d4983c5354
Model: fpadovani/isl-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.5277636271013755,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 7.24715371131897,
|
|
"epoch": 0.005094243504839531,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 7.0659,
|
|
"mean_token_accuracy": 0.09565148800611496,
|
|
"num_tokens": 12441.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 7.236126518249511,
|
|
"epoch": 0.010188487009679063,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 7.1579,
|
|
"mean_token_accuracy": 0.09300948306918144,
|
|
"num_tokens": 23925.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 7.283139085769653,
|
|
"epoch": 0.015282730514518594,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 7.067,
|
|
"mean_token_accuracy": 0.10201717466115952,
|
|
"num_tokens": 33958.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 7.281519556045533,
|
|
"epoch": 0.020376974019358125,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 7.1463,
|
|
"mean_token_accuracy": 0.0955164335668087,
|
|
"num_tokens": 46188.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.289803886413575,
|
|
"epoch": 0.02547121752419766,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 7.081,
|
|
"mean_token_accuracy": 0.09788228124380112,
|
|
"num_tokens": 58080.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 7.318724346160889,
|
|
"epoch": 0.030565461029037188,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 7.1375,
|
|
"mean_token_accuracy": 0.09244033247232437,
|
|
"num_tokens": 72446.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 7.406049919128418,
|
|
"epoch": 0.03565970453387672,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 7.1094,
|
|
"mean_token_accuracy": 0.09800529479980469,
|
|
"num_tokens": 85522.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 7.43074402809143,
|
|
"epoch": 0.04075394803871625,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.09737828224897385,
|
|
"num_tokens": 96970.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 7.456353950500488,
|
|
"epoch": 0.04584819154355578,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 7.1454,
|
|
"mean_token_accuracy": 0.10147813335061073,
|
|
"num_tokens": 108011.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 7.404264450073242,
|
|
"epoch": 0.05094243504839532,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.9644,
|
|
"mean_token_accuracy": 0.10034293979406357,
|
|
"num_tokens": 119673.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 7.425551128387451,
|
|
"epoch": 0.056036678553234846,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 7.1202,
|
|
"mean_token_accuracy": 0.09701324105262757,
|
|
"num_tokens": 133041.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 7.469049024581909,
|
|
"epoch": 0.061130922058074376,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 7.0973,
|
|
"mean_token_accuracy": 0.09784900322556496,
|
|
"num_tokens": 143467.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 7.420843839645386,
|
|
"epoch": 0.06622516556291391,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 7.0709,
|
|
"mean_token_accuracy": 0.09607185646891594,
|
|
"num_tokens": 154745.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 7.428053951263427,
|
|
"epoch": 0.07131940906775344,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.9739,
|
|
"mean_token_accuracy": 0.10642262473702431,
|
|
"num_tokens": 165526.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 7.392958068847657,
|
|
"epoch": 0.07641365257259297,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 7.0348,
|
|
"mean_token_accuracy": 0.10094631165266037,
|
|
"num_tokens": 177735.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 7.331642818450928,
|
|
"epoch": 0.0815078960774325,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.9698,
|
|
"mean_token_accuracy": 0.1012849636375904,
|
|
"num_tokens": 189877.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 7.37539849281311,
|
|
"epoch": 0.08660213958227203,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 7.1194,
|
|
"mean_token_accuracy": 0.09329616203904152,
|
|
"num_tokens": 201740.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 7.420966720581054,
|
|
"epoch": 0.09169638308711156,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 7.0655,
|
|
"mean_token_accuracy": 0.09406941458582878,
|
|
"num_tokens": 213385.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 7.407033824920655,
|
|
"epoch": 0.09679062659195109,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 7.0471,
|
|
"mean_token_accuracy": 0.0981024071574211,
|
|
"num_tokens": 225501.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 7.419355821609497,
|
|
"epoch": 0.10188487009679063,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 7.0654,
|
|
"mean_token_accuracy": 0.09309478253126144,
|
|
"num_tokens": 239512.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 7.417162942886352,
|
|
"epoch": 0.10697911360163016,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 7.0518,
|
|
"mean_token_accuracy": 0.09458114504814148,
|
|
"num_tokens": 252261.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 7.417002153396607,
|
|
"epoch": 0.11207335710646969,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 7.0866,
|
|
"mean_token_accuracy": 0.09766212329268456,
|
|
"num_tokens": 263754.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 7.41152606010437,
|
|
"epoch": 0.11716760061130922,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.9867,
|
|
"mean_token_accuracy": 0.10208452045917511,
|
|
"num_tokens": 274953.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 7.411972093582153,
|
|
"epoch": 0.12226184411614875,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 7.0836,
|
|
"mean_token_accuracy": 0.09611152485013008,
|
|
"num_tokens": 284844.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 7.369313478469849,
|
|
"epoch": 0.1273560876209883,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.0652,
|
|
"mean_token_accuracy": 0.09571715593338012,
|
|
"num_tokens": 296833.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 7.397171831130981,
|
|
"epoch": 0.13245033112582782,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.0606,
|
|
"mean_token_accuracy": 0.10037691295146942,
|
|
"num_tokens": 308528.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 7.37181396484375,
|
|
"epoch": 0.13754457463066735,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.9656,
|
|
"mean_token_accuracy": 0.10391477197408676,
|
|
"num_tokens": 320599.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 7.357561874389648,
|
|
"epoch": 0.14263881813550688,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.0179,
|
|
"mean_token_accuracy": 0.09861662462353707,
|
|
"num_tokens": 332264.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 7.355445146560669,
|
|
"epoch": 0.1477330616403464,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.9961,
|
|
"mean_token_accuracy": 0.10537474304437637,
|
|
"num_tokens": 344233.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 7.4016601085662845,
|
|
"epoch": 0.15282730514518594,
|
|
"grad_norm": 0.8046875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.0702,
|
|
"mean_token_accuracy": 0.1011621817946434,
|
|
"num_tokens": 356200.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 7.38088321685791,
|
|
"epoch": 0.15792154865002547,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.1423,
|
|
"mean_token_accuracy": 0.0946283534169197,
|
|
"num_tokens": 369791.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 7.396672773361206,
|
|
"epoch": 0.163015792154865,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.9715,
|
|
"mean_token_accuracy": 0.10692625269293785,
|
|
"num_tokens": 381127.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 7.438468647003174,
|
|
"epoch": 0.16811003565970453,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.1174,
|
|
"mean_token_accuracy": 0.09891587346792222,
|
|
"num_tokens": 393941.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.398786973953247,
|
|
"epoch": 0.17320427916454406,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.944,
|
|
"mean_token_accuracy": 0.10858480706810951,
|
|
"num_tokens": 404883.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.358920097351074,
|
|
"epoch": 0.1782985226693836,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.0333,
|
|
"mean_token_accuracy": 0.10376504063606262,
|
|
"num_tokens": 417961.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.407248592376709,
|
|
"epoch": 0.18339276617422312,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.0078,
|
|
"mean_token_accuracy": 0.10285666510462761,
|
|
"num_tokens": 430076.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.358011960983276,
|
|
"epoch": 0.18848700967906265,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.0096,
|
|
"mean_token_accuracy": 0.10075726360082626,
|
|
"num_tokens": 442091.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.3895142555236815,
|
|
"epoch": 0.19358125318390218,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.9867,
|
|
"mean_token_accuracy": 0.10208861976861953,
|
|
"num_tokens": 454272.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.400474166870117,
|
|
"epoch": 0.1986754966887417,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.9393,
|
|
"mean_token_accuracy": 0.09943680688738823,
|
|
"num_tokens": 465237.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.340952205657959,
|
|
"epoch": 0.20376974019358127,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.10289869606494903,
|
|
"num_tokens": 476842.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.417330551147461,
|
|
"epoch": 0.2088639836984208,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.091,
|
|
"mean_token_accuracy": 0.10102367475628853,
|
|
"num_tokens": 489198.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.344519233703613,
|
|
"epoch": 0.21395822720326033,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0254,
|
|
"mean_token_accuracy": 0.10164412185549736,
|
|
"num_tokens": 500320.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.442423391342163,
|
|
"epoch": 0.21905247070809986,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.1097095027565956,
|
|
"num_tokens": 512117.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.329393529891968,
|
|
"epoch": 0.22414671421293939,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.0294,
|
|
"mean_token_accuracy": 0.10038029104471206,
|
|
"num_tokens": 524106.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.479124975204468,
|
|
"epoch": 0.22924095771777891,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9523,
|
|
"mean_token_accuracy": 0.09942076131701469,
|
|
"num_tokens": 535815.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.250899314880371,
|
|
"epoch": 0.23433520122261844,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.9819,
|
|
"mean_token_accuracy": 0.09995833933353424,
|
|
"num_tokens": 548217.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.365714073181152,
|
|
"epoch": 0.23942944472745797,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.9956,
|
|
"mean_token_accuracy": 0.10438351854681968,
|
|
"num_tokens": 559218.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.321756649017334,
|
|
"epoch": 0.2445236882322975,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.9717,
|
|
"mean_token_accuracy": 0.10155888423323631,
|
|
"num_tokens": 571021.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.408106660842895,
|
|
"epoch": 0.24961793173713703,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.0651,
|
|
"mean_token_accuracy": 0.09935919940471649,
|
|
"num_tokens": 582023.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.373356723785401,
|
|
"epoch": 0.2547121752419766,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.9923,
|
|
"mean_token_accuracy": 0.10503458455204964,
|
|
"num_tokens": 594938.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.268188571929931,
|
|
"epoch": 0.2598064187468161,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.9128,
|
|
"mean_token_accuracy": 0.10892222672700883,
|
|
"num_tokens": 606978.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.342260932922363,
|
|
"epoch": 0.26490066225165565,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9451,
|
|
"mean_token_accuracy": 0.10842019617557526,
|
|
"num_tokens": 617653.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.347458028793335,
|
|
"epoch": 0.26999490575649515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9279,
|
|
"mean_token_accuracy": 0.10115889385342598,
|
|
"num_tokens": 628807.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.304141664505005,
|
|
"epoch": 0.2750891492613347,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.9881,
|
|
"mean_token_accuracy": 0.09916051551699638,
|
|
"num_tokens": 639512.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.4720312595367435,
|
|
"epoch": 0.2801833927661742,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.0029,
|
|
"mean_token_accuracy": 0.10025146678090095,
|
|
"num_tokens": 652491.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.232586193084717,
|
|
"epoch": 0.28527763627101377,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.9056,
|
|
"mean_token_accuracy": 0.11045916676521302,
|
|
"num_tokens": 664152.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.36027889251709,
|
|
"epoch": 0.29037187977585327,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.9867,
|
|
"mean_token_accuracy": 0.10425695180892944,
|
|
"num_tokens": 676774.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.373179769515991,
|
|
"epoch": 0.2954661232806928,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.9972,
|
|
"mean_token_accuracy": 0.10727740600705146,
|
|
"num_tokens": 689459.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.343478679656982,
|
|
"epoch": 0.30056036678553233,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8707,
|
|
"mean_token_accuracy": 0.11534939855337142,
|
|
"num_tokens": 700338.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.288828611373901,
|
|
"epoch": 0.3056546102903719,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.9304,
|
|
"mean_token_accuracy": 0.10776630565524101,
|
|
"num_tokens": 711764.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.266808032989502,
|
|
"epoch": 0.3107488537952114,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9168,
|
|
"mean_token_accuracy": 0.10816785246133805,
|
|
"num_tokens": 725232.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.320744132995605,
|
|
"epoch": 0.31584309730005095,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8824,
|
|
"mean_token_accuracy": 0.1124703548848629,
|
|
"num_tokens": 737036.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.303640222549438,
|
|
"epoch": 0.32093734080489045,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.9317,
|
|
"mean_token_accuracy": 0.10430295616388321,
|
|
"num_tokens": 747844.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.338167333602906,
|
|
"epoch": 0.32603158430973,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9546,
|
|
"mean_token_accuracy": 0.103739532828331,
|
|
"num_tokens": 760730.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.331214952468872,
|
|
"epoch": 0.33112582781456956,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9831,
|
|
"mean_token_accuracy": 0.10031276121735573,
|
|
"num_tokens": 773020.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.253560304641724,
|
|
"epoch": 0.33622007131940906,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9569,
|
|
"mean_token_accuracy": 0.10467252805829048,
|
|
"num_tokens": 786337.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.3519202709198,
|
|
"epoch": 0.3413143148242486,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9698,
|
|
"mean_token_accuracy": 0.10344023779034614,
|
|
"num_tokens": 798490.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.301313209533691,
|
|
"epoch": 0.3464085583290881,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.8476,
|
|
"mean_token_accuracy": 0.10969852656126022,
|
|
"num_tokens": 810448.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.347597742080689,
|
|
"epoch": 0.3515028018339277,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9274,
|
|
"mean_token_accuracy": 0.10717827007174492,
|
|
"num_tokens": 821443.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.337577152252197,
|
|
"epoch": 0.3565970453387672,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9732,
|
|
"mean_token_accuracy": 0.10012148767709732,
|
|
"num_tokens": 833935.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.234713411331176,
|
|
"epoch": 0.36169128884360674,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.9405,
|
|
"mean_token_accuracy": 0.10321041941642761,
|
|
"num_tokens": 845343.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.291878795623779,
|
|
"epoch": 0.36678553234844624,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.8832,
|
|
"mean_token_accuracy": 0.10981655418872834,
|
|
"num_tokens": 856683.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.328572940826416,
|
|
"epoch": 0.3718797758532858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.9786,
|
|
"mean_token_accuracy": 0.10471872314810753,
|
|
"num_tokens": 869190.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.334085178375244,
|
|
"epoch": 0.3769740193581253,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.9311,
|
|
"mean_token_accuracy": 0.10740116089582444,
|
|
"num_tokens": 880811.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.261279678344726,
|
|
"epoch": 0.38206826286296486,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.884,
|
|
"mean_token_accuracy": 0.1079227589070797,
|
|
"num_tokens": 892559.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.29451060295105,
|
|
"epoch": 0.38716250636780436,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.81,
|
|
"mean_token_accuracy": 0.11197227388620376,
|
|
"num_tokens": 903602.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.267256736755371,
|
|
"epoch": 0.3922567498726439,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9094,
|
|
"mean_token_accuracy": 0.10841504335403443,
|
|
"num_tokens": 915021.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.22667555809021,
|
|
"epoch": 0.3973509933774834,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.9842,
|
|
"mean_token_accuracy": 0.10425722897052765,
|
|
"num_tokens": 927084.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.245050430297852,
|
|
"epoch": 0.402445236882323,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.8916,
|
|
"mean_token_accuracy": 0.10427912101149558,
|
|
"num_tokens": 939900.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.323367309570313,
|
|
"epoch": 0.40753948038716253,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.9759,
|
|
"mean_token_accuracy": 0.0958385668694973,
|
|
"num_tokens": 953956.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.277850532531739,
|
|
"epoch": 0.41263372389200204,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.8991,
|
|
"mean_token_accuracy": 0.1134574793279171,
|
|
"num_tokens": 966526.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.1823567867279055,
|
|
"epoch": 0.4177279673968416,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.799,
|
|
"mean_token_accuracy": 0.10640768259763718,
|
|
"num_tokens": 977889.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.318188762664795,
|
|
"epoch": 0.4228222109016811,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.8991,
|
|
"mean_token_accuracy": 0.10327906757593155,
|
|
"num_tokens": 989737.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.212354612350464,
|
|
"epoch": 0.42791645440652065,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.9267,
|
|
"mean_token_accuracy": 0.10874359682202339,
|
|
"num_tokens": 1000820.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.192477703094482,
|
|
"epoch": 0.43301069791136015,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.8115,
|
|
"mean_token_accuracy": 0.11217232421040535,
|
|
"num_tokens": 1012156.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.23812313079834,
|
|
"epoch": 0.4381049414161997,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.7541,
|
|
"mean_token_accuracy": 0.11500602141022682,
|
|
"num_tokens": 1022825.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.1828369140625,
|
|
"epoch": 0.4431991849210392,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.8394,
|
|
"mean_token_accuracy": 0.10943387299776078,
|
|
"num_tokens": 1034725.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.196564865112305,
|
|
"epoch": 0.44829342842587877,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.816,
|
|
"mean_token_accuracy": 0.10969174206256867,
|
|
"num_tokens": 1045927.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.2752502918243405,
|
|
"epoch": 0.4533876719307183,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.8532,
|
|
"mean_token_accuracy": 0.108445206284523,
|
|
"num_tokens": 1057943.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.276237630844117,
|
|
"epoch": 0.45848191543555783,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9311,
|
|
"mean_token_accuracy": 0.10205709040164948,
|
|
"num_tokens": 1068949.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.20145263671875,
|
|
"epoch": 0.46357615894039733,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.7965,
|
|
"mean_token_accuracy": 0.11336861625313759,
|
|
"num_tokens": 1081298.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.164061260223389,
|
|
"epoch": 0.4686704024452369,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.8602,
|
|
"mean_token_accuracy": 0.11112992614507675,
|
|
"num_tokens": 1092522.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.170074987411499,
|
|
"epoch": 0.4737646459500764,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.7713,
|
|
"mean_token_accuracy": 0.10729397088289261,
|
|
"num_tokens": 1104133.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.155507373809814,
|
|
"epoch": 0.47885888945491595,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.7384,
|
|
"mean_token_accuracy": 0.11368128582835198,
|
|
"num_tokens": 1116057.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.160354423522949,
|
|
"epoch": 0.48395313295975545,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.8402,
|
|
"mean_token_accuracy": 0.10656984969973564,
|
|
"num_tokens": 1128148.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.16317310333252,
|
|
"epoch": 0.489047376464595,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8379,
|
|
"mean_token_accuracy": 0.107074935734272,
|
|
"num_tokens": 1140104.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.168023204803466,
|
|
"epoch": 0.49414161996943456,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.7889,
|
|
"mean_token_accuracy": 0.10834618881344796,
|
|
"num_tokens": 1152711.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.219213676452637,
|
|
"epoch": 0.49923586347427407,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.7814,
|
|
"mean_token_accuracy": 0.11331865563988686,
|
|
"num_tokens": 1164223.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.176358985900879,
|
|
"epoch": 0.5043301069791136,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.7669,
|
|
"mean_token_accuracy": 0.11871317028999329,
|
|
"num_tokens": 1174990.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.062222909927368,
|
|
"epoch": 0.5094243504839532,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.77,
|
|
"mean_token_accuracy": 0.10959928631782531,
|
|
"num_tokens": 1187561.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5094243504839532,
|
|
"eval_entropy": 7.085992576457836,
|
|
"eval_loss": 6.768796920776367,
|
|
"eval_mean_token_accuracy": 0.11561443664041567,
|
|
"eval_num_tokens": 1187561.0,
|
|
"eval_runtime": 2.0637,
|
|
"eval_samples_per_second": 1570.022,
|
|
"eval_steps_per_second": 196.253,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.202742099761963,
|
|
"epoch": 0.5145185939887926,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.7783,
|
|
"mean_token_accuracy": 0.11349713429808617,
|
|
"num_tokens": 1199108.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.226378297805786,
|
|
"epoch": 0.5196128374936322,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.8717,
|
|
"mean_token_accuracy": 0.10977188646793365,
|
|
"num_tokens": 1211944.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.088994693756104,
|
|
"epoch": 0.5247070809984717,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.7808,
|
|
"mean_token_accuracy": 0.10796541795134544,
|
|
"num_tokens": 1223044.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.2095095157623295,
|
|
"epoch": 0.5298013245033113,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.8041,
|
|
"mean_token_accuracy": 0.11091304048895836,
|
|
"num_tokens": 1235161.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.200789976119995,
|
|
"epoch": 0.5348955680081507,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.8577,
|
|
"mean_token_accuracy": 0.10809552669525146,
|
|
"num_tokens": 1247732.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.035283708572388,
|
|
"epoch": 0.5399898115129903,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.6934,
|
|
"mean_token_accuracy": 0.11823734119534493,
|
|
"num_tokens": 1259349.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.104032897949219,
|
|
"epoch": 0.5450840550178299,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.6594,
|
|
"mean_token_accuracy": 0.12075161710381507,
|
|
"num_tokens": 1269992.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.043303966522217,
|
|
"epoch": 0.5501782985226694,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7282,
|
|
"mean_token_accuracy": 0.11604061350226402,
|
|
"num_tokens": 1281502.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.078115224838257,
|
|
"epoch": 0.5552725420275089,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.6639,
|
|
"mean_token_accuracy": 0.11288660168647766,
|
|
"num_tokens": 1293880.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.100582504272461,
|
|
"epoch": 0.5603667855323484,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.7769,
|
|
"mean_token_accuracy": 0.1091226652264595,
|
|
"num_tokens": 1306022.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.05301866531372,
|
|
"epoch": 0.565461029037188,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7466,
|
|
"mean_token_accuracy": 0.11346042230725288,
|
|
"num_tokens": 1317939.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.991235208511353,
|
|
"epoch": 0.5705552725420275,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.12489392086863518,
|
|
"num_tokens": 1329917.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.060615587234497,
|
|
"epoch": 0.5756495160468671,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.7266,
|
|
"mean_token_accuracy": 0.11624963730573654,
|
|
"num_tokens": 1341733.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.117263460159302,
|
|
"epoch": 0.5807437595517065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.6608,
|
|
"mean_token_accuracy": 0.12135659679770469,
|
|
"num_tokens": 1352454.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.126049613952636,
|
|
"epoch": 0.5858380030565461,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8228,
|
|
"mean_token_accuracy": 0.11468813419342042,
|
|
"num_tokens": 1363878.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.052760887145996,
|
|
"epoch": 0.5909322465613857,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7401,
|
|
"mean_token_accuracy": 0.10709904357790948,
|
|
"num_tokens": 1375159.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.03482346534729,
|
|
"epoch": 0.5960264900662252,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.7567,
|
|
"mean_token_accuracy": 0.10899177491664887,
|
|
"num_tokens": 1386646.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.129480218887329,
|
|
"epoch": 0.6011207335710647,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.6824,
|
|
"mean_token_accuracy": 0.12052492946386337,
|
|
"num_tokens": 1398020.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.073828077316284,
|
|
"epoch": 0.6062149770759042,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.6439,
|
|
"mean_token_accuracy": 0.11667829379439354,
|
|
"num_tokens": 1408209.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.060064697265625,
|
|
"epoch": 0.6113092205807438,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.7209,
|
|
"mean_token_accuracy": 0.11234999671578408,
|
|
"num_tokens": 1420391.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.063613843917847,
|
|
"epoch": 0.6164034640855833,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.6879,
|
|
"mean_token_accuracy": 0.11753065437078476,
|
|
"num_tokens": 1432250.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.011401224136352,
|
|
"epoch": 0.6214977075904228,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.6715,
|
|
"mean_token_accuracy": 0.11426473781466484,
|
|
"num_tokens": 1444823.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.012602376937866,
|
|
"epoch": 0.6265919510952623,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.7273,
|
|
"mean_token_accuracy": 0.11627330929040909,
|
|
"num_tokens": 1455997.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.10994553565979,
|
|
"epoch": 0.6316861946001019,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.6835,
|
|
"mean_token_accuracy": 0.12411359027028084,
|
|
"num_tokens": 1467707.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.050820541381836,
|
|
"epoch": 0.6367804381049414,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7042,
|
|
"mean_token_accuracy": 0.1137191615998745,
|
|
"num_tokens": 1477920.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.073276948928833,
|
|
"epoch": 0.6418746816097809,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7249,
|
|
"mean_token_accuracy": 0.11637379750609397,
|
|
"num_tokens": 1488176.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.003222799301147,
|
|
"epoch": 0.6469689251146205,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.6709,
|
|
"mean_token_accuracy": 0.11933397352695466,
|
|
"num_tokens": 1499018.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.051417684555053,
|
|
"epoch": 0.65206316861946,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.722,
|
|
"mean_token_accuracy": 0.11735864207148552,
|
|
"num_tokens": 1510759.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.018922901153564,
|
|
"epoch": 0.6571574121242996,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7104,
|
|
"mean_token_accuracy": 0.11861589848995209,
|
|
"num_tokens": 1523028.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.0437907695770265,
|
|
"epoch": 0.6622516556291391,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7012,
|
|
"mean_token_accuracy": 0.11660568863153457,
|
|
"num_tokens": 1533505.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.063781309127807,
|
|
"epoch": 0.6673458991339786,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.735,
|
|
"mean_token_accuracy": 0.118368199467659,
|
|
"num_tokens": 1545366.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.021435022354126,
|
|
"epoch": 0.6724401426388181,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.597,
|
|
"mean_token_accuracy": 0.1193833477795124,
|
|
"num_tokens": 1557976.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.96804141998291,
|
|
"epoch": 0.6775343861436577,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7011,
|
|
"mean_token_accuracy": 0.11802663505077363,
|
|
"num_tokens": 1569822.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.956974935531616,
|
|
"epoch": 0.6826286296484972,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.6172,
|
|
"mean_token_accuracy": 0.12357318550348281,
|
|
"num_tokens": 1580485.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.067401933670044,
|
|
"epoch": 0.6877228731533367,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.6709,
|
|
"mean_token_accuracy": 0.11605887711048127,
|
|
"num_tokens": 1591865.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.967849397659302,
|
|
"epoch": 0.6928171166581762,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.5586,
|
|
"mean_token_accuracy": 0.1236411914229393,
|
|
"num_tokens": 1604580.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.95851092338562,
|
|
"epoch": 0.6979113601630158,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6996,
|
|
"mean_token_accuracy": 0.11606580689549446,
|
|
"num_tokens": 1615772.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.994662523269653,
|
|
"epoch": 0.7030056036678554,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.6126,
|
|
"mean_token_accuracy": 0.12819717302918435,
|
|
"num_tokens": 1626965.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.929726219177246,
|
|
"epoch": 0.7080998471726948,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6176,
|
|
"mean_token_accuracy": 0.12228650972247124,
|
|
"num_tokens": 1638865.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.930889701843261,
|
|
"epoch": 0.7131940906775344,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.5917,
|
|
"mean_token_accuracy": 0.12742585986852645,
|
|
"num_tokens": 1649200.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.945865154266357,
|
|
"epoch": 0.7182883341823739,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.5849,
|
|
"mean_token_accuracy": 0.12496745511889458,
|
|
"num_tokens": 1660726.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.071734762191772,
|
|
"epoch": 0.7233825776872135,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6231,
|
|
"mean_token_accuracy": 0.11918936520814896,
|
|
"num_tokens": 1672101.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.834320974349976,
|
|
"epoch": 0.7284768211920529,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.5853,
|
|
"mean_token_accuracy": 0.12185745760798454,
|
|
"num_tokens": 1683560.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.901132106781006,
|
|
"epoch": 0.7335710646968925,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.6349,
|
|
"mean_token_accuracy": 0.12105946019291877,
|
|
"num_tokens": 1694084.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.888202762603759,
|
|
"epoch": 0.738665308201732,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.6154,
|
|
"mean_token_accuracy": 0.12401428148150444,
|
|
"num_tokens": 1706686.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.007581329345703,
|
|
"epoch": 0.7437595517065716,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6901,
|
|
"mean_token_accuracy": 0.11879087910056114,
|
|
"num_tokens": 1719661.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.030799865722656,
|
|
"epoch": 0.7488537952114112,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.5694,
|
|
"mean_token_accuracy": 0.11836362406611442,
|
|
"num_tokens": 1730521.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.825137615203857,
|
|
"epoch": 0.7539480387162506,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.62,
|
|
"mean_token_accuracy": 0.12314270883798599,
|
|
"num_tokens": 1742251.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.852825355529785,
|
|
"epoch": 0.7590422822210902,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.4581,
|
|
"mean_token_accuracy": 0.12352288588881492,
|
|
"num_tokens": 1752871.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.885448026657104,
|
|
"epoch": 0.7641365257259297,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.6466,
|
|
"mean_token_accuracy": 0.12006721049547195,
|
|
"num_tokens": 1765690.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.866309595108032,
|
|
"epoch": 0.7692307692307693,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.12556443139910697,
|
|
"num_tokens": 1778398.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.937959241867065,
|
|
"epoch": 0.7743250127356087,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6207,
|
|
"mean_token_accuracy": 0.11847181618213654,
|
|
"num_tokens": 1791496.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.926975250244141,
|
|
"epoch": 0.7794192562404483,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6254,
|
|
"mean_token_accuracy": 0.11798580288887024,
|
|
"num_tokens": 1803152.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.861951160430908,
|
|
"epoch": 0.7845134997452878,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.11828219741582871,
|
|
"num_tokens": 1814775.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.800394582748413,
|
|
"epoch": 0.7896077432501274,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.5339,
|
|
"mean_token_accuracy": 0.11809298321604729,
|
|
"num_tokens": 1826068.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.898525381088257,
|
|
"epoch": 0.7947019867549668,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.5344,
|
|
"mean_token_accuracy": 0.1225415162742138,
|
|
"num_tokens": 1838115.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.940602111816406,
|
|
"epoch": 0.7997962302598064,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6306,
|
|
"mean_token_accuracy": 0.11780340895056725,
|
|
"num_tokens": 1850670.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.854982042312622,
|
|
"epoch": 0.804890473764646,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.616,
|
|
"mean_token_accuracy": 0.11739890798926353,
|
|
"num_tokens": 1863500.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.914686965942383,
|
|
"epoch": 0.8099847172694855,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6154,
|
|
"mean_token_accuracy": 0.12203172892332077,
|
|
"num_tokens": 1875840.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.85515456199646,
|
|
"epoch": 0.8150789607743251,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.5543,
|
|
"mean_token_accuracy": 0.12009180188179017,
|
|
"num_tokens": 1887265.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.828195285797119,
|
|
"epoch": 0.8201732042791645,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.4398,
|
|
"mean_token_accuracy": 0.1308925487101078,
|
|
"num_tokens": 1898550.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.684770154953003,
|
|
"epoch": 0.8252674477840041,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.3752,
|
|
"mean_token_accuracy": 0.13319009616971017,
|
|
"num_tokens": 1908990.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.816861534118653,
|
|
"epoch": 0.8303616912888436,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.4541,
|
|
"mean_token_accuracy": 0.12384122163057328,
|
|
"num_tokens": 1920109.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.798564386367798,
|
|
"epoch": 0.8354559347936832,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5351,
|
|
"mean_token_accuracy": 0.12183218970894813,
|
|
"num_tokens": 1930936.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.896174907684326,
|
|
"epoch": 0.8405501782985226,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.5257,
|
|
"mean_token_accuracy": 0.12270051911473275,
|
|
"num_tokens": 1940968.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.705681896209716,
|
|
"epoch": 0.8456444218033622,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.4705,
|
|
"mean_token_accuracy": 0.12962407395243644,
|
|
"num_tokens": 1952955.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.854102897644043,
|
|
"epoch": 0.8507386653082017,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5501,
|
|
"mean_token_accuracy": 0.1253214158117771,
|
|
"num_tokens": 1966023.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.853938150405884,
|
|
"epoch": 0.8558329088130413,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.5151,
|
|
"mean_token_accuracy": 0.12766359820961953,
|
|
"num_tokens": 1978036.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.737774801254273,
|
|
"epoch": 0.8609271523178808,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.4726,
|
|
"mean_token_accuracy": 0.1248572550714016,
|
|
"num_tokens": 1988870.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.853228282928467,
|
|
"epoch": 0.8660213958227203,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5448,
|
|
"mean_token_accuracy": 0.12996360063552856,
|
|
"num_tokens": 2000010.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.816134738922119,
|
|
"epoch": 0.8711156393275599,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.465,
|
|
"mean_token_accuracy": 0.12555269598960878,
|
|
"num_tokens": 2011127.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.795154285430908,
|
|
"epoch": 0.8762098828323994,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5618,
|
|
"mean_token_accuracy": 0.12125623747706413,
|
|
"num_tokens": 2023483.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.815837430953979,
|
|
"epoch": 0.8813041263372389,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.5735,
|
|
"mean_token_accuracy": 0.12030385881662368,
|
|
"num_tokens": 2034699.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.831874084472656,
|
|
"epoch": 0.8863983698420784,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.497,
|
|
"mean_token_accuracy": 0.11927786245942115,
|
|
"num_tokens": 2046849.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.8040472030639645,
|
|
"epoch": 0.891492613346918,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6073,
|
|
"mean_token_accuracy": 0.12338241189718246,
|
|
"num_tokens": 2060535.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.874739456176758,
|
|
"epoch": 0.8965868568517575,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.4684,
|
|
"mean_token_accuracy": 0.1270739935338497,
|
|
"num_tokens": 2072576.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.7486847877502445,
|
|
"epoch": 0.9016811003565971,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.4933,
|
|
"mean_token_accuracy": 0.12777445167303086,
|
|
"num_tokens": 2083884.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.787013149261474,
|
|
"epoch": 0.9067753438614365,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.5078,
|
|
"mean_token_accuracy": 0.12172376960515977,
|
|
"num_tokens": 2095429.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.852152156829834,
|
|
"epoch": 0.9118695873662761,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.5223,
|
|
"mean_token_accuracy": 0.11706596538424492,
|
|
"num_tokens": 2108351.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.851186800003052,
|
|
"epoch": 0.9169638308711157,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5591,
|
|
"mean_token_accuracy": 0.12036029696464538,
|
|
"num_tokens": 2120227.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.726887655258179,
|
|
"epoch": 0.9220580743759552,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.496,
|
|
"mean_token_accuracy": 0.12714903727173804,
|
|
"num_tokens": 2131624.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.783950710296631,
|
|
"epoch": 0.9271523178807947,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.4446,
|
|
"mean_token_accuracy": 0.1240805096924305,
|
|
"num_tokens": 2143523.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.688472318649292,
|
|
"epoch": 0.9322465613856342,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.4761,
|
|
"mean_token_accuracy": 0.1291765660047531,
|
|
"num_tokens": 2154952.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.781457567214966,
|
|
"epoch": 0.9373408048904738,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.4763,
|
|
"mean_token_accuracy": 0.12729405909776687,
|
|
"num_tokens": 2166407.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.729043579101562,
|
|
"epoch": 0.9424350483953133,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.4797,
|
|
"mean_token_accuracy": 0.12381153106689453,
|
|
"num_tokens": 2179748.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.845701742172241,
|
|
"epoch": 0.9475292919001528,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5444,
|
|
"mean_token_accuracy": 0.1277336798608303,
|
|
"num_tokens": 2191266.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.760814046859741,
|
|
"epoch": 0.9526235354049923,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.4946,
|
|
"mean_token_accuracy": 0.13033583462238313,
|
|
"num_tokens": 2203209.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.764568948745728,
|
|
"epoch": 0.9577177789098319,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4869,
|
|
"mean_token_accuracy": 0.12792996764183046,
|
|
"num_tokens": 2214693.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.672637605667115,
|
|
"epoch": 0.9628120224146715,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4045,
|
|
"mean_token_accuracy": 0.12818877920508384,
|
|
"num_tokens": 2226778.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.691420078277588,
|
|
"epoch": 0.9679062659195109,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.12146158292889594,
|
|
"num_tokens": 2238528.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.759699440002441,
|
|
"epoch": 0.9730005094243505,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5504,
|
|
"mean_token_accuracy": 0.12113381028175355,
|
|
"num_tokens": 2251483.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.650844717025757,
|
|
"epoch": 0.97809475292919,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.414,
|
|
"mean_token_accuracy": 0.1301756575703621,
|
|
"num_tokens": 2262151.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.661144399642945,
|
|
"epoch": 0.9831889964340296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.3917,
|
|
"mean_token_accuracy": 0.1330823615193367,
|
|
"num_tokens": 2274533.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.702968740463257,
|
|
"epoch": 0.9882832399388691,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4181,
|
|
"mean_token_accuracy": 0.13070132881402968,
|
|
"num_tokens": 2285518.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.772857713699341,
|
|
"epoch": 0.9933774834437086,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.501,
|
|
"mean_token_accuracy": 0.12216568812727928,
|
|
"num_tokens": 2297900.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.735703706741333,
|
|
"epoch": 0.9984717269485481,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.13494152277708055,
|
|
"num_tokens": 2311035.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.679241233401829,
|
|
"epoch": 1.0030565461029037,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.3098,
|
|
"mean_token_accuracy": 0.13246113310257593,
|
|
"num_tokens": 2321287.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.696215581893921,
|
|
"epoch": 1.0081507896077433,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.3415,
|
|
"mean_token_accuracy": 0.13118159845471383,
|
|
"num_tokens": 2333276.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.644043588638306,
|
|
"epoch": 1.0132450331125828,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2111,
|
|
"mean_token_accuracy": 0.13802278339862822,
|
|
"num_tokens": 2346554.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.599811792373657,
|
|
"epoch": 1.0183392766174224,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.2414,
|
|
"mean_token_accuracy": 0.13802444264292718,
|
|
"num_tokens": 2357564.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0183392766174224,
|
|
"eval_entropy": 6.431222215110873,
|
|
"eval_loss": 6.403883934020996,
|
|
"eval_mean_token_accuracy": 0.1330218936742088,
|
|
"eval_num_tokens": 2357564.0,
|
|
"eval_runtime": 2.0409,
|
|
"eval_samples_per_second": 1587.571,
|
|
"eval_steps_per_second": 198.446,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.5967113971710205,
|
|
"epoch": 1.0234335201222617,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999997711134041,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.13662363216280937,
|
|
"num_tokens": 2369749.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.6823607921600345,
|
|
"epoch": 1.0285277636271013,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999988412624068,
|
|
"loss": 6.3325,
|
|
"mean_token_accuracy": 0.1326715439558029,
|
|
"num_tokens": 2382622.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.610849332809448,
|
|
"epoch": 1.0336220071319409,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999971961445492,
|
|
"loss": 6.3184,
|
|
"mean_token_accuracy": 0.12889624014496803,
|
|
"num_tokens": 2394749.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.70923490524292,
|
|
"epoch": 1.0387162506367804,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004999948357650613,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.12995868176221848,
|
|
"num_tokens": 2407868.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.561355447769165,
|
|
"epoch": 1.04381049414162,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999917601314468,
|
|
"loss": 6.228,
|
|
"mean_token_accuracy": 0.1391261026263237,
|
|
"num_tokens": 2419045.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.575577259063721,
|
|
"epoch": 1.0489047376464595,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004999879692534828,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.12873076125979424,
|
|
"num_tokens": 2431697.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.686066818237305,
|
|
"epoch": 1.053998981151299,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999834631432208,
|
|
"loss": 6.3127,
|
|
"mean_token_accuracy": 0.13791258484125138,
|
|
"num_tokens": 2442579.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.544984340667725,
|
|
"epoch": 1.0590932246561386,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999782418149852,
|
|
"loss": 6.1506,
|
|
"mean_token_accuracy": 0.14760711789131165,
|
|
"num_tokens": 2453601.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.5693583488464355,
|
|
"epoch": 1.064187468160978,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999723052853747,
|
|
"loss": 6.184,
|
|
"mean_token_accuracy": 0.1383779227733612,
|
|
"num_tokens": 2465500.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.495132970809936,
|
|
"epoch": 1.0692817116658175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999656535732615,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.13339439705014228,
|
|
"num_tokens": 2477948.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.540082406997681,
|
|
"epoch": 1.074375955170657,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999582866997913,
|
|
"loss": 6.2464,
|
|
"mean_token_accuracy": 0.14291542023420334,
|
|
"num_tokens": 2489575.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.657188940048218,
|
|
"epoch": 1.0794701986754967,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499950204688383,
|
|
"loss": 6.2469,
|
|
"mean_token_accuracy": 0.13168998658657075,
|
|
"num_tokens": 2501615.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.490920352935791,
|
|
"epoch": 1.0845644421803362,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999414075647292,
|
|
"loss": 6.241,
|
|
"mean_token_accuracy": 0.13728386163711548,
|
|
"num_tokens": 2513164.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.619253206253052,
|
|
"epoch": 1.0896586856851758,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000499931895356796,
|
|
"loss": 6.3418,
|
|
"mean_token_accuracy": 0.13567643836140633,
|
|
"num_tokens": 2524349.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.5358747959136965,
|
|
"epoch": 1.0947529291900153,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999216680948223,
|
|
"loss": 6.1957,
|
|
"mean_token_accuracy": 0.13942680433392524,
|
|
"num_tokens": 2535834.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.475406217575073,
|
|
"epoch": 1.0998471726948549,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999107258113204,
|
|
"loss": 6.134,
|
|
"mean_token_accuracy": 0.13922262340784072,
|
|
"num_tokens": 2547690.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.5090185642242435,
|
|
"epoch": 1.1049414161996944,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004998990685410759,
|
|
"loss": 6.204,
|
|
"mean_token_accuracy": 0.14063166007399558,
|
|
"num_tokens": 2559249.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.592030763626099,
|
|
"epoch": 1.110035659704534,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998866963211465,
|
|
"loss": 6.3258,
|
|
"mean_token_accuracy": 0.1333697982132435,
|
|
"num_tokens": 2573014.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.544809436798095,
|
|
"epoch": 1.1151299032093733,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998736091908636,
|
|
"loss": 6.2155,
|
|
"mean_token_accuracy": 0.13714805468916894,
|
|
"num_tokens": 2584720.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.537558078765869,
|
|
"epoch": 1.1202241467142129,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998598071918309,
|
|
"loss": 6.2286,
|
|
"mean_token_accuracy": 0.14043956696987153,
|
|
"num_tokens": 2596711.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.607815361022949,
|
|
"epoch": 1.1253183902190524,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998452903679246,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.13805221170186996,
|
|
"num_tokens": 2606709.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.417927694320679,
|
|
"epoch": 1.130412633723892,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998300587652936,
|
|
"loss": 6.2147,
|
|
"mean_token_accuracy": 0.14031558632850646,
|
|
"num_tokens": 2617233.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.602714776992798,
|
|
"epoch": 1.1355068772287316,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004998141124323586,
|
|
"loss": 6.1923,
|
|
"mean_token_accuracy": 0.14089069664478301,
|
|
"num_tokens": 2628313.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.396322965621948,
|
|
"epoch": 1.1406011207335711,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997974514198126,
|
|
"loss": 6.1257,
|
|
"mean_token_accuracy": 0.14331114217638968,
|
|
"num_tokens": 2639198.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.575943279266357,
|
|
"epoch": 1.1456953642384107,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997800757806209,
|
|
"loss": 6.2225,
|
|
"mean_token_accuracy": 0.13633955642580986,
|
|
"num_tokens": 2650914.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.428630304336548,
|
|
"epoch": 1.15078960774325,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004997619855700203,
|
|
"loss": 6.1988,
|
|
"mean_token_accuracy": 0.14106837585568427,
|
|
"num_tokens": 2664335.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.554105043411255,
|
|
"epoch": 1.1558838512480896,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499743180845519,
|
|
"loss": 6.2681,
|
|
"mean_token_accuracy": 0.1395036794245243,
|
|
"num_tokens": 2675839.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.326878070831299,
|
|
"epoch": 1.1609780947529291,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499723661666897,
|
|
"loss": 6.0571,
|
|
"mean_token_accuracy": 0.1559303343296051,
|
|
"num_tokens": 2686793.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.5064756870269775,
|
|
"epoch": 1.1660723382577687,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004997034280962054,
|
|
"loss": 6.2556,
|
|
"mean_token_accuracy": 0.13475740030407907,
|
|
"num_tokens": 2698504.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.4849590301513675,
|
|
"epoch": 1.1711665817626082,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004996824801977663,
|
|
"loss": 6.1637,
|
|
"mean_token_accuracy": 0.1421566501259804,
|
|
"num_tokens": 2710678.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.533821296691895,
|
|
"epoch": 1.1762608252674478,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996608180381727,
|
|
"loss": 6.2663,
|
|
"mean_token_accuracy": 0.14090365990996362,
|
|
"num_tokens": 2722644.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.464822435379029,
|
|
"epoch": 1.1813550687722874,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499638441686288,
|
|
"loss": 6.1544,
|
|
"mean_token_accuracy": 0.1428435891866684,
|
|
"num_tokens": 2735279.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.44498028755188,
|
|
"epoch": 1.186449312277127,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004996153512132462,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.14193697571754454,
|
|
"num_tokens": 2746905.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.569237184524536,
|
|
"epoch": 1.1915435557819665,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004995915466924517,
|
|
"loss": 6.2616,
|
|
"mean_token_accuracy": 0.13707489892840385,
|
|
"num_tokens": 2760081.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.495239686965943,
|
|
"epoch": 1.196637799286806,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995670281995784,
|
|
"loss": 6.2122,
|
|
"mean_token_accuracy": 0.13570564985275269,
|
|
"num_tokens": 2771416.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.555169725418091,
|
|
"epoch": 1.2017320427916454,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004995417958125703,
|
|
"loss": 6.3241,
|
|
"mean_token_accuracy": 0.1286981664597988,
|
|
"num_tokens": 2784216.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.556367540359497,
|
|
"epoch": 1.206826286296485,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004995158496116402,
|
|
"loss": 6.2914,
|
|
"mean_token_accuracy": 0.13823686838150023,
|
|
"num_tokens": 2795706.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.51031346321106,
|
|
"epoch": 1.2119205298013245,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004994891896792712,
|
|
"loss": 6.2255,
|
|
"mean_token_accuracy": 0.14145646169781684,
|
|
"num_tokens": 2806317.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.583649206161499,
|
|
"epoch": 1.217014773306164,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004994618161002142,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.13678759709000587,
|
|
"num_tokens": 2817432.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.558048677444458,
|
|
"epoch": 1.2221090168110036,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004994337289614893,
|
|
"loss": 6.1765,
|
|
"mean_token_accuracy": 0.14228238984942437,
|
|
"num_tokens": 2828145.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.402686882019043,
|
|
"epoch": 1.2272032603158431,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994049283523854,
|
|
"loss": 6.1451,
|
|
"mean_token_accuracy": 0.15038926303386688,
|
|
"num_tokens": 2838962.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.473277854919433,
|
|
"epoch": 1.2322975038206827,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004993754143644585,
|
|
"loss": 6.1926,
|
|
"mean_token_accuracy": 0.14219142869114876,
|
|
"num_tokens": 2850941.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.456243801116943,
|
|
"epoch": 1.237391747325522,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004993451870915333,
|
|
"loss": 6.1468,
|
|
"mean_token_accuracy": 0.1412370204925537,
|
|
"num_tokens": 2863265.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.4249999046325685,
|
|
"epoch": 1.2424859908303616,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993142466297016,
|
|
"loss": 6.1665,
|
|
"mean_token_accuracy": 0.13820707872509957,
|
|
"num_tokens": 2876235.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.4690908908844,
|
|
"epoch": 1.2475802343352012,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004992825930773228,
|
|
"loss": 6.2146,
|
|
"mean_token_accuracy": 0.1370810180902481,
|
|
"num_tokens": 2889518.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.469319200515747,
|
|
"epoch": 1.2526744778400407,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004992502265350224,
|
|
"loss": 6.2284,
|
|
"mean_token_accuracy": 0.14134810566902162,
|
|
"num_tokens": 2902417.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.423447751998902,
|
|
"epoch": 1.2577687213448803,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004992171471056933,
|
|
"loss": 6.1394,
|
|
"mean_token_accuracy": 0.1492676056921482,
|
|
"num_tokens": 2912904.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.4873619556427,
|
|
"epoch": 1.2628629648497198,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004991833548944945,
|
|
"loss": 6.1768,
|
|
"mean_token_accuracy": 0.1468370273709297,
|
|
"num_tokens": 2925583.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.46749906539917,
|
|
"epoch": 1.2679572083545594,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991488500088505,
|
|
"loss": 6.1156,
|
|
"mean_token_accuracy": 0.14610635489225388,
|
|
"num_tokens": 2937916.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.421872282028199,
|
|
"epoch": 1.273051451859399,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004991136325584516,
|
|
"loss": 6.1559,
|
|
"mean_token_accuracy": 0.14431013092398642,
|
|
"num_tokens": 2949583.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.373481225967407,
|
|
"epoch": 1.2781456953642385,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990777026552536,
|
|
"loss": 6.1462,
|
|
"mean_token_accuracy": 0.13999590128660203,
|
|
"num_tokens": 2962044.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.590294075012207,
|
|
"epoch": 1.283239938869078,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004990410604134769,
|
|
"loss": 6.3187,
|
|
"mean_token_accuracy": 0.13254853561520577,
|
|
"num_tokens": 2973570.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.443555641174316,
|
|
"epoch": 1.2883341823739174,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004990037059496064,
|
|
"loss": 6.1888,
|
|
"mean_token_accuracy": 0.13860231786966323,
|
|
"num_tokens": 2983683.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.5187116146087645,
|
|
"epoch": 1.293428425878757,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004989656393823913,
|
|
"loss": 6.2225,
|
|
"mean_token_accuracy": 0.13842222839593887,
|
|
"num_tokens": 2995100.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.389102649688721,
|
|
"epoch": 1.2985226693835965,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004989268608328443,
|
|
"loss": 6.2357,
|
|
"mean_token_accuracy": 0.13710789233446122,
|
|
"num_tokens": 3006887.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.546328496932984,
|
|
"epoch": 1.303616912888436,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004988873704242418,
|
|
"loss": 6.1966,
|
|
"mean_token_accuracy": 0.13944068923592567,
|
|
"num_tokens": 3020444.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.4080541133880615,
|
|
"epoch": 1.3087111563932756,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004988471682821229,
|
|
"loss": 6.1348,
|
|
"mean_token_accuracy": 0.1399799458682537,
|
|
"num_tokens": 3032487.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.4733987808227536,
|
|
"epoch": 1.3138053998981152,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004988062545342894,
|
|
"loss": 6.2313,
|
|
"mean_token_accuracy": 0.14029935523867607,
|
|
"num_tokens": 3044652.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.406413459777832,
|
|
"epoch": 1.3188996434029547,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004987646293108056,
|
|
"loss": 6.1232,
|
|
"mean_token_accuracy": 0.14590274095535277,
|
|
"num_tokens": 3056252.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.361226367950439,
|
|
"epoch": 1.323993886907794,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004987222927439967,
|
|
"loss": 6.1334,
|
|
"mean_token_accuracy": 0.14939191639423371,
|
|
"num_tokens": 3066959.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.506962156295776,
|
|
"epoch": 1.3290881304126336,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986792449684504,
|
|
"loss": 6.2415,
|
|
"mean_token_accuracy": 0.1412517547607422,
|
|
"num_tokens": 3079113.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.437981748580933,
|
|
"epoch": 1.3341823739174732,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004986354861210144,
|
|
"loss": 6.1687,
|
|
"mean_token_accuracy": 0.14122130796313287,
|
|
"num_tokens": 3090290.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.435575103759765,
|
|
"epoch": 1.3392766174223127,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004985910163407973,
|
|
"loss": 6.1055,
|
|
"mean_token_accuracy": 0.1450235664844513,
|
|
"num_tokens": 3101031.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.401088809967041,
|
|
"epoch": 1.3443708609271523,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004985458357691675,
|
|
"loss": 6.1951,
|
|
"mean_token_accuracy": 0.14005887359380723,
|
|
"num_tokens": 3113371.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.431767082214355,
|
|
"epoch": 1.3494651044319919,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004984999445497533,
|
|
"loss": 6.1707,
|
|
"mean_token_accuracy": 0.14353958070278167,
|
|
"num_tokens": 3125167.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.420112466812133,
|
|
"epoch": 1.3545593479368314,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000498453342828442,
|
|
"loss": 6.1993,
|
|
"mean_token_accuracy": 0.14175814837217332,
|
|
"num_tokens": 3136945.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.4782510757446286,
|
|
"epoch": 1.359653591441671,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004984060307533794,
|
|
"loss": 6.1237,
|
|
"mean_token_accuracy": 0.1451432079076767,
|
|
"num_tokens": 3147435.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.4224035263061525,
|
|
"epoch": 1.3647478349465105,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004983580084749698,
|
|
"loss": 6.1504,
|
|
"mean_token_accuracy": 0.1486922547221184,
|
|
"num_tokens": 3157693.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.4051799297332765,
|
|
"epoch": 1.36984207845135,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000498309276145875,
|
|
"loss": 6.0883,
|
|
"mean_token_accuracy": 0.14632491916418075,
|
|
"num_tokens": 3169561.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.3754946231842045,
|
|
"epoch": 1.3749363219561896,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982598339210142,
|
|
"loss": 6.2784,
|
|
"mean_token_accuracy": 0.13376625180244445,
|
|
"num_tokens": 3183142.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.503082227706909,
|
|
"epoch": 1.380030565461029,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004982096819575631,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.13932920768857002,
|
|
"num_tokens": 3195703.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.311429500579834,
|
|
"epoch": 1.3851248089658685,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004981588204149538,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.1440253436565399,
|
|
"num_tokens": 3206603.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.404092025756836,
|
|
"epoch": 1.390219052470708,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004981072494548743,
|
|
"loss": 6.1948,
|
|
"mean_token_accuracy": 0.1399855472147465,
|
|
"num_tokens": 3218909.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.466204738616943,
|
|
"epoch": 1.3953132959755477,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004980549692412676,
|
|
"loss": 6.1223,
|
|
"mean_token_accuracy": 0.147786183655262,
|
|
"num_tokens": 3230252.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.365512180328369,
|
|
"epoch": 1.4004075394803872,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004980019799403313,
|
|
"loss": 6.1815,
|
|
"mean_token_accuracy": 0.14454519003629684,
|
|
"num_tokens": 3241227.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.396402311325073,
|
|
"epoch": 1.4055017829852268,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004979482817205176,
|
|
"loss": 6.0382,
|
|
"mean_token_accuracy": 0.1510227859020233,
|
|
"num_tokens": 3253569.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.410557460784912,
|
|
"epoch": 1.410596026490066,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004978938747525319,
|
|
"loss": 6.1376,
|
|
"mean_token_accuracy": 0.14222131669521332,
|
|
"num_tokens": 3265371.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.4386780738830565,
|
|
"epoch": 1.4156902699949057,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004978387592093329,
|
|
"loss": 6.1589,
|
|
"mean_token_accuracy": 0.1422835037112236,
|
|
"num_tokens": 3276612.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.3645223617553714,
|
|
"epoch": 1.4207845134997452,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004977829352661318,
|
|
"loss": 6.1789,
|
|
"mean_token_accuracy": 0.14313849434256554,
|
|
"num_tokens": 3288847.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.322448587417602,
|
|
"epoch": 1.4258787570045848,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004977264031003919,
|
|
"loss": 6.1331,
|
|
"mean_token_accuracy": 0.14431178718805313,
|
|
"num_tokens": 3300738.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.37766432762146,
|
|
"epoch": 1.4309730005094243,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004976691628918278,
|
|
"loss": 6.1221,
|
|
"mean_token_accuracy": 0.14262940138578414,
|
|
"num_tokens": 3312369.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.376084899902343,
|
|
"epoch": 1.4360672440142639,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000497611214822405,
|
|
"loss": 6.1302,
|
|
"mean_token_accuracy": 0.14152591601014136,
|
|
"num_tokens": 3325277.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.312879371643066,
|
|
"epoch": 1.4411614875191034,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004975525590763393,
|
|
"loss": 6.0101,
|
|
"mean_token_accuracy": 0.15388279408216476,
|
|
"num_tokens": 3335881.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.452538871765137,
|
|
"epoch": 1.446255731023943,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497493195840096,
|
|
"loss": 6.1671,
|
|
"mean_token_accuracy": 0.14172034859657287,
|
|
"num_tokens": 3348942.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.370132780075073,
|
|
"epoch": 1.4513499745287826,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004974331253023898,
|
|
"loss": 6.1682,
|
|
"mean_token_accuracy": 0.1442234508693218,
|
|
"num_tokens": 3361890.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.410511493682861,
|
|
"epoch": 1.4564442180336221,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004973723476541836,
|
|
"loss": 6.1372,
|
|
"mean_token_accuracy": 0.14195466041564941,
|
|
"num_tokens": 3373297.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.244622659683228,
|
|
"epoch": 1.4615384615384617,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004973108630886885,
|
|
"loss": 6.0327,
|
|
"mean_token_accuracy": 0.14957051649689673,
|
|
"num_tokens": 3384623.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.282334423065185,
|
|
"epoch": 1.466632705043301,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004972486718013626,
|
|
"loss": 6.1101,
|
|
"mean_token_accuracy": 0.1433366797864437,
|
|
"num_tokens": 3398379.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.443861722946167,
|
|
"epoch": 1.4717269485481406,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004971857739899109,
|
|
"loss": 6.1147,
|
|
"mean_token_accuracy": 0.13918179646134377,
|
|
"num_tokens": 3410928.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.298733854293824,
|
|
"epoch": 1.4768211920529801,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004971221698542842,
|
|
"loss": 6.0568,
|
|
"mean_token_accuracy": 0.15082672387361526,
|
|
"num_tokens": 3422173.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.286819505691528,
|
|
"epoch": 1.4819154355578197,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004970578595966789,
|
|
"loss": 6.1095,
|
|
"mean_token_accuracy": 0.14911753311753273,
|
|
"num_tokens": 3434341.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.3674551486969,
|
|
"epoch": 1.4870096790626592,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004969928434215358,
|
|
"loss": 6.0905,
|
|
"mean_token_accuracy": 0.14403041228652,
|
|
"num_tokens": 3445605.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.396696519851685,
|
|
"epoch": 1.4921039225674988,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004969271215355402,
|
|
"loss": 6.0867,
|
|
"mean_token_accuracy": 0.14174341931939125,
|
|
"num_tokens": 3456957.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.337380743026733,
|
|
"epoch": 1.4971981660723381,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004968606941476207,
|
|
"loss": 6.1026,
|
|
"mean_token_accuracy": 0.14822276681661606,
|
|
"num_tokens": 3468340.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.369459438323974,
|
|
"epoch": 1.5022924095771777,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004967935614689485,
|
|
"loss": 6.0964,
|
|
"mean_token_accuracy": 0.14861045107245446,
|
|
"num_tokens": 3479137.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.2834419250488285,
|
|
"epoch": 1.5073866530820172,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004967257237129373,
|
|
"loss": 5.9759,
|
|
"mean_token_accuracy": 0.15439686700701713,
|
|
"num_tokens": 3490525.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.234587955474853,
|
|
"epoch": 1.5124808965868568,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004966571810952417,
|
|
"loss": 5.9747,
|
|
"mean_token_accuracy": 0.15747153908014297,
|
|
"num_tokens": 3500731.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.3374734878540036,
|
|
"epoch": 1.5175751400916964,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004965879338337577,
|
|
"loss": 6.0687,
|
|
"mean_token_accuracy": 0.14279309958219527,
|
|
"num_tokens": 3512944.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.316036939620972,
|
|
"epoch": 1.522669383596536,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004965179821486205,
|
|
"loss": 6.0431,
|
|
"mean_token_accuracy": 0.1427792452275753,
|
|
"num_tokens": 3524110.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.204104280471801,
|
|
"epoch": 1.5277636271013755,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004964473262622054,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.15044626891613005,
|
|
"num_tokens": 3536162.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.5277636271013755,
|
|
"eval_entropy": 6.162457740454026,
|
|
"eval_loss": 6.133503437042236,
|
|
"eval_mean_token_accuracy": 0.1476598968476425,
|
|
"eval_num_tokens": 3536162.0,
|
|
"eval_runtime": 2.0431,
|
|
"eval_samples_per_second": 1585.789,
|
|
"eval_steps_per_second": 198.224,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9810,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 833526943580160.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|