1455 lines
41 KiB
JSON
1455 lines
41 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 3.0,
|
||
|
|
"eval_steps": 500,
|
||
|
|
"global_step": 1425,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"entropy": 0.8691443920135498,
|
||
|
|
"epoch": 0.021085925144965736,
|
||
|
|
"grad_norm": 47.5,
|
||
|
|
"learning_rate": 7.2e-07,
|
||
|
|
"loss": 2.1268569946289064,
|
||
|
|
"mean_token_accuracy": 0.5856592752039432,
|
||
|
|
"num_tokens": 53385.0,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9018128983676433,
|
||
|
|
"epoch": 0.04217185028993147,
|
||
|
|
"grad_norm": 49.25,
|
||
|
|
"learning_rate": 1.52e-06,
|
||
|
|
"loss": 2.1235822677612304,
|
||
|
|
"mean_token_accuracy": 0.5826416034251451,
|
||
|
|
"num_tokens": 108949.0,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9313402399420738,
|
||
|
|
"epoch": 0.0632577754348972,
|
||
|
|
"grad_norm": 38.25,
|
||
|
|
"learning_rate": 2.3200000000000002e-06,
|
||
|
|
"loss": 1.9997358322143555,
|
||
|
|
"mean_token_accuracy": 0.5817345686256885,
|
||
|
|
"num_tokens": 162034.0,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9819381944835186,
|
||
|
|
"epoch": 0.08434370057986294,
|
||
|
|
"grad_norm": 28.25,
|
||
|
|
"learning_rate": 3.12e-06,
|
||
|
|
"loss": 1.8760208129882812,
|
||
|
|
"mean_token_accuracy": 0.5874959267675877,
|
||
|
|
"num_tokens": 214165.0,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0624087303876877,
|
||
|
|
"epoch": 0.10542962572482868,
|
||
|
|
"grad_norm": 14.1875,
|
||
|
|
"learning_rate": 3.920000000000001e-06,
|
||
|
|
"loss": 1.6169570922851562,
|
||
|
|
"mean_token_accuracy": 0.601640059798956,
|
||
|
|
"num_tokens": 265527.0,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1721510097384453,
|
||
|
|
"epoch": 0.1265155508697944,
|
||
|
|
"grad_norm": 11.3125,
|
||
|
|
"learning_rate": 4.7200000000000005e-06,
|
||
|
|
"loss": 1.4622878074645995,
|
||
|
|
"mean_token_accuracy": 0.6209174185991287,
|
||
|
|
"num_tokens": 319341.0,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.2009219758212566,
|
||
|
|
"epoch": 0.14760147601476015,
|
||
|
|
"grad_norm": 9.0625,
|
||
|
|
"learning_rate": 5.5200000000000005e-06,
|
||
|
|
"loss": 1.2933337211608886,
|
||
|
|
"mean_token_accuracy": 0.6467074818909169,
|
||
|
|
"num_tokens": 373709.0,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.1439887225627898,
|
||
|
|
"epoch": 0.16868740115972589,
|
||
|
|
"grad_norm": 9.3125,
|
||
|
|
"learning_rate": 6.3200000000000005e-06,
|
||
|
|
"loss": 1.2084498405456543,
|
||
|
|
"mean_token_accuracy": 0.6660686552524566,
|
||
|
|
"num_tokens": 425923.0,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0713621236383915,
|
||
|
|
"epoch": 0.18977332630469163,
|
||
|
|
"grad_norm": 7.84375,
|
||
|
|
"learning_rate": 7.1200000000000004e-06,
|
||
|
|
"loss": 1.1175435066223145,
|
||
|
|
"mean_token_accuracy": 0.6812180563807487,
|
||
|
|
"num_tokens": 480012.0,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 1.0369222231209279,
|
||
|
|
"epoch": 0.21085925144965736,
|
||
|
|
"grad_norm": 7.09375,
|
||
|
|
"learning_rate": 7.92e-06,
|
||
|
|
"loss": 1.062850570678711,
|
||
|
|
"mean_token_accuracy": 0.6899570092558861,
|
||
|
|
"num_tokens": 533890.0,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9894454136490822,
|
||
|
|
"epoch": 0.23194517659462308,
|
||
|
|
"grad_norm": 7.75,
|
||
|
|
"learning_rate": 8.720000000000001e-06,
|
||
|
|
"loss": 1.0150799751281738,
|
||
|
|
"mean_token_accuracy": 0.7009239844977856,
|
||
|
|
"num_tokens": 590654.0,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9410129584372043,
|
||
|
|
"epoch": 0.2530311017395888,
|
||
|
|
"grad_norm": 7.28125,
|
||
|
|
"learning_rate": 9.52e-06,
|
||
|
|
"loss": 0.9810515403747558,
|
||
|
|
"mean_token_accuracy": 0.709618617594242,
|
||
|
|
"num_tokens": 643080.0,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9265115469694137,
|
||
|
|
"epoch": 0.2741170268845546,
|
||
|
|
"grad_norm": 7.46875,
|
||
|
|
"learning_rate": 1.0320000000000001e-05,
|
||
|
|
"loss": 0.9485825538635254,
|
||
|
|
"mean_token_accuracy": 0.7162063598632813,
|
||
|
|
"num_tokens": 697254.0,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9009825974702835,
|
||
|
|
"epoch": 0.2952029520295203,
|
||
|
|
"grad_norm": 6.96875,
|
||
|
|
"learning_rate": 1.1120000000000002e-05,
|
||
|
|
"loss": 0.9119878768920898,
|
||
|
|
"mean_token_accuracy": 0.7238413639366627,
|
||
|
|
"num_tokens": 751774.0,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.9040703341364861,
|
||
|
|
"epoch": 0.316288877174486,
|
||
|
|
"grad_norm": 7.21875,
|
||
|
|
"learning_rate": 1.1920000000000001e-05,
|
||
|
|
"loss": 0.9283308029174805,
|
||
|
|
"mean_token_accuracy": 0.7166316144168376,
|
||
|
|
"num_tokens": 808927.0,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8878034949302673,
|
||
|
|
"epoch": 0.33737480231945177,
|
||
|
|
"grad_norm": 7.375,
|
||
|
|
"learning_rate": 1.2720000000000002e-05,
|
||
|
|
"loss": 0.9068665504455566,
|
||
|
|
"mean_token_accuracy": 0.7254431843757629,
|
||
|
|
"num_tokens": 858559.0,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.887480016052723,
|
||
|
|
"epoch": 0.3584607274644175,
|
||
|
|
"grad_norm": 7.65625,
|
||
|
|
"learning_rate": 1.3520000000000003e-05,
|
||
|
|
"loss": 0.8971159934997559,
|
||
|
|
"mean_token_accuracy": 0.7258390121161937,
|
||
|
|
"num_tokens": 912285.0,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8527634374797344,
|
||
|
|
"epoch": 0.37954665260938325,
|
||
|
|
"grad_norm": 7.6875,
|
||
|
|
"learning_rate": 1.432e-05,
|
||
|
|
"loss": 0.8586221694946289,
|
||
|
|
"mean_token_accuracy": 0.7359919100999832,
|
||
|
|
"num_tokens": 969385.0,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8445627830922604,
|
||
|
|
"epoch": 0.40063257775434896,
|
||
|
|
"grad_norm": 7.03125,
|
||
|
|
"learning_rate": 1.5120000000000001e-05,
|
||
|
|
"loss": 0.855714225769043,
|
||
|
|
"mean_token_accuracy": 0.7341729424893856,
|
||
|
|
"num_tokens": 1026106.0,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.841130618005991,
|
||
|
|
"epoch": 0.42171850289931473,
|
||
|
|
"grad_norm": 7.21875,
|
||
|
|
"learning_rate": 1.5920000000000003e-05,
|
||
|
|
"loss": 0.8620488166809082,
|
||
|
|
"mean_token_accuracy": 0.7361006513237953,
|
||
|
|
"num_tokens": 1079743.0,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.831258487701416,
|
||
|
|
"epoch": 0.44280442804428044,
|
||
|
|
"grad_norm": 7.0,
|
||
|
|
"learning_rate": 1.672e-05,
|
||
|
|
"loss": 0.8441590309143067,
|
||
|
|
"mean_token_accuracy": 0.7379129216074943,
|
||
|
|
"num_tokens": 1131453.0,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.79731200709939,
|
||
|
|
"epoch": 0.46389035318924615,
|
||
|
|
"grad_norm": 7.875,
|
||
|
|
"learning_rate": 1.752e-05,
|
||
|
|
"loss": 0.8135190963745117,
|
||
|
|
"mean_token_accuracy": 0.7497627973556519,
|
||
|
|
"num_tokens": 1182486.0,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8269097700715065,
|
||
|
|
"epoch": 0.4849762783342119,
|
||
|
|
"grad_norm": 7.5625,
|
||
|
|
"learning_rate": 1.832e-05,
|
||
|
|
"loss": 0.8399410247802734,
|
||
|
|
"mean_token_accuracy": 0.7389348305761814,
|
||
|
|
"num_tokens": 1238136.0,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8093138128519058,
|
||
|
|
"epoch": 0.5060622034791776,
|
||
|
|
"grad_norm": 7.40625,
|
||
|
|
"learning_rate": 1.912e-05,
|
||
|
|
"loss": 0.810025691986084,
|
||
|
|
"mean_token_accuracy": 0.7460841409862041,
|
||
|
|
"num_tokens": 1294939.0,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8069386176764966,
|
||
|
|
"epoch": 0.5271481286241434,
|
||
|
|
"grad_norm": 6.84375,
|
||
|
|
"learning_rate": 1.9920000000000002e-05,
|
||
|
|
"loss": 0.8215888023376465,
|
||
|
|
"mean_token_accuracy": 0.7438369721174241,
|
||
|
|
"num_tokens": 1346462.0,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.8088659413158894,
|
||
|
|
"epoch": 0.5482340537691092,
|
||
|
|
"grad_norm": 6.75,
|
||
|
|
"learning_rate": 1.9920000000000002e-05,
|
||
|
|
"loss": 0.8201449394226075,
|
||
|
|
"mean_token_accuracy": 0.7393839374184609,
|
||
|
|
"num_tokens": 1397444.0,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7969542622566224,
|
||
|
|
"epoch": 0.5693199789140748,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.9831111111111113e-05,
|
||
|
|
"loss": 0.8009759902954101,
|
||
|
|
"mean_token_accuracy": 0.7512488305568695,
|
||
|
|
"num_tokens": 1453900.0,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.780416651815176,
|
||
|
|
"epoch": 0.5904059040590406,
|
||
|
|
"grad_norm": 7.21875,
|
||
|
|
"learning_rate": 1.9742222222222223e-05,
|
||
|
|
"loss": 0.7874263763427735,
|
||
|
|
"mean_token_accuracy": 0.7526524804532528,
|
||
|
|
"num_tokens": 1505529.0,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7752278268337249,
|
||
|
|
"epoch": 0.6114918292040064,
|
||
|
|
"grad_norm": 6.875,
|
||
|
|
"learning_rate": 1.9653333333333334e-05,
|
||
|
|
"loss": 0.8075678825378418,
|
||
|
|
"mean_token_accuracy": 0.746569037437439,
|
||
|
|
"num_tokens": 1562473.0,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7895354829728604,
|
||
|
|
"epoch": 0.632577754348972,
|
||
|
|
"grad_norm": 6.6875,
|
||
|
|
"learning_rate": 1.9564444444444444e-05,
|
||
|
|
"loss": 0.7880948543548584,
|
||
|
|
"mean_token_accuracy": 0.7494045957922936,
|
||
|
|
"num_tokens": 1613320.0,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7653828673064709,
|
||
|
|
"epoch": 0.6536636794939378,
|
||
|
|
"grad_norm": 6.78125,
|
||
|
|
"learning_rate": 1.9475555555555558e-05,
|
||
|
|
"loss": 0.7780828475952148,
|
||
|
|
"mean_token_accuracy": 0.7556171745061875,
|
||
|
|
"num_tokens": 1662847.0,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7652887552976608,
|
||
|
|
"epoch": 0.6747496046389035,
|
||
|
|
"grad_norm": 6.375,
|
||
|
|
"learning_rate": 1.938666666666667e-05,
|
||
|
|
"loss": 0.7539787292480469,
|
||
|
|
"mean_token_accuracy": 0.7582654722034932,
|
||
|
|
"num_tokens": 1718553.0,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7076033115386963,
|
||
|
|
"epoch": 0.6958355297838693,
|
||
|
|
"grad_norm": 7.15625,
|
||
|
|
"learning_rate": 1.929777777777778e-05,
|
||
|
|
"loss": 0.7396605491638184,
|
||
|
|
"mean_token_accuracy": 0.7659220136702061,
|
||
|
|
"num_tokens": 1771024.0,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7509547933936119,
|
||
|
|
"epoch": 0.716921454928835,
|
||
|
|
"grad_norm": 6.15625,
|
||
|
|
"learning_rate": 1.920888888888889e-05,
|
||
|
|
"loss": 0.7459388256072998,
|
||
|
|
"mean_token_accuracy": 0.7644543245434761,
|
||
|
|
"num_tokens": 1821303.0,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7591827407479286,
|
||
|
|
"epoch": 0.7380073800738007,
|
||
|
|
"grad_norm": 7.03125,
|
||
|
|
"learning_rate": 1.912e-05,
|
||
|
|
"loss": 0.7627717971801757,
|
||
|
|
"mean_token_accuracy": 0.7567056350409984,
|
||
|
|
"num_tokens": 1868674.0,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7316406652331352,
|
||
|
|
"epoch": 0.7590933052187665,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.9031111111111114e-05,
|
||
|
|
"loss": 0.7425707340240478,
|
||
|
|
"mean_token_accuracy": 0.7653280735015869,
|
||
|
|
"num_tokens": 1924619.0,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7539067789912224,
|
||
|
|
"epoch": 0.7801792303637322,
|
||
|
|
"grad_norm": 7.0625,
|
||
|
|
"learning_rate": 1.8942222222222224e-05,
|
||
|
|
"loss": 0.7716987609863282,
|
||
|
|
"mean_token_accuracy": 0.7524559155106545,
|
||
|
|
"num_tokens": 1978795.0,
|
||
|
|
"step": 370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7687754213809967,
|
||
|
|
"epoch": 0.8012651555086979,
|
||
|
|
"grad_norm": 7.1875,
|
||
|
|
"learning_rate": 1.8853333333333335e-05,
|
||
|
|
"loss": 0.7734373569488525,
|
||
|
|
"mean_token_accuracy": 0.7497503250837326,
|
||
|
|
"num_tokens": 2029994.0,
|
||
|
|
"step": 380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7508607231080532,
|
||
|
|
"epoch": 0.8223510806536637,
|
||
|
|
"grad_norm": 7.09375,
|
||
|
|
"learning_rate": 1.8764444444444445e-05,
|
||
|
|
"loss": 0.7575721740722656,
|
||
|
|
"mean_token_accuracy": 0.7572125047445297,
|
||
|
|
"num_tokens": 2082653.0,
|
||
|
|
"step": 390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7252861328423024,
|
||
|
|
"epoch": 0.8434370057986295,
|
||
|
|
"grad_norm": 6.5625,
|
||
|
|
"learning_rate": 1.867555555555556e-05,
|
||
|
|
"loss": 0.7381186485290527,
|
||
|
|
"mean_token_accuracy": 0.7612978041172027,
|
||
|
|
"num_tokens": 2135276.0,
|
||
|
|
"step": 400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7339485622942448,
|
||
|
|
"epoch": 0.8645229309435951,
|
||
|
|
"grad_norm": 6.625,
|
||
|
|
"learning_rate": 1.858666666666667e-05,
|
||
|
|
"loss": 0.7348906993865967,
|
||
|
|
"mean_token_accuracy": 0.7663561321794987,
|
||
|
|
"num_tokens": 2187466.0,
|
||
|
|
"step": 410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7272624362260103,
|
||
|
|
"epoch": 0.8856088560885609,
|
||
|
|
"grad_norm": 6.90625,
|
||
|
|
"learning_rate": 1.849777777777778e-05,
|
||
|
|
"loss": 0.749025821685791,
|
||
|
|
"mean_token_accuracy": 0.7617379955947399,
|
||
|
|
"num_tokens": 2237434.0,
|
||
|
|
"step": 420
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7477380312979222,
|
||
|
|
"epoch": 0.9066947812335266,
|
||
|
|
"grad_norm": 6.34375,
|
||
|
|
"learning_rate": 1.840888888888889e-05,
|
||
|
|
"loss": 0.7498275756835937,
|
||
|
|
"mean_token_accuracy": 0.7604753144085408,
|
||
|
|
"num_tokens": 2288985.0,
|
||
|
|
"step": 430
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7346535548567772,
|
||
|
|
"epoch": 0.9277807063784923,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.832e-05,
|
||
|
|
"loss": 0.739972448348999,
|
||
|
|
"mean_token_accuracy": 0.7658773727715016,
|
||
|
|
"num_tokens": 2341187.0,
|
||
|
|
"step": 440
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7167556572705507,
|
||
|
|
"epoch": 0.9488666315234581,
|
||
|
|
"grad_norm": 6.4375,
|
||
|
|
"learning_rate": 1.823111111111111e-05,
|
||
|
|
"loss": 0.7350784778594971,
|
||
|
|
"mean_token_accuracy": 0.7644179269671441,
|
||
|
|
"num_tokens": 2393385.0,
|
||
|
|
"step": 450
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.7537519961595536,
|
||
|
|
"epoch": 0.9699525566684238,
|
||
|
|
"grad_norm": 7.28125,
|
||
|
|
"learning_rate": 1.814222222222222e-05,
|
||
|
|
"loss": 0.757652759552002,
|
||
|
|
"mean_token_accuracy": 0.7587068513035774,
|
||
|
|
"num_tokens": 2442248.0,
|
||
|
|
"step": 460
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.72038439437747,
|
||
|
|
"epoch": 0.9910384818133896,
|
||
|
|
"grad_norm": 7.40625,
|
||
|
|
"learning_rate": 1.8053333333333332e-05,
|
||
|
|
"loss": 0.7361824035644531,
|
||
|
|
"mean_token_accuracy": 0.7609050884842873,
|
||
|
|
"num_tokens": 2498394.0,
|
||
|
|
"step": 470
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.735570513718837,
|
||
|
|
"epoch": 1.0105429625724829,
|
||
|
|
"grad_norm": 6.875,
|
||
|
|
"learning_rate": 1.7964444444444446e-05,
|
||
|
|
"loss": 0.730868148803711,
|
||
|
|
"mean_token_accuracy": 0.7710521068121936,
|
||
|
|
"num_tokens": 2545422.0,
|
||
|
|
"step": 480
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6775223582983017,
|
||
|
|
"epoch": 1.0316288877174487,
|
||
|
|
"grad_norm": 7.125,
|
||
|
|
"learning_rate": 1.7875555555555556e-05,
|
||
|
|
"loss": 0.6740085124969483,
|
||
|
|
"mean_token_accuracy": 0.7795436836779117,
|
||
|
|
"num_tokens": 2597144.0,
|
||
|
|
"step": 490
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6628748029470444,
|
||
|
|
"epoch": 1.0527148128624144,
|
||
|
|
"grad_norm": 6.84375,
|
||
|
|
"learning_rate": 1.7786666666666667e-05,
|
||
|
|
"loss": 0.6774313926696778,
|
||
|
|
"mean_token_accuracy": 0.7789938881993294,
|
||
|
|
"num_tokens": 2651938.0,
|
||
|
|
"step": 500
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6695188779383898,
|
||
|
|
"epoch": 1.07380073800738,
|
||
|
|
"grad_norm": 7.0,
|
||
|
|
"learning_rate": 1.7697777777777777e-05,
|
||
|
|
"loss": 0.664512825012207,
|
||
|
|
"mean_token_accuracy": 0.7832319870591163,
|
||
|
|
"num_tokens": 2704948.0,
|
||
|
|
"step": 510
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6812492057681083,
|
||
|
|
"epoch": 1.0948866631523457,
|
||
|
|
"grad_norm": 7.0,
|
||
|
|
"learning_rate": 1.760888888888889e-05,
|
||
|
|
"loss": 0.7036191940307617,
|
||
|
|
"mean_token_accuracy": 0.7707728065550328,
|
||
|
|
"num_tokens": 2757576.0,
|
||
|
|
"step": 520
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6861097872257232,
|
||
|
|
"epoch": 1.1159725882973115,
|
||
|
|
"grad_norm": 7.3125,
|
||
|
|
"learning_rate": 1.752e-05,
|
||
|
|
"loss": 0.6651991844177246,
|
||
|
|
"mean_token_accuracy": 0.7841649115085602,
|
||
|
|
"num_tokens": 2808989.0,
|
||
|
|
"step": 530
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6694958411157131,
|
||
|
|
"epoch": 1.1370585134422773,
|
||
|
|
"grad_norm": 6.78125,
|
||
|
|
"learning_rate": 1.7431111111111112e-05,
|
||
|
|
"loss": 0.6833772659301758,
|
||
|
|
"mean_token_accuracy": 0.7775424145162105,
|
||
|
|
"num_tokens": 2864507.0,
|
||
|
|
"step": 540
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6568067740648985,
|
||
|
|
"epoch": 1.158144438587243,
|
||
|
|
"grad_norm": 6.21875,
|
||
|
|
"learning_rate": 1.7342222222222226e-05,
|
||
|
|
"loss": 0.6552294254302978,
|
||
|
|
"mean_token_accuracy": 0.7832994520664215,
|
||
|
|
"num_tokens": 2919201.0,
|
||
|
|
"step": 550
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6432577680796385,
|
||
|
|
"epoch": 1.1792303637322088,
|
||
|
|
"grad_norm": 6.65625,
|
||
|
|
"learning_rate": 1.7253333333333336e-05,
|
||
|
|
"loss": 0.6560451030731201,
|
||
|
|
"mean_token_accuracy": 0.7863748580217361,
|
||
|
|
"num_tokens": 2972434.0,
|
||
|
|
"step": 560
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6655397050082683,
|
||
|
|
"epoch": 1.2003162888771746,
|
||
|
|
"grad_norm": 6.09375,
|
||
|
|
"learning_rate": 1.7164444444444447e-05,
|
||
|
|
"loss": 0.6878513336181641,
|
||
|
|
"mean_token_accuracy": 0.7753379993140698,
|
||
|
|
"num_tokens": 3025099.0,
|
||
|
|
"step": 570
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.686093359068036,
|
||
|
|
"epoch": 1.2214022140221403,
|
||
|
|
"grad_norm": 7.09375,
|
||
|
|
"learning_rate": 1.7075555555555557e-05,
|
||
|
|
"loss": 0.6911579132080078,
|
||
|
|
"mean_token_accuracy": 0.7787310972809791,
|
||
|
|
"num_tokens": 3076307.0,
|
||
|
|
"step": 580
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6812980778515338,
|
||
|
|
"epoch": 1.2424881391671059,
|
||
|
|
"grad_norm": 7.125,
|
||
|
|
"learning_rate": 1.6986666666666668e-05,
|
||
|
|
"loss": 0.6704156398773193,
|
||
|
|
"mean_token_accuracy": 0.7793935135006904,
|
||
|
|
"num_tokens": 3130053.0,
|
||
|
|
"step": 590
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6409769468009472,
|
||
|
|
"epoch": 1.2635740643120716,
|
||
|
|
"grad_norm": 7.4375,
|
||
|
|
"learning_rate": 1.6897777777777778e-05,
|
||
|
|
"loss": 0.661864185333252,
|
||
|
|
"mean_token_accuracy": 0.7865670621395111,
|
||
|
|
"num_tokens": 3183331.0,
|
||
|
|
"step": 600
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6799454756081105,
|
||
|
|
"epoch": 1.2846599894570374,
|
||
|
|
"grad_norm": 6.65625,
|
||
|
|
"learning_rate": 1.680888888888889e-05,
|
||
|
|
"loss": 0.6818383693695068,
|
||
|
|
"mean_token_accuracy": 0.778287123888731,
|
||
|
|
"num_tokens": 3239561.0,
|
||
|
|
"step": 610
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6814555354416371,
|
||
|
|
"epoch": 1.3057459146020032,
|
||
|
|
"grad_norm": 6.3125,
|
||
|
|
"learning_rate": 1.672e-05,
|
||
|
|
"loss": 0.679498815536499,
|
||
|
|
"mean_token_accuracy": 0.7788635492324829,
|
||
|
|
"num_tokens": 3292996.0,
|
||
|
|
"step": 620
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.667108029127121,
|
||
|
|
"epoch": 1.326831839746969,
|
||
|
|
"grad_norm": 6.90625,
|
||
|
|
"learning_rate": 1.6631111111111113e-05,
|
||
|
|
"loss": 0.6758651733398438,
|
||
|
|
"mean_token_accuracy": 0.7792007848620415,
|
||
|
|
"num_tokens": 3345736.0,
|
||
|
|
"step": 630
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6692709110677242,
|
||
|
|
"epoch": 1.3479177648919347,
|
||
|
|
"grad_norm": 6.96875,
|
||
|
|
"learning_rate": 1.6542222222222223e-05,
|
||
|
|
"loss": 0.6796128273010253,
|
||
|
|
"mean_token_accuracy": 0.7766051791608334,
|
||
|
|
"num_tokens": 3398076.0,
|
||
|
|
"step": 640
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6564048878848553,
|
||
|
|
"epoch": 1.3690036900369003,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.6453333333333334e-05,
|
||
|
|
"loss": 0.6633264064788819,
|
||
|
|
"mean_token_accuracy": 0.7854946702718735,
|
||
|
|
"num_tokens": 3450796.0,
|
||
|
|
"step": 650
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6611979272216558,
|
||
|
|
"epoch": 1.390089615181866,
|
||
|
|
"grad_norm": 6.65625,
|
||
|
|
"learning_rate": 1.6364444444444444e-05,
|
||
|
|
"loss": 0.6661429405212402,
|
||
|
|
"mean_token_accuracy": 0.7835663430392742,
|
||
|
|
"num_tokens": 3503863.0,
|
||
|
|
"step": 660
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6534961331635714,
|
||
|
|
"epoch": 1.4111755403268318,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.6275555555555558e-05,
|
||
|
|
"loss": 0.6586530208587646,
|
||
|
|
"mean_token_accuracy": 0.7844067431986332,
|
||
|
|
"num_tokens": 3556265.0,
|
||
|
|
"step": 670
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6516873501241207,
|
||
|
|
"epoch": 1.4322614654717976,
|
||
|
|
"grad_norm": 6.40625,
|
||
|
|
"learning_rate": 1.618666666666667e-05,
|
||
|
|
"loss": 0.6522529125213623,
|
||
|
|
"mean_token_accuracy": 0.7855845786631107,
|
||
|
|
"num_tokens": 3611423.0,
|
||
|
|
"step": 680
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6764013767242432,
|
||
|
|
"epoch": 1.4533473906167633,
|
||
|
|
"grad_norm": 6.71875,
|
||
|
|
"learning_rate": 1.609777777777778e-05,
|
||
|
|
"loss": 0.6884442806243897,
|
||
|
|
"mean_token_accuracy": 0.7756352722644806,
|
||
|
|
"num_tokens": 3666630.0,
|
||
|
|
"step": 690
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6554486058652401,
|
||
|
|
"epoch": 1.474433315761729,
|
||
|
|
"grad_norm": 7.40625,
|
||
|
|
"learning_rate": 1.600888888888889e-05,
|
||
|
|
"loss": 0.6563316822052002,
|
||
|
|
"mean_token_accuracy": 0.7854672558605671,
|
||
|
|
"num_tokens": 3718719.0,
|
||
|
|
"step": 700
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6385185327380896,
|
||
|
|
"epoch": 1.4955192409066949,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.5920000000000003e-05,
|
||
|
|
"loss": 0.6513954639434815,
|
||
|
|
"mean_token_accuracy": 0.7851097173988819,
|
||
|
|
"num_tokens": 3771822.0,
|
||
|
|
"step": 710
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.654766371101141,
|
||
|
|
"epoch": 1.5166051660516606,
|
||
|
|
"grad_norm": 6.5,
|
||
|
|
"learning_rate": 1.5831111111111114e-05,
|
||
|
|
"loss": 0.6622748851776123,
|
||
|
|
"mean_token_accuracy": 0.7847663886845112,
|
||
|
|
"num_tokens": 3823795.0,
|
||
|
|
"step": 720
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6597310703247785,
|
||
|
|
"epoch": 1.5376910911966264,
|
||
|
|
"grad_norm": 6.375,
|
||
|
|
"learning_rate": 1.5742222222222224e-05,
|
||
|
|
"loss": 0.6634318351745605,
|
||
|
|
"mean_token_accuracy": 0.7861979939043522,
|
||
|
|
"num_tokens": 3875784.0,
|
||
|
|
"step": 730
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6637851595878601,
|
||
|
|
"epoch": 1.558777016341592,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.5653333333333335e-05,
|
||
|
|
"loss": 0.6621220111846924,
|
||
|
|
"mean_token_accuracy": 0.7886320613324642,
|
||
|
|
"num_tokens": 3929412.0,
|
||
|
|
"step": 740
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.637692479044199,
|
||
|
|
"epoch": 1.5798629414865577,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.5564444444444445e-05,
|
||
|
|
"loss": 0.6402386665344239,
|
||
|
|
"mean_token_accuracy": 0.7895873270928859,
|
||
|
|
"num_tokens": 3982472.0,
|
||
|
|
"step": 750
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6578179821372032,
|
||
|
|
"epoch": 1.6009488666315235,
|
||
|
|
"grad_norm": 7.25,
|
||
|
|
"learning_rate": 1.5475555555555556e-05,
|
||
|
|
"loss": 0.6751936435699463,
|
||
|
|
"mean_token_accuracy": 0.7776105716824532,
|
||
|
|
"num_tokens": 4037829.0,
|
||
|
|
"step": 760
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6653612300753593,
|
||
|
|
"epoch": 1.6220347917764892,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.5386666666666666e-05,
|
||
|
|
"loss": 0.6717413425445556,
|
||
|
|
"mean_token_accuracy": 0.777862798422575,
|
||
|
|
"num_tokens": 4093735.0,
|
||
|
|
"step": 770
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6561707239598036,
|
||
|
|
"epoch": 1.6431207169214548,
|
||
|
|
"grad_norm": 7.03125,
|
||
|
|
"learning_rate": 1.5297777777777777e-05,
|
||
|
|
"loss": 0.6501868724822998,
|
||
|
|
"mean_token_accuracy": 0.7863908343017101,
|
||
|
|
"num_tokens": 4148410.0,
|
||
|
|
"step": 780
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6655665650963783,
|
||
|
|
"epoch": 1.6642066420664205,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.520888888888889e-05,
|
||
|
|
"loss": 0.679684829711914,
|
||
|
|
"mean_token_accuracy": 0.778567586839199,
|
||
|
|
"num_tokens": 4202907.0,
|
||
|
|
"step": 790
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6781473312526941,
|
||
|
|
"epoch": 1.6852925672113863,
|
||
|
|
"grad_norm": 6.875,
|
||
|
|
"learning_rate": 1.5120000000000001e-05,
|
||
|
|
"loss": 0.6722404479980468,
|
||
|
|
"mean_token_accuracy": 0.7763243250548839,
|
||
|
|
"num_tokens": 4256213.0,
|
||
|
|
"step": 800
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6321199681609869,
|
||
|
|
"epoch": 1.706378492356352,
|
||
|
|
"grad_norm": 6.78125,
|
||
|
|
"learning_rate": 1.5031111111111111e-05,
|
||
|
|
"loss": 0.6404708385467529,
|
||
|
|
"mean_token_accuracy": 0.7903632469475269,
|
||
|
|
"num_tokens": 4309268.0,
|
||
|
|
"step": 810
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6424700867384672,
|
||
|
|
"epoch": 1.7274644175013179,
|
||
|
|
"grad_norm": 7.15625,
|
||
|
|
"learning_rate": 1.4942222222222222e-05,
|
||
|
|
"loss": 0.6444578170776367,
|
||
|
|
"mean_token_accuracy": 0.7881530813872815,
|
||
|
|
"num_tokens": 4360761.0,
|
||
|
|
"step": 820
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6615507304668427,
|
||
|
|
"epoch": 1.7485503426462836,
|
||
|
|
"grad_norm": 6.0625,
|
||
|
|
"learning_rate": 1.4853333333333336e-05,
|
||
|
|
"loss": 0.6604718208312989,
|
||
|
|
"mean_token_accuracy": 0.782805722206831,
|
||
|
|
"num_tokens": 4411678.0,
|
||
|
|
"step": 830
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6196760050952435,
|
||
|
|
"epoch": 1.7696362677912494,
|
||
|
|
"grad_norm": 6.6875,
|
||
|
|
"learning_rate": 1.4764444444444446e-05,
|
||
|
|
"loss": 0.6261836051940918,
|
||
|
|
"mean_token_accuracy": 0.7926589228212834,
|
||
|
|
"num_tokens": 4467254.0,
|
||
|
|
"step": 840
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6139154363423586,
|
||
|
|
"epoch": 1.7907221929362152,
|
||
|
|
"grad_norm": 7.21875,
|
||
|
|
"learning_rate": 1.4675555555555557e-05,
|
||
|
|
"loss": 0.6301932811737061,
|
||
|
|
"mean_token_accuracy": 0.7918383464217186,
|
||
|
|
"num_tokens": 4517383.0,
|
||
|
|
"step": 850
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6410171519964933,
|
||
|
|
"epoch": 1.811808118081181,
|
||
|
|
"grad_norm": 7.1875,
|
||
|
|
"learning_rate": 1.4586666666666667e-05,
|
||
|
|
"loss": 0.655810022354126,
|
||
|
|
"mean_token_accuracy": 0.7856725618243218,
|
||
|
|
"num_tokens": 4569980.0,
|
||
|
|
"step": 860
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6666022066026926,
|
||
|
|
"epoch": 1.8328940432261467,
|
||
|
|
"grad_norm": 6.5625,
|
||
|
|
"learning_rate": 1.449777777777778e-05,
|
||
|
|
"loss": 0.6677482604980469,
|
||
|
|
"mean_token_accuracy": 0.7798060193657875,
|
||
|
|
"num_tokens": 4622612.0,
|
||
|
|
"step": 870
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6398684818297624,
|
||
|
|
"epoch": 1.8539799683711122,
|
||
|
|
"grad_norm": 7.84375,
|
||
|
|
"learning_rate": 1.440888888888889e-05,
|
||
|
|
"loss": 0.6501501083374024,
|
||
|
|
"mean_token_accuracy": 0.7857734531164169,
|
||
|
|
"num_tokens": 4676525.0,
|
||
|
|
"step": 880
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6285852700471878,
|
||
|
|
"epoch": 1.875065893516078,
|
||
|
|
"grad_norm": 7.0625,
|
||
|
|
"learning_rate": 1.432e-05,
|
||
|
|
"loss": 0.6422877311706543,
|
||
|
|
"mean_token_accuracy": 0.7884316690266132,
|
||
|
|
"num_tokens": 4734090.0,
|
||
|
|
"step": 890
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6409385818988085,
|
||
|
|
"epoch": 1.8961518186610438,
|
||
|
|
"grad_norm": 7.9375,
|
||
|
|
"learning_rate": 1.4231111111111114e-05,
|
||
|
|
"loss": 0.6352181434631348,
|
||
|
|
"mean_token_accuracy": 0.7952412232756615,
|
||
|
|
"num_tokens": 4788537.0,
|
||
|
|
"step": 900
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6456746399402619,
|
||
|
|
"epoch": 1.9172377438060095,
|
||
|
|
"grad_norm": 6.90625,
|
||
|
|
"learning_rate": 1.4142222222222224e-05,
|
||
|
|
"loss": 0.6564863204956055,
|
||
|
|
"mean_token_accuracy": 0.789187741279602,
|
||
|
|
"num_tokens": 4840342.0,
|
||
|
|
"step": 910
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6360713161528111,
|
||
|
|
"epoch": 1.938323668950975,
|
||
|
|
"grad_norm": 6.3125,
|
||
|
|
"learning_rate": 1.4053333333333335e-05,
|
||
|
|
"loss": 0.6300373554229737,
|
||
|
|
"mean_token_accuracy": 0.7876299776136875,
|
||
|
|
"num_tokens": 4891524.0,
|
||
|
|
"step": 920
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6285211775451899,
|
||
|
|
"epoch": 1.9594095940959408,
|
||
|
|
"grad_norm": 7.78125,
|
||
|
|
"learning_rate": 1.3964444444444445e-05,
|
||
|
|
"loss": 0.6337003231048584,
|
||
|
|
"mean_token_accuracy": 0.7889726974070073,
|
||
|
|
"num_tokens": 4939573.0,
|
||
|
|
"step": 930
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6301535632461309,
|
||
|
|
"epoch": 1.9804955192409066,
|
||
|
|
"grad_norm": 6.84375,
|
||
|
|
"learning_rate": 1.3875555555555557e-05,
|
||
|
|
"loss": 0.6374162673950196,
|
||
|
|
"mean_token_accuracy": 0.7898242607712745,
|
||
|
|
"num_tokens": 4991267.0,
|
||
|
|
"step": 940
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6537058566873138,
|
||
|
|
"epoch": 2.0,
|
||
|
|
"grad_norm": 13.3125,
|
||
|
|
"learning_rate": 1.3786666666666668e-05,
|
||
|
|
"loss": 0.641156530380249,
|
||
|
|
"mean_token_accuracy": 0.7822844466647586,
|
||
|
|
"num_tokens": 5036304.0,
|
||
|
|
"step": 950
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6411334682255984,
|
||
|
|
"epoch": 2.0210859251449658,
|
||
|
|
"grad_norm": 7.1875,
|
||
|
|
"learning_rate": 1.3697777777777778e-05,
|
||
|
|
"loss": 0.6083747386932373,
|
||
|
|
"mean_token_accuracy": 0.8002627611160278,
|
||
|
|
"num_tokens": 5089737.0,
|
||
|
|
"step": 960
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5841103594750165,
|
||
|
|
"epoch": 2.0421718502899315,
|
||
|
|
"grad_norm": 6.53125,
|
||
|
|
"learning_rate": 1.3608888888888889e-05,
|
||
|
|
"loss": 0.6012133121490478,
|
||
|
|
"mean_token_accuracy": 0.8027123242616654,
|
||
|
|
"num_tokens": 5143330.0,
|
||
|
|
"step": 970
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.590392829477787,
|
||
|
|
"epoch": 2.0632577754348973,
|
||
|
|
"grad_norm": 6.46875,
|
||
|
|
"learning_rate": 1.3520000000000003e-05,
|
||
|
|
"loss": 0.5829256534576416,
|
||
|
|
"mean_token_accuracy": 0.8068650290369987,
|
||
|
|
"num_tokens": 5198614.0,
|
||
|
|
"step": 980
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6148912344127894,
|
||
|
|
"epoch": 2.084343700579863,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.3431111111111113e-05,
|
||
|
|
"loss": 0.6117685317993165,
|
||
|
|
"mean_token_accuracy": 0.7951427385210991,
|
||
|
|
"num_tokens": 5252755.0,
|
||
|
|
"step": 990
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5893940377980471,
|
||
|
|
"epoch": 2.105429625724829,
|
||
|
|
"grad_norm": 6.625,
|
||
|
|
"learning_rate": 1.3342222222222224e-05,
|
||
|
|
"loss": 0.5790496349334717,
|
||
|
|
"mean_token_accuracy": 0.809234394133091,
|
||
|
|
"num_tokens": 5304859.0,
|
||
|
|
"step": 1000
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.587013079598546,
|
||
|
|
"epoch": 2.1265155508697946,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.3253333333333334e-05,
|
||
|
|
"loss": 0.5892822265625,
|
||
|
|
"mean_token_accuracy": 0.8040889374911785,
|
||
|
|
"num_tokens": 5357877.0,
|
||
|
|
"step": 1010
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5877181258052587,
|
||
|
|
"epoch": 2.14760147601476,
|
||
|
|
"grad_norm": 6.09375,
|
||
|
|
"learning_rate": 1.3164444444444446e-05,
|
||
|
|
"loss": 0.595665693283081,
|
||
|
|
"mean_token_accuracy": 0.8036225430667401,
|
||
|
|
"num_tokens": 5413504.0,
|
||
|
|
"step": 1020
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5848948258906603,
|
||
|
|
"epoch": 2.1686874011597257,
|
||
|
|
"grad_norm": 6.4375,
|
||
|
|
"learning_rate": 1.3075555555555557e-05,
|
||
|
|
"loss": 0.5771171092987061,
|
||
|
|
"mean_token_accuracy": 0.8047491163015366,
|
||
|
|
"num_tokens": 5466549.0,
|
||
|
|
"step": 1030
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5815035384148359,
|
||
|
|
"epoch": 2.1897733263046915,
|
||
|
|
"grad_norm": 6.96875,
|
||
|
|
"learning_rate": 1.2986666666666667e-05,
|
||
|
|
"loss": 0.5768617153167724,
|
||
|
|
"mean_token_accuracy": 0.8057669088244438,
|
||
|
|
"num_tokens": 5519515.0,
|
||
|
|
"step": 1040
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.58335498906672,
|
||
|
|
"epoch": 2.2108592514496572,
|
||
|
|
"grad_norm": 7.6875,
|
||
|
|
"learning_rate": 1.2897777777777778e-05,
|
||
|
|
"loss": 0.6007503509521485,
|
||
|
|
"mean_token_accuracy": 0.8005498602986336,
|
||
|
|
"num_tokens": 5570188.0,
|
||
|
|
"step": 1050
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6068603809922933,
|
||
|
|
"epoch": 2.231945176594623,
|
||
|
|
"grad_norm": 6.5625,
|
||
|
|
"learning_rate": 1.2808888888888891e-05,
|
||
|
|
"loss": 0.6024115562438965,
|
||
|
|
"mean_token_accuracy": 0.7973541341722011,
|
||
|
|
"num_tokens": 5624190.0,
|
||
|
|
"step": 1060
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5774707745760679,
|
||
|
|
"epoch": 2.2530311017395888,
|
||
|
|
"grad_norm": 6.53125,
|
||
|
|
"learning_rate": 1.2720000000000002e-05,
|
||
|
|
"loss": 0.5705479145050049,
|
||
|
|
"mean_token_accuracy": 0.8100002221763134,
|
||
|
|
"num_tokens": 5675656.0,
|
||
|
|
"step": 1070
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5851722996681928,
|
||
|
|
"epoch": 2.2741170268845545,
|
||
|
|
"grad_norm": 7.3125,
|
||
|
|
"learning_rate": 1.2631111111111112e-05,
|
||
|
|
"loss": 0.5905722141265869,
|
||
|
|
"mean_token_accuracy": 0.8069308035075664,
|
||
|
|
"num_tokens": 5728878.0,
|
||
|
|
"step": 1080
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.594813584536314,
|
||
|
|
"epoch": 2.2952029520295203,
|
||
|
|
"grad_norm": 6.875,
|
||
|
|
"learning_rate": 1.2542222222222223e-05,
|
||
|
|
"loss": 0.5923901557922363,
|
||
|
|
"mean_token_accuracy": 0.8034852437675,
|
||
|
|
"num_tokens": 5783257.0,
|
||
|
|
"step": 1090
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5841789487749338,
|
||
|
|
"epoch": 2.316288877174486,
|
||
|
|
"grad_norm": 7.0,
|
||
|
|
"learning_rate": 1.2453333333333335e-05,
|
||
|
|
"loss": 0.5864346504211426,
|
||
|
|
"mean_token_accuracy": 0.8033530995249748,
|
||
|
|
"num_tokens": 5834581.0,
|
||
|
|
"step": 1100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6010521166026592,
|
||
|
|
"epoch": 2.337374802319452,
|
||
|
|
"grad_norm": 6.9375,
|
||
|
|
"learning_rate": 1.2364444444444445e-05,
|
||
|
|
"loss": 0.5941137313842774,
|
||
|
|
"mean_token_accuracy": 0.8033333070576191,
|
||
|
|
"num_tokens": 5887458.0,
|
||
|
|
"step": 1110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5919341973960399,
|
||
|
|
"epoch": 2.3584607274644176,
|
||
|
|
"grad_norm": 8.75,
|
||
|
|
"learning_rate": 1.2275555555555556e-05,
|
||
|
|
"loss": 0.6103182315826416,
|
||
|
|
"mean_token_accuracy": 0.7999237082898617,
|
||
|
|
"num_tokens": 5941284.0,
|
||
|
|
"step": 1120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5942621320486069,
|
||
|
|
"epoch": 2.3795466526093834,
|
||
|
|
"grad_norm": 7.375,
|
||
|
|
"learning_rate": 1.2186666666666666e-05,
|
||
|
|
"loss": 0.5983363151550293,
|
||
|
|
"mean_token_accuracy": 0.8016222022473812,
|
||
|
|
"num_tokens": 5995817.0,
|
||
|
|
"step": 1130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6046227600425482,
|
||
|
|
"epoch": 2.400632577754349,
|
||
|
|
"grad_norm": 6.40625,
|
||
|
|
"learning_rate": 1.209777777777778e-05,
|
||
|
|
"loss": 0.6026669025421143,
|
||
|
|
"mean_token_accuracy": 0.8024684950709343,
|
||
|
|
"num_tokens": 6050783.0,
|
||
|
|
"step": 1140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5998306684195995,
|
||
|
|
"epoch": 2.421718502899315,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.200888888888889e-05,
|
||
|
|
"loss": 0.5893473625183105,
|
||
|
|
"mean_token_accuracy": 0.803314708918333,
|
||
|
|
"num_tokens": 6103371.0,
|
||
|
|
"step": 1150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.593693134188652,
|
||
|
|
"epoch": 2.4428044280442807,
|
||
|
|
"grad_norm": 6.5,
|
||
|
|
"learning_rate": 1.1920000000000001e-05,
|
||
|
|
"loss": 0.6076583862304688,
|
||
|
|
"mean_token_accuracy": 0.7995239071547985,
|
||
|
|
"num_tokens": 6155502.0,
|
||
|
|
"step": 1160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5827030725777149,
|
||
|
|
"epoch": 2.463890353189246,
|
||
|
|
"grad_norm": 6.84375,
|
||
|
|
"learning_rate": 1.1831111111111112e-05,
|
||
|
|
"loss": 0.5882821559906006,
|
||
|
|
"mean_token_accuracy": 0.7929977595806121,
|
||
|
|
"num_tokens": 6208060.0,
|
||
|
|
"step": 1170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5910260166972876,
|
||
|
|
"epoch": 2.4849762783342118,
|
||
|
|
"grad_norm": 7.53125,
|
||
|
|
"learning_rate": 1.1742222222222224e-05,
|
||
|
|
"loss": 0.5860947132110595,
|
||
|
|
"mean_token_accuracy": 0.8049415498971939,
|
||
|
|
"num_tokens": 6261588.0,
|
||
|
|
"step": 1180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5704694345593453,
|
||
|
|
"epoch": 2.5060622034791775,
|
||
|
|
"grad_norm": 6.75,
|
||
|
|
"learning_rate": 1.1653333333333334e-05,
|
||
|
|
"loss": 0.5749064445495605,
|
||
|
|
"mean_token_accuracy": 0.8084491945803165,
|
||
|
|
"num_tokens": 6314963.0,
|
||
|
|
"step": 1190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.597440430521965,
|
||
|
|
"epoch": 2.5271481286241433,
|
||
|
|
"grad_norm": 6.5,
|
||
|
|
"learning_rate": 1.1564444444444445e-05,
|
||
|
|
"loss": 0.5958696842193604,
|
||
|
|
"mean_token_accuracy": 0.8057812102138996,
|
||
|
|
"num_tokens": 6368741.0,
|
||
|
|
"step": 1200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6070981964468956,
|
||
|
|
"epoch": 2.548234053769109,
|
||
|
|
"grad_norm": 7.96875,
|
||
|
|
"learning_rate": 1.1475555555555557e-05,
|
||
|
|
"loss": 0.6143896102905273,
|
||
|
|
"mean_token_accuracy": 0.7980599500238895,
|
||
|
|
"num_tokens": 6425424.0,
|
||
|
|
"step": 1210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5948727998882533,
|
||
|
|
"epoch": 2.569319978914075,
|
||
|
|
"grad_norm": 6.5,
|
||
|
|
"learning_rate": 1.1386666666666669e-05,
|
||
|
|
"loss": 0.5917119026184082,
|
||
|
|
"mean_token_accuracy": 0.8025051333010197,
|
||
|
|
"num_tokens": 6476263.0,
|
||
|
|
"step": 1220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5811738461256027,
|
||
|
|
"epoch": 2.5904059040590406,
|
||
|
|
"grad_norm": 7.21875,
|
||
|
|
"learning_rate": 1.129777777777778e-05,
|
||
|
|
"loss": 0.5865864753723145,
|
||
|
|
"mean_token_accuracy": 0.8044181451201439,
|
||
|
|
"num_tokens": 6528012.0,
|
||
|
|
"step": 1230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5937653455883265,
|
||
|
|
"epoch": 2.6114918292040064,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.120888888888889e-05,
|
||
|
|
"loss": 0.5843213558197021,
|
||
|
|
"mean_token_accuracy": 0.8025668740272522,
|
||
|
|
"num_tokens": 6583421.0,
|
||
|
|
"step": 1240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.573565949127078,
|
||
|
|
"epoch": 2.632577754348972,
|
||
|
|
"grad_norm": 6.5625,
|
||
|
|
"learning_rate": 1.1120000000000002e-05,
|
||
|
|
"loss": 0.5826557159423829,
|
||
|
|
"mean_token_accuracy": 0.8070438347756863,
|
||
|
|
"num_tokens": 6636858.0,
|
||
|
|
"step": 1250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6216809719800949,
|
||
|
|
"epoch": 2.653663679493938,
|
||
|
|
"grad_norm": 7.09375,
|
||
|
|
"learning_rate": 1.1031111111111112e-05,
|
||
|
|
"loss": 0.6125890254974365,
|
||
|
|
"mean_token_accuracy": 0.7970895677804947,
|
||
|
|
"num_tokens": 6692168.0,
|
||
|
|
"step": 1260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5998566258698702,
|
||
|
|
"epoch": 2.6747496046389037,
|
||
|
|
"grad_norm": 7.0625,
|
||
|
|
"learning_rate": 1.0942222222222223e-05,
|
||
|
|
"loss": 0.5921674728393554,
|
||
|
|
"mean_token_accuracy": 0.8061520479619503,
|
||
|
|
"num_tokens": 6745834.0,
|
||
|
|
"step": 1270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5880189586430788,
|
||
|
|
"epoch": 2.6958355297838694,
|
||
|
|
"grad_norm": 7.59375,
|
||
|
|
"learning_rate": 1.0853333333333333e-05,
|
||
|
|
"loss": 0.5986497402191162,
|
||
|
|
"mean_token_accuracy": 0.8006219923496246,
|
||
|
|
"num_tokens": 6800244.0,
|
||
|
|
"step": 1280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5981763809919357,
|
||
|
|
"epoch": 2.716921454928835,
|
||
|
|
"grad_norm": 6.6875,
|
||
|
|
"learning_rate": 1.0764444444444445e-05,
|
||
|
|
"loss": 0.6020579814910889,
|
||
|
|
"mean_token_accuracy": 0.8032843962311744,
|
||
|
|
"num_tokens": 6853454.0,
|
||
|
|
"step": 1290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5963539965450764,
|
||
|
|
"epoch": 2.7380073800738005,
|
||
|
|
"grad_norm": 6.59375,
|
||
|
|
"learning_rate": 1.0675555555555558e-05,
|
||
|
|
"loss": 0.5908561706542969,
|
||
|
|
"mean_token_accuracy": 0.8011696465313435,
|
||
|
|
"num_tokens": 6904473.0,
|
||
|
|
"step": 1300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5899917531758547,
|
||
|
|
"epoch": 2.7590933052187667,
|
||
|
|
"grad_norm": 6.6875,
|
||
|
|
"learning_rate": 1.0586666666666668e-05,
|
||
|
|
"loss": 0.5744058132171631,
|
||
|
|
"mean_token_accuracy": 0.8053683824837208,
|
||
|
|
"num_tokens": 6956756.0,
|
||
|
|
"step": 1310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.582448274642229,
|
||
|
|
"epoch": 2.780179230363732,
|
||
|
|
"grad_norm": 6.78125,
|
||
|
|
"learning_rate": 1.0497777777777779e-05,
|
||
|
|
"loss": 0.5848373889923095,
|
||
|
|
"mean_token_accuracy": 0.8054254874587059,
|
||
|
|
"num_tokens": 7008611.0,
|
||
|
|
"step": 1320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5906594522297383,
|
||
|
|
"epoch": 2.801265155508698,
|
||
|
|
"grad_norm": 7.75,
|
||
|
|
"learning_rate": 1.040888888888889e-05,
|
||
|
|
"loss": 0.6006948471069335,
|
||
|
|
"mean_token_accuracy": 0.7971956506371498,
|
||
|
|
"num_tokens": 7062235.0,
|
||
|
|
"step": 1330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5976275954395532,
|
||
|
|
"epoch": 2.8223510806536636,
|
||
|
|
"grad_norm": 6.34375,
|
||
|
|
"learning_rate": 1.0320000000000001e-05,
|
||
|
|
"loss": 0.5997600078582763,
|
||
|
|
"mean_token_accuracy": 0.7998531877994537,
|
||
|
|
"num_tokens": 7115737.0,
|
||
|
|
"step": 1340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6079881630837918,
|
||
|
|
"epoch": 2.8434370057986293,
|
||
|
|
"grad_norm": 7.0,
|
||
|
|
"learning_rate": 1.0231111111111112e-05,
|
||
|
|
"loss": 0.6024625778198243,
|
||
|
|
"mean_token_accuracy": 0.7988371469080449,
|
||
|
|
"num_tokens": 7166870.0,
|
||
|
|
"step": 1350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.6036868289113044,
|
||
|
|
"epoch": 2.864522930943595,
|
||
|
|
"grad_norm": 6.6875,
|
||
|
|
"learning_rate": 1.0142222222222222e-05,
|
||
|
|
"loss": 0.5962271213531494,
|
||
|
|
"mean_token_accuracy": 0.8011133573949337,
|
||
|
|
"num_tokens": 7219785.0,
|
||
|
|
"step": 1360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5746289525181055,
|
||
|
|
"epoch": 2.885608856088561,
|
||
|
|
"grad_norm": 7.5625,
|
||
|
|
"learning_rate": 1.0053333333333334e-05,
|
||
|
|
"loss": 0.5769401073455811,
|
||
|
|
"mean_token_accuracy": 0.809252617508173,
|
||
|
|
"num_tokens": 7275411.0,
|
||
|
|
"step": 1370
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5846736520528794,
|
||
|
|
"epoch": 2.9066947812335266,
|
||
|
|
"grad_norm": 7.125,
|
||
|
|
"learning_rate": 9.964444444444445e-06,
|
||
|
|
"loss": 0.5899319171905517,
|
||
|
|
"mean_token_accuracy": 0.8028473742306232,
|
||
|
|
"num_tokens": 7326752.0,
|
||
|
|
"step": 1380
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5813163112848997,
|
||
|
|
"epoch": 2.9277807063784924,
|
||
|
|
"grad_norm": 7.46875,
|
||
|
|
"learning_rate": 9.875555555555557e-06,
|
||
|
|
"loss": 0.5799898147583008,
|
||
|
|
"mean_token_accuracy": 0.8026903085410595,
|
||
|
|
"num_tokens": 7375608.0,
|
||
|
|
"step": 1390
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5724870227277279,
|
||
|
|
"epoch": 2.948866631523458,
|
||
|
|
"grad_norm": 6.90625,
|
||
|
|
"learning_rate": 9.786666666666667e-06,
|
||
|
|
"loss": 0.5783823490142822,
|
||
|
|
"mean_token_accuracy": 0.8059644535183906,
|
||
|
|
"num_tokens": 7430145.0,
|
||
|
|
"step": 1400
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5778653729707003,
|
||
|
|
"epoch": 2.969952556668424,
|
||
|
|
"grad_norm": 6.84375,
|
||
|
|
"learning_rate": 9.697777777777778e-06,
|
||
|
|
"loss": 0.5640594005584717,
|
||
|
|
"mean_token_accuracy": 0.8100298568606377,
|
||
|
|
"num_tokens": 7482976.0,
|
||
|
|
"step": 1410
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"entropy": 0.5823766525834799,
|
||
|
|
"epoch": 2.9910384818133897,
|
||
|
|
"grad_norm": 6.5625,
|
||
|
|
"learning_rate": 9.60888888888889e-06,
|
||
|
|
"loss": 0.5931865692138671,
|
||
|
|
"mean_token_accuracy": 0.8023995622992516,
|
||
|
|
"num_tokens": 7533043.0,
|
||
|
|
"step": 1420
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 10,
|
||
|
|
"max_steps": 2500,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 6,
|
||
|
|
"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": 7.73147357152727e+16,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|