2119ed983f
Model: fpadovani/urd-arab-100mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.7793594306049823,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 7627040925696000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|