy": 0.1497044451534748,
|
|
"num_tokens": 5455476.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 6.669409370422363,
|
|
"epoch": 2.7737444057682747,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004579880778457811,
|
|
"loss": 6.3293,
|
|
"mean_token_accuracy": 0.1467943236231804,
|
|
"num_tokens": 5463813.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 6.608752155303955,
|
|
"epoch": 2.77871705619095,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004577605641402056,
|
|
"loss": 6.3333,
|
|
"mean_token_accuracy": 0.14270150810480117,
|
|
"num_tokens": 5473127.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 6.55535888671875,
|
|
"epoch": 2.7836897066136252,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004575324998488681,
|
|
"loss": 6.2767,
|
|
"mean_token_accuracy": 0.15139692574739455,
|
|
"num_tokens": 5481595.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 6.569118070602417,
|
|
"epoch": 2.7886623570363005,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004573038856588364,
|
|
"loss": 6.3261,
|
|
"mean_token_accuracy": 0.14761569127440452,
|
|
"num_tokens": 5491495.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 6.5763898372650145,
|
|
"epoch": 2.7936350074589757,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004570747222588354,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.1482478231191635,
|
|
"num_tokens": 5500702.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 6.639326572418213,
|
|
"epoch": 2.798607657881651,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00045684501033924443,
|
|
"loss": 6.2338,
|
|
"mean_token_accuracy": 0.15206692963838578,
|
|
"num_tokens": 5510223.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 6.672380781173706,
|
|
"epoch": 2.8035803083043263,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00045661475059209503,
|
|
"loss": 6.4191,
|
|
"mean_token_accuracy": 0.13695837780833245,
|
|
"num_tokens": 5522063.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 6.584640026092529,
|
|
"epoch": 2.8085529587270015,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004563839437110695,
|
|
"loss": 6.293,
|
|
"mean_token_accuracy": 0.15412321910262108,
|
|
"num_tokens": 5531323.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 6.618675756454468,
|
|
"epoch": 2.8135256091496768,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045615259039149826,
|
|
"loss": 6.2949,
|
|
"mean_token_accuracy": 0.15227940678596497,
|
|
"num_tokens": 5540637.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 6.630602645874023,
|
|
"epoch": 2.818498259572352,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00045592069133035783,
|
|
"loss": 6.229,
|
|
"mean_token_accuracy": 0.15706085562705993,
|
|
"num_tokens": 5550992.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 6.544808959960937,
|
|
"epoch": 2.8234709099950273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004556882472262691,
|
|
"loss": 6.2617,
|
|
"mean_token_accuracy": 0.14844310879707337,
|
|
"num_tokens": 5561140.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 6.589905023574829,
|
|
"epoch": 2.8284435604177025,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004554552587794948,
|
|
"loss": 6.3147,
|
|
"mean_token_accuracy": 0.14927777722477914,
|
|
"num_tokens": 5570457.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 6.630663967132568,
|
|
"epoch": 2.8334162108403778,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004552217266919375,
|
|
"loss": 6.4772,
|
|
"mean_token_accuracy": 0.1352910540997982,
|
|
"num_tokens": 5579950.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 6.695856094360352,
|
|
"epoch": 2.838388861263053,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004549876516671376,
|
|
"loss": 6.3486,
|
|
"mean_token_accuracy": 0.14736566990613936,
|
|
"num_tokens": 5589441.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 6.572498178482055,
|
|
"epoch": 2.8433615116857283,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00045475303441027114,
|
|
"loss": 6.1749,
|
|
"mean_token_accuracy": 0.15458777770400048,
|
|
"num_tokens": 5598583.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 6.5890871524810795,
|
|
"epoch": 2.8483341621084035,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004545178756281479,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.15485294461250304,
|
|
"num_tokens": 5609198.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 6.607229900360108,
|
|
"epoch": 2.853306812531079,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.00045428217602920876,
|
|
"loss": 6.2885,
|
|
"mean_token_accuracy": 0.14980829805135726,
|
|
"num_tokens": 5620425.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 6.4953470706939695,
|
|
"epoch": 2.8582794629537545,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00045404593632352385,
|
|
"loss": 6.125,
|
|
"mean_token_accuracy": 0.1623014032840729,
|
|
"num_tokens": 5629578.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 6.609415292739868,
|
|
"epoch": 2.8632521133764297,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004538091572227909,
|
|
"loss": 6.3111,
|
|
"mean_token_accuracy": 0.1451733350753784,
|
|
"num_tokens": 5639817.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 6.713958406448365,
|
|
"epoch": 2.868224763799105,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004535718394403319,
|
|
"loss": 6.4119,
|
|
"mean_token_accuracy": 0.14904803037643433,
|
|
"num_tokens": 5649130.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 6.584084415435791,
|
|
"epoch": 2.87319741422178,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004533339836910924,
|
|
"loss": 6.3325,
|
|
"mean_token_accuracy": 0.1534058004617691,
|
|
"num_tokens": 5659672.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 6.641473484039307,
|
|
"epoch": 2.8781700646444555,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004530955906916382,
|
|
"loss": 6.3855,
|
|
"mean_token_accuracy": 0.14638671427965164,
|
|
"num_tokens": 5669491.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 6.49205641746521,
|
|
"epoch": 2.8831427150671307,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004528566611601536,
|
|
"loss": 6.0802,
|
|
"mean_token_accuracy": 0.16959319561719893,
|
|
"num_tokens": 5679444.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 6.630499839782715,
|
|
"epoch": 2.888115365489806,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004526171958164396,
|
|
"loss": 6.4256,
|
|
"mean_token_accuracy": 0.14260466545820236,
|
|
"num_tokens": 5690997.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 6.659808874130249,
|
|
"epoch": 2.893088015912481,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00045237719538191107,
|
|
"loss": 6.3158,
|
|
"mean_token_accuracy": 0.1496461257338524,
|
|
"num_tokens": 5700727.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 6.6553871631622314,
|
|
"epoch": 2.898060666335157,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000452136660579595,
|
|
"loss": 6.2912,
|
|
"mean_token_accuracy": 0.1492832437157631,
|
|
"num_tokens": 5710809.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 6.452506446838379,
|
|
"epoch": 2.903033316757832,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045189559213412825,
|
|
"loss": 6.2022,
|
|
"mean_token_accuracy": 0.16337331980466843,
|
|
"num_tokens": 5720286.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 6.591053771972656,
|
|
"epoch": 2.9080059671805074,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004516539907717553,
|
|
"loss": 6.2924,
|
|
"mean_token_accuracy": 0.15211495161056518,
|
|
"num_tokens": 5730627.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 6.5438313484191895,
|
|
"epoch": 2.9129786176031827,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045141185722032617,
|
|
"loss": 6.195,
|
|
"mean_token_accuracy": 0.15850820243358613,
|
|
"num_tokens": 5739752.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 6.609400606155395,
|
|
"epoch": 2.917951268025858,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045116919220929407,
|
|
"loss": 6.2793,
|
|
"mean_token_accuracy": 0.15536951422691345,
|
|
"num_tokens": 5748277.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 6.502541399002075,
|
|
"epoch": 2.922923918448533,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004509259964697133,
|
|
"loss": 6.1119,
|
|
"mean_token_accuracy": 0.16573403105139733,
|
|
"num_tokens": 5758632.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 6.586759424209594,
|
|
"epoch": 2.9278965688712084,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000450682270734237,
|
|
"loss": 6.2687,
|
|
"mean_token_accuracy": 0.1532439410686493,
|
|
"num_tokens": 5769107.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 6.568906641006469,
|
|
"epoch": 2.9328692192938837,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004504380157371152,
|
|
"loss": 6.2717,
|
|
"mean_token_accuracy": 0.1488373950123787,
|
|
"num_tokens": 5779051.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 6.603536605834961,
|
|
"epoch": 2.937841869716559,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00045019323221419204,
|
|
"loss": 6.2553,
|
|
"mean_token_accuracy": 0.14949641972780228,
|
|
"num_tokens": 5788826.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 6.512678432464599,
|
|
"epoch": 2.942814520139234,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044994792090290425,
|
|
"loss": 6.3043,
|
|
"mean_token_accuracy": 0.14647670686244965,
|
|
"num_tokens": 5798003.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 6.5521715641021725,
|
|
"epoch": 2.9477871705619094,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00044970208254227836,
|
|
"loss": 6.2625,
|
|
"mean_token_accuracy": 0.15389290899038316,
|
|
"num_tokens": 5807215.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 6.609491014480591,
|
|
"epoch": 2.9527598209845847,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044945571787292866,
|
|
"loss": 6.269,
|
|
"mean_token_accuracy": 0.1505482867360115,
|
|
"num_tokens": 5816629.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 6.575523805618286,
|
|
"epoch": 2.95773247140726,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004492088276370552,
|
|
"loss": 6.2654,
|
|
"mean_token_accuracy": 0.1527495115995407,
|
|
"num_tokens": 5825840.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 6.684401988983154,
|
|
"epoch": 2.962705121829935,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004489614125784411,
|
|
"loss": 6.4351,
|
|
"mean_token_accuracy": 0.13940569460392,
|
|
"num_tokens": 5835754.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 6.516243982315063,
|
|
"epoch": 2.9676777722526104,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004487134734424509,
|
|
"loss": 6.3148,
|
|
"mean_token_accuracy": 0.15012251436710358,
|
|
"num_tokens": 5846062.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 6.642323064804077,
|
|
"epoch": 2.9726504226752857,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004484650109760276,
|
|
"loss": 6.3778,
|
|
"mean_token_accuracy": 0.14436433911323548,
|
|
"num_tokens": 5855589.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 6.587514972686767,
|
|
"epoch": 2.9776230730979614,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004482160259276911,
|
|
"loss": 6.167,
|
|
"mean_token_accuracy": 0.16596584916114807,
|
|
"num_tokens": 5864917.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 6.638722610473633,
|
|
"epoch": 2.9825957235206366,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004479665190475355,
|
|
"loss": 6.2921,
|
|
"mean_token_accuracy": 0.1558024726808071,
|
|
"num_tokens": 5875530.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.9825957235206366,
|
|
"eval_entropy": 6.421449354716709,
|
|
"eval_loss": 6.6491522789001465,
|
|
"eval_mean_token_accuracy": 0.13338994867413,
|
|
"eval_num_tokens": 5875530.0,
|
|
"eval_runtime": 1.914,
|
|
"eval_samples_per_second": 1810.378,
|
|
"eval_steps_per_second": 226.754,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 6.62026743888855,
|
|
"epoch": 2.987568373943312,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044771649108722695,
|
|
"loss": 6.3399,
|
|
"mean_token_accuracy": 0.14368895590305328,
|
|
"num_tokens": 5885363.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 6.628533887863159,
|
|
"epoch": 2.992541024365987,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004474659428000014,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.1457657128572464,
|
|
"num_tokens": 5895318.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 6.625722599029541,
|
|
"epoch": 2.9975136747886624,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044721487494066254,
|
|
"loss": 6.3829,
|
|
"mean_token_accuracy": 0.14991095587611197,
|
|
"num_tokens": 5905742.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 6.600646177927653,
|
|
"epoch": 3.00198906016907,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000446963288265579,
|
|
"loss": 6.1373,
|
|
"mean_token_accuracy": 0.16561883522404564,
|
|
"num_tokens": 5913844.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 6.616754531860352,
|
|
"epoch": 3.0069617105917454,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00044671118353268257,
|
|
"loss": 6.0091,
|
|
"mean_token_accuracy": 0.154216206073761,
|
|
"num_tokens": 5923724.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 6.544166231155396,
|
|
"epoch": 3.0119343610144207,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044645856150146593,
|
|
"loss": 6.0441,
|
|
"mean_token_accuracy": 0.15672821253538133,
|
|
"num_tokens": 5933787.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 6.572001218795776,
|
|
"epoch": 3.016907011437096,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00044620542293297974,
|
|
"loss": 6.0367,
|
|
"mean_token_accuracy": 0.1562382996082306,
|
|
"num_tokens": 5944186.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 6.584859704971313,
|
|
"epoch": 3.021879661859771,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004459517685898312,
|
|
"loss": 6.0396,
|
|
"mean_token_accuracy": 0.15635523647069932,
|
|
"num_tokens": 5954465.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 6.634255361557007,
|
|
"epoch": 3.0268523122824464,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044569759923618113,
|
|
"loss": 6.0629,
|
|
"mean_token_accuracy": 0.15396559461951256,
|
|
"num_tokens": 5965047.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 6.508171606063843,
|
|
"epoch": 3.0318249627051217,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004454429156377419,
|
|
"loss": 5.872,
|
|
"mean_token_accuracy": 0.17050796747207642,
|
|
"num_tokens": 5974855.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 6.513836479187011,
|
|
"epoch": 3.036797613127797,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004451877185617751,
|
|
"loss": 5.8592,
|
|
"mean_token_accuracy": 0.16822365671396255,
|
|
"num_tokens": 5985965.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 6.642105484008789,
|
|
"epoch": 3.041770263550472,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004449320087770894,
|
|
"loss": 6.0949,
|
|
"mean_token_accuracy": 0.15949613749980926,
|
|
"num_tokens": 5996646.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 6.55168924331665,
|
|
"epoch": 3.046742913973148,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004446757870540376,
|
|
"loss": 5.8854,
|
|
"mean_token_accuracy": 0.17519350796937944,
|
|
"num_tokens": 6005825.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 6.407686471939087,
|
|
"epoch": 3.051715564395823,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004444190541645153,
|
|
"loss": 5.8325,
|
|
"mean_token_accuracy": 0.1689691051840782,
|
|
"num_tokens": 6016909.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 6.566406011581421,
|
|
"epoch": 3.0566882148184984,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044416181088195776,
|
|
"loss": 6.0348,
|
|
"mean_token_accuracy": 0.1592218518257141,
|
|
"num_tokens": 6025511.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 6.563805341720581,
|
|
"epoch": 3.0616608652411736,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004439040579813379,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.16558490693569183,
|
|
"num_tokens": 6035114.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 6.5152907371521,
|
|
"epoch": 3.066633515663849,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00044364579623916385,
|
|
"loss": 5.9936,
|
|
"mean_token_accuracy": 0.15927704870700837,
|
|
"num_tokens": 6044963.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 6.448471641540527,
|
|
"epoch": 3.071606166086524,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004433870264334767,
|
|
"loss": 5.8463,
|
|
"mean_token_accuracy": 0.17590678930282594,
|
|
"num_tokens": 6054660.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 6.5207359790802,
|
|
"epoch": 3.0765788165091994,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004431277493438484,
|
|
"loss": 5.9442,
|
|
"mean_token_accuracy": 0.16297169625759125,
|
|
"num_tokens": 6064288.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 6.546789741516113,
|
|
"epoch": 3.0815514669318746,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004428679657513789,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.15998447984457015,
|
|
"num_tokens": 6074133.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 6.44585108757019,
|
|
"epoch": 3.08652411735455,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004426076764386939,
|
|
"loss": 5.9876,
|
|
"mean_token_accuracy": 0.16167742758989334,
|
|
"num_tokens": 6084314.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 6.449618864059448,
|
|
"epoch": 3.091496767777225,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00044234688218994293,
|
|
"loss": 5.936,
|
|
"mean_token_accuracy": 0.16978294998407364,
|
|
"num_tokens": 6094616.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 6.536327791213989,
|
|
"epoch": 3.0964694181999004,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004420855837907967,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.1610591694712639,
|
|
"num_tokens": 6104167.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 6.3183704853057865,
|
|
"epoch": 3.1014420686225757,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044182378202844427,
|
|
"loss": 5.7561,
|
|
"mean_token_accuracy": 0.1797334685921669,
|
|
"num_tokens": 6113970.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 6.520317554473877,
|
|
"epoch": 3.106414719045251,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00044156147769159183,
|
|
"loss": 5.9604,
|
|
"mean_token_accuracy": 0.15754238963127137,
|
|
"num_tokens": 6123482.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 6.459848928451538,
|
|
"epoch": 3.1113873694679266,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004412986715704591,
|
|
"loss": 5.9364,
|
|
"mean_token_accuracy": 0.16224179416894913,
|
|
"num_tokens": 6133682.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 6.480337524414063,
|
|
"epoch": 3.116360019890602,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004410353644567777,
|
|
"loss": 5.959,
|
|
"mean_token_accuracy": 0.16604810804128647,
|
|
"num_tokens": 6143876.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 6.527704524993896,
|
|
"epoch": 3.121332670313277,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00044077155714378836,
|
|
"loss": 6.0642,
|
|
"mean_token_accuracy": 0.15960919260978698,
|
|
"num_tokens": 6153615.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 6.4981639862060545,
|
|
"epoch": 3.1263053207359524,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004405072504262391,
|
|
"loss": 5.9401,
|
|
"mean_token_accuracy": 0.1741950899362564,
|
|
"num_tokens": 6163572.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 6.442615032196045,
|
|
"epoch": 3.1312779711586276,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000440242445100382,
|
|
"loss": 5.9372,
|
|
"mean_token_accuracy": 0.16138219684362412,
|
|
"num_tokens": 6173085.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 6.433617639541626,
|
|
"epoch": 3.136250621581303,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043997714196397143,
|
|
"loss": 6.0001,
|
|
"mean_token_accuracy": 0.15629519745707512,
|
|
"num_tokens": 6183124.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 6.564781951904297,
|
|
"epoch": 3.141223272003978,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043971134181626173,
|
|
"loss": 6.0243,
|
|
"mean_token_accuracy": 0.1615046039223671,
|
|
"num_tokens": 6192817.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 6.481431484222412,
|
|
"epoch": 3.1461959224266534,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00043944504545800396,
|
|
"loss": 6.0003,
|
|
"mean_token_accuracy": 0.16734340488910676,
|
|
"num_tokens": 6203643.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 6.498181962966919,
|
|
"epoch": 3.1511685728493286,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004391782536914447,
|
|
"loss": 5.9926,
|
|
"mean_token_accuracy": 0.16047534346580505,
|
|
"num_tokens": 6213712.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 6.499766111373901,
|
|
"epoch": 3.156141223272004,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00043891096732032257,
|
|
"loss": 6.077,
|
|
"mean_token_accuracy": 0.15568471997976302,
|
|
"num_tokens": 6223962.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 6.513842296600342,
|
|
"epoch": 3.161113873694679,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043864318714986625,
|
|
"loss": 6.0637,
|
|
"mean_token_accuracy": 0.1584389850497246,
|
|
"num_tokens": 6235273.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 6.488967990875244,
|
|
"epoch": 3.1660865241173544,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043837491398679234,
|
|
"loss": 5.87,
|
|
"mean_token_accuracy": 0.16557106673717498,
|
|
"num_tokens": 6245161.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 6.458056306838989,
|
|
"epoch": 3.17105917454003,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004381061486393025,
|
|
"loss": 5.9519,
|
|
"mean_token_accuracy": 0.16282296180725098,
|
|
"num_tokens": 6253300.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 6.44740047454834,
|
|
"epoch": 3.1760318249627053,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004378368919170809,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.16672297269105912,
|
|
"num_tokens": 6263525.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 6.450349426269531,
|
|
"epoch": 3.1810044753853806,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043756714463129244,
|
|
"loss": 5.9415,
|
|
"mean_token_accuracy": 0.1665143772959709,
|
|
"num_tokens": 6274010.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 6.488735485076904,
|
|
"epoch": 3.185977125808056,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004372969075945796,
|
|
"loss": 6.0076,
|
|
"mean_token_accuracy": 0.16516091376543046,
|
|
"num_tokens": 6283442.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 6.401978921890259,
|
|
"epoch": 3.190949776230731,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004370261816210604,
|
|
"loss": 5.9595,
|
|
"mean_token_accuracy": 0.16378860473632811,
|
|
"num_tokens": 6293498.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 6.341762256622315,
|
|
"epoch": 3.1959224266534063,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004367549675263258,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.1836935058236122,
|
|
"num_tokens": 6303557.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 6.5663840770721436,
|
|
"epoch": 3.2008950770760816,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004364832661274373,
|
|
"loss": 6.147,
|
|
"mean_token_accuracy": 0.15182227045297622,
|
|
"num_tokens": 6314304.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 6.496912574768066,
|
|
"epoch": 3.205867727498757,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00043621107824292444,
|
|
"loss": 6.0052,
|
|
"mean_token_accuracy": 0.15691158473491668,
|
|
"num_tokens": 6324959.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 6.341780805587769,
|
|
"epoch": 3.210840377921432,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004359384046927824,
|
|
"loss": 5.7994,
|
|
"mean_token_accuracy": 0.18077541887760162,
|
|
"num_tokens": 6335909.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 6.464926910400391,
|
|
"epoch": 3.2158130283441073,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004356652462984693,
|
|
"loss": 6.0476,
|
|
"mean_token_accuracy": 0.15646140947937964,
|
|
"num_tokens": 6346659.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 6.389922952651977,
|
|
"epoch": 3.2207856787667826,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00043539160388290416,
|
|
"loss": 5.821,
|
|
"mean_token_accuracy": 0.18328989446163177,
|
|
"num_tokens": 6357264.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 6.475707197189331,
|
|
"epoch": 3.225758329189458,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004351174782704641,
|
|
"loss": 6.0617,
|
|
"mean_token_accuracy": 0.15647744238376618,
|
|
"num_tokens": 6366107.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 6.420549583435059,
|
|
"epoch": 3.230730979612133,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043484287028698176,
|
|
"loss": 5.8564,
|
|
"mean_token_accuracy": 0.17163854092359543,
|
|
"num_tokens": 6375505.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 6.51609468460083,
|
|
"epoch": 3.2357036300348083,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043456778075974306,
|
|
"loss": 6.0664,
|
|
"mean_token_accuracy": 0.15764744728803634,
|
|
"num_tokens": 6384701.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 6.495784854888916,
|
|
"epoch": 3.240676280457484,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00043429221051748487,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.16849285662174224,
|
|
"num_tokens": 6393707.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 6.420533752441406,
|
|
"epoch": 3.2456489308801593,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043401616039039174,
|
|
"loss": 6.0126,
|
|
"mean_token_accuracy": 0.16080366522073747,
|
|
"num_tokens": 6404277.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 6.48481936454773,
|
|
"epoch": 3.2506215813028345,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004337396312100944,
|
|
"loss": 6.0853,
|
|
"mean_token_accuracy": 0.15636606961488725,
|
|
"num_tokens": 6415275.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 6.520675897598267,
|
|
"epoch": 3.2555942317255098,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004334626238096665,
|
|
"loss": 5.9664,
|
|
"mean_token_accuracy": 0.1631862699985504,
|
|
"num_tokens": 6425598.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 6.430893993377685,
|
|
"epoch": 3.260566882148185,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004331851390236227,
|
|
"loss": 5.9175,
|
|
"mean_token_accuracy": 0.16355313807725907,
|
|
"num_tokens": 6435685.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 6.383790683746338,
|
|
"epoch": 3.2655395325708603,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043290717768791533,
|
|
"loss": 5.9108,
|
|
"mean_token_accuracy": 0.16874608099460603,
|
|
"num_tokens": 6444950.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 6.42014446258545,
|
|
"epoch": 3.2705121829935355,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00043262874063993285,
|
|
"loss": 5.9284,
|
|
"mean_token_accuracy": 0.16107390969991683,
|
|
"num_tokens": 6454066.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 6.4636125564575195,
|
|
"epoch": 3.2754848334162108,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004323498287184967,
|
|
"loss": 6.0548,
|
|
"mean_token_accuracy": 0.15900435745716096,
|
|
"num_tokens": 6464912.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 6.46969084739685,
|
|
"epoch": 3.280457483838886,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043207044276385884,
|
|
"loss": 6.0857,
|
|
"mean_token_accuracy": 0.1524003617465496,
|
|
"num_tokens": 6474178.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 6.345533323287964,
|
|
"epoch": 3.2854301342615613,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00043179058361769924,
|
|
"loss": 5.7685,
|
|
"mean_token_accuracy": 0.1886396273970604,
|
|
"num_tokens": 6485703.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 6.45140380859375,
|
|
"epoch": 3.2904027846842365,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004315102521231237,
|
|
"loss": 5.9338,
|
|
"mean_token_accuracy": 0.16112266033887862,
|
|
"num_tokens": 6495309.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 6.506579637527466,
|
|
"epoch": 3.295375435106912,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004312294491246609,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.16261567324399948,
|
|
"num_tokens": 6505498.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 6.38797402381897,
|
|
"epoch": 3.3003480855295875,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004309481754682597,
|
|
"loss": 5.9288,
|
|
"mean_token_accuracy": 0.16781889349222184,
|
|
"num_tokens": 6514656.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 6.441996288299561,
|
|
"epoch": 3.3053207359522627,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00043066643200128724,
|
|
"loss": 5.9885,
|
|
"mean_token_accuracy": 0.16130710691213607,
|
|
"num_tokens": 6524549.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 6.421645593643189,
|
|
"epoch": 3.310293386374938,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004303842195725259,
|
|
"loss": 5.899,
|
|
"mean_token_accuracy": 0.16915222108364106,
|
|
"num_tokens": 6533861.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 6.5031898021698,
|
|
"epoch": 3.3152660367976132,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00043010153903217074,
|
|
"loss": 6.0722,
|
|
"mean_token_accuracy": 0.15887998342514037,
|
|
"num_tokens": 6543676.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 6.4964241027832035,
|
|
"epoch": 3.3202386872202885,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004298183912318273,
|
|
"loss": 5.9743,
|
|
"mean_token_accuracy": 0.15871274918317796,
|
|
"num_tokens": 6552263.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 6.393415069580078,
|
|
"epoch": 3.3252113376429637,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042953477702450845,
|
|
"loss": 5.9324,
|
|
"mean_token_accuracy": 0.16745486706495286,
|
|
"num_tokens": 6560960.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 6.359142780303955,
|
|
"epoch": 3.330183988065639,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004292506972646324,
|
|
"loss": 5.8488,
|
|
"mean_token_accuracy": 0.17926548719406127,
|
|
"num_tokens": 6572345.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 6.456111145019531,
|
|
"epoch": 3.3351566384883142,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042896615280802006,
|
|
"loss": 5.9981,
|
|
"mean_token_accuracy": 0.16218914091587067,
|
|
"num_tokens": 6582274.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 6.404939317703247,
|
|
"epoch": 3.3401292889109895,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042868114451189164,
|
|
"loss": 5.938,
|
|
"mean_token_accuracy": 0.16270132809877397,
|
|
"num_tokens": 6592486.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 6.373557090759277,
|
|
"epoch": 3.3451019393336647,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004283956732348655,
|
|
"loss": 5.8661,
|
|
"mean_token_accuracy": 0.17641573250293732,
|
|
"num_tokens": 6600876.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 6.412645769119263,
|
|
"epoch": 3.35007458975634,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042810973983695426,
|
|
"loss": 5.9049,
|
|
"mean_token_accuracy": 0.16721379160881042,
|
|
"num_tokens": 6610679.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 6.443577671051026,
|
|
"epoch": 3.3550472401790152,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004278233451795629,
|
|
"loss": 6.0061,
|
|
"mean_token_accuracy": 0.16196780055761337,
|
|
"num_tokens": 6620779.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 6.483259630203247,
|
|
"epoch": 3.3600198906016905,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000427536490125486,
|
|
"loss": 5.9585,
|
|
"mean_token_accuracy": 0.16874787509441375,
|
|
"num_tokens": 6630807.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 6.3367678165435795,
|
|
"epoch": 3.364992541024366,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00042724917553890507,
|
|
"loss": 5.8521,
|
|
"mean_token_accuracy": 0.1761890545487404,
|
|
"num_tokens": 6640530.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 6.379229259490967,
|
|
"epoch": 3.3699651914470414,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042696140228538616,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.1645087793469429,
|
|
"num_tokens": 6650845.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 6.397876453399658,
|
|
"epoch": 3.3749378418697167,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042667317123187696,
|
|
"loss": 5.9598,
|
|
"mean_token_accuracy": 0.15788703113794328,
|
|
"num_tokens": 6660455.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 6.381877088546753,
|
|
"epoch": 3.379910492292392,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042638448324670424,
|
|
"loss": 5.9226,
|
|
"mean_token_accuracy": 0.16544871181249618,
|
|
"num_tokens": 6670493.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 6.426378440856934,
|
|
"epoch": 3.384883142715067,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004260953391995716,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.16044677197933196,
|
|
"num_tokens": 6680598.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 6.277733087539673,
|
|
"epoch": 3.3898557931377424,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00042580573996155646,
|
|
"loss": 5.791,
|
|
"mean_token_accuracy": 0.18272788375616072,
|
|
"num_tokens": 6691569.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 6.486753177642822,
|
|
"epoch": 3.3948284435604177,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004255156864051074,
|
|
"loss": 6.0748,
|
|
"mean_token_accuracy": 0.15900080949068068,
|
|
"num_tokens": 6702570.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 6.3097998142242435,
|
|
"epoch": 3.399801093983093,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00042522517940404186,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.18536251783370972,
|
|
"num_tokens": 6711628.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 6.446558809280395,
|
|
"epoch": 3.404773744405768,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004249342198335433,
|
|
"loss": 5.9455,
|
|
"mean_token_accuracy": 0.16268555223941802,
|
|
"num_tokens": 6721828.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 6.265483808517456,
|
|
"epoch": 3.4097463948284434,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00042464280857015856,
|
|
"loss": 5.8048,
|
|
"mean_token_accuracy": 0.17929885238409043,
|
|
"num_tokens": 6731903.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 6.331756353378296,
|
|
"epoch": 3.4147190452511187,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042435094649179525,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.17659308463335038,
|
|
"num_tokens": 6743236.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 6.295539665222168,
|
|
"epoch": 3.419691695673794,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004240586344777191,
|
|
"loss": 5.785,
|
|
"mean_token_accuracy": 0.18008646816015245,
|
|
"num_tokens": 6753291.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 6.52609076499939,
|
|
"epoch": 3.4246643460964696,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004237658734085514,
|
|
"loss": 5.9903,
|
|
"mean_token_accuracy": 0.1593202605843544,
|
|
"num_tokens": 6762635.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 6.449252128601074,
|
|
"epoch": 3.429636996519145,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042347266416626624,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.16464891135692597,
|
|
"num_tokens": 6772101.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 6.349958610534668,
|
|
"epoch": 3.43460964694182,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004231790076341877,
|
|
"loss": 5.9466,
|
|
"mean_token_accuracy": 0.16573296934366227,
|
|
"num_tokens": 6781563.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 6.439535236358642,
|
|
"epoch": 3.4395822973644954,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004228849046969876,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.16685377210378646,
|
|
"num_tokens": 6790589.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 6.452465200424195,
|
|
"epoch": 3.4445549477871706,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004225903562406826,
|
|
"loss": 5.9537,
|
|
"mean_token_accuracy": 0.16406978368759156,
|
|
"num_tokens": 6799842.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 6.379966783523559,
|
|
"epoch": 3.449527598209846,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004222953631526313,
|
|
"loss": 6.0068,
|
|
"mean_token_accuracy": 0.16496145576238633,
|
|
"num_tokens": 6809166.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 6.373981904983521,
|
|
"epoch": 3.454500248632521,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000421999926321532,
|
|
"loss": 6.0154,
|
|
"mean_token_accuracy": 0.16067953705787658,
|
|
"num_tokens": 6818255.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 6.48245759010315,
|
|
"epoch": 3.4594728990551964,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004217040466374199,
|
|
"loss": 5.9289,
|
|
"mean_token_accuracy": 0.167622210085392,
|
|
"num_tokens": 6828628.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 6.3999439716339115,
|
|
"epoch": 3.4644455494778716,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042140772499166403,
|
|
"loss": 6.0429,
|
|
"mean_token_accuracy": 0.16167934834957123,
|
|
"num_tokens": 6839053.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 6.453637933731079,
|
|
"epoch": 3.469418199900547,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004211109622769652,
|
|
"loss": 5.9845,
|
|
"mean_token_accuracy": 0.16114179342985152,
|
|
"num_tokens": 6849226.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 6.3739683628082275,
|
|
"epoch": 3.474390850323222,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042081375938735283,
|
|
"loss": 5.9456,
|
|
"mean_token_accuracy": 0.1698800265789032,
|
|
"num_tokens": 6858867.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 6.397248697280884,
|
|
"epoch": 3.4793635007458974,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00042051611721818233,
|
|
"loss": 5.9613,
|
|
"mean_token_accuracy": 0.16845120936632157,
|
|
"num_tokens": 6869171.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 6.447233247756958,
|
|
"epoch": 3.4843361511685726,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042021803666613273,
|
|
"loss": 5.9256,
|
|
"mean_token_accuracy": 0.16369603127241134,
|
|
"num_tokens": 6878582.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 6.37340292930603,
|
|
"epoch": 3.489308801591248,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041991951862920365,
|
|
"loss": 5.9851,
|
|
"mean_token_accuracy": 0.16569314152002335,
|
|
"num_tokens": 6887522.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 6.408229827880859,
|
|
"epoch": 3.4942814520139236,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004196205640067124,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.17038443088531494,
|
|
"num_tokens": 6897556.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 6.40773344039917,
|
|
"epoch": 3.499254102436599,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000419321173699292,
|
|
"loss": 5.9478,
|
|
"mean_token_accuracy": 0.1701913520693779,
|
|
"num_tokens": 6908219.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 6.406159591674805,
|
|
"epoch": 3.504226752859274,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004190213486088877,
|
|
"loss": 5.9628,
|
|
"mean_token_accuracy": 0.15989498570561408,
|
|
"num_tokens": 6918175.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 6.41358380317688,
|
|
"epoch": 3.5091994032819493,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004187210896387546,
|
|
"loss": 6.0898,
|
|
"mean_token_accuracy": 0.16336604505777358,
|
|
"num_tokens": 6927076.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 6.452162218093872,
|
|
"epoch": 3.5141720537046246,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004184203976934552,
|
|
"loss": 5.9786,
|
|
"mean_token_accuracy": 0.167048679292202,
|
|
"num_tokens": 6936568.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 6.401066446304322,
|
|
"epoch": 3.5191447041273,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00041811927367885605,
|
|
"loss": 5.9395,
|
|
"mean_token_accuracy": 0.16605142652988433,
|
|
"num_tokens": 6945494.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 6.418497800827026,
|
|
"epoch": 3.524117354549975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004178177185021254,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.16937874257564545,
|
|
"num_tokens": 6955397.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 6.416578531265259,
|
|
"epoch": 3.5290900049726504,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004175157330717305,
|
|
"loss": 6.043,
|
|
"mean_token_accuracy": 0.15985376089811326,
|
|
"num_tokens": 6965431.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 6.3762797832489015,
|
|
"epoch": 3.5340626553953256,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000417213318297435,
|
|
"loss": 5.9032,
|
|
"mean_token_accuracy": 0.17098740339279175,
|
|
"num_tokens": 6975457.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 6.352910566329956,
|
|
"epoch": 3.539035305818001,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041691047509029555,
|
|
"loss": 5.8083,
|
|
"mean_token_accuracy": 0.17906573563814163,
|
|
"num_tokens": 6984691.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 6.26155834197998,
|
|
"epoch": 3.5440079562406765,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041660720436265987,
|
|
"loss": 5.8296,
|
|
"mean_token_accuracy": 0.1815226748585701,
|
|
"num_tokens": 6994052.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 6.41379017829895,
|
|
"epoch": 3.548980606663352,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004163035070281634,
|
|
"loss": 5.9577,
|
|
"mean_token_accuracy": 0.16652003228664397,
|
|
"num_tokens": 7003625.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 6.401768636703491,
|
|
"epoch": 3.553953257086027,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00041599938400172686,
|
|
"loss": 6.0171,
|
|
"mean_token_accuracy": 0.16084549576044083,
|
|
"num_tokens": 7011967.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 6.46328444480896,
|
|
"epoch": 3.5589259075087023,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041569483619955354,
|
|
"loss": 6.0349,
|
|
"mean_token_accuracy": 0.1648852750658989,
|
|
"num_tokens": 7021722.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 6.506385135650635,
|
|
"epoch": 3.5638985579313776,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00041538986453912616,
|
|
"loss": 6.0646,
|
|
"mean_token_accuracy": 0.15883654356002808,
|
|
"num_tokens": 7031720.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 6.371366548538208,
|
|
"epoch": 3.568871208354053,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004150844699392045,
|
|
"loss": 6.023,
|
|
"mean_token_accuracy": 0.1637276753783226,
|
|
"num_tokens": 7040575.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 6.498377513885498,
|
|
"epoch": 3.573843858776728,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041477865331982264,
|
|
"loss": 5.9974,
|
|
"mean_token_accuracy": 0.16109524220228194,
|
|
"num_tokens": 7050035.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 6.304900693893432,
|
|
"epoch": 3.5788165091994033,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004144724156022857,
|
|
"loss": 5.8098,
|
|
"mean_token_accuracy": 0.1859985589981079,
|
|
"num_tokens": 7060950.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 6.374022960662842,
|
|
"epoch": 3.5837891596220786,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041416575770916763,
|
|
"loss": 5.9753,
|
|
"mean_token_accuracy": 0.16270743012428285,
|
|
"num_tokens": 7070384.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 6.469153881072998,
|
|
"epoch": 3.588761810044754,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00041385868056430827,
|
|
"loss": 6.0779,
|
|
"mean_token_accuracy": 0.15291829854249955,
|
|
"num_tokens": 7081118.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 6.415404891967773,
|
|
"epoch": 3.593734460467429,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004135511850928102,
|
|
"loss": 6.04,
|
|
"mean_token_accuracy": 0.15783979892730712,
|
|
"num_tokens": 7091157.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 6.407117462158203,
|
|
"epoch": 3.5987071108901043,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00041324327222103655,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.16366809010505676,
|
|
"num_tokens": 7101005.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 6.333864784240722,
|
|
"epoch": 3.6036797613127796,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00041293494287660785,
|
|
"loss": 5.9673,
|
|
"mean_token_accuracy": 0.16162559241056443,
|
|
"num_tokens": 7109944.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 6.461991596221924,
|
|
"epoch": 3.608652411735455,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00041262619798839914,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.15469216853380202,
|
|
"num_tokens": 7119073.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 6.461965942382813,
|
|
"epoch": 3.61362506215813,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00041231703848653746,
|
|
"loss": 6.0314,
|
|
"mean_token_accuracy": 0.15284736305475236,
|
|
"num_tokens": 7128920.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 6.4453119277954105,
|
|
"epoch": 3.6185977125808053,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00041200746530239903,
|
|
"loss": 6.0308,
|
|
"mean_token_accuracy": 0.16000636518001557,
|
|
"num_tokens": 7138691.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 6.460312795639038,
|
|
"epoch": 3.623570363003481,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004116974793686061,
|
|
"loss": 5.9361,
|
|
"mean_token_accuracy": 0.16632737666368486,
|
|
"num_tokens": 7148362.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 6.377327728271484,
|
|
"epoch": 3.6285430134261563,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004113870816190244,
|
|
"loss": 5.9464,
|
|
"mean_token_accuracy": 0.16557495594024657,
|
|
"num_tokens": 7157138.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 6.42504096031189,
|
|
"epoch": 3.6335156638488315,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004110762729887605,
|
|
"loss": 6.0816,
|
|
"mean_token_accuracy": 0.15720997527241706,
|
|
"num_tokens": 7167425.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 6.426182460784912,
|
|
"epoch": 3.6384883142715068,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00041076505441415873,
|
|
"loss": 5.978,
|
|
"mean_token_accuracy": 0.1654490649700165,
|
|
"num_tokens": 7177059.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 6.453294897079468,
|
|
"epoch": 3.643460964694182,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004104534268327982,
|
|
"loss": 6.0237,
|
|
"mean_token_accuracy": 0.16345555633306502,
|
|
"num_tokens": 7186435.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 6.409654903411865,
|
|
"epoch": 3.6484336151168573,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004101413911834904,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.16464680433273315,
|
|
"num_tokens": 7197248.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 6.361390161514282,
|
|
"epoch": 3.6534062655395325,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040982894840627614,
|
|
"loss": 5.9641,
|
|
"mean_token_accuracy": 0.16056740880012513,
|
|
"num_tokens": 7206042.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 6.4241249561309814,
|
|
"epoch": 3.6583789159622078,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004095160994424228,
|
|
"loss": 6.0235,
|
|
"mean_token_accuracy": 0.165290729701519,
|
|
"num_tokens": 7215822.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 6.4101708889007565,
|
|
"epoch": 3.663351566384883,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040920284523442125,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.1637840062379837,
|
|
"num_tokens": 7226131.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 6.356216621398926,
|
|
"epoch": 3.6683242168075583,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040888918672598337,
|
|
"loss": 5.9345,
|
|
"mean_token_accuracy": 0.16501395255327225,
|
|
"num_tokens": 7234095.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 6.361947727203369,
|
|
"epoch": 3.673296867230234,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000408575124862039,
|
|
"loss": 5.8842,
|
|
"mean_token_accuracy": 0.18133329153060912,
|
|
"num_tokens": 7243184.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 6.444510746002197,
|
|
"epoch": 3.678269517652909,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004082606605887331,
|
|
"loss": 5.9964,
|
|
"mean_token_accuracy": 0.15597949773073197,
|
|
"num_tokens": 7253654.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 6.392144823074341,
|
|
"epoch": 3.6832421680755845,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00040794579485342304,
|
|
"loss": 5.8621,
|
|
"mean_token_accuracy": 0.16191755533218383,
|
|
"num_tokens": 7262063.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 6.320219898223877,
|
|
"epoch": 3.6882148184982597,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040763052860467547,
|
|
"loss": 5.869,
|
|
"mean_token_accuracy": 0.1822064757347107,
|
|
"num_tokens": 7272490.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 6.3486518383026125,
|
|
"epoch": 3.693187468920935,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004073148627922637,
|
|
"loss": 6.0781,
|
|
"mean_token_accuracy": 0.15708443522453308,
|
|
"num_tokens": 7282938.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 6.427023792266846,
|
|
"epoch": 3.69816011934361,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004069987983671648,
|
|
"loss": 5.9317,
|
|
"mean_token_accuracy": 0.1620452344417572,
|
|
"num_tokens": 7293486.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 6.369577360153198,
|
|
"epoch": 3.7031327697662855,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00040668233628155664,
|
|
"loss": 5.9222,
|
|
"mean_token_accuracy": 0.16131380051374436,
|
|
"num_tokens": 7302806.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 6.343423175811767,
|
|
"epoch": 3.7081054201889607,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004063654774888153,
|
|
"loss": 6.0144,
|
|
"mean_token_accuracy": 0.1599692165851593,
|
|
"num_tokens": 7312250.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 6.489184045791626,
|
|
"epoch": 3.713078070611636,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004060482229435116,
|
|
"loss": 5.9407,
|
|
"mean_token_accuracy": 0.16845152974128724,
|
|
"num_tokens": 7321512.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 6.38984055519104,
|
|
"epoch": 3.7180507210343112,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004057305736014089,
|
|
"loss": 5.9881,
|
|
"mean_token_accuracy": 0.17086610794067383,
|
|
"num_tokens": 7331467.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 6.348163223266601,
|
|
"epoch": 3.7230233714569865,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004054125304194598,
|
|
"loss": 5.9381,
|
|
"mean_token_accuracy": 0.1622950702905655,
|
|
"num_tokens": 7340064.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 6.346292400360108,
|
|
"epoch": 3.7279960218796617,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040509409435580355,
|
|
"loss": 5.9239,
|
|
"mean_token_accuracy": 0.17224024087190629,
|
|
"num_tokens": 7350138.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 6.423901176452636,
|
|
"epoch": 3.732968672302337,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004047752663697627,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.16486766040325165,
|
|
"num_tokens": 7359613.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 6.43776569366455,
|
|
"epoch": 3.7379413227250122,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004044560474218409,
|
|
"loss": 5.9196,
|
|
"mean_token_accuracy": 0.171380877494812,
|
|
"num_tokens": 7368754.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 6.402848577499389,
|
|
"epoch": 3.7429139731476875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040413643847371923,
|
|
"loss": 5.9142,
|
|
"mean_token_accuracy": 0.16331055089831353,
|
|
"num_tokens": 7378930.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 6.430608224868775,
|
|
"epoch": 3.7478866235703627,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004038164404882539,
|
|
"loss": 6.0004,
|
|
"mean_token_accuracy": 0.15954973697662353,
|
|
"num_tokens": 7387281.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 6.452896642684936,
|
|
"epoch": 3.7528592739930384,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004034960544294731,
|
|
"loss": 5.8832,
|
|
"mean_token_accuracy": 0.17040052711963655,
|
|
"num_tokens": 7397039.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 6.323165988922119,
|
|
"epoch": 3.7578319244157137,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00040317528126257413,
|
|
"loss": 5.9545,
|
|
"mean_token_accuracy": 0.1684935584664345,
|
|
"num_tokens": 7406771.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 6.319827508926392,
|
|
"epoch": 3.762804574838389,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004028541219539205,
|
|
"loss": 5.9073,
|
|
"mean_token_accuracy": 0.17502459585666658,
|
|
"num_tokens": 7415743.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 6.2032078266143795,
|
|
"epoch": 3.767777225261064,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004025325774710391,
|
|
"loss": 5.7079,
|
|
"mean_token_accuracy": 0.17992841750383376,
|
|
"num_tokens": 7424243.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 6.357032060623169,
|
|
"epoch": 3.7727498756837394,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00040221064878261684,
|
|
"loss": 6.0191,
|
|
"mean_token_accuracy": 0.15919943079352378,
|
|
"num_tokens": 7434273.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 6.422726154327393,
|
|
"epoch": 3.7777225261064147,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004018883368584985,
|
|
"loss": 5.9727,
|
|
"mean_token_accuracy": 0.16274519860744477,
|
|
"num_tokens": 7444640.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 6.339961910247803,
|
|
"epoch": 3.78269517652909,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00040156564266968336,
|
|
"loss": 5.8236,
|
|
"mean_token_accuracy": 0.16762163937091829,
|
|
"num_tokens": 7453774.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 6.414996147155762,
|
|
"epoch": 3.787667826951765,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00040124256718832214,
|
|
"loss": 5.9499,
|
|
"mean_token_accuracy": 0.16798485219478607,
|
|
"num_tokens": 7464288.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 6.417632246017456,
|
|
"epoch": 3.7926404773744404,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004009191113877141,
|
|
"loss": 5.9995,
|
|
"mean_token_accuracy": 0.1584121435880661,
|
|
"num_tokens": 7474626.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 6.405249357223511,
|
|
"epoch": 3.797613127797116,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004005952762423048,
|
|
"loss": 5.9864,
|
|
"mean_token_accuracy": 0.1727811098098755,
|
|
"num_tokens": 7483393.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 6.409138345718384,
|
|
"epoch": 3.8025857782197914,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004002710627276821,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.1617937535047531,
|
|
"num_tokens": 7492197.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 6.379812622070313,
|
|
"epoch": 3.8075584286424666,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003999464718205739,
|
|
"loss": 5.9674,
|
|
"mean_token_accuracy": 0.16627068668603898,
|
|
"num_tokens": 7501822.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 6.2739269733428955,
|
|
"epoch": 3.812531079065142,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003996215044988451,
|
|
"loss": 5.8538,
|
|
"mean_token_accuracy": 0.17567533403635024,
|
|
"num_tokens": 7512686.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 6.472317123413086,
|
|
"epoch": 3.817503729487817,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00039929616174149454,
|
|
"loss": 5.9846,
|
|
"mean_token_accuracy": 0.15762512087821962,
|
|
"num_tokens": 7522593.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 6.378535175323487,
|
|
"epoch": 3.8224763799104924,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039897044452865207,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.15941504687070845,
|
|
"num_tokens": 7532448.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 6.262896966934204,
|
|
"epoch": 3.8274490303331676,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0003986443538415756,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.194780832529068,
|
|
"num_tokens": 7543008.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 6.3897003650665285,
|
|
"epoch": 3.832421680755843,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003983178906626484,
|
|
"loss": 5.8672,
|
|
"mean_token_accuracy": 0.17277711629867554,
|
|
"num_tokens": 7553236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 6.3837518215179445,
|
|
"epoch": 3.837394331178518,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00039799105597537564,
|
|
"loss": 5.9976,
|
|
"mean_token_accuracy": 0.1678009957075119,
|
|
"num_tokens": 7561583.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 6.338859748840332,
|
|
"epoch": 3.8423669816011934,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003976638507643818,
|
|
"loss": 5.9566,
|
|
"mean_token_accuracy": 0.16133857965469361,
|
|
"num_tokens": 7571290.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 6.363219404220581,
|
|
"epoch": 3.8473396320238686,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003973362760154076,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.15912483185529708,
|
|
"num_tokens": 7579976.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 6.380836296081543,
|
|
"epoch": 3.852312282446544,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003970083327153071,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.1628986269235611,
|
|
"num_tokens": 7589796.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 6.375042390823364,
|
|
"epoch": 3.857284932869219,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003966800218520445,
|
|
"loss": 5.8913,
|
|
"mean_token_accuracy": 0.17471663057804107,
|
|
"num_tokens": 7598976.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 6.3400500297546385,
|
|
"epoch": 3.8622575832918944,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003963513444146914,
|
|
"loss": 6.0419,
|
|
"mean_token_accuracy": 0.15903210192918776,
|
|
"num_tokens": 7607955.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 6.325604104995728,
|
|
"epoch": 3.8672302337145696,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0003960223013934239,
|
|
"loss": 5.9333,
|
|
"mean_token_accuracy": 0.1680869773030281,
|
|
"num_tokens": 7619822.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 6.282397031784058,
|
|
"epoch": 3.872202884137245,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003956928937795192,
|
|
"loss": 5.8323,
|
|
"mean_token_accuracy": 0.17623539119958878,
|
|
"num_tokens": 7630481.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 6.37961859703064,
|
|
"epoch": 3.8771755345599206,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000395363122565353,
|
|
"loss": 5.9489,
|
|
"mean_token_accuracy": 0.16511812061071396,
|
|
"num_tokens": 7640312.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 6.505743217468262,
|
|
"epoch": 3.882148184982596,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039503298874439655,
|
|
"loss": 6.2,
|
|
"mean_token_accuracy": 0.1468212939798832,
|
|
"num_tokens": 7649792.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 6.397310638427735,
|
|
"epoch": 3.887120835405271,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003947024933112131,
|
|
"loss": 5.9236,
|
|
"mean_token_accuracy": 0.16863808631896973,
|
|
"num_tokens": 7659458.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 6.364229726791382,
|
|
"epoch": 3.8920934858279463,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003943716372614556,
|
|
"loss": 5.9336,
|
|
"mean_token_accuracy": 0.16976955235004426,
|
|
"num_tokens": 7668770.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 6.288913774490356,
|
|
"epoch": 3.8970661362506216,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003940404215918634,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.17347243428230286,
|
|
"num_tokens": 7677067.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 6.257749319076538,
|
|
"epoch": 3.902038786673297,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00039370884730025897,
|
|
"loss": 5.9164,
|
|
"mean_token_accuracy": 0.16588278263807296,
|
|
"num_tokens": 7687723.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 6.429076766967773,
|
|
"epoch": 3.907011437095972,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003933769153855455,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.16010961532592774,
|
|
"num_tokens": 7696706.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 6.411354637145996,
|
|
"epoch": 3.9119840875186473,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003930446268477034,
|
|
"loss": 5.9741,
|
|
"mean_token_accuracy": 0.16135814636945725,
|
|
"num_tokens": 7706356.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 6.398904657363891,
|
|
"epoch": 3.9169567379413226,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003927119826877875,
|
|
"loss": 6.0208,
|
|
"mean_token_accuracy": 0.16343193426728247,
|
|
"num_tokens": 7717255.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 6.355080699920654,
|
|
"epoch": 3.921929388363998,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003923789839079238,
|
|
"loss": 5.9978,
|
|
"mean_token_accuracy": 0.15839992314577103,
|
|
"num_tokens": 7728493.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 6.437539958953858,
|
|
"epoch": 3.9269020387866735,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00039204563151130683,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.16172994971275328,
|
|
"num_tokens": 7738582.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 6.403608894348144,
|
|
"epoch": 3.931874689209349,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003917119265021965,
|
|
"loss": 6.0284,
|
|
"mean_token_accuracy": 0.1646544873714447,
|
|
"num_tokens": 7748486.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 6.2333310604095455,
|
|
"epoch": 3.936847339632024,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003913778698859146,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.18182142302393914,
|
|
"num_tokens": 7758033.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 6.368918561935425,
|
|
"epoch": 3.9418199900546993,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003910434626688427,
|
|
"loss": 5.8539,
|
|
"mean_token_accuracy": 0.16706655770540238,
|
|
"num_tokens": 7768577.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 6.350898885726929,
|
|
"epoch": 3.9467926404773745,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003907087058584183,
|
|
"loss": 5.9818,
|
|
"mean_token_accuracy": 0.16246742457151414,
|
|
"num_tokens": 7778546.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 6.354884910583496,
|
|
"epoch": 3.95176529090005,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039037360046313186,
|
|
"loss": 5.9727,
|
|
"mean_token_accuracy": 0.16801785975694655,
|
|
"num_tokens": 7787523.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 6.441542768478394,
|
|
"epoch": 3.956737941322725,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.00039003814749252454,
|
|
"loss": 5.9473,
|
|
"mean_token_accuracy": 0.16152272522449493,
|
|
"num_tokens": 7798198.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 6.448786115646362,
|
|
"epoch": 3.9617105917454003,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038970234795718424,
|
|
"loss": 6.0787,
|
|
"mean_token_accuracy": 0.15616578757762908,
|
|
"num_tokens": 7808137.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 6.374927806854248,
|
|
"epoch": 3.9666832421680756,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000389366202868743,
|
|
"loss": 5.951,
|
|
"mean_token_accuracy": 0.1709970861673355,
|
|
"num_tokens": 7817367.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 6.234365320205688,
|
|
"epoch": 3.971655892590751,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003890297132398739,
|
|
"loss": 5.7597,
|
|
"mean_token_accuracy": 0.18199113607406617,
|
|
"num_tokens": 7826419.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 6.373406457901001,
|
|
"epoch": 3.976628543013426,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003886928800842879,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.16350869908928872,
|
|
"num_tokens": 7834984.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.976628543013426,
|
|
"eval_entropy": 6.256970253957581,
|
|
"eval_loss": 6.525735378265381,
|
|
"eval_mean_token_accuracy": 0.14444928023157305,
|
|
"eval_num_tokens": 7834984.0,
|
|
"eval_runtime": 2.1142,
|
|
"eval_samples_per_second": 1638.9,
|
|
"eval_steps_per_second": 205.276,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 6.40118727684021,
|
|
"epoch": 3.9816011934361013,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038835570441673095,
|
|
"loss": 5.9624,
|
|
"mean_token_accuracy": 0.16644908785820006,
|
|
"num_tokens": 7844781.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 6.347452974319458,
|
|
"epoch": 3.9865738438587766,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003880181872529809,
|
|
"loss": 6.0007,
|
|
"mean_token_accuracy": 0.16935638189315796,
|
|
"num_tokens": 7854553.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 6.337394285202026,
|
|
"epoch": 3.991546494281452,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003876803296098443,
|
|
"loss": 5.9334,
|
|
"mean_token_accuracy": 0.1646946743130684,
|
|
"num_tokens": 7863900.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 6.433191013336182,
|
|
"epoch": 3.996519144704127,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003873421325051535,
|
|
"loss": 5.9106,
|
|
"mean_token_accuracy": 0.17189281731843947,
|
|
"num_tokens": 7873008.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 6.354134294721815,
|
|
"epoch": 4.000994530084535,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00038700359695776346,
|
|
"loss": 5.9105,
|
|
"mean_token_accuracy": 0.16437304019927979,
|
|
"num_tokens": 7882489.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 6.357184648513794,
|
|
"epoch": 4.00596718050721,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003866647239875487,
|
|
"loss": 5.7265,
|
|
"mean_token_accuracy": 0.17285685539245604,
|
|
"num_tokens": 7893370.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 6.400729417800903,
|
|
"epoch": 4.010939830929885,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00038632551461540023,
|
|
"loss": 5.6289,
|
|
"mean_token_accuracy": 0.18883104920387267,
|
|
"num_tokens": 7903815.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 6.432913970947266,
|
|
"epoch": 4.015912481352561,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003859859698632228,
|
|
"loss": 5.738,
|
|
"mean_token_accuracy": 0.1733608588576317,
|
|
"num_tokens": 7913365.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 6.396171522140503,
|
|
"epoch": 4.020885131775236,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003856460907539312,
|
|
"loss": 5.707,
|
|
"mean_token_accuracy": 0.17802942544221878,
|
|
"num_tokens": 7922973.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 6.427915763854981,
|
|
"epoch": 4.025857782197911,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003853058783114478,
|
|
"loss": 5.6731,
|
|
"mean_token_accuracy": 0.1754648670554161,
|
|
"num_tokens": 7932860.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 6.355028295516968,
|
|
"epoch": 4.030830432620586,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003849653335606987,
|
|
"loss": 5.6859,
|
|
"mean_token_accuracy": 0.1801391825079918,
|
|
"num_tokens": 7942524.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 6.404474258422852,
|
|
"epoch": 4.035803083043262,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003846244575276117,
|
|
"loss": 5.773,
|
|
"mean_token_accuracy": 0.1693308137357235,
|
|
"num_tokens": 7952289.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 6.294119167327881,
|
|
"epoch": 4.040775733465938,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038428325123911205,
|
|
"loss": 5.6108,
|
|
"mean_token_accuracy": 0.18533276468515397,
|
|
"num_tokens": 7962176.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 6.373195791244507,
|
|
"epoch": 4.045748383888613,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003839417157231205,
|
|
"loss": 5.7075,
|
|
"mean_token_accuracy": 0.17791460305452347,
|
|
"num_tokens": 7971300.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 6.266904973983765,
|
|
"epoch": 4.050721034311288,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003835998520085492,
|
|
"loss": 5.6097,
|
|
"mean_token_accuracy": 0.18678545355796813,
|
|
"num_tokens": 7981135.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 6.3125221729278564,
|
|
"epoch": 4.0556936847339635,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003832576611252993,
|
|
"loss": 5.6456,
|
|
"mean_token_accuracy": 0.17920376509428024,
|
|
"num_tokens": 7991327.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 6.408531188964844,
|
|
"epoch": 4.060666335156639,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003829151441042572,
|
|
"loss": 5.711,
|
|
"mean_token_accuracy": 0.17492799013853072,
|
|
"num_tokens": 8001318.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 6.259915113449097,
|
|
"epoch": 4.065638985579314,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038257230197729237,
|
|
"loss": 5.5251,
|
|
"mean_token_accuracy": 0.18921719938516618,
|
|
"num_tokens": 8010964.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 6.244263982772827,
|
|
"epoch": 4.070611636001989,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00038222913577725325,
|
|
"loss": 5.6716,
|
|
"mean_token_accuracy": 0.1841033160686493,
|
|
"num_tokens": 8020894.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 6.314126110076904,
|
|
"epoch": 4.0755842864246645,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00038188564653796487,
|
|
"loss": 5.653,
|
|
"mean_token_accuracy": 0.18169257342815398,
|
|
"num_tokens": 8030635.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 6.3429749488830565,
|
|
"epoch": 4.08055693684734,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003815418352942252,
|
|
"loss": 5.5929,
|
|
"mean_token_accuracy": 0.18369527012109757,
|
|
"num_tokens": 8040626.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 6.080220794677734,
|
|
"epoch": 4.085529587270015,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003811977030818026,
|
|
"loss": 5.328,
|
|
"mean_token_accuracy": 0.21303489059209824,
|
|
"num_tokens": 8051835.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 6.266879224777222,
|
|
"epoch": 4.09050223769269,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000380853250937432,
|
|
"loss": 5.6798,
|
|
"mean_token_accuracy": 0.1817305788397789,
|
|
"num_tokens": 8061818.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 6.344353103637696,
|
|
"epoch": 4.0954748881153655,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038050847989881236,
|
|
"loss": 5.7622,
|
|
"mean_token_accuracy": 0.1740456685423851,
|
|
"num_tokens": 8071693.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 6.222349500656128,
|
|
"epoch": 4.100447538538041,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003801633910046035,
|
|
"loss": 5.5162,
|
|
"mean_token_accuracy": 0.19879076033830642,
|
|
"num_tokens": 8082304.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 6.309303331375122,
|
|
"epoch": 4.105420188960716,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003798179852944225,
|
|
"loss": 5.6273,
|
|
"mean_token_accuracy": 0.17892390489578247,
|
|
"num_tokens": 8092319.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 6.3586259365081785,
|
|
"epoch": 4.110392839383391,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037947226380884115,
|
|
"loss": 5.6375,
|
|
"mean_token_accuracy": 0.18469706624746324,
|
|
"num_tokens": 8100894.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 6.269500017166138,
|
|
"epoch": 4.1153654898060665,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00037912622758938243,
|
|
"loss": 5.7638,
|
|
"mean_token_accuracy": 0.16581641733646393,
|
|
"num_tokens": 8110029.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 6.347639560699463,
|
|
"epoch": 4.120338140228742,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003787798776785174,
|
|
"loss": 5.6624,
|
|
"mean_token_accuracy": 0.17395906895399094,
|
|
"num_tokens": 8118715.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 6.25194149017334,
|
|
"epoch": 4.125310790651417,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003784332151196622,
|
|
"loss": 5.6428,
|
|
"mean_token_accuracy": 0.18580525517463684,
|
|
"num_tokens": 8128778.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 6.28365306854248,
|
|
"epoch": 4.130283441074092,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00037808624095717503,
|
|
"loss": 5.6612,
|
|
"mean_token_accuracy": 0.1774149015545845,
|
|
"num_tokens": 8138126.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 6.28666090965271,
|
|
"epoch": 4.1352560914967675,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003777389562363526,
|
|
"loss": 5.6416,
|
|
"mean_token_accuracy": 0.18997323662042617,
|
|
"num_tokens": 8147971.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 6.347967672348022,
|
|
"epoch": 4.140228741919443,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00037739136200342743,
|
|
"loss": 5.7218,
|
|
"mean_token_accuracy": 0.1752214178442955,
|
|
"num_tokens": 8157853.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 6.242703104019165,
|
|
"epoch": 4.145201392342118,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037704345930556426,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.1849085956811905,
|
|
"num_tokens": 8167187.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 6.2325104713439945,
|
|
"epoch": 4.150174042764793,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037669524919085716,
|
|
"loss": 5.5989,
|
|
"mean_token_accuracy": 0.1889001563191414,
|
|
"num_tokens": 8175492.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 6.249692344665528,
|
|
"epoch": 4.1551466931874685,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003763467327083264,
|
|
"loss": 5.6877,
|
|
"mean_token_accuracy": 0.17719288766384125,
|
|
"num_tokens": 8185356.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 6.242445755004883,
|
|
"epoch": 4.160119343610144,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00037599791090791527,
|
|
"loss": 5.4999,
|
|
"mean_token_accuracy": 0.19405680596828462,
|
|
"num_tokens": 8194833.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 6.395481252670288,
|
|
"epoch": 4.165091994032819,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003756487848404868,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.17507463991641997,
|
|
"num_tokens": 8205040.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 6.336773443222046,
|
|
"epoch": 4.170064644455495,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003752993555578205,
|
|
"loss": 5.7244,
|
|
"mean_token_accuracy": 0.17834113538265228,
|
|
"num_tokens": 8214769.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 6.178786563873291,
|
|
"epoch": 4.17503729487817,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003749496241126095,
|
|
"loss": 5.5342,
|
|
"mean_token_accuracy": 0.19815549701452256,
|
|
"num_tokens": 8225007.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 6.356832551956177,
|
|
"epoch": 4.180009945300846,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00037459959155845726,
|
|
"loss": 5.7083,
|
|
"mean_token_accuracy": 0.17546304315328598,
|
|
"num_tokens": 8234543.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 6.259526872634888,
|
|
"epoch": 4.184982595723521,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003742492589498744,
|
|
"loss": 5.5809,
|
|
"mean_token_accuracy": 0.18772460967302323,
|
|
"num_tokens": 8243790.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 6.243878412246704,
|
|
"epoch": 4.189955246146196,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037389862734227525,
|
|
"loss": 5.681,
|
|
"mean_token_accuracy": 0.1821441173553467,
|
|
"num_tokens": 8253144.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 6.287298536300659,
|
|
"epoch": 4.194927896568871,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00037354769779197514,
|
|
"loss": 5.6536,
|
|
"mean_token_accuracy": 0.1807618796825409,
|
|
"num_tokens": 8262782.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 6.320519018173218,
|
|
"epoch": 4.199900546991547,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037319647135618695,
|
|
"loss": 5.6729,
|
|
"mean_token_accuracy": 0.17970350086688996,
|
|
"num_tokens": 8271544.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 6.276837396621704,
|
|
"epoch": 4.204873197414222,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037284494909301793,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.18220667839050292,
|
|
"num_tokens": 8280574.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 6.2737256526947025,
|
|
"epoch": 4.209845847836897,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003724931320614663,
|
|
"loss": 5.6217,
|
|
"mean_token_accuracy": 0.18642164170742034,
|
|
"num_tokens": 8290318.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 6.287080383300781,
|
|
"epoch": 4.214818498259572,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00037214102132141874,
|
|
"loss": 5.652,
|
|
"mean_token_accuracy": 0.18077209889888762,
|
|
"num_tokens": 8299619.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 6.232118082046509,
|
|
"epoch": 4.219791148682248,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003717886179336465,
|
|
"loss": 5.608,
|
|
"mean_token_accuracy": 0.19260363280773163,
|
|
"num_tokens": 8310085.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 6.257794141769409,
|
|
"epoch": 4.224763799104923,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00037143592295980243,
|
|
"loss": 5.6655,
|
|
"mean_token_accuracy": 0.18731352537870408,
|
|
"num_tokens": 8320673.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 6.285166263580322,
|
|
"epoch": 4.229736449527598,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0003710829374624181,
|
|
"loss": 5.6308,
|
|
"mean_token_accuracy": 0.18265155255794524,
|
|
"num_tokens": 8331073.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 6.320485925674438,
|
|
"epoch": 4.2347090999502734,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003707296625048999,
|
|
"loss": 5.7231,
|
|
"mean_token_accuracy": 0.16957855746150016,
|
|
"num_tokens": 8341313.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 6.315844631195068,
|
|
"epoch": 4.239681750372949,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003703760991515264,
|
|
"loss": 5.7403,
|
|
"mean_token_accuracy": 0.1675717130303383,
|
|
"num_tokens": 8351004.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 6.124544763565064,
|
|
"epoch": 4.244654400795624,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037002224846744526,
|
|
"loss": 5.4733,
|
|
"mean_token_accuracy": 0.19497003257274628,
|
|
"num_tokens": 8361083.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 6.289008951187133,
|
|
"epoch": 4.249627051218299,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036966811151866943,
|
|
"loss": 5.7696,
|
|
"mean_token_accuracy": 0.17505478411912917,
|
|
"num_tokens": 8372162.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 6.3649365425109865,
|
|
"epoch": 4.2545997016409745,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036931368937207436,
|
|
"loss": 5.7519,
|
|
"mean_token_accuracy": 0.17303352952003478,
|
|
"num_tokens": 8382259.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 6.307545518875122,
|
|
"epoch": 4.25957235206365,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003689589830953948,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.18378251940011978,
|
|
"num_tokens": 8392275.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 6.262635040283203,
|
|
"epoch": 4.264545002486325,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003686039937572213,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.18501282930374147,
|
|
"num_tokens": 8402481.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 6.2491401672363285,
|
|
"epoch": 4.269517652909,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003682487224269972,
|
|
"loss": 5.6788,
|
|
"mean_token_accuracy": 0.18000535368919374,
|
|
"num_tokens": 8412609.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 6.259639406204224,
|
|
"epoch": 4.2744903033316755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003678931701750155,
|
|
"loss": 5.7005,
|
|
"mean_token_accuracy": 0.17414348125457763,
|
|
"num_tokens": 8421396.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 6.241744804382324,
|
|
"epoch": 4.279462953754351,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00036753733807241536,
|
|
"loss": 5.6616,
|
|
"mean_token_accuracy": 0.18139863461256028,
|
|
"num_tokens": 8430076.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 6.215416479110718,
|
|
"epoch": 4.284435604177026,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00036718122719117907,
|
|
"loss": 5.5158,
|
|
"mean_token_accuracy": 0.19809538573026658,
|
|
"num_tokens": 8440199.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 6.29616150856018,
|
|
"epoch": 4.289408254599701,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003668248386041288,
|
|
"loss": 5.633,
|
|
"mean_token_accuracy": 0.1883287951350212,
|
|
"num_tokens": 8449463.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 6.300569629669189,
|
|
"epoch": 4.294380905022377,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00036646817338492334,
|
|
"loss": 5.5335,
|
|
"mean_token_accuracy": 0.1879596471786499,
|
|
"num_tokens": 8459824.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 6.32459864616394,
|
|
"epoch": 4.299353555445053,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003661112326080548,
|
|
"loss": 5.677,
|
|
"mean_token_accuracy": 0.17830198705196382,
|
|
"num_tokens": 8468778.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 6.240612363815307,
|
|
"epoch": 4.304326205867728,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003657540173488455,
|
|
"loss": 5.6083,
|
|
"mean_token_accuracy": 0.1836107611656189,
|
|
"num_tokens": 8478542.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 6.163512516021728,
|
|
"epoch": 4.309298856290403,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003653965286834446,
|
|
"loss": 5.5295,
|
|
"mean_token_accuracy": 0.19456052929162979,
|
|
"num_tokens": 8489652.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 6.292203044891357,
|
|
"epoch": 4.314271506713078,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003650387676888251,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.17510022670030595,
|
|
"num_tokens": 8499951.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 6.347727584838867,
|
|
"epoch": 4.319244157135754,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003646807354427802,
|
|
"loss": 5.8624,
|
|
"mean_token_accuracy": 0.16346115469932557,
|
|
"num_tokens": 8510664.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 6.366744327545166,
|
|
"epoch": 4.324216807558429,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00036432243302392025,
|
|
"loss": 5.7072,
|
|
"mean_token_accuracy": 0.17763352543115615,
|
|
"num_tokens": 8519768.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 6.3413454532623295,
|
|
"epoch": 4.329189457981104,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00036396386151166975,
|
|
"loss": 5.7745,
|
|
"mean_token_accuracy": 0.17181522101163865,
|
|
"num_tokens": 8530909.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 6.291078519821167,
|
|
"epoch": 4.334162108403779,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003636050219862638,
|
|
"loss": 5.7876,
|
|
"mean_token_accuracy": 0.17546870559453964,
|
|
"num_tokens": 8541357.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 6.3333117961883545,
|
|
"epoch": 4.339134758826455,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036324591552874493,
|
|
"loss": 5.695,
|
|
"mean_token_accuracy": 0.17524062246084213,
|
|
"num_tokens": 8550819.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 6.335867261886596,
|
|
"epoch": 4.34410740924913,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00036288654322095973,
|
|
"loss": 5.7493,
|
|
"mean_token_accuracy": 0.1789591446518898,
|
|
"num_tokens": 8561577.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 6.382355880737305,
|
|
"epoch": 4.349080059671805,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00036252690614555557,
|
|
"loss": 5.7505,
|
|
"mean_token_accuracy": 0.16679588630795478,
|
|
"num_tokens": 8572590.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 6.268166542053223,
|
|
"epoch": 4.35405271009448,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003621670053859778,
|
|
"loss": 5.735,
|
|
"mean_token_accuracy": 0.177352237701416,
|
|
"num_tokens": 8583559.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 6.291941213607788,
|
|
"epoch": 4.359025360517156,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000361806842026466,
|
|
"loss": 5.7377,
|
|
"mean_token_accuracy": 0.1735794797539711,
|
|
"num_tokens": 8592845.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 6.1204370021820065,
|
|
"epoch": 4.363998010939831,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00036144641715205077,
|
|
"loss": 5.4736,
|
|
"mean_token_accuracy": 0.19839803129434586,
|
|
"num_tokens": 8602862.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 6.155761289596557,
|
|
"epoch": 4.368970661362506,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003610857318485507,
|
|
"loss": 5.5925,
|
|
"mean_token_accuracy": 0.1996888369321823,
|
|
"num_tokens": 8613793.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 6.126488447189331,
|
|
"epoch": 4.373943311785181,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00036072478720256885,
|
|
"loss": 5.452,
|
|
"mean_token_accuracy": 0.2003104269504547,
|
|
"num_tokens": 8623086.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 6.146832847595215,
|
|
"epoch": 4.378915962207857,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003603635843014895,
|
|
"loss": 5.5318,
|
|
"mean_token_accuracy": 0.19949927181005478,
|
|
"num_tokens": 8632392.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 6.2877930164337155,
|
|
"epoch": 4.383888612630532,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00036000212423347523,
|
|
"loss": 5.7766,
|
|
"mean_token_accuracy": 0.17095604166388512,
|
|
"num_tokens": 8642452.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 6.281671333312988,
|
|
"epoch": 4.388861263053207,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000359640408087463,
|
|
"loss": 5.7315,
|
|
"mean_token_accuracy": 0.1769225925207138,
|
|
"num_tokens": 8651431.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 6.333138942718506,
|
|
"epoch": 4.393833913475882,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003592784369531615,
|
|
"loss": 5.8105,
|
|
"mean_token_accuracy": 0.17103372365236283,
|
|
"num_tokens": 8661348.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 6.341537141799927,
|
|
"epoch": 4.398806563898558,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00035891621192104764,
|
|
"loss": 5.7467,
|
|
"mean_token_accuracy": 0.17713722288608552,
|
|
"num_tokens": 8671038.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 6.274081420898438,
|
|
"epoch": 4.403779214321233,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00035855373408236294,
|
|
"loss": 5.6785,
|
|
"mean_token_accuracy": 0.17936553955078124,
|
|
"num_tokens": 8680115.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 6.295895433425903,
|
|
"epoch": 4.408751864743908,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003581910045291107,
|
|
"loss": 5.7401,
|
|
"mean_token_accuracy": 0.17397327721118927,
|
|
"num_tokens": 8689665.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 6.315898704528808,
|
|
"epoch": 4.413724515166583,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003578280243540526,
|
|
"loss": 5.7156,
|
|
"mean_token_accuracy": 0.1769414246082306,
|
|
"num_tokens": 8700114.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 6.134942150115966,
|
|
"epoch": 4.4186971655892595,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00035746479465070515,
|
|
"loss": 5.4706,
|
|
"mean_token_accuracy": 0.20106033533811568,
|
|
"num_tokens": 8710154.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 6.284492874145508,
|
|
"epoch": 4.423669816011935,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00035710131651333683,
|
|
"loss": 5.815,
|
|
"mean_token_accuracy": 0.17258088737726213,
|
|
"num_tokens": 8720787.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 6.2351048469543455,
|
|
"epoch": 4.42864246643461,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00035673759103696447,
|
|
"loss": 5.6652,
|
|
"mean_token_accuracy": 0.18054416179656982,
|
|
"num_tokens": 8731069.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 6.285560226440429,
|
|
"epoch": 4.433615116857285,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003563736193173499,
|
|
"loss": 5.6595,
|
|
"mean_token_accuracy": 0.18274943977594377,
|
|
"num_tokens": 8740149.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 6.341178321838379,
|
|
"epoch": 4.4385877672799605,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003560094024509969,
|
|
"loss": 5.7589,
|
|
"mean_token_accuracy": 0.17266804426908494,
|
|
"num_tokens": 8749622.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 6.238858699798584,
|
|
"epoch": 4.443560417702636,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00035564494153514777,
|
|
"loss": 5.6554,
|
|
"mean_token_accuracy": 0.17559193819761276,
|
|
"num_tokens": 8759521.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 6.277806186676026,
|
|
"epoch": 4.448533068125311,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00035528023766778,
|
|
"loss": 5.7218,
|
|
"mean_token_accuracy": 0.183507938683033,
|
|
"num_tokens": 8769387.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 6.317159509658813,
|
|
"epoch": 4.453505718547986,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003549152919476031,
|
|
"loss": 5.6993,
|
|
"mean_token_accuracy": 0.1805525854229927,
|
|
"num_tokens": 8777871.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 6.283565044403076,
|
|
"epoch": 4.4584783689706615,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035455010547405513,
|
|
"loss": 5.758,
|
|
"mean_token_accuracy": 0.17433000355958939,
|
|
"num_tokens": 8787613.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 6.304928541183472,
|
|
"epoch": 4.463451019393337,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035418467934729934,
|
|
"loss": 5.7144,
|
|
"mean_token_accuracy": 0.16153289675712584,
|
|
"num_tokens": 8797075.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 6.284492874145508,
|
|
"epoch": 4.468423669816012,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00035381901466822113,
|
|
"loss": 5.7862,
|
|
"mean_token_accuracy": 0.17289407402276993,
|
|
"num_tokens": 8806087.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 6.338675022125244,
|
|
"epoch": 4.473396320238687,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003534531125384244,
|
|
"loss": 5.7389,
|
|
"mean_token_accuracy": 0.17178150862455369,
|
|
"num_tokens": 8815358.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 6.306356906890869,
|
|
"epoch": 4.4783689706613625,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00035308697406022865,
|
|
"loss": 5.7363,
|
|
"mean_token_accuracy": 0.18374397307634355,
|
|
"num_tokens": 8825559.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 6.272407674789429,
|
|
"epoch": 4.483341621084038,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00035272060033666503,
|
|
"loss": 5.632,
|
|
"mean_token_accuracy": 0.17891733050346376,
|
|
"num_tokens": 8835136.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 6.24525032043457,
|
|
"epoch": 4.488314271506713,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003523539924714738,
|
|
"loss": 5.6866,
|
|
"mean_token_accuracy": 0.1813421830534935,
|
|
"num_tokens": 8843936.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 6.300760793685913,
|
|
"epoch": 4.493286921929388,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00035198715156910016,
|
|
"loss": 5.8247,
|
|
"mean_token_accuracy": 0.16563732028007508,
|
|
"num_tokens": 8853990.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 6.209895133972168,
|
|
"epoch": 4.4982595723520635,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00035162007873469175,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.17522012144327165,
|
|
"num_tokens": 8863856.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 6.367771577835083,
|
|
"epoch": 4.503232222774739,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00035125277507409483,
|
|
"loss": 5.722,
|
|
"mean_token_accuracy": 0.17550479620695114,
|
|
"num_tokens": 8872846.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 6.165759611129761,
|
|
"epoch": 4.508204873197414,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00035088524169385075,
|
|
"loss": 5.5547,
|
|
"mean_token_accuracy": 0.19458480477333068,
|
|
"num_tokens": 8883323.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 6.2669336795806885,
|
|
"epoch": 4.513177523620089,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00035051747970119354,
|
|
"loss": 5.8158,
|
|
"mean_token_accuracy": 0.17045991271734237,
|
|
"num_tokens": 8893346.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 6.3207403182983395,
|
|
"epoch": 4.5181501740427645,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00035014949020404523,
|
|
"loss": 5.7332,
|
|
"mean_token_accuracy": 0.1744983673095703,
|
|
"num_tokens": 8902480.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 6.225380325317383,
|
|
"epoch": 4.52312282446544,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00034978127431101365,
|
|
"loss": 5.6116,
|
|
"mean_token_accuracy": 0.1824115976691246,
|
|
"num_tokens": 8913496.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 6.29714617729187,
|
|
"epoch": 4.528095474888115,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00034941283313138865,
|
|
"loss": 5.7561,
|
|
"mean_token_accuracy": 0.17123856246471406,
|
|
"num_tokens": 8924344.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 6.244879484176636,
|
|
"epoch": 4.53306812531079,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003490441677751387,
|
|
"loss": 5.6883,
|
|
"mean_token_accuracy": 0.17647765576839447,
|
|
"num_tokens": 8933602.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 6.304931545257569,
|
|
"epoch": 4.5380407757334655,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00034867527935290773,
|
|
"loss": 5.7385,
|
|
"mean_token_accuracy": 0.1770091623067856,
|
|
"num_tokens": 8943722.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 6.279794359207154,
|
|
"epoch": 4.543013426156142,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00034830616897601146,
|
|
"loss": 5.6307,
|
|
"mean_token_accuracy": 0.1868995189666748,
|
|
"num_tokens": 8953535.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 6.21785717010498,
|
|
"epoch": 4.547986076578816,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003479368377564344,
|
|
"loss": 5.4761,
|
|
"mean_token_accuracy": 0.19697033166885375,
|
|
"num_tokens": 8963775.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 6.219197940826416,
|
|
"epoch": 4.552958727001492,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003475672868068265,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.18603908121585847,
|
|
"num_tokens": 8972822.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 6.245578050613403,
|
|
"epoch": 4.557931377424167,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00034719751724049955,
|
|
"loss": 5.7485,
|
|
"mean_token_accuracy": 0.17705858498811722,
|
|
"num_tokens": 8983058.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 6.2300248622894285,
|
|
"epoch": 4.562904027846843,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00034682753017142387,
|
|
"loss": 5.7604,
|
|
"mean_token_accuracy": 0.17659793347120284,
|
|
"num_tokens": 8993069.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 6.286673831939697,
|
|
"epoch": 4.567876678269518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034645732671422517,
|
|
"loss": 5.767,
|
|
"mean_token_accuracy": 0.1731798082590103,
|
|
"num_tokens": 9001471.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 6.297007703781128,
|
|
"epoch": 4.572849328692193,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003460869079841809,
|
|
"loss": 5.724,
|
|
"mean_token_accuracy": 0.1777175098657608,
|
|
"num_tokens": 9012411.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 6.288017845153808,
|
|
"epoch": 4.577821979114868,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0003457162750972172,
|
|
"loss": 5.6143,
|
|
"mean_token_accuracy": 0.1874997243285179,
|
|
"num_tokens": 9024126.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 6.200018882751465,
|
|
"epoch": 4.582794629537544,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034534542916990534,
|
|
"loss": 5.6261,
|
|
"mean_token_accuracy": 0.18158834129571916,
|
|
"num_tokens": 9034355.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 6.250564765930176,
|
|
"epoch": 4.587767279960219,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00034497437131945837,
|
|
"loss": 5.6646,
|
|
"mean_token_accuracy": 0.1796063795685768,
|
|
"num_tokens": 9043772.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 6.291866397857666,
|
|
"epoch": 4.592739930382894,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003446031026637277,
|
|
"loss": 5.6659,
|
|
"mean_token_accuracy": 0.18503499478101731,
|
|
"num_tokens": 9053397.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 6.2510154247283936,
|
|
"epoch": 4.597712580805569,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00034423162432119995,
|
|
"loss": 5.6218,
|
|
"mean_token_accuracy": 0.17992168962955474,
|
|
"num_tokens": 9063904.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 6.253576374053955,
|
|
"epoch": 4.602685231228245,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003438599374109934,
|
|
"loss": 5.6439,
|
|
"mean_token_accuracy": 0.1815967932343483,
|
|
"num_tokens": 9073146.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 6.091863107681275,
|
|
"epoch": 4.60765788165092,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00034348804305285466,
|
|
"loss": 5.5084,
|
|
"mean_token_accuracy": 0.2017839640378952,
|
|
"num_tokens": 9083617.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 6.289822912216186,
|
|
"epoch": 4.612630532073595,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003431159423671552,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.1726579800248146,
|
|
"num_tokens": 9092639.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 6.268873453140259,
|
|
"epoch": 4.61760318249627,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034274363647488826,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.1714460387825966,
|
|
"num_tokens": 9103377.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 6.2134175300598145,
|
|
"epoch": 4.622575832918946,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003423711264976652,
|
|
"loss": 5.6088,
|
|
"mean_token_accuracy": 0.18925182670354843,
|
|
"num_tokens": 9112681.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 6.236137390136719,
|
|
"epoch": 4.627548483341621,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00034199841355771226,
|
|
"loss": 5.6399,
|
|
"mean_token_accuracy": 0.1900264799594879,
|
|
"num_tokens": 9121079.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 6.267116260528565,
|
|
"epoch": 4.632521133764296,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00034162549877786685,
|
|
"loss": 5.7275,
|
|
"mean_token_accuracy": 0.1857730820775032,
|
|
"num_tokens": 9131121.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 6.327781867980957,
|
|
"epoch": 4.6374937841869714,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003412523832815748,
|
|
"loss": 5.7798,
|
|
"mean_token_accuracy": 0.17013072669506074,
|
|
"num_tokens": 9140352.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 6.324205160140991,
|
|
"epoch": 4.642466434609647,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003408790681928865,
|
|
"loss": 5.7427,
|
|
"mean_token_accuracy": 0.172853085398674,
|
|
"num_tokens": 9150892.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 6.247129678726196,
|
|
"epoch": 4.647439085032322,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003405055546364535,
|
|
"loss": 5.775,
|
|
"mean_token_accuracy": 0.17285481244325637,
|
|
"num_tokens": 9160943.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 6.326165390014649,
|
|
"epoch": 4.652411735454997,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00034013184373752557,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.17004822492599486,
|
|
"num_tokens": 9171461.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 6.282613897323609,
|
|
"epoch": 4.6573843858776725,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003397579366219466,
|
|
"loss": 5.7341,
|
|
"mean_token_accuracy": 0.17605248391628264,
|
|
"num_tokens": 9181991.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 6.265548944473267,
|
|
"epoch": 4.662357036300348,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003393838344161519,
|
|
"loss": 5.5618,
|
|
"mean_token_accuracy": 0.1912733718752861,
|
|
"num_tokens": 9191715.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 6.252676248550415,
|
|
"epoch": 4.667329686723024,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00033900953824716443,
|
|
"loss": 5.6349,
|
|
"mean_token_accuracy": 0.18039775043725967,
|
|
"num_tokens": 9201338.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 6.298232650756836,
|
|
"epoch": 4.672302337145698,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003386350492425913,
|
|
"loss": 5.7583,
|
|
"mean_token_accuracy": 0.17435802668333053,
|
|
"num_tokens": 9210970.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 6.281692266464233,
|
|
"epoch": 4.677274987568374,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033826036853062095,
|
|
"loss": 5.7199,
|
|
"mean_token_accuracy": 0.1749281704425812,
|
|
"num_tokens": 9221304.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 6.317141485214234,
|
|
"epoch": 4.68224763799105,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033788549724001897,
|
|
"loss": 5.6765,
|
|
"mean_token_accuracy": 0.17569528818130492,
|
|
"num_tokens": 9231386.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 6.305721616744995,
|
|
"epoch": 4.687220288413725,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00033751043650012534,
|
|
"loss": 5.7537,
|
|
"mean_token_accuracy": 0.17719806581735612,
|
|
"num_tokens": 9240203.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 6.26689829826355,
|
|
"epoch": 4.6921929388364,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003371351874408506,
|
|
"loss": 5.6948,
|
|
"mean_token_accuracy": 0.18024620413780212,
|
|
"num_tokens": 9249623.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 6.280479812622071,
|
|
"epoch": 4.697165589259075,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003367597511926728,
|
|
"loss": 5.787,
|
|
"mean_token_accuracy": 0.16911298036575317,
|
|
"num_tokens": 9259359.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 6.2634992599487305,
|
|
"epoch": 4.702138239681751,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033638412888663374,
|
|
"loss": 5.6046,
|
|
"mean_token_accuracy": 0.1874556064605713,
|
|
"num_tokens": 9267897.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 6.214230155944824,
|
|
"epoch": 4.707110890104426,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003360083216543358,
|
|
"loss": 5.6113,
|
|
"mean_token_accuracy": 0.18815526813268663,
|
|
"num_tokens": 9277520.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 6.37258505821228,
|
|
"epoch": 4.712083540527101,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003356323306279387,
|
|
"loss": 5.9142,
|
|
"mean_token_accuracy": 0.1563256487250328,
|
|
"num_tokens": 9288157.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 6.298085689544678,
|
|
"epoch": 4.717056190949776,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00033525615694015545,
|
|
"loss": 5.6503,
|
|
"mean_token_accuracy": 0.18661709278821945,
|
|
"num_tokens": 9296698.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 6.300492000579834,
|
|
"epoch": 4.722028841372452,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033487980172424955,
|
|
"loss": 5.774,
|
|
"mean_token_accuracy": 0.17444614619016646,
|
|
"num_tokens": 9305880.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 6.340297698974609,
|
|
"epoch": 4.727001491795127,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00033450326611403136,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.1716974124312401,
|
|
"num_tokens": 9316123.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 6.330162239074707,
|
|
"epoch": 4.731974142217802,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033412655124385494,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.17066662311553954,
|
|
"num_tokens": 9325914.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 6.117185926437378,
|
|
"epoch": 4.736946792640477,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003337496582486139,
|
|
"loss": 5.6563,
|
|
"mean_token_accuracy": 0.1917197063565254,
|
|
"num_tokens": 9336330.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 6.2731953144073485,
|
|
"epoch": 4.741919443063153,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033337258826373896,
|
|
"loss": 5.7194,
|
|
"mean_token_accuracy": 0.17292167544364928,
|
|
"num_tokens": 9345358.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 6.279635715484619,
|
|
"epoch": 4.746892093485828,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00033299534242519355,
|
|
"loss": 5.646,
|
|
"mean_token_accuracy": 0.18476144671440126,
|
|
"num_tokens": 9355610.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 6.249187994003296,
|
|
"epoch": 4.751864743908503,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00033261792186947133,
|
|
"loss": 5.7304,
|
|
"mean_token_accuracy": 0.17904064059257507,
|
|
"num_tokens": 9364450.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 6.273398351669312,
|
|
"epoch": 4.756837394331178,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00033224032773359203,
|
|
"loss": 5.7604,
|
|
"mean_token_accuracy": 0.17725847959518432,
|
|
"num_tokens": 9374426.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 6.110723209381104,
|
|
"epoch": 4.761810044753854,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00033186256115509845,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.19618281871080398,
|
|
"num_tokens": 9384067.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 6.195953702926635,
|
|
"epoch": 4.766782695176529,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003314846232720528,
|
|
"loss": 5.4821,
|
|
"mean_token_accuracy": 0.19459174275398256,
|
|
"num_tokens": 9394269.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 6.292109394073487,
|
|
"epoch": 4.771755345599204,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00033110651522303346,
|
|
"loss": 5.6397,
|
|
"mean_token_accuracy": 0.1860302522778511,
|
|
"num_tokens": 9403930.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 6.257152032852173,
|
|
"epoch": 4.776727996021879,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003307282381471314,
|
|
"loss": 5.7067,
|
|
"mean_token_accuracy": 0.1708251029253006,
|
|
"num_tokens": 9413520.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 6.210833978652954,
|
|
"epoch": 4.781700646444555,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00033034979318394676,
|
|
"loss": 5.6764,
|
|
"mean_token_accuracy": 0.17804316729307174,
|
|
"num_tokens": 9423945.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 6.257463073730468,
|
|
"epoch": 4.78667329686723,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00032997118147358554,
|
|
"loss": 5.7306,
|
|
"mean_token_accuracy": 0.18188440948724746,
|
|
"num_tokens": 9434139.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 6.2802948474884035,
|
|
"epoch": 4.791645947289906,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000329592404156656,
|
|
"loss": 5.7305,
|
|
"mean_token_accuracy": 0.18140054047107695,
|
|
"num_tokens": 9443807.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 6.29645676612854,
|
|
"epoch": 4.79661859771258,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032921346237426536,
|
|
"loss": 5.7373,
|
|
"mean_token_accuracy": 0.16916095316410065,
|
|
"num_tokens": 9452608.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 6.252938652038575,
|
|
"epoch": 4.8015912481352565,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032883435726801635,
|
|
"loss": 5.7407,
|
|
"mean_token_accuracy": 0.17163490504026413,
|
|
"num_tokens": 9462233.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 6.290170335769654,
|
|
"epoch": 4.806563898557932,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003284550899800036,
|
|
"loss": 5.7554,
|
|
"mean_token_accuracy": 0.17069380581378937,
|
|
"num_tokens": 9472926.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 6.2936845302581785,
|
|
"epoch": 4.811536548980607,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00032807566165281055,
|
|
"loss": 5.6715,
|
|
"mean_token_accuracy": 0.18531303703784943,
|
|
"num_tokens": 9481711.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 6.276135015487671,
|
|
"epoch": 4.816509199403282,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003276960734295054,
|
|
"loss": 5.6257,
|
|
"mean_token_accuracy": 0.17929538190364838,
|
|
"num_tokens": 9492655.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 6.289324188232422,
|
|
"epoch": 4.8214818498259575,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00032731632645363856,
|
|
"loss": 5.6296,
|
|
"mean_token_accuracy": 0.1859590455889702,
|
|
"num_tokens": 9502852.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 6.198389911651612,
|
|
"epoch": 4.826454500248633,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003269364218692384,
|
|
"loss": 5.5957,
|
|
"mean_token_accuracy": 0.18984483629465104,
|
|
"num_tokens": 9511781.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 6.252685022354126,
|
|
"epoch": 4.831427150671308,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003265563608208081,
|
|
"loss": 5.6974,
|
|
"mean_token_accuracy": 0.1846608579158783,
|
|
"num_tokens": 9521436.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 6.2685243606567385,
|
|
"epoch": 4.836399801093983,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003261761444533221,
|
|
"loss": 5.6421,
|
|
"mean_token_accuracy": 0.1865024134516716,
|
|
"num_tokens": 9531432.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 6.281853580474854,
|
|
"epoch": 4.8413724515166585,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003257957739122231,
|
|
"loss": 5.7979,
|
|
"mean_token_accuracy": 0.17684038132429122,
|
|
"num_tokens": 9540976.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 6.248445272445679,
|
|
"epoch": 4.846345101939334,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00032541525034341805,
|
|
"loss": 5.5605,
|
|
"mean_token_accuracy": 0.186386276781559,
|
|
"num_tokens": 9549566.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 6.1702240943908695,
|
|
"epoch": 4.851317752362009,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003250345748932749,
|
|
"loss": 5.6697,
|
|
"mean_token_accuracy": 0.1850678652524948,
|
|
"num_tokens": 9558937.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 6.178045701980591,
|
|
"epoch": 4.856290402784684,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00032465374870861913,
|
|
"loss": 5.66,
|
|
"mean_token_accuracy": 0.18690076172351838,
|
|
"num_tokens": 9568374.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 6.180189943313598,
|
|
"epoch": 4.8612630532073595,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00032427277293673053,
|
|
"loss": 5.5338,
|
|
"mean_token_accuracy": 0.18867705017328262,
|
|
"num_tokens": 9577835.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 6.259511947631836,
|
|
"epoch": 4.866235703630035,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003238916487253393,
|
|
"loss": 5.6925,
|
|
"mean_token_accuracy": 0.1759900137782097,
|
|
"num_tokens": 9586787.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 6.235188245773315,
|
|
"epoch": 4.87120835405271,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003235103772226229,
|
|
"loss": 5.7182,
|
|
"mean_token_accuracy": 0.18234290480613707,
|
|
"num_tokens": 9597669.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 6.265584564208984,
|
|
"epoch": 4.876181004475385,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003231289595772026,
|
|
"loss": 5.6141,
|
|
"mean_token_accuracy": 0.18462282419204712,
|
|
"num_tokens": 9607535.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 6.26797513961792,
|
|
"epoch": 4.8811536548980605,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003227473969381398,
|
|
"loss": 5.7038,
|
|
"mean_token_accuracy": 0.17398782670497895,
|
|
"num_tokens": 9616745.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 6.2442967891693115,
|
|
"epoch": 4.886126305320736,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032236569045493283,
|
|
"loss": 5.791,
|
|
"mean_token_accuracy": 0.17143307477235795,
|
|
"num_tokens": 9626445.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 6.265368366241455,
|
|
"epoch": 4.891098955743411,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00032198384127751344,
|
|
"loss": 5.6212,
|
|
"mean_token_accuracy": 0.1873790815472603,
|
|
"num_tokens": 9635083.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 6.2759510517120365,
|
|
"epoch": 4.896071606166086,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000321601850556243,
|
|
"loss": 5.7911,
|
|
"mean_token_accuracy": 0.17120405286550522,
|
|
"num_tokens": 9645410.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 6.298156690597534,
|
|
"epoch": 4.9010442565887615,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003212197194419095,
|
|
"loss": 5.6987,
|
|
"mean_token_accuracy": 0.17279478162527084,
|
|
"num_tokens": 9654826.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 6.27287335395813,
|
|
"epoch": 4.906016907011437,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003208374490857238,
|
|
"loss": 5.7198,
|
|
"mean_token_accuracy": 0.17037489712238313,
|
|
"num_tokens": 9664609.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 6.109054183959961,
|
|
"epoch": 4.910989557434112,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003204550406393164,
|
|
"loss": 5.5256,
|
|
"mean_token_accuracy": 0.19707108289003372,
|
|
"num_tokens": 9672963.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 6.261562967300415,
|
|
"epoch": 4.915962207856787,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003200724952547335,
|
|
"loss": 5.6072,
|
|
"mean_token_accuracy": 0.18578533232212066,
|
|
"num_tokens": 9681662.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 6.186133766174317,
|
|
"epoch": 4.9209348582794625,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003196898140844341,
|
|
"loss": 5.5328,
|
|
"mean_token_accuracy": 0.19103305041790009,
|
|
"num_tokens": 9693204.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 6.237615776062012,
|
|
"epoch": 4.925907508702139,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00031930699828128614,
|
|
"loss": 5.6115,
|
|
"mean_token_accuracy": 0.17995752096176149,
|
|
"num_tokens": 9702964.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 6.255939912796021,
|
|
"epoch": 4.930880159124813,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003189240489985633,
|
|
"loss": 5.6713,
|
|
"mean_token_accuracy": 0.18880916833877565,
|
|
"num_tokens": 9712407.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 6.29522557258606,
|
|
"epoch": 4.935852809547489,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003185409673899412,
|
|
"loss": 5.7273,
|
|
"mean_token_accuracy": 0.17456728219985962,
|
|
"num_tokens": 9722123.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 6.2634683609008786,
|
|
"epoch": 4.940825459970164,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00031815775460949426,
|
|
"loss": 5.6726,
|
|
"mean_token_accuracy": 0.18369710445404053,
|
|
"num_tokens": 9732641.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 6.3147131443023685,
|
|
"epoch": 4.94579811039284,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003177744118116919,
|
|
"loss": 5.7164,
|
|
"mean_token_accuracy": 0.1725757300853729,
|
|
"num_tokens": 9741498.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 6.32067494392395,
|
|
"epoch": 4.950770760815515,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003173909401513954,
|
|
"loss": 5.7264,
|
|
"mean_token_accuracy": 0.17549206763505937,
|
|
"num_tokens": 9750811.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 6.076323175430298,
|
|
"epoch": 4.95574341123819,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00031700734078385416,
|
|
"loss": 5.5491,
|
|
"mean_token_accuracy": 0.19996912330389022,
|
|
"num_tokens": 9761233.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 6.201333379745483,
|
|
"epoch": 4.960716061660865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00031662361486470243,
|
|
"loss": 5.6229,
|
|
"mean_token_accuracy": 0.18014758080244064,
|
|
"num_tokens": 9771258.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 6.196002197265625,
|
|
"epoch": 4.965688712083541,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0003162397635499555,
|
|
"loss": 5.6248,
|
|
"mean_token_accuracy": 0.18833557665348052,
|
|
"num_tokens": 9781501.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 6.282912731170654,
|
|
"epoch": 4.970661362506216,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031585578799600653,
|
|
"loss": 5.8765,
|
|
"mean_token_accuracy": 0.16619472354650497,
|
|
"num_tokens": 9792488.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.970661362506216,
|
|
"eval_entropy": 6.173630982500067,
|
|
"eval_loss": 6.462355613708496,
|
|
"eval_mean_token_accuracy": 0.14795834861249418,
|
|
"eval_num_tokens": 9792488.0,
|
|
"eval_runtime": 1.9298,
|
|
"eval_samples_per_second": 1795.528,
|
|
"eval_steps_per_second": 224.894,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 6.287742710113525,
|
|
"epoch": 4.975634012928891,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031547168935962304,
|
|
"loss": 5.6264,
|
|
"mean_token_accuracy": 0.18396837264299393,
|
|
"num_tokens": 9802711.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 6.3044774532318115,
|
|
"epoch": 4.980606663351566,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003150874687979433,
|
|
"loss": 5.6794,
|
|
"mean_token_accuracy": 0.1792687565088272,
|
|
"num_tokens": 9812668.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 6.2200157165527346,
|
|
"epoch": 4.985579313774242,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031470312746847284,
|
|
"loss": 5.6841,
|
|
"mean_token_accuracy": 0.18543685376644134,
|
|
"num_tokens": 9822466.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 6.2429756164550785,
|
|
"epoch": 4.990551964196917,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00031431866652908107,
|
|
"loss": 5.674,
|
|
"mean_token_accuracy": 0.18790722638368607,
|
|
"num_tokens": 9831529.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 6.243465375900269,
|
|
"epoch": 4.995524614619592,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003139340871379977,
|
|
"loss": 5.7481,
|
|
"mean_token_accuracy": 0.1841896265745163,
|
|
"num_tokens": 9842321.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 6.322022438049316,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 0.0003135493904538093,
|
|
"loss": 5.7545,
|
|
"mean_token_accuracy": 0.18238132529788548,
|
|
"num_tokens": 9850520.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 6.308475589752197,
|
|
"epoch": 5.004972650422675,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003131645776354559,
|
|
"loss": 5.327,
|
|
"mean_token_accuracy": 0.2130538895726204,
|
|
"num_tokens": 9859880.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 6.219911575317383,
|
|
"epoch": 5.0099453008453505,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003127796498422272,
|
|
"loss": 5.4872,
|
|
"mean_token_accuracy": 0.19472162872552873,
|
|
"num_tokens": 9869409.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 6.189472913742065,
|
|
"epoch": 5.014917951268026,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00031239460823375933,
|
|
"loss": 5.4801,
|
|
"mean_token_accuracy": 0.18600523620843887,
|
|
"num_tokens": 9879558.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 6.250964736938476,
|
|
"epoch": 5.019890601690701,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00031200945397003157,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.19435054808855057,
|
|
"num_tokens": 9890612.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 6.275197982788086,
|
|
"epoch": 5.024863252113376,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003116241882113623,
|
|
"loss": 5.4732,
|
|
"mean_token_accuracy": 0.1901419445872307,
|
|
"num_tokens": 9900928.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 6.294151878356933,
|
|
"epoch": 5.0298359025360515,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00031123881211840565,
|
|
"loss": 5.5079,
|
|
"mean_token_accuracy": 0.18897881358861923,
|
|
"num_tokens": 9910627.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 6.266399908065796,
|
|
"epoch": 5.034808552958727,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003108533268521485,
|
|
"loss": 5.4481,
|
|
"mean_token_accuracy": 0.19327375441789627,
|
|
"num_tokens": 9921250.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 6.168449735641479,
|
|
"epoch": 5.039781203381402,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003104677335739065,
|
|
"loss": 5.3712,
|
|
"mean_token_accuracy": 0.20772831588983537,
|
|
"num_tokens": 9931213.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 6.221916055679321,
|
|
"epoch": 5.044753853804077,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00031008203344532067,
|
|
"loss": 5.4445,
|
|
"mean_token_accuracy": 0.19876363277435302,
|
|
"num_tokens": 9941385.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 6.19802417755127,
|
|
"epoch": 5.0497265042267525,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000309696227628354,
|
|
"loss": 5.3574,
|
|
"mean_token_accuracy": 0.1983446002006531,
|
|
"num_tokens": 9951547.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 6.260295677185058,
|
|
"epoch": 5.054699154649428,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00030931031728528754,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.18359490633010864,
|
|
"num_tokens": 9960855.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 6.249000072479248,
|
|
"epoch": 5.059671805072103,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003089243035787178,
|
|
"loss": 5.5482,
|
|
"mean_token_accuracy": 0.18599290251731873,
|
|
"num_tokens": 9970116.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 6.062478828430176,
|
|
"epoch": 5.064644455494779,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00030853818767155225,
|
|
"loss": 5.2531,
|
|
"mean_token_accuracy": 0.21984208673238753,
|
|
"num_tokens": 9980488.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 6.190887784957885,
|
|
"epoch": 5.069617105917454,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003081519707270065,
|
|
"loss": 5.5374,
|
|
"mean_token_accuracy": 0.18874291479587554,
|
|
"num_tokens": 9990677.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 6.081090927124023,
|
|
"epoch": 5.07458975634013,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030776565390860017,
|
|
"loss": 5.2164,
|
|
"mean_token_accuracy": 0.21669256091117858,
|
|
"num_tokens": 10000218.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 6.216696834564209,
|
|
"epoch": 5.079562406762805,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003073792383801543,
|
|
"loss": 5.4332,
|
|
"mean_token_accuracy": 0.19844959527254105,
|
|
"num_tokens": 10008720.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 6.228510856628418,
|
|
"epoch": 5.08453505718548,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00030699272530578697,
|
|
"loss": 5.5009,
|
|
"mean_token_accuracy": 0.19840269833803176,
|
|
"num_tokens": 10018648.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 6.189662599563599,
|
|
"epoch": 5.089507707608155,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00030660611584991013,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.1995312139391899,
|
|
"num_tokens": 10028639.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 6.211550188064575,
|
|
"epoch": 5.094480358030831,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00030621941117722615,
|
|
"loss": 5.4516,
|
|
"mean_token_accuracy": 0.19814881533384324,
|
|
"num_tokens": 10038136.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 6.169123744964599,
|
|
"epoch": 5.099453008453506,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003058326124527243,
|
|
"loss": 5.4242,
|
|
"mean_token_accuracy": 0.198819100856781,
|
|
"num_tokens": 10048503.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 6.152795696258545,
|
|
"epoch": 5.104425658876181,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003054457208416771,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.1983527958393097,
|
|
"num_tokens": 10059088.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 6.2142784118652346,
|
|
"epoch": 5.109398309298856,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030505873750963705,
|
|
"loss": 5.3936,
|
|
"mean_token_accuracy": 0.1978098839521408,
|
|
"num_tokens": 10069655.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 6.269792461395264,
|
|
"epoch": 5.114370959721532,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003046716636224327,
|
|
"loss": 5.4109,
|
|
"mean_token_accuracy": 0.19529213160276412,
|
|
"num_tokens": 10079748.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 6.22677788734436,
|
|
"epoch": 5.119343610144207,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003042845003461657,
|
|
"loss": 5.4145,
|
|
"mean_token_accuracy": 0.1957701787352562,
|
|
"num_tokens": 10088877.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 6.22240138053894,
|
|
"epoch": 5.124316260566882,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00030389724884720664,
|
|
"loss": 5.4951,
|
|
"mean_token_accuracy": 0.19334201365709305,
|
|
"num_tokens": 10098699.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 5.941190147399903,
|
|
"epoch": 5.129288910989557,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00030350991029219236,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.23784599155187608,
|
|
"num_tokens": 10110084.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 6.174799966812134,
|
|
"epoch": 5.134261561412233,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030312248584802166,
|
|
"loss": 5.4782,
|
|
"mean_token_accuracy": 0.1938440278172493,
|
|
"num_tokens": 10119553.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 6.256517124176026,
|
|
"epoch": 5.139234211834908,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003027349766818521,
|
|
"loss": 5.424,
|
|
"mean_token_accuracy": 0.19743411242961884,
|
|
"num_tokens": 10128332.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 6.2417723655700685,
|
|
"epoch": 5.144206862257583,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003023473839610965,
|
|
"loss": 5.5189,
|
|
"mean_token_accuracy": 0.18771152049303055,
|
|
"num_tokens": 10138109.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 6.270859670639038,
|
|
"epoch": 5.149179512680258,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003019597088534195,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.18481241315603256,
|
|
"num_tokens": 10148103.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 6.343954610824585,
|
|
"epoch": 5.154152163102934,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003015719525267338,
|
|
"loss": 5.5851,
|
|
"mean_token_accuracy": 0.18354683518409728,
|
|
"num_tokens": 10158189.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 6.232406044006348,
|
|
"epoch": 5.159124813525609,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003011841161491969,
|
|
"loss": 5.5238,
|
|
"mean_token_accuracy": 0.19161584824323655,
|
|
"num_tokens": 10169048.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 6.174866628646851,
|
|
"epoch": 5.164097463948284,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00030079620088920727,
|
|
"loss": 5.4448,
|
|
"mean_token_accuracy": 0.2006381019949913,
|
|
"num_tokens": 10178067.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 6.254225397109986,
|
|
"epoch": 5.169070114370959,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003004082079154013,
|
|
"loss": 5.5368,
|
|
"mean_token_accuracy": 0.18376923352479935,
|
|
"num_tokens": 10186952.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 6.21033239364624,
|
|
"epoch": 5.174042764793635,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030002013839664925,
|
|
"loss": 5.5427,
|
|
"mean_token_accuracy": 0.18650542348623275,
|
|
"num_tokens": 10196699.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 6.181064796447754,
|
|
"epoch": 5.17901541521631,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002996319935020522,
|
|
"loss": 5.3882,
|
|
"mean_token_accuracy": 0.19638360887765885,
|
|
"num_tokens": 10206198.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 6.210749292373658,
|
|
"epoch": 5.183988065638985,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002992437744009381,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.1885737195611,
|
|
"num_tokens": 10215698.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 6.342647552490234,
|
|
"epoch": 5.188960716061661,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00029885548226285855,
|
|
"loss": 5.547,
|
|
"mean_token_accuracy": 0.1901486799120903,
|
|
"num_tokens": 10225732.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 6.233848619461059,
|
|
"epoch": 5.193933366484337,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002984671182575851,
|
|
"loss": 5.4809,
|
|
"mean_token_accuracy": 0.19677677750587463,
|
|
"num_tokens": 10235691.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 6.187122392654419,
|
|
"epoch": 5.198906016907012,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002980786835551061,
|
|
"loss": 5.4875,
|
|
"mean_token_accuracy": 0.20081337094306945,
|
|
"num_tokens": 10246066.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 6.274002647399902,
|
|
"epoch": 5.203878667329687,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002976901793256225,
|
|
"loss": 5.4319,
|
|
"mean_token_accuracy": 0.1975369021296501,
|
|
"num_tokens": 10254968.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 6.2490629196167,
|
|
"epoch": 5.208851317752362,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000297301606739545,
|
|
"loss": 5.4594,
|
|
"mean_token_accuracy": 0.1957896262407303,
|
|
"num_tokens": 10263800.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 6.194499158859253,
|
|
"epoch": 5.213823968175038,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00029691296696748994,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.191917884349823,
|
|
"num_tokens": 10274044.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 6.203491163253784,
|
|
"epoch": 5.218796618597713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00029652426118027635,
|
|
"loss": 5.5485,
|
|
"mean_token_accuracy": 0.18982027620077133,
|
|
"num_tokens": 10284429.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 6.2401045799255375,
|
|
"epoch": 5.223769269020388,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00029613549054892206,
|
|
"loss": 5.5202,
|
|
"mean_token_accuracy": 0.1909793496131897,
|
|
"num_tokens": 10293968.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 6.051461029052734,
|
|
"epoch": 5.228741919443063,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002957466562446401,
|
|
"loss": 5.2843,
|
|
"mean_token_accuracy": 0.2161101832985878,
|
|
"num_tokens": 10303497.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 6.232101106643677,
|
|
"epoch": 5.233714569865739,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002953577594388355,
|
|
"loss": 5.5473,
|
|
"mean_token_accuracy": 0.1833020865917206,
|
|
"num_tokens": 10312766.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 6.178042078018189,
|
|
"epoch": 5.238687220288414,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00029496880130310145,
|
|
"loss": 5.4039,
|
|
"mean_token_accuracy": 0.19440586268901824,
|
|
"num_tokens": 10323604.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 6.238149976730346,
|
|
"epoch": 5.243659870711089,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002945797830092161,
|
|
"loss": 5.4737,
|
|
"mean_token_accuracy": 0.19265810251235962,
|
|
"num_tokens": 10333620.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 6.195771884918213,
|
|
"epoch": 5.248632521133764,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002941907057291386,
|
|
"loss": 5.4676,
|
|
"mean_token_accuracy": 0.18848675787448882,
|
|
"num_tokens": 10344078.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 6.236386156082153,
|
|
"epoch": 5.25360517155644,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00029380157063500594,
|
|
"loss": 5.4419,
|
|
"mean_token_accuracy": 0.19538459777832032,
|
|
"num_tokens": 10353254.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 6.173181056976318,
|
|
"epoch": 5.258577821979115,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002934123788991291,
|
|
"loss": 5.4768,
|
|
"mean_token_accuracy": 0.19660588800907136,
|
|
"num_tokens": 10362310.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 6.2702678680419925,
|
|
"epoch": 5.26355047240179,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00029302313169399,
|
|
"loss": 5.6565,
|
|
"mean_token_accuracy": 0.17811751514673232,
|
|
"num_tokens": 10373252.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 6.1967692375183105,
|
|
"epoch": 5.268523122824465,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002926338301922376,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.19628230184316636,
|
|
"num_tokens": 10381682.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 6.266997766494751,
|
|
"epoch": 5.273495773247141,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00029224447556668417,
|
|
"loss": 5.6209,
|
|
"mean_token_accuracy": 0.17701685577630996,
|
|
"num_tokens": 10391567.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 6.273753833770752,
|
|
"epoch": 5.278468423669816,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002918550689903022,
|
|
"loss": 5.5529,
|
|
"mean_token_accuracy": 0.1899106502532959,
|
|
"num_tokens": 10401820.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 6.253249168395996,
|
|
"epoch": 5.283441074092491,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00029146561163622076,
|
|
"loss": 5.4952,
|
|
"mean_token_accuracy": 0.19662512093782425,
|
|
"num_tokens": 10413378.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 6.193878126144409,
|
|
"epoch": 5.288413724515166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002910761046777218,
|
|
"loss": 5.4247,
|
|
"mean_token_accuracy": 0.20357256829738618,
|
|
"num_tokens": 10423447.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 6.144731712341309,
|
|
"epoch": 5.293386374937842,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00029068654928823666,
|
|
"loss": 5.3908,
|
|
"mean_token_accuracy": 0.1991468384861946,
|
|
"num_tokens": 10433785.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 6.173011684417725,
|
|
"epoch": 5.298359025360517,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00029029694664134284,
|
|
"loss": 5.52,
|
|
"mean_token_accuracy": 0.17952480912208557,
|
|
"num_tokens": 10444748.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 6.29718337059021,
|
|
"epoch": 5.303331675783192,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002899072979107597,
|
|
"loss": 5.5702,
|
|
"mean_token_accuracy": 0.18071280270814896,
|
|
"num_tokens": 10455778.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 6.233965444564819,
|
|
"epoch": 5.308304326205867,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002895176042703461,
|
|
"loss": 5.5472,
|
|
"mean_token_accuracy": 0.18933781683444978,
|
|
"num_tokens": 10465915.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 6.143447351455689,
|
|
"epoch": 5.3132769766285435,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00028912786689409554,
|
|
"loss": 5.354,
|
|
"mean_token_accuracy": 0.20719279646873473,
|
|
"num_tokens": 10475838.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 6.005608177185058,
|
|
"epoch": 5.318249627051218,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002887380869561338,
|
|
"loss": 5.2269,
|
|
"mean_token_accuracy": 0.21692942976951599,
|
|
"num_tokens": 10485719.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 6.241970825195312,
|
|
"epoch": 5.323222277473894,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002883482656307145,
|
|
"loss": 5.444,
|
|
"mean_token_accuracy": 0.19545145332813263,
|
|
"num_tokens": 10494695.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 6.271560859680176,
|
|
"epoch": 5.328194927896569,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002879584040922162,
|
|
"loss": 5.5606,
|
|
"mean_token_accuracy": 0.18719894737005233,
|
|
"num_tokens": 10505356.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 6.159087514877319,
|
|
"epoch": 5.3331675783192445,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002875685035151386,
|
|
"loss": 5.3378,
|
|
"mean_token_accuracy": 0.19615171551704408,
|
|
"num_tokens": 10515118.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 6.129981565475464,
|
|
"epoch": 5.33814022874192,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00028717856507409886,
|
|
"loss": 5.4536,
|
|
"mean_token_accuracy": 0.19922214150428771,
|
|
"num_tokens": 10525713.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 6.221442651748657,
|
|
"epoch": 5.343112879164595,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002867885899438281,
|
|
"loss": 5.4301,
|
|
"mean_token_accuracy": 0.20060539543628692,
|
|
"num_tokens": 10534363.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 6.088037061691284,
|
|
"epoch": 5.34808552958727,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002863985792991683,
|
|
"loss": 5.3082,
|
|
"mean_token_accuracy": 0.2042089581489563,
|
|
"num_tokens": 10544302.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 6.22057695388794,
|
|
"epoch": 5.3530581800099455,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002860085343150683,
|
|
"loss": 5.4784,
|
|
"mean_token_accuracy": 0.19250127524137498,
|
|
"num_tokens": 10553742.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 6.247465705871582,
|
|
"epoch": 5.358030830432621,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028561845616658033,
|
|
"loss": 5.5807,
|
|
"mean_token_accuracy": 0.18292634636163713,
|
|
"num_tokens": 10564061.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 6.201936531066894,
|
|
"epoch": 5.363003480855296,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00028522834602885656,
|
|
"loss": 5.5641,
|
|
"mean_token_accuracy": 0.18645524084568024,
|
|
"num_tokens": 10573439.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 6.267965364456177,
|
|
"epoch": 5.367976131277971,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002848382050771455,
|
|
"loss": 5.5237,
|
|
"mean_token_accuracy": 0.19233374744653703,
|
|
"num_tokens": 10583694.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 6.246150255203247,
|
|
"epoch": 5.3729487817006465,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002844480344867886,
|
|
"loss": 5.5767,
|
|
"mean_token_accuracy": 0.1866806209087372,
|
|
"num_tokens": 10593432.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 5.988351774215698,
|
|
"epoch": 5.377921432123322,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002840578354332165,
|
|
"loss": 5.2513,
|
|
"mean_token_accuracy": 0.21707857698202132,
|
|
"num_tokens": 10603050.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 6.172908115386963,
|
|
"epoch": 5.382894082545997,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0002836676090919456,
|
|
"loss": 5.4803,
|
|
"mean_token_accuracy": 0.19822771847248077,
|
|
"num_tokens": 10613842.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 6.225389623641968,
|
|
"epoch": 5.387866732968672,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002832773566385746,
|
|
"loss": 5.4798,
|
|
"mean_token_accuracy": 0.19098188132047653,
|
|
"num_tokens": 10624220.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 6.139774990081787,
|
|
"epoch": 5.3928393833913475,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00028288707924878056,
|
|
"loss": 5.3083,
|
|
"mean_token_accuracy": 0.20708127468824386,
|
|
"num_tokens": 10633797.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 6.1448524475097654,
|
|
"epoch": 5.397812033814023,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002824967780983162,
|
|
"loss": 5.4108,
|
|
"mean_token_accuracy": 0.2033515229821205,
|
|
"num_tokens": 10642718.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 6.277849578857422,
|
|
"epoch": 5.402784684236698,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00028210645436300524,
|
|
"loss": 5.577,
|
|
"mean_token_accuracy": 0.17948753237724305,
|
|
"num_tokens": 10653733.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 6.196979904174805,
|
|
"epoch": 5.407757334659373,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028171610921873995,
|
|
"loss": 5.4835,
|
|
"mean_token_accuracy": 0.19333163797855377,
|
|
"num_tokens": 10663238.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 6.20132884979248,
|
|
"epoch": 5.4127299850820485,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0002813257438414766,
|
|
"loss": 5.5253,
|
|
"mean_token_accuracy": 0.18715387284755708,
|
|
"num_tokens": 10674037.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 6.243438100814819,
|
|
"epoch": 5.417702635504724,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000280935359407233,
|
|
"loss": 5.5427,
|
|
"mean_token_accuracy": 0.1821499839425087,
|
|
"num_tokens": 10684394.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 6.175432348251343,
|
|
"epoch": 5.422675285927399,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002805449570920838,
|
|
"loss": 5.5518,
|
|
"mean_token_accuracy": 0.18757390677928926,
|
|
"num_tokens": 10693396.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 6.169561672210693,
|
|
"epoch": 5.427647936350074,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00028015453807215786,
|
|
"loss": 5.4749,
|
|
"mean_token_accuracy": 0.19486436992883682,
|
|
"num_tokens": 10703643.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 6.217954826354981,
|
|
"epoch": 5.4326205867727495,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00027976410352363416,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.1933427169919014,
|
|
"num_tokens": 10712883.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 6.175315952301025,
|
|
"epoch": 5.437593237195425,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027937365462273864,
|
|
"loss": 5.4095,
|
|
"mean_token_accuracy": 0.19473367929458618,
|
|
"num_tokens": 10721926.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 6.168694067001343,
|
|
"epoch": 5.4425658876181,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00027898319254574043,
|
|
"loss": 5.4558,
|
|
"mean_token_accuracy": 0.19669611155986785,
|
|
"num_tokens": 10731772.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 6.248317098617553,
|
|
"epoch": 5.447538538040776,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002785927184689483,
|
|
"loss": 5.628,
|
|
"mean_token_accuracy": 0.18476845920085908,
|
|
"num_tokens": 10742815.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 6.218686914443969,
|
|
"epoch": 5.452511188463451,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002782022335687073,
|
|
"loss": 5.4455,
|
|
"mean_token_accuracy": 0.2007911428809166,
|
|
"num_tokens": 10753958.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 6.260257863998413,
|
|
"epoch": 5.457483838886127,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00027781173902139485,
|
|
"loss": 5.5471,
|
|
"mean_token_accuracy": 0.1848934218287468,
|
|
"num_tokens": 10763808.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 6.170033121109009,
|
|
"epoch": 5.462456489308802,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002774212360034176,
|
|
"loss": 5.5163,
|
|
"mean_token_accuracy": 0.18955819457769393,
|
|
"num_tokens": 10773688.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 6.234539127349853,
|
|
"epoch": 5.467429139731477,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002770307256912078,
|
|
"loss": 5.5562,
|
|
"mean_token_accuracy": 0.18901694864034652,
|
|
"num_tokens": 10783902.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 6.0569891929626465,
|
|
"epoch": 5.472401790154152,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00027664020926121944,
|
|
"loss": 5.2184,
|
|
"mean_token_accuracy": 0.22152826189994812,
|
|
"num_tokens": 10794138.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 6.195235395431519,
|
|
"epoch": 5.477374440576828,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002762496878899251,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.192214997112751,
|
|
"num_tokens": 10804599.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 6.209974670410157,
|
|
"epoch": 5.482347090999503,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027585916275381226,
|
|
"loss": 5.4793,
|
|
"mean_token_accuracy": 0.19507736563682557,
|
|
"num_tokens": 10813632.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 6.177565288543701,
|
|
"epoch": 5.487319741422178,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002754686350293797,
|
|
"loss": 5.5474,
|
|
"mean_token_accuracy": 0.1862577185034752,
|
|
"num_tokens": 10824287.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 6.223867750167846,
|
|
"epoch": 5.492292391844853,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002750781058931339,
|
|
"loss": 5.5709,
|
|
"mean_token_accuracy": 0.18647487014532088,
|
|
"num_tokens": 10834079.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 6.2317811965942385,
|
|
"epoch": 5.497265042267529,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002746875765215857,
|
|
"loss": 5.5513,
|
|
"mean_token_accuracy": 0.184711816906929,
|
|
"num_tokens": 10843029.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 6.215256452560425,
|
|
"epoch": 5.502237692690204,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027429704809124665,
|
|
"loss": 5.4864,
|
|
"mean_token_accuracy": 0.19604783803224562,
|
|
"num_tokens": 10852749.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 6.095922183990479,
|
|
"epoch": 5.507210343112879,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002739065217786255,
|
|
"loss": 5.3893,
|
|
"mean_token_accuracy": 0.20161352604627608,
|
|
"num_tokens": 10862911.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 6.2243987083435055,
|
|
"epoch": 5.512182993535554,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002735159987602246,
|
|
"loss": 5.4601,
|
|
"mean_token_accuracy": 0.2010153651237488,
|
|
"num_tokens": 10872389.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 6.163063001632691,
|
|
"epoch": 5.51715564395823,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002731254802125362,
|
|
"loss": 5.4204,
|
|
"mean_token_accuracy": 0.20291622132062911,
|
|
"num_tokens": 10881280.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 6.092837762832642,
|
|
"epoch": 5.522128294380905,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00027273496731203945,
|
|
"loss": 5.5046,
|
|
"mean_token_accuracy": 0.196041339635849,
|
|
"num_tokens": 10891154.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 6.230661821365357,
|
|
"epoch": 5.52710094480358,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000272344461235196,
|
|
"loss": 5.4269,
|
|
"mean_token_accuracy": 0.19658998548984527,
|
|
"num_tokens": 10900029.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 6.005603265762329,
|
|
"epoch": 5.532073595226255,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00027195396315844733,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.22333486378192902,
|
|
"num_tokens": 10909616.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 5.998079919815064,
|
|
"epoch": 5.537046245648931,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002715634742582107,
|
|
"loss": 5.2274,
|
|
"mean_token_accuracy": 0.2255914106965065,
|
|
"num_tokens": 10920754.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 6.181475877761841,
|
|
"epoch": 5.542018896071606,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027117299571087576,
|
|
"loss": 5.4614,
|
|
"mean_token_accuracy": 0.19166520982980728,
|
|
"num_tokens": 10929918.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 6.213248586654663,
|
|
"epoch": 5.546991546494281,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002707825286928008,
|
|
"loss": 5.4873,
|
|
"mean_token_accuracy": 0.19400571882724763,
|
|
"num_tokens": 10938932.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 6.177088642120362,
|
|
"epoch": 5.551964196916956,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002703920743803097,
|
|
"loss": 5.3785,
|
|
"mean_token_accuracy": 0.1945304200053215,
|
|
"num_tokens": 10948896.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 6.208967018127441,
|
|
"epoch": 5.556936847339632,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027000163394968763,
|
|
"loss": 5.544,
|
|
"mean_token_accuracy": 0.18625040352344513,
|
|
"num_tokens": 10958245.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 6.214950895309448,
|
|
"epoch": 5.561909497762308,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026961120857717843,
|
|
"loss": 5.4907,
|
|
"mean_token_accuracy": 0.20012946873903276,
|
|
"num_tokens": 10967917.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 6.17227520942688,
|
|
"epoch": 5.566882148184982,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00026922079943898016,
|
|
"loss": 5.4645,
|
|
"mean_token_accuracy": 0.19646290838718414,
|
|
"num_tokens": 10977062.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 6.224547433853149,
|
|
"epoch": 5.571854798607658,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002688304077112421,
|
|
"loss": 5.5831,
|
|
"mean_token_accuracy": 0.18622443377971648,
|
|
"num_tokens": 10987009.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 6.193683910369873,
|
|
"epoch": 5.576827449030333,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002684400345700613,
|
|
"loss": 5.3319,
|
|
"mean_token_accuracy": 0.20565827786922455,
|
|
"num_tokens": 10996077.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 6.203093481063843,
|
|
"epoch": 5.581800099453009,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002680496811914786,
|
|
"loss": 5.4422,
|
|
"mean_token_accuracy": 0.19437175542116164,
|
|
"num_tokens": 11005240.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 6.179389190673828,
|
|
"epoch": 5.586772749875684,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00026765934875147513,
|
|
"loss": 5.5522,
|
|
"mean_token_accuracy": 0.1822713002562523,
|
|
"num_tokens": 11014373.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 6.012269687652588,
|
|
"epoch": 5.591745400298359,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002672690384259693,
|
|
"loss": 5.2655,
|
|
"mean_token_accuracy": 0.22190622687339784,
|
|
"num_tokens": 11024676.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 6.137647390365601,
|
|
"epoch": 5.596718050721035,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026687875139081287,
|
|
"loss": 5.4417,
|
|
"mean_token_accuracy": 0.19626105278730394,
|
|
"num_tokens": 11033722.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 6.167677688598633,
|
|
"epoch": 5.60169070114371,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00026648848882178703,
|
|
"loss": 5.4578,
|
|
"mean_token_accuracy": 0.19803825616836548,
|
|
"num_tokens": 11043405.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 6.177847480773925,
|
|
"epoch": 5.606663351566385,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002660982518945998,
|
|
"loss": 5.4631,
|
|
"mean_token_accuracy": 0.19092631936073304,
|
|
"num_tokens": 11053577.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 6.148113346099853,
|
|
"epoch": 5.61163600198906,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00026570804178488146,
|
|
"loss": 5.4604,
|
|
"mean_token_accuracy": 0.19672759771347045,
|
|
"num_tokens": 11062890.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 6.0788078784942625,
|
|
"epoch": 5.616608652411736,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002653178596681819,
|
|
"loss": 5.3318,
|
|
"mean_token_accuracy": 0.20952214002609254,
|
|
"num_tokens": 11073445.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 6.1549766063690186,
|
|
"epoch": 5.621581302834411,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002649277067199665,
|
|
"loss": 5.5019,
|
|
"mean_token_accuracy": 0.18847260475158692,
|
|
"num_tokens": 11083159.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 6.18242712020874,
|
|
"epoch": 5.626553953257086,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002645375841156127,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.18415375798940659,
|
|
"num_tokens": 11093417.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 6.224481010437012,
|
|
"epoch": 5.631526603679761,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002641474930304067,
|
|
"loss": 5.4804,
|
|
"mean_token_accuracy": 0.1974414438009262,
|
|
"num_tokens": 11102775.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 6.28015661239624,
|
|
"epoch": 5.636499254102437,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002637574346395396,
|
|
"loss": 5.5624,
|
|
"mean_token_accuracy": 0.18258007168769835,
|
|
"num_tokens": 11111952.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 6.16537036895752,
|
|
"epoch": 5.641471904525112,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000263367410118104,
|
|
"loss": 5.3609,
|
|
"mean_token_accuracy": 0.1973536878824234,
|
|
"num_tokens": 11120391.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 6.185120010375977,
|
|
"epoch": 5.646444554947787,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002629774206410906,
|
|
"loss": 5.5165,
|
|
"mean_token_accuracy": 0.1889160916209221,
|
|
"num_tokens": 11129905.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 6.163305807113647,
|
|
"epoch": 5.651417205370462,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002625874673833844,
|
|
"loss": 5.4984,
|
|
"mean_token_accuracy": 0.19599420875310897,
|
|
"num_tokens": 11139068.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 6.188838291168213,
|
|
"epoch": 5.656389855793138,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002621975515197613,
|
|
"loss": 5.4049,
|
|
"mean_token_accuracy": 0.1997423827648163,
|
|
"num_tokens": 11147718.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 6.231201696395874,
|
|
"epoch": 5.661362506215813,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00026180767422488466,
|
|
"loss": 5.5846,
|
|
"mean_token_accuracy": 0.18448083996772766,
|
|
"num_tokens": 11157621.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 6.271498441696167,
|
|
"epoch": 5.666335156638488,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00026141783667330156,
|
|
"loss": 5.4948,
|
|
"mean_token_accuracy": 0.1891541674733162,
|
|
"num_tokens": 11168061.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 6.254337453842163,
|
|
"epoch": 5.671307807061163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002610280400394393,
|
|
"loss": 5.5661,
|
|
"mean_token_accuracy": 0.1819332018494606,
|
|
"num_tokens": 11177240.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 6.206892728805542,
|
|
"epoch": 5.676280457483839,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00026063828549760195,
|
|
"loss": 5.5806,
|
|
"mean_token_accuracy": 0.19183616936206818,
|
|
"num_tokens": 11187280.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 6.1544352054595945,
|
|
"epoch": 5.681253107906514,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002602485742219667,
|
|
"loss": 5.4226,
|
|
"mean_token_accuracy": 0.19681801348924638,
|
|
"num_tokens": 11197207.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 6.19719090461731,
|
|
"epoch": 5.686225758329189,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00025985890738658077,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.18535284250974654,
|
|
"num_tokens": 11206373.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 6.173356771469116,
|
|
"epoch": 5.691198408751864,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000259469286165357,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.19136733412742615,
|
|
"num_tokens": 11216799.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 6.1993265628814695,
|
|
"epoch": 5.6961710591745405,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00025907971173207107,
|
|
"loss": 5.5414,
|
|
"mean_token_accuracy": 0.17875082939863204,
|
|
"num_tokens": 11226968.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 6.2059674739837645,
|
|
"epoch": 5.701143709597215,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002586901852603576,
|
|
"loss": 5.5399,
|
|
"mean_token_accuracy": 0.19038939028978347,
|
|
"num_tokens": 11235741.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 6.2154436111450195,
|
|
"epoch": 5.706116360019891,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0002583007079237069,
|
|
"loss": 5.5048,
|
|
"mean_token_accuracy": 0.19233092069625854,
|
|
"num_tokens": 11245804.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 6.17623200416565,
|
|
"epoch": 5.711089010442566,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00025791128089546107,
|
|
"loss": 5.5343,
|
|
"mean_token_accuracy": 0.18729630410671233,
|
|
"num_tokens": 11256304.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 6.2297107696533205,
|
|
"epoch": 5.7160616608652415,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00025752190534881065,
|
|
"loss": 5.551,
|
|
"mean_token_accuracy": 0.18987454771995543,
|
|
"num_tokens": 11265800.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 6.289957904815674,
|
|
"epoch": 5.721034311287917,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00025713258245679127,
|
|
"loss": 5.5525,
|
|
"mean_token_accuracy": 0.1858096554875374,
|
|
"num_tokens": 11275983.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 6.279061508178711,
|
|
"epoch": 5.726006961710592,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00025674331339227966,
|
|
"loss": 5.4888,
|
|
"mean_token_accuracy": 0.19437556117773055,
|
|
"num_tokens": 11286375.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 6.22482271194458,
|
|
"epoch": 5.730979612133267,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00025635409932799074,
|
|
"loss": 5.5537,
|
|
"mean_token_accuracy": 0.19246532917022705,
|
|
"num_tokens": 11297359.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 6.224590396881103,
|
|
"epoch": 5.7359522625559425,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002559649414364735,
|
|
"loss": 5.5881,
|
|
"mean_token_accuracy": 0.18374993652105331,
|
|
"num_tokens": 11307235.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 6.284792423248291,
|
|
"epoch": 5.740924912978618,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002555758408901076,
|
|
"loss": 5.5486,
|
|
"mean_token_accuracy": 0.180190809071064,
|
|
"num_tokens": 11316618.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 6.293300819396973,
|
|
"epoch": 5.745897563401293,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00025518679886110023,
|
|
"loss": 5.6008,
|
|
"mean_token_accuracy": 0.18046425729990007,
|
|
"num_tokens": 11327178.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 6.047530317306519,
|
|
"epoch": 5.750870213823968,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0002547978165214822,
|
|
"loss": 5.2909,
|
|
"mean_token_accuracy": 0.2175559550523758,
|
|
"num_tokens": 11337788.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 6.177235460281372,
|
|
"epoch": 5.7558428642466435,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025440889504310416,
|
|
"loss": 5.4305,
|
|
"mean_token_accuracy": 0.2050589844584465,
|
|
"num_tokens": 11346958.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 6.215101432800293,
|
|
"epoch": 5.760815514669319,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00025402003559763393,
|
|
"loss": 5.4781,
|
|
"mean_token_accuracy": 0.19783546775579453,
|
|
"num_tokens": 11355817.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 6.2255054950714115,
|
|
"epoch": 5.765788165091994,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00025363123935655214,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.192696650326252,
|
|
"num_tokens": 11366557.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 6.1506195068359375,
|
|
"epoch": 5.770760815514669,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002532425074911491,
|
|
"loss": 5.5312,
|
|
"mean_token_accuracy": 0.18546296358108522,
|
|
"num_tokens": 11376931.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 6.211869764328003,
|
|
"epoch": 5.7757334659373445,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002528538411725212,
|
|
"loss": 5.5591,
|
|
"mean_token_accuracy": 0.18311526328325273,
|
|
"num_tokens": 11387430.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 6.192480087280273,
|
|
"epoch": 5.78070611636002,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002524652415715672,
|
|
"loss": 5.4347,
|
|
"mean_token_accuracy": 0.19979879707098008,
|
|
"num_tokens": 11396957.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 6.235910654067993,
|
|
"epoch": 5.785678766782695,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002520767098589852,
|
|
"loss": 5.4452,
|
|
"mean_token_accuracy": 0.1945034682750702,
|
|
"num_tokens": 11405313.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 6.204665994644165,
|
|
"epoch": 5.79065141720537,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002516882472052683,
|
|
"loss": 5.5454,
|
|
"mean_token_accuracy": 0.18731661289930343,
|
|
"num_tokens": 11414437.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 6.220718002319336,
|
|
"epoch": 5.7956240676280455,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002512998547807022,
|
|
"loss": 5.647,
|
|
"mean_token_accuracy": 0.18317157924175262,
|
|
"num_tokens": 11425498.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 6.313884305953979,
|
|
"epoch": 5.800596718050721,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00025091153375536026,
|
|
"loss": 5.567,
|
|
"mean_token_accuracy": 0.1833800792694092,
|
|
"num_tokens": 11435333.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 6.070698165893555,
|
|
"epoch": 5.805569368473396,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025052328529910135,
|
|
"loss": 5.2973,
|
|
"mean_token_accuracy": 0.21049655973911285,
|
|
"num_tokens": 11445366.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 6.142288875579834,
|
|
"epoch": 5.810542018896071,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002501351105815655,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.2036324605345726,
|
|
"num_tokens": 11455229.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 6.20145616531372,
|
|
"epoch": 5.8155146693187465,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00024974701077217057,
|
|
"loss": 5.5308,
|
|
"mean_token_accuracy": 0.18817630559206008,
|
|
"num_tokens": 11464789.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 6.187739419937134,
|
|
"epoch": 5.820487319741423,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00024935898704010875,
|
|
"loss": 5.4795,
|
|
"mean_token_accuracy": 0.19468995183706284,
|
|
"num_tokens": 11473694.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 6.253514528274536,
|
|
"epoch": 5.825459970164097,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024897104055434304,
|
|
"loss": 5.5379,
|
|
"mean_token_accuracy": 0.18862048387527466,
|
|
"num_tokens": 11482701.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 6.223541641235352,
|
|
"epoch": 5.830432620586773,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00024858317248360387,
|
|
"loss": 5.5865,
|
|
"mean_token_accuracy": 0.18316777795553207,
|
|
"num_tokens": 11493917.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 6.171255350112915,
|
|
"epoch": 5.835405271009448,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002481953839963851,
|
|
"loss": 5.3805,
|
|
"mean_token_accuracy": 0.201360222697258,
|
|
"num_tokens": 11504001.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 6.197660112380982,
|
|
"epoch": 5.840377921432124,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002478076762609413,
|
|
"loss": 5.5529,
|
|
"mean_token_accuracy": 0.18849510550498963,
|
|
"num_tokens": 11513010.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 6.2395592212677,
|
|
"epoch": 5.845350571854799,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002474200504452834,
|
|
"loss": 5.4797,
|
|
"mean_token_accuracy": 0.19790483117103577,
|
|
"num_tokens": 11522190.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 6.243426561355591,
|
|
"epoch": 5.850323222277474,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024703250771717576,
|
|
"loss": 5.5519,
|
|
"mean_token_accuracy": 0.1910725563764572,
|
|
"num_tokens": 11531180.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 6.243704175949096,
|
|
"epoch": 5.855295872700149,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00024664504924413224,
|
|
"loss": 5.5387,
|
|
"mean_token_accuracy": 0.186438749730587,
|
|
"num_tokens": 11540829.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 6.211485528945923,
|
|
"epoch": 5.860268523122825,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002462576761934129,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.18230901062488555,
|
|
"num_tokens": 11549118.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 6.1626921653747555,
|
|
"epoch": 5.8652411735455,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002458703897320208,
|
|
"loss": 5.49,
|
|
"mean_token_accuracy": 0.19440292418003083,
|
|
"num_tokens": 11557942.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 6.225880098342896,
|
|
"epoch": 5.870213823968175,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024548319102669745,
|
|
"loss": 5.4488,
|
|
"mean_token_accuracy": 0.19550804942846298,
|
|
"num_tokens": 11567273.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 6.223353672027588,
|
|
"epoch": 5.87518647439085,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00024509608124392065,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.19354547560214996,
|
|
"num_tokens": 11577383.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 6.1919537544250485,
|
|
"epoch": 5.880159124813526,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002447090615499,
|
|
"loss": 5.5447,
|
|
"mean_token_accuracy": 0.1909714788198471,
|
|
"num_tokens": 11586899.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 6.194301795959473,
|
|
"epoch": 5.885131775236201,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00024432213311057383,
|
|
"loss": 5.5271,
|
|
"mean_token_accuracy": 0.1862612247467041,
|
|
"num_tokens": 11596884.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 6.208656597137451,
|
|
"epoch": 5.890104425658876,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00024393529709160544,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.19804628044366837,
|
|
"num_tokens": 11606603.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 6.173265600204468,
|
|
"epoch": 5.895077076081551,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00024354855465837967,
|
|
"loss": 5.4884,
|
|
"mean_token_accuracy": 0.19526475071907043,
|
|
"num_tokens": 11616202.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 6.166230916976929,
|
|
"epoch": 5.900049726504227,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00024316190697599955,
|
|
"loss": 5.3771,
|
|
"mean_token_accuracy": 0.20703060626983644,
|
|
"num_tokens": 11625054.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 6.1557121753692625,
|
|
"epoch": 5.905022376926902,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00024277535520928257,
|
|
"loss": 5.4783,
|
|
"mean_token_accuracy": 0.1943609356880188,
|
|
"num_tokens": 11634636.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 6.1998481273651125,
|
|
"epoch": 5.909995027349577,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00024238890052275742,
|
|
"loss": 5.5443,
|
|
"mean_token_accuracy": 0.18736673891544342,
|
|
"num_tokens": 11643949.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 6.208349132537842,
|
|
"epoch": 5.914967677772252,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024200254408066014,
|
|
"loss": 5.4949,
|
|
"mean_token_accuracy": 0.19262496381998062,
|
|
"num_tokens": 11653964.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 6.216341209411621,
|
|
"epoch": 5.919940328194928,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002416162870469308,
|
|
"loss": 5.5003,
|
|
"mean_token_accuracy": 0.1977234661579132,
|
|
"num_tokens": 11662774.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 6.212335062026978,
|
|
"epoch": 5.924912978617603,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002412301305852102,
|
|
"loss": 5.5317,
|
|
"mean_token_accuracy": 0.18417190462350846,
|
|
"num_tokens": 11672504.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 6.2016270637512205,
|
|
"epoch": 5.929885629040278,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024084407585883588,
|
|
"loss": 5.5595,
|
|
"mean_token_accuracy": 0.19079016298055648,
|
|
"num_tokens": 11682225.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 6.198901081085205,
|
|
"epoch": 5.934858279462953,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002404581240308391,
|
|
"loss": 5.5351,
|
|
"mean_token_accuracy": 0.18596795648336412,
|
|
"num_tokens": 11691886.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 6.149114561080933,
|
|
"epoch": 5.939830929885629,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000240072276263941,
|
|
"loss": 5.4385,
|
|
"mean_token_accuracy": 0.1972994714975357,
|
|
"num_tokens": 11701366.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 6.14278736114502,
|
|
"epoch": 5.944803580308305,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023968653372054933,
|
|
"loss": 5.3602,
|
|
"mean_token_accuracy": 0.20109702348709108,
|
|
"num_tokens": 11710309.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 6.2338783740997314,
|
|
"epoch": 5.949776230730979,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023930089756275474,
|
|
"loss": 5.5267,
|
|
"mean_token_accuracy": 0.1863902062177658,
|
|
"num_tokens": 11719123.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 6.150992774963379,
|
|
"epoch": 5.954748881153655,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002389153689523275,
|
|
"loss": 5.4686,
|
|
"mean_token_accuracy": 0.19715814888477326,
|
|
"num_tokens": 11728931.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 6.205615663528443,
|
|
"epoch": 5.9597215315763306,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023852994905071385,
|
|
"loss": 5.4616,
|
|
"mean_token_accuracy": 0.19389537423849107,
|
|
"num_tokens": 11738666.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 6.177970743179321,
|
|
"epoch": 5.964694181999006,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00023814463901903238,
|
|
"loss": 5.5,
|
|
"mean_token_accuracy": 0.1945701539516449,
|
|
"num_tokens": 11748530.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.964694181999006,
|
|
"eval_entropy": 6.061989250271002,
|
|
"eval_loss": 6.436110019683838,
|
|
"eval_mean_token_accuracy": 0.15027962455929425,
|
|
"eval_num_tokens": 11748530.0,
|
|
"eval_runtime": 1.9276,
|
|
"eval_samples_per_second": 1797.611,
|
|
"eval_steps_per_second": 225.155,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 6.192976808547973,
|
|
"epoch": 5.969666832421681,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023775944001807077,
|
|
"loss": 5.5115,
|
|
"mean_token_accuracy": 0.18898769915103913,
|
|
"num_tokens": 11759611.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 6.2512797832489015,
|
|
"epoch": 5.974639482844356,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00023737435320828244,
|
|
"loss": 5.4704,
|
|
"mean_token_accuracy": 0.19322008192539214,
|
|
"num_tokens": 11769427.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 6.121019744873047,
|
|
"epoch": 5.979612133267032,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00023698937974978257,
|
|
"loss": 5.3084,
|
|
"mean_token_accuracy": 0.21014534831047058,
|
|
"num_tokens": 11778992.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 6.059270000457763,
|
|
"epoch": 5.984584783689707,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00023660452080234473,
|
|
"loss": 5.365,
|
|
"mean_token_accuracy": 0.21044591218233108,
|
|
"num_tokens": 11789844.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 6.180132246017456,
|
|
"epoch": 5.989557434112382,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023621977752539785,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.1940154477953911,
|
|
"num_tokens": 11800336.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 6.2097114562988285,
|
|
"epoch": 5.994530084535057,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002358351510780223,
|
|
"loss": 5.5206,
|
|
"mean_token_accuracy": 0.19156551361083984,
|
|
"num_tokens": 11810081.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 6.215184688568115,
|
|
"epoch": 5.999502734957733,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00023545064261894623,
|
|
"loss": 5.4829,
|
|
"mean_token_accuracy": 0.1928404986858368,
|
|
"num_tokens": 11820046.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 6.198449081844753,
|
|
"epoch": 6.00397812033814,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00023506625330654263,
|
|
"loss": 5.2614,
|
|
"mean_token_accuracy": 0.2163579977220959,
|
|
"num_tokens": 11828769.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 6.242174911499023,
|
|
"epoch": 6.008950770760816,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00023468198429882548,
|
|
"loss": 5.2562,
|
|
"mean_token_accuracy": 0.20609267055988312,
|
|
"num_tokens": 11838217.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 6.201699876785279,
|
|
"epoch": 6.013923421183491,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002342978367534462,
|
|
"loss": 5.3802,
|
|
"mean_token_accuracy": 0.19845359921455383,
|
|
"num_tokens": 11847268.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 6.180997276306153,
|
|
"epoch": 6.018896071606166,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002339138118276905,
|
|
"loss": 5.3701,
|
|
"mean_token_accuracy": 0.2022520661354065,
|
|
"num_tokens": 11857781.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 6.145877361297607,
|
|
"epoch": 6.023868722028841,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023352991067847456,
|
|
"loss": 5.1981,
|
|
"mean_token_accuracy": 0.21169434934854509,
|
|
"num_tokens": 11867810.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 6.125358057022095,
|
|
"epoch": 6.028841372451517,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023314613446234162,
|
|
"loss": 5.2495,
|
|
"mean_token_accuracy": 0.2119219794869423,
|
|
"num_tokens": 11877698.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 6.142247438430786,
|
|
"epoch": 6.033814022874192,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023276248433545867,
|
|
"loss": 5.3469,
|
|
"mean_token_accuracy": 0.20616139322519303,
|
|
"num_tokens": 11888696.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 6.2534980297088625,
|
|
"epoch": 6.038786673296867,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00023237896145361276,
|
|
"loss": 5.4098,
|
|
"mean_token_accuracy": 0.19901968836784362,
|
|
"num_tokens": 11898865.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 6.1437276840209964,
|
|
"epoch": 6.043759323719542,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00023199556697220758,
|
|
"loss": 5.2547,
|
|
"mean_token_accuracy": 0.2137102246284485,
|
|
"num_tokens": 11909506.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 6.157366514205933,
|
|
"epoch": 6.048731974142218,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023161230204626015,
|
|
"loss": 5.2761,
|
|
"mean_token_accuracy": 0.20969801396131516,
|
|
"num_tokens": 11918663.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 6.250073099136353,
|
|
"epoch": 6.053704624564893,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00023122916783039694,
|
|
"loss": 5.4147,
|
|
"mean_token_accuracy": 0.20144670605659484,
|
|
"num_tokens": 11928683.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 6.160496091842651,
|
|
"epoch": 6.058677274987568,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002308461654788509,
|
|
"loss": 5.2485,
|
|
"mean_token_accuracy": 0.21229866296052932,
|
|
"num_tokens": 11937251.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 6.158193302154541,
|
|
"epoch": 6.063649925410243,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00023046329614545758,
|
|
"loss": 5.264,
|
|
"mean_token_accuracy": 0.20873568803071976,
|
|
"num_tokens": 11946061.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 6.151159048080444,
|
|
"epoch": 6.068622575832919,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00023008056098365183,
|
|
"loss": 5.2427,
|
|
"mean_token_accuracy": 0.206567744910717,
|
|
"num_tokens": 11956512.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 6.159635210037232,
|
|
"epoch": 6.073595226255594,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022969796114646424,
|
|
"loss": 5.3535,
|
|
"mean_token_accuracy": 0.2004166752099991,
|
|
"num_tokens": 11966052.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 6.1625025272369385,
|
|
"epoch": 6.078567876678269,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022931549778651783,
|
|
"loss": 5.2959,
|
|
"mean_token_accuracy": 0.20723561197519302,
|
|
"num_tokens": 11974742.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 6.1945319175720215,
|
|
"epoch": 6.083540527100944,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022893317205602444,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.21162859201431275,
|
|
"num_tokens": 11985009.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 6.166810655593872,
|
|
"epoch": 6.08851317752362,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00022855098510678125,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.20758309662342073,
|
|
"num_tokens": 11994685.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 6.220981502532959,
|
|
"epoch": 6.093485827946296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00022816893809016737,
|
|
"loss": 5.3625,
|
|
"mean_token_accuracy": 0.20074956566095353,
|
|
"num_tokens": 12004225.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 6.092534971237183,
|
|
"epoch": 6.098458478368971,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00022778703215714014,
|
|
"loss": 5.2892,
|
|
"mean_token_accuracy": 0.20928363054990767,
|
|
"num_tokens": 12013422.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 6.045114946365357,
|
|
"epoch": 6.103431128791646,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00022740526845823223,
|
|
"loss": 5.1541,
|
|
"mean_token_accuracy": 0.22685232013463974,
|
|
"num_tokens": 12023739.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 6.251814270019532,
|
|
"epoch": 6.1084037792143215,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002270236481435477,
|
|
"loss": 5.403,
|
|
"mean_token_accuracy": 0.19793127924203874,
|
|
"num_tokens": 12034189.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 6.238321352005005,
|
|
"epoch": 6.113376429636997,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022664217236275847,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.1951762095093727,
|
|
"num_tokens": 12043134.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 6.202394199371338,
|
|
"epoch": 6.118349080059672,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022626084226510124,
|
|
"loss": 5.2695,
|
|
"mean_token_accuracy": 0.20055439621210097,
|
|
"num_tokens": 12052843.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 6.2221291065216064,
|
|
"epoch": 6.123321730482347,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022587965899937364,
|
|
"loss": 5.3594,
|
|
"mean_token_accuracy": 0.20139932483434678,
|
|
"num_tokens": 12063260.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 6.159873723983765,
|
|
"epoch": 6.1282943809050225,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002254986237139312,
|
|
"loss": 5.352,
|
|
"mean_token_accuracy": 0.20042279064655305,
|
|
"num_tokens": 12072534.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 6.193654632568359,
|
|
"epoch": 6.133267031327698,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022511773755668342,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.18903882056474686,
|
|
"num_tokens": 12081796.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 6.062978649139405,
|
|
"epoch": 6.138239681750373,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002247370016750907,
|
|
"loss": 5.1871,
|
|
"mean_token_accuracy": 0.2163309335708618,
|
|
"num_tokens": 12092885.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 6.19430251121521,
|
|
"epoch": 6.143212332173048,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002243564172161605,
|
|
"loss": 5.3087,
|
|
"mean_token_accuracy": 0.20299756973981858,
|
|
"num_tokens": 12102432.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 6.048294162750244,
|
|
"epoch": 6.1481849825957235,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00022397598532644441,
|
|
"loss": 5.1672,
|
|
"mean_token_accuracy": 0.22341015040874482,
|
|
"num_tokens": 12112245.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 6.146071577072144,
|
|
"epoch": 6.153157633018399,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022359570715203408,
|
|
"loss": 5.273,
|
|
"mean_token_accuracy": 0.20776588320732117,
|
|
"num_tokens": 12121850.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 6.155256128311157,
|
|
"epoch": 6.158130283441074,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00022321558383855828,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.22041698098182677,
|
|
"num_tokens": 12132538.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 5.996930027008057,
|
|
"epoch": 6.163102933863749,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002228356165311791,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.22774107456207277,
|
|
"num_tokens": 12142041.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 6.245426988601684,
|
|
"epoch": 6.1680755842864246,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022245580637458883,
|
|
"loss": 5.4572,
|
|
"mean_token_accuracy": 0.1951980859041214,
|
|
"num_tokens": 12151075.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 6.264968585968018,
|
|
"epoch": 6.1730482347091,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00022207615451300616,
|
|
"loss": 5.3741,
|
|
"mean_token_accuracy": 0.19857178181409835,
|
|
"num_tokens": 12160657.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 6.224822902679444,
|
|
"epoch": 6.178020885131775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022169666209017304,
|
|
"loss": 5.4875,
|
|
"mean_token_accuracy": 0.19328429400920868,
|
|
"num_tokens": 12171692.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 6.06883544921875,
|
|
"epoch": 6.18299353555445,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002213173302493509,
|
|
"loss": 5.2162,
|
|
"mean_token_accuracy": 0.2132483348250389,
|
|
"num_tokens": 12181429.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 6.133856153488159,
|
|
"epoch": 6.1879661859771256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00022093816013331763,
|
|
"loss": 5.2768,
|
|
"mean_token_accuracy": 0.21482496708631516,
|
|
"num_tokens": 12191992.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 6.065379190444946,
|
|
"epoch": 6.192938836399801,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022055915288436362,
|
|
"loss": 5.2613,
|
|
"mean_token_accuracy": 0.21494257152080537,
|
|
"num_tokens": 12200804.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 6.273176431655884,
|
|
"epoch": 6.197911486822476,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00022018030964428897,
|
|
"loss": 5.4137,
|
|
"mean_token_accuracy": 0.190847946703434,
|
|
"num_tokens": 12211774.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 6.186079597473144,
|
|
"epoch": 6.202884137245151,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021980163155439943,
|
|
"loss": 5.3221,
|
|
"mean_token_accuracy": 0.20729541033506393,
|
|
"num_tokens": 12220270.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 6.176749658584595,
|
|
"epoch": 6.207856787667827,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021942311975550323,
|
|
"loss": 5.3925,
|
|
"mean_token_accuracy": 0.19734973311424256,
|
|
"num_tokens": 12230662.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 6.193095445632935,
|
|
"epoch": 6.212829438090502,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00021904477538790774,
|
|
"loss": 5.4366,
|
|
"mean_token_accuracy": 0.19687764048576356,
|
|
"num_tokens": 12240383.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 6.1750411033630375,
|
|
"epoch": 6.217802088513178,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00021866659959141578,
|
|
"loss": 5.3868,
|
|
"mean_token_accuracy": 0.20115018486976624,
|
|
"num_tokens": 12250247.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 6.084330797195435,
|
|
"epoch": 6.222774738935853,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002182885935053225,
|
|
"loss": 5.2423,
|
|
"mean_token_accuracy": 0.22114407122135163,
|
|
"num_tokens": 12261245.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 6.204240083694458,
|
|
"epoch": 6.2277473893585285,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00021791075826841166,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.19137222617864608,
|
|
"num_tokens": 12271596.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 6.29051570892334,
|
|
"epoch": 6.232720039781204,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002175330950189523,
|
|
"loss": 5.4701,
|
|
"mean_token_accuracy": 0.19414646327495574,
|
|
"num_tokens": 12283124.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 6.160266590118408,
|
|
"epoch": 6.237692690203879,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002171556048946955,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.20849465280771257,
|
|
"num_tokens": 12293399.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 6.02526364326477,
|
|
"epoch": 6.242665340626554,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00021677828903287063,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.22288449853658676,
|
|
"num_tokens": 12303396.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 6.1786637783050535,
|
|
"epoch": 6.2476379910492295,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021640114857018205,
|
|
"loss": 5.3454,
|
|
"mean_token_accuracy": 0.2000336229801178,
|
|
"num_tokens": 12312483.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 6.1429096221923825,
|
|
"epoch": 6.252610641471905,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00021602418464280586,
|
|
"loss": 5.2279,
|
|
"mean_token_accuracy": 0.21664342135190964,
|
|
"num_tokens": 12322420.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 6.294628143310547,
|
|
"epoch": 6.25758329189458,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021564739838638615,
|
|
"loss": 5.5284,
|
|
"mean_token_accuracy": 0.18682329654693602,
|
|
"num_tokens": 12331740.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 6.225471830368042,
|
|
"epoch": 6.262555942317255,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00021527079093603202,
|
|
"loss": 5.3926,
|
|
"mean_token_accuracy": 0.19420452117919923,
|
|
"num_tokens": 12341803.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 6.1433844566345215,
|
|
"epoch": 6.2675285927399305,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00021489436342631352,
|
|
"loss": 5.297,
|
|
"mean_token_accuracy": 0.20837516486644744,
|
|
"num_tokens": 12350824.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 6.214679718017578,
|
|
"epoch": 6.272501243162606,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021451811699125883,
|
|
"loss": 5.3701,
|
|
"mean_token_accuracy": 0.19645046591758727,
|
|
"num_tokens": 12360116.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 6.18390007019043,
|
|
"epoch": 6.277473893585281,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00021414205276435073,
|
|
"loss": 5.3286,
|
|
"mean_token_accuracy": 0.20339615792036056,
|
|
"num_tokens": 12370486.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 6.135993719100952,
|
|
"epoch": 6.282446544007956,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002137661718785229,
|
|
"loss": 5.2405,
|
|
"mean_token_accuracy": 0.2120935246348381,
|
|
"num_tokens": 12380178.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 6.189220666885376,
|
|
"epoch": 6.2874191944306315,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00021339047546615675,
|
|
"loss": 5.3864,
|
|
"mean_token_accuracy": 0.19760001003742217,
|
|
"num_tokens": 12389844.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 6.226894092559815,
|
|
"epoch": 6.292391844853307,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00021301496465907794,
|
|
"loss": 5.4027,
|
|
"mean_token_accuracy": 0.20374062955379485,
|
|
"num_tokens": 12400880.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 6.231021118164063,
|
|
"epoch": 6.297364495275982,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021263964058855294,
|
|
"loss": 5.3964,
|
|
"mean_token_accuracy": 0.19737664461135865,
|
|
"num_tokens": 12410833.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 6.245246124267578,
|
|
"epoch": 6.302337145698657,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00021226450438528572,
|
|
"loss": 5.2975,
|
|
"mean_token_accuracy": 0.21298856735229493,
|
|
"num_tokens": 12419993.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 6.183073377609253,
|
|
"epoch": 6.3073097961213325,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00021188955717941428,
|
|
"loss": 5.4339,
|
|
"mean_token_accuracy": 0.19749951660633086,
|
|
"num_tokens": 12431395.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 6.088893175125122,
|
|
"epoch": 6.312282446544008,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002115148001005071,
|
|
"loss": 5.2444,
|
|
"mean_token_accuracy": 0.20660543739795684,
|
|
"num_tokens": 12440961.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 6.20079026222229,
|
|
"epoch": 6.317255096966683,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021114023427756012,
|
|
"loss": 5.4154,
|
|
"mean_token_accuracy": 0.2045823574066162,
|
|
"num_tokens": 12450333.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 6.212509346008301,
|
|
"epoch": 6.322227747389358,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00021076586083899296,
|
|
"loss": 5.4144,
|
|
"mean_token_accuracy": 0.19474466741085053,
|
|
"num_tokens": 12459373.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 6.132611179351807,
|
|
"epoch": 6.3272003978120335,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002103916809126456,
|
|
"loss": 5.2887,
|
|
"mean_token_accuracy": 0.21082127839326859,
|
|
"num_tokens": 12469366.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 6.223371601104736,
|
|
"epoch": 6.332173048234709,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021001769562577517,
|
|
"loss": 5.3267,
|
|
"mean_token_accuracy": 0.2061912775039673,
|
|
"num_tokens": 12478650.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 5.957190132141113,
|
|
"epoch": 6.337145698657384,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002096439061050523,
|
|
"loss": 5.0416,
|
|
"mean_token_accuracy": 0.2312224179506302,
|
|
"num_tokens": 12490607.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 6.14608678817749,
|
|
"epoch": 6.34211834908006,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000209270313476558,
|
|
"loss": 5.3638,
|
|
"mean_token_accuracy": 0.1985326364636421,
|
|
"num_tokens": 12502893.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 6.022424030303955,
|
|
"epoch": 6.347090999502735,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00020889691886578004,
|
|
"loss": 5.165,
|
|
"mean_token_accuracy": 0.22769471108913422,
|
|
"num_tokens": 12513470.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 6.229014205932617,
|
|
"epoch": 6.352063649925411,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020852372339760973,
|
|
"loss": 5.373,
|
|
"mean_token_accuracy": 0.2003431349992752,
|
|
"num_tokens": 12522982.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 6.048313474655151,
|
|
"epoch": 6.357036300348086,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002081507281963383,
|
|
"loss": 5.2826,
|
|
"mean_token_accuracy": 0.2067160040140152,
|
|
"num_tokens": 12533247.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 6.094562244415283,
|
|
"epoch": 6.362008950770761,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00020777793438565368,
|
|
"loss": 5.3996,
|
|
"mean_token_accuracy": 0.19904104620218277,
|
|
"num_tokens": 12543488.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 6.225197458267212,
|
|
"epoch": 6.366981601193436,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00020740534308863717,
|
|
"loss": 5.4632,
|
|
"mean_token_accuracy": 0.19361710101366042,
|
|
"num_tokens": 12553782.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 6.257753705978393,
|
|
"epoch": 6.371954251616112,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020703295542775992,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.1903871178627014,
|
|
"num_tokens": 12562913.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 6.159871625900268,
|
|
"epoch": 6.376926902038787,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00020666077252487968,
|
|
"loss": 5.2366,
|
|
"mean_token_accuracy": 0.21530804485082627,
|
|
"num_tokens": 12571979.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 5.9264411449432375,
|
|
"epoch": 6.381899552461462,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002062887955012372,
|
|
"loss": 5.0767,
|
|
"mean_token_accuracy": 0.23475490957498552,
|
|
"num_tokens": 12581445.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 6.207914781570435,
|
|
"epoch": 6.386872202884137,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00020591702547745306,
|
|
"loss": 5.4566,
|
|
"mean_token_accuracy": 0.1940680354833603,
|
|
"num_tokens": 12592437.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 6.1588245868682865,
|
|
"epoch": 6.391844853306813,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00020554546357352433,
|
|
"loss": 5.3249,
|
|
"mean_token_accuracy": 0.20703023672103882,
|
|
"num_tokens": 12602946.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 6.2462725162506105,
|
|
"epoch": 6.396817503729488,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000205174110908821,
|
|
"loss": 5.4069,
|
|
"mean_token_accuracy": 0.2001415193080902,
|
|
"num_tokens": 12612620.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 6.1421479225158695,
|
|
"epoch": 6.401790154152163,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002048029686020827,
|
|
"loss": 5.3388,
|
|
"mean_token_accuracy": 0.19522664397954942,
|
|
"num_tokens": 12621588.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 6.107249784469604,
|
|
"epoch": 6.406762804574838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002044320377714154,
|
|
"loss": 5.2677,
|
|
"mean_token_accuracy": 0.20796940922737123,
|
|
"num_tokens": 12630802.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 6.179610633850098,
|
|
"epoch": 6.411735454997514,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00020406131953428785,
|
|
"loss": 5.2754,
|
|
"mean_token_accuracy": 0.20968370735645295,
|
|
"num_tokens": 12640358.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 6.225587224960327,
|
|
"epoch": 6.416708105420189,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002036908150075285,
|
|
"loss": 5.4508,
|
|
"mean_token_accuracy": 0.1989014685153961,
|
|
"num_tokens": 12650598.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 6.160083055496216,
|
|
"epoch": 6.421680755842864,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0002033205253073218,
|
|
"loss": 5.3112,
|
|
"mean_token_accuracy": 0.20788884907960892,
|
|
"num_tokens": 12661213.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 6.209637117385864,
|
|
"epoch": 6.426653406265539,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020295045154920528,
|
|
"loss": 5.3631,
|
|
"mean_token_accuracy": 0.20595581978559493,
|
|
"num_tokens": 12670374.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 6.1385337829589846,
|
|
"epoch": 6.431626056688215,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00020258059484806564,
|
|
"loss": 5.3357,
|
|
"mean_token_accuracy": 0.20517788231372833,
|
|
"num_tokens": 12679725.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 6.129979467391967,
|
|
"epoch": 6.43659870711089,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00020221095631813574,
|
|
"loss": 5.2865,
|
|
"mean_token_accuracy": 0.20605688095092772,
|
|
"num_tokens": 12688391.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 6.219012594223022,
|
|
"epoch": 6.441571357533565,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002018415370729913,
|
|
"loss": 5.4133,
|
|
"mean_token_accuracy": 0.1988206148147583,
|
|
"num_tokens": 12698030.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 6.159636354446411,
|
|
"epoch": 6.44654400795624,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00020147233822554722,
|
|
"loss": 5.4101,
|
|
"mean_token_accuracy": 0.20323293209075927,
|
|
"num_tokens": 12708577.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 6.214919519424439,
|
|
"epoch": 6.451516658378916,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00020110336088805463,
|
|
"loss": 5.4179,
|
|
"mean_token_accuracy": 0.19207817167043686,
|
|
"num_tokens": 12718573.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 6.204273319244384,
|
|
"epoch": 6.456489308801591,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00020073460617209732,
|
|
"loss": 5.2471,
|
|
"mean_token_accuracy": 0.20768800377845764,
|
|
"num_tokens": 12728246.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 6.173808574676514,
|
|
"epoch": 6.461461959224266,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00020036607518858824,
|
|
"loss": 5.3421,
|
|
"mean_token_accuracy": 0.21279857009649278,
|
|
"num_tokens": 12737807.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 6.213733053207397,
|
|
"epoch": 6.466434609646942,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019999776904776646,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.1919709712266922,
|
|
"num_tokens": 12747463.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 6.229540729522705,
|
|
"epoch": 6.471407260069617,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001996296888591937,
|
|
"loss": 5.3459,
|
|
"mean_token_accuracy": 0.1997175395488739,
|
|
"num_tokens": 12757449.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 6.1976385593414305,
|
|
"epoch": 6.476379910492293,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019926183573175084,
|
|
"loss": 5.3108,
|
|
"mean_token_accuracy": 0.21077080965042114,
|
|
"num_tokens": 12767698.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 6.1255340576171875,
|
|
"epoch": 6.481352560914968,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001988942107736348,
|
|
"loss": 5.3263,
|
|
"mean_token_accuracy": 0.19955979883670807,
|
|
"num_tokens": 12776252.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 6.1220300674438475,
|
|
"epoch": 6.486325211337643,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019852681509235516,
|
|
"loss": 5.2647,
|
|
"mean_token_accuracy": 0.20946802645921708,
|
|
"num_tokens": 12785001.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 6.186177730560303,
|
|
"epoch": 6.4912978617603185,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019815964979473088,
|
|
"loss": 5.4007,
|
|
"mean_token_accuracy": 0.19769272953271866,
|
|
"num_tokens": 12794587.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 6.19054069519043,
|
|
"epoch": 6.496270512182994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019779271598688642,
|
|
"loss": 5.3469,
|
|
"mean_token_accuracy": 0.2017733111977577,
|
|
"num_tokens": 12803439.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 6.157043933868408,
|
|
"epoch": 6.501243162605669,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019742601477424937,
|
|
"loss": 5.401,
|
|
"mean_token_accuracy": 0.19467286616563798,
|
|
"num_tokens": 12813039.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 6.120511484146118,
|
|
"epoch": 6.506215813028344,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001970595472615463,
|
|
"loss": 5.3177,
|
|
"mean_token_accuracy": 0.2070774480700493,
|
|
"num_tokens": 12823105.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 6.08782958984375,
|
|
"epoch": 6.5111884634510195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019669331455279989,
|
|
"loss": 5.3158,
|
|
"mean_token_accuracy": 0.2083218663930893,
|
|
"num_tokens": 12833576.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 6.204066467285156,
|
|
"epoch": 6.516161113873695,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00019632731775132522,
|
|
"loss": 5.4466,
|
|
"mean_token_accuracy": 0.19658162742853164,
|
|
"num_tokens": 12843570.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 6.208154344558716,
|
|
"epoch": 6.52113376429637,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019596155795972698,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.20629088431596757,
|
|
"num_tokens": 12852751.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 6.211516809463501,
|
|
"epoch": 6.526106414719045,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001955960362798957,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.20079130977392196,
|
|
"num_tokens": 12863114.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 6.123029136657715,
|
|
"epoch": 6.5310790651417205,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001952307538130045,
|
|
"loss": 5.2465,
|
|
"mean_token_accuracy": 0.20475418567657472,
|
|
"num_tokens": 12871731.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 6.131535148620605,
|
|
"epoch": 6.536051715564396,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001948657116595059,
|
|
"loss": 5.3242,
|
|
"mean_token_accuracy": 0.20023371726274491,
|
|
"num_tokens": 12880261.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 6.142472267150879,
|
|
"epoch": 6.541024365987071,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019450091091912847,
|
|
"loss": 5.2725,
|
|
"mean_token_accuracy": 0.20938150584697723,
|
|
"num_tokens": 12889528.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 6.007477474212647,
|
|
"epoch": 6.545997016409746,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001941363526908736,
|
|
"loss": 5.2165,
|
|
"mean_token_accuracy": 0.22507892102003096,
|
|
"num_tokens": 12899734.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 6.175390005111694,
|
|
"epoch": 6.5509696668324215,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00019377203807301185,
|
|
"loss": 5.3148,
|
|
"mean_token_accuracy": 0.2023821696639061,
|
|
"num_tokens": 12909960.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 6.118598365783692,
|
|
"epoch": 6.555942317255097,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00019340796816308,
|
|
"loss": 5.3262,
|
|
"mean_token_accuracy": 0.21056386530399324,
|
|
"num_tokens": 12919411.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 6.200818061828613,
|
|
"epoch": 6.560914967677772,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019304414405787768,
|
|
"loss": 5.4283,
|
|
"mean_token_accuracy": 0.20197722762823106,
|
|
"num_tokens": 12929470.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 6.1788777828216555,
|
|
"epoch": 6.565887618100447,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001926805668534639,
|
|
"loss": 5.379,
|
|
"mean_token_accuracy": 0.199052694439888,
|
|
"num_tokens": 12940292.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 6.195863437652588,
|
|
"epoch": 6.5708602685231225,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00019231723764515397,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.1962520018219948,
|
|
"num_tokens": 12950218.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 6.202704048156738,
|
|
"epoch": 6.575832918945798,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00019195415752751595,
|
|
"loss": 5.4365,
|
|
"mean_token_accuracy": 0.19866038411855697,
|
|
"num_tokens": 12961082.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 6.211267280578613,
|
|
"epoch": 6.580805569368473,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00019159132759436764,
|
|
"loss": 5.4271,
|
|
"mean_token_accuracy": 0.18873112499713898,
|
|
"num_tokens": 12970913.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 6.254213905334472,
|
|
"epoch": 6.585778219791148,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019122874893877294,
|
|
"loss": 5.41,
|
|
"mean_token_accuracy": 0.19683656096458435,
|
|
"num_tokens": 12981201.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 6.226009559631348,
|
|
"epoch": 6.590750870213824,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00019086642265303894,
|
|
"loss": 5.4421,
|
|
"mean_token_accuracy": 0.19422940909862518,
|
|
"num_tokens": 12991283.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 6.15442910194397,
|
|
"epoch": 6.595723520636499,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001905043498287124,
|
|
"loss": 5.2273,
|
|
"mean_token_accuracy": 0.22058405578136445,
|
|
"num_tokens": 13001201.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 6.009544229507446,
|
|
"epoch": 6.600696171059175,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001901425315565764,
|
|
"loss": 5.1631,
|
|
"mean_token_accuracy": 0.22629863917827606,
|
|
"num_tokens": 13011119.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 6.18225588798523,
|
|
"epoch": 6.60566882148185,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00018978096892664726,
|
|
"loss": 5.349,
|
|
"mean_token_accuracy": 0.2033852905035019,
|
|
"num_tokens": 13021459.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 6.0900530338287355,
|
|
"epoch": 6.6106414719045254,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000189419663028171,
|
|
"loss": 5.2713,
|
|
"mean_token_accuracy": 0.21112505793571473,
|
|
"num_tokens": 13032021.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 6.178182363510132,
|
|
"epoch": 6.615614122327201,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00018905861494962034,
|
|
"loss": 5.4092,
|
|
"mean_token_accuracy": 0.1974985972046852,
|
|
"num_tokens": 13041418.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 6.218775606155395,
|
|
"epoch": 6.620586772749876,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00018869782577869132,
|
|
"loss": 5.3824,
|
|
"mean_token_accuracy": 0.19866434037685393,
|
|
"num_tokens": 13051535.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 6.19223690032959,
|
|
"epoch": 6.625559423172551,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00018833729660229997,
|
|
"loss": 5.2951,
|
|
"mean_token_accuracy": 0.20487680435180664,
|
|
"num_tokens": 13060854.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 6.171342372894287,
|
|
"epoch": 6.6305320735952264,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018797702850657889,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.20410254448652268,
|
|
"num_tokens": 13069429.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 6.146954011917114,
|
|
"epoch": 6.635504724017902,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001876170225768744,
|
|
"loss": 5.3361,
|
|
"mean_token_accuracy": 0.20816441476345063,
|
|
"num_tokens": 13078990.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 6.0871994972229,
|
|
"epoch": 6.640477374440577,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0001872572798977429,
|
|
"loss": 5.2563,
|
|
"mean_token_accuracy": 0.22178249210119247,
|
|
"num_tokens": 13089702.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 6.183096122741699,
|
|
"epoch": 6.645450024863252,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001868978015529476,
|
|
"loss": 5.3477,
|
|
"mean_token_accuracy": 0.19685374796390534,
|
|
"num_tokens": 13099890.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 6.192337608337402,
|
|
"epoch": 6.6504226752859275,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018653858862545565,
|
|
"loss": 5.431,
|
|
"mean_token_accuracy": 0.19664229601621627,
|
|
"num_tokens": 13108767.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 5.959743595123291,
|
|
"epoch": 6.655395325708603,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00018617964219743428,
|
|
"loss": 5.0454,
|
|
"mean_token_accuracy": 0.23613035678863525,
|
|
"num_tokens": 13118548.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 6.114861488342285,
|
|
"epoch": 6.660367976131278,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00018582096335024813,
|
|
"loss": 5.37,
|
|
"mean_token_accuracy": 0.20695649236440658,
|
|
"num_tokens": 13129451.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 6.178614091873169,
|
|
"epoch": 6.665340626553953,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001854625531644557,
|
|
"loss": 5.4122,
|
|
"mean_token_accuracy": 0.19407947063446046,
|
|
"num_tokens": 13138679.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 6.2284962177276615,
|
|
"epoch": 6.6703132769766285,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00018510441271980588,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.19095547050237655,
|
|
"num_tokens": 13148980.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 6.14340181350708,
|
|
"epoch": 6.675285927399304,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00018474654309523508,
|
|
"loss": 5.2919,
|
|
"mean_token_accuracy": 0.21084808558225632,
|
|
"num_tokens": 13158743.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 6.122425603866577,
|
|
"epoch": 6.680258577821979,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00018438894536886396,
|
|
"loss": 5.2991,
|
|
"mean_token_accuracy": 0.2108449012041092,
|
|
"num_tokens": 13168371.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 6.182071256637573,
|
|
"epoch": 6.685231228244654,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018403162061799395,
|
|
"loss": 5.3563,
|
|
"mean_token_accuracy": 0.19799830615520478,
|
|
"num_tokens": 13177399.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 6.188065004348755,
|
|
"epoch": 6.6902038786673295,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018367456991910403,
|
|
"loss": 5.2806,
|
|
"mean_token_accuracy": 0.20680064260959624,
|
|
"num_tokens": 13186689.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 6.160720729827881,
|
|
"epoch": 6.695176529090005,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00018331779434784773,
|
|
"loss": 5.2721,
|
|
"mean_token_accuracy": 0.2070927545428276,
|
|
"num_tokens": 13198859.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 6.205449962615967,
|
|
"epoch": 6.70014917951268,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00018296129497904955,
|
|
"loss": 5.3351,
|
|
"mean_token_accuracy": 0.19956362396478652,
|
|
"num_tokens": 13207237.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 6.206780576705933,
|
|
"epoch": 6.705121829935355,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001826050728867021,
|
|
"loss": 5.3467,
|
|
"mean_token_accuracy": 0.20436042547225952,
|
|
"num_tokens": 13216163.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 6.248223924636841,
|
|
"epoch": 6.7100944803580305,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00018224912914396253,
|
|
"loss": 5.474,
|
|
"mean_token_accuracy": 0.19069626033306122,
|
|
"num_tokens": 13225714.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 6.111670303344726,
|
|
"epoch": 6.715067130780707,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00018189346482314945,
|
|
"loss": 5.326,
|
|
"mean_token_accuracy": 0.20460679978132248,
|
|
"num_tokens": 13234226.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 6.178861284255982,
|
|
"epoch": 6.720039781203381,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018153808099573976,
|
|
"loss": 5.4278,
|
|
"mean_token_accuracy": 0.201892127096653,
|
|
"num_tokens": 13244286.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 6.1532810688018795,
|
|
"epoch": 6.725012431626057,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00018118297873236517,
|
|
"loss": 5.3658,
|
|
"mean_token_accuracy": 0.20363544821739196,
|
|
"num_tokens": 13254283.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 6.1738364696502686,
|
|
"epoch": 6.729985082048732,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00018082815910280925,
|
|
"loss": 5.3224,
|
|
"mean_token_accuracy": 0.20088783651590347,
|
|
"num_tokens": 13263886.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 6.223603105545044,
|
|
"epoch": 6.734957732471408,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00018047362317600424,
|
|
"loss": 5.3917,
|
|
"mean_token_accuracy": 0.19975163787603378,
|
|
"num_tokens": 13273406.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 6.0805824279785154,
|
|
"epoch": 6.739930382894083,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00018011937202002748,
|
|
"loss": 5.23,
|
|
"mean_token_accuracy": 0.22292280495166777,
|
|
"num_tokens": 13282980.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 6.134560585021973,
|
|
"epoch": 6.744903033316758,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017976540670209856,
|
|
"loss": 5.3054,
|
|
"mean_token_accuracy": 0.20249229073524475,
|
|
"num_tokens": 13292791.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 6.017916059494018,
|
|
"epoch": 6.749875683739433,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017941172828857572,
|
|
"loss": 5.185,
|
|
"mean_token_accuracy": 0.22657199054956437,
|
|
"num_tokens": 13302440.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 6.128577375411988,
|
|
"epoch": 6.754848334162109,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017905833784495325,
|
|
"loss": 5.361,
|
|
"mean_token_accuracy": 0.20422105938196183,
|
|
"num_tokens": 13312465.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 6.199922561645508,
|
|
"epoch": 6.759820984584784,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017870523643585769,
|
|
"loss": 5.3692,
|
|
"mean_token_accuracy": 0.20029908865690232,
|
|
"num_tokens": 13321991.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 6.086440563201904,
|
|
"epoch": 6.764793635007459,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00017835242512504463,
|
|
"loss": 5.2476,
|
|
"mean_token_accuracy": 0.20824384093284606,
|
|
"num_tokens": 13331671.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 6.176172590255737,
|
|
"epoch": 6.769766285430134,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00017799990497539608,
|
|
"loss": 5.3861,
|
|
"mean_token_accuracy": 0.20386532545089722,
|
|
"num_tokens": 13341196.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 6.236951732635498,
|
|
"epoch": 6.77473893585281,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001776476770489167,
|
|
"loss": 5.3974,
|
|
"mean_token_accuracy": 0.19220943748950958,
|
|
"num_tokens": 13350993.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 6.196377611160278,
|
|
"epoch": 6.779711586275485,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00017729574240673069,
|
|
"loss": 5.3337,
|
|
"mean_token_accuracy": 0.20534877926111222,
|
|
"num_tokens": 13360385.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 6.140643882751465,
|
|
"epoch": 6.78468423669816,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00017694410210907906,
|
|
"loss": 5.3496,
|
|
"mean_token_accuracy": 0.20782495886087418,
|
|
"num_tokens": 13370057.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 6.011312103271484,
|
|
"epoch": 6.789656887120835,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001765927572153156,
|
|
"loss": 5.1771,
|
|
"mean_token_accuracy": 0.22363151907920836,
|
|
"num_tokens": 13380379.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 6.211725187301636,
|
|
"epoch": 6.794629537543511,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001762417087839046,
|
|
"loss": 5.4107,
|
|
"mean_token_accuracy": 0.20119830071926117,
|
|
"num_tokens": 13389795.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 6.197131013870239,
|
|
"epoch": 6.799602187966186,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00017589095787241688,
|
|
"loss": 5.3465,
|
|
"mean_token_accuracy": 0.20323589742183684,
|
|
"num_tokens": 13400630.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 6.202950572967529,
|
|
"epoch": 6.804574838388861,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001755405055375272,
|
|
"loss": 5.3515,
|
|
"mean_token_accuracy": 0.19733516722917557,
|
|
"num_tokens": 13411044.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 6.205971527099609,
|
|
"epoch": 6.809547488811536,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00017519035283501084,
|
|
"loss": 5.3405,
|
|
"mean_token_accuracy": 0.1959124267101288,
|
|
"num_tokens": 13420092.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 6.1507401943206785,
|
|
"epoch": 6.814520139234212,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00017484050081974004,
|
|
"loss": 5.3726,
|
|
"mean_token_accuracy": 0.2063729792833328,
|
|
"num_tokens": 13430058.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 6.168980264663697,
|
|
"epoch": 6.819492789656887,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001744909505456817,
|
|
"loss": 5.3633,
|
|
"mean_token_accuracy": 0.20177532285451888,
|
|
"num_tokens": 13439061.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 6.214245796203613,
|
|
"epoch": 6.824465440079562,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017414170306589318,
|
|
"loss": 5.426,
|
|
"mean_token_accuracy": 0.20245388001203538,
|
|
"num_tokens": 13449341.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 6.2386761665344235,
|
|
"epoch": 6.829438090502237,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0001737927594325201,
|
|
"loss": 5.3779,
|
|
"mean_token_accuracy": 0.19970094710588454,
|
|
"num_tokens": 13457896.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 6.178305006027221,
|
|
"epoch": 6.834410740924913,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017344412069679257,
|
|
"loss": 5.3412,
|
|
"mean_token_accuracy": 0.20102737545967103,
|
|
"num_tokens": 13468404.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 6.186580944061279,
|
|
"epoch": 6.839383391347588,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001730957879090218,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.1979622721672058,
|
|
"num_tokens": 13477530.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 6.170603799819946,
|
|
"epoch": 6.844356041770263,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00017274776211859778,
|
|
"loss": 5.4463,
|
|
"mean_token_accuracy": 0.19679221361875535,
|
|
"num_tokens": 13487250.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 6.1768029689788815,
|
|
"epoch": 6.849328692192939,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00017240004437398554,
|
|
"loss": 5.4089,
|
|
"mean_token_accuracy": 0.19964559376239777,
|
|
"num_tokens": 13497342.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 6.246775484085083,
|
|
"epoch": 6.854301342615614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0001720526357227218,
|
|
"loss": 5.4655,
|
|
"mean_token_accuracy": 0.1932525247335434,
|
|
"num_tokens": 13507644.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 6.176058292388916,
|
|
"epoch": 6.85927399303829,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017170553721141253,
|
|
"loss": 5.3787,
|
|
"mean_token_accuracy": 0.20007410198450087,
|
|
"num_tokens": 13516713.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 6.182245588302612,
|
|
"epoch": 6.864246643460965,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001713587498857289,
|
|
"loss": 5.3294,
|
|
"mean_token_accuracy": 0.2022675335407257,
|
|
"num_tokens": 13525603.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 6.194425058364868,
|
|
"epoch": 6.86921929388364,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00017101227479040498,
|
|
"loss": 5.3394,
|
|
"mean_token_accuracy": 0.19556166380643844,
|
|
"num_tokens": 13534809.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 6.13035101890564,
|
|
"epoch": 6.8741919443063155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017066611296923424,
|
|
"loss": 5.3149,
|
|
"mean_token_accuracy": 0.2007948026061058,
|
|
"num_tokens": 13544482.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 6.134300470352173,
|
|
"epoch": 6.879164594728991,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00017032026546506595,
|
|
"loss": 5.3664,
|
|
"mean_token_accuracy": 0.19503656029701233,
|
|
"num_tokens": 13553583.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 6.180714702606201,
|
|
"epoch": 6.884137245151666,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016997473331980295,
|
|
"loss": 5.4506,
|
|
"mean_token_accuracy": 0.1887023702263832,
|
|
"num_tokens": 13562387.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 6.237661123275757,
|
|
"epoch": 6.889109895574341,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001696295175743977,
|
|
"loss": 5.4661,
|
|
"mean_token_accuracy": 0.19605302959680557,
|
|
"num_tokens": 13572654.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 6.223883390426636,
|
|
"epoch": 6.8940825459970165,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016928461926884954,
|
|
"loss": 5.3247,
|
|
"mean_token_accuracy": 0.20719764083623887,
|
|
"num_tokens": 13582066.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 6.197817850112915,
|
|
"epoch": 6.899055196419692,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00016894003944220167,
|
|
"loss": 5.3164,
|
|
"mean_token_accuracy": 0.20580175518989563,
|
|
"num_tokens": 13591702.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 6.16766586303711,
|
|
"epoch": 6.904027846842367,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001685957791325375,
|
|
"loss": 5.3737,
|
|
"mean_token_accuracy": 0.20324709862470627,
|
|
"num_tokens": 13601822.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 5.997172832489014,
|
|
"epoch": 6.909000497265042,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001682518393769782,
|
|
"loss": 5.1481,
|
|
"mean_token_accuracy": 0.2216852381825447,
|
|
"num_tokens": 13611389.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 6.078057432174683,
|
|
"epoch": 6.9139731476877175,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00016790822121167913,
|
|
"loss": 5.3548,
|
|
"mean_token_accuracy": 0.20874399095773696,
|
|
"num_tokens": 13621590.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 6.140646982192993,
|
|
"epoch": 6.918945798110393,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00016756492567182674,
|
|
"loss": 5.2631,
|
|
"mean_token_accuracy": 0.20648375898599625,
|
|
"num_tokens": 13630319.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 6.1383952617645265,
|
|
"epoch": 6.923918448533068,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016722195379163568,
|
|
"loss": 5.2672,
|
|
"mean_token_accuracy": 0.20301622599363328,
|
|
"num_tokens": 13639807.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 6.151399755477906,
|
|
"epoch": 6.928891098955743,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00016687930660434528,
|
|
"loss": 5.2904,
|
|
"mean_token_accuracy": 0.21019121408462524,
|
|
"num_tokens": 13649522.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 6.135382652282715,
|
|
"epoch": 6.9338637493784185,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00016653698514221694,
|
|
"loss": 5.2933,
|
|
"mean_token_accuracy": 0.20786001682281494,
|
|
"num_tokens": 13658855.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 6.030326223373413,
|
|
"epoch": 6.938836399801094,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001661949904365308,
|
|
"loss": 5.1932,
|
|
"mean_token_accuracy": 0.21700284481048585,
|
|
"num_tokens": 13669196.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 6.136969327926636,
|
|
"epoch": 6.943809050223769,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001658533235175824,
|
|
"loss": 5.2431,
|
|
"mean_token_accuracy": 0.2077976018190384,
|
|
"num_tokens": 13680222.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 6.16507363319397,
|
|
"epoch": 6.948781700646444,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00016551198541468,
|
|
"loss": 5.3605,
|
|
"mean_token_accuracy": 0.20974992960691452,
|
|
"num_tokens": 13689392.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 6.139012145996094,
|
|
"epoch": 6.9537543510691195,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00016517097715614115,
|
|
"loss": 5.2489,
|
|
"mean_token_accuracy": 0.21108735203742982,
|
|
"num_tokens": 13698104.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 6.1807029247283936,
|
|
"epoch": 6.958727001491795,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016483029976928958,
|
|
"loss": 5.3724,
|
|
"mean_token_accuracy": 0.2092432215809822,
|
|
"num_tokens": 13707386.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 6.958727001491795,
|
|
"eval_entropy": 6.044168541508336,
|
|
"eval_loss": 6.4264655113220215,
|
|
"eval_mean_token_accuracy": 0.15142131730708108,
|
|
"eval_num_tokens": 13707386.0,
|
|
"eval_runtime": 1.9201,
|
|
"eval_samples_per_second": 1804.548,
|
|
"eval_steps_per_second": 226.024,
|
|
"step": 7000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 10050,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 1000,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 3331782639267840.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|