d3bff78c52
Model: fpadovani/tur-latn-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed10 Source: Original Platform
2046 lines
54 KiB
JSON
2046 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.9945300845350572,
|
|
"eval_steps": 1000,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 477704557854720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|