2119ed983f
Model: fpadovani/urd-arab-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
7112 lines
193 KiB
JSON
7112 lines
193 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.1518386714116255,
|
|
"eval_steps": 500,
|
|
"global_step": 3500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.805317306518555,
|
|
"epoch": 0.005931198102016607,
|
|
"grad_norm": 16.25,
|
|
"learning_rate": 2e-06,
|
|
"loss": 14.4312,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12647.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.825564956665039,
|
|
"epoch": 0.011862396204033215,
|
|
"grad_norm": 16.875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 14.4169,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 26668.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.845508098602295,
|
|
"epoch": 0.017793594306049824,
|
|
"grad_norm": 24.75,
|
|
"learning_rate": 7e-06,
|
|
"loss": 14.1564,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 41021.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 5.1169312477111815,
|
|
"epoch": 0.02372479240806643,
|
|
"grad_norm": 33.75,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.5366,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 56626.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 7.628379774093628,
|
|
"epoch": 0.02965599051008304,
|
|
"grad_norm": 15.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 11.8077,
|
|
"mean_token_accuracy": 0.00010419512109365315,
|
|
"num_tokens": 72914.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.455063152313233,
|
|
"epoch": 0.03558718861209965,
|
|
"grad_norm": 3.21875,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.841,
|
|
"mean_token_accuracy": 0.0054948762990534306,
|
|
"num_tokens": 88514.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.646884059906006,
|
|
"epoch": 0.04151838671411625,
|
|
"grad_norm": 3.03125,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.5696,
|
|
"mean_token_accuracy": 0.01355562973767519,
|
|
"num_tokens": 101256.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.619774627685548,
|
|
"epoch": 0.04744958481613286,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.282,
|
|
"mean_token_accuracy": 0.012707393150776625,
|
|
"num_tokens": 114459.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.61210651397705,
|
|
"epoch": 0.05338078291814947,
|
|
"grad_norm": 2.28125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0287,
|
|
"mean_token_accuracy": 0.01491355374455452,
|
|
"num_tokens": 129858.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.597074794769288,
|
|
"epoch": 0.05931198102016608,
|
|
"grad_norm": 2.1875,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.8185,
|
|
"mean_token_accuracy": 0.0331686332821846,
|
|
"num_tokens": 143486.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.579236507415771,
|
|
"epoch": 0.06524317912218268,
|
|
"grad_norm": 1.9609375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.6912,
|
|
"mean_token_accuracy": 0.035220497474074366,
|
|
"num_tokens": 156003.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.577684020996093,
|
|
"epoch": 0.0711743772241993,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.6506,
|
|
"mean_token_accuracy": 0.034727090038359165,
|
|
"num_tokens": 170988.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.5711012840271,
|
|
"epoch": 0.0771055753262159,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.5328,
|
|
"mean_token_accuracy": 0.03448135014623403,
|
|
"num_tokens": 187165.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.558809852600097,
|
|
"epoch": 0.0830367734282325,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.4798,
|
|
"mean_token_accuracy": 0.03228826243430376,
|
|
"num_tokens": 202072.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.549299335479736,
|
|
"epoch": 0.08896797153024912,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.3281,
|
|
"mean_token_accuracy": 0.034324941039085385,
|
|
"num_tokens": 215375.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.532067012786865,
|
|
"epoch": 0.09489916963226572,
|
|
"grad_norm": 1.734375,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.2492,
|
|
"mean_token_accuracy": 0.03353528436273336,
|
|
"num_tokens": 228749.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.505972194671632,
|
|
"epoch": 0.10083036773428232,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.1494,
|
|
"mean_token_accuracy": 0.03314766474068165,
|
|
"num_tokens": 242387.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.478415584564209,
|
|
"epoch": 0.10676156583629894,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0705,
|
|
"mean_token_accuracy": 0.032026905380189416,
|
|
"num_tokens": 257793.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.454620742797852,
|
|
"epoch": 0.11269276393831554,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9712,
|
|
"mean_token_accuracy": 0.0325484286993742,
|
|
"num_tokens": 271187.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.40716381072998,
|
|
"epoch": 0.11862396204033215,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8474,
|
|
"mean_token_accuracy": 0.03217122312635183,
|
|
"num_tokens": 284937.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.335246372222901,
|
|
"epoch": 0.12455516014234876,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6893,
|
|
"mean_token_accuracy": 0.03255892116576433,
|
|
"num_tokens": 300108.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.246189594268799,
|
|
"epoch": 0.13048635824436536,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.5681,
|
|
"mean_token_accuracy": 0.031065495498478413,
|
|
"num_tokens": 316760.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.13212080001831,
|
|
"epoch": 0.13641755634638197,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3819,
|
|
"mean_token_accuracy": 0.03049417305737734,
|
|
"num_tokens": 329705.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.97801332473755,
|
|
"epoch": 0.1423487544483986,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.244,
|
|
"mean_token_accuracy": 0.03179512321949005,
|
|
"num_tokens": 342947.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.771716022491455,
|
|
"epoch": 0.14827995255041518,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.0865,
|
|
"mean_token_accuracy": 0.03442887142300606,
|
|
"num_tokens": 358397.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.533180236816406,
|
|
"epoch": 0.1542111506524318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9643,
|
|
"mean_token_accuracy": 0.03413730468600988,
|
|
"num_tokens": 370369.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.256175804138184,
|
|
"epoch": 0.1601423487544484,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.8421,
|
|
"mean_token_accuracy": 0.03181219045072794,
|
|
"num_tokens": 386896.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 8.96011724472046,
|
|
"epoch": 0.166073546856465,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.6981,
|
|
"mean_token_accuracy": 0.03533113058656454,
|
|
"num_tokens": 400433.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.675125217437744,
|
|
"epoch": 0.17200474495848161,
|
|
"grad_norm": 0.7890625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.6507,
|
|
"mean_token_accuracy": 0.030981262400746345,
|
|
"num_tokens": 415092.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.432178497314453,
|
|
"epoch": 0.17793594306049823,
|
|
"grad_norm": 0.7109375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.5489,
|
|
"mean_token_accuracy": 0.033769184909760955,
|
|
"num_tokens": 429793.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.235133266448974,
|
|
"epoch": 0.18386714116251482,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.5575,
|
|
"mean_token_accuracy": 0.03205921929329634,
|
|
"num_tokens": 442618.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.142054271697997,
|
|
"epoch": 0.18979833926453143,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.4449,
|
|
"mean_token_accuracy": 0.03399460650980472,
|
|
"num_tokens": 454976.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.025918197631835,
|
|
"epoch": 0.19572953736654805,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5037,
|
|
"mean_token_accuracy": 0.03331909943372011,
|
|
"num_tokens": 467523.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 7.9379983901977536,
|
|
"epoch": 0.20166073546856464,
|
|
"grad_norm": 0.6796875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.4373,
|
|
"mean_token_accuracy": 0.03239843733608723,
|
|
"num_tokens": 481391.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 7.869046401977539,
|
|
"epoch": 0.20759193357058126,
|
|
"grad_norm": 0.67578125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.4399,
|
|
"mean_token_accuracy": 0.0311466746032238,
|
|
"num_tokens": 496697.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 7.866212034225464,
|
|
"epoch": 0.21352313167259787,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.4743,
|
|
"mean_token_accuracy": 0.027226756513118743,
|
|
"num_tokens": 510091.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 7.788731241226197,
|
|
"epoch": 0.21945432977461446,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.426,
|
|
"mean_token_accuracy": 0.03152833003550768,
|
|
"num_tokens": 522168.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.754296016693115,
|
|
"epoch": 0.22538552787663108,
|
|
"grad_norm": 0.71875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3943,
|
|
"mean_token_accuracy": 0.03427231907844543,
|
|
"num_tokens": 535853.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.75240159034729,
|
|
"epoch": 0.2313167259786477,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4255,
|
|
"mean_token_accuracy": 0.03329353462904692,
|
|
"num_tokens": 551082.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.700812101364136,
|
|
"epoch": 0.2372479240806643,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4629,
|
|
"mean_token_accuracy": 0.035965221747756,
|
|
"num_tokens": 567181.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.696683025360107,
|
|
"epoch": 0.2431791221826809,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.4112,
|
|
"mean_token_accuracy": 0.035206197388470174,
|
|
"num_tokens": 580689.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.712547016143799,
|
|
"epoch": 0.2491103202846975,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3345,
|
|
"mean_token_accuracy": 0.03781173955649138,
|
|
"num_tokens": 594590.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.769494247436524,
|
|
"epoch": 0.25504151838671413,
|
|
"grad_norm": 10.5,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.3905,
|
|
"mean_token_accuracy": 0.03799592889845371,
|
|
"num_tokens": 610577.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.6959569454193115,
|
|
"epoch": 0.2609727164887307,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.384,
|
|
"mean_token_accuracy": 0.037612661719322205,
|
|
"num_tokens": 627066.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.781623411178589,
|
|
"epoch": 0.2669039145907473,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.3672,
|
|
"mean_token_accuracy": 0.04785094037652016,
|
|
"num_tokens": 642601.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.77043514251709,
|
|
"epoch": 0.27283511269276395,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.2531,
|
|
"mean_token_accuracy": 0.05052996054291725,
|
|
"num_tokens": 655615.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.747775220870972,
|
|
"epoch": 0.27876631079478054,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2647,
|
|
"mean_token_accuracy": 0.05221507623791695,
|
|
"num_tokens": 670218.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.743488121032715,
|
|
"epoch": 0.2846975088967972,
|
|
"grad_norm": 0.75390625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.2593,
|
|
"mean_token_accuracy": 0.05101203434169292,
|
|
"num_tokens": 683661.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.713605117797852,
|
|
"epoch": 0.29062870699881377,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2135,
|
|
"mean_token_accuracy": 0.05287150628864765,
|
|
"num_tokens": 699767.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.695834493637085,
|
|
"epoch": 0.29655990510083036,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.1762,
|
|
"mean_token_accuracy": 0.05634650327265263,
|
|
"num_tokens": 713579.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.629615783691406,
|
|
"epoch": 0.302491103202847,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.1678,
|
|
"mean_token_accuracy": 0.05482946485280991,
|
|
"num_tokens": 727490.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.607522439956665,
|
|
"epoch": 0.3084223013048636,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.2348,
|
|
"mean_token_accuracy": 0.05509466864168644,
|
|
"num_tokens": 742477.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.624451637268066,
|
|
"epoch": 0.3143534994068802,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.113,
|
|
"mean_token_accuracy": 0.05751051269471645,
|
|
"num_tokens": 757772.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.647220134735107,
|
|
"epoch": 0.3202846975088968,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.1596,
|
|
"mean_token_accuracy": 0.05049732141196728,
|
|
"num_tokens": 772898.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.600799989700318,
|
|
"epoch": 0.3262158956109134,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.1154,
|
|
"mean_token_accuracy": 0.05433249212801457,
|
|
"num_tokens": 788395.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.563078927993774,
|
|
"epoch": 0.33214709371293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.1648,
|
|
"mean_token_accuracy": 0.06050356179475784,
|
|
"num_tokens": 803413.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.517913770675659,
|
|
"epoch": 0.33807829181494664,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.0248,
|
|
"mean_token_accuracy": 0.06933637075126171,
|
|
"num_tokens": 817775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.566510915756226,
|
|
"epoch": 0.34400948991696323,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.1421,
|
|
"mean_token_accuracy": 0.06264986135065556,
|
|
"num_tokens": 831379.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.551775217056274,
|
|
"epoch": 0.3499406880189798,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.0759,
|
|
"mean_token_accuracy": 0.05862759165465832,
|
|
"num_tokens": 846596.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.553944540023804,
|
|
"epoch": 0.35587188612099646,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.0666,
|
|
"mean_token_accuracy": 0.05912046544253826,
|
|
"num_tokens": 862063.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.582914543151856,
|
|
"epoch": 0.36180308422301305,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.9961,
|
|
"mean_token_accuracy": 0.06986662931740284,
|
|
"num_tokens": 876841.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.488122177124024,
|
|
"epoch": 0.36773428232502964,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.0603,
|
|
"mean_token_accuracy": 0.06794393062591553,
|
|
"num_tokens": 890728.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.442899990081787,
|
|
"epoch": 0.3736654804270463,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.983,
|
|
"mean_token_accuracy": 0.0702366728335619,
|
|
"num_tokens": 905599.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.451259899139404,
|
|
"epoch": 0.37959667852906287,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.9367,
|
|
"mean_token_accuracy": 0.07826149612665176,
|
|
"num_tokens": 918139.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.405993700027466,
|
|
"epoch": 0.38552787663107946,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.9943,
|
|
"mean_token_accuracy": 0.0733895305544138,
|
|
"num_tokens": 931666.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.446602392196655,
|
|
"epoch": 0.3914590747330961,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.9418,
|
|
"mean_token_accuracy": 0.0699479129165411,
|
|
"num_tokens": 946286.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.44321928024292,
|
|
"epoch": 0.3973902728351127,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.9048,
|
|
"mean_token_accuracy": 0.0711387813091278,
|
|
"num_tokens": 959689.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.383141279220581,
|
|
"epoch": 0.4033214709371293,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.9099,
|
|
"mean_token_accuracy": 0.07436129115521908,
|
|
"num_tokens": 972707.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.326433992385864,
|
|
"epoch": 0.4092526690391459,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.9298,
|
|
"mean_token_accuracy": 0.07282942235469818,
|
|
"num_tokens": 988010.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.3913586139678955,
|
|
"epoch": 0.4151838671411625,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.9494,
|
|
"mean_token_accuracy": 0.0702156439423561,
|
|
"num_tokens": 1003823.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.271407747268677,
|
|
"epoch": 0.4211150652431791,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7257,
|
|
"mean_token_accuracy": 0.08785916194319725,
|
|
"num_tokens": 1018276.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.315984010696411,
|
|
"epoch": 0.42704626334519574,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.9177,
|
|
"mean_token_accuracy": 0.07590937204658985,
|
|
"num_tokens": 1031242.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.36967716217041,
|
|
"epoch": 0.43297746144721233,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.826,
|
|
"mean_token_accuracy": 0.08421060293912888,
|
|
"num_tokens": 1043940.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.209809446334839,
|
|
"epoch": 0.4389086595492289,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.7337,
|
|
"mean_token_accuracy": 0.08058795779943466,
|
|
"num_tokens": 1058656.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.326260042190552,
|
|
"epoch": 0.44483985765124556,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.8599,
|
|
"mean_token_accuracy": 0.07443175241351127,
|
|
"num_tokens": 1073128.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.216764450073242,
|
|
"epoch": 0.45077105575326215,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.7831,
|
|
"mean_token_accuracy": 0.08439998105168342,
|
|
"num_tokens": 1086467.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.274557256698609,
|
|
"epoch": 0.45670225385527874,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.9052,
|
|
"mean_token_accuracy": 0.08290659114718438,
|
|
"num_tokens": 1100356.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.288721132278442,
|
|
"epoch": 0.4626334519572954,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.8286,
|
|
"mean_token_accuracy": 0.08163000270724297,
|
|
"num_tokens": 1115877.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.271324348449707,
|
|
"epoch": 0.46856465005931197,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.7874,
|
|
"mean_token_accuracy": 0.08328207209706306,
|
|
"num_tokens": 1131415.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.159351873397827,
|
|
"epoch": 0.4744958481613286,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.7132,
|
|
"mean_token_accuracy": 0.08293095007538795,
|
|
"num_tokens": 1145903.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.103127574920654,
|
|
"epoch": 0.4804270462633452,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.67,
|
|
"mean_token_accuracy": 0.08018745929002762,
|
|
"num_tokens": 1161168.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.14245753288269,
|
|
"epoch": 0.4863582443653618,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.7324,
|
|
"mean_token_accuracy": 0.08255215212702752,
|
|
"num_tokens": 1178155.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.131292104721069,
|
|
"epoch": 0.49228944246737844,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.5728,
|
|
"mean_token_accuracy": 0.09345637336373329,
|
|
"num_tokens": 1190830.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.046381425857544,
|
|
"epoch": 0.498220640569395,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.08820187523961068,
|
|
"num_tokens": 1205260.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.083016061782837,
|
|
"epoch": 0.5041518386714117,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.6316,
|
|
"mean_token_accuracy": 0.09104111194610595,
|
|
"num_tokens": 1218856.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.034869623184204,
|
|
"epoch": 0.5100830367734283,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.6934,
|
|
"mean_token_accuracy": 0.09010339081287384,
|
|
"num_tokens": 1234669.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.032301235198974,
|
|
"epoch": 0.5160142348754448,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.6293,
|
|
"mean_token_accuracy": 0.09128119722008705,
|
|
"num_tokens": 1249514.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.032708358764649,
|
|
"epoch": 0.5219454329774614,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5626,
|
|
"mean_token_accuracy": 0.0961005687713623,
|
|
"num_tokens": 1262311.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.9413275718688965,
|
|
"epoch": 0.527876631079478,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.5488,
|
|
"mean_token_accuracy": 0.09156981036067009,
|
|
"num_tokens": 1277681.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.003400182723999,
|
|
"epoch": 0.5338078291814946,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5879,
|
|
"mean_token_accuracy": 0.09069180116057396,
|
|
"num_tokens": 1292300.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.015186309814453,
|
|
"epoch": 0.5397390272835113,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.595,
|
|
"mean_token_accuracy": 0.08999119028449058,
|
|
"num_tokens": 1307426.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.900823402404785,
|
|
"epoch": 0.5456702253855279,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.6162,
|
|
"mean_token_accuracy": 0.09043457433581352,
|
|
"num_tokens": 1322167.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.999922370910644,
|
|
"epoch": 0.5516014234875445,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.09610431045293807,
|
|
"num_tokens": 1335343.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.857682752609253,
|
|
"epoch": 0.5575326215895611,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.4976,
|
|
"mean_token_accuracy": 0.09826786443591118,
|
|
"num_tokens": 1349078.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.899160289764405,
|
|
"epoch": 0.5634638196915777,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.4224,
|
|
"mean_token_accuracy": 0.1049638457596302,
|
|
"num_tokens": 1362030.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.993446207046508,
|
|
"epoch": 0.5693950177935944,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.5763,
|
|
"mean_token_accuracy": 0.08985362276434898,
|
|
"num_tokens": 1378303.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.935125350952148,
|
|
"epoch": 0.575326215895611,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5767,
|
|
"mean_token_accuracy": 0.09316942617297172,
|
|
"num_tokens": 1392585.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.9703733921051025,
|
|
"epoch": 0.5812574139976275,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.5814,
|
|
"mean_token_accuracy": 0.09292712658643723,
|
|
"num_tokens": 1406962.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.836954164505005,
|
|
"epoch": 0.5871886120996441,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.4605,
|
|
"mean_token_accuracy": 0.09848191663622856,
|
|
"num_tokens": 1419493.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.868243980407715,
|
|
"epoch": 0.5931198102016607,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4801,
|
|
"mean_token_accuracy": 0.09710933044552802,
|
|
"num_tokens": 1435373.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5931198102016607,
|
|
"eval_entropy": 6.868733419231111,
|
|
"eval_loss": 6.604183673858643,
|
|
"eval_mean_token_accuracy": 0.09870429635276458,
|
|
"eval_num_tokens": 1435373.0,
|
|
"eval_runtime": 2.5304,
|
|
"eval_samples_per_second": 1030.674,
|
|
"eval_steps_per_second": 128.834,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.877611589431763,
|
|
"epoch": 0.5990510083036773,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4385,
|
|
"mean_token_accuracy": 0.10158890634775161,
|
|
"num_tokens": 1449805.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.82432370185852,
|
|
"epoch": 0.604982206405694,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.09393253624439239,
|
|
"num_tokens": 1463657.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.9438111782073975,
|
|
"epoch": 0.6109134045077106,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4667,
|
|
"mean_token_accuracy": 0.10057351440191269,
|
|
"num_tokens": 1478089.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.7661386013031,
|
|
"epoch": 0.6168446026097272,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4698,
|
|
"mean_token_accuracy": 0.10367124900221825,
|
|
"num_tokens": 1492225.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.823407316207886,
|
|
"epoch": 0.6227758007117438,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.4847,
|
|
"mean_token_accuracy": 0.0941373310983181,
|
|
"num_tokens": 1506751.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.816655206680298,
|
|
"epoch": 0.6287069988137604,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.3687,
|
|
"mean_token_accuracy": 0.10237260162830353,
|
|
"num_tokens": 1520590.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.7571464538574215,
|
|
"epoch": 0.6346381969157769,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4174,
|
|
"mean_token_accuracy": 0.1001747913658619,
|
|
"num_tokens": 1534752.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.783740854263305,
|
|
"epoch": 0.6405693950177936,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.3769,
|
|
"mean_token_accuracy": 0.10932414084672928,
|
|
"num_tokens": 1547891.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.724957847595215,
|
|
"epoch": 0.6465005931198102,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3247,
|
|
"mean_token_accuracy": 0.10839766561985016,
|
|
"num_tokens": 1562560.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.809849119186401,
|
|
"epoch": 0.6524317912218268,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.10033882036805153,
|
|
"num_tokens": 1576896.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.732705163955688,
|
|
"epoch": 0.6583629893238434,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.3496,
|
|
"mean_token_accuracy": 0.10360537618398666,
|
|
"num_tokens": 1590806.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.740630769729615,
|
|
"epoch": 0.66429418742586,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4697,
|
|
"mean_token_accuracy": 0.09847484454512596,
|
|
"num_tokens": 1605874.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.772911071777344,
|
|
"epoch": 0.6702253855278766,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.3582,
|
|
"mean_token_accuracy": 0.10450243502855301,
|
|
"num_tokens": 1620947.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.672332429885865,
|
|
"epoch": 0.6761565836298933,
|
|
"grad_norm": 1.6796875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.393,
|
|
"mean_token_accuracy": 0.0936239667236805,
|
|
"num_tokens": 1635974.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.685958909988403,
|
|
"epoch": 0.6820877817319099,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.10207983180880546,
|
|
"num_tokens": 1651502.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.668765211105347,
|
|
"epoch": 0.6880189798339265,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3031,
|
|
"mean_token_accuracy": 0.10794700905680657,
|
|
"num_tokens": 1665547.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.6357416152954105,
|
|
"epoch": 0.693950177935943,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.2873,
|
|
"mean_token_accuracy": 0.10822367817163467,
|
|
"num_tokens": 1678875.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.671806287765503,
|
|
"epoch": 0.6998813760379596,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.3113,
|
|
"mean_token_accuracy": 0.10574031621217728,
|
|
"num_tokens": 1692673.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.649136257171631,
|
|
"epoch": 0.7058125741399762,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3549,
|
|
"mean_token_accuracy": 0.10422232896089553,
|
|
"num_tokens": 1707318.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.6383771896362305,
|
|
"epoch": 0.7117437722419929,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.3908,
|
|
"mean_token_accuracy": 0.09859651327133179,
|
|
"num_tokens": 1724331.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.54994387626648,
|
|
"epoch": 0.7176749703440095,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.1816,
|
|
"mean_token_accuracy": 0.11202944293618203,
|
|
"num_tokens": 1736151.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.531128454208374,
|
|
"epoch": 0.7236061684460261,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.11085337772965431,
|
|
"num_tokens": 1749831.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.670119380950927,
|
|
"epoch": 0.7295373665480427,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.3033,
|
|
"mean_token_accuracy": 0.10853635594248771,
|
|
"num_tokens": 1764296.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.622691011428833,
|
|
"epoch": 0.7354685646500593,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.274,
|
|
"mean_token_accuracy": 0.11076630353927612,
|
|
"num_tokens": 1778042.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.569571828842163,
|
|
"epoch": 0.741399762752076,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.2567,
|
|
"mean_token_accuracy": 0.11235617473721504,
|
|
"num_tokens": 1791614.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.480262708663941,
|
|
"epoch": 0.7473309608540926,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.2203,
|
|
"mean_token_accuracy": 0.10766129791736603,
|
|
"num_tokens": 1807388.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.692783164978027,
|
|
"epoch": 0.7532621589561092,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.10339825823903084,
|
|
"num_tokens": 1820407.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.493888711929321,
|
|
"epoch": 0.7591933570581257,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.2041,
|
|
"mean_token_accuracy": 0.11163504943251609,
|
|
"num_tokens": 1834666.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.516778469085693,
|
|
"epoch": 0.7651245551601423,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.2915,
|
|
"mean_token_accuracy": 0.10899070277810097,
|
|
"num_tokens": 1848160.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.624192810058593,
|
|
"epoch": 0.7710557532621589,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3004,
|
|
"mean_token_accuracy": 0.11020564585924149,
|
|
"num_tokens": 1865140.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.473660230636597,
|
|
"epoch": 0.7769869513641756,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.1704,
|
|
"mean_token_accuracy": 0.11627639755606652,
|
|
"num_tokens": 1878447.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.505875301361084,
|
|
"epoch": 0.7829181494661922,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.2289,
|
|
"mean_token_accuracy": 0.11153926998376847,
|
|
"num_tokens": 1893539.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.592386484146118,
|
|
"epoch": 0.7888493475682088,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2853,
|
|
"mean_token_accuracy": 0.11337760239839553,
|
|
"num_tokens": 1908052.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.493582057952881,
|
|
"epoch": 0.7947805456702254,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.1767,
|
|
"mean_token_accuracy": 0.11641135811805725,
|
|
"num_tokens": 1921528.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.502324628829956,
|
|
"epoch": 0.800711743772242,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.1962,
|
|
"mean_token_accuracy": 0.11467067450284958,
|
|
"num_tokens": 1935089.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.515084457397461,
|
|
"epoch": 0.8066429418742586,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3085,
|
|
"mean_token_accuracy": 0.10803200528025628,
|
|
"num_tokens": 1949717.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.477928495407104,
|
|
"epoch": 0.8125741399762753,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.171,
|
|
"mean_token_accuracy": 0.11270654797554017,
|
|
"num_tokens": 1962619.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.589840030670166,
|
|
"epoch": 0.8185053380782918,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2633,
|
|
"mean_token_accuracy": 0.1094804771244526,
|
|
"num_tokens": 1977094.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.4256690502166744,
|
|
"epoch": 0.8244365361803084,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2222,
|
|
"mean_token_accuracy": 0.10567561089992523,
|
|
"num_tokens": 1992446.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.449682807922363,
|
|
"epoch": 0.830367734282325,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.1737,
|
|
"mean_token_accuracy": 0.11305198296904564,
|
|
"num_tokens": 2008753.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.401576709747315,
|
|
"epoch": 0.8362989323843416,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.1256,
|
|
"mean_token_accuracy": 0.11778818741440773,
|
|
"num_tokens": 2024088.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.476001882553101,
|
|
"epoch": 0.8422301304863582,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.1461,
|
|
"mean_token_accuracy": 0.11078088879585266,
|
|
"num_tokens": 2038449.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.4743701934814455,
|
|
"epoch": 0.8481613285883749,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.131,
|
|
"mean_token_accuracy": 0.12273426279425621,
|
|
"num_tokens": 2052038.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.402703428268433,
|
|
"epoch": 0.8540925266903915,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.071,
|
|
"mean_token_accuracy": 0.11885945498943329,
|
|
"num_tokens": 2064907.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.448020601272583,
|
|
"epoch": 0.8600237247924081,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.2535,
|
|
"mean_token_accuracy": 0.11163201704621314,
|
|
"num_tokens": 2081295.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.399222040176392,
|
|
"epoch": 0.8659549228944247,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.178,
|
|
"mean_token_accuracy": 0.11559945717453957,
|
|
"num_tokens": 2098077.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.422470808029175,
|
|
"epoch": 0.8718861209964412,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1573,
|
|
"mean_token_accuracy": 0.11710702180862427,
|
|
"num_tokens": 2111716.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.352961158752441,
|
|
"epoch": 0.8778173190984578,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.0612,
|
|
"mean_token_accuracy": 0.12471788600087166,
|
|
"num_tokens": 2125757.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.467966890335083,
|
|
"epoch": 0.8837485172004745,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2669,
|
|
"mean_token_accuracy": 0.11346666887402534,
|
|
"num_tokens": 2142929.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.385439872741699,
|
|
"epoch": 0.8896797153024911,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.1379,
|
|
"mean_token_accuracy": 0.11529970616102218,
|
|
"num_tokens": 2155206.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.316713285446167,
|
|
"epoch": 0.8956109134045077,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.0204,
|
|
"mean_token_accuracy": 0.1224408395588398,
|
|
"num_tokens": 2169816.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.3729979515075685,
|
|
"epoch": 0.9015421115065243,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.106,
|
|
"mean_token_accuracy": 0.11994262486696243,
|
|
"num_tokens": 2184827.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.261362934112549,
|
|
"epoch": 0.9074733096085409,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.0392,
|
|
"mean_token_accuracy": 0.11882986202836036,
|
|
"num_tokens": 2199333.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.316654014587402,
|
|
"epoch": 0.9134045077105575,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.12039939314126968,
|
|
"num_tokens": 2215239.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.3938861846923825,
|
|
"epoch": 0.9193357058125742,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.085,
|
|
"mean_token_accuracy": 0.1213831715285778,
|
|
"num_tokens": 2229369.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.262855195999146,
|
|
"epoch": 0.9252669039145908,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.0793,
|
|
"mean_token_accuracy": 0.12699812799692153,
|
|
"num_tokens": 2242709.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.378830051422119,
|
|
"epoch": 0.9311981020166074,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.0655,
|
|
"mean_token_accuracy": 0.11725454851984977,
|
|
"num_tokens": 2257067.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.268696022033692,
|
|
"epoch": 0.9371293001186239,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.0146,
|
|
"mean_token_accuracy": 0.12406265735626221,
|
|
"num_tokens": 2270932.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.312669181823731,
|
|
"epoch": 0.9430604982206405,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.109,
|
|
"mean_token_accuracy": 0.11879168227314948,
|
|
"num_tokens": 2285517.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.1997991561889645,
|
|
"epoch": 0.9489916963226572,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 5.9847,
|
|
"mean_token_accuracy": 0.12326976284384727,
|
|
"num_tokens": 2298657.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.30251636505127,
|
|
"epoch": 0.9549228944246738,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000402,
|
|
"loss": 5.9947,
|
|
"mean_token_accuracy": 0.1275511786341667,
|
|
"num_tokens": 2312521.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.245606517791748,
|
|
"epoch": 0.9608540925266904,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 5.98,
|
|
"mean_token_accuracy": 0.12642857506871225,
|
|
"num_tokens": 2325849.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.244391679763794,
|
|
"epoch": 0.966785290628707,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.0049,
|
|
"mean_token_accuracy": 0.12391034886240959,
|
|
"num_tokens": 2339717.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.243370532989502,
|
|
"epoch": 0.9727164887307236,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.0679,
|
|
"mean_token_accuracy": 0.12291354686021805,
|
|
"num_tokens": 2353536.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.279402780532837,
|
|
"epoch": 0.9786476868327402,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.0335,
|
|
"mean_token_accuracy": 0.11777281686663628,
|
|
"num_tokens": 2368626.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.269067335128784,
|
|
"epoch": 0.9845788849347569,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.0985,
|
|
"mean_token_accuracy": 0.12023787796497346,
|
|
"num_tokens": 2383768.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.300761365890503,
|
|
"epoch": 0.9905100830367735,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1254,
|
|
"mean_token_accuracy": 0.12162568196654319,
|
|
"num_tokens": 2398089.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.220589017868042,
|
|
"epoch": 0.99644128113879,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 5.992,
|
|
"mean_token_accuracy": 0.12641523778438568,
|
|
"num_tokens": 2411354.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.249312019348144,
|
|
"epoch": 1.0023724792408066,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 5.9945,
|
|
"mean_token_accuracy": 0.12419759780168534,
|
|
"num_tokens": 2425071.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.2975963115692135,
|
|
"epoch": 1.0083036773428233,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 5.9569,
|
|
"mean_token_accuracy": 0.13322616517543792,
|
|
"num_tokens": 2439143.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.093033933639527,
|
|
"epoch": 1.0142348754448398,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 5.7989,
|
|
"mean_token_accuracy": 0.12915606275200844,
|
|
"num_tokens": 2453466.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.169495487213135,
|
|
"epoch": 1.0201660735468565,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 5.9269,
|
|
"mean_token_accuracy": 0.13161411434412001,
|
|
"num_tokens": 2466802.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.1576333999633786,
|
|
"epoch": 1.026097271648873,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 5.8612,
|
|
"mean_token_accuracy": 0.13197067752480507,
|
|
"num_tokens": 2480080.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.120718669891358,
|
|
"epoch": 1.0320284697508897,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 5.9076,
|
|
"mean_token_accuracy": 0.13211096227169036,
|
|
"num_tokens": 2495196.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.083641386032104,
|
|
"epoch": 1.0379596678529064,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 5.8176,
|
|
"mean_token_accuracy": 0.12702380120754242,
|
|
"num_tokens": 2510084.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.16945514678955,
|
|
"epoch": 1.0438908659549229,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 5.8956,
|
|
"mean_token_accuracy": 0.1281700126826763,
|
|
"num_tokens": 2522834.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.146343135833741,
|
|
"epoch": 1.0498220640569396,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 5.8657,
|
|
"mean_token_accuracy": 0.13543630614876748,
|
|
"num_tokens": 2536778.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.07814359664917,
|
|
"epoch": 1.055753262158956,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 5.9005,
|
|
"mean_token_accuracy": 0.13273447826504708,
|
|
"num_tokens": 2551220.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.273546886444092,
|
|
"epoch": 1.0616844602609727,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 5.9534,
|
|
"mean_token_accuracy": 0.12553919553756715,
|
|
"num_tokens": 2565686.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.094685745239258,
|
|
"epoch": 1.0676156583629894,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 5.8687,
|
|
"mean_token_accuracy": 0.12925135344266891,
|
|
"num_tokens": 2579569.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.097094774246216,
|
|
"epoch": 1.073546856465006,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.12844056114554406,
|
|
"num_tokens": 2593226.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.054626512527466,
|
|
"epoch": 1.0794780545670226,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 5.8381,
|
|
"mean_token_accuracy": 0.13379990607500075,
|
|
"num_tokens": 2607711.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.059130144119263,
|
|
"epoch": 1.085409252669039,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 5.8697,
|
|
"mean_token_accuracy": 0.130379880964756,
|
|
"num_tokens": 2623024.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.079943466186523,
|
|
"epoch": 1.0913404507710558,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 5.8499,
|
|
"mean_token_accuracy": 0.12744099125266076,
|
|
"num_tokens": 2637824.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.051183843612671,
|
|
"epoch": 1.0972716488730723,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000462,
|
|
"loss": 5.8486,
|
|
"mean_token_accuracy": 0.13178792968392372,
|
|
"num_tokens": 2652077.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.085985422134399,
|
|
"epoch": 1.103202846975089,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 5.8555,
|
|
"mean_token_accuracy": 0.1309271603822708,
|
|
"num_tokens": 2666791.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.091299629211425,
|
|
"epoch": 1.1091340450771057,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000467,
|
|
"loss": 5.9439,
|
|
"mean_token_accuracy": 0.12866507172584535,
|
|
"num_tokens": 2681482.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.0315086364746096,
|
|
"epoch": 1.1150652431791221,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 5.7415,
|
|
"mean_token_accuracy": 0.13911210745573044,
|
|
"num_tokens": 2695305.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.070809412002563,
|
|
"epoch": 1.1209964412811388,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 5.81,
|
|
"mean_token_accuracy": 0.13352661579847336,
|
|
"num_tokens": 2712211.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.029483652114868,
|
|
"epoch": 1.1269276393831553,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 5.8556,
|
|
"mean_token_accuracy": 0.13038185760378837,
|
|
"num_tokens": 2726888.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.026515007019043,
|
|
"epoch": 1.132858837485172,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000477,
|
|
"loss": 5.8209,
|
|
"mean_token_accuracy": 0.1355987899005413,
|
|
"num_tokens": 2741734.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.002088499069214,
|
|
"epoch": 1.1387900355871885,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 5.8273,
|
|
"mean_token_accuracy": 0.13999779522418976,
|
|
"num_tokens": 2754680.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.053714990615845,
|
|
"epoch": 1.1447212336892052,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 5.8717,
|
|
"mean_token_accuracy": 0.12814902439713477,
|
|
"num_tokens": 2769572.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.109354639053345,
|
|
"epoch": 1.150652431791222,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 5.8165,
|
|
"mean_token_accuracy": 0.12989651635289193,
|
|
"num_tokens": 2782948.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 5.920255517959594,
|
|
"epoch": 1.1565836298932384,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.7569,
|
|
"mean_token_accuracy": 0.13256713449954988,
|
|
"num_tokens": 2795399.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 5.974841451644897,
|
|
"epoch": 1.162514827995255,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.7628,
|
|
"mean_token_accuracy": 0.13333625793457032,
|
|
"num_tokens": 2809591.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.011464214324951,
|
|
"epoch": 1.1684460260972718,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000492,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.13518815189599992,
|
|
"num_tokens": 2824106.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.024096918106079,
|
|
"epoch": 1.1743772241992882,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 5.9184,
|
|
"mean_token_accuracy": 0.12920955419540406,
|
|
"num_tokens": 2838616.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 5.981972026824951,
|
|
"epoch": 1.180308422301305,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 5.798,
|
|
"mean_token_accuracy": 0.13885542377829552,
|
|
"num_tokens": 2851707.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.011092615127564,
|
|
"epoch": 1.1862396204033214,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 5.8358,
|
|
"mean_token_accuracy": 0.1295635662972927,
|
|
"num_tokens": 2867644.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.1862396204033214,
|
|
"eval_entropy": 5.97049690024253,
|
|
"eval_loss": 6.048816204071045,
|
|
"eval_mean_token_accuracy": 0.130689339495144,
|
|
"eval_num_tokens": 2867644.0,
|
|
"eval_runtime": 2.4358,
|
|
"eval_samples_per_second": 1070.705,
|
|
"eval_steps_per_second": 133.838,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.003067874908448,
|
|
"epoch": 1.1921708185053381,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00049999967732573,
|
|
"loss": 5.8347,
|
|
"mean_token_accuracy": 0.12673901543021202,
|
|
"num_tokens": 2882663.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.0496416091918945,
|
|
"epoch": 1.1981020166073546,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999983664630945,
|
|
"loss": 5.8902,
|
|
"mean_token_accuracy": 0.1319477692246437,
|
|
"num_tokens": 2898791.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 5.922034502029419,
|
|
"epoch": 1.2040332147093713,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999960472508219,
|
|
"loss": 5.7771,
|
|
"mean_token_accuracy": 0.13882209360599518,
|
|
"num_tokens": 2912997.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 5.958587026596069,
|
|
"epoch": 1.209964412811388,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999927196993059,
|
|
"loss": 5.7702,
|
|
"mean_token_accuracy": 0.13983350694179536,
|
|
"num_tokens": 2925064.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.004720687866211,
|
|
"epoch": 1.2158956109134045,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999883838234592,
|
|
"loss": 5.8331,
|
|
"mean_token_accuracy": 0.1291055254638195,
|
|
"num_tokens": 2941400.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 5.947526121139527,
|
|
"epoch": 1.2218268090154212,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999830396427133,
|
|
"loss": 5.8063,
|
|
"mean_token_accuracy": 0.1374014362692833,
|
|
"num_tokens": 2955541.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 5.969132947921753,
|
|
"epoch": 1.2277580071174377,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999766871810188,
|
|
"loss": 5.8038,
|
|
"mean_token_accuracy": 0.13658639341592788,
|
|
"num_tokens": 2968928.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.005309391021728,
|
|
"epoch": 1.2336892052194544,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999693264668446,
|
|
"loss": 5.748,
|
|
"mean_token_accuracy": 0.13989535421133042,
|
|
"num_tokens": 2984072.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 5.915805292129517,
|
|
"epoch": 1.2396204033214708,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004999609575331786,
|
|
"loss": 5.8379,
|
|
"mean_token_accuracy": 0.13512861132621765,
|
|
"num_tokens": 2999083.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 5.936822700500488,
|
|
"epoch": 1.2455516014234875,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999515804175268,
|
|
"loss": 5.7715,
|
|
"mean_token_accuracy": 0.13389404788613318,
|
|
"num_tokens": 3012738.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 5.919717836380005,
|
|
"epoch": 1.2514827995255042,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999411951619134,
|
|
"loss": 5.7231,
|
|
"mean_token_accuracy": 0.14047607854008676,
|
|
"num_tokens": 3027929.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 5.967278051376343,
|
|
"epoch": 1.2574139976275207,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000499929801812881,
|
|
"loss": 5.8651,
|
|
"mean_token_accuracy": 0.1260008007287979,
|
|
"num_tokens": 3043576.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 5.859627962112427,
|
|
"epoch": 1.2633451957295374,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999174004214899,
|
|
"loss": 5.7148,
|
|
"mean_token_accuracy": 0.13711703568696976,
|
|
"num_tokens": 3059352.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.000982999801636,
|
|
"epoch": 1.269276393831554,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999039910433179,
|
|
"loss": 5.8426,
|
|
"mean_token_accuracy": 0.14122046902775764,
|
|
"num_tokens": 3073140.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 5.896023321151733,
|
|
"epoch": 1.2752075919335706,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004998895737384605,
|
|
"loss": 5.818,
|
|
"mean_token_accuracy": 0.137306509912014,
|
|
"num_tokens": 3087710.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 5.961836051940918,
|
|
"epoch": 1.281138790035587,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004998741485715297,
|
|
"loss": 5.7594,
|
|
"mean_token_accuracy": 0.13758910968899726,
|
|
"num_tokens": 3102035.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.8731317043304445,
|
|
"epoch": 1.2870699881376038,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004998577156116551,
|
|
"loss": 5.8223,
|
|
"mean_token_accuracy": 0.13428573533892632,
|
|
"num_tokens": 3118534.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 5.9692919731140135,
|
|
"epoch": 1.2930011862396205,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004998402749324823,
|
|
"loss": 5.7653,
|
|
"mean_token_accuracy": 0.12841569781303405,
|
|
"num_tokens": 3133419.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 5.975205659866333,
|
|
"epoch": 1.298932384341637,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004998218266121731,
|
|
"loss": 5.861,
|
|
"mean_token_accuracy": 0.1412831462919712,
|
|
"num_tokens": 3145903.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 5.925574588775635,
|
|
"epoch": 1.3048635824436536,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998023707334055,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.13825963586568832,
|
|
"num_tokens": 3159938.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.843659114837647,
|
|
"epoch": 1.3107947805456703,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004997819073833724,
|
|
"loss": 5.6987,
|
|
"mean_token_accuracy": 0.14315861761569976,
|
|
"num_tokens": 3175089.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.886387014389038,
|
|
"epoch": 1.3167259786476868,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997604366537823,
|
|
"loss": 5.833,
|
|
"mean_token_accuracy": 0.13882801905274392,
|
|
"num_tokens": 3191027.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 5.824345541000366,
|
|
"epoch": 1.3226571767497035,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499737958640858,
|
|
"loss": 5.6801,
|
|
"mean_token_accuracy": 0.1453348807990551,
|
|
"num_tokens": 3205489.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 5.800811624526977,
|
|
"epoch": 1.32858837485172,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004997144734453367,
|
|
"loss": 5.6496,
|
|
"mean_token_accuracy": 0.14170826822519303,
|
|
"num_tokens": 3218764.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.964564657211303,
|
|
"epoch": 1.3345195729537367,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004996899811724693,
|
|
"loss": 5.7727,
|
|
"mean_token_accuracy": 0.13682308942079544,
|
|
"num_tokens": 3232953.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 5.803732538223267,
|
|
"epoch": 1.3404507710557532,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004996644819320201,
|
|
"loss": 5.676,
|
|
"mean_token_accuracy": 0.1429899662733078,
|
|
"num_tokens": 3246977.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 5.727921009063721,
|
|
"epoch": 1.3463819691577699,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996379758382662,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.14849494844675065,
|
|
"num_tokens": 3260668.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 5.879596281051636,
|
|
"epoch": 1.3523131672597866,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996104630099969,
|
|
"loss": 5.7056,
|
|
"mean_token_accuracy": 0.14046500623226166,
|
|
"num_tokens": 3274213.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 5.843897438049316,
|
|
"epoch": 1.358244365361803,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995819435705133,
|
|
"loss": 5.6817,
|
|
"mean_token_accuracy": 0.13680549412965776,
|
|
"num_tokens": 3288011.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 5.844601249694824,
|
|
"epoch": 1.3641755634638197,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004995524176476277,
|
|
"loss": 5.7551,
|
|
"mean_token_accuracy": 0.13818126022815705,
|
|
"num_tokens": 3303263.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.869444084167481,
|
|
"epoch": 1.3701067615658362,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004995218853736631,
|
|
"loss": 5.731,
|
|
"mean_token_accuracy": 0.14651385322213173,
|
|
"num_tokens": 3316843.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 5.764492511749268,
|
|
"epoch": 1.376037959667853,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004994903468854527,
|
|
"loss": 5.5899,
|
|
"mean_token_accuracy": 0.14461414963006974,
|
|
"num_tokens": 3331527.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 5.831653451919555,
|
|
"epoch": 1.3819691577698694,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994578023243386,
|
|
"loss": 5.7027,
|
|
"mean_token_accuracy": 0.14239056706428527,
|
|
"num_tokens": 3347795.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.8456896305084225,
|
|
"epoch": 1.387900355871886,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004994242518361723,
|
|
"loss": 5.7528,
|
|
"mean_token_accuracy": 0.13648397997021675,
|
|
"num_tokens": 3363532.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.773947715759277,
|
|
"epoch": 1.3938315539739028,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004993896955713133,
|
|
"loss": 5.7159,
|
|
"mean_token_accuracy": 0.14554002955555917,
|
|
"num_tokens": 3377040.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.8154942989349365,
|
|
"epoch": 1.3997627520759193,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004993541336846283,
|
|
"loss": 5.666,
|
|
"mean_token_accuracy": 0.1508113294839859,
|
|
"num_tokens": 3390663.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 5.816216230392456,
|
|
"epoch": 1.405693950177936,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.000499317566335491,
|
|
"loss": 5.6963,
|
|
"mean_token_accuracy": 0.1484333947300911,
|
|
"num_tokens": 3405860.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.797020339965821,
|
|
"epoch": 1.4116251482799527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992799936877813,
|
|
"loss": 5.6734,
|
|
"mean_token_accuracy": 0.13814794421195983,
|
|
"num_tokens": 3420248.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.837377595901489,
|
|
"epoch": 1.4175563463819691,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004992414159098841,
|
|
"loss": 5.6779,
|
|
"mean_token_accuracy": 0.144036927819252,
|
|
"num_tokens": 3434126.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.865153551101685,
|
|
"epoch": 1.4234875444839858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004992018331746891,
|
|
"loss": 5.728,
|
|
"mean_token_accuracy": 0.13874396309256554,
|
|
"num_tokens": 3449462.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.815957403182983,
|
|
"epoch": 1.4294187425860023,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004991612456595896,
|
|
"loss": 5.7856,
|
|
"mean_token_accuracy": 0.1379348635673523,
|
|
"num_tokens": 3465536.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 5.77282943725586,
|
|
"epoch": 1.435349940688019,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004991196535464824,
|
|
"loss": 5.675,
|
|
"mean_token_accuracy": 0.1448883593082428,
|
|
"num_tokens": 3479129.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.719824647903442,
|
|
"epoch": 1.4412811387900355,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004990770570217659,
|
|
"loss": 5.5833,
|
|
"mean_token_accuracy": 0.15740741640329362,
|
|
"num_tokens": 3492191.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.840599250793457,
|
|
"epoch": 1.4472123368920522,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004990334562763401,
|
|
"loss": 5.7191,
|
|
"mean_token_accuracy": 0.14363316372036933,
|
|
"num_tokens": 3507459.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.758994436264038,
|
|
"epoch": 1.453143534994069,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004989888515056057,
|
|
"loss": 5.6846,
|
|
"mean_token_accuracy": 0.14048208221793174,
|
|
"num_tokens": 3521517.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.83376293182373,
|
|
"epoch": 1.4590747330960854,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004989432429094627,
|
|
"loss": 5.645,
|
|
"mean_token_accuracy": 0.14501417130231858,
|
|
"num_tokens": 3535279.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.730698728561402,
|
|
"epoch": 1.465005931198102,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004988966306923099,
|
|
"loss": 5.5423,
|
|
"mean_token_accuracy": 0.15444459170103073,
|
|
"num_tokens": 3549960.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.7588770389556885,
|
|
"epoch": 1.4709371293001186,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004988490150630441,
|
|
"loss": 5.6762,
|
|
"mean_token_accuracy": 0.1423516869544983,
|
|
"num_tokens": 3563814.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.824617910385132,
|
|
"epoch": 1.4768683274021353,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004988003962350589,
|
|
"loss": 5.6802,
|
|
"mean_token_accuracy": 0.14822768270969391,
|
|
"num_tokens": 3577529.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.733047246932983,
|
|
"epoch": 1.4827995255041517,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004987507744262436,
|
|
"loss": 5.6131,
|
|
"mean_token_accuracy": 0.1476161688566208,
|
|
"num_tokens": 3591812.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.68289361000061,
|
|
"epoch": 1.4887307236061684,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004987001498589828,
|
|
"loss": 5.5893,
|
|
"mean_token_accuracy": 0.14960140883922576,
|
|
"num_tokens": 3604954.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.745348644256592,
|
|
"epoch": 1.4946619217081851,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004986485227601548,
|
|
"loss": 5.6545,
|
|
"mean_token_accuracy": 0.14511096552014352,
|
|
"num_tokens": 3620486.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.739377021789551,
|
|
"epoch": 1.5005931198102016,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.000498595893361131,
|
|
"loss": 5.5731,
|
|
"mean_token_accuracy": 0.15375369787216187,
|
|
"num_tokens": 3633972.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.809413433074951,
|
|
"epoch": 1.5065243179122183,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004985422618977744,
|
|
"loss": 5.6275,
|
|
"mean_token_accuracy": 0.15041894391179084,
|
|
"num_tokens": 3649306.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 5.7898821353912355,
|
|
"epoch": 1.512455516014235,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004984876286104392,
|
|
"loss": 5.7006,
|
|
"mean_token_accuracy": 0.14360713064670563,
|
|
"num_tokens": 3664809.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.734278774261474,
|
|
"epoch": 1.5183867141162515,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004984319937439693,
|
|
"loss": 5.6463,
|
|
"mean_token_accuracy": 0.14773811772465706,
|
|
"num_tokens": 3677669.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.791883563995361,
|
|
"epoch": 1.524317912218268,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004983753575476969,
|
|
"loss": 5.6371,
|
|
"mean_token_accuracy": 0.1450825333595276,
|
|
"num_tokens": 3690588.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.810116958618164,
|
|
"epoch": 1.5302491103202847,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004983177202754421,
|
|
"loss": 5.6821,
|
|
"mean_token_accuracy": 0.14587436318397523,
|
|
"num_tokens": 3703749.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.796082973480225,
|
|
"epoch": 1.5361803084223014,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004982590821855115,
|
|
"loss": 5.6986,
|
|
"mean_token_accuracy": 0.1416331112384796,
|
|
"num_tokens": 3720181.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.7356054306030275,
|
|
"epoch": 1.5421115065243178,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004981994435406965,
|
|
"loss": 5.6172,
|
|
"mean_token_accuracy": 0.15273793041706085,
|
|
"num_tokens": 3734918.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.694397592544556,
|
|
"epoch": 1.5480427046263345,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004981388046082731,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.15164828896522523,
|
|
"num_tokens": 3748660.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.7075213432312015,
|
|
"epoch": 1.5539739027283512,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004980771656599998,
|
|
"loss": 5.6362,
|
|
"mean_token_accuracy": 0.14845281541347505,
|
|
"num_tokens": 3763481.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.803341007232666,
|
|
"epoch": 1.5599051008303677,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004980145269721168,
|
|
"loss": 5.6529,
|
|
"mean_token_accuracy": 0.14686342626810073,
|
|
"num_tokens": 3777077.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.699376630783081,
|
|
"epoch": 1.5658362989323842,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000497950888825345,
|
|
"loss": 5.6558,
|
|
"mean_token_accuracy": 0.14442251324653627,
|
|
"num_tokens": 3793649.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.706108856201172,
|
|
"epoch": 1.571767497034401,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000497886251504884,
|
|
"loss": 5.5359,
|
|
"mean_token_accuracy": 0.15608448088169097,
|
|
"num_tokens": 3807712.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.644539451599121,
|
|
"epoch": 1.5776986951364176,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004978206153004117,
|
|
"loss": 5.5782,
|
|
"mean_token_accuracy": 0.15468101650476457,
|
|
"num_tokens": 3821637.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.83319206237793,
|
|
"epoch": 1.583629893238434,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977539805060824,
|
|
"loss": 5.7103,
|
|
"mean_token_accuracy": 0.1453540086746216,
|
|
"num_tokens": 3836272.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.648958444595337,
|
|
"epoch": 1.5895610913404508,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004976863474205257,
|
|
"loss": 5.5273,
|
|
"mean_token_accuracy": 0.1518918976187706,
|
|
"num_tokens": 3850018.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.8174097537994385,
|
|
"epoch": 1.5954922894424675,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004976177163468451,
|
|
"loss": 5.6052,
|
|
"mean_token_accuracy": 0.1518917240202427,
|
|
"num_tokens": 3865061.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.687358045578003,
|
|
"epoch": 1.601423487544484,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004975480875926168,
|
|
"loss": 5.6984,
|
|
"mean_token_accuracy": 0.14721440970897676,
|
|
"num_tokens": 3879697.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.69095287322998,
|
|
"epoch": 1.6073546856465006,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000497477461469888,
|
|
"loss": 5.5008,
|
|
"mean_token_accuracy": 0.15820710659027098,
|
|
"num_tokens": 3891889.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.717138576507568,
|
|
"epoch": 1.6132858837485173,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004974058382951757,
|
|
"loss": 5.6056,
|
|
"mean_token_accuracy": 0.14453255534172058,
|
|
"num_tokens": 3907356.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.704308366775512,
|
|
"epoch": 1.6192170818505338,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004973332183894653,
|
|
"loss": 5.5935,
|
|
"mean_token_accuracy": 0.15117160528898238,
|
|
"num_tokens": 3921839.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.614565849304199,
|
|
"epoch": 1.6251482799525503,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004972596020782096,
|
|
"loss": 5.4939,
|
|
"mean_token_accuracy": 0.1566345527768135,
|
|
"num_tokens": 3937266.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.742886257171631,
|
|
"epoch": 1.631079478054567,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004971849896913259,
|
|
"loss": 5.631,
|
|
"mean_token_accuracy": 0.14571579843759536,
|
|
"num_tokens": 3952721.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.74390230178833,
|
|
"epoch": 1.6370106761565837,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004971093815631965,
|
|
"loss": 5.64,
|
|
"mean_token_accuracy": 0.14603027999401091,
|
|
"num_tokens": 3966557.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.584549236297607,
|
|
"epoch": 1.6429418742586002,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004970327780326655,
|
|
"loss": 5.4739,
|
|
"mean_token_accuracy": 0.15459775924682617,
|
|
"num_tokens": 3978739.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.584955310821533,
|
|
"epoch": 1.6488730723606169,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004969551794430383,
|
|
"loss": 5.4729,
|
|
"mean_token_accuracy": 0.1543826073408127,
|
|
"num_tokens": 3993868.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.754845428466797,
|
|
"epoch": 1.6548042704626336,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004968765861420799,
|
|
"loss": 5.6204,
|
|
"mean_token_accuracy": 0.15352840721607208,
|
|
"num_tokens": 4006661.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.612430429458618,
|
|
"epoch": 1.66073546856465,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004967969984820128,
|
|
"loss": 5.5979,
|
|
"mean_token_accuracy": 0.15308566838502885,
|
|
"num_tokens": 4021670.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.754387235641479,
|
|
"epoch": 1.6666666666666665,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004967164168195159,
|
|
"loss": 5.5708,
|
|
"mean_token_accuracy": 0.15037682205438613,
|
|
"num_tokens": 4036995.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.684358930587768,
|
|
"epoch": 1.6725978647686834,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000496634841515723,
|
|
"loss": 5.6189,
|
|
"mean_token_accuracy": 0.1510004997253418,
|
|
"num_tokens": 4049464.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.670193815231324,
|
|
"epoch": 1.6785290628707,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004965522729362209,
|
|
"loss": 5.5126,
|
|
"mean_token_accuracy": 0.1586139217019081,
|
|
"num_tokens": 4063630.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.625356912612915,
|
|
"epoch": 1.6844602609727164,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004964687114510478,
|
|
"loss": 5.5268,
|
|
"mean_token_accuracy": 0.15081458985805513,
|
|
"num_tokens": 4079890.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.592355966567993,
|
|
"epoch": 1.690391459074733,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004963841574346917,
|
|
"loss": 5.5931,
|
|
"mean_token_accuracy": 0.1528349280357361,
|
|
"num_tokens": 4095693.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.809736585617065,
|
|
"epoch": 1.6963226571767498,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004962986112660887,
|
|
"loss": 5.6263,
|
|
"mean_token_accuracy": 0.14935975819826125,
|
|
"num_tokens": 4109655.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.607567262649536,
|
|
"epoch": 1.7022538552787663,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004962120733286215,
|
|
"loss": 5.5123,
|
|
"mean_token_accuracy": 0.15562189668416976,
|
|
"num_tokens": 4123870.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.743023347854614,
|
|
"epoch": 1.708185053380783,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004961245440101171,
|
|
"loss": 5.6319,
|
|
"mean_token_accuracy": 0.15589529424905776,
|
|
"num_tokens": 4138009.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.7484697818756105,
|
|
"epoch": 1.7141162514827997,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496036023702846,
|
|
"loss": 5.729,
|
|
"mean_token_accuracy": 0.14504553452134133,
|
|
"num_tokens": 4151929.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.704558658599853,
|
|
"epoch": 1.7200474495848161,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004959465128035195,
|
|
"loss": 5.5439,
|
|
"mean_token_accuracy": 0.1563679590821266,
|
|
"num_tokens": 4164626.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.6194459915161135,
|
|
"epoch": 1.7259786476868326,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004958560117132884,
|
|
"loss": 5.4823,
|
|
"mean_token_accuracy": 0.15618694126605986,
|
|
"num_tokens": 4179418.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.661043882369995,
|
|
"epoch": 1.7319098457888493,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004957645208377411,
|
|
"loss": 5.6078,
|
|
"mean_token_accuracy": 0.14679401218891144,
|
|
"num_tokens": 4194385.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.695458889007568,
|
|
"epoch": 1.737841043890866,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000495672040586902,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.1486389935016632,
|
|
"num_tokens": 4209246.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.590035724639892,
|
|
"epoch": 1.7437722419928825,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004955785713752292,
|
|
"loss": 5.512,
|
|
"mean_token_accuracy": 0.1497904747724533,
|
|
"num_tokens": 4225142.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.62952733039856,
|
|
"epoch": 1.7497034400948992,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004954841136216132,
|
|
"loss": 5.5127,
|
|
"mean_token_accuracy": 0.1571367785334587,
|
|
"num_tokens": 4239495.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.631760358810425,
|
|
"epoch": 1.755634638196916,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004953886677493746,
|
|
"loss": 5.4409,
|
|
"mean_token_accuracy": 0.16082352101802827,
|
|
"num_tokens": 4251780.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.635686779022217,
|
|
"epoch": 1.7615658362989324,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004952922341862622,
|
|
"loss": 5.6989,
|
|
"mean_token_accuracy": 0.1460478313267231,
|
|
"num_tokens": 4265490.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.635773324966431,
|
|
"epoch": 1.7674970344009489,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004951948133644514,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.1595569834113121,
|
|
"num_tokens": 4280478.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.658703470230103,
|
|
"epoch": 1.7734282325029656,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004950964057205421,
|
|
"loss": 5.5446,
|
|
"mean_token_accuracy": 0.15271539092063904,
|
|
"num_tokens": 4296649.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.559925746917725,
|
|
"epoch": 1.7793594306049823,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004949970116955566,
|
|
"loss": 5.5282,
|
|
"mean_token_accuracy": 0.15817976742982864,
|
|
"num_tokens": 4310470.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.7793594306049823,
|
|
"eval_entropy": 5.509380918338986,
|
|
"eval_loss": 5.712648868560791,
|
|
"eval_mean_token_accuracy": 0.15203302554740497,
|
|
"eval_num_tokens": 4310470.0,
|
|
"eval_runtime": 2.4337,
|
|
"eval_samples_per_second": 1071.632,
|
|
"eval_steps_per_second": 133.954,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.659485387802124,
|
|
"epoch": 1.7852906287069987,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004948966317349379,
|
|
"loss": 5.5592,
|
|
"mean_token_accuracy": 0.1531466454267502,
|
|
"num_tokens": 4324996.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.5955507278442385,
|
|
"epoch": 1.7912218268090154,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004947952662885473,
|
|
"loss": 5.494,
|
|
"mean_token_accuracy": 0.15311797112226486,
|
|
"num_tokens": 4337194.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.518834209442138,
|
|
"epoch": 1.7971530249110321,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000494692915810663,
|
|
"loss": 5.4711,
|
|
"mean_token_accuracy": 0.15367639809846878,
|
|
"num_tokens": 4352516.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.6385064125061035,
|
|
"epoch": 1.8030842230130486,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004945895807599775,
|
|
"loss": 5.5102,
|
|
"mean_token_accuracy": 0.16052137017250062,
|
|
"num_tokens": 4367654.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.646355199813843,
|
|
"epoch": 1.809015421115065,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004944852615995956,
|
|
"loss": 5.497,
|
|
"mean_token_accuracy": 0.16428107917308807,
|
|
"num_tokens": 4381275.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.626601076126098,
|
|
"epoch": 1.814946619217082,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004943799587970328,
|
|
"loss": 5.5146,
|
|
"mean_token_accuracy": 0.15436722487211227,
|
|
"num_tokens": 4394392.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.512476968765259,
|
|
"epoch": 1.8208778173190985,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004942736728242127,
|
|
"loss": 5.4123,
|
|
"mean_token_accuracy": 0.15786397457122803,
|
|
"num_tokens": 4407512.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.644066619873047,
|
|
"epoch": 1.826809015421115,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004941664041574651,
|
|
"loss": 5.5096,
|
|
"mean_token_accuracy": 0.16130108386278152,
|
|
"num_tokens": 4422223.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.567513799667358,
|
|
"epoch": 1.8327402135231317,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004940581532775237,
|
|
"loss": 5.5023,
|
|
"mean_token_accuracy": 0.15360118597745895,
|
|
"num_tokens": 4436994.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.558993625640869,
|
|
"epoch": 1.8386714116251484,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004939489206695241,
|
|
"loss": 5.413,
|
|
"mean_token_accuracy": 0.1617772325873375,
|
|
"num_tokens": 4450429.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.645631742477417,
|
|
"epoch": 1.8446026097271648,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004938387068230021,
|
|
"loss": 5.5741,
|
|
"mean_token_accuracy": 0.15571037530899048,
|
|
"num_tokens": 4465991.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.595277976989746,
|
|
"epoch": 1.8505338078291815,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004937275122318902,
|
|
"loss": 5.4806,
|
|
"mean_token_accuracy": 0.160708387196064,
|
|
"num_tokens": 4480873.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.6150984287261965,
|
|
"epoch": 1.8564650059311982,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004936153373945165,
|
|
"loss": 5.5424,
|
|
"mean_token_accuracy": 0.16084490567445756,
|
|
"num_tokens": 4493443.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.64046573638916,
|
|
"epoch": 1.8623962040332147,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004935021828136025,
|
|
"loss": 5.5217,
|
|
"mean_token_accuracy": 0.1507852181792259,
|
|
"num_tokens": 4506457.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.53718523979187,
|
|
"epoch": 1.8683274021352312,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004933880489962602,
|
|
"loss": 5.445,
|
|
"mean_token_accuracy": 0.15812863707542418,
|
|
"num_tokens": 4520529.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.561793470382691,
|
|
"epoch": 1.8742586002372479,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004932729364539901,
|
|
"loss": 5.4115,
|
|
"mean_token_accuracy": 0.15655311942100525,
|
|
"num_tokens": 4535100.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.543294382095337,
|
|
"epoch": 1.8801897983392646,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004931568457026789,
|
|
"loss": 5.5379,
|
|
"mean_token_accuracy": 0.1601419448852539,
|
|
"num_tokens": 4549152.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.5530016899108885,
|
|
"epoch": 1.886120996441281,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004930397772625976,
|
|
"loss": 5.4056,
|
|
"mean_token_accuracy": 0.1641964390873909,
|
|
"num_tokens": 4563302.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.490928220748901,
|
|
"epoch": 1.8920521945432978,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004929217316583985,
|
|
"loss": 5.34,
|
|
"mean_token_accuracy": 0.1666228875517845,
|
|
"num_tokens": 4576770.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.5428135871887205,
|
|
"epoch": 1.8979833926453145,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000492802709419113,
|
|
"loss": 5.5009,
|
|
"mean_token_accuracy": 0.15701414197683333,
|
|
"num_tokens": 4590264.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.607468557357788,
|
|
"epoch": 1.903914590747331,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004926827110781496,
|
|
"loss": 5.4478,
|
|
"mean_token_accuracy": 0.15441692173480986,
|
|
"num_tokens": 4606217.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.482126998901367,
|
|
"epoch": 1.9098457888493474,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004925617371732914,
|
|
"loss": 5.3745,
|
|
"mean_token_accuracy": 0.16097915470600127,
|
|
"num_tokens": 4620460.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.577009773254394,
|
|
"epoch": 1.9157769869513643,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004924397882466933,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.16377745270729066,
|
|
"num_tokens": 4635915.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.577189922332764,
|
|
"epoch": 1.9217081850533808,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004923168648448799,
|
|
"loss": 5.4498,
|
|
"mean_token_accuracy": 0.15955513715744019,
|
|
"num_tokens": 4650789.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.5927036762237545,
|
|
"epoch": 1.9276393831553973,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004921929675187433,
|
|
"loss": 5.5179,
|
|
"mean_token_accuracy": 0.1542101040482521,
|
|
"num_tokens": 4666306.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.58389105796814,
|
|
"epoch": 1.933570581257414,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004920680968235397,
|
|
"loss": 5.5087,
|
|
"mean_token_accuracy": 0.1588361606001854,
|
|
"num_tokens": 4679587.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.609423875808716,
|
|
"epoch": 1.9395017793594307,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004919422533188882,
|
|
"loss": 5.4381,
|
|
"mean_token_accuracy": 0.15847491174936296,
|
|
"num_tokens": 4694237.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.590464782714844,
|
|
"epoch": 1.9454329774614472,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004918154375687673,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.14613832235336305,
|
|
"num_tokens": 4708423.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.63156361579895,
|
|
"epoch": 1.9513641755634639,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004916876501415129,
|
|
"loss": 5.4942,
|
|
"mean_token_accuracy": 0.16041443049907683,
|
|
"num_tokens": 4723039.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.530616712570191,
|
|
"epoch": 1.9572953736654806,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004915588916098151,
|
|
"loss": 5.4245,
|
|
"mean_token_accuracy": 0.15755415111780166,
|
|
"num_tokens": 4735882.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.508607053756714,
|
|
"epoch": 1.963226571767497,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004914291625507164,
|
|
"loss": 5.445,
|
|
"mean_token_accuracy": 0.16593677997589112,
|
|
"num_tokens": 4748810.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.600621747970581,
|
|
"epoch": 1.9691577698695135,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000491298463545609,
|
|
"loss": 5.4631,
|
|
"mean_token_accuracy": 0.15750820338726043,
|
|
"num_tokens": 4763563.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.5426243305206295,
|
|
"epoch": 1.9750889679715302,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004911667951802318,
|
|
"loss": 5.455,
|
|
"mean_token_accuracy": 0.1601806804537773,
|
|
"num_tokens": 4775483.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.510920190811158,
|
|
"epoch": 1.981020166073547,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000491034158044668,
|
|
"loss": 5.4364,
|
|
"mean_token_accuracy": 0.15871407985687255,
|
|
"num_tokens": 4792507.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.565320634841919,
|
|
"epoch": 1.9869513641755634,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004909005527333421,
|
|
"loss": 5.4966,
|
|
"mean_token_accuracy": 0.15744143724441528,
|
|
"num_tokens": 4807722.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.542731189727784,
|
|
"epoch": 1.99288256227758,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000490765979845018,
|
|
"loss": 5.3227,
|
|
"mean_token_accuracy": 0.17180967330932617,
|
|
"num_tokens": 4821129.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.488026714324951,
|
|
"epoch": 1.9988137603795968,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000490630439982796,
|
|
"loss": 5.3816,
|
|
"mean_token_accuracy": 0.16403862237930297,
|
|
"num_tokens": 4835483.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.5563037395477295,
|
|
"epoch": 2.0047449584816133,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004904939337541093,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.17830317765474318,
|
|
"num_tokens": 4848607.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.380071020126342,
|
|
"epoch": 2.0106761565836297,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004903564617707227,
|
|
"loss": 5.1126,
|
|
"mean_token_accuracy": 0.17633208483457566,
|
|
"num_tokens": 4863324.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.556159830093383,
|
|
"epoch": 2.0166073546856467,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004902180246487285,
|
|
"loss": 5.185,
|
|
"mean_token_accuracy": 0.1745128035545349,
|
|
"num_tokens": 4876603.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.477500009536743,
|
|
"epoch": 2.022538552787663,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.000490078623008545,
|
|
"loss": 5.2556,
|
|
"mean_token_accuracy": 0.16603254973888398,
|
|
"num_tokens": 4893733.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.527109003067016,
|
|
"epoch": 2.0284697508896796,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004899382574749124,
|
|
"loss": 5.1965,
|
|
"mean_token_accuracy": 0.16696007400751114,
|
|
"num_tokens": 4908476.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.477317190170288,
|
|
"epoch": 2.0344009489916965,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000489796928676891,
|
|
"loss": 5.1449,
|
|
"mean_token_accuracy": 0.17870160490274428,
|
|
"num_tokens": 4920471.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.451275587081909,
|
|
"epoch": 2.040332147093713,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004896546372478581,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.17599333822727203,
|
|
"num_tokens": 4934984.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.42772421836853,
|
|
"epoch": 2.0462633451957295,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004895113838255049,
|
|
"loss": 5.1951,
|
|
"mean_token_accuracy": 0.17033085972070694,
|
|
"num_tokens": 4949709.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.489536190032959,
|
|
"epoch": 2.052194543297746,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004893671690518343,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.1719989776611328,
|
|
"num_tokens": 4964439.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.408801555633545,
|
|
"epoch": 2.058125741399763,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004892219935731571,
|
|
"loss": 5.1197,
|
|
"mean_token_accuracy": 0.1718905135989189,
|
|
"num_tokens": 4979958.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.451976299285889,
|
|
"epoch": 2.0640569395017794,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004890758580400898,
|
|
"loss": 5.2094,
|
|
"mean_token_accuracy": 0.17419177293777466,
|
|
"num_tokens": 4993034.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.38391227722168,
|
|
"epoch": 2.069988137603796,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004889287631075513,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.1770681068301201,
|
|
"num_tokens": 5008580.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.406116962432861,
|
|
"epoch": 2.0759193357058128,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004887807094347606,
|
|
"loss": 5.1683,
|
|
"mean_token_accuracy": 0.17439881563186646,
|
|
"num_tokens": 5022322.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.54651403427124,
|
|
"epoch": 2.0818505338078293,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004886316976852326,
|
|
"loss": 5.202,
|
|
"mean_token_accuracy": 0.16747494488954545,
|
|
"num_tokens": 5036511.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.410180139541626,
|
|
"epoch": 2.0877817319098457,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004884817285267766,
|
|
"loss": 5.1669,
|
|
"mean_token_accuracy": 0.1763795480132103,
|
|
"num_tokens": 5050129.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.358382844924927,
|
|
"epoch": 2.093712930011862,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004883308026314923,
|
|
"loss": 5.0373,
|
|
"mean_token_accuracy": 0.18603201508522033,
|
|
"num_tokens": 5062299.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.396565103530884,
|
|
"epoch": 2.099644128113879,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048817892067576726,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.1693236470222473,
|
|
"num_tokens": 5077665.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.330175447463989,
|
|
"epoch": 2.1055753262158956,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004880260833402734,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.17749213576316833,
|
|
"num_tokens": 5092005.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.410255479812622,
|
|
"epoch": 2.111506524317912,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004878722913099645,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.16958235204219818,
|
|
"num_tokens": 5109155.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.37639307975769,
|
|
"epoch": 2.117437722419929,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000487717545274073,
|
|
"loss": 5.1166,
|
|
"mean_token_accuracy": 0.17338277846574784,
|
|
"num_tokens": 5125791.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.381539440155029,
|
|
"epoch": 2.1233689205219455,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004875618459261065,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.17168417274951936,
|
|
"num_tokens": 5139908.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.37563419342041,
|
|
"epoch": 2.129300118623962,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00048740519396384515,
|
|
"loss": 5.1454,
|
|
"mean_token_accuracy": 0.1780998319387436,
|
|
"num_tokens": 5154922.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.335973501205444,
|
|
"epoch": 2.135231316725979,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004872475900893382,
|
|
"loss": 5.1869,
|
|
"mean_token_accuracy": 0.17797199934720992,
|
|
"num_tokens": 5169919.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.478775691986084,
|
|
"epoch": 2.1411625148279954,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004870890350089011,
|
|
"loss": 5.168,
|
|
"mean_token_accuracy": 0.16982053816318513,
|
|
"num_tokens": 5186687.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.416435575485229,
|
|
"epoch": 2.147093712930012,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048692952943311214,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.1737753450870514,
|
|
"num_tokens": 5199489.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.461081218719483,
|
|
"epoch": 2.1530249110320283,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004867690740768094,
|
|
"loss": 5.2248,
|
|
"mean_token_accuracy": 0.17353710383176804,
|
|
"num_tokens": 5214549.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.421732711791992,
|
|
"epoch": 2.1589561091340452,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00048660766965908736,
|
|
"loss": 5.205,
|
|
"mean_token_accuracy": 0.17148554623126983,
|
|
"num_tokens": 5228571.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.390435361862183,
|
|
"epoch": 2.1648873072360617,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000486445316903294,
|
|
"loss": 5.1314,
|
|
"mean_token_accuracy": 0.17618492543697356,
|
|
"num_tokens": 5242797.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.382180500030517,
|
|
"epoch": 2.170818505338078,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000486282016537027,
|
|
"loss": 5.217,
|
|
"mean_token_accuracy": 0.17333617210388183,
|
|
"num_tokens": 5257692.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.446983528137207,
|
|
"epoch": 2.176749703440095,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004861177692921313,
|
|
"loss": 5.2139,
|
|
"mean_token_accuracy": 0.18043139427900315,
|
|
"num_tokens": 5273117.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.36657567024231,
|
|
"epoch": 2.1826809015421116,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000485952575904695,
|
|
"loss": 5.1616,
|
|
"mean_token_accuracy": 0.1691793143749237,
|
|
"num_tokens": 5287232.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.382761287689209,
|
|
"epoch": 2.188612099644128,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004857864371150467,
|
|
"loss": 5.1611,
|
|
"mean_token_accuracy": 0.17534703761339188,
|
|
"num_tokens": 5300049.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.423830842971801,
|
|
"epoch": 2.1945432977461445,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004856193536677515,
|
|
"loss": 5.1635,
|
|
"mean_token_accuracy": 0.174823035299778,
|
|
"num_tokens": 5313811.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.392314052581787,
|
|
"epoch": 2.2004744958481615,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048545132631160856,
|
|
"loss": 5.2345,
|
|
"mean_token_accuracy": 0.17085386961698532,
|
|
"num_tokens": 5326402.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.3459028720855715,
|
|
"epoch": 2.206405693950178,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000485282355799647,
|
|
"loss": 5.0983,
|
|
"mean_token_accuracy": 0.18093341439962388,
|
|
"num_tokens": 5340440.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.335662603378296,
|
|
"epoch": 2.2123368920521944,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004851124428891226,
|
|
"loss": 5.1865,
|
|
"mean_token_accuracy": 0.17553413212299346,
|
|
"num_tokens": 5355271.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.411830902099609,
|
|
"epoch": 2.2182680901542113,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048494158834151485,
|
|
"loss": 5.148,
|
|
"mean_token_accuracy": 0.175830639898777,
|
|
"num_tokens": 5371939.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.293512201309204,
|
|
"epoch": 2.224199288256228,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00048476979292252323,
|
|
"loss": 5.0534,
|
|
"mean_token_accuracy": 0.1873798117041588,
|
|
"num_tokens": 5386242.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.341111183166504,
|
|
"epoch": 2.2301304863582443,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004845970574020637,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.17497328966856002,
|
|
"num_tokens": 5399867.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.310100793838501,
|
|
"epoch": 2.2360616844602608,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048442338255426555,
|
|
"loss": 5.0383,
|
|
"mean_token_accuracy": 0.18003387302160262,
|
|
"num_tokens": 5414236.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.361874961853028,
|
|
"epoch": 2.2419928825622777,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004842487691574675,
|
|
"loss": 5.1712,
|
|
"mean_token_accuracy": 0.17581834644079208,
|
|
"num_tokens": 5428510.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.443099546432495,
|
|
"epoch": 2.247924080664294,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004840732179942146,
|
|
"loss": 5.1684,
|
|
"mean_token_accuracy": 0.17391909658908844,
|
|
"num_tokens": 5443137.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.333987522125244,
|
|
"epoch": 2.2538552787663106,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00048389672985125467,
|
|
"loss": 5.1889,
|
|
"mean_token_accuracy": 0.17397797256708145,
|
|
"num_tokens": 5456574.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.3122971534729,
|
|
"epoch": 2.2597864768683276,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004837193055195344,
|
|
"loss": 5.1688,
|
|
"mean_token_accuracy": 0.1741654947400093,
|
|
"num_tokens": 5472789.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.433049488067627,
|
|
"epoch": 2.265717674970344,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00048354094579419656,
|
|
"loss": 5.1572,
|
|
"mean_token_accuracy": 0.16992865055799483,
|
|
"num_tokens": 5488202.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.345348739624024,
|
|
"epoch": 2.2716488730723605,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00048336165147457537,
|
|
"loss": 5.2325,
|
|
"mean_token_accuracy": 0.1682398408651352,
|
|
"num_tokens": 5502661.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.399437189102173,
|
|
"epoch": 2.277580071174377,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00048318142336419413,
|
|
"loss": 5.1577,
|
|
"mean_token_accuracy": 0.17407260537147523,
|
|
"num_tokens": 5516371.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.323335647583008,
|
|
"epoch": 2.283511269276394,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004830002622707604,
|
|
"loss": 5.1092,
|
|
"mean_token_accuracy": 0.18390558362007142,
|
|
"num_tokens": 5529339.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.33511791229248,
|
|
"epoch": 2.2894424673784104,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048281816900616354,
|
|
"loss": 5.0979,
|
|
"mean_token_accuracy": 0.1778499886393547,
|
|
"num_tokens": 5543173.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.2589012622833256,
|
|
"epoch": 2.295373665480427,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048263514438647015,
|
|
"loss": 5.1118,
|
|
"mean_token_accuracy": 0.17924153059720993,
|
|
"num_tokens": 5557080.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.395978164672852,
|
|
"epoch": 2.301304863582444,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004824511892319209,
|
|
"loss": 5.1571,
|
|
"mean_token_accuracy": 0.1766973093152046,
|
|
"num_tokens": 5571422.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.259165811538696,
|
|
"epoch": 2.3072360616844603,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004822663043669267,
|
|
"loss": 4.9588,
|
|
"mean_token_accuracy": 0.1932467743754387,
|
|
"num_tokens": 5583057.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.324064111709594,
|
|
"epoch": 2.3131672597864767,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048208049062006505,
|
|
"loss": 5.1963,
|
|
"mean_token_accuracy": 0.17319023609161377,
|
|
"num_tokens": 5596201.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.360679006576538,
|
|
"epoch": 2.3190984578884937,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004818937488240764,
|
|
"loss": 5.1391,
|
|
"mean_token_accuracy": 0.17664920836687087,
|
|
"num_tokens": 5612461.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.299524879455566,
|
|
"epoch": 2.32502965599051,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048170607981586023,
|
|
"loss": 5.2067,
|
|
"mean_token_accuracy": 0.17111538350582123,
|
|
"num_tokens": 5628148.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.320257139205933,
|
|
"epoch": 2.3309608540925266,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004815174844364714,
|
|
"loss": 5.0699,
|
|
"mean_token_accuracy": 0.17787946462631227,
|
|
"num_tokens": 5641695.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.34710054397583,
|
|
"epoch": 2.3368920521945435,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004813279635311165,
|
|
"loss": 5.1008,
|
|
"mean_token_accuracy": 0.1845395416021347,
|
|
"num_tokens": 5654902.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.288831996917724,
|
|
"epoch": 2.34282325029656,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048113751794914974,
|
|
"loss": 5.1023,
|
|
"mean_token_accuracy": 0.18733354657888412,
|
|
"num_tokens": 5666899.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.404209423065185,
|
|
"epoch": 2.3487544483985765,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004809461485440694,
|
|
"loss": 5.2098,
|
|
"mean_token_accuracy": 0.17810151129961013,
|
|
"num_tokens": 5682525.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.2773528575897215,
|
|
"epoch": 2.354685646500593,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004807538561735143,
|
|
"loss": 5.1282,
|
|
"mean_token_accuracy": 0.17133836448192596,
|
|
"num_tokens": 5697097.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.2945380210876465,
|
|
"epoch": 2.36061684460261,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048056064169925904,
|
|
"loss": 5.0673,
|
|
"mean_token_accuracy": 0.17804103642702102,
|
|
"num_tokens": 5712340.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.324316120147705,
|
|
"epoch": 2.3665480427046264,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048036650598721125,
|
|
"loss": 5.1811,
|
|
"mean_token_accuracy": 0.1743994101881981,
|
|
"num_tokens": 5726509.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.3393980979919435,
|
|
"epoch": 2.372479240806643,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00048017144990740673,
|
|
"loss": 5.0981,
|
|
"mean_token_accuracy": 0.18119961768388748,
|
|
"num_tokens": 5740977.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.372479240806643,
|
|
"eval_entropy": 5.177864376021309,
|
|
"eval_loss": 5.554720401763916,
|
|
"eval_mean_token_accuracy": 0.163550903300749,
|
|
"eval_num_tokens": 5740977.0,
|
|
"eval_runtime": 2.4421,
|
|
"eval_samples_per_second": 1067.951,
|
|
"eval_steps_per_second": 133.494,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.240979337692261,
|
|
"epoch": 2.3784104389086593,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047997547433400615,
|
|
"loss": 5.0726,
|
|
"mean_token_accuracy": 0.18319687992334366,
|
|
"num_tokens": 5756000.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.333919858932495,
|
|
"epoch": 2.3843416370106763,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004797785801452911,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.18067850768566132,
|
|
"num_tokens": 5771438.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.343159294128418,
|
|
"epoch": 2.3902728351126927,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047958076822365994,
|
|
"loss": 5.1256,
|
|
"mean_token_accuracy": 0.17654556334018706,
|
|
"num_tokens": 5784686.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.298641729354858,
|
|
"epoch": 2.396204033214709,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004793820394556238,
|
|
"loss": 5.1877,
|
|
"mean_token_accuracy": 0.17813650369644166,
|
|
"num_tokens": 5799363.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.330705642700195,
|
|
"epoch": 2.402135231316726,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004791823947318029,
|
|
"loss": 5.0637,
|
|
"mean_token_accuracy": 0.18452088087797164,
|
|
"num_tokens": 5812109.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.241121292114258,
|
|
"epoch": 2.4080664294187426,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004789818349469222,
|
|
"loss": 5.0892,
|
|
"mean_token_accuracy": 0.18050483465194703,
|
|
"num_tokens": 5825780.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.275242710113526,
|
|
"epoch": 2.413997627520759,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047878036099980795,
|
|
"loss": 5.111,
|
|
"mean_token_accuracy": 0.1737181141972542,
|
|
"num_tokens": 5841184.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.307240915298462,
|
|
"epoch": 2.419928825622776,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047857797379338284,
|
|
"loss": 5.0999,
|
|
"mean_token_accuracy": 0.18021198958158494,
|
|
"num_tokens": 5854422.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.3930497646331785,
|
|
"epoch": 2.4258600237247925,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00047837467423466284,
|
|
"loss": 5.2512,
|
|
"mean_token_accuracy": 0.16936945021152497,
|
|
"num_tokens": 5870878.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.318941688537597,
|
|
"epoch": 2.431791221826809,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00047817046323475246,
|
|
"loss": 5.1895,
|
|
"mean_token_accuracy": 0.16959331035614014,
|
|
"num_tokens": 5886137.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.306102466583252,
|
|
"epoch": 2.4377224199288254,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004779653417088408,
|
|
"loss": 5.1172,
|
|
"mean_token_accuracy": 0.17597496807575225,
|
|
"num_tokens": 5900300.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.342070722579956,
|
|
"epoch": 2.4436536180308424,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00047775931057619786,
|
|
"loss": 5.1785,
|
|
"mean_token_accuracy": 0.17496780306100845,
|
|
"num_tokens": 5917933.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.405538606643677,
|
|
"epoch": 2.449584816132859,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00047755237076016997,
|
|
"loss": 5.1268,
|
|
"mean_token_accuracy": 0.17876010686159133,
|
|
"num_tokens": 5932189.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.274222469329834,
|
|
"epoch": 2.4555160142348753,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004773445231881757,
|
|
"loss": 5.1389,
|
|
"mean_token_accuracy": 0.17788672298192978,
|
|
"num_tokens": 5945999.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.319956541061401,
|
|
"epoch": 2.4614472123368922,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004771357687917021,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.1796871840953827,
|
|
"num_tokens": 5960187.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.2961443901062015,
|
|
"epoch": 2.4673784104389087,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004769261085063,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.1815377578139305,
|
|
"num_tokens": 5975211.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.280208873748779,
|
|
"epoch": 2.473309608540925,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047671554327157996,
|
|
"loss": 5.0969,
|
|
"mean_token_accuracy": 0.1803988590836525,
|
|
"num_tokens": 5988904.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.210264778137207,
|
|
"epoch": 2.4792408066429417,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047650407403120845,
|
|
"loss": 5.0779,
|
|
"mean_token_accuracy": 0.18145141005516052,
|
|
"num_tokens": 6004691.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.305342960357666,
|
|
"epoch": 2.4851720047449586,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047629170173290296,
|
|
"loss": 5.1265,
|
|
"mean_token_accuracy": 0.18025910407304763,
|
|
"num_tokens": 6017859.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.318156719207764,
|
|
"epoch": 2.491103202846975,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004760784273284286,
|
|
"loss": 5.146,
|
|
"mean_token_accuracy": 0.17370295524597168,
|
|
"num_tokens": 6032042.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.273529529571533,
|
|
"epoch": 2.4970344009489915,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004758642517735928,
|
|
"loss": 5.1485,
|
|
"mean_token_accuracy": 0.17880896776914595,
|
|
"num_tokens": 6048591.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.281426286697387,
|
|
"epoch": 2.5029655990510085,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00047564917602824203,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.1841832086443901,
|
|
"num_tokens": 6062436.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.225084066390991,
|
|
"epoch": 2.508896797153025,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004754332010562567,
|
|
"loss": 5.0298,
|
|
"mean_token_accuracy": 0.18615067899227142,
|
|
"num_tokens": 6075564.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.296521472930908,
|
|
"epoch": 2.5148279952550414,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004752163278255473,
|
|
"loss": 5.1303,
|
|
"mean_token_accuracy": 0.17207663059234618,
|
|
"num_tokens": 6088939.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.275126361846924,
|
|
"epoch": 2.5207591933570583,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004749985573080501,
|
|
"loss": 5.0755,
|
|
"mean_token_accuracy": 0.1820480704307556,
|
|
"num_tokens": 6101855.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.281578493118286,
|
|
"epoch": 2.526690391459075,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004747798904797225,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.17975178360939026,
|
|
"num_tokens": 6116192.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.183947134017944,
|
|
"epoch": 2.5326215895610913,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004745603283205387,
|
|
"loss": 5.0631,
|
|
"mean_token_accuracy": 0.18271563649177552,
|
|
"num_tokens": 6130741.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.2799170970916744,
|
|
"epoch": 2.538552787663108,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047433987181448546,
|
|
"loss": 5.0443,
|
|
"mean_token_accuracy": 0.17934627681970597,
|
|
"num_tokens": 6144937.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.258137655258179,
|
|
"epoch": 2.5444839857651247,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004741185219495579,
|
|
"loss": 5.1345,
|
|
"mean_token_accuracy": 0.18690065443515777,
|
|
"num_tokens": 6156747.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.19084677696228,
|
|
"epoch": 2.550415183867141,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004738962797177544,
|
|
"loss": 5.0243,
|
|
"mean_token_accuracy": 0.18932819068431855,
|
|
"num_tokens": 6169148.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.3175774097442625,
|
|
"epoch": 2.5563463819691576,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00047367314611507276,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.1801987111568451,
|
|
"num_tokens": 6181881.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.204585886001587,
|
|
"epoch": 2.562277580071174,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047344912214150565,
|
|
"loss": 5.0433,
|
|
"mean_token_accuracy": 0.18094256818294524,
|
|
"num_tokens": 6196155.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.271319341659546,
|
|
"epoch": 2.568208778173191,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047322420880103566,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.1778367802500725,
|
|
"num_tokens": 6210918.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.210383653640747,
|
|
"epoch": 2.5741399762752075,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00047299840710163157,
|
|
"loss": 5.0483,
|
|
"mean_token_accuracy": 0.18602339327335357,
|
|
"num_tokens": 6225747.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.2173426151275635,
|
|
"epoch": 2.580071174377224,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004727717180552431,
|
|
"loss": 5.1171,
|
|
"mean_token_accuracy": 0.17571157366037368,
|
|
"num_tokens": 6239038.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.3905346393585205,
|
|
"epoch": 2.586002372479241,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00047254414267779686,
|
|
"loss": 5.2252,
|
|
"mean_token_accuracy": 0.17802518457174302,
|
|
"num_tokens": 6253346.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.2353105545043945,
|
|
"epoch": 2.5919335705812574,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004723156819891917,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.17749161124229432,
|
|
"num_tokens": 6268060.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.297154712677002,
|
|
"epoch": 2.597864768683274,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004720863370132937,
|
|
"loss": 5.0243,
|
|
"mean_token_accuracy": 0.18695527017116548,
|
|
"num_tokens": 6283192.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.211258411407471,
|
|
"epoch": 2.603795966785291,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00047185610877793245,
|
|
"loss": 5.086,
|
|
"mean_token_accuracy": 0.1903633087873459,
|
|
"num_tokens": 6297511.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.289752435684204,
|
|
"epoch": 2.6097271648873073,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004716249983148955,
|
|
"loss": 5.0939,
|
|
"mean_token_accuracy": 0.17909670919179915,
|
|
"num_tokens": 6313257.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.252411222457885,
|
|
"epoch": 2.6156583629893237,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004713930066599246,
|
|
"loss": 5.0355,
|
|
"mean_token_accuracy": 0.1826240077614784,
|
|
"num_tokens": 6326437.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.19045057296753,
|
|
"epoch": 2.6215895610913407,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00047116013485271014,
|
|
"loss": 5.0707,
|
|
"mean_token_accuracy": 0.18425143808126448,
|
|
"num_tokens": 6341704.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.245875835418701,
|
|
"epoch": 2.627520759193357,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004709263839368875,
|
|
"loss": 5.084,
|
|
"mean_token_accuracy": 0.18271270245313645,
|
|
"num_tokens": 6355783.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.329879665374756,
|
|
"epoch": 2.6334519572953736,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.00047069175496003147,
|
|
"loss": 5.0741,
|
|
"mean_token_accuracy": 0.17595567405223847,
|
|
"num_tokens": 6370601.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.212981367111206,
|
|
"epoch": 2.6393831553973905,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004704562489736522,
|
|
"loss": 5.12,
|
|
"mean_token_accuracy": 0.17609403133392335,
|
|
"num_tokens": 6388124.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.318413352966308,
|
|
"epoch": 2.645314353499407,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004702198670331901,
|
|
"loss": 5.14,
|
|
"mean_token_accuracy": 0.17445059567689897,
|
|
"num_tokens": 6403018.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.293129968643188,
|
|
"epoch": 2.6512455516014235,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004699826101980112,
|
|
"loss": 5.0751,
|
|
"mean_token_accuracy": 0.18371366560459138,
|
|
"num_tokens": 6417403.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.257935523986816,
|
|
"epoch": 2.65717674970344,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004697444795314027,
|
|
"loss": 5.1255,
|
|
"mean_token_accuracy": 0.17589530646800994,
|
|
"num_tokens": 6432950.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.271436405181885,
|
|
"epoch": 2.6631079478054565,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00046950547610056766,
|
|
"loss": 5.1231,
|
|
"mean_token_accuracy": 0.17779676765203475,
|
|
"num_tokens": 6447735.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.259174013137818,
|
|
"epoch": 2.6690391459074734,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004692656009766206,
|
|
"loss": 5.021,
|
|
"mean_token_accuracy": 0.18835458606481553,
|
|
"num_tokens": 6460066.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.208312892913819,
|
|
"epoch": 2.67497034400949,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046902485523458266,
|
|
"loss": 5.1425,
|
|
"mean_token_accuracy": 0.17693556547164918,
|
|
"num_tokens": 6473880.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.311737155914306,
|
|
"epoch": 2.6809015421115063,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004687832399533767,
|
|
"loss": 5.0631,
|
|
"mean_token_accuracy": 0.1831899255514145,
|
|
"num_tokens": 6485678.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.178772783279419,
|
|
"epoch": 2.6868327402135233,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004685407562158226,
|
|
"loss": 5.1357,
|
|
"mean_token_accuracy": 0.17821813970804215,
|
|
"num_tokens": 6498606.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.229574394226074,
|
|
"epoch": 2.6927639383155397,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004682974051086321,
|
|
"loss": 5.0759,
|
|
"mean_token_accuracy": 0.1833651065826416,
|
|
"num_tokens": 6511590.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.297993230819702,
|
|
"epoch": 2.698695136417556,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004680531877224042,
|
|
"loss": 5.1395,
|
|
"mean_token_accuracy": 0.17965688854455947,
|
|
"num_tokens": 6525391.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.201884508132935,
|
|
"epoch": 2.704626334519573,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00046780810515162027,
|
|
"loss": 5.0758,
|
|
"mean_token_accuracy": 0.18399910628795624,
|
|
"num_tokens": 6540749.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.218594646453857,
|
|
"epoch": 2.7105575326215896,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00046756215849463905,
|
|
"loss": 5.1115,
|
|
"mean_token_accuracy": 0.18162870407104492,
|
|
"num_tokens": 6556607.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.230248975753784,
|
|
"epoch": 2.716488730723606,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046731534885369175,
|
|
"loss": 5.0282,
|
|
"mean_token_accuracy": 0.19147792160511018,
|
|
"num_tokens": 6569340.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.198195695877075,
|
|
"epoch": 2.722419928825623,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046706767733487714,
|
|
"loss": 5.119,
|
|
"mean_token_accuracy": 0.17285093665122986,
|
|
"num_tokens": 6586070.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.373974180221557,
|
|
"epoch": 2.7283511269276395,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00046681914504815653,
|
|
"loss": 5.2172,
|
|
"mean_token_accuracy": 0.1737759843468666,
|
|
"num_tokens": 6601441.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.243975973129272,
|
|
"epoch": 2.734282325029656,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004665697531073489,
|
|
"loss": 5.0781,
|
|
"mean_token_accuracy": 0.18189757764339448,
|
|
"num_tokens": 6616478.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.265233325958252,
|
|
"epoch": 2.7402135231316724,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004663195026301258,
|
|
"loss": 5.1639,
|
|
"mean_token_accuracy": 0.17295905351638793,
|
|
"num_tokens": 6631519.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.298001480102539,
|
|
"epoch": 2.7461447212336894,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00046606839473800626,
|
|
"loss": 5.1306,
|
|
"mean_token_accuracy": 0.17749191373586654,
|
|
"num_tokens": 6646862.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.256069469451904,
|
|
"epoch": 2.752075919335706,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00046581643055635215,
|
|
"loss": 5.0557,
|
|
"mean_token_accuracy": 0.18286117166280746,
|
|
"num_tokens": 6660731.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.240893125534058,
|
|
"epoch": 2.7580071174377223,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046556361121436265,
|
|
"loss": 5.0646,
|
|
"mean_token_accuracy": 0.17607052475214005,
|
|
"num_tokens": 6675922.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.2206097602844235,
|
|
"epoch": 2.763938315539739,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004653099378450696,
|
|
"loss": 5.0654,
|
|
"mean_token_accuracy": 0.17890889644622804,
|
|
"num_tokens": 6689363.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.279681396484375,
|
|
"epoch": 2.7698695136417557,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004650554115853321,
|
|
"loss": 5.0703,
|
|
"mean_token_accuracy": 0.1791287586092949,
|
|
"num_tokens": 6702655.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.2157464027404785,
|
|
"epoch": 2.775800711743772,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.00046480003357583154,
|
|
"loss": 5.0983,
|
|
"mean_token_accuracy": 0.17856834977865219,
|
|
"num_tokens": 6718101.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.2404945373535154,
|
|
"epoch": 2.7817319098457887,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00046454380496106665,
|
|
"loss": 5.0155,
|
|
"mean_token_accuracy": 0.18914393633604049,
|
|
"num_tokens": 6731410.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.237841176986694,
|
|
"epoch": 2.7876631079478056,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004642867268893481,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.18432362377643585,
|
|
"num_tokens": 6744544.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.202240800857544,
|
|
"epoch": 2.793594306049822,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004640288005127933,
|
|
"loss": 4.9961,
|
|
"mean_token_accuracy": 0.19183183163404466,
|
|
"num_tokens": 6757559.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.20791597366333,
|
|
"epoch": 2.7995255041518385,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000463770026987322,
|
|
"loss": 5.0623,
|
|
"mean_token_accuracy": 0.18101912438869477,
|
|
"num_tokens": 6772655.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.163735818862915,
|
|
"epoch": 2.8054567022538555,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004635104074726498,
|
|
"loss": 5.0017,
|
|
"mean_token_accuracy": 0.1892199322581291,
|
|
"num_tokens": 6786038.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.217639446258545,
|
|
"epoch": 2.811387900355872,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004632499431322842,
|
|
"loss": 5.0376,
|
|
"mean_token_accuracy": 0.18138809502124786,
|
|
"num_tokens": 6799759.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.1645301342010494,
|
|
"epoch": 2.8173190984578884,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00046298863513351854,
|
|
"loss": 5.0757,
|
|
"mean_token_accuracy": 0.18388784527778626,
|
|
"num_tokens": 6814242.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.3358196258544925,
|
|
"epoch": 2.8232502965599053,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046272648464742733,
|
|
"loss": 5.1034,
|
|
"mean_token_accuracy": 0.17454032301902772,
|
|
"num_tokens": 6828697.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.126727533340454,
|
|
"epoch": 2.829181494661922,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004624634928488605,
|
|
"loss": 5.0525,
|
|
"mean_token_accuracy": 0.18808865398168564,
|
|
"num_tokens": 6842549.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.260295009613037,
|
|
"epoch": 2.8351126927639383,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046219966091643865,
|
|
"loss": 5.1537,
|
|
"mean_token_accuracy": 0.17728508859872819,
|
|
"num_tokens": 6857599.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.228496980667114,
|
|
"epoch": 2.8410438908659548,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046193499003254737,
|
|
"loss": 5.055,
|
|
"mean_token_accuracy": 0.18687221258878708,
|
|
"num_tokens": 6872621.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.201246309280395,
|
|
"epoch": 2.8469750889679717,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004616694813833322,
|
|
"loss": 5.1423,
|
|
"mean_token_accuracy": 0.1792610228061676,
|
|
"num_tokens": 6887297.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.247128534317016,
|
|
"epoch": 2.852906287069988,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046140313615869287,
|
|
"loss": 5.031,
|
|
"mean_token_accuracy": 0.1847183734178543,
|
|
"num_tokens": 6900857.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.14920654296875,
|
|
"epoch": 2.8588374851720046,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00046113595555227885,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.18829631805419922,
|
|
"num_tokens": 6917351.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.136858177185059,
|
|
"epoch": 2.864768683274021,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000460867940761483,
|
|
"loss": 4.9899,
|
|
"mean_token_accuracy": 0.19460772275924682,
|
|
"num_tokens": 6930619.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.225667905807495,
|
|
"epoch": 2.870699881376038,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004605990929874366,
|
|
"loss": 5.0183,
|
|
"mean_token_accuracy": 0.18654988706111908,
|
|
"num_tokens": 6944842.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.15550389289856,
|
|
"epoch": 2.8766310794780545,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00046032941343500443,
|
|
"loss": 5.016,
|
|
"mean_token_accuracy": 0.1845167189836502,
|
|
"num_tokens": 6957699.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.214073324203492,
|
|
"epoch": 2.882562277580071,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00046005890331277875,
|
|
"loss": 5.1496,
|
|
"mean_token_accuracy": 0.17965820133686067,
|
|
"num_tokens": 6973373.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.286633729934692,
|
|
"epoch": 2.888493475682088,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004597875638330739,
|
|
"loss": 5.0847,
|
|
"mean_token_accuracy": 0.18050620704889297,
|
|
"num_tokens": 6985943.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.2187779426574705,
|
|
"epoch": 2.8944246737841044,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045951539621192137,
|
|
"loss": 5.1014,
|
|
"mean_token_accuracy": 0.19231599122285842,
|
|
"num_tokens": 6999655.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.231914472579956,
|
|
"epoch": 2.900355871886121,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004592424016690637,
|
|
"loss": 5.0696,
|
|
"mean_token_accuracy": 0.18619441241025925,
|
|
"num_tokens": 7014278.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.1600054740905765,
|
|
"epoch": 2.906287069988138,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00045896858142794976,
|
|
"loss": 5.0572,
|
|
"mean_token_accuracy": 0.1820664018392563,
|
|
"num_tokens": 7029998.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.263396883010865,
|
|
"epoch": 2.9122182680901543,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004586939367157285,
|
|
"loss": 5.0585,
|
|
"mean_token_accuracy": 0.18225264996290208,
|
|
"num_tokens": 7043212.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.159766960144043,
|
|
"epoch": 2.9181494661921707,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000458418468763244,
|
|
"loss": 5.0024,
|
|
"mean_token_accuracy": 0.18569402545690536,
|
|
"num_tokens": 7057723.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.208741140365601,
|
|
"epoch": 2.9240806642941877,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045814217880502973,
|
|
"loss": 5.134,
|
|
"mean_token_accuracy": 0.17713429480791093,
|
|
"num_tokens": 7073334.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.21648473739624,
|
|
"epoch": 2.930011862396204,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.000457865068079303,
|
|
"loss": 5.0734,
|
|
"mean_token_accuracy": 0.18073422461748123,
|
|
"num_tokens": 7087671.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.162028169631958,
|
|
"epoch": 2.9359430604982206,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045758713782795946,
|
|
"loss": 4.9354,
|
|
"mean_token_accuracy": 0.19076858162879945,
|
|
"num_tokens": 7101168.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.191947412490845,
|
|
"epoch": 2.941874258600237,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004573083892965677,
|
|
"loss": 5.0882,
|
|
"mean_token_accuracy": 0.18465250134468078,
|
|
"num_tokens": 7115581.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.192853307723999,
|
|
"epoch": 2.9478054567022536,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00045702882373436317,
|
|
"loss": 5.0657,
|
|
"mean_token_accuracy": 0.18381879925727845,
|
|
"num_tokens": 7129948.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.1133428573608395,
|
|
"epoch": 2.9537366548042705,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004567484423942433,
|
|
"loss": 4.9725,
|
|
"mean_token_accuracy": 0.19180593490600586,
|
|
"num_tokens": 7144122.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.158716344833374,
|
|
"epoch": 2.959667852906287,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045646724653276117,
|
|
"loss": 5.0013,
|
|
"mean_token_accuracy": 0.1959048628807068,
|
|
"num_tokens": 7157892.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.198528671264649,
|
|
"epoch": 2.9655990510083035,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00045618523741012044,
|
|
"loss": 5.094,
|
|
"mean_token_accuracy": 0.1803452491760254,
|
|
"num_tokens": 7172581.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.9655990510083035,
|
|
"eval_entropy": 5.126248273381426,
|
|
"eval_loss": 5.422834873199463,
|
|
"eval_mean_token_accuracy": 0.17372721712464936,
|
|
"eval_num_tokens": 7172581.0,
|
|
"eval_runtime": 2.4384,
|
|
"eval_samples_per_second": 1069.552,
|
|
"eval_steps_per_second": 133.694,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.20638074874878,
|
|
"epoch": 2.9715302491103204,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00045590241629016933,
|
|
"loss": 5.0573,
|
|
"mean_token_accuracy": 0.18101746439933777,
|
|
"num_tokens": 7187180.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.160393905639649,
|
|
"epoch": 2.977461447212337,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00045561878444039526,
|
|
"loss": 5.0678,
|
|
"mean_token_accuracy": 0.18904184103012084,
|
|
"num_tokens": 7202153.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.216562271118164,
|
|
"epoch": 2.9833926453143533,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00045533434313191866,
|
|
"loss": 5.0389,
|
|
"mean_token_accuracy": 0.18389637321233748,
|
|
"num_tokens": 7217714.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.061032962799072,
|
|
"epoch": 2.9893238434163703,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004550490936394881,
|
|
"loss": 5.0056,
|
|
"mean_token_accuracy": 0.1876061424612999,
|
|
"num_tokens": 7232366.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.267653942108154,
|
|
"epoch": 2.9952550415183867,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004547630372414737,
|
|
"loss": 5.0137,
|
|
"mean_token_accuracy": 0.18774788379669188,
|
|
"num_tokens": 7245542.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.122905683517456,
|
|
"epoch": 3.001186239620403,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00045447617521986205,
|
|
"loss": 4.9339,
|
|
"mean_token_accuracy": 0.18823694437742233,
|
|
"num_tokens": 7260437.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.162771606445313,
|
|
"epoch": 3.00711743772242,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045418850886025006,
|
|
"loss": 4.7414,
|
|
"mean_token_accuracy": 0.20306676477193833,
|
|
"num_tokens": 7273470.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.117113351821899,
|
|
"epoch": 3.0130486358244366,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004539000394518393,
|
|
"loss": 4.7834,
|
|
"mean_token_accuracy": 0.20288531333208085,
|
|
"num_tokens": 7286515.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.046678256988526,
|
|
"epoch": 3.018979833926453,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045361076828743054,
|
|
"loss": 4.6065,
|
|
"mean_token_accuracy": 0.21320339143276215,
|
|
"num_tokens": 7299011.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.106463003158569,
|
|
"epoch": 3.0249110320284696,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045332069666341733,
|
|
"loss": 4.7103,
|
|
"mean_token_accuracy": 0.2048000767827034,
|
|
"num_tokens": 7312421.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.146386575698853,
|
|
"epoch": 3.0308422301304865,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000453029825879781,
|
|
"loss": 4.7965,
|
|
"mean_token_accuracy": 0.19482676535844803,
|
|
"num_tokens": 7326298.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.225305557250977,
|
|
"epoch": 3.036773428232503,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004527381572400838,
|
|
"loss": 4.7953,
|
|
"mean_token_accuracy": 0.20457307249307632,
|
|
"num_tokens": 7340411.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.1848314762115475,
|
|
"epoch": 3.0427046263345194,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00045244569205146425,
|
|
"loss": 4.7486,
|
|
"mean_token_accuracy": 0.20244074910879134,
|
|
"num_tokens": 7355853.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.010654067993164,
|
|
"epoch": 3.0486358244365364,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00045215243162463023,
|
|
"loss": 4.7147,
|
|
"mean_token_accuracy": 0.21057500094175338,
|
|
"num_tokens": 7370462.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.089919757843018,
|
|
"epoch": 3.054567022538553,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045185837727385373,
|
|
"loss": 4.7126,
|
|
"mean_token_accuracy": 0.20583755373954774,
|
|
"num_tokens": 7382283.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.152230310440063,
|
|
"epoch": 3.0604982206405693,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004515635303169648,
|
|
"loss": 4.7655,
|
|
"mean_token_accuracy": 0.1998978689312935,
|
|
"num_tokens": 7397282.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.08672776222229,
|
|
"epoch": 3.066429418742586,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045126789207534544,
|
|
"loss": 4.7294,
|
|
"mean_token_accuracy": 0.20503358244895936,
|
|
"num_tokens": 7410471.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.0893319606781,
|
|
"epoch": 3.0723606168446027,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00045097146387392423,
|
|
"loss": 4.7244,
|
|
"mean_token_accuracy": 0.20449725836515426,
|
|
"num_tokens": 7426151.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.199353790283203,
|
|
"epoch": 3.078291814946619,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00045067424704116956,
|
|
"loss": 4.7679,
|
|
"mean_token_accuracy": 0.20008099675178528,
|
|
"num_tokens": 7441336.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.0868443012237545,
|
|
"epoch": 3.0842230130486357,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004503762429090845,
|
|
"loss": 4.807,
|
|
"mean_token_accuracy": 0.1966621071100235,
|
|
"num_tokens": 7456152.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.053614568710327,
|
|
"epoch": 3.0901542111506526,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045007745281320035,
|
|
"loss": 4.7119,
|
|
"mean_token_accuracy": 0.20085344314575196,
|
|
"num_tokens": 7469477.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.117205619812012,
|
|
"epoch": 3.096085409252669,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004497778780925706,
|
|
"loss": 4.7087,
|
|
"mean_token_accuracy": 0.20839252173900605,
|
|
"num_tokens": 7482859.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.147538852691651,
|
|
"epoch": 3.1020166073546855,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00044947752008976546,
|
|
"loss": 4.7198,
|
|
"mean_token_accuracy": 0.20925938636064528,
|
|
"num_tokens": 7497021.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.0605854988098145,
|
|
"epoch": 3.1079478054567025,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004491763801508652,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.19931816309690475,
|
|
"num_tokens": 7512528.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.0907529354095455,
|
|
"epoch": 3.113879003558719,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004488744596254545,
|
|
"loss": 4.7824,
|
|
"mean_token_accuracy": 0.19935620278120042,
|
|
"num_tokens": 7529492.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.0850004196167,
|
|
"epoch": 3.1198102016607354,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044857175986661634,
|
|
"loss": 4.7893,
|
|
"mean_token_accuracy": 0.1967274472117424,
|
|
"num_tokens": 7544750.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.144971799850464,
|
|
"epoch": 3.125741399762752,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00044826828223092584,
|
|
"loss": 4.7585,
|
|
"mean_token_accuracy": 0.20286923944950103,
|
|
"num_tokens": 7560436.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.038550949096679,
|
|
"epoch": 3.131672597864769,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004479640280784443,
|
|
"loss": 4.7097,
|
|
"mean_token_accuracy": 0.204975363612175,
|
|
"num_tokens": 7572784.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.098921012878418,
|
|
"epoch": 3.1376037959667853,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044765899877271306,
|
|
"loss": 4.8205,
|
|
"mean_token_accuracy": 0.19938212931156157,
|
|
"num_tokens": 7585677.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.140252733230591,
|
|
"epoch": 3.1435349940688018,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004473531956807471,
|
|
"loss": 4.8292,
|
|
"mean_token_accuracy": 0.19890899807214737,
|
|
"num_tokens": 7600099.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.025925922393799,
|
|
"epoch": 3.1494661921708187,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004470466201730298,
|
|
"loss": 4.748,
|
|
"mean_token_accuracy": 0.19855117797851562,
|
|
"num_tokens": 7615907.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.139107322692871,
|
|
"epoch": 3.155397390272835,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044673927362350545,
|
|
"loss": 4.7794,
|
|
"mean_token_accuracy": 0.20262508243322372,
|
|
"num_tokens": 7629322.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.103339624404907,
|
|
"epoch": 3.1613285883748516,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00044643115740957437,
|
|
"loss": 4.758,
|
|
"mean_token_accuracy": 0.20006387382745744,
|
|
"num_tokens": 7642876.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.127402353286743,
|
|
"epoch": 3.167259786476868,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000446122272912086,
|
|
"loss": 4.6918,
|
|
"mean_token_accuracy": 0.21341730654239655,
|
|
"num_tokens": 7655252.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.068348407745361,
|
|
"epoch": 3.173190984578885,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004458126215153329,
|
|
"loss": 4.7605,
|
|
"mean_token_accuracy": 0.19759204536676406,
|
|
"num_tokens": 7669858.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.0621391296386715,
|
|
"epoch": 3.1791221826809015,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004455022046070446,
|
|
"loss": 4.8024,
|
|
"mean_token_accuracy": 0.19463968724012376,
|
|
"num_tokens": 7684268.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.034128999710083,
|
|
"epoch": 3.185053380782918,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004451910235783814,
|
|
"loss": 4.6857,
|
|
"mean_token_accuracy": 0.21760487854480742,
|
|
"num_tokens": 7696579.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.08457932472229,
|
|
"epoch": 3.190984578884935,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004448790798239278,
|
|
"loss": 4.8018,
|
|
"mean_token_accuracy": 0.19982813745737077,
|
|
"num_tokens": 7712167.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.09724988937378,
|
|
"epoch": 3.1969157769869514,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004445663747416868,
|
|
"loss": 4.7742,
|
|
"mean_token_accuracy": 0.19927359521389007,
|
|
"num_tokens": 7729466.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.006623554229736,
|
|
"epoch": 3.202846975088968,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044425290973307345,
|
|
"loss": 4.664,
|
|
"mean_token_accuracy": 0.2080710217356682,
|
|
"num_tokens": 7741327.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.016892910003662,
|
|
"epoch": 3.208778173190985,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004439386862029081,
|
|
"loss": 4.7402,
|
|
"mean_token_accuracy": 0.21002513021230698,
|
|
"num_tokens": 7757212.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.095934009552002,
|
|
"epoch": 3.2147093712930013,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00044362370555941057,
|
|
"loss": 4.7882,
|
|
"mean_token_accuracy": 0.20233893245458603,
|
|
"num_tokens": 7772726.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.1834417343139645,
|
|
"epoch": 3.2206405693950177,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004433079692141941,
|
|
"loss": 4.8964,
|
|
"mean_token_accuracy": 0.19452748447656631,
|
|
"num_tokens": 7788185.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.090254116058349,
|
|
"epoch": 3.2265717674970342,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00044299147858225835,
|
|
"loss": 4.7831,
|
|
"mean_token_accuracy": 0.200778092443943,
|
|
"num_tokens": 7802653.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.085210752487183,
|
|
"epoch": 3.232502965599051,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004426742350819834,
|
|
"loss": 4.7432,
|
|
"mean_token_accuracy": 0.20108419209718703,
|
|
"num_tokens": 7816898.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.104282283782959,
|
|
"epoch": 3.2384341637010676,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004423562401351234,
|
|
"loss": 4.8442,
|
|
"mean_token_accuracy": 0.19418878853321075,
|
|
"num_tokens": 7833509.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.1720446109771725,
|
|
"epoch": 3.244365361803084,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00044203749516680035,
|
|
"loss": 4.793,
|
|
"mean_token_accuracy": 0.20495709478855134,
|
|
"num_tokens": 7849910.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.0430737972259525,
|
|
"epoch": 3.250296559905101,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004417180016054972,
|
|
"loss": 4.7742,
|
|
"mean_token_accuracy": 0.1986553490161896,
|
|
"num_tokens": 7864555.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.06261420249939,
|
|
"epoch": 3.2562277580071175,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044139776088305215,
|
|
"loss": 4.73,
|
|
"mean_token_accuracy": 0.20505828112363816,
|
|
"num_tokens": 7878850.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.076127576828003,
|
|
"epoch": 3.262158956109134,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00044107677443465165,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.1994879275560379,
|
|
"num_tokens": 7894200.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.0823582172393795,
|
|
"epoch": 3.2680901542111505,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044075504369882416,
|
|
"loss": 4.7966,
|
|
"mean_token_accuracy": 0.1985312283039093,
|
|
"num_tokens": 7908939.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.121370410919189,
|
|
"epoch": 3.2740213523131674,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004404325701174338,
|
|
"loss": 4.827,
|
|
"mean_token_accuracy": 0.1943907380104065,
|
|
"num_tokens": 7924223.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.0246189594268795,
|
|
"epoch": 3.279952550415184,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004401093551356739,
|
|
"loss": 4.7144,
|
|
"mean_token_accuracy": 0.20681486129760743,
|
|
"num_tokens": 7938323.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.062094736099243,
|
|
"epoch": 3.2858837485172003,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004397854002020604,
|
|
"loss": 4.7546,
|
|
"mean_token_accuracy": 0.20174836218357087,
|
|
"num_tokens": 7952126.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.059618139266968,
|
|
"epoch": 3.2918149466192173,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004394607067684251,
|
|
"loss": 4.7671,
|
|
"mean_token_accuracy": 0.20267260819673538,
|
|
"num_tokens": 7965150.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.128559350967407,
|
|
"epoch": 3.2977461447212337,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043913527628990984,
|
|
"loss": 4.8056,
|
|
"mean_token_accuracy": 0.1984943524003029,
|
|
"num_tokens": 7980532.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.027470207214355,
|
|
"epoch": 3.30367734282325,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00043880911022495955,
|
|
"loss": 4.7623,
|
|
"mean_token_accuracy": 0.20342125743627548,
|
|
"num_tokens": 7994731.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.0275311946868895,
|
|
"epoch": 3.309608540925267,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004384822100353155,
|
|
"loss": 4.7388,
|
|
"mean_token_accuracy": 0.20851449072360992,
|
|
"num_tokens": 8009405.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 4.976219892501831,
|
|
"epoch": 3.3155397390272836,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043815457718600933,
|
|
"loss": 4.6718,
|
|
"mean_token_accuracy": 0.20388849526643754,
|
|
"num_tokens": 8023683.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 4.947156524658203,
|
|
"epoch": 3.3214709371293,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004378262131453558,
|
|
"loss": 4.7079,
|
|
"mean_token_accuracy": 0.20666873902082444,
|
|
"num_tokens": 8038219.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.0434857368469235,
|
|
"epoch": 3.3274021352313166,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043749711938494693,
|
|
"loss": 4.7111,
|
|
"mean_token_accuracy": 0.20576132237911224,
|
|
"num_tokens": 8052037.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.043458557128906,
|
|
"epoch": 3.3333333333333335,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00043716729737964475,
|
|
"loss": 4.7646,
|
|
"mean_token_accuracy": 0.2000156119465828,
|
|
"num_tokens": 8065346.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.0696159362792965,
|
|
"epoch": 3.33926453143535,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000436836748607575,
|
|
"loss": 4.7663,
|
|
"mean_token_accuracy": 0.19980355650186538,
|
|
"num_tokens": 8081411.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.084588527679443,
|
|
"epoch": 3.3451957295373664,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004365054745501209,
|
|
"loss": 4.823,
|
|
"mean_token_accuracy": 0.20078919380903243,
|
|
"num_tokens": 8096979.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.026589012145996,
|
|
"epoch": 3.351126927639383,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004361734766919156,
|
|
"loss": 4.8067,
|
|
"mean_token_accuracy": 0.1944137543439865,
|
|
"num_tokens": 8110573.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.040169191360474,
|
|
"epoch": 3.3570581257414,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004358407565208361,
|
|
"loss": 4.8322,
|
|
"mean_token_accuracy": 0.20268825739622115,
|
|
"num_tokens": 8125529.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.094412231445313,
|
|
"epoch": 3.3629893238434163,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00043550731552799676,
|
|
"loss": 4.8085,
|
|
"mean_token_accuracy": 0.19715405702590943,
|
|
"num_tokens": 8141042.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.031404590606689,
|
|
"epoch": 3.368920521945433,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004351731552077423,
|
|
"loss": 4.7318,
|
|
"mean_token_accuracy": 0.20163578242063523,
|
|
"num_tokens": 8153157.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 4.992622947692871,
|
|
"epoch": 3.3748517200474497,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00043483827705764094,
|
|
"loss": 4.7184,
|
|
"mean_token_accuracy": 0.21184460073709488,
|
|
"num_tokens": 8168570.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.079294204711914,
|
|
"epoch": 3.380782918149466,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043450268257847814,
|
|
"loss": 4.7905,
|
|
"mean_token_accuracy": 0.20020936131477357,
|
|
"num_tokens": 8182041.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.038019514083862,
|
|
"epoch": 3.3867141162514827,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00043416637327424955,
|
|
"loss": 4.7922,
|
|
"mean_token_accuracy": 0.20409162491559982,
|
|
"num_tokens": 8196466.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.047460842132568,
|
|
"epoch": 3.3926453143534996,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043382935065215444,
|
|
"loss": 4.7372,
|
|
"mean_token_accuracy": 0.20662083625793456,
|
|
"num_tokens": 8210847.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 4.973863172531128,
|
|
"epoch": 3.398576512455516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004334916162225889,
|
|
"loss": 4.6462,
|
|
"mean_token_accuracy": 0.21235304921865464,
|
|
"num_tokens": 8225575.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.033106279373169,
|
|
"epoch": 3.4045077105575325,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00043315317149913876,
|
|
"loss": 4.782,
|
|
"mean_token_accuracy": 0.2004644051194191,
|
|
"num_tokens": 8239881.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.020281982421875,
|
|
"epoch": 3.4104389086595495,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043281401799857356,
|
|
"loss": 4.7943,
|
|
"mean_token_accuracy": 0.20126855075359346,
|
|
"num_tokens": 8253692.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.035511779785156,
|
|
"epoch": 3.416370106761566,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00043247415724083886,
|
|
"loss": 4.7438,
|
|
"mean_token_accuracy": 0.2059497982263565,
|
|
"num_tokens": 8268322.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.014690589904785,
|
|
"epoch": 3.4223013048635824,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004321335907490502,
|
|
"loss": 4.7463,
|
|
"mean_token_accuracy": 0.20121788382530212,
|
|
"num_tokens": 8285317.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.039463233947754,
|
|
"epoch": 3.428232502965599,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004317923200494856,
|
|
"loss": 4.7651,
|
|
"mean_token_accuracy": 0.20686009228229524,
|
|
"num_tokens": 8299429.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 4.999642324447632,
|
|
"epoch": 3.434163701067616,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00043145034667157935,
|
|
"loss": 4.7808,
|
|
"mean_token_accuracy": 0.19736639112234117,
|
|
"num_tokens": 8314039.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.063088703155517,
|
|
"epoch": 3.4400948991696323,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00043110767214791466,
|
|
"loss": 4.8314,
|
|
"mean_token_accuracy": 0.19301424771547318,
|
|
"num_tokens": 8328634.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.02191686630249,
|
|
"epoch": 3.4460260972716488,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004307642980142171,
|
|
"loss": 4.7383,
|
|
"mean_token_accuracy": 0.20328510850667952,
|
|
"num_tokens": 8342448.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.046097278594971,
|
|
"epoch": 3.4519572953736652,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004304202258093473,
|
|
"loss": 4.7129,
|
|
"mean_token_accuracy": 0.20647248327732087,
|
|
"num_tokens": 8354450.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 4.973519659042358,
|
|
"epoch": 3.457888493475682,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00043007545707529496,
|
|
"loss": 4.7734,
|
|
"mean_token_accuracy": 0.20650199204683303,
|
|
"num_tokens": 8369054.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.056032657623291,
|
|
"epoch": 3.4638196915776986,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004297299933571708,
|
|
"loss": 4.7878,
|
|
"mean_token_accuracy": 0.20445864796638488,
|
|
"num_tokens": 8383401.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.056894445419312,
|
|
"epoch": 3.469750889679715,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004293838362032003,
|
|
"loss": 4.8094,
|
|
"mean_token_accuracy": 0.19751561880111695,
|
|
"num_tokens": 8396990.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.085438299179077,
|
|
"epoch": 3.475682087781732,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00042903698716471677,
|
|
"loss": 4.7315,
|
|
"mean_token_accuracy": 0.20365943461656572,
|
|
"num_tokens": 8411212.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.031349086761475,
|
|
"epoch": 3.4816132858837485,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004286894477961541,
|
|
"loss": 4.7843,
|
|
"mean_token_accuracy": 0.19990857392549516,
|
|
"num_tokens": 8424669.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 4.983600616455078,
|
|
"epoch": 3.487544483985765,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042834121965504014,
|
|
"loss": 4.75,
|
|
"mean_token_accuracy": 0.2036602795124054,
|
|
"num_tokens": 8437122.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.066770315170288,
|
|
"epoch": 3.493475682087782,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00042799230430198935,
|
|
"loss": 4.7912,
|
|
"mean_token_accuracy": 0.1955005034804344,
|
|
"num_tokens": 8450543.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.028520488739014,
|
|
"epoch": 3.4994068801897984,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000427642703300696,
|
|
"loss": 4.741,
|
|
"mean_token_accuracy": 0.20334245413541793,
|
|
"num_tokens": 8464720.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.056568956375122,
|
|
"epoch": 3.505338078291815,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00042729241821792726,
|
|
"loss": 4.8167,
|
|
"mean_token_accuracy": 0.20227830111980438,
|
|
"num_tokens": 8478547.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.136375427246094,
|
|
"epoch": 3.511269276393832,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000426941450623516,
|
|
"loss": 4.8482,
|
|
"mean_token_accuracy": 0.20175531059503554,
|
|
"num_tokens": 8494126.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.042528486251831,
|
|
"epoch": 3.5172004744958483,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042658980209035394,
|
|
"loss": 4.8203,
|
|
"mean_token_accuracy": 0.19100423008203507,
|
|
"num_tokens": 8508575.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 4.972239303588867,
|
|
"epoch": 3.5231316725978647,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00042623747419438427,
|
|
"loss": 4.7206,
|
|
"mean_token_accuracy": 0.20315259099006652,
|
|
"num_tokens": 8523589.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.1017388820648195,
|
|
"epoch": 3.5290628706998812,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000425884468514595,
|
|
"loss": 4.8072,
|
|
"mean_token_accuracy": 0.2059174284338951,
|
|
"num_tokens": 8538238.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 4.981739139556884,
|
|
"epoch": 3.534994068801898,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042553078663301163,
|
|
"loss": 4.7648,
|
|
"mean_token_accuracy": 0.21039492785930633,
|
|
"num_tokens": 8551959.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.042350244522095,
|
|
"epoch": 3.5409252669039146,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00042517643013469,
|
|
"loss": 4.7703,
|
|
"mean_token_accuracy": 0.20805414766073227,
|
|
"num_tokens": 8564628.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.043861818313599,
|
|
"epoch": 3.546856465005931,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004248214006077095,
|
|
"loss": 4.7922,
|
|
"mean_token_accuracy": 0.2033117890357971,
|
|
"num_tokens": 8579342.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.042752981185913,
|
|
"epoch": 3.5527876631079476,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042446569964316557,
|
|
"loss": 4.761,
|
|
"mean_token_accuracy": 0.20291885286569594,
|
|
"num_tokens": 8595096.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.157862043380737,
|
|
"epoch": 3.5587188612099645,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00042410932883516297,
|
|
"loss": 4.8657,
|
|
"mean_token_accuracy": 0.19814113080501555,
|
|
"num_tokens": 8608023.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 3.5587188612099645,
|
|
"eval_entropy": 4.920296645603297,
|
|
"eval_loss": 5.375751972198486,
|
|
"eval_mean_token_accuracy": 0.1804342555350687,
|
|
"eval_num_tokens": 8608023.0,
|
|
"eval_runtime": 2.4405,
|
|
"eval_samples_per_second": 1068.642,
|
|
"eval_steps_per_second": 133.58,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.008726119995117,
|
|
"epoch": 3.564650059311981,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00042375228978080813,
|
|
"loss": 4.7275,
|
|
"mean_token_accuracy": 0.2046857789158821,
|
|
"num_tokens": 8623529.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.011298274993896,
|
|
"epoch": 3.5705812574139975,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042339458408020267,
|
|
"loss": 4.797,
|
|
"mean_token_accuracy": 0.20083173513412475,
|
|
"num_tokens": 8639007.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 4.97677264213562,
|
|
"epoch": 3.5765124555160144,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004230362133364354,
|
|
"loss": 4.708,
|
|
"mean_token_accuracy": 0.20730987638235093,
|
|
"num_tokens": 8653678.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.092321729660034,
|
|
"epoch": 3.582443653618031,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004226771791555758,
|
|
"loss": 4.8702,
|
|
"mean_token_accuracy": 0.19500762373209,
|
|
"num_tokens": 8668663.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.058094787597656,
|
|
"epoch": 3.5883748517200473,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00042231748314666684,
|
|
"loss": 4.7956,
|
|
"mean_token_accuracy": 0.19681732058525087,
|
|
"num_tokens": 8682121.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 4.961966848373413,
|
|
"epoch": 3.5943060498220643,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00042195712692171705,
|
|
"loss": 4.7837,
|
|
"mean_token_accuracy": 0.20123348832130433,
|
|
"num_tokens": 8698409.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.056216430664063,
|
|
"epoch": 3.6002372479240807,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00042159611209569416,
|
|
"loss": 4.8095,
|
|
"mean_token_accuracy": 0.1971350535750389,
|
|
"num_tokens": 8714076.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.043015623092652,
|
|
"epoch": 3.606168446026097,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00042123444028651725,
|
|
"loss": 4.742,
|
|
"mean_token_accuracy": 0.20921751856803894,
|
|
"num_tokens": 8730482.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 4.960888481140136,
|
|
"epoch": 3.612099644128114,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00042087211311504986,
|
|
"loss": 4.7434,
|
|
"mean_token_accuracy": 0.21072260439395904,
|
|
"num_tokens": 8744036.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.049616146087646,
|
|
"epoch": 3.6180308422301306,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042050913220509253,
|
|
"loss": 4.7923,
|
|
"mean_token_accuracy": 0.19877566695213317,
|
|
"num_tokens": 8760458.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.006571483612061,
|
|
"epoch": 3.623962040332147,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042014549918337554,
|
|
"loss": 4.7508,
|
|
"mean_token_accuracy": 0.20669716000556945,
|
|
"num_tokens": 8774946.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 4.989468812942505,
|
|
"epoch": 3.6298932384341636,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004197812156795518,
|
|
"loss": 4.7373,
|
|
"mean_token_accuracy": 0.20692423433065416,
|
|
"num_tokens": 8790918.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.002850151062011,
|
|
"epoch": 3.63582443653618,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004194162833261893,
|
|
"loss": 4.7954,
|
|
"mean_token_accuracy": 0.19880906939506532,
|
|
"num_tokens": 8803505.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.040811491012573,
|
|
"epoch": 3.641755634638197,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004190507037587639,
|
|
"loss": 4.8397,
|
|
"mean_token_accuracy": 0.1974925011396408,
|
|
"num_tokens": 8819567.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.102786254882813,
|
|
"epoch": 3.6476868327402134,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004186844786156522,
|
|
"loss": 4.7532,
|
|
"mean_token_accuracy": 0.19830316305160522,
|
|
"num_tokens": 8834257.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 4.980304431915283,
|
|
"epoch": 3.65361803084223,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004183176095381238,
|
|
"loss": 4.7665,
|
|
"mean_token_accuracy": 0.21026202738285066,
|
|
"num_tokens": 8847866.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.039072322845459,
|
|
"epoch": 3.659549228944247,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000417950098170334,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.2039116784930229,
|
|
"num_tokens": 8863297.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 4.99145097732544,
|
|
"epoch": 3.6654804270462633,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000417581946159317,
|
|
"loss": 4.7819,
|
|
"mean_token_accuracy": 0.20517824739217758,
|
|
"num_tokens": 8876605.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 4.928086805343628,
|
|
"epoch": 3.67141162514828,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004172131551549778,
|
|
"loss": 4.7079,
|
|
"mean_token_accuracy": 0.20931409150362015,
|
|
"num_tokens": 8892040.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 4.984323310852051,
|
|
"epoch": 3.6773428232502967,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00041684372681008506,
|
|
"loss": 4.7896,
|
|
"mean_token_accuracy": 0.20277808755636215,
|
|
"num_tokens": 8905806.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 4.949889278411865,
|
|
"epoch": 3.683274021352313,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041647366278026415,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.20926615446805955,
|
|
"num_tokens": 8920638.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 4.986402797698974,
|
|
"epoch": 3.6892052194543297,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00041610296472398875,
|
|
"loss": 4.7272,
|
|
"mean_token_accuracy": 0.20068549513816833,
|
|
"num_tokens": 8933501.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.0357715606689455,
|
|
"epoch": 3.6951364175563466,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.00041573163430257435,
|
|
"loss": 4.8195,
|
|
"mean_token_accuracy": 0.20384510308504106,
|
|
"num_tokens": 8948909.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.051932430267334,
|
|
"epoch": 3.701067615658363,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041535967318017037,
|
|
"loss": 4.7816,
|
|
"mean_token_accuracy": 0.20459156930446626,
|
|
"num_tokens": 8961611.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 4.94138617515564,
|
|
"epoch": 3.7069988137603795,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00041498708302375277,
|
|
"loss": 4.6844,
|
|
"mean_token_accuracy": 0.21273948550224303,
|
|
"num_tokens": 8975993.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 4.9991271018981935,
|
|
"epoch": 3.7129300118623965,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004146138655031165,
|
|
"loss": 4.7616,
|
|
"mean_token_accuracy": 0.19694194197654724,
|
|
"num_tokens": 8989198.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 4.9471393585205075,
|
|
"epoch": 3.718861209964413,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004142400222908681,
|
|
"loss": 4.78,
|
|
"mean_token_accuracy": 0.20367717146873474,
|
|
"num_tokens": 9004592.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.004219198226929,
|
|
"epoch": 3.7247924080664294,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004138655550624183,
|
|
"loss": 4.755,
|
|
"mean_token_accuracy": 0.19702138751745224,
|
|
"num_tokens": 9017772.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 4.996582460403443,
|
|
"epoch": 3.730723606168446,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004134904654959745,
|
|
"loss": 4.6673,
|
|
"mean_token_accuracy": 0.21102655231952666,
|
|
"num_tokens": 9031182.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 4.990418148040772,
|
|
"epoch": 3.7366548042704624,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00041311475527253284,
|
|
"loss": 4.7594,
|
|
"mean_token_accuracy": 0.1996576502919197,
|
|
"num_tokens": 9044934.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 4.958402919769287,
|
|
"epoch": 3.7425860023724793,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041273842607587137,
|
|
"loss": 4.7585,
|
|
"mean_token_accuracy": 0.20037429183721542,
|
|
"num_tokens": 9059795.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 4.948233509063721,
|
|
"epoch": 3.7485172004744958,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00041236147959254166,
|
|
"loss": 4.7427,
|
|
"mean_token_accuracy": 0.20549604296684265,
|
|
"num_tokens": 9073410.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 4.99368896484375,
|
|
"epoch": 3.7544483985765122,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00041198391751186227,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.1967238664627075,
|
|
"num_tokens": 9089338.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.035021591186523,
|
|
"epoch": 3.760379596678529,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004116057415259101,
|
|
"loss": 4.7213,
|
|
"mean_token_accuracy": 0.20836372971534728,
|
|
"num_tokens": 9103520.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.016361999511719,
|
|
"epoch": 3.7663107947805456,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00041122695332951363,
|
|
"loss": 4.7366,
|
|
"mean_token_accuracy": 0.20817891359329224,
|
|
"num_tokens": 9117935.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 4.941559505462647,
|
|
"epoch": 3.772241992882562,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004108475546202449,
|
|
"loss": 4.724,
|
|
"mean_token_accuracy": 0.210002501308918,
|
|
"num_tokens": 9132457.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.0450352191925045,
|
|
"epoch": 3.778173190984579,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004104675470984121,
|
|
"loss": 4.7728,
|
|
"mean_token_accuracy": 0.20576235949993132,
|
|
"num_tokens": 9147329.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 4.97810378074646,
|
|
"epoch": 3.7841043890865955,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004100869324670517,
|
|
"loss": 4.7348,
|
|
"mean_token_accuracy": 0.2090249091386795,
|
|
"num_tokens": 9161052.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 4.998898649215699,
|
|
"epoch": 3.790035587188612,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004097057124319212,
|
|
"loss": 4.7872,
|
|
"mean_token_accuracy": 0.20106951892375946,
|
|
"num_tokens": 9175900.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 4.991289758682251,
|
|
"epoch": 3.795966785290629,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000409323888701491,
|
|
"loss": 4.74,
|
|
"mean_token_accuracy": 0.20916445255279542,
|
|
"num_tokens": 9189706.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 4.998389005661011,
|
|
"epoch": 3.8018979833926454,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004089414629869373,
|
|
"loss": 4.8134,
|
|
"mean_token_accuracy": 0.1990466296672821,
|
|
"num_tokens": 9203727.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.003602933883667,
|
|
"epoch": 3.807829181494662,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00040855843700213386,
|
|
"loss": 4.758,
|
|
"mean_token_accuracy": 0.20540229827165604,
|
|
"num_tokens": 9217664.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 4.920092439651489,
|
|
"epoch": 3.813760379596679,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004081748124636449,
|
|
"loss": 4.6756,
|
|
"mean_token_accuracy": 0.2066286250948906,
|
|
"num_tokens": 9233474.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 4.991271114349365,
|
|
"epoch": 3.8196915776986953,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00040779059109071706,
|
|
"loss": 4.7515,
|
|
"mean_token_accuracy": 0.20587924718856812,
|
|
"num_tokens": 9247567.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 4.999785566329956,
|
|
"epoch": 3.8256227758007118,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004074057746052713,
|
|
"loss": 4.7512,
|
|
"mean_token_accuracy": 0.20081204026937485,
|
|
"num_tokens": 9262264.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.077322340011596,
|
|
"epoch": 3.8315539739027282,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004070203647318963,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.18915841430425645,
|
|
"num_tokens": 9276893.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.001492691040039,
|
|
"epoch": 3.8374851720047447,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00040663436319783965,
|
|
"loss": 4.7647,
|
|
"mean_token_accuracy": 0.20067933052778245,
|
|
"num_tokens": 9291646.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 4.994247627258301,
|
|
"epoch": 3.8434163701067616,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004062477717330006,
|
|
"loss": 4.7783,
|
|
"mean_token_accuracy": 0.2093625172972679,
|
|
"num_tokens": 9305789.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.001421499252319,
|
|
"epoch": 3.849347568208778,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040586059206992214,
|
|
"loss": 4.7153,
|
|
"mean_token_accuracy": 0.20331584066152572,
|
|
"num_tokens": 9321039.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 4.9822968482971195,
|
|
"epoch": 3.8552787663107946,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004054728259437836,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.20271218568086624,
|
|
"num_tokens": 9337105.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.014848947525024,
|
|
"epoch": 3.8612099644128115,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040508447509239224,
|
|
"loss": 4.7738,
|
|
"mean_token_accuracy": 0.20225811898708343,
|
|
"num_tokens": 9350684.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.001141357421875,
|
|
"epoch": 3.867141162514828,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000404695541256176,
|
|
"loss": 4.7764,
|
|
"mean_token_accuracy": 0.19928564578294755,
|
|
"num_tokens": 9365074.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.038498115539551,
|
|
"epoch": 3.8730723606168445,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004043060261781756,
|
|
"loss": 4.8761,
|
|
"mean_token_accuracy": 0.18968692272901536,
|
|
"num_tokens": 9380912.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.037493324279785,
|
|
"epoch": 3.8790035587188614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040391593160403656,
|
|
"loss": 4.7994,
|
|
"mean_token_accuracy": 0.20988689661026,
|
|
"num_tokens": 9393812.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 4.93535304069519,
|
|
"epoch": 3.884934756820878,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040352525928200137,
|
|
"loss": 4.7582,
|
|
"mean_token_accuracy": 0.20945370346307754,
|
|
"num_tokens": 9409731.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 4.952421045303344,
|
|
"epoch": 3.8908659549228943,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004031340109629017,
|
|
"loss": 4.736,
|
|
"mean_token_accuracy": 0.20832393765449525,
|
|
"num_tokens": 9422743.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 4.997214126586914,
|
|
"epoch": 3.8967971530249113,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000402742188400151,
|
|
"loss": 4.7868,
|
|
"mean_token_accuracy": 0.2028869390487671,
|
|
"num_tokens": 9436440.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 4.9821700096130375,
|
|
"epoch": 3.9027283511269277,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00040234979334973577,
|
|
"loss": 4.814,
|
|
"mean_token_accuracy": 0.19977682679891587,
|
|
"num_tokens": 9451668.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.011359119415284,
|
|
"epoch": 3.908659549228944,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040195682757020846,
|
|
"loss": 4.7295,
|
|
"mean_token_accuracy": 0.2123936116695404,
|
|
"num_tokens": 9464700.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 4.895818471908569,
|
|
"epoch": 3.914590747330961,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00040156329282267915,
|
|
"loss": 4.69,
|
|
"mean_token_accuracy": 0.20839773416519164,
|
|
"num_tokens": 9479003.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 4.989343452453613,
|
|
"epoch": 3.9205219454329776,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00040116919087080787,
|
|
"loss": 4.7299,
|
|
"mean_token_accuracy": 0.2052148088812828,
|
|
"num_tokens": 9493134.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 4.9859777927398685,
|
|
"epoch": 3.926453143534994,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00040077452348079664,
|
|
"loss": 4.728,
|
|
"mean_token_accuracy": 0.20534912198781968,
|
|
"num_tokens": 9507087.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.005017566680908,
|
|
"epoch": 3.9323843416370106,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004003792924213814,
|
|
"loss": 4.731,
|
|
"mean_token_accuracy": 0.21763905882835388,
|
|
"num_tokens": 9518328.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 4.975371599197388,
|
|
"epoch": 3.938315539739027,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0003999834994638243,
|
|
"loss": 4.7808,
|
|
"mean_token_accuracy": 0.20904267728328704,
|
|
"num_tokens": 9532413.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 4.9734986305236815,
|
|
"epoch": 3.944246737841044,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003995871463819059,
|
|
"loss": 4.7308,
|
|
"mean_token_accuracy": 0.20565488934516907,
|
|
"num_tokens": 9547681.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 4.94563102722168,
|
|
"epoch": 3.9501779359430604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039919023495191665,
|
|
"loss": 4.6864,
|
|
"mean_token_accuracy": 0.2076362982392311,
|
|
"num_tokens": 9560198.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 4.953912162780762,
|
|
"epoch": 3.956109134045077,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0003987927669526494,
|
|
"loss": 4.7803,
|
|
"mean_token_accuracy": 0.2033618852496147,
|
|
"num_tokens": 9574347.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 4.976535940170288,
|
|
"epoch": 3.962040332147094,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039839474416539163,
|
|
"loss": 4.6694,
|
|
"mean_token_accuracy": 0.21107939928770064,
|
|
"num_tokens": 9587337.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 4.960287427902221,
|
|
"epoch": 3.9679715302491103,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039799616837391655,
|
|
"loss": 4.7503,
|
|
"mean_token_accuracy": 0.20462618321180343,
|
|
"num_tokens": 9601047.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.009324312210083,
|
|
"epoch": 3.973902728351127,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039759704136447627,
|
|
"loss": 4.7242,
|
|
"mean_token_accuracy": 0.20350200384855271,
|
|
"num_tokens": 9614164.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 4.973122644424438,
|
|
"epoch": 3.9798339264531437,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003971973649257929,
|
|
"loss": 4.7649,
|
|
"mean_token_accuracy": 0.2030460312962532,
|
|
"num_tokens": 9628179.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 4.955025863647461,
|
|
"epoch": 3.98576512455516,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000396797140849051,
|
|
"loss": 4.7786,
|
|
"mean_token_accuracy": 0.20531781762838364,
|
|
"num_tokens": 9641575.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 4.971503114700317,
|
|
"epoch": 3.9916963226571767,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039639637092788935,
|
|
"loss": 4.6749,
|
|
"mean_token_accuracy": 0.21360119432210922,
|
|
"num_tokens": 9655381.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 4.975626707077026,
|
|
"epoch": 3.9976275207591936,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000395995056958393,
|
|
"loss": 4.7882,
|
|
"mean_token_accuracy": 0.2019601806998253,
|
|
"num_tokens": 9670477.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 4.898117542266846,
|
|
"epoch": 4.00355871886121,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039559320073908526,
|
|
"loss": 4.5278,
|
|
"mean_token_accuracy": 0.22469987720251083,
|
|
"num_tokens": 9683964.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.007152366638183,
|
|
"epoch": 4.0094899169632265,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00039519080407091956,
|
|
"loss": 4.5287,
|
|
"mean_token_accuracy": 0.22204413414001464,
|
|
"num_tokens": 9698415.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 4.971146965026856,
|
|
"epoch": 4.0154211150652435,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.00039478786875727116,
|
|
"loss": 4.5088,
|
|
"mean_token_accuracy": 0.2267781004309654,
|
|
"num_tokens": 9712895.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 4.877899503707885,
|
|
"epoch": 4.0213523131672595,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00039438439660392983,
|
|
"loss": 4.4028,
|
|
"mean_token_accuracy": 0.2324035331606865,
|
|
"num_tokens": 9728872.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 4.971367359161377,
|
|
"epoch": 4.027283511269276,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039398038941909066,
|
|
"loss": 4.4763,
|
|
"mean_token_accuracy": 0.2239493578672409,
|
|
"num_tokens": 9744284.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 4.9583673000335695,
|
|
"epoch": 4.033214709371293,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000393575849013347,
|
|
"loss": 4.4797,
|
|
"mean_token_accuracy": 0.22193355560302735,
|
|
"num_tokens": 9759374.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 4.913420343399048,
|
|
"epoch": 4.039145907473309,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039317077719968143,
|
|
"loss": 4.4557,
|
|
"mean_token_accuracy": 0.22923122644424437,
|
|
"num_tokens": 9773919.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 4.945164203643799,
|
|
"epoch": 4.045077105575326,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039276517579345846,
|
|
"loss": 4.4985,
|
|
"mean_token_accuracy": 0.22622151225805281,
|
|
"num_tokens": 9788023.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 4.96866455078125,
|
|
"epoch": 4.051008303677343,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003923590466124159,
|
|
"loss": 4.5084,
|
|
"mean_token_accuracy": 0.22214894890785217,
|
|
"num_tokens": 9803037.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 4.934285116195679,
|
|
"epoch": 4.056939501779359,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00039195239147665664,
|
|
"loss": 4.4634,
|
|
"mean_token_accuracy": 0.2275235190987587,
|
|
"num_tokens": 9816782.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 4.944174718856812,
|
|
"epoch": 4.062870699881376,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039154521220864094,
|
|
"loss": 4.4868,
|
|
"mean_token_accuracy": 0.22822190821170807,
|
|
"num_tokens": 9830471.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 4.976789569854736,
|
|
"epoch": 4.068801897983393,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00039113751063317795,
|
|
"loss": 4.5377,
|
|
"mean_token_accuracy": 0.21644991934299468,
|
|
"num_tokens": 9844303.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 4.912378787994385,
|
|
"epoch": 4.074733096085409,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003907292885774175,
|
|
"loss": 4.4636,
|
|
"mean_token_accuracy": 0.22525488436222077,
|
|
"num_tokens": 9861002.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 4.828717756271362,
|
|
"epoch": 4.080664294187426,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003903205478708419,
|
|
"loss": 4.4367,
|
|
"mean_token_accuracy": 0.22953445911407472,
|
|
"num_tokens": 9876338.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 4.982810688018799,
|
|
"epoch": 4.086595492289442,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003899112903452582,
|
|
"loss": 4.5384,
|
|
"mean_token_accuracy": 0.2221672788262367,
|
|
"num_tokens": 9891217.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 4.93398118019104,
|
|
"epoch": 4.092526690391459,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038950151783478926,
|
|
"loss": 4.4857,
|
|
"mean_token_accuracy": 0.2276325508952141,
|
|
"num_tokens": 9905309.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 4.932527875900268,
|
|
"epoch": 4.098457888493476,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.000389091232175866,
|
|
"loss": 4.4543,
|
|
"mean_token_accuracy": 0.22840490341186523,
|
|
"num_tokens": 9918347.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 4.88067216873169,
|
|
"epoch": 4.104389086595492,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003886804352072192,
|
|
"loss": 4.4841,
|
|
"mean_token_accuracy": 0.23047109842300414,
|
|
"num_tokens": 9933068.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 4.864267015457154,
|
|
"epoch": 4.110320284697509,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0003882691287698709,
|
|
"loss": 4.4818,
|
|
"mean_token_accuracy": 0.2262014165520668,
|
|
"num_tokens": 9948149.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 4.919790840148925,
|
|
"epoch": 4.116251482799526,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003878573147071265,
|
|
"loss": 4.4688,
|
|
"mean_token_accuracy": 0.2208899065852165,
|
|
"num_tokens": 9961494.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 4.871296215057373,
|
|
"epoch": 4.122182680901542,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0003874449948645664,
|
|
"loss": 4.4381,
|
|
"mean_token_accuracy": 0.2287437066435814,
|
|
"num_tokens": 9977270.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 4.859757661819458,
|
|
"epoch": 4.128113879003559,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0003870321710900375,
|
|
"loss": 4.3824,
|
|
"mean_token_accuracy": 0.22972583323717116,
|
|
"num_tokens": 9991102.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 4.930823421478271,
|
|
"epoch": 4.134045077105576,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038661884523364537,
|
|
"loss": 4.5163,
|
|
"mean_token_accuracy": 0.22619430869817733,
|
|
"num_tokens": 10004991.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 4.893933343887329,
|
|
"epoch": 4.139976275207592,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038620501914774554,
|
|
"loss": 4.5282,
|
|
"mean_token_accuracy": 0.22893610149621962,
|
|
"num_tokens": 10019654.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 4.885422086715698,
|
|
"epoch": 4.145907473309609,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00038579069468693526,
|
|
"loss": 4.4481,
|
|
"mean_token_accuracy": 0.22396969348192214,
|
|
"num_tokens": 10033831.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 4.897957801818848,
|
|
"epoch": 4.1518386714116255,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00038537587370804567,
|
|
"loss": 4.4511,
|
|
"mean_token_accuracy": 0.22712469696998597,
|
|
"num_tokens": 10049549.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 4.1518386714116255,
|
|
"eval_entropy": 4.770081594677791,
|
|
"eval_loss": 5.347801685333252,
|
|
"eval_mean_token_accuracy": 0.1851496258122058,
|
|
"eval_num_tokens": 10049549.0,
|
|
"eval_runtime": 2.4419,
|
|
"eval_samples_per_second": 1068.032,
|
|
"eval_steps_per_second": 133.504,
|
|
"step": 3500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 8420,
|
|
"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": 1.771285416192e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|