d3bff78c52
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
4057 lines
109 KiB
JSON
4057 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.9885629040278467,
|
|
"eval_steps": 1000,
|
|
"global_step": 2000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 956740866508800.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|