d4983c5354
Model: fpadovani/isl-latn-10mb-after-ppt-Dp-10mb-ckpt500_seed3407 Source: Original Platform
1046 lines
28 KiB
JSON
1046 lines
28 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.5094243504839532,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 282752160399360.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|