22277 lines
609 KiB
JSON
22277 lines
609 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 9.205020920502092,
|
|
"eval_steps": 500,
|
|
"global_step": 11000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.692013263702393,
|
|
"epoch": 0.0041841004184100415,
|
|
"grad_norm": 14.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8216,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9848.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69194860458374,
|
|
"epoch": 0.008368200836820083,
|
|
"grad_norm": 13.4375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7246,
|
|
"mean_token_accuracy": 0.00012674271129071714,
|
|
"num_tokens": 18196.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.691147422790527,
|
|
"epoch": 0.012552301255230125,
|
|
"grad_norm": 9.9375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4677,
|
|
"mean_token_accuracy": 0.0241286511416547,
|
|
"num_tokens": 26927.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.680734634399414,
|
|
"epoch": 0.016736401673640166,
|
|
"grad_norm": 6.6875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.115,
|
|
"mean_token_accuracy": 0.04429602064192295,
|
|
"num_tokens": 34987.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.62258882522583,
|
|
"epoch": 0.02092050209205021,
|
|
"grad_norm": 4.25,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.8294,
|
|
"mean_token_accuracy": 0.04711197260767221,
|
|
"num_tokens": 44936.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.580855178833009,
|
|
"epoch": 0.02510460251046025,
|
|
"grad_norm": 3.515625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.6731,
|
|
"mean_token_accuracy": 0.046292875334620474,
|
|
"num_tokens": 54562.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.583562088012695,
|
|
"epoch": 0.029288702928870293,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.6171,
|
|
"mean_token_accuracy": 0.04133799690753222,
|
|
"num_tokens": 63844.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.599079990386963,
|
|
"epoch": 0.03347280334728033,
|
|
"grad_norm": 2.8125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6032,
|
|
"mean_token_accuracy": 0.040498660318553446,
|
|
"num_tokens": 74331.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.584051704406738,
|
|
"epoch": 0.03765690376569038,
|
|
"grad_norm": 2.796875,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5052,
|
|
"mean_token_accuracy": 0.04810179658234119,
|
|
"num_tokens": 83028.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.553222179412842,
|
|
"epoch": 0.04184100418410042,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.4503,
|
|
"mean_token_accuracy": 0.04860256612300873,
|
|
"num_tokens": 91395.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.53250617980957,
|
|
"epoch": 0.04602510460251046,
|
|
"grad_norm": 2.859375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.3929,
|
|
"mean_token_accuracy": 0.05782372578978538,
|
|
"num_tokens": 101868.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.429417991638184,
|
|
"epoch": 0.0502092050209205,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.2812,
|
|
"mean_token_accuracy": 0.05993606857955456,
|
|
"num_tokens": 110403.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.376260757446289,
|
|
"epoch": 0.05439330543933055,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.3299,
|
|
"mean_token_accuracy": 0.05404467172920704,
|
|
"num_tokens": 119557.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.482521724700927,
|
|
"epoch": 0.058577405857740586,
|
|
"grad_norm": 2.515625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.1462,
|
|
"mean_token_accuracy": 0.05712815783917904,
|
|
"num_tokens": 128903.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.432404041290283,
|
|
"epoch": 0.06276150627615062,
|
|
"grad_norm": 2.546875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.1137,
|
|
"mean_token_accuracy": 0.05686149187386036,
|
|
"num_tokens": 137630.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.388175106048584,
|
|
"epoch": 0.06694560669456066,
|
|
"grad_norm": 2.890625,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 8.8947,
|
|
"mean_token_accuracy": 0.06613010838627816,
|
|
"num_tokens": 146335.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.388596439361573,
|
|
"epoch": 0.07112970711297072,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 8.8136,
|
|
"mean_token_accuracy": 0.06454679258167743,
|
|
"num_tokens": 155766.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.335438442230224,
|
|
"epoch": 0.07531380753138076,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.7316,
|
|
"mean_token_accuracy": 0.06548036001622677,
|
|
"num_tokens": 165844.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.317347526550293,
|
|
"epoch": 0.0794979079497908,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.6896,
|
|
"mean_token_accuracy": 0.06795907951891422,
|
|
"num_tokens": 174858.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.24217710494995,
|
|
"epoch": 0.08368200836820083,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.5453,
|
|
"mean_token_accuracy": 0.07007661424577236,
|
|
"num_tokens": 184020.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.19391098022461,
|
|
"epoch": 0.08786610878661087,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.4004,
|
|
"mean_token_accuracy": 0.06834917515516281,
|
|
"num_tokens": 193714.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.143691921234131,
|
|
"epoch": 0.09205020920502092,
|
|
"grad_norm": 2.265625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.3625,
|
|
"mean_token_accuracy": 0.07014007195830345,
|
|
"num_tokens": 203139.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.033193874359132,
|
|
"epoch": 0.09623430962343096,
|
|
"grad_norm": 2.75,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.2108,
|
|
"mean_token_accuracy": 0.0716083101928234,
|
|
"num_tokens": 212446.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.93264389038086,
|
|
"epoch": 0.100418410041841,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.0305,
|
|
"mean_token_accuracy": 0.08365688249468803,
|
|
"num_tokens": 220910.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.831494235992432,
|
|
"epoch": 0.10460251046025104,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 7.975,
|
|
"mean_token_accuracy": 0.07431616149842739,
|
|
"num_tokens": 230522.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.695558643341064,
|
|
"epoch": 0.1087866108786611,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.8131,
|
|
"mean_token_accuracy": 0.07820754498243332,
|
|
"num_tokens": 240092.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.519502639770508,
|
|
"epoch": 0.11297071129707113,
|
|
"grad_norm": 1.859375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.07841893658041954,
|
|
"num_tokens": 249478.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.289987373352051,
|
|
"epoch": 0.11715481171548117,
|
|
"grad_norm": 2.578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.522,
|
|
"mean_token_accuracy": 0.09086068868637084,
|
|
"num_tokens": 257714.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.143731307983398,
|
|
"epoch": 0.12133891213389121,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.5578,
|
|
"mean_token_accuracy": 0.08316774666309357,
|
|
"num_tokens": 266569.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.99811429977417,
|
|
"epoch": 0.12552301255230125,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.4546,
|
|
"mean_token_accuracy": 0.08273414000868798,
|
|
"num_tokens": 276688.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.839478302001954,
|
|
"epoch": 0.1297071129707113,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.3044,
|
|
"mean_token_accuracy": 0.08153865486383438,
|
|
"num_tokens": 285430.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.572208499908447,
|
|
"epoch": 0.13389121338912133,
|
|
"grad_norm": 1.8203125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.2963,
|
|
"mean_token_accuracy": 0.08305188342928886,
|
|
"num_tokens": 295727.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.346776962280273,
|
|
"epoch": 0.13807531380753138,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.2537,
|
|
"mean_token_accuracy": 0.0843418687582016,
|
|
"num_tokens": 304854.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.35010871887207,
|
|
"epoch": 0.14225941422594143,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.2224,
|
|
"mean_token_accuracy": 0.08320909515023231,
|
|
"num_tokens": 313579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.109733581542969,
|
|
"epoch": 0.14644351464435146,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.2637,
|
|
"mean_token_accuracy": 0.08496511876583099,
|
|
"num_tokens": 323310.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.118343544006347,
|
|
"epoch": 0.1506276150627615,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.1236,
|
|
"mean_token_accuracy": 0.0888123981654644,
|
|
"num_tokens": 331720.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.003835868835449,
|
|
"epoch": 0.15481171548117154,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.0728,
|
|
"mean_token_accuracy": 0.0916049025952816,
|
|
"num_tokens": 340331.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 7.846502780914307,
|
|
"epoch": 0.1589958158995816,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.0803,
|
|
"mean_token_accuracy": 0.09315933063626289,
|
|
"num_tokens": 349677.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 7.870132970809936,
|
|
"epoch": 0.16317991631799164,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.0721,
|
|
"mean_token_accuracy": 0.08652734383940697,
|
|
"num_tokens": 358674.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 7.712577533721924,
|
|
"epoch": 0.16736401673640167,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.0811,
|
|
"mean_token_accuracy": 0.08503658622503281,
|
|
"num_tokens": 368258.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.745207118988037,
|
|
"epoch": 0.17154811715481172,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.9726,
|
|
"mean_token_accuracy": 0.09250840395689011,
|
|
"num_tokens": 377083.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.754372835159302,
|
|
"epoch": 0.17573221757322174,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.0446,
|
|
"mean_token_accuracy": 0.0841131079941988,
|
|
"num_tokens": 385456.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.639361429214477,
|
|
"epoch": 0.1799163179916318,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.9265,
|
|
"mean_token_accuracy": 0.0942225843667984,
|
|
"num_tokens": 395295.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.558081722259521,
|
|
"epoch": 0.18410041841004185,
|
|
"grad_norm": 1.796875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.9772,
|
|
"mean_token_accuracy": 0.08982849642634391,
|
|
"num_tokens": 405023.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.534724760055542,
|
|
"epoch": 0.18828451882845187,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.9996,
|
|
"mean_token_accuracy": 0.08760679811239243,
|
|
"num_tokens": 414224.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.567825078964233,
|
|
"epoch": 0.19246861924686193,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.8948,
|
|
"mean_token_accuracy": 0.08697729334235191,
|
|
"num_tokens": 423951.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.65638952255249,
|
|
"epoch": 0.19665271966527198,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.9173,
|
|
"mean_token_accuracy": 0.08747867494821548,
|
|
"num_tokens": 432427.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.392250013351441,
|
|
"epoch": 0.200836820083682,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.8495,
|
|
"mean_token_accuracy": 0.09490769654512406,
|
|
"num_tokens": 441560.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.542767906188965,
|
|
"epoch": 0.20502092050209206,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.9423,
|
|
"mean_token_accuracy": 0.08346537575125694,
|
|
"num_tokens": 450364.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.399922561645508,
|
|
"epoch": 0.20920502092050208,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.8922,
|
|
"mean_token_accuracy": 0.08459680303931236,
|
|
"num_tokens": 460076.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.435028028488159,
|
|
"epoch": 0.21338912133891214,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.8343,
|
|
"mean_token_accuracy": 0.09127931594848633,
|
|
"num_tokens": 468718.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.456234979629516,
|
|
"epoch": 0.2175732217573222,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.9113,
|
|
"mean_token_accuracy": 0.08921206891536712,
|
|
"num_tokens": 478436.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.390152215957642,
|
|
"epoch": 0.2217573221757322,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.9278,
|
|
"mean_token_accuracy": 0.0872597984969616,
|
|
"num_tokens": 486833.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.41746563911438,
|
|
"epoch": 0.22594142259414227,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.7959,
|
|
"mean_token_accuracy": 0.09919953644275666,
|
|
"num_tokens": 495119.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.363225364685059,
|
|
"epoch": 0.2301255230125523,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.9685,
|
|
"mean_token_accuracy": 0.09823801144957542,
|
|
"num_tokens": 505396.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.373254871368408,
|
|
"epoch": 0.23430962343096234,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.8264,
|
|
"mean_token_accuracy": 0.08988286629319191,
|
|
"num_tokens": 515259.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.381401538848877,
|
|
"epoch": 0.2384937238493724,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.8366,
|
|
"mean_token_accuracy": 0.08750834614038468,
|
|
"num_tokens": 523608.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.272293424606323,
|
|
"epoch": 0.24267782426778242,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.8138,
|
|
"mean_token_accuracy": 0.09048807919025421,
|
|
"num_tokens": 534613.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.320169878005982,
|
|
"epoch": 0.24686192468619247,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.8579,
|
|
"mean_token_accuracy": 0.09750684052705764,
|
|
"num_tokens": 543347.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.278151416778565,
|
|
"epoch": 0.2510460251046025,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.8097,
|
|
"mean_token_accuracy": 0.08896707370877266,
|
|
"num_tokens": 552371.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.234066581726074,
|
|
"epoch": 0.25523012552301255,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.7401,
|
|
"mean_token_accuracy": 0.09521848261356354,
|
|
"num_tokens": 560742.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.309448719024658,
|
|
"epoch": 0.2594142259414226,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.8506,
|
|
"mean_token_accuracy": 0.09421026632189751,
|
|
"num_tokens": 569960.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.26940336227417,
|
|
"epoch": 0.26359832635983266,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.7952,
|
|
"mean_token_accuracy": 0.09265207946300506,
|
|
"num_tokens": 578859.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.1658327102661135,
|
|
"epoch": 0.26778242677824265,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.802,
|
|
"mean_token_accuracy": 0.09623523280024529,
|
|
"num_tokens": 588146.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.232771396636963,
|
|
"epoch": 0.2719665271966527,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.7437,
|
|
"mean_token_accuracy": 0.09764528870582581,
|
|
"num_tokens": 597268.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.080677461624146,
|
|
"epoch": 0.27615062761506276,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.676,
|
|
"mean_token_accuracy": 0.0969509594142437,
|
|
"num_tokens": 606199.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.3103187561035154,
|
|
"epoch": 0.2803347280334728,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.8102,
|
|
"mean_token_accuracy": 0.08201513588428497,
|
|
"num_tokens": 615966.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.242901706695557,
|
|
"epoch": 0.28451882845188287,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.7714,
|
|
"mean_token_accuracy": 0.0932280994951725,
|
|
"num_tokens": 625689.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.17906231880188,
|
|
"epoch": 0.28870292887029286,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.7957,
|
|
"mean_token_accuracy": 0.08773538246750831,
|
|
"num_tokens": 634877.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.1589775562286375,
|
|
"epoch": 0.2928870292887029,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.7211,
|
|
"mean_token_accuracy": 0.09845930486917495,
|
|
"num_tokens": 644042.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.122976160049438,
|
|
"epoch": 0.29707112970711297,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.7157,
|
|
"mean_token_accuracy": 0.09408211931586266,
|
|
"num_tokens": 653072.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.08035979270935,
|
|
"epoch": 0.301255230125523,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.7206,
|
|
"mean_token_accuracy": 0.09660579040646552,
|
|
"num_tokens": 663686.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.087335777282715,
|
|
"epoch": 0.3054393305439331,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.655,
|
|
"mean_token_accuracy": 0.09818260595202447,
|
|
"num_tokens": 672293.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.929145002365113,
|
|
"epoch": 0.30962343096234307,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.617,
|
|
"mean_token_accuracy": 0.10118300542235374,
|
|
"num_tokens": 681758.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.119603776931763,
|
|
"epoch": 0.3138075313807531,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.6496,
|
|
"mean_token_accuracy": 0.10070758163928986,
|
|
"num_tokens": 690727.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.051980209350586,
|
|
"epoch": 0.3179916317991632,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.6052,
|
|
"mean_token_accuracy": 0.09968686252832412,
|
|
"num_tokens": 701210.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.929204320907592,
|
|
"epoch": 0.32217573221757323,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.5924,
|
|
"mean_token_accuracy": 0.10255414322018623,
|
|
"num_tokens": 710869.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.104875373840332,
|
|
"epoch": 0.3263598326359833,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.7186,
|
|
"mean_token_accuracy": 0.095282444357872,
|
|
"num_tokens": 721455.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.909361505508423,
|
|
"epoch": 0.3305439330543933,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.5982,
|
|
"mean_token_accuracy": 0.10492872297763825,
|
|
"num_tokens": 730917.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.0232825756073,
|
|
"epoch": 0.33472803347280333,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.6235,
|
|
"mean_token_accuracy": 0.10053492337465286,
|
|
"num_tokens": 739302.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.034136819839477,
|
|
"epoch": 0.3389121338912134,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.5922,
|
|
"mean_token_accuracy": 0.09860573858022689,
|
|
"num_tokens": 748785.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.9551417350769045,
|
|
"epoch": 0.34309623430962344,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.10383330285549164,
|
|
"num_tokens": 757215.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.833004713058472,
|
|
"epoch": 0.3472803347280335,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.4628,
|
|
"mean_token_accuracy": 0.10961654931306838,
|
|
"num_tokens": 765682.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.001798439025879,
|
|
"epoch": 0.3514644351464435,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.6685,
|
|
"mean_token_accuracy": 0.09785127714276314,
|
|
"num_tokens": 774814.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.951812839508056,
|
|
"epoch": 0.35564853556485354,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.5856,
|
|
"mean_token_accuracy": 0.10029251947999,
|
|
"num_tokens": 784019.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.814007663726807,
|
|
"epoch": 0.3598326359832636,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.4521,
|
|
"mean_token_accuracy": 0.10874552577733994,
|
|
"num_tokens": 793237.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.967527246475219,
|
|
"epoch": 0.36401673640167365,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.5148,
|
|
"mean_token_accuracy": 0.1004909910261631,
|
|
"num_tokens": 801548.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.829876708984375,
|
|
"epoch": 0.3682008368200837,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.5539,
|
|
"mean_token_accuracy": 0.10507597774267197,
|
|
"num_tokens": 810974.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.837385892868042,
|
|
"epoch": 0.3723849372384937,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.4384,
|
|
"mean_token_accuracy": 0.10976337790489196,
|
|
"num_tokens": 820023.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.836634254455566,
|
|
"epoch": 0.37656903765690375,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.5294,
|
|
"mean_token_accuracy": 0.10367678925395012,
|
|
"num_tokens": 828297.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.901482248306275,
|
|
"epoch": 0.3807531380753138,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.5793,
|
|
"mean_token_accuracy": 0.09453905671834946,
|
|
"num_tokens": 838268.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.938200902938843,
|
|
"epoch": 0.38493723849372385,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.4984,
|
|
"mean_token_accuracy": 0.10036915764212609,
|
|
"num_tokens": 846491.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.742056369781494,
|
|
"epoch": 0.3891213389121339,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4374,
|
|
"mean_token_accuracy": 0.10895640999078751,
|
|
"num_tokens": 856501.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.806368494033814,
|
|
"epoch": 0.39330543933054396,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.5002,
|
|
"mean_token_accuracy": 0.10559085234999657,
|
|
"num_tokens": 867054.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.738630914688111,
|
|
"epoch": 0.39748953974895396,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.521,
|
|
"mean_token_accuracy": 0.10245606228709221,
|
|
"num_tokens": 875956.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.697034120559692,
|
|
"epoch": 0.401673640167364,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.10774662867188453,
|
|
"num_tokens": 884427.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.911193704605102,
|
|
"epoch": 0.40585774058577406,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.5239,
|
|
"mean_token_accuracy": 0.1094229944050312,
|
|
"num_tokens": 894181.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.690055084228516,
|
|
"epoch": 0.4100418410041841,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.3689,
|
|
"mean_token_accuracy": 0.11442587077617646,
|
|
"num_tokens": 904076.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.700846529006958,
|
|
"epoch": 0.41422594142259417,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5172,
|
|
"mean_token_accuracy": 0.10785161033272743,
|
|
"num_tokens": 913915.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.8667717456817625,
|
|
"epoch": 0.41841004184100417,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.5828,
|
|
"mean_token_accuracy": 0.10332566574215889,
|
|
"num_tokens": 922876.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.41841004184100417,
|
|
"eval_entropy": 6.700536389536166,
|
|
"eval_loss": 6.505311965942383,
|
|
"eval_mean_token_accuracy": 0.10990596924487554,
|
|
"eval_num_tokens": 922876.0,
|
|
"eval_runtime": 2.7776,
|
|
"eval_samples_per_second": 1406.597,
|
|
"eval_steps_per_second": 176.05,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.676968288421631,
|
|
"epoch": 0.4225941422594142,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.3727,
|
|
"mean_token_accuracy": 0.11308198049664497,
|
|
"num_tokens": 931998.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.805904197692871,
|
|
"epoch": 0.42677824267782427,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.4411,
|
|
"mean_token_accuracy": 0.10866107791662216,
|
|
"num_tokens": 941387.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.625282573699951,
|
|
"epoch": 0.4309623430962343,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.4789,
|
|
"mean_token_accuracy": 0.10362649038434028,
|
|
"num_tokens": 951866.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.657781028747559,
|
|
"epoch": 0.4351464435146444,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.4117,
|
|
"mean_token_accuracy": 0.10945287123322486,
|
|
"num_tokens": 961252.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.821563005447388,
|
|
"epoch": 0.4393305439330544,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.4206,
|
|
"mean_token_accuracy": 0.11085275113582611,
|
|
"num_tokens": 969857.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.679222440719604,
|
|
"epoch": 0.4435146443514644,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4176,
|
|
"mean_token_accuracy": 0.11243949085474014,
|
|
"num_tokens": 978754.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.723608922958374,
|
|
"epoch": 0.4476987447698745,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.391,
|
|
"mean_token_accuracy": 0.1103011853992939,
|
|
"num_tokens": 987935.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.682740306854248,
|
|
"epoch": 0.45188284518828453,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.3588,
|
|
"mean_token_accuracy": 0.11257985085248948,
|
|
"num_tokens": 996873.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.741128253936767,
|
|
"epoch": 0.4560669456066946,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.4534,
|
|
"mean_token_accuracy": 0.10576379746198654,
|
|
"num_tokens": 1006487.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.482868623733521,
|
|
"epoch": 0.4602510460251046,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.2856,
|
|
"mean_token_accuracy": 0.11558721140027046,
|
|
"num_tokens": 1015715.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.656096649169922,
|
|
"epoch": 0.46443514644351463,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4102,
|
|
"mean_token_accuracy": 0.11223409548401833,
|
|
"num_tokens": 1025091.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.6403412342071535,
|
|
"epoch": 0.4686192468619247,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.342,
|
|
"mean_token_accuracy": 0.10930742770433426,
|
|
"num_tokens": 1034174.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.697489976882935,
|
|
"epoch": 0.47280334728033474,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.549,
|
|
"mean_token_accuracy": 0.09722351431846618,
|
|
"num_tokens": 1043777.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.652381849288941,
|
|
"epoch": 0.4769874476987448,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3832,
|
|
"mean_token_accuracy": 0.11293103024363518,
|
|
"num_tokens": 1052749.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.7199958801269535,
|
|
"epoch": 0.4811715481171548,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4071,
|
|
"mean_token_accuracy": 0.11299470067024231,
|
|
"num_tokens": 1061564.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.540546655654907,
|
|
"epoch": 0.48535564853556484,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.3267,
|
|
"mean_token_accuracy": 0.11745149567723275,
|
|
"num_tokens": 1070006.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.592943572998047,
|
|
"epoch": 0.4895397489539749,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3903,
|
|
"mean_token_accuracy": 0.1095091424882412,
|
|
"num_tokens": 1079799.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.655527019500733,
|
|
"epoch": 0.49372384937238495,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.2219,
|
|
"mean_token_accuracy": 0.120665642619133,
|
|
"num_tokens": 1088842.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.531441926956177,
|
|
"epoch": 0.497907949790795,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.3064,
|
|
"mean_token_accuracy": 0.12399455532431602,
|
|
"num_tokens": 1097960.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.42484769821167,
|
|
"epoch": 0.502092050209205,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.2793,
|
|
"mean_token_accuracy": 0.11625320315361024,
|
|
"num_tokens": 1106797.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.580197429656982,
|
|
"epoch": 0.5062761506276151,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.4432,
|
|
"mean_token_accuracy": 0.11806596219539642,
|
|
"num_tokens": 1116384.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.462054252624512,
|
|
"epoch": 0.5104602510460251,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.2024,
|
|
"mean_token_accuracy": 0.12296005710959435,
|
|
"num_tokens": 1125206.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.673148584365845,
|
|
"epoch": 0.5146443514644351,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.4218,
|
|
"mean_token_accuracy": 0.11573300361633301,
|
|
"num_tokens": 1134110.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.7004382610321045,
|
|
"epoch": 0.5188284518828452,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.417,
|
|
"mean_token_accuracy": 0.11261885166168213,
|
|
"num_tokens": 1142833.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.535337591171265,
|
|
"epoch": 0.5230125523012552,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.3455,
|
|
"mean_token_accuracy": 0.11737915500998497,
|
|
"num_tokens": 1152006.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.4866430282592775,
|
|
"epoch": 0.5271966527196653,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.2844,
|
|
"mean_token_accuracy": 0.11385276094079018,
|
|
"num_tokens": 1161061.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.597733354568481,
|
|
"epoch": 0.5313807531380753,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.3047,
|
|
"mean_token_accuracy": 0.11783810779452324,
|
|
"num_tokens": 1170754.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.399986743927002,
|
|
"epoch": 0.5355648535564853,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.2454,
|
|
"mean_token_accuracy": 0.11883974969387054,
|
|
"num_tokens": 1180854.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.562164402008056,
|
|
"epoch": 0.5397489539748954,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.11495971083641052,
|
|
"num_tokens": 1190567.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.4136192321777346,
|
|
"epoch": 0.5439330543933054,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.1772,
|
|
"mean_token_accuracy": 0.12412636205554009,
|
|
"num_tokens": 1200247.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.411277961730957,
|
|
"epoch": 0.5481171548117155,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.29,
|
|
"mean_token_accuracy": 0.11657113656401634,
|
|
"num_tokens": 1209358.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.452598428726196,
|
|
"epoch": 0.5523012552301255,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.3091,
|
|
"mean_token_accuracy": 0.12059221267700196,
|
|
"num_tokens": 1217746.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.468178224563599,
|
|
"epoch": 0.5564853556485355,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.273,
|
|
"mean_token_accuracy": 0.12242133840918541,
|
|
"num_tokens": 1226829.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.409869623184204,
|
|
"epoch": 0.5606694560669456,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.1811,
|
|
"mean_token_accuracy": 0.1252070352435112,
|
|
"num_tokens": 1236642.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.611316394805908,
|
|
"epoch": 0.5648535564853556,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.3054,
|
|
"mean_token_accuracy": 0.12332511767745018,
|
|
"num_tokens": 1244934.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.345842742919922,
|
|
"epoch": 0.5690376569037657,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.1933,
|
|
"mean_token_accuracy": 0.11780603304505348,
|
|
"num_tokens": 1254786.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.343827581405639,
|
|
"epoch": 0.5732217573221757,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.12387140616774558,
|
|
"num_tokens": 1263745.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.4398839473724365,
|
|
"epoch": 0.5774058577405857,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.3144,
|
|
"mean_token_accuracy": 0.12714288905262947,
|
|
"num_tokens": 1273101.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.485600519180298,
|
|
"epoch": 0.5815899581589958,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2931,
|
|
"mean_token_accuracy": 0.11583595350384712,
|
|
"num_tokens": 1282547.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.455876684188842,
|
|
"epoch": 0.5857740585774058,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3152,
|
|
"mean_token_accuracy": 0.11679297238588333,
|
|
"num_tokens": 1291594.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.429898071289062,
|
|
"epoch": 0.5899581589958159,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.2199,
|
|
"mean_token_accuracy": 0.12549455389380454,
|
|
"num_tokens": 1300282.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.407459688186646,
|
|
"epoch": 0.5941422594142259,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.3096,
|
|
"mean_token_accuracy": 0.12226130589842796,
|
|
"num_tokens": 1309304.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.392927360534668,
|
|
"epoch": 0.5983263598326359,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.1472,
|
|
"mean_token_accuracy": 0.12595504522323608,
|
|
"num_tokens": 1317638.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.5016227722167965,
|
|
"epoch": 0.602510460251046,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.353,
|
|
"mean_token_accuracy": 0.11902206242084504,
|
|
"num_tokens": 1327242.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.401812124252319,
|
|
"epoch": 0.606694560669456,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.1596,
|
|
"mean_token_accuracy": 0.12535706236958505,
|
|
"num_tokens": 1335880.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.383457088470459,
|
|
"epoch": 0.6108786610878661,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2361,
|
|
"mean_token_accuracy": 0.12056306451559066,
|
|
"num_tokens": 1344909.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.346205854415894,
|
|
"epoch": 0.6150627615062761,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.1104,
|
|
"mean_token_accuracy": 0.12955476492643356,
|
|
"num_tokens": 1353997.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.379322576522827,
|
|
"epoch": 0.6192468619246861,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.12443547993898392,
|
|
"num_tokens": 1363293.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.49610013961792,
|
|
"epoch": 0.6234309623430963,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.2785,
|
|
"mean_token_accuracy": 0.11846815049648285,
|
|
"num_tokens": 1372174.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.406333255767822,
|
|
"epoch": 0.6276150627615062,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.1697,
|
|
"mean_token_accuracy": 0.12304460108280182,
|
|
"num_tokens": 1381067.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.325730276107788,
|
|
"epoch": 0.6317991631799164,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.1643,
|
|
"mean_token_accuracy": 0.12812735885381699,
|
|
"num_tokens": 1390072.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.292504358291626,
|
|
"epoch": 0.6359832635983264,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.156,
|
|
"mean_token_accuracy": 0.12865231558680534,
|
|
"num_tokens": 1399391.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.304660320281982,
|
|
"epoch": 0.6401673640167364,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.12419576495885849,
|
|
"num_tokens": 1408054.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.39186725616455,
|
|
"epoch": 0.6443514644351465,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.1785,
|
|
"mean_token_accuracy": 0.11784627810120582,
|
|
"num_tokens": 1417406.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.323220872879029,
|
|
"epoch": 0.6485355648535565,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.237,
|
|
"mean_token_accuracy": 0.11933397874236107,
|
|
"num_tokens": 1426907.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.324512672424317,
|
|
"epoch": 0.6527196652719666,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3007,
|
|
"mean_token_accuracy": 0.11848932132124901,
|
|
"num_tokens": 1436085.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.432321214675904,
|
|
"epoch": 0.6569037656903766,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.208,
|
|
"mean_token_accuracy": 0.12037489041686059,
|
|
"num_tokens": 1444823.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.354400157928467,
|
|
"epoch": 0.6610878661087866,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.2424,
|
|
"mean_token_accuracy": 0.1207665555179119,
|
|
"num_tokens": 1454621.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.24983959197998,
|
|
"epoch": 0.6652719665271967,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.1116,
|
|
"mean_token_accuracy": 0.12615672051906585,
|
|
"num_tokens": 1463784.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.329709482192993,
|
|
"epoch": 0.6694560669456067,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.2278,
|
|
"mean_token_accuracy": 0.12423735633492469,
|
|
"num_tokens": 1473257.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.327875566482544,
|
|
"epoch": 0.6736401673640168,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.1715,
|
|
"mean_token_accuracy": 0.13215391784906388,
|
|
"num_tokens": 1482195.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.267716884613037,
|
|
"epoch": 0.6778242677824268,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.2104,
|
|
"mean_token_accuracy": 0.12745170965790747,
|
|
"num_tokens": 1491590.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.32491455078125,
|
|
"epoch": 0.6820083682008368,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.0945,
|
|
"mean_token_accuracy": 0.1320909932255745,
|
|
"num_tokens": 1500208.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.329631280899048,
|
|
"epoch": 0.6861924686192469,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.2269,
|
|
"mean_token_accuracy": 0.12872799932956697,
|
|
"num_tokens": 1509412.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.2926709175109865,
|
|
"epoch": 0.6903765690376569,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2233,
|
|
"mean_token_accuracy": 0.12587290704250337,
|
|
"num_tokens": 1518426.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.372130298614502,
|
|
"epoch": 0.694560669456067,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.173,
|
|
"mean_token_accuracy": 0.12464739456772804,
|
|
"num_tokens": 1528605.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.198592662811279,
|
|
"epoch": 0.698744769874477,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.1198,
|
|
"mean_token_accuracy": 0.12950538620352745,
|
|
"num_tokens": 1537449.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.2445250988006595,
|
|
"epoch": 0.702928870292887,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1531,
|
|
"mean_token_accuracy": 0.12678662985563277,
|
|
"num_tokens": 1546743.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.198931741714477,
|
|
"epoch": 0.7071129707112971,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1512,
|
|
"mean_token_accuracy": 0.12723555639386178,
|
|
"num_tokens": 1555691.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.251468992233276,
|
|
"epoch": 0.7112970711297071,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.1,
|
|
"mean_token_accuracy": 0.13119761049747466,
|
|
"num_tokens": 1564957.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.330463075637818,
|
|
"epoch": 0.7154811715481172,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1739,
|
|
"mean_token_accuracy": 0.12831180393695832,
|
|
"num_tokens": 1573901.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.1841747760772705,
|
|
"epoch": 0.7196652719665272,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.1514,
|
|
"mean_token_accuracy": 0.13142475932836534,
|
|
"num_tokens": 1582644.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.314925765991211,
|
|
"epoch": 0.7238493723849372,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2846,
|
|
"mean_token_accuracy": 0.12040768340229988,
|
|
"num_tokens": 1592363.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.307796049118042,
|
|
"epoch": 0.7280334728033473,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.2164,
|
|
"mean_token_accuracy": 0.12550526708364487,
|
|
"num_tokens": 1602012.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.193496656417847,
|
|
"epoch": 0.7322175732217573,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.1262,
|
|
"mean_token_accuracy": 0.12875815853476524,
|
|
"num_tokens": 1611625.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.1236217498779295,
|
|
"epoch": 0.7364016736401674,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.0812,
|
|
"mean_token_accuracy": 0.12527592107653618,
|
|
"num_tokens": 1620941.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.263450574874878,
|
|
"epoch": 0.7405857740585774,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.1654,
|
|
"mean_token_accuracy": 0.12486932873725891,
|
|
"num_tokens": 1629225.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.240986633300781,
|
|
"epoch": 0.7447698744769874,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.1445,
|
|
"mean_token_accuracy": 0.12441490367054939,
|
|
"num_tokens": 1639551.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.223584127426148,
|
|
"epoch": 0.7489539748953975,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.1279,
|
|
"mean_token_accuracy": 0.12395419999957084,
|
|
"num_tokens": 1648963.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.18436336517334,
|
|
"epoch": 0.7531380753138075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.0355,
|
|
"mean_token_accuracy": 0.13170942068099975,
|
|
"num_tokens": 1657735.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.318353271484375,
|
|
"epoch": 0.7573221757322176,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.0891,
|
|
"mean_token_accuracy": 0.1331306591629982,
|
|
"num_tokens": 1666349.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.016112899780273,
|
|
"epoch": 0.7615062761506276,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.0271,
|
|
"mean_token_accuracy": 0.1322978489100933,
|
|
"num_tokens": 1675920.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.141036748886108,
|
|
"epoch": 0.7656903765690377,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.0628,
|
|
"mean_token_accuracy": 0.13066673129796982,
|
|
"num_tokens": 1685626.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.278197765350342,
|
|
"epoch": 0.7698744769874477,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.1888,
|
|
"mean_token_accuracy": 0.1338045746088028,
|
|
"num_tokens": 1694972.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.141257333755493,
|
|
"epoch": 0.7740585774058577,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.0668,
|
|
"mean_token_accuracy": 0.13685579001903533,
|
|
"num_tokens": 1703806.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.192854118347168,
|
|
"epoch": 0.7782426778242678,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.1659,
|
|
"mean_token_accuracy": 0.13321068361401558,
|
|
"num_tokens": 1712949.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.347252273559571,
|
|
"epoch": 0.7824267782426778,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.287,
|
|
"mean_token_accuracy": 0.12331678792834282,
|
|
"num_tokens": 1722328.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.2889646053314205,
|
|
"epoch": 0.7866108786610879,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.253,
|
|
"mean_token_accuracy": 0.13125553727149963,
|
|
"num_tokens": 1731880.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.093116664886475,
|
|
"epoch": 0.7907949790794979,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.0693,
|
|
"mean_token_accuracy": 0.13057696744799613,
|
|
"num_tokens": 1740719.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.186368799209594,
|
|
"epoch": 0.7949790794979079,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.1042,
|
|
"mean_token_accuracy": 0.12656014263629914,
|
|
"num_tokens": 1749753.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.255837249755859,
|
|
"epoch": 0.799163179916318,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.1266303114593029,
|
|
"num_tokens": 1758913.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.229157781600952,
|
|
"epoch": 0.803347280334728,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1442,
|
|
"mean_token_accuracy": 0.12871945798397064,
|
|
"num_tokens": 1768536.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.113988447189331,
|
|
"epoch": 0.8075313807531381,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.12936384379863738,
|
|
"num_tokens": 1777291.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.186689996719361,
|
|
"epoch": 0.8117154811715481,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.1634,
|
|
"mean_token_accuracy": 0.12880500480532647,
|
|
"num_tokens": 1787186.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.11544098854065,
|
|
"epoch": 0.8158995815899581,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000487,
|
|
"loss": 5.9697,
|
|
"mean_token_accuracy": 0.13747073709964752,
|
|
"num_tokens": 1795581.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 5.999046802520752,
|
|
"epoch": 0.8200836820083682,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 5.9825,
|
|
"mean_token_accuracy": 0.1357624500989914,
|
|
"num_tokens": 1804627.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.198841238021851,
|
|
"epoch": 0.8242677824267782,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.1134,
|
|
"mean_token_accuracy": 0.13310517519712448,
|
|
"num_tokens": 1813808.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.064494943618774,
|
|
"epoch": 0.8284518828451883,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.1108,
|
|
"mean_token_accuracy": 0.1350795842707157,
|
|
"num_tokens": 1823422.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.163973569869995,
|
|
"epoch": 0.8326359832635983,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.1526,
|
|
"mean_token_accuracy": 0.13110492676496505,
|
|
"num_tokens": 1832437.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.180834436416626,
|
|
"epoch": 0.8368200836820083,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0707,
|
|
"mean_token_accuracy": 0.12942072749137878,
|
|
"num_tokens": 1842063.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.8368200836820083,
|
|
"eval_entropy": 5.93860879558727,
|
|
"eval_loss": 6.10989236831665,
|
|
"eval_mean_token_accuracy": 0.14120837797721228,
|
|
"eval_num_tokens": 1842063.0,
|
|
"eval_runtime": 2.7411,
|
|
"eval_samples_per_second": 1425.332,
|
|
"eval_steps_per_second": 178.395,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.116349458694458,
|
|
"epoch": 0.8410041841004184,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999998518355666,
|
|
"loss": 6.2017,
|
|
"mean_token_accuracy": 0.13061145469546317,
|
|
"num_tokens": 1851043.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.059402561187744,
|
|
"epoch": 0.8451882845188284,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999992499178905,
|
|
"loss": 6.0564,
|
|
"mean_token_accuracy": 0.13553041368722915,
|
|
"num_tokens": 1860268.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.27886700630188,
|
|
"epoch": 0.8493723849372385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999981849879322,
|
|
"loss": 6.1793,
|
|
"mean_token_accuracy": 0.133279487490654,
|
|
"num_tokens": 1870096.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.025306558609008,
|
|
"epoch": 0.8535564853556485,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999966570478833,
|
|
"loss": 6.031,
|
|
"mean_token_accuracy": 0.13792885467410088,
|
|
"num_tokens": 1879164.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.075544977188111,
|
|
"epoch": 0.8577405857740585,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999946661008878,
|
|
"loss": 6.0018,
|
|
"mean_token_accuracy": 0.14045969545841216,
|
|
"num_tokens": 1888135.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.10041766166687,
|
|
"epoch": 0.8619246861924686,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000499992212151043,
|
|
"loss": 6.0593,
|
|
"mean_token_accuracy": 0.13504695892333984,
|
|
"num_tokens": 1898302.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.017566347122193,
|
|
"epoch": 0.8661087866108786,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999892952033987,
|
|
"loss": 6.0455,
|
|
"mean_token_accuracy": 0.13498899713158607,
|
|
"num_tokens": 1907753.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.111660099029541,
|
|
"epoch": 0.8702928870292888,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999859152639573,
|
|
"loss": 6.0849,
|
|
"mean_token_accuracy": 0.12984225451946257,
|
|
"num_tokens": 1917543.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.189549350738526,
|
|
"epoch": 0.8744769874476988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999820723396745,
|
|
"loss": 6.1305,
|
|
"mean_token_accuracy": 0.1277165487408638,
|
|
"num_tokens": 1927335.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.131802034378052,
|
|
"epoch": 0.8786610878661087,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999777664384581,
|
|
"loss": 6.0468,
|
|
"mean_token_accuracy": 0.13682940900325774,
|
|
"num_tokens": 1936381.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.053474760055542,
|
|
"epoch": 0.8828451882845189,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999729975691692,
|
|
"loss": 6.0878,
|
|
"mean_token_accuracy": 0.13545934334397317,
|
|
"num_tokens": 1945559.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.001845169067383,
|
|
"epoch": 0.8870292887029289,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999677657416212,
|
|
"loss": 5.9453,
|
|
"mean_token_accuracy": 0.13565901443362235,
|
|
"num_tokens": 1954709.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.0640534400939945,
|
|
"epoch": 0.891213389121339,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999620709665804,
|
|
"loss": 6.0854,
|
|
"mean_token_accuracy": 0.13231634050607682,
|
|
"num_tokens": 1963664.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.13521728515625,
|
|
"epoch": 0.895397489539749,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999559132557659,
|
|
"loss": 6.1175,
|
|
"mean_token_accuracy": 0.13134895339608194,
|
|
"num_tokens": 1974103.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.173500108718872,
|
|
"epoch": 0.899581589958159,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004999492926218489,
|
|
"loss": 6.0113,
|
|
"mean_token_accuracy": 0.1355544939637184,
|
|
"num_tokens": 1982924.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 5.965735149383545,
|
|
"epoch": 0.9037656903765691,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499942209078454,
|
|
"loss": 5.9378,
|
|
"mean_token_accuracy": 0.1465287446975708,
|
|
"num_tokens": 1992033.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 5.998280811309814,
|
|
"epoch": 0.9079497907949791,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999346626401578,
|
|
"loss": 5.9444,
|
|
"mean_token_accuracy": 0.14567727744579315,
|
|
"num_tokens": 2001218.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 5.986832046508789,
|
|
"epoch": 0.9121338912133892,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999266533224897,
|
|
"loss": 6.1065,
|
|
"mean_token_accuracy": 0.1391438327729702,
|
|
"num_tokens": 2010463.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.14104437828064,
|
|
"epoch": 0.9163179916317992,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999181811419314,
|
|
"loss": 5.9562,
|
|
"mean_token_accuracy": 0.13891242519021035,
|
|
"num_tokens": 2019246.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.079975891113281,
|
|
"epoch": 0.9205020920502092,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999092461159177,
|
|
"loss": 6.0431,
|
|
"mean_token_accuracy": 0.13607420697808265,
|
|
"num_tokens": 2028083.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 5.9877180576324465,
|
|
"epoch": 0.9246861924686193,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998998482628351,
|
|
"loss": 5.8649,
|
|
"mean_token_accuracy": 0.14478956088423728,
|
|
"num_tokens": 2036631.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 5.946710157394409,
|
|
"epoch": 0.9288702928870293,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004998899876020229,
|
|
"loss": 5.9929,
|
|
"mean_token_accuracy": 0.13790326938033104,
|
|
"num_tokens": 2045910.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 5.976225805282593,
|
|
"epoch": 0.9330543933054394,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499879664153773,
|
|
"loss": 6.0313,
|
|
"mean_token_accuracy": 0.1479876421391964,
|
|
"num_tokens": 2055959.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.234367895126343,
|
|
"epoch": 0.9372384937238494,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998688779393291,
|
|
"loss": 6.1265,
|
|
"mean_token_accuracy": 0.13584480583667755,
|
|
"num_tokens": 2065475.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 5.7713428974151615,
|
|
"epoch": 0.9414225941422594,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998576289808877,
|
|
"loss": 5.8793,
|
|
"mean_token_accuracy": 0.14183773547410966,
|
|
"num_tokens": 2074469.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.010820722579956,
|
|
"epoch": 0.9456066945606695,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004998459173015972,
|
|
"loss": 6.0604,
|
|
"mean_token_accuracy": 0.13986758440732955,
|
|
"num_tokens": 2084604.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.090184593200684,
|
|
"epoch": 0.9497907949790795,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004998337429255584,
|
|
"loss": 6.0115,
|
|
"mean_token_accuracy": 0.1413180112838745,
|
|
"num_tokens": 2093377.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.078766250610352,
|
|
"epoch": 0.9539748953974896,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998211058778243,
|
|
"loss": 6.045,
|
|
"mean_token_accuracy": 0.13628122508525847,
|
|
"num_tokens": 2102798.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.039411830902099,
|
|
"epoch": 0.9581589958158996,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998080061843999,
|
|
"loss": 6.0185,
|
|
"mean_token_accuracy": 0.1370515465736389,
|
|
"num_tokens": 2112516.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.11058464050293,
|
|
"epoch": 0.9623430962343096,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000499794443872242,
|
|
"loss": 6.1106,
|
|
"mean_token_accuracy": 0.13264601677656174,
|
|
"num_tokens": 2121677.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.006878089904785,
|
|
"epoch": 0.9665271966527197,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997804189692601,
|
|
"loss": 5.9499,
|
|
"mean_token_accuracy": 0.14087534546852112,
|
|
"num_tokens": 2131220.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.021556997299195,
|
|
"epoch": 0.9707112970711297,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004997659315043149,
|
|
"loss": 6.0939,
|
|
"mean_token_accuracy": 0.13201906755566598,
|
|
"num_tokens": 2142104.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.0106141567230225,
|
|
"epoch": 0.9748953974895398,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004997509815072191,
|
|
"loss": 6.0054,
|
|
"mean_token_accuracy": 0.13905177861452103,
|
|
"num_tokens": 2151970.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 5.97538275718689,
|
|
"epoch": 0.9790794979079498,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004997355690087379,
|
|
"loss": 5.9543,
|
|
"mean_token_accuracy": 0.14026949256658555,
|
|
"num_tokens": 2161371.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 5.91616268157959,
|
|
"epoch": 0.9832635983263598,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004997196940405874,
|
|
"loss": 5.9131,
|
|
"mean_token_accuracy": 0.1355208821594715,
|
|
"num_tokens": 2170650.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.999844169616699,
|
|
"epoch": 0.9874476987447699,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004997033566354358,
|
|
"loss": 5.993,
|
|
"mean_token_accuracy": 0.14383321404457092,
|
|
"num_tokens": 2180934.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.03171181678772,
|
|
"epoch": 0.9916317991631799,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996865568269028,
|
|
"loss": 5.9362,
|
|
"mean_token_accuracy": 0.14685204923152922,
|
|
"num_tokens": 2189894.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.93413290977478,
|
|
"epoch": 0.99581589958159,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004996692946495597,
|
|
"loss": 5.8959,
|
|
"mean_token_accuracy": 0.1524686649441719,
|
|
"num_tokens": 2200208.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 5.956554412841797,
|
|
"epoch": 1.0,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004996515701389295,
|
|
"loss": 5.9548,
|
|
"mean_token_accuracy": 0.1403520457446575,
|
|
"num_tokens": 2209322.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 5.894754028320312,
|
|
"epoch": 1.00418410041841,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004996333833314862,
|
|
"loss": 5.7446,
|
|
"mean_token_accuracy": 0.15427049696445466,
|
|
"num_tokens": 2217663.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 5.828617906570434,
|
|
"epoch": 1.00836820083682,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996147342646554,
|
|
"loss": 5.7845,
|
|
"mean_token_accuracy": 0.14702118411660195,
|
|
"num_tokens": 2226559.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.070319509506225,
|
|
"epoch": 1.0125523012552302,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995956229768137,
|
|
"loss": 5.819,
|
|
"mean_token_accuracy": 0.1527664080262184,
|
|
"num_tokens": 2235468.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 5.813411045074463,
|
|
"epoch": 1.0167364016736402,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004995760495072893,
|
|
"loss": 5.6136,
|
|
"mean_token_accuracy": 0.1567228153347969,
|
|
"num_tokens": 2245176.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 5.895093011856079,
|
|
"epoch": 1.0209205020920502,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004995560138963611,
|
|
"loss": 5.7825,
|
|
"mean_token_accuracy": 0.14997974932193756,
|
|
"num_tokens": 2255123.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 5.8670361042022705,
|
|
"epoch": 1.0251046025104602,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004995355161852593,
|
|
"loss": 5.702,
|
|
"mean_token_accuracy": 0.15056501626968383,
|
|
"num_tokens": 2263596.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 5.887784576416015,
|
|
"epoch": 1.0292887029288702,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004995145564161648,
|
|
"loss": 5.7436,
|
|
"mean_token_accuracy": 0.14897513389587402,
|
|
"num_tokens": 2272228.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.846974182128906,
|
|
"epoch": 1.0334728033472804,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994931346322094,
|
|
"loss": 5.7995,
|
|
"mean_token_accuracy": 0.1486528918147087,
|
|
"num_tokens": 2281248.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 5.952018976211548,
|
|
"epoch": 1.0376569037656904,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004994712508774757,
|
|
"loss": 5.76,
|
|
"mean_token_accuracy": 0.14765810891985892,
|
|
"num_tokens": 2290958.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.779097032546997,
|
|
"epoch": 1.0418410041841004,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000499448905196997,
|
|
"loss": 5.714,
|
|
"mean_token_accuracy": 0.15628314018249512,
|
|
"num_tokens": 2299763.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 5.858342695236206,
|
|
"epoch": 1.0460251046025104,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994260976367573,
|
|
"loss": 5.7272,
|
|
"mean_token_accuracy": 0.15564334839582444,
|
|
"num_tokens": 2309847.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 5.9082927227020265,
|
|
"epoch": 1.0502092050209204,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994028282436905,
|
|
"loss": 5.7491,
|
|
"mean_token_accuracy": 0.14895465224981308,
|
|
"num_tokens": 2317556.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 5.7722269058227536,
|
|
"epoch": 1.0543933054393306,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993790970656814,
|
|
"loss": 5.6812,
|
|
"mean_token_accuracy": 0.15420028567314148,
|
|
"num_tokens": 2327151.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.7930036067962645,
|
|
"epoch": 1.0585774058577406,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993549041515652,
|
|
"loss": 5.6802,
|
|
"mean_token_accuracy": 0.15717751085758208,
|
|
"num_tokens": 2336119.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 5.781204509735107,
|
|
"epoch": 1.0627615062761506,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993302495511268,
|
|
"loss": 5.7134,
|
|
"mean_token_accuracy": 0.15141206830739976,
|
|
"num_tokens": 2345262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.001182174682617,
|
|
"epoch": 1.0669456066945606,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004993051333151014,
|
|
"loss": 5.847,
|
|
"mean_token_accuracy": 0.1470116265118122,
|
|
"num_tokens": 2355210.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 5.718861150741577,
|
|
"epoch": 1.0711297071129706,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992795554951742,
|
|
"loss": 5.5887,
|
|
"mean_token_accuracy": 0.15690547823905945,
|
|
"num_tokens": 2364599.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 5.716666555404663,
|
|
"epoch": 1.0753138075313808,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004992535161439803,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.1451686941087246,
|
|
"num_tokens": 2374143.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 5.969261169433594,
|
|
"epoch": 1.0794979079497908,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004992270153151045,
|
|
"loss": 5.7635,
|
|
"mean_token_accuracy": 0.15289389342069626,
|
|
"num_tokens": 2383155.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.92108302116394,
|
|
"epoch": 1.0836820083682008,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992000530630812,
|
|
"loss": 5.8095,
|
|
"mean_token_accuracy": 0.1509893462061882,
|
|
"num_tokens": 2392542.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 5.737433242797851,
|
|
"epoch": 1.0878661087866108,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004991726294433943,
|
|
"loss": 5.7792,
|
|
"mean_token_accuracy": 0.14687865376472473,
|
|
"num_tokens": 2401851.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.892284393310547,
|
|
"epoch": 1.0920502092050208,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004991447445124774,
|
|
"loss": 5.7048,
|
|
"mean_token_accuracy": 0.1529296711087227,
|
|
"num_tokens": 2410726.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.729634380340576,
|
|
"epoch": 1.096234309623431,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499116398327713,
|
|
"loss": 5.7891,
|
|
"mean_token_accuracy": 0.14780578166246414,
|
|
"num_tokens": 2420702.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 5.774093294143677,
|
|
"epoch": 1.100418410041841,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004990875909474331,
|
|
"loss": 5.7044,
|
|
"mean_token_accuracy": 0.1555623084306717,
|
|
"num_tokens": 2431335.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 5.8864298343658445,
|
|
"epoch": 1.104602510460251,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004990583224309187,
|
|
"loss": 5.6981,
|
|
"mean_token_accuracy": 0.15561241209506987,
|
|
"num_tokens": 2440722.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 5.802323818206787,
|
|
"epoch": 1.108786610878661,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004990285928383997,
|
|
"loss": 5.7621,
|
|
"mean_token_accuracy": 0.1484719768166542,
|
|
"num_tokens": 2451230.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 5.884494256973267,
|
|
"epoch": 1.112970711297071,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004989984022310545,
|
|
"loss": 5.7884,
|
|
"mean_token_accuracy": 0.14925398379564286,
|
|
"num_tokens": 2461442.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.787790966033936,
|
|
"epoch": 1.1171548117154813,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004989677506710109,
|
|
"loss": 5.6094,
|
|
"mean_token_accuracy": 0.16102316826581956,
|
|
"num_tokens": 2471381.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.780320405960083,
|
|
"epoch": 1.1213389121338913,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004989366382213449,
|
|
"loss": 5.6627,
|
|
"mean_token_accuracy": 0.15939401984214782,
|
|
"num_tokens": 2479937.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 5.739678001403808,
|
|
"epoch": 1.1255230125523012,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989050649460806,
|
|
"loss": 5.6362,
|
|
"mean_token_accuracy": 0.15941027402877808,
|
|
"num_tokens": 2488974.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 5.791116237640381,
|
|
"epoch": 1.1297071129707112,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000498873030910191,
|
|
"loss": 5.781,
|
|
"mean_token_accuracy": 0.150092051923275,
|
|
"num_tokens": 2498800.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.717321586608887,
|
|
"epoch": 1.1338912133891212,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004988405361795969,
|
|
"loss": 5.6848,
|
|
"mean_token_accuracy": 0.14925537556409835,
|
|
"num_tokens": 2506580.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.770847225189209,
|
|
"epoch": 1.1380753138075315,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004988075808211672,
|
|
"loss": 5.739,
|
|
"mean_token_accuracy": 0.1563566356897354,
|
|
"num_tokens": 2516342.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.770087480545044,
|
|
"epoch": 1.1422594142259415,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004987741649027189,
|
|
"loss": 5.7688,
|
|
"mean_token_accuracy": 0.15330522656440734,
|
|
"num_tokens": 2526054.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.847585964202881,
|
|
"epoch": 1.1464435146443515,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004987402884930163,
|
|
"loss": 5.7187,
|
|
"mean_token_accuracy": 0.15662140548229217,
|
|
"num_tokens": 2535556.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.832192087173462,
|
|
"epoch": 1.1506276150627615,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004987059516617717,
|
|
"loss": 5.6932,
|
|
"mean_token_accuracy": 0.15749878734350203,
|
|
"num_tokens": 2544930.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 5.700648450851441,
|
|
"epoch": 1.1548117154811715,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004986711544796449,
|
|
"loss": 5.6074,
|
|
"mean_token_accuracy": 0.16140320003032685,
|
|
"num_tokens": 2554362.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.79828109741211,
|
|
"epoch": 1.1589958158995817,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000498635897018243,
|
|
"loss": 5.7171,
|
|
"mean_token_accuracy": 0.15348803699016572,
|
|
"num_tokens": 2563499.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 5.725475692749024,
|
|
"epoch": 1.1631799163179917,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004986001793501198,
|
|
"loss": 5.741,
|
|
"mean_token_accuracy": 0.15232713371515275,
|
|
"num_tokens": 2573346.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.78861083984375,
|
|
"epoch": 1.1673640167364017,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004985640015487768,
|
|
"loss": 5.6204,
|
|
"mean_token_accuracy": 0.16266809701919555,
|
|
"num_tokens": 2582541.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.745808744430542,
|
|
"epoch": 1.1715481171548117,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004985273636886621,
|
|
"loss": 5.6995,
|
|
"mean_token_accuracy": 0.1560098946094513,
|
|
"num_tokens": 2592015.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.726541948318482,
|
|
"epoch": 1.1757322175732217,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004984902658451704,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.15896441042423248,
|
|
"num_tokens": 2601998.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.72317328453064,
|
|
"epoch": 1.1799163179916319,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984527080946432,
|
|
"loss": 5.6508,
|
|
"mean_token_accuracy": 0.15956526398658752,
|
|
"num_tokens": 2611435.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.779344701766968,
|
|
"epoch": 1.1841004184100419,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004984146905143683,
|
|
"loss": 5.7451,
|
|
"mean_token_accuracy": 0.15034509003162383,
|
|
"num_tokens": 2621639.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 5.777336454391479,
|
|
"epoch": 1.1882845188284519,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004983762131825797,
|
|
"loss": 5.7047,
|
|
"mean_token_accuracy": 0.1531509578227997,
|
|
"num_tokens": 2631390.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.752439022064209,
|
|
"epoch": 1.1924686192468619,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004983372761784576,
|
|
"loss": 5.7487,
|
|
"mean_token_accuracy": 0.14954135715961456,
|
|
"num_tokens": 2640769.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.865439939498901,
|
|
"epoch": 1.196652719665272,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004982978795821281,
|
|
"loss": 5.7567,
|
|
"mean_token_accuracy": 0.1547407925128937,
|
|
"num_tokens": 2651137.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.814016962051392,
|
|
"epoch": 1.200836820083682,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004982580234746632,
|
|
"loss": 5.6655,
|
|
"mean_token_accuracy": 0.15743645280599594,
|
|
"num_tokens": 2660530.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.694674301147461,
|
|
"epoch": 1.205020920502092,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00049821770793808,
|
|
"loss": 5.7379,
|
|
"mean_token_accuracy": 0.15339982360601426,
|
|
"num_tokens": 2669541.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.861599159240723,
|
|
"epoch": 1.209205020920502,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004981769330553418,
|
|
"loss": 5.737,
|
|
"mean_token_accuracy": 0.15259334444999695,
|
|
"num_tokens": 2678542.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.763570022583008,
|
|
"epoch": 1.213389121338912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004981356989103566,
|
|
"loss": 5.7951,
|
|
"mean_token_accuracy": 0.15657299458980561,
|
|
"num_tokens": 2687720.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.803497743606568,
|
|
"epoch": 1.217573221757322,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004980940055879774,
|
|
"loss": 5.6471,
|
|
"mean_token_accuracy": 0.15510508120059968,
|
|
"num_tokens": 2697641.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.740324783325195,
|
|
"epoch": 1.2217573221757323,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004980518531740027,
|
|
"loss": 5.7073,
|
|
"mean_token_accuracy": 0.1536521553993225,
|
|
"num_tokens": 2706600.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.839646625518799,
|
|
"epoch": 1.2259414225941423,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004980092417551752,
|
|
"loss": 5.818,
|
|
"mean_token_accuracy": 0.15001391619443893,
|
|
"num_tokens": 2716633.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.8136616230010985,
|
|
"epoch": 1.2301255230125523,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004979661714191825,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.16975591331720352,
|
|
"num_tokens": 2724829.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.69421501159668,
|
|
"epoch": 1.2343096234309623,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004979226422546563,
|
|
"loss": 5.7013,
|
|
"mean_token_accuracy": 0.1594356283545494,
|
|
"num_tokens": 2734276.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.816491365432739,
|
|
"epoch": 1.2384937238493725,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004978786543511726,
|
|
"loss": 5.8119,
|
|
"mean_token_accuracy": 0.15787640661001207,
|
|
"num_tokens": 2744413.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.775937271118164,
|
|
"epoch": 1.2426778242677825,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004978342077992515,
|
|
"loss": 5.6762,
|
|
"mean_token_accuracy": 0.15499028116464614,
|
|
"num_tokens": 2753242.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.606312656402588,
|
|
"epoch": 1.2468619246861925,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004977893026903568,
|
|
"loss": 5.6161,
|
|
"mean_token_accuracy": 0.166127747297287,
|
|
"num_tokens": 2761811.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.812732315063476,
|
|
"epoch": 1.2510460251046025,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004977439391168963,
|
|
"loss": 5.7151,
|
|
"mean_token_accuracy": 0.16037967205047607,
|
|
"num_tokens": 2770560.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.781078672409057,
|
|
"epoch": 1.2552301255230125,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004976981171722205,
|
|
"loss": 5.7108,
|
|
"mean_token_accuracy": 0.15685239136219026,
|
|
"num_tokens": 2780131.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.2552301255230125,
|
|
"eval_entropy": 5.562492416436443,
|
|
"eval_loss": 5.858166217803955,
|
|
"eval_mean_token_accuracy": 0.15818902674932908,
|
|
"eval_num_tokens": 2780131.0,
|
|
"eval_runtime": 2.9229,
|
|
"eval_samples_per_second": 1336.691,
|
|
"eval_steps_per_second": 167.3,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.832910346984863,
|
|
"epoch": 1.2594142259414225,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976518369506237,
|
|
"loss": 5.8164,
|
|
"mean_token_accuracy": 0.14845228791236878,
|
|
"num_tokens": 2789319.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.669151735305786,
|
|
"epoch": 1.2635983263598327,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004976050985473433,
|
|
"loss": 5.6901,
|
|
"mean_token_accuracy": 0.1474151998758316,
|
|
"num_tokens": 2799253.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.815981101989746,
|
|
"epoch": 1.2677824267782427,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004975579020585593,
|
|
"loss": 5.7889,
|
|
"mean_token_accuracy": 0.1519041746854782,
|
|
"num_tokens": 2808968.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 5.869286918640137,
|
|
"epoch": 1.2719665271966527,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004975102475813947,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.15941894352436065,
|
|
"num_tokens": 2817282.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.687379598617554,
|
|
"epoch": 1.2761506276150627,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004974621352139144,
|
|
"loss": 5.7378,
|
|
"mean_token_accuracy": 0.1533280447125435,
|
|
"num_tokens": 2826395.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.823107051849365,
|
|
"epoch": 1.280334728033473,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004974135650551262,
|
|
"loss": 5.8028,
|
|
"mean_token_accuracy": 0.15243592411279677,
|
|
"num_tokens": 2836195.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.762291669845581,
|
|
"epoch": 1.284518828451883,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004973645372049796,
|
|
"loss": 5.6422,
|
|
"mean_token_accuracy": 0.15816431641578674,
|
|
"num_tokens": 2846020.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.687832546234131,
|
|
"epoch": 1.288702928870293,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004973150517643659,
|
|
"loss": 5.5872,
|
|
"mean_token_accuracy": 0.16854178458452224,
|
|
"num_tokens": 2854895.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 5.724603605270386,
|
|
"epoch": 1.292887029288703,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004972651088351185,
|
|
"loss": 5.6319,
|
|
"mean_token_accuracy": 0.15647992193698884,
|
|
"num_tokens": 2864491.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.78979229927063,
|
|
"epoch": 1.297071129707113,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004972147085200116,
|
|
"loss": 5.6483,
|
|
"mean_token_accuracy": 0.1545269563794136,
|
|
"num_tokens": 2873962.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 5.721522378921509,
|
|
"epoch": 1.301255230125523,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004971638509227611,
|
|
"loss": 5.6967,
|
|
"mean_token_accuracy": 0.16045853793621062,
|
|
"num_tokens": 2883530.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.705851316452026,
|
|
"epoch": 1.3054393305439331,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004971125361480236,
|
|
"loss": 5.6853,
|
|
"mean_token_accuracy": 0.15562771111726761,
|
|
"num_tokens": 2892867.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 5.719560289382935,
|
|
"epoch": 1.3096234309623431,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004970607643013969,
|
|
"loss": 5.643,
|
|
"mean_token_accuracy": 0.16159158051013947,
|
|
"num_tokens": 2902320.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.679278230667114,
|
|
"epoch": 1.3138075313807531,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000497008535489419,
|
|
"loss": 5.5838,
|
|
"mean_token_accuracy": 0.16410900056362152,
|
|
"num_tokens": 2911390.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.678142166137695,
|
|
"epoch": 1.3179916317991631,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004969558498195684,
|
|
"loss": 5.6136,
|
|
"mean_token_accuracy": 0.16564353704452514,
|
|
"num_tokens": 2921295.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 5.726165390014648,
|
|
"epoch": 1.3221757322175733,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004969027074002638,
|
|
"loss": 5.6419,
|
|
"mean_token_accuracy": 0.16103103309869765,
|
|
"num_tokens": 2931328.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.626484727859497,
|
|
"epoch": 1.3263598326359833,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968491083408636,
|
|
"loss": 5.5921,
|
|
"mean_token_accuracy": 0.1605369418859482,
|
|
"num_tokens": 2940193.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.644883728027343,
|
|
"epoch": 1.3305439330543933,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004967950527516663,
|
|
"loss": 5.6319,
|
|
"mean_token_accuracy": 0.16434539556503297,
|
|
"num_tokens": 2949759.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.6134930610656735,
|
|
"epoch": 1.3347280334728033,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004967405407439092,
|
|
"loss": 5.556,
|
|
"mean_token_accuracy": 0.15818172544240952,
|
|
"num_tokens": 2959185.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.77109956741333,
|
|
"epoch": 1.3389121338912133,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004966855724297698,
|
|
"loss": 5.7123,
|
|
"mean_token_accuracy": 0.15126804038882255,
|
|
"num_tokens": 2968945.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.717288303375244,
|
|
"epoch": 1.3430962343096233,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004966301479223637,
|
|
"loss": 5.6578,
|
|
"mean_token_accuracy": 0.15509079843759538,
|
|
"num_tokens": 2978395.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 5.650655460357666,
|
|
"epoch": 1.3472803347280335,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004965742673357456,
|
|
"loss": 5.5628,
|
|
"mean_token_accuracy": 0.1712281510233879,
|
|
"num_tokens": 2987030.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.700111675262451,
|
|
"epoch": 1.3514644351464435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965179307849088,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.16350831985473632,
|
|
"num_tokens": 2995736.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.741107034683227,
|
|
"epoch": 1.3556485355648535,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000496461138385785,
|
|
"loss": 5.722,
|
|
"mean_token_accuracy": 0.1551881417632103,
|
|
"num_tokens": 3005105.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.6597717761993405,
|
|
"epoch": 1.3598326359832635,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004964038902552438,
|
|
"loss": 5.6405,
|
|
"mean_token_accuracy": 0.15983598679304123,
|
|
"num_tokens": 3013857.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.71797742843628,
|
|
"epoch": 1.3640167364016738,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004963461865110926,
|
|
"loss": 5.6753,
|
|
"mean_token_accuracy": 0.16151416003704072,
|
|
"num_tokens": 3023972.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.615366840362549,
|
|
"epoch": 1.3682008368200838,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004962880272720764,
|
|
"loss": 5.6281,
|
|
"mean_token_accuracy": 0.16673413515090943,
|
|
"num_tokens": 3032705.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.748799800872803,
|
|
"epoch": 1.3723849372384938,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004962294126578775,
|
|
"loss": 5.5674,
|
|
"mean_token_accuracy": 0.16736229360103608,
|
|
"num_tokens": 3040900.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.635876178741455,
|
|
"epoch": 1.3765690376569037,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004961703427891154,
|
|
"loss": 5.6241,
|
|
"mean_token_accuracy": 0.1661652982234955,
|
|
"num_tokens": 3050477.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.716712760925293,
|
|
"epoch": 1.3807531380753137,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004961108177873466,
|
|
"loss": 5.6781,
|
|
"mean_token_accuracy": 0.1577182188630104,
|
|
"num_tokens": 3059743.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.63826847076416,
|
|
"epoch": 1.3849372384937237,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004960508377750638,
|
|
"loss": 5.4816,
|
|
"mean_token_accuracy": 0.1699427306652069,
|
|
"num_tokens": 3067801.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.599407529830932,
|
|
"epoch": 1.389121338912134,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004959904028756961,
|
|
"loss": 5.6768,
|
|
"mean_token_accuracy": 0.15741273611783982,
|
|
"num_tokens": 3077404.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.752321434020996,
|
|
"epoch": 1.393305439330544,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004959295132136091,
|
|
"loss": 5.7073,
|
|
"mean_token_accuracy": 0.1570734903216362,
|
|
"num_tokens": 3087130.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.7756424903869625,
|
|
"epoch": 1.397489539748954,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004958681689141036,
|
|
"loss": 5.7149,
|
|
"mean_token_accuracy": 0.15712554901838302,
|
|
"num_tokens": 3096038.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.627508974075317,
|
|
"epoch": 1.401673640167364,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004958063701034167,
|
|
"loss": 5.569,
|
|
"mean_token_accuracy": 0.1632245361804962,
|
|
"num_tokens": 3105668.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.595325088500976,
|
|
"epoch": 1.4058577405857742,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00049574411690872,
|
|
"loss": 5.5417,
|
|
"mean_token_accuracy": 0.15734163001179696,
|
|
"num_tokens": 3114020.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.577217531204224,
|
|
"epoch": 1.4100418410041842,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004956814094581208,
|
|
"loss": 5.5852,
|
|
"mean_token_accuracy": 0.1656106784939766,
|
|
"num_tokens": 3124096.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.7230309009552,
|
|
"epoch": 1.4142259414225942,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000495618247880661,
|
|
"loss": 5.6452,
|
|
"mean_token_accuracy": 0.15970809012651443,
|
|
"num_tokens": 3132806.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.648838520050049,
|
|
"epoch": 1.4184100418410042,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004955546323063168,
|
|
"loss": 5.574,
|
|
"mean_token_accuracy": 0.17397267371416092,
|
|
"num_tokens": 3142001.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.563747406005859,
|
|
"epoch": 1.4225941422594142,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004954905628659989,
|
|
"loss": 5.5495,
|
|
"mean_token_accuracy": 0.1571080982685089,
|
|
"num_tokens": 3151658.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.665132617950439,
|
|
"epoch": 1.4267782426778242,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004954260396915521,
|
|
"loss": 5.5774,
|
|
"mean_token_accuracy": 0.16344356238842012,
|
|
"num_tokens": 3159573.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.464680528640747,
|
|
"epoch": 1.4309623430962344,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004953610629157544,
|
|
"loss": 5.4537,
|
|
"mean_token_accuracy": 0.1730719119310379,
|
|
"num_tokens": 3168312.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.727207994461059,
|
|
"epoch": 1.4351464435146444,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004952956326723177,
|
|
"loss": 5.5523,
|
|
"mean_token_accuracy": 0.1717923477292061,
|
|
"num_tokens": 3176811.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.65036792755127,
|
|
"epoch": 1.4393305439330544,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000495229749095887,
|
|
"loss": 5.6479,
|
|
"mean_token_accuracy": 0.15595420598983764,
|
|
"num_tokens": 3186599.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.775017213821411,
|
|
"epoch": 1.4435146443514644,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00049516341232204,
|
|
"loss": 5.7534,
|
|
"mean_token_accuracy": 0.15901043713092805,
|
|
"num_tokens": 3195905.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.671367931365967,
|
|
"epoch": 1.4476987447698746,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000495096622487287,
|
|
"loss": 5.7076,
|
|
"mean_token_accuracy": 0.15824892222881318,
|
|
"num_tokens": 3205132.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.791279268264771,
|
|
"epoch": 1.4518828451882846,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000495029379729071,
|
|
"loss": 5.704,
|
|
"mean_token_accuracy": 0.1495993286371231,
|
|
"num_tokens": 3214362.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.713367319107055,
|
|
"epoch": 1.4560669456066946,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004949616841857666,
|
|
"loss": 5.7001,
|
|
"mean_token_accuracy": 0.15628902167081832,
|
|
"num_tokens": 3224248.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.627431154251099,
|
|
"epoch": 1.4602510460251046,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004948935359966806,
|
|
"loss": 5.57,
|
|
"mean_token_accuracy": 0.1658832088112831,
|
|
"num_tokens": 3233416.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.533376026153564,
|
|
"epoch": 1.4644351464435146,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004948249353020506,
|
|
"loss": 5.5031,
|
|
"mean_token_accuracy": 0.1715339496731758,
|
|
"num_tokens": 3242354.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.683704471588134,
|
|
"epoch": 1.4686192468619246,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004947558822430461,
|
|
"loss": 5.5964,
|
|
"mean_token_accuracy": 0.1610857516527176,
|
|
"num_tokens": 3251026.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.659394121170044,
|
|
"epoch": 1.4728033472803348,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004946863769617672,
|
|
"loss": 5.5468,
|
|
"mean_token_accuracy": 0.16529789119958876,
|
|
"num_tokens": 3259429.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.613833808898926,
|
|
"epoch": 1.4769874476987448,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004946164196012445,
|
|
"loss": 5.5968,
|
|
"mean_token_accuracy": 0.16423317044973373,
|
|
"num_tokens": 3269305.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.672654628753662,
|
|
"epoch": 1.4811715481171548,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004945460103054389,
|
|
"loss": 5.6816,
|
|
"mean_token_accuracy": 0.15662355720996857,
|
|
"num_tokens": 3278846.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.669299554824829,
|
|
"epoch": 1.4853556485355648,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004944751492192416,
|
|
"loss": 5.5884,
|
|
"mean_token_accuracy": 0.16536742597818374,
|
|
"num_tokens": 3287200.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.683395767211914,
|
|
"epoch": 1.489539748953975,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004944038364884734,
|
|
"loss": 5.6288,
|
|
"mean_token_accuracy": 0.16207243353128434,
|
|
"num_tokens": 3296651.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.641139030456543,
|
|
"epoch": 1.493723849372385,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004943320722598842,
|
|
"loss": 5.4855,
|
|
"mean_token_accuracy": 0.16841897070407869,
|
|
"num_tokens": 3304887.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.611622476577759,
|
|
"epoch": 1.497907949790795,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004942598566811532,
|
|
"loss": 5.5241,
|
|
"mean_token_accuracy": 0.16441086083650588,
|
|
"num_tokens": 3313925.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.564924716949463,
|
|
"epoch": 1.502092050209205,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004941871899008887,
|
|
"loss": 5.523,
|
|
"mean_token_accuracy": 0.1679730162024498,
|
|
"num_tokens": 3322943.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.561237525939942,
|
|
"epoch": 1.506276150627615,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000494114072068627,
|
|
"loss": 5.5396,
|
|
"mean_token_accuracy": 0.15930552929639816,
|
|
"num_tokens": 3332742.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.60180697441101,
|
|
"epoch": 1.510460251046025,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004940405033348332,
|
|
"loss": 5.6209,
|
|
"mean_token_accuracy": 0.16569100469350814,
|
|
"num_tokens": 3342228.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.643794870376587,
|
|
"epoch": 1.514644351464435,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004939664838508996,
|
|
"loss": 5.5844,
|
|
"mean_token_accuracy": 0.16332708448171615,
|
|
"num_tokens": 3351854.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.556795120239258,
|
|
"epoch": 1.5188284518828452,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004938920137691465,
|
|
"loss": 5.5729,
|
|
"mean_token_accuracy": 0.16669173836708068,
|
|
"num_tokens": 3360809.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.631477355957031,
|
|
"epoch": 1.5230125523012552,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004938170932428214,
|
|
"loss": 5.611,
|
|
"mean_token_accuracy": 0.1575515404343605,
|
|
"num_tokens": 3371033.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.687102317810059,
|
|
"epoch": 1.5271966527196654,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004937417224260984,
|
|
"loss": 5.4723,
|
|
"mean_token_accuracy": 0.16213055700063705,
|
|
"num_tokens": 3379943.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.527942228317261,
|
|
"epoch": 1.5313807531380754,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004936659014740791,
|
|
"loss": 5.6114,
|
|
"mean_token_accuracy": 0.16545302420854568,
|
|
"num_tokens": 3389316.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.69856481552124,
|
|
"epoch": 1.5355648535564854,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004935896305427904,
|
|
"loss": 5.6451,
|
|
"mean_token_accuracy": 0.1602216750383377,
|
|
"num_tokens": 3398914.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.756103801727295,
|
|
"epoch": 1.5397489539748954,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004935129097891857,
|
|
"loss": 5.6111,
|
|
"mean_token_accuracy": 0.16129681766033171,
|
|
"num_tokens": 3407967.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.65551176071167,
|
|
"epoch": 1.5439330543933054,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000493435739371144,
|
|
"loss": 5.6406,
|
|
"mean_token_accuracy": 0.1634376108646393,
|
|
"num_tokens": 3416530.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.584914255142212,
|
|
"epoch": 1.5481171548117154,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004933581194474695,
|
|
"loss": 5.5043,
|
|
"mean_token_accuracy": 0.16534085869789122,
|
|
"num_tokens": 3425762.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.579939603805542,
|
|
"epoch": 1.5523012552301254,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004932800501778918,
|
|
"loss": 5.5956,
|
|
"mean_token_accuracy": 0.16289042830467224,
|
|
"num_tokens": 3434799.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.686808443069458,
|
|
"epoch": 1.5564853556485354,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004932015317230646,
|
|
"loss": 5.5969,
|
|
"mean_token_accuracy": 0.1612599492073059,
|
|
"num_tokens": 3442986.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.627131843566895,
|
|
"epoch": 1.5606694560669456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004931225642445665,
|
|
"loss": 5.6385,
|
|
"mean_token_accuracy": 0.16176490783691405,
|
|
"num_tokens": 3452914.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.601911830902099,
|
|
"epoch": 1.5648535564853556,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004930431479048998,
|
|
"loss": 5.4773,
|
|
"mean_token_accuracy": 0.1727757453918457,
|
|
"num_tokens": 3462093.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.599895477294922,
|
|
"epoch": 1.5690376569037658,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004929632828674904,
|
|
"loss": 5.6701,
|
|
"mean_token_accuracy": 0.15993697494268416,
|
|
"num_tokens": 3471259.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.64299430847168,
|
|
"epoch": 1.5732217573221758,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004928829692966879,
|
|
"loss": 5.5967,
|
|
"mean_token_accuracy": 0.17002829760313035,
|
|
"num_tokens": 3480071.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.7399860382080075,
|
|
"epoch": 1.5774058577405858,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004928022073577646,
|
|
"loss": 5.6332,
|
|
"mean_token_accuracy": 0.16624970436096193,
|
|
"num_tokens": 3489075.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.606470966339112,
|
|
"epoch": 1.5815899581589958,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004927209972169157,
|
|
"loss": 5.4697,
|
|
"mean_token_accuracy": 0.1706315368413925,
|
|
"num_tokens": 3498495.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.556340599060059,
|
|
"epoch": 1.5857740585774058,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004926393390412586,
|
|
"loss": 5.6112,
|
|
"mean_token_accuracy": 0.16138895153999328,
|
|
"num_tokens": 3508590.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.680856943130493,
|
|
"epoch": 1.5899581589958158,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004925572329988325,
|
|
"loss": 5.6171,
|
|
"mean_token_accuracy": 0.15914507061243058,
|
|
"num_tokens": 3518397.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.604571723937989,
|
|
"epoch": 1.5941422594142258,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004924746792585987,
|
|
"loss": 5.6019,
|
|
"mean_token_accuracy": 0.1657917842268944,
|
|
"num_tokens": 3526620.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.760931587219238,
|
|
"epoch": 1.5983263598326358,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004923916779904394,
|
|
"loss": 5.6215,
|
|
"mean_token_accuracy": 0.16468919068574905,
|
|
"num_tokens": 3535534.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.53223729133606,
|
|
"epoch": 1.602510460251046,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004923082293651578,
|
|
"loss": 5.568,
|
|
"mean_token_accuracy": 0.16554071754217148,
|
|
"num_tokens": 3544771.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.640694427490234,
|
|
"epoch": 1.606694560669456,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000492224333554478,
|
|
"loss": 5.5337,
|
|
"mean_token_accuracy": 0.15835429430007936,
|
|
"num_tokens": 3553970.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.546091079711914,
|
|
"epoch": 1.6108786610878663,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004921399907310437,
|
|
"loss": 5.4523,
|
|
"mean_token_accuracy": 0.1624826580286026,
|
|
"num_tokens": 3563116.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.54065899848938,
|
|
"epoch": 1.6150627615062763,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004920552010684191,
|
|
"loss": 5.5843,
|
|
"mean_token_accuracy": 0.1632082238793373,
|
|
"num_tokens": 3571605.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.579107475280762,
|
|
"epoch": 1.6192468619246863,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004919699647410878,
|
|
"loss": 5.6733,
|
|
"mean_token_accuracy": 0.15940951108932494,
|
|
"num_tokens": 3581837.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.651897621154785,
|
|
"epoch": 1.6234309623430963,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004918842819244523,
|
|
"loss": 5.6108,
|
|
"mean_token_accuracy": 0.15646165311336518,
|
|
"num_tokens": 3590853.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.609458684921265,
|
|
"epoch": 1.6276150627615062,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004917981527948342,
|
|
"loss": 5.5049,
|
|
"mean_token_accuracy": 0.16811413764953614,
|
|
"num_tokens": 3599708.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.549583625793457,
|
|
"epoch": 1.6317991631799162,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004917115775294731,
|
|
"loss": 5.5331,
|
|
"mean_token_accuracy": 0.1643170267343521,
|
|
"num_tokens": 3608634.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.594006633758545,
|
|
"epoch": 1.6359832635983262,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004916245563065273,
|
|
"loss": 5.5907,
|
|
"mean_token_accuracy": 0.1617286831140518,
|
|
"num_tokens": 3618776.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.669147920608521,
|
|
"epoch": 1.6401673640167362,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004915370893050722,
|
|
"loss": 5.6027,
|
|
"mean_token_accuracy": 0.1599277600646019,
|
|
"num_tokens": 3628595.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.733761310577393,
|
|
"epoch": 1.6443514644351465,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004914491767051012,
|
|
"loss": 5.7264,
|
|
"mean_token_accuracy": 0.15489014610648155,
|
|
"num_tokens": 3638133.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.570251989364624,
|
|
"epoch": 1.6485355648535565,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004913608186875241,
|
|
"loss": 5.5056,
|
|
"mean_token_accuracy": 0.1667278602719307,
|
|
"num_tokens": 3646679.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.62856240272522,
|
|
"epoch": 1.6527196652719667,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004912720154341674,
|
|
"loss": 5.6283,
|
|
"mean_token_accuracy": 0.15991436094045638,
|
|
"num_tokens": 3656131.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.649011325836182,
|
|
"epoch": 1.6569037656903767,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004911827671277739,
|
|
"loss": 5.531,
|
|
"mean_token_accuracy": 0.16518089473247527,
|
|
"num_tokens": 3665187.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.646500158309936,
|
|
"epoch": 1.6610878661087867,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004910930739520025,
|
|
"loss": 5.5849,
|
|
"mean_token_accuracy": 0.1645905703306198,
|
|
"num_tokens": 3674649.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.67808575630188,
|
|
"epoch": 1.6652719665271967,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004910029360914271,
|
|
"loss": 5.6256,
|
|
"mean_token_accuracy": 0.15767135918140412,
|
|
"num_tokens": 3684680.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.539081144332886,
|
|
"epoch": 1.6694560669456067,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000490912353731537,
|
|
"loss": 5.5464,
|
|
"mean_token_accuracy": 0.1714841827750206,
|
|
"num_tokens": 3693560.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.563750696182251,
|
|
"epoch": 1.6736401673640167,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004908213270587363,
|
|
"loss": 5.5372,
|
|
"mean_token_accuracy": 0.16540804207324983,
|
|
"num_tokens": 3702632.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.6736401673640167,
|
|
"eval_entropy": 5.491124003942759,
|
|
"eval_loss": 5.678599834442139,
|
|
"eval_mean_token_accuracy": 0.16818918819503062,
|
|
"eval_num_tokens": 3702632.0,
|
|
"eval_runtime": 2.7304,
|
|
"eval_samples_per_second": 1430.95,
|
|
"eval_steps_per_second": 179.098,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.572523546218872,
|
|
"epoch": 1.6778242677824267,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004907298562603431,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.1779036343097687,
|
|
"num_tokens": 3711022.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.51769700050354,
|
|
"epoch": 1.6820083682008367,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004906379415245896,
|
|
"loss": 5.5388,
|
|
"mean_token_accuracy": 0.16877284198999404,
|
|
"num_tokens": 3719778.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.562564992904663,
|
|
"epoch": 1.6861924686192469,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004905455830406215,
|
|
"loss": 5.5346,
|
|
"mean_token_accuracy": 0.1705291375517845,
|
|
"num_tokens": 3728773.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.590969371795654,
|
|
"epoch": 1.6903765690376569,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004904527809984978,
|
|
"loss": 5.58,
|
|
"mean_token_accuracy": 0.1744068220257759,
|
|
"num_tokens": 3737781.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.690853548049927,
|
|
"epoch": 1.694560669456067,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004903595355891902,
|
|
"loss": 5.6152,
|
|
"mean_token_accuracy": 0.1568044126033783,
|
|
"num_tokens": 3748010.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.597555637359619,
|
|
"epoch": 1.698744769874477,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004902658470045826,
|
|
"loss": 5.6003,
|
|
"mean_token_accuracy": 0.16391026079654694,
|
|
"num_tokens": 3757894.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.673963737487793,
|
|
"epoch": 1.702928870292887,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004901717154374713,
|
|
"loss": 5.6991,
|
|
"mean_token_accuracy": 0.16143399029970168,
|
|
"num_tokens": 3768760.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.624513435363769,
|
|
"epoch": 1.707112970711297,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004900771410815638,
|
|
"loss": 5.5315,
|
|
"mean_token_accuracy": 0.16945276856422425,
|
|
"num_tokens": 3778273.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.517771816253662,
|
|
"epoch": 1.711297071129707,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004899821241314788,
|
|
"loss": 5.5358,
|
|
"mean_token_accuracy": 0.1683587208390236,
|
|
"num_tokens": 3787012.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.567167663574219,
|
|
"epoch": 1.715481171548117,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000489886664782746,
|
|
"loss": 5.486,
|
|
"mean_token_accuracy": 0.1716827154159546,
|
|
"num_tokens": 3795144.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.686856889724732,
|
|
"epoch": 1.719665271966527,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004897907632318055,
|
|
"loss": 5.7107,
|
|
"mean_token_accuracy": 0.15753279775381088,
|
|
"num_tokens": 3804534.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.55545563697815,
|
|
"epoch": 1.723849372384937,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004896944196760071,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.16995557993650437,
|
|
"num_tokens": 3813757.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.505803346633911,
|
|
"epoch": 1.7280334728033473,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895976343136106,
|
|
"loss": 5.5339,
|
|
"mean_token_accuracy": 0.17111619263887407,
|
|
"num_tokens": 3823157.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.46083288192749,
|
|
"epoch": 1.7322175732217573,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004895004073437844,
|
|
"loss": 5.4172,
|
|
"mean_token_accuracy": 0.17137759774923325,
|
|
"num_tokens": 3831661.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.466241359710693,
|
|
"epoch": 1.7364016736401675,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004894027389666061,
|
|
"loss": 5.4048,
|
|
"mean_token_accuracy": 0.16757647842168807,
|
|
"num_tokens": 3840591.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.573554372787475,
|
|
"epoch": 1.7405857740585775,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004893046293830616,
|
|
"loss": 5.5502,
|
|
"mean_token_accuracy": 0.16351725161075592,
|
|
"num_tokens": 3850094.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.571382141113281,
|
|
"epoch": 1.7447698744769875,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004892060787950447,
|
|
"loss": 5.5728,
|
|
"mean_token_accuracy": 0.16660072952508925,
|
|
"num_tokens": 3858973.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.496927070617676,
|
|
"epoch": 1.7489539748953975,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004891070874053564,
|
|
"loss": 5.3678,
|
|
"mean_token_accuracy": 0.17261970937252044,
|
|
"num_tokens": 3868315.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.528550434112549,
|
|
"epoch": 1.7531380753138075,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004890076554177051,
|
|
"loss": 5.5204,
|
|
"mean_token_accuracy": 0.16873810589313507,
|
|
"num_tokens": 3877605.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.6158366203308105,
|
|
"epoch": 1.7573221757322175,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004889077830367061,
|
|
"loss": 5.5403,
|
|
"mean_token_accuracy": 0.16531874537467955,
|
|
"num_tokens": 3886998.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.541978931427002,
|
|
"epoch": 1.7615062761506275,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004888074704678806,
|
|
"loss": 5.4911,
|
|
"mean_token_accuracy": 0.1717954158782959,
|
|
"num_tokens": 3896018.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.581347227096558,
|
|
"epoch": 1.7656903765690377,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004887067179176558,
|
|
"loss": 5.483,
|
|
"mean_token_accuracy": 0.16950565576553345,
|
|
"num_tokens": 3904957.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.531962156295776,
|
|
"epoch": 1.7698744769874477,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004886055255933639,
|
|
"loss": 5.5971,
|
|
"mean_token_accuracy": 0.16620343998074533,
|
|
"num_tokens": 3913427.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.583233118057251,
|
|
"epoch": 1.7740585774058577,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004885038937032429,
|
|
"loss": 5.5093,
|
|
"mean_token_accuracy": 0.16479990780353546,
|
|
"num_tokens": 3923134.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.555366945266724,
|
|
"epoch": 1.778242677824268,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004884018224564345,
|
|
"loss": 5.4355,
|
|
"mean_token_accuracy": 0.16853926777839662,
|
|
"num_tokens": 3931867.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.533804702758789,
|
|
"epoch": 1.782426778242678,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004882993120629852,
|
|
"loss": 5.5219,
|
|
"mean_token_accuracy": 0.1705673635005951,
|
|
"num_tokens": 3941057.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.687018394470215,
|
|
"epoch": 1.786610878661088,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004881963627338449,
|
|
"loss": 5.6362,
|
|
"mean_token_accuracy": 0.15854295045137407,
|
|
"num_tokens": 3950511.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.586686897277832,
|
|
"epoch": 1.790794979079498,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048809297468086654,
|
|
"loss": 5.5561,
|
|
"mean_token_accuracy": 0.17459627687931062,
|
|
"num_tokens": 3959769.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.593465280532837,
|
|
"epoch": 1.794979079497908,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004879891481168063,
|
|
"loss": 5.4756,
|
|
"mean_token_accuracy": 0.17072359174489976,
|
|
"num_tokens": 3968462.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.463370895385742,
|
|
"epoch": 1.799163179916318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048788488325532235,
|
|
"loss": 5.5393,
|
|
"mean_token_accuracy": 0.1667232632637024,
|
|
"num_tokens": 3977464.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.515686464309693,
|
|
"epoch": 1.803347280334728,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004877801803109753,
|
|
"loss": 5.4495,
|
|
"mean_token_accuracy": 0.17411705553531648,
|
|
"num_tokens": 3986303.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.621507215499878,
|
|
"epoch": 1.8075313807531381,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048767503949922683,
|
|
"loss": 5.4763,
|
|
"mean_token_accuracy": 0.17094481438398362,
|
|
"num_tokens": 3994970.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.613283586502075,
|
|
"epoch": 1.8117154811715481,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004875694610364398,
|
|
"loss": 5.5803,
|
|
"mean_token_accuracy": 0.16584169268608093,
|
|
"num_tokens": 4004679.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.642657518386841,
|
|
"epoch": 1.8158995815899581,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048746344513987775,
|
|
"loss": 5.5997,
|
|
"mean_token_accuracy": 0.16097164154052734,
|
|
"num_tokens": 4013516.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.576339340209961,
|
|
"epoch": 1.8200836820083683,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004873569920277043,
|
|
"loss": 5.4456,
|
|
"mean_token_accuracy": 0.17589643746614456,
|
|
"num_tokens": 4022335.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.451902580261231,
|
|
"epoch": 1.8242677824267783,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048725010191898296,
|
|
"loss": 5.433,
|
|
"mean_token_accuracy": 0.17280954718589783,
|
|
"num_tokens": 4031295.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.596143627166748,
|
|
"epoch": 1.8284518828451883,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048714277503367627,
|
|
"loss": 5.6556,
|
|
"mean_token_accuracy": 0.16676488071680068,
|
|
"num_tokens": 4040378.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.647720813751221,
|
|
"epoch": 1.8326359832635983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048703501159264576,
|
|
"loss": 5.5182,
|
|
"mean_token_accuracy": 0.15616264939308167,
|
|
"num_tokens": 4050217.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.399335479736328,
|
|
"epoch": 1.8368200836820083,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004869268118176512,
|
|
"loss": 5.39,
|
|
"mean_token_accuracy": 0.16944845616817475,
|
|
"num_tokens": 4059088.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.6183709621429445,
|
|
"epoch": 1.8410041841004183,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048681817593135046,
|
|
"loss": 5.4924,
|
|
"mean_token_accuracy": 0.18004305362701417,
|
|
"num_tokens": 4067609.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.551707029342651,
|
|
"epoch": 1.8451882845188283,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048670910415729855,
|
|
"loss": 5.4916,
|
|
"mean_token_accuracy": 0.1724642962217331,
|
|
"num_tokens": 4076763.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.59165768623352,
|
|
"epoch": 1.8493723849372385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048659959671994786,
|
|
"loss": 5.6457,
|
|
"mean_token_accuracy": 0.16204374581575393,
|
|
"num_tokens": 4086552.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.455987501144409,
|
|
"epoch": 1.8535564853556485,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00048648965384464693,
|
|
"loss": 5.406,
|
|
"mean_token_accuracy": 0.1716286361217499,
|
|
"num_tokens": 4096200.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.506076335906982,
|
|
"epoch": 1.8577405857740585,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048637927575764057,
|
|
"loss": 5.4556,
|
|
"mean_token_accuracy": 0.17082349807024003,
|
|
"num_tokens": 4105080.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.517089176177978,
|
|
"epoch": 1.8619246861924688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048626846268606913,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.16985356509685517,
|
|
"num_tokens": 4113972.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.6436656475067135,
|
|
"epoch": 1.8661087866108788,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048615721485796803,
|
|
"loss": 5.6571,
|
|
"mean_token_accuracy": 0.16025995314121247,
|
|
"num_tokens": 4123615.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.553581619262696,
|
|
"epoch": 1.8702928870292888,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004860455325022676,
|
|
"loss": 5.4155,
|
|
"mean_token_accuracy": 0.1689600870013237,
|
|
"num_tokens": 4132128.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.622593832015991,
|
|
"epoch": 1.8744769874476988,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000485933415848792,
|
|
"loss": 5.6146,
|
|
"mean_token_accuracy": 0.16623764038085936,
|
|
"num_tokens": 4141575.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.566574192047119,
|
|
"epoch": 1.8786610878661087,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004858208651282593,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.1704362750053406,
|
|
"num_tokens": 4150372.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.508659744262696,
|
|
"epoch": 1.8828451882845187,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000485707880572281,
|
|
"loss": 5.5718,
|
|
"mean_token_accuracy": 0.17235517501831055,
|
|
"num_tokens": 4159007.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.592802858352661,
|
|
"epoch": 1.8870292887029287,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004855944624133608,
|
|
"loss": 5.6236,
|
|
"mean_token_accuracy": 0.16393871158361434,
|
|
"num_tokens": 4168646.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.489811658859253,
|
|
"epoch": 1.891213389121339,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048548061088489523,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.16455192416906356,
|
|
"num_tokens": 4177979.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.58298807144165,
|
|
"epoch": 1.895397489539749,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048536632622117246,
|
|
"loss": 5.4378,
|
|
"mean_token_accuracy": 0.17333355098962783,
|
|
"num_tokens": 4186781.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.503539419174194,
|
|
"epoch": 1.899581589958159,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00048525160865737186,
|
|
"loss": 5.5073,
|
|
"mean_token_accuracy": 0.16708964258432388,
|
|
"num_tokens": 4195661.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.515585327148438,
|
|
"epoch": 1.9037656903765692,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048513645842956384,
|
|
"loss": 5.4108,
|
|
"mean_token_accuracy": 0.17160611748695373,
|
|
"num_tokens": 4203902.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.459004068374634,
|
|
"epoch": 1.9079497907949792,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048502087577470895,
|
|
"loss": 5.4666,
|
|
"mean_token_accuracy": 0.16583277732133866,
|
|
"num_tokens": 4213663.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.629412889480591,
|
|
"epoch": 1.9121338912133892,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00048490486093065774,
|
|
"loss": 5.514,
|
|
"mean_token_accuracy": 0.1735001504421234,
|
|
"num_tokens": 4222152.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.527402353286743,
|
|
"epoch": 1.9163179916317992,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00048478841413615017,
|
|
"loss": 5.529,
|
|
"mean_token_accuracy": 0.17351837158203126,
|
|
"num_tokens": 4232347.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.60933895111084,
|
|
"epoch": 1.9205020920502092,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048467153563081487,
|
|
"loss": 5.6077,
|
|
"mean_token_accuracy": 0.1692125305533409,
|
|
"num_tokens": 4240607.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.599947690963745,
|
|
"epoch": 1.9246861924686192,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004845542256551692,
|
|
"loss": 5.6023,
|
|
"mean_token_accuracy": 0.16642989665269853,
|
|
"num_tokens": 4251131.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.583032417297363,
|
|
"epoch": 1.9288702928870292,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004844364844506181,
|
|
"loss": 5.5281,
|
|
"mean_token_accuracy": 0.16871487051248552,
|
|
"num_tokens": 4259955.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.600530576705933,
|
|
"epoch": 1.9330543933054394,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004843183122594541,
|
|
"loss": 5.5134,
|
|
"mean_token_accuracy": 0.16982486397027968,
|
|
"num_tokens": 4268686.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.535481405258179,
|
|
"epoch": 1.9372384937238494,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004841997093248566,
|
|
"loss": 5.4926,
|
|
"mean_token_accuracy": 0.17420293092727662,
|
|
"num_tokens": 4277377.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.471238851547241,
|
|
"epoch": 1.9414225941422594,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004840806758908913,
|
|
"loss": 5.4791,
|
|
"mean_token_accuracy": 0.16643480509519576,
|
|
"num_tokens": 4287037.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.6762055397033695,
|
|
"epoch": 1.9456066945606696,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048396121220250983,
|
|
"loss": 5.6222,
|
|
"mean_token_accuracy": 0.15888616740703582,
|
|
"num_tokens": 4296923.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.445138311386108,
|
|
"epoch": 1.9497907949790796,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004838413185055494,
|
|
"loss": 5.3837,
|
|
"mean_token_accuracy": 0.17829248905181885,
|
|
"num_tokens": 4305873.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.485724210739136,
|
|
"epoch": 1.9539748953974896,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004837209950467318,
|
|
"loss": 5.5408,
|
|
"mean_token_accuracy": 0.16909062415361403,
|
|
"num_tokens": 4314912.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.572770309448242,
|
|
"epoch": 1.9581589958158996,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004836002420736635,
|
|
"loss": 5.5479,
|
|
"mean_token_accuracy": 0.160999795794487,
|
|
"num_tokens": 4324606.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.4958367347717285,
|
|
"epoch": 1.9623430962343096,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004834790598348347,
|
|
"loss": 5.4023,
|
|
"mean_token_accuracy": 0.17136014699935914,
|
|
"num_tokens": 4333076.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.666383218765259,
|
|
"epoch": 1.9665271966527196,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004833574485796188,
|
|
"loss": 5.5494,
|
|
"mean_token_accuracy": 0.17031686305999755,
|
|
"num_tokens": 4341893.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.475465250015259,
|
|
"epoch": 1.9707112970711296,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048323540855827244,
|
|
"loss": 5.5148,
|
|
"mean_token_accuracy": 0.16724376380443573,
|
|
"num_tokens": 4352303.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.614174842834473,
|
|
"epoch": 1.9748953974895398,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004831129400219341,
|
|
"loss": 5.4649,
|
|
"mean_token_accuracy": 0.17016783058643342,
|
|
"num_tokens": 4361166.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.578010511398316,
|
|
"epoch": 1.9790794979079498,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048299004322262457,
|
|
"loss": 5.5273,
|
|
"mean_token_accuracy": 0.1644423395395279,
|
|
"num_tokens": 4369999.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.497027206420898,
|
|
"epoch": 1.9832635983263598,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048286671841324565,
|
|
"loss": 5.4836,
|
|
"mean_token_accuracy": 0.17087049186229705,
|
|
"num_tokens": 4379373.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.576445913314819,
|
|
"epoch": 1.98744769874477,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004827429658475799,
|
|
"loss": 5.4755,
|
|
"mean_token_accuracy": 0.1715611144900322,
|
|
"num_tokens": 4389035.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.533624887466431,
|
|
"epoch": 1.99163179916318,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00048261878578029026,
|
|
"loss": 5.6441,
|
|
"mean_token_accuracy": 0.1629743292927742,
|
|
"num_tokens": 4399366.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.575339126586914,
|
|
"epoch": 1.99581589958159,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004824941784669193,
|
|
"loss": 5.4744,
|
|
"mean_token_accuracy": 0.1669633612036705,
|
|
"num_tokens": 4408183.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.541588544845581,
|
|
"epoch": 2.0,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004823691441638889,
|
|
"loss": 5.5844,
|
|
"mean_token_accuracy": 0.15914448723196983,
|
|
"num_tokens": 4418644.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.465671730041504,
|
|
"epoch": 2.00418410041841,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004822436831284995,
|
|
"loss": 5.1718,
|
|
"mean_token_accuracy": 0.1804201766848564,
|
|
"num_tokens": 4427828.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.462167692184448,
|
|
"epoch": 2.00836820083682,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004821177956189297,
|
|
"loss": 5.0932,
|
|
"mean_token_accuracy": 0.18268589228391646,
|
|
"num_tokens": 4436194.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.3815234184265135,
|
|
"epoch": 2.01255230125523,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004819914818942359,
|
|
"loss": 5.1587,
|
|
"mean_token_accuracy": 0.18811518847942352,
|
|
"num_tokens": 4444468.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.480875778198242,
|
|
"epoch": 2.01673640167364,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048186474221435127,
|
|
"loss": 5.0801,
|
|
"mean_token_accuracy": 0.19296729266643525,
|
|
"num_tokens": 4452760.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.3785402297973635,
|
|
"epoch": 2.02092050209205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048173757684008583,
|
|
"loss": 5.1482,
|
|
"mean_token_accuracy": 0.1877833530306816,
|
|
"num_tokens": 4462154.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.477201223373413,
|
|
"epoch": 2.0251046025104604,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004816099860331255,
|
|
"loss": 5.2378,
|
|
"mean_token_accuracy": 0.18000938147306442,
|
|
"num_tokens": 4471987.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.446134471893311,
|
|
"epoch": 2.0292887029288704,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00048148197005603145,
|
|
"loss": 5.1628,
|
|
"mean_token_accuracy": 0.18375986963510513,
|
|
"num_tokens": 4481645.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.28934006690979,
|
|
"epoch": 2.0334728033472804,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048135352917224005,
|
|
"loss": 5.132,
|
|
"mean_token_accuracy": 0.19004039466381073,
|
|
"num_tokens": 4491166.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.52780499458313,
|
|
"epoch": 2.0376569037656904,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00048122466364606214,
|
|
"loss": 5.1386,
|
|
"mean_token_accuracy": 0.18404159545898438,
|
|
"num_tokens": 4500159.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.346362209320068,
|
|
"epoch": 2.0418410041841004,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048109537374268216,
|
|
"loss": 5.0872,
|
|
"mean_token_accuracy": 0.19677301347255707,
|
|
"num_tokens": 4509737.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.292774248123169,
|
|
"epoch": 2.0460251046025104,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048096565972815793,
|
|
"loss": 5.0221,
|
|
"mean_token_accuracy": 0.19164910465478896,
|
|
"num_tokens": 4518379.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.365337228775024,
|
|
"epoch": 2.0502092050209204,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004808355218694199,
|
|
"loss": 5.1611,
|
|
"mean_token_accuracy": 0.18420323133468627,
|
|
"num_tokens": 4527885.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.407716226577759,
|
|
"epoch": 2.0543933054393304,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00048070496043427116,
|
|
"loss": 5.1975,
|
|
"mean_token_accuracy": 0.1824670508503914,
|
|
"num_tokens": 4537894.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.380266380310059,
|
|
"epoch": 2.0585774058577404,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00048057397569138596,
|
|
"loss": 5.0945,
|
|
"mean_token_accuracy": 0.19576826244592666,
|
|
"num_tokens": 4546462.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.378341960906982,
|
|
"epoch": 2.062761506276151,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004804425679103099,
|
|
"loss": 5.1205,
|
|
"mean_token_accuracy": 0.18275196850299835,
|
|
"num_tokens": 4555970.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.402397394180298,
|
|
"epoch": 2.066945606694561,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004803107373614589,
|
|
"loss": 5.1896,
|
|
"mean_token_accuracy": 0.17988977134227752,
|
|
"num_tokens": 4565799.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.4651405811309814,
|
|
"epoch": 2.071129707112971,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004801784843161194,
|
|
"loss": 5.2097,
|
|
"mean_token_accuracy": 0.17294996231794357,
|
|
"num_tokens": 4575571.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.369615983963013,
|
|
"epoch": 2.075313807531381,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004800458090464465,
|
|
"loss": 5.1138,
|
|
"mean_token_accuracy": 0.18762540370225905,
|
|
"num_tokens": 4584263.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.3507678508758545,
|
|
"epoch": 2.079497907949791,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047991271182546476,
|
|
"loss": 5.0984,
|
|
"mean_token_accuracy": 0.1907823845744133,
|
|
"num_tokens": 4592979.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.3727912425994875,
|
|
"epoch": 2.083682008368201,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047977919292706696,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.1879875510931015,
|
|
"num_tokens": 4602042.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.352688646316528,
|
|
"epoch": 2.087866108786611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004796452526260134,
|
|
"loss": 5.1756,
|
|
"mean_token_accuracy": 0.18628377914428712,
|
|
"num_tokens": 4611197.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.2905971050262455,
|
|
"epoch": 2.092050209205021,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047951089119793164,
|
|
"loss": 5.1278,
|
|
"mean_token_accuracy": 0.1851946383714676,
|
|
"num_tokens": 4621612.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.092050209205021,
|
|
"eval_entropy": 5.239512876498918,
|
|
"eval_loss": 5.599039077758789,
|
|
"eval_mean_token_accuracy": 0.17449360253986407,
|
|
"eval_num_tokens": 4621612.0,
|
|
"eval_runtime": 2.7292,
|
|
"eval_samples_per_second": 1431.545,
|
|
"eval_steps_per_second": 179.172,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.28721981048584,
|
|
"epoch": 2.096234309623431,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000479376108919316,
|
|
"loss": 5.0147,
|
|
"mean_token_accuracy": 0.19302805066108703,
|
|
"num_tokens": 4630198.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.339468908309937,
|
|
"epoch": 2.100418410041841,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004792409060675267,
|
|
"loss": 5.1156,
|
|
"mean_token_accuracy": 0.18671335577964782,
|
|
"num_tokens": 4638729.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.3387229442596436,
|
|
"epoch": 2.104602510460251,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004791052829207895,
|
|
"loss": 5.2312,
|
|
"mean_token_accuracy": 0.1840809628367424,
|
|
"num_tokens": 4648759.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.413992691040039,
|
|
"epoch": 2.1087866108786613,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000478969239758195,
|
|
"loss": 5.175,
|
|
"mean_token_accuracy": 0.1829567328095436,
|
|
"num_tokens": 4657744.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.335267686843872,
|
|
"epoch": 2.1129707112970713,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047883277685969815,
|
|
"loss": 5.1367,
|
|
"mean_token_accuracy": 0.1837652623653412,
|
|
"num_tokens": 4666154.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.357482194900513,
|
|
"epoch": 2.1171548117154813,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00047869589450611773,
|
|
"loss": 5.1786,
|
|
"mean_token_accuracy": 0.18171548396348952,
|
|
"num_tokens": 4675677.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.311690902709961,
|
|
"epoch": 2.1213389121338913,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004785585929791355,
|
|
"loss": 5.1512,
|
|
"mean_token_accuracy": 0.18145935833454133,
|
|
"num_tokens": 4684733.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.415076541900635,
|
|
"epoch": 2.1255230125523012,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000478420872561296,
|
|
"loss": 5.0715,
|
|
"mean_token_accuracy": 0.18859368413686753,
|
|
"num_tokens": 4694155.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.291185712814331,
|
|
"epoch": 2.1297071129707112,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047828273353600577,
|
|
"loss": 5.1058,
|
|
"mean_token_accuracy": 0.19461557418107986,
|
|
"num_tokens": 4703286.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.355195236206055,
|
|
"epoch": 2.1338912133891212,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047814417618753256,
|
|
"loss": 5.1218,
|
|
"mean_token_accuracy": 0.18194112032651902,
|
|
"num_tokens": 4711666.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.325894451141357,
|
|
"epoch": 2.1380753138075312,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047800520080100526,
|
|
"loss": 5.1644,
|
|
"mean_token_accuracy": 0.1786948025226593,
|
|
"num_tokens": 4721644.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.337622499465942,
|
|
"epoch": 2.1422594142259412,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047786580766241265,
|
|
"loss": 5.1024,
|
|
"mean_token_accuracy": 0.19383088648319244,
|
|
"num_tokens": 4730885.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.25313811302185,
|
|
"epoch": 2.1464435146443517,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004777259970586036,
|
|
"loss": 5.0937,
|
|
"mean_token_accuracy": 0.1866190418601036,
|
|
"num_tokens": 4740871.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.308656549453735,
|
|
"epoch": 2.1506276150627617,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004775857692772859,
|
|
"loss": 5.067,
|
|
"mean_token_accuracy": 0.18714566230773927,
|
|
"num_tokens": 4750603.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.346760892868042,
|
|
"epoch": 2.1548117154811717,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004774451246070257,
|
|
"loss": 5.1993,
|
|
"mean_token_accuracy": 0.1845145434141159,
|
|
"num_tokens": 4760681.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.39498872756958,
|
|
"epoch": 2.1589958158995817,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004773040633372471,
|
|
"loss": 5.1711,
|
|
"mean_token_accuracy": 0.19251902550458908,
|
|
"num_tokens": 4770237.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.339705848693848,
|
|
"epoch": 2.1631799163179917,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047716258575823147,
|
|
"loss": 5.1857,
|
|
"mean_token_accuracy": 0.18849558085203172,
|
|
"num_tokens": 4779022.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.370788526535034,
|
|
"epoch": 2.1673640167364017,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000477020692161117,
|
|
"loss": 5.2216,
|
|
"mean_token_accuracy": 0.18457528799772263,
|
|
"num_tokens": 4788644.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.320277547836303,
|
|
"epoch": 2.1715481171548117,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004768783828378978,
|
|
"loss": 5.2146,
|
|
"mean_token_accuracy": 0.1754754438996315,
|
|
"num_tokens": 4798628.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.5206005573272705,
|
|
"epoch": 2.1757322175732217,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004767356580814237,
|
|
"loss": 5.315,
|
|
"mean_token_accuracy": 0.17956572473049165,
|
|
"num_tokens": 4807727.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.378936100006103,
|
|
"epoch": 2.1799163179916317,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004765925181853993,
|
|
"loss": 5.2003,
|
|
"mean_token_accuracy": 0.1807018294930458,
|
|
"num_tokens": 4817896.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.424268579483032,
|
|
"epoch": 2.1841004184100417,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004764489634443834,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.18551215380430222,
|
|
"num_tokens": 4827135.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.374562454223633,
|
|
"epoch": 2.1882845188284517,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047630499415378863,
|
|
"loss": 5.2499,
|
|
"mean_token_accuracy": 0.17581205815076828,
|
|
"num_tokens": 4837554.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.28589940071106,
|
|
"epoch": 2.192468619246862,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004761606106098806,
|
|
"loss": 5.1101,
|
|
"mean_token_accuracy": 0.18689924031496047,
|
|
"num_tokens": 4846739.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.334984683990479,
|
|
"epoch": 2.196652719665272,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047601581310977747,
|
|
"loss": 5.1643,
|
|
"mean_token_accuracy": 0.18399628549814223,
|
|
"num_tokens": 4855707.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.362114810943604,
|
|
"epoch": 2.200836820083682,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004758706019514493,
|
|
"loss": 5.1704,
|
|
"mean_token_accuracy": 0.18662786334753037,
|
|
"num_tokens": 4865098.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.416744756698608,
|
|
"epoch": 2.205020920502092,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004757249774337172,
|
|
"loss": 5.1806,
|
|
"mean_token_accuracy": 0.18706023395061494,
|
|
"num_tokens": 4874426.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.33036322593689,
|
|
"epoch": 2.209205020920502,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047557893985625293,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.18973471969366074,
|
|
"num_tokens": 4883240.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.298898506164551,
|
|
"epoch": 2.213389121338912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004754324895195786,
|
|
"loss": 5.1355,
|
|
"mean_token_accuracy": 0.184861022233963,
|
|
"num_tokens": 4891989.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.392232990264892,
|
|
"epoch": 2.217573221757322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004752856267250652,
|
|
"loss": 5.2462,
|
|
"mean_token_accuracy": 0.18160006254911423,
|
|
"num_tokens": 4901546.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.27634105682373,
|
|
"epoch": 2.221757322175732,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047513835177493286,
|
|
"loss": 5.0503,
|
|
"mean_token_accuracy": 0.19290486574172974,
|
|
"num_tokens": 4910222.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.338057374954223,
|
|
"epoch": 2.225941422594142,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004749906649722498,
|
|
"loss": 5.2734,
|
|
"mean_token_accuracy": 0.1754023849964142,
|
|
"num_tokens": 4920184.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.364781236648559,
|
|
"epoch": 2.2301255230125525,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004748425666209316,
|
|
"loss": 5.1815,
|
|
"mean_token_accuracy": 0.18442230075597763,
|
|
"num_tokens": 4930014.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.353585243225098,
|
|
"epoch": 2.2343096234309625,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004746940570257409,
|
|
"loss": 5.0934,
|
|
"mean_token_accuracy": 0.18913581520318984,
|
|
"num_tokens": 4939556.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.264898443222046,
|
|
"epoch": 2.2384937238493725,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004745451364922867,
|
|
"loss": 5.157,
|
|
"mean_token_accuracy": 0.18276126384735109,
|
|
"num_tokens": 4949066.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.400563907623291,
|
|
"epoch": 2.2426778242677825,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004743958053270233,
|
|
"loss": 5.1582,
|
|
"mean_token_accuracy": 0.1834804370999336,
|
|
"num_tokens": 4957881.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.3733145236969,
|
|
"epoch": 2.2468619246861925,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00047424606383725037,
|
|
"loss": 5.204,
|
|
"mean_token_accuracy": 0.1812547981739044,
|
|
"num_tokens": 4967105.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.277979040145874,
|
|
"epoch": 2.2510460251046025,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004740959123311118,
|
|
"loss": 5.1127,
|
|
"mean_token_accuracy": 0.19214510917663574,
|
|
"num_tokens": 4977676.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.2787762641906735,
|
|
"epoch": 2.2552301255230125,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047394535111759526,
|
|
"loss": 5.1306,
|
|
"mean_token_accuracy": 0.19049037843942643,
|
|
"num_tokens": 4986163.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.378134489059448,
|
|
"epoch": 2.2594142259414225,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004737943805065314,
|
|
"loss": 5.1316,
|
|
"mean_token_accuracy": 0.1849228858947754,
|
|
"num_tokens": 4995179.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.347646808624267,
|
|
"epoch": 2.2635983263598325,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047364300080859363,
|
|
"loss": 5.2238,
|
|
"mean_token_accuracy": 0.17981732934713363,
|
|
"num_tokens": 5003856.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.3950119972229,
|
|
"epoch": 2.2677824267782425,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047349121233529695,
|
|
"loss": 5.2562,
|
|
"mean_token_accuracy": 0.17975369840860367,
|
|
"num_tokens": 5013506.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.358010244369507,
|
|
"epoch": 2.2719665271966525,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004733390153989976,
|
|
"loss": 5.1859,
|
|
"mean_token_accuracy": 0.1891730934381485,
|
|
"num_tokens": 5022159.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.342816638946533,
|
|
"epoch": 2.276150627615063,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047318641031289245,
|
|
"loss": 5.1398,
|
|
"mean_token_accuracy": 0.18648490756750108,
|
|
"num_tokens": 5031869.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.201842546463013,
|
|
"epoch": 2.280334728033473,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004730333973910183,
|
|
"loss": 5.0041,
|
|
"mean_token_accuracy": 0.1997317850589752,
|
|
"num_tokens": 5039904.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.368862152099609,
|
|
"epoch": 2.284518828451883,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004728799769482511,
|
|
"loss": 5.2992,
|
|
"mean_token_accuracy": 0.17865955531597139,
|
|
"num_tokens": 5049144.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.295654439926148,
|
|
"epoch": 2.288702928870293,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047272614930030533,
|
|
"loss": 5.0556,
|
|
"mean_token_accuracy": 0.1902414456009865,
|
|
"num_tokens": 5058302.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.243958425521851,
|
|
"epoch": 2.292887029288703,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047257191476373375,
|
|
"loss": 5.0985,
|
|
"mean_token_accuracy": 0.18769923150539397,
|
|
"num_tokens": 5067443.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.298603916168213,
|
|
"epoch": 2.297071129707113,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00047241727365592603,
|
|
"loss": 5.2392,
|
|
"mean_token_accuracy": 0.17964693009853364,
|
|
"num_tokens": 5076415.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.382549667358399,
|
|
"epoch": 2.301255230125523,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004722622262951089,
|
|
"loss": 5.1812,
|
|
"mean_token_accuracy": 0.18127255886793137,
|
|
"num_tokens": 5085882.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.394537258148193,
|
|
"epoch": 2.305439330543933,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004721067730003448,
|
|
"loss": 5.1455,
|
|
"mean_token_accuracy": 0.19172731935977935,
|
|
"num_tokens": 5094685.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.251836252212525,
|
|
"epoch": 2.309623430962343,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004719509140915317,
|
|
"loss": 5.1642,
|
|
"mean_token_accuracy": 0.18699409663677216,
|
|
"num_tokens": 5103885.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.316407489776611,
|
|
"epoch": 2.3138075313807533,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047179464988940205,
|
|
"loss": 5.0901,
|
|
"mean_token_accuracy": 0.19509715288877488,
|
|
"num_tokens": 5112164.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.253286981582642,
|
|
"epoch": 2.3179916317991633,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047163798071552253,
|
|
"loss": 5.0557,
|
|
"mean_token_accuracy": 0.1919330596923828,
|
|
"num_tokens": 5121913.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.354563283920288,
|
|
"epoch": 2.3221757322175733,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004714809068922931,
|
|
"loss": 5.064,
|
|
"mean_token_accuracy": 0.19694381207227707,
|
|
"num_tokens": 5130447.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.2508293151855465,
|
|
"epoch": 2.3263598326359833,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004713234287429465,
|
|
"loss": 5.1883,
|
|
"mean_token_accuracy": 0.1843970760703087,
|
|
"num_tokens": 5139863.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.266225385665893,
|
|
"epoch": 2.3305439330543933,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004711655465915473,
|
|
"loss": 5.1607,
|
|
"mean_token_accuracy": 0.18224266469478606,
|
|
"num_tokens": 5149945.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.367474031448364,
|
|
"epoch": 2.3347280334728033,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004710072607629917,
|
|
"loss": 5.0995,
|
|
"mean_token_accuracy": 0.18524859994649887,
|
|
"num_tokens": 5160293.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.230232858657837,
|
|
"epoch": 2.3389121338912133,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047084857158300646,
|
|
"loss": 5.1369,
|
|
"mean_token_accuracy": 0.17988671660423278,
|
|
"num_tokens": 5169896.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.3255023002624515,
|
|
"epoch": 2.3430962343096233,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047068947937814844,
|
|
"loss": 5.1708,
|
|
"mean_token_accuracy": 0.1853892683982849,
|
|
"num_tokens": 5178898.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.364515209197998,
|
|
"epoch": 2.3472803347280333,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00047052998447580374,
|
|
"loss": 5.241,
|
|
"mean_token_accuracy": 0.18364589363336564,
|
|
"num_tokens": 5188733.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.256385135650635,
|
|
"epoch": 2.3514644351464433,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047037008720418724,
|
|
"loss": 5.1216,
|
|
"mean_token_accuracy": 0.1908215194940567,
|
|
"num_tokens": 5198117.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.366924047470093,
|
|
"epoch": 2.3556485355648533,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004702097878923419,
|
|
"loss": 5.176,
|
|
"mean_token_accuracy": 0.18560814559459687,
|
|
"num_tokens": 5206791.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.248342752456665,
|
|
"epoch": 2.3598326359832638,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004700490868701377,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.19201775193214415,
|
|
"num_tokens": 5216427.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.246069145202637,
|
|
"epoch": 2.3640167364016738,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004698879844682718,
|
|
"loss": 5.0657,
|
|
"mean_token_accuracy": 0.19573935866355896,
|
|
"num_tokens": 5225983.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.283814764022827,
|
|
"epoch": 2.3682008368200838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046972648101826695,
|
|
"loss": 5.1763,
|
|
"mean_token_accuracy": 0.19329917281866074,
|
|
"num_tokens": 5234547.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.360036993026734,
|
|
"epoch": 2.3723849372384938,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004695645768524712,
|
|
"loss": 5.1996,
|
|
"mean_token_accuracy": 0.1927380457520485,
|
|
"num_tokens": 5243281.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.32995982170105,
|
|
"epoch": 2.3765690376569037,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004694022723040573,
|
|
"loss": 5.0958,
|
|
"mean_token_accuracy": 0.1876293659210205,
|
|
"num_tokens": 5252318.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.3196028709411625,
|
|
"epoch": 2.3807531380753137,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046923956770702206,
|
|
"loss": 5.1844,
|
|
"mean_token_accuracy": 0.18977564722299575,
|
|
"num_tokens": 5261781.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.385820102691651,
|
|
"epoch": 2.3849372384937237,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004690764633961852,
|
|
"loss": 5.0845,
|
|
"mean_token_accuracy": 0.1870831221342087,
|
|
"num_tokens": 5270409.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.194503498077393,
|
|
"epoch": 2.3891213389121337,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004689129597071893,
|
|
"loss": 5.0864,
|
|
"mean_token_accuracy": 0.20467019975185394,
|
|
"num_tokens": 5278766.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.3033617496490475,
|
|
"epoch": 2.393305439330544,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046874905697649857,
|
|
"loss": 5.2798,
|
|
"mean_token_accuracy": 0.1811196655035019,
|
|
"num_tokens": 5287572.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.376643705368042,
|
|
"epoch": 2.397489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046858475554139857,
|
|
"loss": 5.1774,
|
|
"mean_token_accuracy": 0.19020477384328843,
|
|
"num_tokens": 5297348.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.30971393585205,
|
|
"epoch": 2.401673640167364,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004684200557399952,
|
|
"loss": 5.17,
|
|
"mean_token_accuracy": 0.18733545392751694,
|
|
"num_tokens": 5305748.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.253690767288208,
|
|
"epoch": 2.405857740585774,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046825495791121414,
|
|
"loss": 5.1127,
|
|
"mean_token_accuracy": 0.19702039659023285,
|
|
"num_tokens": 5314482.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.342981004714966,
|
|
"epoch": 2.410041841004184,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046808946239480017,
|
|
"loss": 5.2155,
|
|
"mean_token_accuracy": 0.1836487054824829,
|
|
"num_tokens": 5325161.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.290003299713135,
|
|
"epoch": 2.414225941422594,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004679235695313165,
|
|
"loss": 5.1972,
|
|
"mean_token_accuracy": 0.18110067993402482,
|
|
"num_tokens": 5335262.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.294895935058594,
|
|
"epoch": 2.418410041841004,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00046775727966214393,
|
|
"loss": 5.1094,
|
|
"mean_token_accuracy": 0.1824950486421585,
|
|
"num_tokens": 5345145.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.224514198303223,
|
|
"epoch": 2.422594142259414,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004675905931294803,
|
|
"loss": 5.094,
|
|
"mean_token_accuracy": 0.1889493152499199,
|
|
"num_tokens": 5354715.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.274765205383301,
|
|
"epoch": 2.426778242677824,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004674235102763398,
|
|
"loss": 5.1451,
|
|
"mean_token_accuracy": 0.19259584397077562,
|
|
"num_tokens": 5364180.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.251356887817383,
|
|
"epoch": 2.430962343096234,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00046725603144655195,
|
|
"loss": 5.1061,
|
|
"mean_token_accuracy": 0.19033802896738053,
|
|
"num_tokens": 5372919.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.2445591449737545,
|
|
"epoch": 2.435146443514644,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004670881569847612,
|
|
"loss": 5.0977,
|
|
"mean_token_accuracy": 0.1990473598241806,
|
|
"num_tokens": 5383002.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.357643556594849,
|
|
"epoch": 2.439330543933054,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046691988723642636,
|
|
"loss": 5.2855,
|
|
"mean_token_accuracy": 0.1770305499434471,
|
|
"num_tokens": 5393622.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.35295991897583,
|
|
"epoch": 2.4435146443514646,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004667512225478195,
|
|
"loss": 5.1631,
|
|
"mean_token_accuracy": 0.1849386990070343,
|
|
"num_tokens": 5403106.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.357168865203858,
|
|
"epoch": 2.4476987447698746,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046658216326602526,
|
|
"loss": 5.1612,
|
|
"mean_token_accuracy": 0.1885208159685135,
|
|
"num_tokens": 5412596.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.360099649429321,
|
|
"epoch": 2.4518828451882846,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004664127097389406,
|
|
"loss": 5.2035,
|
|
"mean_token_accuracy": 0.18829190880060195,
|
|
"num_tokens": 5422301.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.334564256668091,
|
|
"epoch": 2.4560669456066946,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046624286231527367,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.18864951580762862,
|
|
"num_tokens": 5432178.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.250463533401489,
|
|
"epoch": 2.4602510460251046,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004660726213445431,
|
|
"loss": 5.1678,
|
|
"mean_token_accuracy": 0.19103987216949464,
|
|
"num_tokens": 5441162.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.267671680450439,
|
|
"epoch": 2.4644351464435146,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004659019871770774,
|
|
"loss": 5.0925,
|
|
"mean_token_accuracy": 0.19838272631168366,
|
|
"num_tokens": 5450566.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.311909198760986,
|
|
"epoch": 2.4686192468619246,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046573096016401435,
|
|
"loss": 5.1437,
|
|
"mean_token_accuracy": 0.19732541888952254,
|
|
"num_tokens": 5460289.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.229812288284302,
|
|
"epoch": 2.4728033472803346,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004655595406572999,
|
|
"loss": 5.1217,
|
|
"mean_token_accuracy": 0.1915070191025734,
|
|
"num_tokens": 5469581.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.276672077178955,
|
|
"epoch": 2.476987447698745,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000465387729009688,
|
|
"loss": 5.2106,
|
|
"mean_token_accuracy": 0.19285676926374434,
|
|
"num_tokens": 5478005.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.279580068588257,
|
|
"epoch": 2.481171548117155,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046521552557473927,
|
|
"loss": 5.191,
|
|
"mean_token_accuracy": 0.18668065518140792,
|
|
"num_tokens": 5486963.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.3233075618743895,
|
|
"epoch": 2.485355648535565,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00046504293070682073,
|
|
"loss": 5.089,
|
|
"mean_token_accuracy": 0.1827392965555191,
|
|
"num_tokens": 5496368.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.237191534042358,
|
|
"epoch": 2.489539748953975,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00046486994476110493,
|
|
"loss": 5.0792,
|
|
"mean_token_accuracy": 0.19299820810556412,
|
|
"num_tokens": 5505706.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.196697092056274,
|
|
"epoch": 2.493723849372385,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046469656809356916,
|
|
"loss": 5.0668,
|
|
"mean_token_accuracy": 0.1921019211411476,
|
|
"num_tokens": 5513945.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.200759792327881,
|
|
"epoch": 2.497907949790795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00046452280106099454,
|
|
"loss": 5.1095,
|
|
"mean_token_accuracy": 0.1871996432542801,
|
|
"num_tokens": 5523831.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.315826892852783,
|
|
"epoch": 2.502092050209205,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004643486440209659,
|
|
"loss": 5.2538,
|
|
"mean_token_accuracy": 0.18520038872957229,
|
|
"num_tokens": 5532705.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.38626766204834,
|
|
"epoch": 2.506276150627615,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004641740973318702,
|
|
"loss": 5.1605,
|
|
"mean_token_accuracy": 0.19704229682683944,
|
|
"num_tokens": 5542793.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.264936208724976,
|
|
"epoch": 2.510460251046025,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004639991613528967,
|
|
"loss": 5.1533,
|
|
"mean_token_accuracy": 0.1924260824918747,
|
|
"num_tokens": 5551938.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.510460251046025,
|
|
"eval_entropy": 5.071391386488464,
|
|
"eval_loss": 5.53163480758667,
|
|
"eval_mean_token_accuracy": 0.18035019420712028,
|
|
"eval_num_tokens": 5551938.0,
|
|
"eval_runtime": 2.9512,
|
|
"eval_samples_per_second": 1323.855,
|
|
"eval_steps_per_second": 165.694,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"entropy": 5.292342901229858,
|
|
"epoch": 2.514644351464435,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046382383644403543,
|
|
"loss": 5.1944,
|
|
"mean_token_accuracy": 0.18540506362915038,
|
|
"num_tokens": 5561374.0,
|
|
"step": 3005
|
|
},
|
|
{
|
|
"entropy": 5.252744817733765,
|
|
"epoch": 2.518828451882845,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046364812296607676,
|
|
"loss": 5.0649,
|
|
"mean_token_accuracy": 0.18859452307224273,
|
|
"num_tokens": 5570459.0,
|
|
"step": 3010
|
|
},
|
|
{
|
|
"entropy": 5.198776197433472,
|
|
"epoch": 2.523012552301255,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046347202128061094,
|
|
"loss": 5.1679,
|
|
"mean_token_accuracy": 0.18339806497097016,
|
|
"num_tokens": 5579692.0,
|
|
"step": 3015
|
|
},
|
|
{
|
|
"entropy": 5.352016592025757,
|
|
"epoch": 2.5271966527196654,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004632955317500268,
|
|
"loss": 5.1233,
|
|
"mean_token_accuracy": 0.19220564365386963,
|
|
"num_tokens": 5588338.0,
|
|
"step": 3020
|
|
},
|
|
{
|
|
"entropy": 5.2927733898162845,
|
|
"epoch": 2.5313807531380754,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046311865473751154,
|
|
"loss": 5.1978,
|
|
"mean_token_accuracy": 0.1851389855146408,
|
|
"num_tokens": 5598589.0,
|
|
"step": 3025
|
|
},
|
|
{
|
|
"entropy": 5.333362579345703,
|
|
"epoch": 2.5355648535564854,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004629413906070495,
|
|
"loss": 5.1867,
|
|
"mean_token_accuracy": 0.18730844408273697,
|
|
"num_tokens": 5608127.0,
|
|
"step": 3030
|
|
},
|
|
{
|
|
"entropy": 5.322610855102539,
|
|
"epoch": 2.5397489539748954,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046276373972342177,
|
|
"loss": 5.2177,
|
|
"mean_token_accuracy": 0.18922489285469055,
|
|
"num_tokens": 5617728.0,
|
|
"step": 3035
|
|
},
|
|
{
|
|
"entropy": 5.319589424133301,
|
|
"epoch": 2.5439330543933054,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004625857024522054,
|
|
"loss": 5.1818,
|
|
"mean_token_accuracy": 0.18781887739896774,
|
|
"num_tokens": 5626823.0,
|
|
"step": 3040
|
|
},
|
|
{
|
|
"entropy": 5.260791206359864,
|
|
"epoch": 2.5481171548117154,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046240727915977236,
|
|
"loss": 5.1493,
|
|
"mean_token_accuracy": 0.18897433280944825,
|
|
"num_tokens": 5636284.0,
|
|
"step": 3045
|
|
},
|
|
{
|
|
"entropy": 5.3233942031860355,
|
|
"epoch": 2.5523012552301254,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004622284702132893,
|
|
"loss": 5.2784,
|
|
"mean_token_accuracy": 0.18023642897605896,
|
|
"num_tokens": 5645721.0,
|
|
"step": 3050
|
|
},
|
|
{
|
|
"entropy": 5.339594221115112,
|
|
"epoch": 2.5564853556485354,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00046204927598071595,
|
|
"loss": 5.202,
|
|
"mean_token_accuracy": 0.1866790235042572,
|
|
"num_tokens": 5654734.0,
|
|
"step": 3055
|
|
},
|
|
{
|
|
"entropy": 5.320618057250977,
|
|
"epoch": 2.560669456066946,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046186969683080547,
|
|
"loss": 5.1747,
|
|
"mean_token_accuracy": 0.19123451709747313,
|
|
"num_tokens": 5664010.0,
|
|
"step": 3060
|
|
},
|
|
{
|
|
"entropy": 5.247310018539428,
|
|
"epoch": 2.564853556485356,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046168973313310294,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.18832643777132035,
|
|
"num_tokens": 5673028.0,
|
|
"step": 3065
|
|
},
|
|
{
|
|
"entropy": 5.256870317459106,
|
|
"epoch": 2.569037656903766,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004615093852579445,
|
|
"loss": 5.1313,
|
|
"mean_token_accuracy": 0.1948254331946373,
|
|
"num_tokens": 5682517.0,
|
|
"step": 3070
|
|
},
|
|
{
|
|
"entropy": 5.1915984630584715,
|
|
"epoch": 2.573221757322176,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046132865357645717,
|
|
"loss": 5.0571,
|
|
"mean_token_accuracy": 0.19928653985261918,
|
|
"num_tokens": 5690920.0,
|
|
"step": 3075
|
|
},
|
|
{
|
|
"entropy": 5.217821025848389,
|
|
"epoch": 2.577405857740586,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00046114753846055774,
|
|
"loss": 5.1346,
|
|
"mean_token_accuracy": 0.1855100527405739,
|
|
"num_tokens": 5700336.0,
|
|
"step": 3080
|
|
},
|
|
{
|
|
"entropy": 5.363028907775879,
|
|
"epoch": 2.581589958158996,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000460966040282952,
|
|
"loss": 5.2708,
|
|
"mean_token_accuracy": 0.17967168092727662,
|
|
"num_tokens": 5711085.0,
|
|
"step": 3085
|
|
},
|
|
{
|
|
"entropy": 5.325925683975219,
|
|
"epoch": 2.585774058577406,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046078415941713384,
|
|
"loss": 5.0733,
|
|
"mean_token_accuracy": 0.19305233061313629,
|
|
"num_tokens": 5719560.0,
|
|
"step": 3090
|
|
},
|
|
{
|
|
"entropy": 5.213385915756225,
|
|
"epoch": 2.589958158995816,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004606018962373852,
|
|
"loss": 5.0424,
|
|
"mean_token_accuracy": 0.19770486056804656,
|
|
"num_tokens": 5727865.0,
|
|
"step": 3095
|
|
},
|
|
{
|
|
"entropy": 5.262719345092774,
|
|
"epoch": 2.594142259414226,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004604192511187741,
|
|
"loss": 5.1575,
|
|
"mean_token_accuracy": 0.18986293971538543,
|
|
"num_tokens": 5737705.0,
|
|
"step": 3100
|
|
},
|
|
{
|
|
"entropy": 5.290724277496338,
|
|
"epoch": 2.598326359832636,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00046023622443715483,
|
|
"loss": 5.1825,
|
|
"mean_token_accuracy": 0.17730821371078492,
|
|
"num_tokens": 5747057.0,
|
|
"step": 3105
|
|
},
|
|
{
|
|
"entropy": 5.290460109710693,
|
|
"epoch": 2.602510460251046,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00046005281656916707,
|
|
"loss": 5.1435,
|
|
"mean_token_accuracy": 0.1906292349100113,
|
|
"num_tokens": 5755793.0,
|
|
"step": 3110
|
|
},
|
|
{
|
|
"entropy": 5.308032512664795,
|
|
"epoch": 2.606694560669456,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004598690278922346,
|
|
"loss": 5.1555,
|
|
"mean_token_accuracy": 0.18686850368976593,
|
|
"num_tokens": 5765385.0,
|
|
"step": 3115
|
|
},
|
|
{
|
|
"entropy": 5.174343538284302,
|
|
"epoch": 2.6108786610878663,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045968485878456497,
|
|
"loss": 5.0638,
|
|
"mean_token_accuracy": 0.1933064341545105,
|
|
"num_tokens": 5773947.0,
|
|
"step": 3120
|
|
},
|
|
{
|
|
"entropy": 5.261218500137329,
|
|
"epoch": 2.6150627615062763,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004595003096251486,
|
|
"loss": 5.0674,
|
|
"mean_token_accuracy": 0.19476365000009538,
|
|
"num_tokens": 5782700.0,
|
|
"step": 3125
|
|
},
|
|
{
|
|
"entropy": 5.330811834335327,
|
|
"epoch": 2.6192468619246863,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00045931538079375803,
|
|
"loss": 5.2227,
|
|
"mean_token_accuracy": 0.18306794613599778,
|
|
"num_tokens": 5792452.0,
|
|
"step": 3130
|
|
},
|
|
{
|
|
"entropy": 5.3118369579315186,
|
|
"epoch": 2.6234309623430963,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045913007267094704,
|
|
"loss": 5.1593,
|
|
"mean_token_accuracy": 0.18939217180013657,
|
|
"num_tokens": 5801270.0,
|
|
"step": 3135
|
|
},
|
|
{
|
|
"entropy": 5.282789945602417,
|
|
"epoch": 2.6276150627615062,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004589443856380499,
|
|
"loss": 5.1985,
|
|
"mean_token_accuracy": 0.19086408019065856,
|
|
"num_tokens": 5810704.0,
|
|
"step": 3140
|
|
},
|
|
{
|
|
"entropy": 5.302239179611206,
|
|
"epoch": 2.6317991631799162,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004587583200771807,
|
|
"loss": 5.1343,
|
|
"mean_token_accuracy": 0.18619504123926162,
|
|
"num_tokens": 5819690.0,
|
|
"step": 3145
|
|
},
|
|
{
|
|
"entropy": 5.145789575576782,
|
|
"epoch": 2.6359832635983262,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004585718763712325,
|
|
"loss": 5.0737,
|
|
"mean_token_accuracy": 0.18743123710155488,
|
|
"num_tokens": 5828912.0,
|
|
"step": 3150
|
|
},
|
|
{
|
|
"entropy": 5.172671127319336,
|
|
"epoch": 2.6401673640167362,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004583850549038764,
|
|
"loss": 4.9916,
|
|
"mean_token_accuracy": 0.19601633697748183,
|
|
"num_tokens": 5836914.0,
|
|
"step": 3155
|
|
},
|
|
{
|
|
"entropy": 5.252508640289307,
|
|
"epoch": 2.6443514644351467,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045819785605956105,
|
|
"loss": 5.1656,
|
|
"mean_token_accuracy": 0.19016501307487488,
|
|
"num_tokens": 5846268.0,
|
|
"step": 3160
|
|
},
|
|
{
|
|
"entropy": 5.242251014709472,
|
|
"epoch": 2.6485355648535567,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004580102802235115,
|
|
"loss": 5.1621,
|
|
"mean_token_accuracy": 0.19489868134260177,
|
|
"num_tokens": 5855029.0,
|
|
"step": 3165
|
|
},
|
|
{
|
|
"entropy": 5.319009685516358,
|
|
"epoch": 2.6527196652719667,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004578223277817286,
|
|
"loss": 5.1608,
|
|
"mean_token_accuracy": 0.18146876245737076,
|
|
"num_tokens": 5865049.0,
|
|
"step": 3170
|
|
},
|
|
{
|
|
"entropy": 5.3414257049560545,
|
|
"epoch": 2.6569037656903767,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045763399912098835,
|
|
"loss": 5.1667,
|
|
"mean_token_accuracy": 0.18784808069467546,
|
|
"num_tokens": 5873593.0,
|
|
"step": 3175
|
|
},
|
|
{
|
|
"entropy": 5.179458951950073,
|
|
"epoch": 2.6610878661087867,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00045744529462884086,
|
|
"loss": 5.0692,
|
|
"mean_token_accuracy": 0.189924156665802,
|
|
"num_tokens": 5882562.0,
|
|
"step": 3180
|
|
},
|
|
{
|
|
"entropy": 5.275640201568604,
|
|
"epoch": 2.6652719665271967,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045725621469360964,
|
|
"loss": 5.2603,
|
|
"mean_token_accuracy": 0.17695028930902482,
|
|
"num_tokens": 5892516.0,
|
|
"step": 3185
|
|
},
|
|
{
|
|
"entropy": 5.27607274055481,
|
|
"epoch": 2.6694560669456067,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004570667597043907,
|
|
"loss": 5.1484,
|
|
"mean_token_accuracy": 0.19291958808898926,
|
|
"num_tokens": 5901494.0,
|
|
"step": 3190
|
|
},
|
|
{
|
|
"entropy": 5.2923718929290775,
|
|
"epoch": 2.6736401673640167,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000456876930051052,
|
|
"loss": 5.1307,
|
|
"mean_token_accuracy": 0.18793087750673293,
|
|
"num_tokens": 5910750.0,
|
|
"step": 3195
|
|
},
|
|
{
|
|
"entropy": 5.321951484680175,
|
|
"epoch": 2.6778242677824267,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004566867261242325,
|
|
"loss": 5.1863,
|
|
"mean_token_accuracy": 0.19099028408527374,
|
|
"num_tokens": 5920277.0,
|
|
"step": 3200
|
|
},
|
|
{
|
|
"entropy": 5.236455678939819,
|
|
"epoch": 2.6820083682008367,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045649614831534127,
|
|
"loss": 5.1547,
|
|
"mean_token_accuracy": 0.18574196845293045,
|
|
"num_tokens": 5929880.0,
|
|
"step": 3205
|
|
},
|
|
{
|
|
"entropy": 5.421674203872681,
|
|
"epoch": 2.6861924686192467,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045630519701655677,
|
|
"loss": 5.3141,
|
|
"mean_token_accuracy": 0.18591759502887725,
|
|
"num_tokens": 5939538.0,
|
|
"step": 3210
|
|
},
|
|
{
|
|
"entropy": 5.269159460067749,
|
|
"epoch": 2.6903765690376567,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045611387262082604,
|
|
"loss": 5.1458,
|
|
"mean_token_accuracy": 0.18764640390872955,
|
|
"num_tokens": 5948747.0,
|
|
"step": 3215
|
|
},
|
|
{
|
|
"entropy": 5.210002279281616,
|
|
"epoch": 2.694560669456067,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045592217552186407,
|
|
"loss": 5.1366,
|
|
"mean_token_accuracy": 0.1887553408741951,
|
|
"num_tokens": 5958209.0,
|
|
"step": 3220
|
|
},
|
|
{
|
|
"entropy": 5.334749937057495,
|
|
"epoch": 2.698744769874477,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004557301061141525,
|
|
"loss": 5.1765,
|
|
"mean_token_accuracy": 0.184686116874218,
|
|
"num_tokens": 5967184.0,
|
|
"step": 3225
|
|
},
|
|
{
|
|
"entropy": 5.253610706329345,
|
|
"epoch": 2.702928870292887,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00045553766479293945,
|
|
"loss": 5.1898,
|
|
"mean_token_accuracy": 0.18726228028535843,
|
|
"num_tokens": 5976006.0,
|
|
"step": 3230
|
|
},
|
|
{
|
|
"entropy": 5.16000804901123,
|
|
"epoch": 2.707112970711297,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00045534485195423806,
|
|
"loss": 5.0375,
|
|
"mean_token_accuracy": 0.19638874381780624,
|
|
"num_tokens": 5984671.0,
|
|
"step": 3235
|
|
},
|
|
{
|
|
"entropy": 5.338690757751465,
|
|
"epoch": 2.711297071129707,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045515166799482633,
|
|
"loss": 5.1543,
|
|
"mean_token_accuracy": 0.1887945681810379,
|
|
"num_tokens": 5993190.0,
|
|
"step": 3240
|
|
},
|
|
{
|
|
"entropy": 5.255921268463135,
|
|
"epoch": 2.715481171548117,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00045495811331224564,
|
|
"loss": 5.1492,
|
|
"mean_token_accuracy": 0.18723828792572023,
|
|
"num_tokens": 6004080.0,
|
|
"step": 3245
|
|
},
|
|
{
|
|
"entropy": 5.252399635314942,
|
|
"epoch": 2.719665271966527,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004547641883048005,
|
|
"loss": 5.1552,
|
|
"mean_token_accuracy": 0.18935271948575974,
|
|
"num_tokens": 6013212.0,
|
|
"step": 3250
|
|
},
|
|
{
|
|
"entropy": 5.255982971191406,
|
|
"epoch": 2.723849372384937,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004545698933715573,
|
|
"loss": 5.1035,
|
|
"mean_token_accuracy": 0.19102832973003386,
|
|
"num_tokens": 6023444.0,
|
|
"step": 3255
|
|
},
|
|
{
|
|
"entropy": 5.272893381118775,
|
|
"epoch": 2.7280334728033475,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045437522891234386,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.1912008211016655,
|
|
"num_tokens": 6032217.0,
|
|
"step": 3260
|
|
},
|
|
{
|
|
"entropy": 5.215366220474243,
|
|
"epoch": 2.7322175732217575,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045418019532774827,
|
|
"loss": 5.0544,
|
|
"mean_token_accuracy": 0.19953375905752183,
|
|
"num_tokens": 6040246.0,
|
|
"step": 3265
|
|
},
|
|
{
|
|
"entropy": 5.208912658691406,
|
|
"epoch": 2.7364016736401675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045398479301911836,
|
|
"loss": 5.1659,
|
|
"mean_token_accuracy": 0.18322770148515702,
|
|
"num_tokens": 6049447.0,
|
|
"step": 3270
|
|
},
|
|
{
|
|
"entropy": 5.158584499359131,
|
|
"epoch": 2.7405857740585775,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045378902238856064,
|
|
"loss": 5.0487,
|
|
"mean_token_accuracy": 0.19468893259763717,
|
|
"num_tokens": 6058294.0,
|
|
"step": 3275
|
|
},
|
|
{
|
|
"entropy": 5.282944250106811,
|
|
"epoch": 2.7447698744769875,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004535928838389396,
|
|
"loss": 5.1726,
|
|
"mean_token_accuracy": 0.19354380518198014,
|
|
"num_tokens": 6067237.0,
|
|
"step": 3280
|
|
},
|
|
{
|
|
"entropy": 5.317514514923095,
|
|
"epoch": 2.7489539748953975,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004533963777738768,
|
|
"loss": 5.1742,
|
|
"mean_token_accuracy": 0.18680903315544128,
|
|
"num_tokens": 6075599.0,
|
|
"step": 3285
|
|
},
|
|
{
|
|
"entropy": 5.292194890975952,
|
|
"epoch": 2.7531380753138075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00045319950459775024,
|
|
"loss": 5.1535,
|
|
"mean_token_accuracy": 0.1817997068166733,
|
|
"num_tokens": 6085131.0,
|
|
"step": 3290
|
|
},
|
|
{
|
|
"entropy": 5.271710014343261,
|
|
"epoch": 2.7573221757322175,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004530022647156931,
|
|
"loss": 5.0994,
|
|
"mean_token_accuracy": 0.19030941724777223,
|
|
"num_tokens": 6094745.0,
|
|
"step": 3295
|
|
},
|
|
{
|
|
"entropy": 5.216010618209839,
|
|
"epoch": 2.7615062761506275,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004528046585335935,
|
|
"loss": 5.1664,
|
|
"mean_token_accuracy": 0.1816794067621231,
|
|
"num_tokens": 6104182.0,
|
|
"step": 3300
|
|
},
|
|
{
|
|
"entropy": 5.274562120437622,
|
|
"epoch": 2.7656903765690375,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00045260668645809304,
|
|
"loss": 5.0309,
|
|
"mean_token_accuracy": 0.19805881083011628,
|
|
"num_tokens": 6113370.0,
|
|
"step": 3305
|
|
},
|
|
{
|
|
"entropy": 5.183791971206665,
|
|
"epoch": 2.7698744769874475,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004524083488965866,
|
|
"loss": 5.1193,
|
|
"mean_token_accuracy": 0.19656048566102982,
|
|
"num_tokens": 6122202.0,
|
|
"step": 3310
|
|
},
|
|
{
|
|
"entropy": 5.258046197891235,
|
|
"epoch": 2.7740585774058575,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004522096462572208,
|
|
"loss": 5.1677,
|
|
"mean_token_accuracy": 0.18603909462690355,
|
|
"num_tokens": 6131081.0,
|
|
"step": 3315
|
|
},
|
|
{
|
|
"entropy": 5.375565719604492,
|
|
"epoch": 2.778242677824268,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004520105789488939,
|
|
"loss": 5.2482,
|
|
"mean_token_accuracy": 0.1816224291920662,
|
|
"num_tokens": 6141707.0,
|
|
"step": 3320
|
|
},
|
|
{
|
|
"entropy": 5.219994783401489,
|
|
"epoch": 2.782426778242678,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00045181114738125427,
|
|
"loss": 5.0883,
|
|
"mean_token_accuracy": 0.19167425930500032,
|
|
"num_tokens": 6151651.0,
|
|
"step": 3325
|
|
},
|
|
{
|
|
"entropy": 5.2585258960723875,
|
|
"epoch": 2.786610878661088,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045161135196470015,
|
|
"loss": 5.0854,
|
|
"mean_token_accuracy": 0.19647460281848908,
|
|
"num_tokens": 6159928.0,
|
|
"step": 3330
|
|
},
|
|
{
|
|
"entropy": 5.22525954246521,
|
|
"epoch": 2.790794979079498,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004514111931103783,
|
|
"loss": 5.1599,
|
|
"mean_token_accuracy": 0.1832011714577675,
|
|
"num_tokens": 6169115.0,
|
|
"step": 3335
|
|
},
|
|
{
|
|
"entropy": 5.246417474746704,
|
|
"epoch": 2.794979079497908,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00045121067123018353,
|
|
"loss": 5.1782,
|
|
"mean_token_accuracy": 0.18664777427911758,
|
|
"num_tokens": 6177995.0,
|
|
"step": 3340
|
|
},
|
|
{
|
|
"entropy": 5.317909908294678,
|
|
"epoch": 2.799163179916318,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045100978673675755,
|
|
"loss": 5.1669,
|
|
"mean_token_accuracy": 0.18307592272758483,
|
|
"num_tokens": 6188081.0,
|
|
"step": 3345
|
|
},
|
|
{
|
|
"entropy": 5.327097654342651,
|
|
"epoch": 2.803347280334728,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045080854004348836,
|
|
"loss": 5.1301,
|
|
"mean_token_accuracy": 0.18444235771894454,
|
|
"num_tokens": 6198010.0,
|
|
"step": 3350
|
|
},
|
|
{
|
|
"entropy": 5.156101369857788,
|
|
"epoch": 2.8075313807531384,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045060693156450933,
|
|
"loss": 5.1072,
|
|
"mean_token_accuracy": 0.19760254472494126,
|
|
"num_tokens": 6207164.0,
|
|
"step": 3355
|
|
},
|
|
{
|
|
"entropy": 5.3418936252594,
|
|
"epoch": 2.8117154811715483,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004504049617146983,
|
|
"loss": 5.229,
|
|
"mean_token_accuracy": 0.1856566309928894,
|
|
"num_tokens": 6216512.0,
|
|
"step": 3360
|
|
},
|
|
{
|
|
"entropy": 5.300665712356567,
|
|
"epoch": 2.8158995815899583,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045020263090967674,
|
|
"loss": 5.199,
|
|
"mean_token_accuracy": 0.19643250554800035,
|
|
"num_tokens": 6226196.0,
|
|
"step": 3365
|
|
},
|
|
{
|
|
"entropy": 5.239013671875,
|
|
"epoch": 2.8200836820083683,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004499999395658088,
|
|
"loss": 5.0575,
|
|
"mean_token_accuracy": 0.18818544298410417,
|
|
"num_tokens": 6234820.0,
|
|
"step": 3370
|
|
},
|
|
{
|
|
"entropy": 5.232560491561889,
|
|
"epoch": 2.8242677824267783,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000449796888100201,
|
|
"loss": 5.1225,
|
|
"mean_token_accuracy": 0.18812528103590012,
|
|
"num_tokens": 6243813.0,
|
|
"step": 3375
|
|
},
|
|
{
|
|
"entropy": 5.379570436477661,
|
|
"epoch": 2.8284518828451883,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044959347693070026,
|
|
"loss": 5.2175,
|
|
"mean_token_accuracy": 0.1857367128133774,
|
|
"num_tokens": 6253376.0,
|
|
"step": 3380
|
|
},
|
|
{
|
|
"entropy": 5.306171989440918,
|
|
"epoch": 2.8326359832635983,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004493897064758943,
|
|
"loss": 5.2418,
|
|
"mean_token_accuracy": 0.18497615456581115,
|
|
"num_tokens": 6262640.0,
|
|
"step": 3385
|
|
},
|
|
{
|
|
"entropy": 5.297601509094238,
|
|
"epoch": 2.8368200836820083,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004491855771551098,
|
|
"loss": 5.1818,
|
|
"mean_token_accuracy": 0.19384098201990127,
|
|
"num_tokens": 6270908.0,
|
|
"step": 3390
|
|
},
|
|
{
|
|
"entropy": 5.268275451660156,
|
|
"epoch": 2.8410041841004183,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004489810893884121,
|
|
"loss": 5.0911,
|
|
"mean_token_accuracy": 0.18632905334234237,
|
|
"num_tokens": 6280361.0,
|
|
"step": 3395
|
|
},
|
|
{
|
|
"entropy": 5.2687921047210695,
|
|
"epoch": 2.8451882845188283,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004487762435966042,
|
|
"loss": 5.1139,
|
|
"mean_token_accuracy": 0.18623328059911728,
|
|
"num_tokens": 6288832.0,
|
|
"step": 3400
|
|
},
|
|
{
|
|
"entropy": 5.22910885810852,
|
|
"epoch": 2.8493723849372383,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044857104020122585,
|
|
"loss": 5.0722,
|
|
"mean_token_accuracy": 0.1906009793281555,
|
|
"num_tokens": 6298553.0,
|
|
"step": 3405
|
|
},
|
|
{
|
|
"entropy": 5.178006649017334,
|
|
"epoch": 2.8535564853556483,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004483654796245526,
|
|
"loss": 5.057,
|
|
"mean_token_accuracy": 0.1915947452187538,
|
|
"num_tokens": 6307692.0,
|
|
"step": 3410
|
|
},
|
|
{
|
|
"entropy": 5.329270124435425,
|
|
"epoch": 2.8577405857740583,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000448159562289595,
|
|
"loss": 5.1468,
|
|
"mean_token_accuracy": 0.1889610692858696,
|
|
"num_tokens": 6316729.0,
|
|
"step": 3415
|
|
},
|
|
{
|
|
"entropy": 5.346681594848633,
|
|
"epoch": 2.8619246861924688,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044795328862009794,
|
|
"loss": 5.2339,
|
|
"mean_token_accuracy": 0.18380739092826842,
|
|
"num_tokens": 6326210.0,
|
|
"step": 3420
|
|
},
|
|
{
|
|
"entropy": 5.185216903686523,
|
|
"epoch": 2.8661087866108788,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044774665904053943,
|
|
"loss": 5.1183,
|
|
"mean_token_accuracy": 0.19041745066642762,
|
|
"num_tokens": 6335156.0,
|
|
"step": 3425
|
|
},
|
|
{
|
|
"entropy": 5.138974666595459,
|
|
"epoch": 2.8702928870292888,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004475396739761299,
|
|
"loss": 5.02,
|
|
"mean_token_accuracy": 0.1941646233201027,
|
|
"num_tokens": 6344264.0,
|
|
"step": 3430
|
|
},
|
|
{
|
|
"entropy": 5.239344882965088,
|
|
"epoch": 2.8744769874476988,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00044733233385281117,
|
|
"loss": 5.0785,
|
|
"mean_token_accuracy": 0.19590935558080674,
|
|
"num_tokens": 6352997.0,
|
|
"step": 3435
|
|
},
|
|
{
|
|
"entropy": 5.187147998809815,
|
|
"epoch": 2.8786610878661087,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00044712463909725626,
|
|
"loss": 5.0965,
|
|
"mean_token_accuracy": 0.19587049186229705,
|
|
"num_tokens": 6362312.0,
|
|
"step": 3440
|
|
},
|
|
{
|
|
"entropy": 5.312726163864136,
|
|
"epoch": 2.8828451882845187,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00044691659013686727,
|
|
"loss": 5.2189,
|
|
"mean_token_accuracy": 0.1809667631983757,
|
|
"num_tokens": 6372107.0,
|
|
"step": 3445
|
|
},
|
|
{
|
|
"entropy": 5.2199914932250975,
|
|
"epoch": 2.8870292887029287,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004467081873997756,
|
|
"loss": 5.0499,
|
|
"mean_token_accuracy": 0.20045217275619506,
|
|
"num_tokens": 6381306.0,
|
|
"step": 3450
|
|
},
|
|
{
|
|
"entropy": 5.187382984161377,
|
|
"epoch": 2.891213389121339,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044649943131484056,
|
|
"loss": 4.9838,
|
|
"mean_token_accuracy": 0.19676827192306517,
|
|
"num_tokens": 6390404.0,
|
|
"step": 3455
|
|
},
|
|
{
|
|
"entropy": 5.328663730621338,
|
|
"epoch": 2.895397489539749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044629032231164874,
|
|
"loss": 5.2171,
|
|
"mean_token_accuracy": 0.18264185041189193,
|
|
"num_tokens": 6400366.0,
|
|
"step": 3460
|
|
},
|
|
{
|
|
"entropy": 5.221409893035888,
|
|
"epoch": 2.899581589958159,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004460808608205127,
|
|
"loss": 5.0359,
|
|
"mean_token_accuracy": 0.19969803243875503,
|
|
"num_tokens": 6408955.0,
|
|
"step": 3465
|
|
},
|
|
{
|
|
"entropy": 5.120856666564942,
|
|
"epoch": 2.903765690376569,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004458710472724706,
|
|
"loss": 5.0829,
|
|
"mean_token_accuracy": 0.19425891637802123,
|
|
"num_tokens": 6417172.0,
|
|
"step": 3470
|
|
},
|
|
{
|
|
"entropy": 5.25793046951294,
|
|
"epoch": 2.907949790794979,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044566088209928485,
|
|
"loss": 5.1188,
|
|
"mean_token_accuracy": 0.19275038093328475,
|
|
"num_tokens": 6425898.0,
|
|
"step": 3475
|
|
},
|
|
{
|
|
"entropy": 5.213550615310669,
|
|
"epoch": 2.912133891213389,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00044545036573344175,
|
|
"loss": 5.1346,
|
|
"mean_token_accuracy": 0.18566753566265107,
|
|
"num_tokens": 6435940.0,
|
|
"step": 3480
|
|
},
|
|
{
|
|
"entropy": 5.196372985839844,
|
|
"epoch": 2.916317991631799,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044523949860815015,
|
|
"loss": 5.0784,
|
|
"mean_token_accuracy": 0.19582689106464385,
|
|
"num_tokens": 6445119.0,
|
|
"step": 3485
|
|
},
|
|
{
|
|
"entropy": 5.231039810180664,
|
|
"epoch": 2.920502092050209,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004450282811573405,
|
|
"loss": 5.1007,
|
|
"mean_token_accuracy": 0.19189773201942445,
|
|
"num_tokens": 6453986.0,
|
|
"step": 3490
|
|
},
|
|
{
|
|
"entropy": 5.200790023803711,
|
|
"epoch": 2.924686192468619,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044481671381566446,
|
|
"loss": 5.0776,
|
|
"mean_token_accuracy": 0.19045459479093552,
|
|
"num_tokens": 6463658.0,
|
|
"step": 3495
|
|
},
|
|
{
|
|
"entropy": 5.230255842208862,
|
|
"epoch": 2.928870292887029,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004446047970184937,
|
|
"loss": 5.1212,
|
|
"mean_token_accuracy": 0.1928677961230278,
|
|
"num_tokens": 6472370.0,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"epoch": 2.928870292887029,
|
|
"eval_entropy": 5.1023782503873045,
|
|
"eval_loss": 5.454714775085449,
|
|
"eval_mean_token_accuracy": 0.18507229949677162,
|
|
"eval_num_tokens": 6472370.0,
|
|
"eval_runtime": 2.7385,
|
|
"eval_samples_per_second": 1426.699,
|
|
"eval_steps_per_second": 178.566,
|
|
"step": 3500
|
|
},
|
|
{
|
|
"entropy": 5.289586400985717,
|
|
"epoch": 2.933054393305439,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00044439253120191877,
|
|
"loss": 5.2474,
|
|
"mean_token_accuracy": 0.1766175702214241,
|
|
"num_tokens": 6482642.0,
|
|
"step": 3505
|
|
},
|
|
{
|
|
"entropy": 5.2938282012939455,
|
|
"epoch": 2.937238493723849,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044417991680274874,
|
|
"loss": 5.1557,
|
|
"mean_token_accuracy": 0.19141605347394944,
|
|
"num_tokens": 6491818.0,
|
|
"step": 3510
|
|
},
|
|
{
|
|
"entropy": 5.238175296783448,
|
|
"epoch": 2.941422594142259,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004439669542585097,
|
|
"loss": 5.1424,
|
|
"mean_token_accuracy": 0.18577272742986678,
|
|
"num_tokens": 6500908.0,
|
|
"step": 3515
|
|
},
|
|
{
|
|
"entropy": 5.198642253875732,
|
|
"epoch": 2.9456066945606696,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004437536440074444,
|
|
"loss": 5.1208,
|
|
"mean_token_accuracy": 0.19453504383563996,
|
|
"num_tokens": 6510182.0,
|
|
"step": 3520
|
|
},
|
|
{
|
|
"entropy": 5.196361923217774,
|
|
"epoch": 2.9497907949790796,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004435399864885111,
|
|
"loss": 5.0929,
|
|
"mean_token_accuracy": 0.19343974441289902,
|
|
"num_tokens": 6519051.0,
|
|
"step": 3525
|
|
},
|
|
{
|
|
"entropy": 5.1653931617736815,
|
|
"epoch": 2.9539748953974896,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004433259821413825,
|
|
"loss": 5.0259,
|
|
"mean_token_accuracy": 0.20487706661224364,
|
|
"num_tokens": 6527949.0,
|
|
"step": 3530
|
|
},
|
|
{
|
|
"entropy": 5.379975843429565,
|
|
"epoch": 2.9581589958158996,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044311163140644524,
|
|
"loss": 5.2889,
|
|
"mean_token_accuracy": 0.18154271692037582,
|
|
"num_tokens": 6537227.0,
|
|
"step": 3535
|
|
},
|
|
{
|
|
"entropy": 5.203045082092285,
|
|
"epoch": 2.9623430962343096,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044289693472479867,
|
|
"loss": 5.0896,
|
|
"mean_token_accuracy": 0.1889070376753807,
|
|
"num_tokens": 6546938.0,
|
|
"step": 3540
|
|
},
|
|
{
|
|
"entropy": 5.22341251373291,
|
|
"epoch": 2.9665271966527196,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044268189253825397,
|
|
"loss": 5.1005,
|
|
"mean_token_accuracy": 0.19170305579900743,
|
|
"num_tokens": 6555750.0,
|
|
"step": 3545
|
|
},
|
|
{
|
|
"entropy": 5.1741471767425535,
|
|
"epoch": 2.9707112970711296,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004424665052893334,
|
|
"loss": 5.0377,
|
|
"mean_token_accuracy": 0.19787123501300813,
|
|
"num_tokens": 6565680.0,
|
|
"step": 3550
|
|
},
|
|
{
|
|
"entropy": 5.095081043243408,
|
|
"epoch": 2.97489539748954,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00044225077342126935,
|
|
"loss": 5.0187,
|
|
"mean_token_accuracy": 0.19338533133268357,
|
|
"num_tokens": 6574595.0,
|
|
"step": 3555
|
|
},
|
|
{
|
|
"entropy": 5.268528985977173,
|
|
"epoch": 2.97907949790795,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004420346973780033,
|
|
"loss": 5.1403,
|
|
"mean_token_accuracy": 0.1893066793680191,
|
|
"num_tokens": 6583333.0,
|
|
"step": 3560
|
|
},
|
|
{
|
|
"entropy": 5.29896993637085,
|
|
"epoch": 2.98326359832636,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000441818277604185,
|
|
"loss": 5.1634,
|
|
"mean_token_accuracy": 0.19332201927900314,
|
|
"num_tokens": 6592252.0,
|
|
"step": 3565
|
|
},
|
|
{
|
|
"entropy": 5.248419380187988,
|
|
"epoch": 2.98744769874477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004416015145451715,
|
|
"loss": 5.0151,
|
|
"mean_token_accuracy": 0.1980673924088478,
|
|
"num_tokens": 6600675.0,
|
|
"step": 3570
|
|
},
|
|
{
|
|
"entropy": 5.184472894668579,
|
|
"epoch": 2.99163179916318,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004413844086470264,
|
|
"loss": 5.0997,
|
|
"mean_token_accuracy": 0.1909830391407013,
|
|
"num_tokens": 6609722.0,
|
|
"step": 3575
|
|
},
|
|
{
|
|
"entropy": 5.22449893951416,
|
|
"epoch": 2.99581589958159,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004411669603565187,
|
|
"loss": 5.0824,
|
|
"mean_token_accuracy": 0.1947742760181427,
|
|
"num_tokens": 6618415.0,
|
|
"step": 3580
|
|
},
|
|
{
|
|
"entropy": 5.222368383407593,
|
|
"epoch": 3.0,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000440949170121122,
|
|
"loss": 5.0946,
|
|
"mean_token_accuracy": 0.1937455415725708,
|
|
"num_tokens": 6627966.0,
|
|
"step": 3585
|
|
},
|
|
{
|
|
"entropy": 5.221515703201294,
|
|
"epoch": 3.00418410041841,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044073103838901373,
|
|
"loss": 4.7313,
|
|
"mean_token_accuracy": 0.2179719626903534,
|
|
"num_tokens": 6636820.0,
|
|
"step": 3590
|
|
},
|
|
{
|
|
"entropy": 5.115354728698731,
|
|
"epoch": 3.00836820083682,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00044051256560907374,
|
|
"loss": 4.7634,
|
|
"mean_token_accuracy": 0.20080406963825226,
|
|
"num_tokens": 6645792.0,
|
|
"step": 3595
|
|
},
|
|
{
|
|
"entropy": 5.272567415237427,
|
|
"epoch": 3.01255230125523,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00044029375223088394,
|
|
"loss": 4.8738,
|
|
"mean_token_accuracy": 0.2002882704138756,
|
|
"num_tokens": 6654850.0,
|
|
"step": 3600
|
|
},
|
|
{
|
|
"entropy": 5.1043055057525635,
|
|
"epoch": 3.01673640167364,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00044007459870472706,
|
|
"loss": 4.7996,
|
|
"mean_token_accuracy": 0.20851542800664902,
|
|
"num_tokens": 6664524.0,
|
|
"step": 3605
|
|
},
|
|
{
|
|
"entropy": 5.063422203063965,
|
|
"epoch": 3.02092050209205,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004398551054815859,
|
|
"loss": 4.7462,
|
|
"mean_token_accuracy": 0.21160941421985627,
|
|
"num_tokens": 6674330.0,
|
|
"step": 3610
|
|
},
|
|
{
|
|
"entropy": 5.186771106719971,
|
|
"epoch": 3.0251046025104604,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004396352730131422,
|
|
"loss": 4.8139,
|
|
"mean_token_accuracy": 0.2042303502559662,
|
|
"num_tokens": 6683121.0,
|
|
"step": 3615
|
|
},
|
|
{
|
|
"entropy": 5.164731836318969,
|
|
"epoch": 3.0292887029288704,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004394151017517758,
|
|
"loss": 4.7725,
|
|
"mean_token_accuracy": 0.20801787227392196,
|
|
"num_tokens": 6692566.0,
|
|
"step": 3620
|
|
},
|
|
{
|
|
"entropy": 5.201624250411987,
|
|
"epoch": 3.0334728033472804,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004391945921505638,
|
|
"loss": 4.8651,
|
|
"mean_token_accuracy": 0.200774884223938,
|
|
"num_tokens": 6701577.0,
|
|
"step": 3625
|
|
},
|
|
{
|
|
"entropy": 5.084546136856079,
|
|
"epoch": 3.0376569037656904,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00043897374466327944,
|
|
"loss": 4.746,
|
|
"mean_token_accuracy": 0.21395038068294525,
|
|
"num_tokens": 6710672.0,
|
|
"step": 3630
|
|
},
|
|
{
|
|
"entropy": 5.098412609100341,
|
|
"epoch": 3.0418410041841004,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043875255974439145,
|
|
"loss": 4.7452,
|
|
"mean_token_accuracy": 0.21385350823402405,
|
|
"num_tokens": 6719412.0,
|
|
"step": 3635
|
|
},
|
|
{
|
|
"entropy": 5.060048151016235,
|
|
"epoch": 3.0460251046025104,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004385310378490627,
|
|
"loss": 4.7301,
|
|
"mean_token_accuracy": 0.20718805640935897,
|
|
"num_tokens": 6729169.0,
|
|
"step": 3640
|
|
},
|
|
{
|
|
"entropy": 5.1146400451660154,
|
|
"epoch": 3.0502092050209204,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043830917943314977,
|
|
"loss": 4.6793,
|
|
"mean_token_accuracy": 0.2218550533056259,
|
|
"num_tokens": 6738549.0,
|
|
"step": 3645
|
|
},
|
|
{
|
|
"entropy": 5.10999026298523,
|
|
"epoch": 3.0543933054393304,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043808698495320154,
|
|
"loss": 4.7457,
|
|
"mean_token_accuracy": 0.20681171864271164,
|
|
"num_tokens": 6747909.0,
|
|
"step": 3650
|
|
},
|
|
{
|
|
"entropy": 5.094274282455444,
|
|
"epoch": 3.0585774058577404,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00043786445486645846,
|
|
"loss": 4.7929,
|
|
"mean_token_accuracy": 0.21617049127817153,
|
|
"num_tokens": 6757125.0,
|
|
"step": 3655
|
|
},
|
|
{
|
|
"entropy": 5.131888103485108,
|
|
"epoch": 3.062761506276151,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004376415896308518,
|
|
"loss": 4.7622,
|
|
"mean_token_accuracy": 0.21757681369781495,
|
|
"num_tokens": 6765774.0,
|
|
"step": 3660
|
|
},
|
|
{
|
|
"entropy": 5.1514064311981205,
|
|
"epoch": 3.066945606694561,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043741838970500234,
|
|
"loss": 4.8389,
|
|
"mean_token_accuracy": 0.21301922649145127,
|
|
"num_tokens": 6774980.0,
|
|
"step": 3665
|
|
},
|
|
{
|
|
"entropy": 5.146578407287597,
|
|
"epoch": 3.071129707112971,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004371948555482196,
|
|
"loss": 4.8403,
|
|
"mean_token_accuracy": 0.20335593521595002,
|
|
"num_tokens": 6784187.0,
|
|
"step": 3670
|
|
},
|
|
{
|
|
"entropy": 5.15595588684082,
|
|
"epoch": 3.075313807531381,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000436970987620501,
|
|
"loss": 4.8511,
|
|
"mean_token_accuracy": 0.20775990188121796,
|
|
"num_tokens": 6792953.0,
|
|
"step": 3675
|
|
},
|
|
{
|
|
"entropy": 5.148085451126098,
|
|
"epoch": 3.079497907949791,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00043674678638253066,
|
|
"loss": 4.7258,
|
|
"mean_token_accuracy": 0.21285004317760467,
|
|
"num_tokens": 6803004.0,
|
|
"step": 3680
|
|
},
|
|
{
|
|
"entropy": 5.067785120010376,
|
|
"epoch": 3.083682008368201,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004365222522956788,
|
|
"loss": 4.7829,
|
|
"mean_token_accuracy": 0.2070772871375084,
|
|
"num_tokens": 6811883.0,
|
|
"step": 3685
|
|
},
|
|
{
|
|
"entropy": 5.081501483917236,
|
|
"epoch": 3.087866108786611,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004362973858220004,
|
|
"loss": 4.7942,
|
|
"mean_token_accuracy": 0.21725988686084746,
|
|
"num_tokens": 6820327.0,
|
|
"step": 3690
|
|
},
|
|
{
|
|
"entropy": 5.119329977035522,
|
|
"epoch": 3.092050209205021,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043607218742423447,
|
|
"loss": 4.7859,
|
|
"mean_token_accuracy": 0.20845946818590164,
|
|
"num_tokens": 6829412.0,
|
|
"step": 3695
|
|
},
|
|
{
|
|
"entropy": 5.076244688034057,
|
|
"epoch": 3.096234309623431,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043584665756580323,
|
|
"loss": 4.779,
|
|
"mean_token_accuracy": 0.20705614238977432,
|
|
"num_tokens": 6838997.0,
|
|
"step": 3700
|
|
},
|
|
{
|
|
"entropy": 5.189417886734009,
|
|
"epoch": 3.100418410041841,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004356207967108106,
|
|
"loss": 4.8377,
|
|
"mean_token_accuracy": 0.19680506885051727,
|
|
"num_tokens": 6848957.0,
|
|
"step": 3705
|
|
},
|
|
{
|
|
"entropy": 5.122786378860473,
|
|
"epoch": 3.104602510460251,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004353946053240422,
|
|
"loss": 4.8893,
|
|
"mean_token_accuracy": 0.2002638354897499,
|
|
"num_tokens": 6859688.0,
|
|
"step": 3710
|
|
},
|
|
{
|
|
"entropy": 5.160658884048462,
|
|
"epoch": 3.1087866108786613,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004351680838709635,
|
|
"loss": 4.9205,
|
|
"mean_token_accuracy": 0.19382916539907455,
|
|
"num_tokens": 6869149.0,
|
|
"step": 3715
|
|
},
|
|
{
|
|
"entropy": 5.148589992523194,
|
|
"epoch": 3.1129707112970713,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043494123281771915,
|
|
"loss": 4.8765,
|
|
"mean_token_accuracy": 0.20164572298526764,
|
|
"num_tokens": 6879432.0,
|
|
"step": 3720
|
|
},
|
|
{
|
|
"entropy": 5.105045413970947,
|
|
"epoch": 3.1171548117154813,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043471405263113214,
|
|
"loss": 4.841,
|
|
"mean_token_accuracy": 0.2052438274025917,
|
|
"num_tokens": 6888398.0,
|
|
"step": 3725
|
|
},
|
|
{
|
|
"entropy": 5.187251329421997,
|
|
"epoch": 3.1213389121338913,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004344865437787027,
|
|
"loss": 4.7729,
|
|
"mean_token_accuracy": 0.20841041952371597,
|
|
"num_tokens": 6898082.0,
|
|
"step": 3730
|
|
},
|
|
{
|
|
"entropy": 4.990958833694458,
|
|
"epoch": 3.1255230125523012,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00043425870672860774,
|
|
"loss": 4.7493,
|
|
"mean_token_accuracy": 0.20974217653274535,
|
|
"num_tokens": 6907693.0,
|
|
"step": 3735
|
|
},
|
|
{
|
|
"entropy": 5.2403295040130615,
|
|
"epoch": 3.1297071129707112,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004340305419496989,
|
|
"loss": 4.9496,
|
|
"mean_token_accuracy": 0.19715171456336975,
|
|
"num_tokens": 6917095.0,
|
|
"step": 3740
|
|
},
|
|
{
|
|
"entropy": 5.15607762336731,
|
|
"epoch": 3.1338912133891212,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004338020499115029,
|
|
"loss": 4.8805,
|
|
"mean_token_accuracy": 0.2006925404071808,
|
|
"num_tokens": 6926952.0,
|
|
"step": 3745
|
|
},
|
|
{
|
|
"entropy": 5.082753658294678,
|
|
"epoch": 3.1380753138075312,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004335732310842196,
|
|
"loss": 4.8244,
|
|
"mean_token_accuracy": 0.2041364461183548,
|
|
"num_tokens": 6936704.0,
|
|
"step": 3750
|
|
},
|
|
{
|
|
"entropy": 5.167978429794312,
|
|
"epoch": 3.1422594142259412,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004333440859387212,
|
|
"loss": 4.8359,
|
|
"mean_token_accuracy": 0.2067200645804405,
|
|
"num_tokens": 6946614.0,
|
|
"step": 3755
|
|
},
|
|
{
|
|
"entropy": 5.161498022079468,
|
|
"epoch": 3.1464435146443517,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00043311461494655165,
|
|
"loss": 4.8197,
|
|
"mean_token_accuracy": 0.21429964900016785,
|
|
"num_tokens": 6955980.0,
|
|
"step": 3760
|
|
},
|
|
{
|
|
"entropy": 5.109651279449463,
|
|
"epoch": 3.1506276150627617,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004328848185799254,
|
|
"loss": 4.8281,
|
|
"mean_token_accuracy": 0.20562287420034409,
|
|
"num_tokens": 6964108.0,
|
|
"step": 3765
|
|
},
|
|
{
|
|
"entropy": 5.128606367111206,
|
|
"epoch": 3.1548117154811717,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004326546973117264,
|
|
"loss": 4.8102,
|
|
"mean_token_accuracy": 0.21551830768585206,
|
|
"num_tokens": 6972895.0,
|
|
"step": 3770
|
|
},
|
|
{
|
|
"entropy": 5.06142783164978,
|
|
"epoch": 3.1589958158995817,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00043242425161550723,
|
|
"loss": 4.8289,
|
|
"mean_token_accuracy": 0.20230160504579545,
|
|
"num_tokens": 6982222.0,
|
|
"step": 3775
|
|
},
|
|
{
|
|
"entropy": 5.202808284759522,
|
|
"epoch": 3.1631799163179917,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043219348196548807,
|
|
"loss": 4.825,
|
|
"mean_token_accuracy": 0.20703910291194916,
|
|
"num_tokens": 6990820.0,
|
|
"step": 3780
|
|
},
|
|
{
|
|
"entropy": 5.123299598693848,
|
|
"epoch": 3.1673640167364017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004319623888365557,
|
|
"loss": 4.8516,
|
|
"mean_token_accuracy": 0.20734433680772782,
|
|
"num_tokens": 7000846.0,
|
|
"step": 3785
|
|
},
|
|
{
|
|
"entropy": 5.049972152709961,
|
|
"epoch": 3.1715481171548117,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00043173097270426275,
|
|
"loss": 4.7947,
|
|
"mean_token_accuracy": 0.20025773495435714,
|
|
"num_tokens": 7009086.0,
|
|
"step": 3790
|
|
},
|
|
{
|
|
"entropy": 5.211647653579712,
|
|
"epoch": 3.1757322175732217,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00043149923404482634,
|
|
"loss": 4.8521,
|
|
"mean_token_accuracy": 0.2014863073825836,
|
|
"num_tokens": 7017752.0,
|
|
"step": 3795
|
|
},
|
|
{
|
|
"entropy": 5.08097882270813,
|
|
"epoch": 3.1799163179916317,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00043126717333512733,
|
|
"loss": 4.8387,
|
|
"mean_token_accuracy": 0.2177015706896782,
|
|
"num_tokens": 7026535.0,
|
|
"step": 3800
|
|
},
|
|
{
|
|
"entropy": 5.128823947906494,
|
|
"epoch": 3.1841004184100417,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004310347910527094,
|
|
"loss": 4.8763,
|
|
"mean_token_accuracy": 0.20947964638471603,
|
|
"num_tokens": 7035462.0,
|
|
"step": 3805
|
|
},
|
|
{
|
|
"entropy": 5.128912544250488,
|
|
"epoch": 3.1882845188284517,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004308020876757778,
|
|
"loss": 4.7777,
|
|
"mean_token_accuracy": 0.21443202644586562,
|
|
"num_tokens": 7043978.0,
|
|
"step": 3810
|
|
},
|
|
{
|
|
"entropy": 5.0324770450592045,
|
|
"epoch": 3.192468619246862,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000430569063683199,
|
|
"loss": 4.7884,
|
|
"mean_token_accuracy": 0.21533980071544648,
|
|
"num_tokens": 7053757.0,
|
|
"step": 3815
|
|
},
|
|
{
|
|
"entropy": 5.01102614402771,
|
|
"epoch": 3.196652719665272,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004303357195544986,
|
|
"loss": 4.7331,
|
|
"mean_token_accuracy": 0.20302986055612565,
|
|
"num_tokens": 7062491.0,
|
|
"step": 3820
|
|
},
|
|
{
|
|
"entropy": 5.074666452407837,
|
|
"epoch": 3.200836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004301020557698615,
|
|
"loss": 4.777,
|
|
"mean_token_accuracy": 0.21599315851926804,
|
|
"num_tokens": 7072157.0,
|
|
"step": 3825
|
|
},
|
|
{
|
|
"entropy": 5.031312131881714,
|
|
"epoch": 3.205020920502092,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00042986807281013013,
|
|
"loss": 4.7469,
|
|
"mean_token_accuracy": 0.2086035668849945,
|
|
"num_tokens": 7080631.0,
|
|
"step": 3830
|
|
},
|
|
{
|
|
"entropy": 5.139399814605713,
|
|
"epoch": 3.209205020920502,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004296337711568037,
|
|
"loss": 4.8141,
|
|
"mean_token_accuracy": 0.21343361735343933,
|
|
"num_tokens": 7089944.0,
|
|
"step": 3835
|
|
},
|
|
{
|
|
"entropy": 5.104015207290649,
|
|
"epoch": 3.213389121338912,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042939915129203745,
|
|
"loss": 4.8667,
|
|
"mean_token_accuracy": 0.205406454205513,
|
|
"num_tokens": 7099190.0,
|
|
"step": 3840
|
|
},
|
|
{
|
|
"entropy": 4.993949460983276,
|
|
"epoch": 3.217573221757322,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004291642136986413,
|
|
"loss": 4.8324,
|
|
"mean_token_accuracy": 0.201272714138031,
|
|
"num_tokens": 7107944.0,
|
|
"step": 3845
|
|
},
|
|
{
|
|
"entropy": 5.050441884994507,
|
|
"epoch": 3.221757322175732,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000428928958860079,
|
|
"loss": 4.6937,
|
|
"mean_token_accuracy": 0.2107575535774231,
|
|
"num_tokens": 7116953.0,
|
|
"step": 3850
|
|
},
|
|
{
|
|
"entropy": 5.019984340667724,
|
|
"epoch": 3.225941422594142,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004286933872604673,
|
|
"loss": 4.7471,
|
|
"mean_token_accuracy": 0.21281925290822984,
|
|
"num_tokens": 7126236.0,
|
|
"step": 3855
|
|
},
|
|
{
|
|
"entropy": 5.087703037261963,
|
|
"epoch": 3.2301255230125525,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0004284574993845746,
|
|
"loss": 4.796,
|
|
"mean_token_accuracy": 0.2036880612373352,
|
|
"num_tokens": 7136286.0,
|
|
"step": 3860
|
|
},
|
|
{
|
|
"entropy": 5.139914608001709,
|
|
"epoch": 3.2343096234309625,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00042822129571782025,
|
|
"loss": 4.8893,
|
|
"mean_token_accuracy": 0.2061291828751564,
|
|
"num_tokens": 7145551.0,
|
|
"step": 3865
|
|
},
|
|
{
|
|
"entropy": 5.14477686882019,
|
|
"epoch": 3.2384937238493725,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00042798477674627343,
|
|
"loss": 4.8104,
|
|
"mean_token_accuracy": 0.2095671236515045,
|
|
"num_tokens": 7155486.0,
|
|
"step": 3870
|
|
},
|
|
{
|
|
"entropy": 5.092449808120728,
|
|
"epoch": 3.2426778242677825,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042774794295665213,
|
|
"loss": 4.8111,
|
|
"mean_token_accuracy": 0.20950937569141387,
|
|
"num_tokens": 7163984.0,
|
|
"step": 3875
|
|
},
|
|
{
|
|
"entropy": 5.053369474411011,
|
|
"epoch": 3.2468619246861925,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004275107948363223,
|
|
"loss": 4.7864,
|
|
"mean_token_accuracy": 0.20992782562971116,
|
|
"num_tokens": 7172854.0,
|
|
"step": 3880
|
|
},
|
|
{
|
|
"entropy": 5.0084789276123045,
|
|
"epoch": 3.2510460251046025,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004272733328732965,
|
|
"loss": 4.7736,
|
|
"mean_token_accuracy": 0.20957640409469605,
|
|
"num_tokens": 7182202.0,
|
|
"step": 3885
|
|
},
|
|
{
|
|
"entropy": 5.114617490768433,
|
|
"epoch": 3.2552301255230125,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042703555755623337,
|
|
"loss": 4.9001,
|
|
"mean_token_accuracy": 0.20189402848482133,
|
|
"num_tokens": 7192062.0,
|
|
"step": 3890
|
|
},
|
|
{
|
|
"entropy": 5.2213215827941895,
|
|
"epoch": 3.2594142259414225,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004267974693744362,
|
|
"loss": 4.9003,
|
|
"mean_token_accuracy": 0.20662265866994858,
|
|
"num_tokens": 7201766.0,
|
|
"step": 3895
|
|
},
|
|
{
|
|
"entropy": 5.102724409103393,
|
|
"epoch": 3.2635983263598325,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004265590688178523,
|
|
"loss": 4.8171,
|
|
"mean_token_accuracy": 0.20365506261587143,
|
|
"num_tokens": 7211361.0,
|
|
"step": 3900
|
|
},
|
|
{
|
|
"entropy": 5.059186172485352,
|
|
"epoch": 3.2677824267782425,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004263203563770716,
|
|
"loss": 4.7952,
|
|
"mean_token_accuracy": 0.21072521954774856,
|
|
"num_tokens": 7220221.0,
|
|
"step": 3905
|
|
},
|
|
{
|
|
"entropy": 5.0689839839935305,
|
|
"epoch": 3.2719665271966525,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000426081332543326,
|
|
"loss": 4.9079,
|
|
"mean_token_accuracy": 0.20144714713096618,
|
|
"num_tokens": 7230134.0,
|
|
"step": 3910
|
|
},
|
|
{
|
|
"entropy": 5.054864883422852,
|
|
"epoch": 3.276150627615063,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004258419978084881,
|
|
"loss": 4.7363,
|
|
"mean_token_accuracy": 0.21276534497737884,
|
|
"num_tokens": 7238713.0,
|
|
"step": 3915
|
|
},
|
|
{
|
|
"entropy": 5.15344648361206,
|
|
"epoch": 3.280334728033473,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004256023526650702,
|
|
"loss": 4.8408,
|
|
"mean_token_accuracy": 0.21071268022060394,
|
|
"num_tokens": 7247603.0,
|
|
"step": 3920
|
|
},
|
|
{
|
|
"entropy": 5.0383727073669435,
|
|
"epoch": 3.284518828451883,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004253623976062236,
|
|
"loss": 4.8706,
|
|
"mean_token_accuracy": 0.20507752448320388,
|
|
"num_tokens": 7257452.0,
|
|
"step": 3925
|
|
},
|
|
{
|
|
"entropy": 5.173134851455688,
|
|
"epoch": 3.288702928870293,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004251221331257372,
|
|
"loss": 4.8753,
|
|
"mean_token_accuracy": 0.20658525079488754,
|
|
"num_tokens": 7266401.0,
|
|
"step": 3930
|
|
},
|
|
{
|
|
"entropy": 5.100352716445923,
|
|
"epoch": 3.292887029288703,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004248815597180366,
|
|
"loss": 4.8202,
|
|
"mean_token_accuracy": 0.21314020305871964,
|
|
"num_tokens": 7275836.0,
|
|
"step": 3935
|
|
},
|
|
{
|
|
"entropy": 5.121883535385132,
|
|
"epoch": 3.297071129707113,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00042464067787818333,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.20382918119430543,
|
|
"num_tokens": 7284854.0,
|
|
"step": 3940
|
|
},
|
|
{
|
|
"entropy": 5.0534083366394045,
|
|
"epoch": 3.301255230125523,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004243994881018735,
|
|
"loss": 4.8283,
|
|
"mean_token_accuracy": 0.2006744235754013,
|
|
"num_tokens": 7293421.0,
|
|
"step": 3945
|
|
},
|
|
{
|
|
"entropy": 5.053618288040161,
|
|
"epoch": 3.305439330543933,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004241579908854367,
|
|
"loss": 4.7635,
|
|
"mean_token_accuracy": 0.2105734646320343,
|
|
"num_tokens": 7302184.0,
|
|
"step": 3950
|
|
},
|
|
{
|
|
"entropy": 5.153102922439575,
|
|
"epoch": 3.309623430962343,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00042391618672583546,
|
|
"loss": 4.9227,
|
|
"mean_token_accuracy": 0.19989618957042693,
|
|
"num_tokens": 7311596.0,
|
|
"step": 3955
|
|
},
|
|
{
|
|
"entropy": 5.11073637008667,
|
|
"epoch": 3.3138075313807533,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00042367407612066383,
|
|
"loss": 4.7787,
|
|
"mean_token_accuracy": 0.20876786708831788,
|
|
"num_tokens": 7319866.0,
|
|
"step": 3960
|
|
},
|
|
{
|
|
"entropy": 5.097158002853393,
|
|
"epoch": 3.3179916317991633,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004234316595681468,
|
|
"loss": 4.83,
|
|
"mean_token_accuracy": 0.206571127474308,
|
|
"num_tokens": 7329327.0,
|
|
"step": 3965
|
|
},
|
|
{
|
|
"entropy": 5.203888273239135,
|
|
"epoch": 3.3221757322175733,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004231889375671382,
|
|
"loss": 4.9789,
|
|
"mean_token_accuracy": 0.19189484268426896,
|
|
"num_tokens": 7339717.0,
|
|
"step": 3970
|
|
},
|
|
{
|
|
"entropy": 5.143088006973267,
|
|
"epoch": 3.3263598326359833,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00042294591061712124,
|
|
"loss": 4.8241,
|
|
"mean_token_accuracy": 0.20905834883451463,
|
|
"num_tokens": 7348850.0,
|
|
"step": 3975
|
|
},
|
|
{
|
|
"entropy": 5.06374716758728,
|
|
"epoch": 3.3305439330543933,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042270257921820623,
|
|
"loss": 4.7578,
|
|
"mean_token_accuracy": 0.2122231051325798,
|
|
"num_tokens": 7357081.0,
|
|
"step": 3980
|
|
},
|
|
{
|
|
"entropy": 4.939764070510864,
|
|
"epoch": 3.3347280334728033,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00042245894387112987,
|
|
"loss": 4.7788,
|
|
"mean_token_accuracy": 0.21034014225006104,
|
|
"num_tokens": 7365762.0,
|
|
"step": 3985
|
|
},
|
|
{
|
|
"entropy": 5.099068021774292,
|
|
"epoch": 3.3389121338912133,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004222150050772548,
|
|
"loss": 4.8426,
|
|
"mean_token_accuracy": 0.20577617585659028,
|
|
"num_tokens": 7374870.0,
|
|
"step": 3990
|
|
},
|
|
{
|
|
"entropy": 5.078297710418701,
|
|
"epoch": 3.3430962343096233,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004219707633385677,
|
|
"loss": 4.7408,
|
|
"mean_token_accuracy": 0.21203900277614593,
|
|
"num_tokens": 7383836.0,
|
|
"step": 3995
|
|
},
|
|
{
|
|
"entropy": 5.019320487976074,
|
|
"epoch": 3.3472803347280333,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042172621915767886,
|
|
"loss": 4.8191,
|
|
"mean_token_accuracy": 0.20933733135461807,
|
|
"num_tokens": 7394059.0,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"epoch": 3.3472803347280333,
|
|
"eval_entropy": 4.918908658934517,
|
|
"eval_loss": 5.45020055770874,
|
|
"eval_mean_token_accuracy": 0.18719054693032383,
|
|
"eval_num_tokens": 7394059.0,
|
|
"eval_runtime": 2.7397,
|
|
"eval_samples_per_second": 1426.087,
|
|
"eval_steps_per_second": 178.489,
|
|
"step": 4000
|
|
},
|
|
{
|
|
"entropy": 5.026743650436401,
|
|
"epoch": 3.3514644351464433,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00042148137303782077,
|
|
"loss": 4.7846,
|
|
"mean_token_accuracy": 0.21195760369300842,
|
|
"num_tokens": 7402947.0,
|
|
"step": 4005
|
|
},
|
|
{
|
|
"entropy": 5.113661575317383,
|
|
"epoch": 3.3556485355648533,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004212362254828475,
|
|
"loss": 4.8525,
|
|
"mean_token_accuracy": 0.20901040285825728,
|
|
"num_tokens": 7413430.0,
|
|
"step": 4010
|
|
},
|
|
{
|
|
"entropy": 5.15121865272522,
|
|
"epoch": 3.3598326359832638,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00042099077699723326,
|
|
"loss": 4.8662,
|
|
"mean_token_accuracy": 0.20262160003185273,
|
|
"num_tokens": 7423239.0,
|
|
"step": 4015
|
|
},
|
|
{
|
|
"entropy": 5.116614866256714,
|
|
"epoch": 3.3640167364016738,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004207450280860716,
|
|
"loss": 4.8382,
|
|
"mean_token_accuracy": 0.20936330854892732,
|
|
"num_tokens": 7432519.0,
|
|
"step": 4020
|
|
},
|
|
{
|
|
"entropy": 5.016262483596802,
|
|
"epoch": 3.3682008368200838,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004204989792550742,
|
|
"loss": 4.8089,
|
|
"mean_token_accuracy": 0.20921995788812636,
|
|
"num_tokens": 7442089.0,
|
|
"step": 4025
|
|
},
|
|
{
|
|
"entropy": 5.077898120880127,
|
|
"epoch": 3.3723849372384938,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00042025263101057016,
|
|
"loss": 4.8183,
|
|
"mean_token_accuracy": 0.20580985099077226,
|
|
"num_tokens": 7451059.0,
|
|
"step": 4030
|
|
},
|
|
{
|
|
"entropy": 5.093995189666748,
|
|
"epoch": 3.3765690376569037,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00042000598385950427,
|
|
"loss": 4.7766,
|
|
"mean_token_accuracy": 0.20833854079246522,
|
|
"num_tokens": 7458960.0,
|
|
"step": 4035
|
|
},
|
|
{
|
|
"entropy": 5.083715963363647,
|
|
"epoch": 3.3807531380753137,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000419759038309437,
|
|
"loss": 4.9734,
|
|
"mean_token_accuracy": 0.19399410784244536,
|
|
"num_tokens": 7468863.0,
|
|
"step": 4040
|
|
},
|
|
{
|
|
"entropy": 5.115178108215332,
|
|
"epoch": 3.3849372384937237,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00041951179486854253,
|
|
"loss": 4.9201,
|
|
"mean_token_accuracy": 0.20426041036844253,
|
|
"num_tokens": 7477204.0,
|
|
"step": 4045
|
|
},
|
|
{
|
|
"entropy": 5.095623874664307,
|
|
"epoch": 3.3891213389121337,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000419264254045608,
|
|
"loss": 4.745,
|
|
"mean_token_accuracy": 0.20784950852394105,
|
|
"num_tokens": 7487367.0,
|
|
"step": 4050
|
|
},
|
|
{
|
|
"entropy": 4.9586732387542725,
|
|
"epoch": 3.393305439330544,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004190164163500326,
|
|
"loss": 4.7298,
|
|
"mean_token_accuracy": 0.21990751773118972,
|
|
"num_tokens": 7495808.0,
|
|
"step": 4055
|
|
},
|
|
{
|
|
"entropy": 5.054813003540039,
|
|
"epoch": 3.397489539748954,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004187682822918266,
|
|
"loss": 4.8847,
|
|
"mean_token_accuracy": 0.19259270876646042,
|
|
"num_tokens": 7506119.0,
|
|
"step": 4060
|
|
},
|
|
{
|
|
"entropy": 5.079545164108277,
|
|
"epoch": 3.401673640167364,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00041851985238161,
|
|
"loss": 4.8558,
|
|
"mean_token_accuracy": 0.21103399246931076,
|
|
"num_tokens": 7514931.0,
|
|
"step": 4065
|
|
},
|
|
{
|
|
"entropy": 5.005889892578125,
|
|
"epoch": 3.405857740585774,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00041827112713061155,
|
|
"loss": 4.6473,
|
|
"mean_token_accuracy": 0.22515798062086106,
|
|
"num_tokens": 7523753.0,
|
|
"step": 4070
|
|
},
|
|
{
|
|
"entropy": 5.045839500427246,
|
|
"epoch": 3.410041841004184,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00041802210705066775,
|
|
"loss": 4.7358,
|
|
"mean_token_accuracy": 0.2144197940826416,
|
|
"num_tokens": 7532447.0,
|
|
"step": 4075
|
|
},
|
|
{
|
|
"entropy": 5.026084136962891,
|
|
"epoch": 3.414225941422594,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000417772792654222,
|
|
"loss": 4.8939,
|
|
"mean_token_accuracy": 0.19934459924697875,
|
|
"num_tokens": 7541388.0,
|
|
"step": 4080
|
|
},
|
|
{
|
|
"entropy": 5.10627760887146,
|
|
"epoch": 3.418410041841004,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00041752318445432323,
|
|
"loss": 4.8946,
|
|
"mean_token_accuracy": 0.19759735763072966,
|
|
"num_tokens": 7550422.0,
|
|
"step": 4085
|
|
},
|
|
{
|
|
"entropy": 5.081887817382812,
|
|
"epoch": 3.422594142259414,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000417273282964625,
|
|
"loss": 4.7504,
|
|
"mean_token_accuracy": 0.21667537093162537,
|
|
"num_tokens": 7559522.0,
|
|
"step": 4090
|
|
},
|
|
{
|
|
"entropy": 5.101574993133545,
|
|
"epoch": 3.426778242677824,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004170230886993844,
|
|
"loss": 4.881,
|
|
"mean_token_accuracy": 0.19691682010889053,
|
|
"num_tokens": 7568785.0,
|
|
"step": 4095
|
|
},
|
|
{
|
|
"entropy": 5.041030550003052,
|
|
"epoch": 3.430962343096234,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041677260217346093,
|
|
"loss": 4.7897,
|
|
"mean_token_accuracy": 0.20719113498926162,
|
|
"num_tokens": 7579013.0,
|
|
"step": 4100
|
|
},
|
|
{
|
|
"entropy": 5.075591659545898,
|
|
"epoch": 3.435146443514644,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00041652182390231556,
|
|
"loss": 4.8246,
|
|
"mean_token_accuracy": 0.207845076918602,
|
|
"num_tokens": 7588577.0,
|
|
"step": 4105
|
|
},
|
|
{
|
|
"entropy": 5.070200443267822,
|
|
"epoch": 3.439330543933054,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00041627075440200977,
|
|
"loss": 4.7545,
|
|
"mean_token_accuracy": 0.21534578055143355,
|
|
"num_tokens": 7597702.0,
|
|
"step": 4110
|
|
},
|
|
{
|
|
"entropy": 5.003120088577271,
|
|
"epoch": 3.4435146443514646,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004160193941892043,
|
|
"loss": 4.8009,
|
|
"mean_token_accuracy": 0.21220875382423401,
|
|
"num_tokens": 7607156.0,
|
|
"step": 4115
|
|
},
|
|
{
|
|
"entropy": 5.070883941650391,
|
|
"epoch": 3.4476987447698746,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004157677437811578,
|
|
"loss": 4.8947,
|
|
"mean_token_accuracy": 0.2164589583873749,
|
|
"num_tokens": 7615936.0,
|
|
"step": 4120
|
|
},
|
|
{
|
|
"entropy": 5.099539279937744,
|
|
"epoch": 3.4518828451882846,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004155158036957265,
|
|
"loss": 4.8704,
|
|
"mean_token_accuracy": 0.2049857795238495,
|
|
"num_tokens": 7626568.0,
|
|
"step": 4125
|
|
},
|
|
{
|
|
"entropy": 5.087817811965943,
|
|
"epoch": 3.4560669456066946,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004152635744513626,
|
|
"loss": 4.7877,
|
|
"mean_token_accuracy": 0.21546396017074584,
|
|
"num_tokens": 7635102.0,
|
|
"step": 4130
|
|
},
|
|
{
|
|
"entropy": 5.043559598922729,
|
|
"epoch": 3.4602510460251046,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004150110565671133,
|
|
"loss": 4.7344,
|
|
"mean_token_accuracy": 0.21734167039394378,
|
|
"num_tokens": 7644204.0,
|
|
"step": 4135
|
|
},
|
|
{
|
|
"entropy": 5.085167598724365,
|
|
"epoch": 3.4644351464435146,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00041475825056261985,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.20746599435806273,
|
|
"num_tokens": 7653585.0,
|
|
"step": 4140
|
|
},
|
|
{
|
|
"entropy": 5.074635410308838,
|
|
"epoch": 3.4686192468619246,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004145051569581162,
|
|
"loss": 4.8527,
|
|
"mean_token_accuracy": 0.20142358541488647,
|
|
"num_tokens": 7662734.0,
|
|
"step": 4145
|
|
},
|
|
{
|
|
"entropy": 5.157631778717041,
|
|
"epoch": 3.4728033472803346,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004142517762744286,
|
|
"loss": 4.8016,
|
|
"mean_token_accuracy": 0.20872812271118163,
|
|
"num_tokens": 7671682.0,
|
|
"step": 4150
|
|
},
|
|
{
|
|
"entropy": 5.112520694732666,
|
|
"epoch": 3.476987447698745,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004139981090329735,
|
|
"loss": 4.9159,
|
|
"mean_token_accuracy": 0.20473155975341797,
|
|
"num_tokens": 7681551.0,
|
|
"step": 4155
|
|
},
|
|
{
|
|
"entropy": 5.103758525848389,
|
|
"epoch": 3.481171548117155,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004137441557557575,
|
|
"loss": 4.8648,
|
|
"mean_token_accuracy": 0.2094658225774765,
|
|
"num_tokens": 7690943.0,
|
|
"step": 4160
|
|
},
|
|
{
|
|
"entropy": 5.034735584259034,
|
|
"epoch": 3.485355648535565,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00041348991696537533,
|
|
"loss": 4.7521,
|
|
"mean_token_accuracy": 0.2184034615755081,
|
|
"num_tokens": 7700376.0,
|
|
"step": 4165
|
|
},
|
|
{
|
|
"entropy": 5.14447054862976,
|
|
"epoch": 3.489539748953975,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004132353931850098,
|
|
"loss": 4.8528,
|
|
"mean_token_accuracy": 0.20408765375614166,
|
|
"num_tokens": 7709845.0,
|
|
"step": 4170
|
|
},
|
|
{
|
|
"entropy": 5.182646036148071,
|
|
"epoch": 3.493723849372385,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004129805849384299,
|
|
"loss": 4.9354,
|
|
"mean_token_accuracy": 0.20410424917936326,
|
|
"num_tokens": 7719908.0,
|
|
"step": 4175
|
|
},
|
|
{
|
|
"entropy": 5.024223184585571,
|
|
"epoch": 3.497907949790795,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004127254927499898,
|
|
"loss": 4.8743,
|
|
"mean_token_accuracy": 0.20478571057319642,
|
|
"num_tokens": 7729543.0,
|
|
"step": 4180
|
|
},
|
|
{
|
|
"entropy": 5.127690839767456,
|
|
"epoch": 3.502092050209205,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004124701171446286,
|
|
"loss": 4.8852,
|
|
"mean_token_accuracy": 0.2095595195889473,
|
|
"num_tokens": 7739371.0,
|
|
"step": 4185
|
|
},
|
|
{
|
|
"entropy": 5.078702688217163,
|
|
"epoch": 3.506276150627615,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00041221445864786794,
|
|
"loss": 4.8006,
|
|
"mean_token_accuracy": 0.20972957611083984,
|
|
"num_tokens": 7748027.0,
|
|
"step": 4190
|
|
},
|
|
{
|
|
"entropy": 4.986270380020142,
|
|
"epoch": 3.510460251046025,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000411958517785812,
|
|
"loss": 4.7529,
|
|
"mean_token_accuracy": 0.21445594727993011,
|
|
"num_tokens": 7756588.0,
|
|
"step": 4195
|
|
},
|
|
{
|
|
"entropy": 5.06249098777771,
|
|
"epoch": 3.514644351464435,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00041170229508514584,
|
|
"loss": 4.7225,
|
|
"mean_token_accuracy": 0.21254072934389115,
|
|
"num_tokens": 7765222.0,
|
|
"step": 4200
|
|
},
|
|
{
|
|
"entropy": 5.075818109512329,
|
|
"epoch": 3.518828451882845,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004114457910731347,
|
|
"loss": 4.8336,
|
|
"mean_token_accuracy": 0.21234757602214813,
|
|
"num_tokens": 7774139.0,
|
|
"step": 4205
|
|
},
|
|
{
|
|
"entropy": 5.0034605979919435,
|
|
"epoch": 3.523012552301255,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004111890062776226,
|
|
"loss": 4.7204,
|
|
"mean_token_accuracy": 0.2159600406885147,
|
|
"num_tokens": 7783597.0,
|
|
"step": 4210
|
|
},
|
|
{
|
|
"entropy": 4.991421365737915,
|
|
"epoch": 3.5271966527196654,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004109319412270314,
|
|
"loss": 4.8036,
|
|
"mean_token_accuracy": 0.21447610408067702,
|
|
"num_tokens": 7792711.0,
|
|
"step": 4215
|
|
},
|
|
{
|
|
"entropy": 5.11831750869751,
|
|
"epoch": 3.5313807531380754,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0004106745964503594,
|
|
"loss": 4.871,
|
|
"mean_token_accuracy": 0.20587026476860046,
|
|
"num_tokens": 7801878.0,
|
|
"step": 4220
|
|
},
|
|
{
|
|
"entropy": 5.0122026920318605,
|
|
"epoch": 3.5355648535564854,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000410416972477181,
|
|
"loss": 4.7707,
|
|
"mean_token_accuracy": 0.2146313339471817,
|
|
"num_tokens": 7810484.0,
|
|
"step": 4225
|
|
},
|
|
{
|
|
"entropy": 5.067341995239258,
|
|
"epoch": 3.5397489539748954,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00041015906983764495,
|
|
"loss": 4.8682,
|
|
"mean_token_accuracy": 0.2098026379942894,
|
|
"num_tokens": 7820521.0,
|
|
"step": 4230
|
|
},
|
|
{
|
|
"entropy": 5.108982181549072,
|
|
"epoch": 3.5439330543933054,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040990088906247326,
|
|
"loss": 4.8025,
|
|
"mean_token_accuracy": 0.21214094310998916,
|
|
"num_tokens": 7829218.0,
|
|
"step": 4235
|
|
},
|
|
{
|
|
"entropy": 5.057541418075561,
|
|
"epoch": 3.5481171548117154,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004096424306829605,
|
|
"loss": 4.8334,
|
|
"mean_token_accuracy": 0.20323596745729447,
|
|
"num_tokens": 7837795.0,
|
|
"step": 4240
|
|
},
|
|
{
|
|
"entropy": 5.06811637878418,
|
|
"epoch": 3.5523012552301254,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004093836952309726,
|
|
"loss": 4.8287,
|
|
"mean_token_accuracy": 0.20759256184101105,
|
|
"num_tokens": 7846973.0,
|
|
"step": 4245
|
|
},
|
|
{
|
|
"entropy": 5.1008216381073,
|
|
"epoch": 3.5564853556485354,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004091246832389453,
|
|
"loss": 4.8977,
|
|
"mean_token_accuracy": 0.2029066264629364,
|
|
"num_tokens": 7856627.0,
|
|
"step": 4250
|
|
},
|
|
{
|
|
"entropy": 5.057607984542846,
|
|
"epoch": 3.560669456066946,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004088653952398837,
|
|
"loss": 4.7982,
|
|
"mean_token_accuracy": 0.2106788218021393,
|
|
"num_tokens": 7866308.0,
|
|
"step": 4255
|
|
},
|
|
{
|
|
"entropy": 5.021573734283447,
|
|
"epoch": 3.564853556485356,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00040860583176736097,
|
|
"loss": 4.7695,
|
|
"mean_token_accuracy": 0.21330823749303818,
|
|
"num_tokens": 7875350.0,
|
|
"step": 4260
|
|
},
|
|
{
|
|
"entropy": 5.03112826347351,
|
|
"epoch": 3.569037656903766,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004083459933555169,
|
|
"loss": 4.7888,
|
|
"mean_token_accuracy": 0.21724218279123306,
|
|
"num_tokens": 7884276.0,
|
|
"step": 4265
|
|
},
|
|
{
|
|
"entropy": 5.059065151214599,
|
|
"epoch": 3.573221757322176,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004080858805390571,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.20559176057577133,
|
|
"num_tokens": 7893836.0,
|
|
"step": 4270
|
|
},
|
|
{
|
|
"entropy": 5.114040422439575,
|
|
"epoch": 3.577405857740586,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000407825493853252,
|
|
"loss": 4.8342,
|
|
"mean_token_accuracy": 0.21475007832050325,
|
|
"num_tokens": 7903249.0,
|
|
"step": 4275
|
|
},
|
|
{
|
|
"entropy": 5.092215299606323,
|
|
"epoch": 3.581589958158996,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040756483383393556,
|
|
"loss": 4.8945,
|
|
"mean_token_accuracy": 0.2081194445490837,
|
|
"num_tokens": 7912217.0,
|
|
"step": 4280
|
|
},
|
|
{
|
|
"entropy": 5.05587911605835,
|
|
"epoch": 3.585774058577406,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004073039010175041,
|
|
"loss": 4.9195,
|
|
"mean_token_accuracy": 0.20009565055370332,
|
|
"num_tokens": 7922241.0,
|
|
"step": 4285
|
|
},
|
|
{
|
|
"entropy": 5.128195905685425,
|
|
"epoch": 3.589958158995816,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00040704269594091556,
|
|
"loss": 4.8382,
|
|
"mean_token_accuracy": 0.209613598883152,
|
|
"num_tokens": 7931866.0,
|
|
"step": 4290
|
|
},
|
|
{
|
|
"entropy": 5.112419748306275,
|
|
"epoch": 3.594142259414226,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004067812191416878,
|
|
"loss": 4.8695,
|
|
"mean_token_accuracy": 0.20142448842525482,
|
|
"num_tokens": 7941787.0,
|
|
"step": 4295
|
|
},
|
|
{
|
|
"entropy": 5.121737432479859,
|
|
"epoch": 3.598326359832636,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004065194711578982,
|
|
"loss": 4.8798,
|
|
"mean_token_accuracy": 0.20556933879852296,
|
|
"num_tokens": 7950832.0,
|
|
"step": 4300
|
|
},
|
|
{
|
|
"entropy": 5.109773588180542,
|
|
"epoch": 3.602510460251046,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004062574525281819,
|
|
"loss": 4.8635,
|
|
"mean_token_accuracy": 0.2077685624361038,
|
|
"num_tokens": 7960059.0,
|
|
"step": 4305
|
|
},
|
|
{
|
|
"entropy": 5.0760462284088135,
|
|
"epoch": 3.606694560669456,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00040599516379173113,
|
|
"loss": 4.848,
|
|
"mean_token_accuracy": 0.20024775117635726,
|
|
"num_tokens": 7969411.0,
|
|
"step": 4310
|
|
},
|
|
{
|
|
"entropy": 5.063443183898926,
|
|
"epoch": 3.6108786610878663,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00040573260548829417,
|
|
"loss": 4.8413,
|
|
"mean_token_accuracy": 0.20782430768013,
|
|
"num_tokens": 7978446.0,
|
|
"step": 4315
|
|
},
|
|
{
|
|
"entropy": 4.962226295471192,
|
|
"epoch": 3.6150627615062763,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004054697781581734,
|
|
"loss": 4.7319,
|
|
"mean_token_accuracy": 0.2118672251701355,
|
|
"num_tokens": 7987117.0,
|
|
"step": 4320
|
|
},
|
|
{
|
|
"entropy": 5.088425016403198,
|
|
"epoch": 3.6192468619246863,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004052066823422255,
|
|
"loss": 4.8356,
|
|
"mean_token_accuracy": 0.20978725403547288,
|
|
"num_tokens": 7995989.0,
|
|
"step": 4325
|
|
},
|
|
{
|
|
"entropy": 5.049139595031738,
|
|
"epoch": 3.6234309623430963,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004049433185818593,
|
|
"loss": 4.8906,
|
|
"mean_token_accuracy": 0.20367469936609267,
|
|
"num_tokens": 8005486.0,
|
|
"step": 4330
|
|
},
|
|
{
|
|
"entropy": 5.0111713886260985,
|
|
"epoch": 3.6276150627615062,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040467968741903507,
|
|
"loss": 4.7642,
|
|
"mean_token_accuracy": 0.20974033772945405,
|
|
"num_tokens": 8014853.0,
|
|
"step": 4335
|
|
},
|
|
{
|
|
"entropy": 5.084183549880981,
|
|
"epoch": 3.6317991631799162,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004044157893962633,
|
|
"loss": 4.8059,
|
|
"mean_token_accuracy": 0.209882690012455,
|
|
"num_tokens": 8023931.0,
|
|
"step": 4340
|
|
},
|
|
{
|
|
"entropy": 5.049024868011474,
|
|
"epoch": 3.6359832635983262,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004041516250566037,
|
|
"loss": 4.85,
|
|
"mean_token_accuracy": 0.2115127697587013,
|
|
"num_tokens": 8032659.0,
|
|
"step": 4345
|
|
},
|
|
{
|
|
"entropy": 5.075996732711792,
|
|
"epoch": 3.6401673640167362,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040388719494366404,
|
|
"loss": 4.9155,
|
|
"mean_token_accuracy": 0.2042904168367386,
|
|
"num_tokens": 8041857.0,
|
|
"step": 4350
|
|
},
|
|
{
|
|
"entropy": 5.170190048217774,
|
|
"epoch": 3.6443514644351467,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00040362249960159904,
|
|
"loss": 4.9435,
|
|
"mean_token_accuracy": 0.20341576784849166,
|
|
"num_tokens": 8051186.0,
|
|
"step": 4355
|
|
},
|
|
{
|
|
"entropy": 5.050500440597534,
|
|
"epoch": 3.6485355648535567,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00040335753957510906,
|
|
"loss": 4.8552,
|
|
"mean_token_accuracy": 0.19800822883844377,
|
|
"num_tokens": 8060396.0,
|
|
"step": 4360
|
|
},
|
|
{
|
|
"entropy": 5.050659704208374,
|
|
"epoch": 3.6527196652719667,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00040309231540943937,
|
|
"loss": 4.7709,
|
|
"mean_token_accuracy": 0.2178446263074875,
|
|
"num_tokens": 8068106.0,
|
|
"step": 4365
|
|
},
|
|
{
|
|
"entropy": 5.061417102813721,
|
|
"epoch": 3.6569037656903767,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004028268276503784,
|
|
"loss": 4.8127,
|
|
"mean_token_accuracy": 0.2099913015961647,
|
|
"num_tokens": 8077878.0,
|
|
"step": 4370
|
|
},
|
|
{
|
|
"entropy": 5.105002117156983,
|
|
"epoch": 3.6610878661087867,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004025610768442576,
|
|
"loss": 4.9382,
|
|
"mean_token_accuracy": 0.1990278869867325,
|
|
"num_tokens": 8086968.0,
|
|
"step": 4375
|
|
},
|
|
{
|
|
"entropy": 5.073428630828857,
|
|
"epoch": 3.6652719665271967,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004022950635379492,
|
|
"loss": 4.7461,
|
|
"mean_token_accuracy": 0.22072159945964814,
|
|
"num_tokens": 8095999.0,
|
|
"step": 4380
|
|
},
|
|
{
|
|
"entropy": 4.953249454498291,
|
|
"epoch": 3.6694560669456067,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00040202878827886587,
|
|
"loss": 4.7538,
|
|
"mean_token_accuracy": 0.2230379804968834,
|
|
"num_tokens": 8105376.0,
|
|
"step": 4385
|
|
},
|
|
{
|
|
"entropy": 5.0866615772247314,
|
|
"epoch": 3.6736401673640167,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004017622516149592,
|
|
"loss": 4.8512,
|
|
"mean_token_accuracy": 0.20940975695848466,
|
|
"num_tokens": 8115024.0,
|
|
"step": 4390
|
|
},
|
|
{
|
|
"entropy": 5.063612270355224,
|
|
"epoch": 3.6778242677824267,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004014954540947189,
|
|
"loss": 4.8496,
|
|
"mean_token_accuracy": 0.20381466001272203,
|
|
"num_tokens": 8124306.0,
|
|
"step": 4395
|
|
},
|
|
{
|
|
"entropy": 5.092057466506958,
|
|
"epoch": 3.6820083682008367,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00040122839626717137,
|
|
"loss": 4.8606,
|
|
"mean_token_accuracy": 0.212374047935009,
|
|
"num_tokens": 8133151.0,
|
|
"step": 4400
|
|
},
|
|
{
|
|
"entropy": 5.0674127578735355,
|
|
"epoch": 3.6861924686192467,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004009610786818786,
|
|
"loss": 4.8506,
|
|
"mean_token_accuracy": 0.20975900888442994,
|
|
"num_tokens": 8141745.0,
|
|
"step": 4405
|
|
},
|
|
{
|
|
"entropy": 4.994757795333863,
|
|
"epoch": 3.6903765690376567,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00040069350188893735,
|
|
"loss": 4.7025,
|
|
"mean_token_accuracy": 0.22032392472028733,
|
|
"num_tokens": 8150884.0,
|
|
"step": 4410
|
|
},
|
|
{
|
|
"entropy": 5.0607987403869625,
|
|
"epoch": 3.694560669456067,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040042566643897745,
|
|
"loss": 4.8189,
|
|
"mean_token_accuracy": 0.20442269444465638,
|
|
"num_tokens": 8160138.0,
|
|
"step": 4415
|
|
},
|
|
{
|
|
"entropy": 5.09253716468811,
|
|
"epoch": 3.698744769874477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004001575728831613,
|
|
"loss": 4.8774,
|
|
"mean_token_accuracy": 0.21017420142889023,
|
|
"num_tokens": 8169655.0,
|
|
"step": 4420
|
|
},
|
|
{
|
|
"entropy": 5.17042727470398,
|
|
"epoch": 3.702928870292887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003998892217731824,
|
|
"loss": 4.8856,
|
|
"mean_token_accuracy": 0.20997167229652405,
|
|
"num_tokens": 8178846.0,
|
|
"step": 4425
|
|
},
|
|
{
|
|
"entropy": 5.020892477035522,
|
|
"epoch": 3.707112970711297,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000399620613661264,
|
|
"loss": 4.808,
|
|
"mean_token_accuracy": 0.20630120038986205,
|
|
"num_tokens": 8187525.0,
|
|
"step": 4430
|
|
},
|
|
{
|
|
"entropy": 5.041886997222901,
|
|
"epoch": 3.711297071129707,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003993517491001586,
|
|
"loss": 4.8935,
|
|
"mean_token_accuracy": 0.20309384316205978,
|
|
"num_tokens": 8196855.0,
|
|
"step": 4435
|
|
},
|
|
{
|
|
"entropy": 5.031271696090698,
|
|
"epoch": 3.715481171548117,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003990826286431462,
|
|
"loss": 4.7574,
|
|
"mean_token_accuracy": 0.21235278695821763,
|
|
"num_tokens": 8205530.0,
|
|
"step": 4440
|
|
},
|
|
{
|
|
"entropy": 5.085037803649902,
|
|
"epoch": 3.719665271966527,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003988132528440333,
|
|
"loss": 4.844,
|
|
"mean_token_accuracy": 0.20562432408332826,
|
|
"num_tokens": 8215066.0,
|
|
"step": 4445
|
|
},
|
|
{
|
|
"entropy": 5.089331436157226,
|
|
"epoch": 3.723849372384937,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003985436222571521,
|
|
"loss": 4.8364,
|
|
"mean_token_accuracy": 0.2009859189391136,
|
|
"num_tokens": 8224585.0,
|
|
"step": 4450
|
|
},
|
|
{
|
|
"entropy": 5.12259464263916,
|
|
"epoch": 3.7280334728033475,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000398273737437359,
|
|
"loss": 4.9161,
|
|
"mean_token_accuracy": 0.2001411572098732,
|
|
"num_tokens": 8233334.0,
|
|
"step": 4455
|
|
},
|
|
{
|
|
"entropy": 5.12815375328064,
|
|
"epoch": 3.7322175732217575,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00039800359894003355,
|
|
"loss": 4.8357,
|
|
"mean_token_accuracy": 0.2120933249592781,
|
|
"num_tokens": 8241636.0,
|
|
"step": 4460
|
|
},
|
|
{
|
|
"entropy": 5.076593923568725,
|
|
"epoch": 3.7364016736401675,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003977332073210774,
|
|
"loss": 4.8704,
|
|
"mean_token_accuracy": 0.20853939801454544,
|
|
"num_tokens": 8250273.0,
|
|
"step": 4465
|
|
},
|
|
{
|
|
"entropy": 5.088836812973023,
|
|
"epoch": 3.7405857740585775,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003974625631369129,
|
|
"loss": 4.8908,
|
|
"mean_token_accuracy": 0.20182004272937776,
|
|
"num_tokens": 8260349.0,
|
|
"step": 4470
|
|
},
|
|
{
|
|
"entropy": 5.093937683105469,
|
|
"epoch": 3.7447698744769875,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003971916669444824,
|
|
"loss": 4.8907,
|
|
"mean_token_accuracy": 0.20949985086917877,
|
|
"num_tokens": 8270505.0,
|
|
"step": 4475
|
|
},
|
|
{
|
|
"entropy": 5.142940092086792,
|
|
"epoch": 3.7489539748953975,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00039692051930124663,
|
|
"loss": 4.8612,
|
|
"mean_token_accuracy": 0.19923475235700608,
|
|
"num_tokens": 8278651.0,
|
|
"step": 4480
|
|
},
|
|
{
|
|
"entropy": 5.048276472091675,
|
|
"epoch": 3.7531380753138075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003966491207651838,
|
|
"loss": 4.8089,
|
|
"mean_token_accuracy": 0.21328677088022233,
|
|
"num_tokens": 8287173.0,
|
|
"step": 4485
|
|
},
|
|
{
|
|
"entropy": 5.103929233551026,
|
|
"epoch": 3.7573221757322175,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00039637747189478866,
|
|
"loss": 4.8205,
|
|
"mean_token_accuracy": 0.20645579546689988,
|
|
"num_tokens": 8295785.0,
|
|
"step": 4490
|
|
},
|
|
{
|
|
"entropy": 5.141931486129761,
|
|
"epoch": 3.7615062761506275,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003961055732490708,
|
|
"loss": 4.9302,
|
|
"mean_token_accuracy": 0.20601388663053513,
|
|
"num_tokens": 8305534.0,
|
|
"step": 4495
|
|
},
|
|
{
|
|
"entropy": 5.037916946411133,
|
|
"epoch": 3.7656903765690375,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00039583342538755374,
|
|
"loss": 4.8134,
|
|
"mean_token_accuracy": 0.21269698739051818,
|
|
"num_tokens": 8314232.0,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"epoch": 3.7656903765690375,
|
|
"eval_entropy": 4.884806910906833,
|
|
"eval_loss": 5.398492336273193,
|
|
"eval_mean_token_accuracy": 0.19212210286553408,
|
|
"eval_num_tokens": 8314232.0,
|
|
"eval_runtime": 2.928,
|
|
"eval_samples_per_second": 1334.376,
|
|
"eval_steps_per_second": 167.01,
|
|
"step": 4500
|
|
},
|
|
{
|
|
"entropy": 5.0887449264526365,
|
|
"epoch": 3.7698744769874475,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039556102887027423,
|
|
"loss": 4.8708,
|
|
"mean_token_accuracy": 0.2007088452577591,
|
|
"num_tokens": 8323230.0,
|
|
"step": 4505
|
|
},
|
|
{
|
|
"entropy": 5.169616603851319,
|
|
"epoch": 3.7740585774058575,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039528838425778053,
|
|
"loss": 4.9089,
|
|
"mean_token_accuracy": 0.20350997298955917,
|
|
"num_tokens": 8333072.0,
|
|
"step": 4510
|
|
},
|
|
{
|
|
"entropy": 5.091904067993164,
|
|
"epoch": 3.778242677824268,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003950154921111313,
|
|
"loss": 4.8708,
|
|
"mean_token_accuracy": 0.2033441960811615,
|
|
"num_tokens": 8342953.0,
|
|
"step": 4515
|
|
},
|
|
{
|
|
"entropy": 5.062140703201294,
|
|
"epoch": 3.782426778242678,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00039474235299189484,
|
|
"loss": 4.8666,
|
|
"mean_token_accuracy": 0.20563971102237702,
|
|
"num_tokens": 8351729.0,
|
|
"step": 4520
|
|
},
|
|
{
|
|
"entropy": 5.128932237625122,
|
|
"epoch": 3.786610878661088,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003944689674621474,
|
|
"loss": 4.9198,
|
|
"mean_token_accuracy": 0.20895106047391893,
|
|
"num_tokens": 8360929.0,
|
|
"step": 4525
|
|
},
|
|
{
|
|
"entropy": 5.09924464225769,
|
|
"epoch": 3.790794979079498,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00039419533608447244,
|
|
"loss": 4.8996,
|
|
"mean_token_accuracy": 0.20189084559679032,
|
|
"num_tokens": 8369506.0,
|
|
"step": 4530
|
|
},
|
|
{
|
|
"entropy": 5.100954294204712,
|
|
"epoch": 3.794979079497908,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003939214594219595,
|
|
"loss": 4.8144,
|
|
"mean_token_accuracy": 0.21108427047729492,
|
|
"num_tokens": 8378583.0,
|
|
"step": 4535
|
|
},
|
|
{
|
|
"entropy": 5.102209949493409,
|
|
"epoch": 3.799163179916318,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039364733803820263,
|
|
"loss": 4.8041,
|
|
"mean_token_accuracy": 0.21209674030542375,
|
|
"num_tokens": 8388078.0,
|
|
"step": 4540
|
|
},
|
|
{
|
|
"entropy": 4.987361812591553,
|
|
"epoch": 3.803347280334728,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003933729724972996,
|
|
"loss": 4.8058,
|
|
"mean_token_accuracy": 0.2119000256061554,
|
|
"num_tokens": 8397086.0,
|
|
"step": 4545
|
|
},
|
|
{
|
|
"entropy": 5.016415309906006,
|
|
"epoch": 3.8075313807531384,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00039309836336385044,
|
|
"loss": 4.7933,
|
|
"mean_token_accuracy": 0.22351413369178771,
|
|
"num_tokens": 8405621.0,
|
|
"step": 4550
|
|
},
|
|
{
|
|
"entropy": 5.127601242065429,
|
|
"epoch": 3.8117154811715483,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003928235112029568,
|
|
"loss": 4.8583,
|
|
"mean_token_accuracy": 0.20968536883592606,
|
|
"num_tokens": 8415220.0,
|
|
"step": 4555
|
|
},
|
|
{
|
|
"entropy": 5.0672749996185305,
|
|
"epoch": 3.8158995815899583,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003925484165802201,
|
|
"loss": 4.887,
|
|
"mean_token_accuracy": 0.20485952794551848,
|
|
"num_tokens": 8424719.0,
|
|
"step": 4560
|
|
},
|
|
{
|
|
"entropy": 5.059260702133178,
|
|
"epoch": 3.8200836820083683,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003922730800617408,
|
|
"loss": 4.8207,
|
|
"mean_token_accuracy": 0.20452101826667785,
|
|
"num_tokens": 8433844.0,
|
|
"step": 4565
|
|
},
|
|
{
|
|
"entropy": 4.95822491645813,
|
|
"epoch": 3.8242677824267783,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003919975022141174,
|
|
"loss": 4.7451,
|
|
"mean_token_accuracy": 0.21766358166933059,
|
|
"num_tokens": 8442700.0,
|
|
"step": 4570
|
|
},
|
|
{
|
|
"entropy": 5.146186256408692,
|
|
"epoch": 3.8284518828451883,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00039172168360444463,
|
|
"loss": 4.9827,
|
|
"mean_token_accuracy": 0.19318521171808242,
|
|
"num_tokens": 8452246.0,
|
|
"step": 4575
|
|
},
|
|
{
|
|
"entropy": 5.134763431549072,
|
|
"epoch": 3.8326359832635983,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039144562480031297,
|
|
"loss": 4.8319,
|
|
"mean_token_accuracy": 0.20804019570350646,
|
|
"num_tokens": 8461977.0,
|
|
"step": 4580
|
|
},
|
|
{
|
|
"entropy": 5.030867433547973,
|
|
"epoch": 3.8368200836820083,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00039116932636980685,
|
|
"loss": 4.7258,
|
|
"mean_token_accuracy": 0.21764087527990342,
|
|
"num_tokens": 8470124.0,
|
|
"step": 4585
|
|
},
|
|
{
|
|
"entropy": 5.03757791519165,
|
|
"epoch": 3.8410041841004183,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00039089278888150417,
|
|
"loss": 4.8887,
|
|
"mean_token_accuracy": 0.2029036521911621,
|
|
"num_tokens": 8479408.0,
|
|
"step": 4590
|
|
},
|
|
{
|
|
"entropy": 5.059219121932983,
|
|
"epoch": 3.8451882845188283,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003906160129044746,
|
|
"loss": 4.8161,
|
|
"mean_token_accuracy": 0.20938904583454132,
|
|
"num_tokens": 8488239.0,
|
|
"step": 4595
|
|
},
|
|
{
|
|
"entropy": 5.143855285644531,
|
|
"epoch": 3.8493723849372383,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003903389990082787,
|
|
"loss": 4.8665,
|
|
"mean_token_accuracy": 0.2034483477473259,
|
|
"num_tokens": 8497690.0,
|
|
"step": 4600
|
|
},
|
|
{
|
|
"entropy": 5.131691598892212,
|
|
"epoch": 3.8535564853556483,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003900617477629663,
|
|
"loss": 4.9391,
|
|
"mean_token_accuracy": 0.207201486825943,
|
|
"num_tokens": 8506974.0,
|
|
"step": 4605
|
|
},
|
|
{
|
|
"entropy": 5.065772294998169,
|
|
"epoch": 3.8577405857740583,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003897842597390761,
|
|
"loss": 4.8715,
|
|
"mean_token_accuracy": 0.20888876765966416,
|
|
"num_tokens": 8516066.0,
|
|
"step": 4610
|
|
},
|
|
{
|
|
"entropy": 5.064631938934326,
|
|
"epoch": 3.8619246861924688,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003895065355076338,
|
|
"loss": 4.8299,
|
|
"mean_token_accuracy": 0.21090545058250426,
|
|
"num_tokens": 8524747.0,
|
|
"step": 4615
|
|
},
|
|
{
|
|
"entropy": 5.006565856933594,
|
|
"epoch": 3.8661087866108788,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003892285756401511,
|
|
"loss": 4.833,
|
|
"mean_token_accuracy": 0.21545778065919877,
|
|
"num_tokens": 8534274.0,
|
|
"step": 4620
|
|
},
|
|
{
|
|
"entropy": 5.077407550811768,
|
|
"epoch": 3.8702928870292888,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003889503807086248,
|
|
"loss": 4.8309,
|
|
"mean_token_accuracy": 0.20187753438949585,
|
|
"num_tokens": 8543453.0,
|
|
"step": 4625
|
|
},
|
|
{
|
|
"entropy": 5.070255374908447,
|
|
"epoch": 3.8744769874476988,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00038867195128553536,
|
|
"loss": 4.9294,
|
|
"mean_token_accuracy": 0.2065426990389824,
|
|
"num_tokens": 8553190.0,
|
|
"step": 4630
|
|
},
|
|
{
|
|
"entropy": 5.136327409744263,
|
|
"epoch": 3.8786610878661087,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003883932879438457,
|
|
"loss": 4.8039,
|
|
"mean_token_accuracy": 0.20469482541084288,
|
|
"num_tokens": 8562814.0,
|
|
"step": 4635
|
|
},
|
|
{
|
|
"entropy": 5.025051259994507,
|
|
"epoch": 3.8828451882845187,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003881143912570002,
|
|
"loss": 4.8103,
|
|
"mean_token_accuracy": 0.20663148462772368,
|
|
"num_tokens": 8572131.0,
|
|
"step": 4640
|
|
},
|
|
{
|
|
"entropy": 4.959716320037842,
|
|
"epoch": 3.8870292887029287,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003878352617989235,
|
|
"loss": 4.8839,
|
|
"mean_token_accuracy": 0.19846435785293579,
|
|
"num_tokens": 8582325.0,
|
|
"step": 4645
|
|
},
|
|
{
|
|
"entropy": 5.069037389755249,
|
|
"epoch": 3.891213389121339,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003875559001440191,
|
|
"loss": 4.7838,
|
|
"mean_token_accuracy": 0.2163553774356842,
|
|
"num_tokens": 8591704.0,
|
|
"step": 4650
|
|
},
|
|
{
|
|
"entropy": 5.100965356826782,
|
|
"epoch": 3.895397489539749,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003872763068671682,
|
|
"loss": 4.8624,
|
|
"mean_token_accuracy": 0.20466250628232957,
|
|
"num_tokens": 8601136.0,
|
|
"step": 4655
|
|
},
|
|
{
|
|
"entropy": 4.997474527359008,
|
|
"epoch": 3.899581589958159,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003869964825437291,
|
|
"loss": 4.8204,
|
|
"mean_token_accuracy": 0.2090546652674675,
|
|
"num_tokens": 8610491.0,
|
|
"step": 4660
|
|
},
|
|
{
|
|
"entropy": 5.088261079788208,
|
|
"epoch": 3.903765690376569,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038671642774953507,
|
|
"loss": 4.8549,
|
|
"mean_token_accuracy": 0.20822233706712723,
|
|
"num_tokens": 8619199.0,
|
|
"step": 4665
|
|
},
|
|
{
|
|
"entropy": 5.074900436401367,
|
|
"epoch": 3.907949790794979,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003864361430608939,
|
|
"loss": 4.91,
|
|
"mean_token_accuracy": 0.2046773537993431,
|
|
"num_tokens": 8629291.0,
|
|
"step": 4670
|
|
},
|
|
{
|
|
"entropy": 5.0980140209198,
|
|
"epoch": 3.912133891213389,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003861556290545864,
|
|
"loss": 4.8755,
|
|
"mean_token_accuracy": 0.20484975278377532,
|
|
"num_tokens": 8639212.0,
|
|
"step": 4675
|
|
},
|
|
{
|
|
"entropy": 5.051729726791382,
|
|
"epoch": 3.916317991631799,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038587488630786534,
|
|
"loss": 4.825,
|
|
"mean_token_accuracy": 0.20686403661966324,
|
|
"num_tokens": 8648351.0,
|
|
"step": 4680
|
|
},
|
|
{
|
|
"entropy": 5.064916372299194,
|
|
"epoch": 3.920502092050209,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038559391539845413,
|
|
"loss": 4.8734,
|
|
"mean_token_accuracy": 0.204984550178051,
|
|
"num_tokens": 8657267.0,
|
|
"step": 4685
|
|
},
|
|
{
|
|
"entropy": 5.094161081314087,
|
|
"epoch": 3.924686192468619,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00038531271690454567,
|
|
"loss": 5.0139,
|
|
"mean_token_accuracy": 0.20341511070728302,
|
|
"num_tokens": 8666982.0,
|
|
"step": 4690
|
|
},
|
|
{
|
|
"entropy": 5.111768627166748,
|
|
"epoch": 3.928870292887029,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003850312914048015,
|
|
"loss": 4.8047,
|
|
"mean_token_accuracy": 0.21520207822322845,
|
|
"num_tokens": 8676067.0,
|
|
"step": 4695
|
|
},
|
|
{
|
|
"entropy": 5.069075727462769,
|
|
"epoch": 3.933054393305439,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003847496394783496,
|
|
"loss": 4.8602,
|
|
"mean_token_accuracy": 0.2114288792014122,
|
|
"num_tokens": 8685106.0,
|
|
"step": 4700
|
|
},
|
|
{
|
|
"entropy": 5.078355598449707,
|
|
"epoch": 3.937238493723849,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003844677617047848,
|
|
"loss": 4.9108,
|
|
"mean_token_accuracy": 0.19890486150979997,
|
|
"num_tokens": 8695203.0,
|
|
"step": 4705
|
|
},
|
|
{
|
|
"entropy": 4.988590335845947,
|
|
"epoch": 3.941422594142259,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000384185658664166,
|
|
"loss": 4.7497,
|
|
"mean_token_accuracy": 0.2151346832513809,
|
|
"num_tokens": 8703700.0,
|
|
"step": 4710
|
|
},
|
|
{
|
|
"entropy": 5.109715175628662,
|
|
"epoch": 3.9456066945606696,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038390333093701596,
|
|
"loss": 4.8841,
|
|
"mean_token_accuracy": 0.20251901000738143,
|
|
"num_tokens": 8713393.0,
|
|
"step": 4715
|
|
},
|
|
{
|
|
"entropy": 5.071229267120361,
|
|
"epoch": 3.9497907949790796,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038362077910431973,
|
|
"loss": 4.8428,
|
|
"mean_token_accuracy": 0.20612818300724028,
|
|
"num_tokens": 8722263.0,
|
|
"step": 4720
|
|
},
|
|
{
|
|
"entropy": 5.073835134506226,
|
|
"epoch": 3.9539748953974896,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003833380037475236,
|
|
"loss": 4.8465,
|
|
"mean_token_accuracy": 0.21168048083782195,
|
|
"num_tokens": 8732190.0,
|
|
"step": 4725
|
|
},
|
|
{
|
|
"entropy": 5.104987955093383,
|
|
"epoch": 3.9581589958158996,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00038305500544853375,
|
|
"loss": 4.8055,
|
|
"mean_token_accuracy": 0.2169350042939186,
|
|
"num_tokens": 8740995.0,
|
|
"step": 4730
|
|
},
|
|
{
|
|
"entropy": 5.152263498306274,
|
|
"epoch": 3.9623430962343096,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00038277178478971527,
|
|
"loss": 4.9471,
|
|
"mean_token_accuracy": 0.2005343645811081,
|
|
"num_tokens": 8750768.0,
|
|
"step": 4735
|
|
},
|
|
{
|
|
"entropy": 5.04274697303772,
|
|
"epoch": 3.9665271966527196,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003824883423538905,
|
|
"loss": 4.8601,
|
|
"mean_token_accuracy": 0.20793256163597107,
|
|
"num_tokens": 8760070.0,
|
|
"step": 4740
|
|
},
|
|
{
|
|
"entropy": 5.042676162719727,
|
|
"epoch": 3.9707112970711296,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00038220467872433875,
|
|
"loss": 4.8394,
|
|
"mean_token_accuracy": 0.21444716304540634,
|
|
"num_tokens": 8769313.0,
|
|
"step": 4745
|
|
},
|
|
{
|
|
"entropy": 5.059892702102661,
|
|
"epoch": 3.97489539748954,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003819207944847939,
|
|
"loss": 4.8111,
|
|
"mean_token_accuracy": 0.21010124683380127,
|
|
"num_tokens": 8778328.0,
|
|
"step": 4750
|
|
},
|
|
{
|
|
"entropy": 5.057912349700928,
|
|
"epoch": 3.97907949790795,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00038163669021944413,
|
|
"loss": 4.8407,
|
|
"mean_token_accuracy": 0.2133011907339096,
|
|
"num_tokens": 8787408.0,
|
|
"step": 4755
|
|
},
|
|
{
|
|
"entropy": 5.097698831558228,
|
|
"epoch": 3.98326359832636,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003813523665129304,
|
|
"loss": 4.7984,
|
|
"mean_token_accuracy": 0.21019234210252763,
|
|
"num_tokens": 8796781.0,
|
|
"step": 4760
|
|
},
|
|
{
|
|
"entropy": 5.0480865955352785,
|
|
"epoch": 3.98744769874477,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003810678239503452,
|
|
"loss": 4.8515,
|
|
"mean_token_accuracy": 0.2045007899403572,
|
|
"num_tokens": 8806965.0,
|
|
"step": 4765
|
|
},
|
|
{
|
|
"entropy": 5.032988452911377,
|
|
"epoch": 3.99163179916318,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003807830631172312,
|
|
"loss": 4.8689,
|
|
"mean_token_accuracy": 0.2041998401284218,
|
|
"num_tokens": 8817348.0,
|
|
"step": 4770
|
|
},
|
|
{
|
|
"entropy": 5.048079490661621,
|
|
"epoch": 3.99581589958159,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00038049808459958065,
|
|
"loss": 4.9004,
|
|
"mean_token_accuracy": 0.20696699172258376,
|
|
"num_tokens": 8827304.0,
|
|
"step": 4775
|
|
},
|
|
{
|
|
"entropy": 5.064014482498169,
|
|
"epoch": 4.0,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038021288898383343,
|
|
"loss": 4.8133,
|
|
"mean_token_accuracy": 0.21239476948976516,
|
|
"num_tokens": 8837288.0,
|
|
"step": 4780
|
|
},
|
|
{
|
|
"entropy": 5.099370050430298,
|
|
"epoch": 4.00418410041841,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003799274768568761,
|
|
"loss": 4.6005,
|
|
"mean_token_accuracy": 0.22966532409191132,
|
|
"num_tokens": 8846447.0,
|
|
"step": 4785
|
|
},
|
|
{
|
|
"entropy": 5.0109912872314455,
|
|
"epoch": 4.00836820083682,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00037964184880604125,
|
|
"loss": 4.5184,
|
|
"mean_token_accuracy": 0.23245137482881545,
|
|
"num_tokens": 8855990.0,
|
|
"step": 4790
|
|
},
|
|
{
|
|
"entropy": 4.955549669265747,
|
|
"epoch": 4.01255230125523,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037935600541910523,
|
|
"loss": 4.5224,
|
|
"mean_token_accuracy": 0.22821322679519654,
|
|
"num_tokens": 8864949.0,
|
|
"step": 4795
|
|
},
|
|
{
|
|
"entropy": 4.945098543167115,
|
|
"epoch": 4.01673640167364,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003790699472842879,
|
|
"loss": 4.4632,
|
|
"mean_token_accuracy": 0.23718201667070388,
|
|
"num_tokens": 8873501.0,
|
|
"step": 4800
|
|
},
|
|
{
|
|
"entropy": 4.970974540710449,
|
|
"epoch": 4.02092050209205,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037878367499025086,
|
|
"loss": 4.5862,
|
|
"mean_token_accuracy": 0.22623080462217332,
|
|
"num_tokens": 8882920.0,
|
|
"step": 4805
|
|
},
|
|
{
|
|
"entropy": 5.004230594635009,
|
|
"epoch": 4.02510460251046,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003784971891260965,
|
|
"loss": 4.5253,
|
|
"mean_token_accuracy": 0.2292540192604065,
|
|
"num_tokens": 8893617.0,
|
|
"step": 4810
|
|
},
|
|
{
|
|
"entropy": 5.063896703720093,
|
|
"epoch": 4.02928870292887,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00037821049028136664,
|
|
"loss": 4.5188,
|
|
"mean_token_accuracy": 0.22279528975486756,
|
|
"num_tokens": 8902558.0,
|
|
"step": 4815
|
|
},
|
|
{
|
|
"entropy": 4.9914785861969,
|
|
"epoch": 4.03347280334728,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003779235790460414,
|
|
"loss": 4.5541,
|
|
"mean_token_accuracy": 0.22863524854183198,
|
|
"num_tokens": 8912477.0,
|
|
"step": 4820
|
|
},
|
|
{
|
|
"entropy": 4.967843675613404,
|
|
"epoch": 4.03765690376569,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003776364560105379,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.22944119870662688,
|
|
"num_tokens": 8922313.0,
|
|
"step": 4825
|
|
},
|
|
{
|
|
"entropy": 4.993152093887329,
|
|
"epoch": 4.0418410041841,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037734912176570935,
|
|
"loss": 4.493,
|
|
"mean_token_accuracy": 0.23747980445623398,
|
|
"num_tokens": 8932011.0,
|
|
"step": 4830
|
|
},
|
|
{
|
|
"entropy": 5.010499286651611,
|
|
"epoch": 4.046025104602511,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00037706157690284335,
|
|
"loss": 4.5563,
|
|
"mean_token_accuracy": 0.22447819262742996,
|
|
"num_tokens": 8941022.0,
|
|
"step": 4835
|
|
},
|
|
{
|
|
"entropy": 4.967879343032837,
|
|
"epoch": 4.050209205020921,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00037677382201366103,
|
|
"loss": 4.5581,
|
|
"mean_token_accuracy": 0.2326229214668274,
|
|
"num_tokens": 8950825.0,
|
|
"step": 4840
|
|
},
|
|
{
|
|
"entropy": 5.002725601196289,
|
|
"epoch": 4.054393305439331,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00037648585769031555,
|
|
"loss": 4.5919,
|
|
"mean_token_accuracy": 0.21739500910043716,
|
|
"num_tokens": 8961035.0,
|
|
"step": 4845
|
|
},
|
|
{
|
|
"entropy": 5.044254207611084,
|
|
"epoch": 4.058577405857741,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003761976845253915,
|
|
"loss": 4.6102,
|
|
"mean_token_accuracy": 0.22492135018110276,
|
|
"num_tokens": 8970535.0,
|
|
"step": 4850
|
|
},
|
|
{
|
|
"entropy": 4.953998041152954,
|
|
"epoch": 4.062761506276151,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003759093031119025,
|
|
"loss": 4.5693,
|
|
"mean_token_accuracy": 0.22199997305870056,
|
|
"num_tokens": 8979404.0,
|
|
"step": 4855
|
|
},
|
|
{
|
|
"entropy": 5.0523707389831545,
|
|
"epoch": 4.066945606694561,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00037562071404329147,
|
|
"loss": 4.548,
|
|
"mean_token_accuracy": 0.23430201411247253,
|
|
"num_tokens": 8989164.0,
|
|
"step": 4860
|
|
},
|
|
{
|
|
"entropy": 5.061492013931274,
|
|
"epoch": 4.071129707112971,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00037533191791342817,
|
|
"loss": 4.6097,
|
|
"mean_token_accuracy": 0.22056894153356552,
|
|
"num_tokens": 8998630.0,
|
|
"step": 4865
|
|
},
|
|
{
|
|
"entropy": 5.012836170196533,
|
|
"epoch": 4.075313807531381,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00037504291531660866,
|
|
"loss": 4.5324,
|
|
"mean_token_accuracy": 0.23262185752391815,
|
|
"num_tokens": 9008235.0,
|
|
"step": 4870
|
|
},
|
|
{
|
|
"entropy": 5.011108493804931,
|
|
"epoch": 4.079497907949791,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003747537068475538,
|
|
"loss": 4.6448,
|
|
"mean_token_accuracy": 0.22776305675506592,
|
|
"num_tokens": 9018175.0,
|
|
"step": 4875
|
|
},
|
|
{
|
|
"entropy": 5.085087823867798,
|
|
"epoch": 4.083682008368201,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003744642931014082,
|
|
"loss": 4.6615,
|
|
"mean_token_accuracy": 0.21800965517759324,
|
|
"num_tokens": 9028159.0,
|
|
"step": 4880
|
|
},
|
|
{
|
|
"entropy": 4.935358476638794,
|
|
"epoch": 4.087866108786611,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003741746746737386,
|
|
"loss": 4.5828,
|
|
"mean_token_accuracy": 0.2178384006023407,
|
|
"num_tokens": 9038263.0,
|
|
"step": 4885
|
|
},
|
|
{
|
|
"entropy": 5.008071947097778,
|
|
"epoch": 4.092050209205021,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003738848521605333,
|
|
"loss": 4.5673,
|
|
"mean_token_accuracy": 0.23014640659093857,
|
|
"num_tokens": 9048023.0,
|
|
"step": 4890
|
|
},
|
|
{
|
|
"entropy": 5.02693247795105,
|
|
"epoch": 4.096234309623431,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003735948261582005,
|
|
"loss": 4.6302,
|
|
"mean_token_accuracy": 0.21947073936462402,
|
|
"num_tokens": 9056988.0,
|
|
"step": 4895
|
|
},
|
|
{
|
|
"entropy": 4.984006786346436,
|
|
"epoch": 4.100418410041841,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000373304597263567,
|
|
"loss": 4.5962,
|
|
"mean_token_accuracy": 0.2262573331594467,
|
|
"num_tokens": 9066595.0,
|
|
"step": 4900
|
|
},
|
|
{
|
|
"entropy": 5.044378137588501,
|
|
"epoch": 4.104602510460251,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00037301416607387716,
|
|
"loss": 4.6178,
|
|
"mean_token_accuracy": 0.22509003281593323,
|
|
"num_tokens": 9075034.0,
|
|
"step": 4905
|
|
},
|
|
{
|
|
"entropy": 4.936192798614502,
|
|
"epoch": 4.108786610878661,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00037272353318679185,
|
|
"loss": 4.5642,
|
|
"mean_token_accuracy": 0.23097352534532548,
|
|
"num_tokens": 9084574.0,
|
|
"step": 4910
|
|
},
|
|
{
|
|
"entropy": 4.947194290161133,
|
|
"epoch": 4.112970711297071,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003724326992003867,
|
|
"loss": 4.5339,
|
|
"mean_token_accuracy": 0.2299984350800514,
|
|
"num_tokens": 9093591.0,
|
|
"step": 4915
|
|
},
|
|
{
|
|
"entropy": 4.988737773895264,
|
|
"epoch": 4.117154811715481,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003721416647131512,
|
|
"loss": 4.5691,
|
|
"mean_token_accuracy": 0.23388854861259462,
|
|
"num_tokens": 9102439.0,
|
|
"step": 4920
|
|
},
|
|
{
|
|
"entropy": 4.902376174926758,
|
|
"epoch": 4.121338912133891,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003718504303239876,
|
|
"loss": 4.4453,
|
|
"mean_token_accuracy": 0.23571163713932036,
|
|
"num_tokens": 9111363.0,
|
|
"step": 4925
|
|
},
|
|
{
|
|
"entropy": 4.99338812828064,
|
|
"epoch": 4.125523012552302,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003715589966322097,
|
|
"loss": 4.6294,
|
|
"mean_token_accuracy": 0.22100960910320283,
|
|
"num_tokens": 9120417.0,
|
|
"step": 4930
|
|
},
|
|
{
|
|
"entropy": 5.04291181564331,
|
|
"epoch": 4.129707112970712,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003712673642375409,
|
|
"loss": 4.5981,
|
|
"mean_token_accuracy": 0.2220792755484581,
|
|
"num_tokens": 9130622.0,
|
|
"step": 4935
|
|
},
|
|
{
|
|
"entropy": 4.999504852294922,
|
|
"epoch": 4.133891213389122,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003709755337401138,
|
|
"loss": 4.5935,
|
|
"mean_token_accuracy": 0.22236400544643403,
|
|
"num_tokens": 9139992.0,
|
|
"step": 4940
|
|
},
|
|
{
|
|
"entropy": 4.907515144348144,
|
|
"epoch": 4.138075313807532,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00037068350574046867,
|
|
"loss": 4.5711,
|
|
"mean_token_accuracy": 0.2272783100605011,
|
|
"num_tokens": 9149883.0,
|
|
"step": 4945
|
|
},
|
|
{
|
|
"entropy": 4.936377382278442,
|
|
"epoch": 4.142259414225942,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00037039128083955227,
|
|
"loss": 4.5251,
|
|
"mean_token_accuracy": 0.2315220907330513,
|
|
"num_tokens": 9159633.0,
|
|
"step": 4950
|
|
},
|
|
{
|
|
"entropy": 4.986016368865966,
|
|
"epoch": 4.146443514644352,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003700988596387163,
|
|
"loss": 4.5118,
|
|
"mean_token_accuracy": 0.22911328971385955,
|
|
"num_tokens": 9167780.0,
|
|
"step": 4955
|
|
},
|
|
{
|
|
"entropy": 4.989432525634766,
|
|
"epoch": 4.150627615062762,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003698062427397168,
|
|
"loss": 4.5551,
|
|
"mean_token_accuracy": 0.22838603407144548,
|
|
"num_tokens": 9176808.0,
|
|
"step": 4960
|
|
},
|
|
{
|
|
"entropy": 5.008588123321533,
|
|
"epoch": 4.154811715481172,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036951343074471217,
|
|
"loss": 4.5822,
|
|
"mean_token_accuracy": 0.22847014516592026,
|
|
"num_tokens": 9186097.0,
|
|
"step": 4965
|
|
},
|
|
{
|
|
"entropy": 4.995897197723389,
|
|
"epoch": 4.158995815899582,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003692204242562624,
|
|
"loss": 4.6132,
|
|
"mean_token_accuracy": 0.2167380541563034,
|
|
"num_tokens": 9194428.0,
|
|
"step": 4970
|
|
},
|
|
{
|
|
"entropy": 4.968228435516357,
|
|
"epoch": 4.163179916317992,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003689272238773278,
|
|
"loss": 4.6121,
|
|
"mean_token_accuracy": 0.2233587920665741,
|
|
"num_tokens": 9203156.0,
|
|
"step": 4975
|
|
},
|
|
{
|
|
"entropy": 4.976684665679931,
|
|
"epoch": 4.167364016736402,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00036863383021126756,
|
|
"loss": 4.5832,
|
|
"mean_token_accuracy": 0.22845959961414336,
|
|
"num_tokens": 9212102.0,
|
|
"step": 4980
|
|
},
|
|
{
|
|
"entropy": 4.976931762695313,
|
|
"epoch": 4.171548117154812,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00036834024386183875,
|
|
"loss": 4.6111,
|
|
"mean_token_accuracy": 0.2249277114868164,
|
|
"num_tokens": 9221507.0,
|
|
"step": 4985
|
|
},
|
|
{
|
|
"entropy": 4.9744750499725345,
|
|
"epoch": 4.175732217573222,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003680464654331947,
|
|
"loss": 4.5188,
|
|
"mean_token_accuracy": 0.22952794432640075,
|
|
"num_tokens": 9230543.0,
|
|
"step": 4990
|
|
},
|
|
{
|
|
"entropy": 5.016287803649902,
|
|
"epoch": 4.179916317991632,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003677524955298844,
|
|
"loss": 4.6429,
|
|
"mean_token_accuracy": 0.22388558685779572,
|
|
"num_tokens": 9240092.0,
|
|
"step": 4995
|
|
},
|
|
{
|
|
"entropy": 4.9820661544799805,
|
|
"epoch": 4.184100418410042,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003674583347568505,
|
|
"loss": 4.5903,
|
|
"mean_token_accuracy": 0.22233077883720398,
|
|
"num_tokens": 9250145.0,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"epoch": 4.184100418410042,
|
|
"eval_entropy": 4.855961594357325,
|
|
"eval_loss": 5.400161266326904,
|
|
"eval_mean_token_accuracy": 0.19320476475668832,
|
|
"eval_num_tokens": 9250145.0,
|
|
"eval_runtime": 2.7534,
|
|
"eval_samples_per_second": 1418.967,
|
|
"eval_steps_per_second": 177.598,
|
|
"step": 5000
|
|
},
|
|
{
|
|
"entropy": 5.007485437393188,
|
|
"epoch": 4.188284518828452,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00036716398371942866,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.2279185250401497,
|
|
"num_tokens": 9259643.0,
|
|
"step": 5005
|
|
},
|
|
{
|
|
"entropy": 4.996592426300049,
|
|
"epoch": 4.192468619246862,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003668694430233459,
|
|
"loss": 4.6075,
|
|
"mean_token_accuracy": 0.2190637469291687,
|
|
"num_tokens": 9268471.0,
|
|
"step": 5010
|
|
},
|
|
{
|
|
"entropy": 5.024766349792481,
|
|
"epoch": 4.196652719665272,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003665747132747198,
|
|
"loss": 4.54,
|
|
"mean_token_accuracy": 0.2208954095840454,
|
|
"num_tokens": 9276916.0,
|
|
"step": 5015
|
|
},
|
|
{
|
|
"entropy": 4.893040800094605,
|
|
"epoch": 4.200836820083682,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036627979508005654,
|
|
"loss": 4.5602,
|
|
"mean_token_accuracy": 0.23013072609901428,
|
|
"num_tokens": 9285867.0,
|
|
"step": 5020
|
|
},
|
|
{
|
|
"entropy": 5.011058044433594,
|
|
"epoch": 4.205020920502092,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003659846890462506,
|
|
"loss": 4.6713,
|
|
"mean_token_accuracy": 0.22012120634317398,
|
|
"num_tokens": 9295044.0,
|
|
"step": 5025
|
|
},
|
|
{
|
|
"entropy": 5.080321311950684,
|
|
"epoch": 4.209205020920502,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003656893957805826,
|
|
"loss": 4.6881,
|
|
"mean_token_accuracy": 0.22354225665330887,
|
|
"num_tokens": 9303744.0,
|
|
"step": 5030
|
|
},
|
|
{
|
|
"entropy": 5.019132852554321,
|
|
"epoch": 4.2133891213389125,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003653939158907186,
|
|
"loss": 4.5667,
|
|
"mean_token_accuracy": 0.2205440506339073,
|
|
"num_tokens": 9312996.0,
|
|
"step": 5035
|
|
},
|
|
{
|
|
"entropy": 4.91114501953125,
|
|
"epoch": 4.2175732217573225,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003650982499847088,
|
|
"loss": 4.5685,
|
|
"mean_token_accuracy": 0.2260834276676178,
|
|
"num_tokens": 9321912.0,
|
|
"step": 5040
|
|
},
|
|
{
|
|
"entropy": 4.936156845092773,
|
|
"epoch": 4.2217573221757325,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000364802398670986,
|
|
"loss": 4.5789,
|
|
"mean_token_accuracy": 0.2273917868733406,
|
|
"num_tokens": 9331798.0,
|
|
"step": 5045
|
|
},
|
|
{
|
|
"entropy": 4.973091697692871,
|
|
"epoch": 4.2259414225941425,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00036450636255836474,
|
|
"loss": 4.6523,
|
|
"mean_token_accuracy": 0.21732382625341415,
|
|
"num_tokens": 9342064.0,
|
|
"step": 5050
|
|
},
|
|
{
|
|
"entropy": 5.04557785987854,
|
|
"epoch": 4.2301255230125525,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00036421014225603964,
|
|
"loss": 4.5564,
|
|
"mean_token_accuracy": 0.22651734054088593,
|
|
"num_tokens": 9351494.0,
|
|
"step": 5055
|
|
},
|
|
{
|
|
"entropy": 4.932202863693237,
|
|
"epoch": 4.2343096234309625,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036391373837358455,
|
|
"loss": 4.5092,
|
|
"mean_token_accuracy": 0.23322861790657043,
|
|
"num_tokens": 9360683.0,
|
|
"step": 5060
|
|
},
|
|
{
|
|
"entropy": 4.886192035675049,
|
|
"epoch": 4.2384937238493725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00036361715152095095,
|
|
"loss": 4.5907,
|
|
"mean_token_accuracy": 0.21822239309549332,
|
|
"num_tokens": 9370784.0,
|
|
"step": 5065
|
|
},
|
|
{
|
|
"entropy": 5.034599161148071,
|
|
"epoch": 4.2426778242677825,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003633203823084669,
|
|
"loss": 4.6421,
|
|
"mean_token_accuracy": 0.2282520741224289,
|
|
"num_tokens": 9380673.0,
|
|
"step": 5070
|
|
},
|
|
{
|
|
"entropy": 5.0176492691040036,
|
|
"epoch": 4.2468619246861925,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036302343134683566,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.22692845165729522,
|
|
"num_tokens": 9390260.0,
|
|
"step": 5075
|
|
},
|
|
{
|
|
"entropy": 5.021199464797974,
|
|
"epoch": 4.2510460251046025,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0003627262992471348,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.2236812084913254,
|
|
"num_tokens": 9400643.0,
|
|
"step": 5080
|
|
},
|
|
{
|
|
"entropy": 4.959815645217896,
|
|
"epoch": 4.2552301255230125,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003624289866208142,
|
|
"loss": 4.5399,
|
|
"mean_token_accuracy": 0.2264810547232628,
|
|
"num_tokens": 9408898.0,
|
|
"step": 5085
|
|
},
|
|
{
|
|
"entropy": 4.991559791564941,
|
|
"epoch": 4.2594142259414225,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00036213149407969547,
|
|
"loss": 4.598,
|
|
"mean_token_accuracy": 0.21969303488731384,
|
|
"num_tokens": 9418024.0,
|
|
"step": 5090
|
|
},
|
|
{
|
|
"entropy": 5.0589834690094,
|
|
"epoch": 4.2635983263598325,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003618338222359705,
|
|
"loss": 4.6885,
|
|
"mean_token_accuracy": 0.21614789515733718,
|
|
"num_tokens": 9427548.0,
|
|
"step": 5095
|
|
},
|
|
{
|
|
"entropy": 4.984819221496582,
|
|
"epoch": 4.2677824267782425,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00036153597170220005,
|
|
"loss": 4.5144,
|
|
"mean_token_accuracy": 0.22999501824378968,
|
|
"num_tokens": 9436370.0,
|
|
"step": 5100
|
|
},
|
|
{
|
|
"entropy": 4.978184843063355,
|
|
"epoch": 4.2719665271966525,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003612379430913126,
|
|
"loss": 4.587,
|
|
"mean_token_accuracy": 0.227661395072937,
|
|
"num_tokens": 9445506.0,
|
|
"step": 5105
|
|
},
|
|
{
|
|
"entropy": 4.987661552429199,
|
|
"epoch": 4.2761506276150625,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0003609397370166031,
|
|
"loss": 4.5252,
|
|
"mean_token_accuracy": 0.22972508072853087,
|
|
"num_tokens": 9455288.0,
|
|
"step": 5110
|
|
},
|
|
{
|
|
"entropy": 4.93526029586792,
|
|
"epoch": 4.2803347280334725,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003606413540917319,
|
|
"loss": 4.6454,
|
|
"mean_token_accuracy": 0.2212807446718216,
|
|
"num_tokens": 9464366.0,
|
|
"step": 5115
|
|
},
|
|
{
|
|
"entropy": 4.974261283874512,
|
|
"epoch": 4.2845188284518825,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003603427949307228,
|
|
"loss": 4.57,
|
|
"mean_token_accuracy": 0.2302352234721184,
|
|
"num_tokens": 9474205.0,
|
|
"step": 5120
|
|
},
|
|
{
|
|
"entropy": 4.910864591598511,
|
|
"epoch": 4.2887029288702925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00036004406014796274,
|
|
"loss": 4.5493,
|
|
"mean_token_accuracy": 0.22894600182771682,
|
|
"num_tokens": 9483181.0,
|
|
"step": 5125
|
|
},
|
|
{
|
|
"entropy": 4.910321617126465,
|
|
"epoch": 4.292887029288703,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003597451503581998,
|
|
"loss": 4.5607,
|
|
"mean_token_accuracy": 0.23195962011814117,
|
|
"num_tokens": 9492394.0,
|
|
"step": 5130
|
|
},
|
|
{
|
|
"entropy": 5.034613752365113,
|
|
"epoch": 4.297071129707113,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00035944606617654236,
|
|
"loss": 4.6538,
|
|
"mean_token_accuracy": 0.2164781853556633,
|
|
"num_tokens": 9502184.0,
|
|
"step": 5135
|
|
},
|
|
{
|
|
"entropy": 4.962450838088989,
|
|
"epoch": 4.301255230125523,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035914680821845746,
|
|
"loss": 4.6141,
|
|
"mean_token_accuracy": 0.22477517575025557,
|
|
"num_tokens": 9511399.0,
|
|
"step": 5140
|
|
},
|
|
{
|
|
"entropy": 4.9716709613800045,
|
|
"epoch": 4.305439330543933,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035884737709976995,
|
|
"loss": 4.6383,
|
|
"mean_token_accuracy": 0.22763275355100632,
|
|
"num_tokens": 9520710.0,
|
|
"step": 5145
|
|
},
|
|
{
|
|
"entropy": 4.967987537384033,
|
|
"epoch": 4.309623430962343,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00035854777343666093,
|
|
"loss": 4.4896,
|
|
"mean_token_accuracy": 0.2403864398598671,
|
|
"num_tokens": 9529992.0,
|
|
"step": 5150
|
|
},
|
|
{
|
|
"entropy": 4.9390504360198975,
|
|
"epoch": 4.313807531380753,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003582479978456665,
|
|
"loss": 4.5063,
|
|
"mean_token_accuracy": 0.23048575818538666,
|
|
"num_tokens": 9538279.0,
|
|
"step": 5155
|
|
},
|
|
{
|
|
"entropy": 4.934126567840576,
|
|
"epoch": 4.317991631799163,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003579480509436766,
|
|
"loss": 4.522,
|
|
"mean_token_accuracy": 0.23249053061008454,
|
|
"num_tokens": 9547425.0,
|
|
"step": 5160
|
|
},
|
|
{
|
|
"entropy": 4.949304103851318,
|
|
"epoch": 4.322175732217573,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035764793334793393,
|
|
"loss": 4.5533,
|
|
"mean_token_accuracy": 0.22813486754894258,
|
|
"num_tokens": 9556657.0,
|
|
"step": 5165
|
|
},
|
|
{
|
|
"entropy": 4.987043285369873,
|
|
"epoch": 4.326359832635983,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035734764567603216,
|
|
"loss": 4.6376,
|
|
"mean_token_accuracy": 0.22299837172031403,
|
|
"num_tokens": 9565989.0,
|
|
"step": 5170
|
|
},
|
|
{
|
|
"entropy": 4.928238153457642,
|
|
"epoch": 4.330543933054393,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003570471885459149,
|
|
"loss": 4.5766,
|
|
"mean_token_accuracy": 0.22828584909439087,
|
|
"num_tokens": 9575246.0,
|
|
"step": 5175
|
|
},
|
|
{
|
|
"entropy": 4.935373449325562,
|
|
"epoch": 4.334728033472803,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003567465625758748,
|
|
"loss": 4.5196,
|
|
"mean_token_accuracy": 0.2307463064789772,
|
|
"num_tokens": 9586185.0,
|
|
"step": 5180
|
|
},
|
|
{
|
|
"entropy": 4.917463541030884,
|
|
"epoch": 4.338912133891213,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003564457683845516,
|
|
"loss": 4.5149,
|
|
"mean_token_accuracy": 0.23191220313310623,
|
|
"num_tokens": 9595222.0,
|
|
"step": 5185
|
|
},
|
|
{
|
|
"entropy": 5.025051307678223,
|
|
"epoch": 4.343096234309623,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003561448065909315,
|
|
"loss": 4.6731,
|
|
"mean_token_accuracy": 0.22581359148025512,
|
|
"num_tokens": 9604866.0,
|
|
"step": 5190
|
|
},
|
|
{
|
|
"entropy": 4.936069583892822,
|
|
"epoch": 4.347280334728033,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003558436778143455,
|
|
"loss": 4.5167,
|
|
"mean_token_accuracy": 0.23848295211791992,
|
|
"num_tokens": 9614522.0,
|
|
"step": 5195
|
|
},
|
|
{
|
|
"entropy": 4.996898031234741,
|
|
"epoch": 4.351464435146443,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003555423826744682,
|
|
"loss": 4.6764,
|
|
"mean_token_accuracy": 0.21184370219707488,
|
|
"num_tokens": 9623308.0,
|
|
"step": 5200
|
|
},
|
|
{
|
|
"entropy": 5.003992986679077,
|
|
"epoch": 4.355648535564853,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00035524092179131645,
|
|
"loss": 4.664,
|
|
"mean_token_accuracy": 0.2172957330942154,
|
|
"num_tokens": 9632887.0,
|
|
"step": 5205
|
|
},
|
|
{
|
|
"entropy": 4.965266704559326,
|
|
"epoch": 4.359832635983263,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003549392957852485,
|
|
"loss": 4.6246,
|
|
"mean_token_accuracy": 0.21999699771404266,
|
|
"num_tokens": 9643218.0,
|
|
"step": 5210
|
|
},
|
|
{
|
|
"entropy": 4.970787477493286,
|
|
"epoch": 4.364016736401673,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003546375052769619,
|
|
"loss": 4.6286,
|
|
"mean_token_accuracy": 0.2146576702594757,
|
|
"num_tokens": 9653175.0,
|
|
"step": 5215
|
|
},
|
|
{
|
|
"entropy": 4.917086505889893,
|
|
"epoch": 4.368200836820083,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003543355508874932,
|
|
"loss": 4.5263,
|
|
"mean_token_accuracy": 0.2292853683233261,
|
|
"num_tokens": 9662431.0,
|
|
"step": 5220
|
|
},
|
|
{
|
|
"entropy": 4.992163038253784,
|
|
"epoch": 4.372384937238493,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003540334332382159,
|
|
"loss": 4.5741,
|
|
"mean_token_accuracy": 0.2309194728732109,
|
|
"num_tokens": 9671191.0,
|
|
"step": 5225
|
|
},
|
|
{
|
|
"entropy": 4.89826979637146,
|
|
"epoch": 4.376569037656903,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003537311529508396,
|
|
"loss": 4.5311,
|
|
"mean_token_accuracy": 0.22980049401521682,
|
|
"num_tokens": 9680637.0,
|
|
"step": 5230
|
|
},
|
|
{
|
|
"entropy": 4.928881978988647,
|
|
"epoch": 4.380753138075314,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00035342871064740846,
|
|
"loss": 4.48,
|
|
"mean_token_accuracy": 0.22648437470197677,
|
|
"num_tokens": 9689325.0,
|
|
"step": 5235
|
|
},
|
|
{
|
|
"entropy": 4.965658664703369,
|
|
"epoch": 4.384937238493724,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00035312610695030017,
|
|
"loss": 4.6133,
|
|
"mean_token_accuracy": 0.22405683100223542,
|
|
"num_tokens": 9698335.0,
|
|
"step": 5240
|
|
},
|
|
{
|
|
"entropy": 4.914887619018555,
|
|
"epoch": 4.389121338912134,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0003528233424822245,
|
|
"loss": 4.5443,
|
|
"mean_token_accuracy": 0.2252321347594261,
|
|
"num_tokens": 9706647.0,
|
|
"step": 5245
|
|
},
|
|
{
|
|
"entropy": 5.058056259155274,
|
|
"epoch": 4.393305439330544,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000352520417866222,
|
|
"loss": 4.5899,
|
|
"mean_token_accuracy": 0.2283244401216507,
|
|
"num_tokens": 9715941.0,
|
|
"step": 5250
|
|
},
|
|
{
|
|
"entropy": 4.907711458206177,
|
|
"epoch": 4.397489539748954,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0003522173337256628,
|
|
"loss": 4.6294,
|
|
"mean_token_accuracy": 0.21934579014778138,
|
|
"num_tokens": 9724693.0,
|
|
"step": 5255
|
|
},
|
|
{
|
|
"entropy": 4.953387689590454,
|
|
"epoch": 4.401673640167364,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003519140906842455,
|
|
"loss": 4.6384,
|
|
"mean_token_accuracy": 0.22570136338472366,
|
|
"num_tokens": 9733314.0,
|
|
"step": 5260
|
|
},
|
|
{
|
|
"entropy": 5.081447696685791,
|
|
"epoch": 4.405857740585774,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003516106893659953,
|
|
"loss": 4.6665,
|
|
"mean_token_accuracy": 0.22074659764766694,
|
|
"num_tokens": 9742844.0,
|
|
"step": 5265
|
|
},
|
|
{
|
|
"entropy": 4.912633752822876,
|
|
"epoch": 4.410041841004184,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003513071303952635,
|
|
"loss": 4.5419,
|
|
"mean_token_accuracy": 0.22619218677282332,
|
|
"num_tokens": 9752164.0,
|
|
"step": 5270
|
|
},
|
|
{
|
|
"entropy": 4.849109268188476,
|
|
"epoch": 4.414225941422594,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003510034143967256,
|
|
"loss": 4.5533,
|
|
"mean_token_accuracy": 0.22914798259735109,
|
|
"num_tokens": 9761483.0,
|
|
"step": 5275
|
|
},
|
|
{
|
|
"entropy": 4.955806922912598,
|
|
"epoch": 4.418410041841004,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00035069954199538027,
|
|
"loss": 4.5129,
|
|
"mean_token_accuracy": 0.22946388572454451,
|
|
"num_tokens": 9769998.0,
|
|
"step": 5280
|
|
},
|
|
{
|
|
"entropy": 5.010292959213257,
|
|
"epoch": 4.422594142259414,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00035039551381654816,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.22881415784358977,
|
|
"num_tokens": 9778965.0,
|
|
"step": 5285
|
|
},
|
|
{
|
|
"entropy": 4.9871738910675045,
|
|
"epoch": 4.426778242677824,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00035009133048587024,
|
|
"loss": 4.62,
|
|
"mean_token_accuracy": 0.22628631591796874,
|
|
"num_tokens": 9788543.0,
|
|
"step": 5290
|
|
},
|
|
{
|
|
"entropy": 4.936344194412231,
|
|
"epoch": 4.430962343096234,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003497869926293069,
|
|
"loss": 4.4999,
|
|
"mean_token_accuracy": 0.22995912730693818,
|
|
"num_tokens": 9797744.0,
|
|
"step": 5295
|
|
},
|
|
{
|
|
"entropy": 4.860135316848755,
|
|
"epoch": 4.435146443514644,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003494825008731367,
|
|
"loss": 4.5435,
|
|
"mean_token_accuracy": 0.22995650470256807,
|
|
"num_tokens": 9807528.0,
|
|
"step": 5300
|
|
},
|
|
{
|
|
"entropy": 4.858149576187134,
|
|
"epoch": 4.439330543933054,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00034917785584395457,
|
|
"loss": 4.5814,
|
|
"mean_token_accuracy": 0.21641302555799485,
|
|
"num_tokens": 9816078.0,
|
|
"step": 5305
|
|
},
|
|
{
|
|
"entropy": 4.986522626876831,
|
|
"epoch": 4.443514644351464,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000348873058168671,
|
|
"loss": 4.5243,
|
|
"mean_token_accuracy": 0.22923026829957963,
|
|
"num_tokens": 9825199.0,
|
|
"step": 5310
|
|
},
|
|
{
|
|
"entropy": 5.037327861785888,
|
|
"epoch": 4.447698744769874,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003485681084745106,
|
|
"loss": 4.6796,
|
|
"mean_token_accuracy": 0.2220054268836975,
|
|
"num_tokens": 9835294.0,
|
|
"step": 5315
|
|
},
|
|
{
|
|
"entropy": 4.888971042633057,
|
|
"epoch": 4.451882845188284,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000348263007389011,
|
|
"loss": 4.5468,
|
|
"mean_token_accuracy": 0.22776620239019393,
|
|
"num_tokens": 9844449.0,
|
|
"step": 5320
|
|
},
|
|
{
|
|
"entropy": 4.965271472930908,
|
|
"epoch": 4.456066945606695,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00034795775554002093,
|
|
"loss": 4.579,
|
|
"mean_token_accuracy": 0.22886669635772705,
|
|
"num_tokens": 9852930.0,
|
|
"step": 5325
|
|
},
|
|
{
|
|
"entropy": 4.960309028625488,
|
|
"epoch": 4.460251046025105,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034765235355569974,
|
|
"loss": 4.5114,
|
|
"mean_token_accuracy": 0.23735589534044266,
|
|
"num_tokens": 9862168.0,
|
|
"step": 5330
|
|
},
|
|
{
|
|
"entropy": 4.9397139072418215,
|
|
"epoch": 4.464435146443515,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003473468020645155,
|
|
"loss": 4.65,
|
|
"mean_token_accuracy": 0.22247984260320663,
|
|
"num_tokens": 9871702.0,
|
|
"step": 5335
|
|
},
|
|
{
|
|
"entropy": 5.026308345794678,
|
|
"epoch": 4.468619246861925,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003470411016952442,
|
|
"loss": 4.651,
|
|
"mean_token_accuracy": 0.22267691791057587,
|
|
"num_tokens": 9880502.0,
|
|
"step": 5340
|
|
},
|
|
{
|
|
"entropy": 5.03424038887024,
|
|
"epoch": 4.472803347280335,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000346735253076968,
|
|
"loss": 4.5827,
|
|
"mean_token_accuracy": 0.22348593920469284,
|
|
"num_tokens": 9889093.0,
|
|
"step": 5345
|
|
},
|
|
{
|
|
"entropy": 4.895762395858765,
|
|
"epoch": 4.476987447698745,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003464292568390743,
|
|
"loss": 4.5022,
|
|
"mean_token_accuracy": 0.22919296771287917,
|
|
"num_tokens": 9897113.0,
|
|
"step": 5350
|
|
},
|
|
{
|
|
"entropy": 4.848031091690063,
|
|
"epoch": 4.481171548117155,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003461231136112539,
|
|
"loss": 4.5473,
|
|
"mean_token_accuracy": 0.22644301056861876,
|
|
"num_tokens": 9906478.0,
|
|
"step": 5355
|
|
},
|
|
{
|
|
"entropy": 4.990407133102417,
|
|
"epoch": 4.485355648535565,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034581682402350057,
|
|
"loss": 4.6052,
|
|
"mean_token_accuracy": 0.2182432547211647,
|
|
"num_tokens": 9916143.0,
|
|
"step": 5360
|
|
},
|
|
{
|
|
"entropy": 5.014854001998901,
|
|
"epoch": 4.489539748953975,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00034551038870610885,
|
|
"loss": 4.5774,
|
|
"mean_token_accuracy": 0.22360296100378035,
|
|
"num_tokens": 9925675.0,
|
|
"step": 5365
|
|
},
|
|
{
|
|
"entropy": 4.895714998245239,
|
|
"epoch": 4.493723849372385,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00034520380828967346,
|
|
"loss": 4.5073,
|
|
"mean_token_accuracy": 0.23399453461170197,
|
|
"num_tokens": 9934398.0,
|
|
"step": 5370
|
|
},
|
|
{
|
|
"entropy": 4.866776657104492,
|
|
"epoch": 4.497907949790795,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003448970834050877,
|
|
"loss": 4.5683,
|
|
"mean_token_accuracy": 0.22729673385620117,
|
|
"num_tokens": 9943066.0,
|
|
"step": 5375
|
|
},
|
|
{
|
|
"entropy": 4.8104407787323,
|
|
"epoch": 4.502092050209205,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003445902146835419,
|
|
"loss": 4.4682,
|
|
"mean_token_accuracy": 0.23471412658691407,
|
|
"num_tokens": 9951565.0,
|
|
"step": 5380
|
|
},
|
|
{
|
|
"entropy": 4.941408252716064,
|
|
"epoch": 4.506276150627615,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003442832027565227,
|
|
"loss": 4.6224,
|
|
"mean_token_accuracy": 0.2161246970295906,
|
|
"num_tokens": 9960997.0,
|
|
"step": 5385
|
|
},
|
|
{
|
|
"entropy": 5.058872890472412,
|
|
"epoch": 4.510460251046025,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034397604825581115,
|
|
"loss": 4.5944,
|
|
"mean_token_accuracy": 0.22595355659723282,
|
|
"num_tokens": 9970188.0,
|
|
"step": 5390
|
|
},
|
|
{
|
|
"entropy": 4.906104230880738,
|
|
"epoch": 4.514644351464435,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003436687518134819,
|
|
"loss": 4.5932,
|
|
"mean_token_accuracy": 0.23194923996925354,
|
|
"num_tokens": 9978749.0,
|
|
"step": 5395
|
|
},
|
|
{
|
|
"entropy": 5.022583961486816,
|
|
"epoch": 4.518828451882845,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0003433613140619016,
|
|
"loss": 4.6676,
|
|
"mean_token_accuracy": 0.22514884024858475,
|
|
"num_tokens": 9988107.0,
|
|
"step": 5400
|
|
},
|
|
{
|
|
"entropy": 4.912742662429809,
|
|
"epoch": 4.523012552301255,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003430537356337277,
|
|
"loss": 4.4883,
|
|
"mean_token_accuracy": 0.2317280188202858,
|
|
"num_tokens": 9996127.0,
|
|
"step": 5405
|
|
},
|
|
{
|
|
"entropy": 4.966116046905517,
|
|
"epoch": 4.527196652719665,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003427460171619073,
|
|
"loss": 4.5992,
|
|
"mean_token_accuracy": 0.2261272132396698,
|
|
"num_tokens": 10005482.0,
|
|
"step": 5410
|
|
},
|
|
{
|
|
"entropy": 5.0073549270629885,
|
|
"epoch": 4.531380753138075,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034243815927967536,
|
|
"loss": 4.656,
|
|
"mean_token_accuracy": 0.21443560421466829,
|
|
"num_tokens": 10014765.0,
|
|
"step": 5415
|
|
},
|
|
{
|
|
"entropy": 4.961246395111084,
|
|
"epoch": 4.535564853556485,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034213016262055396,
|
|
"loss": 4.5992,
|
|
"mean_token_accuracy": 0.22173235267400743,
|
|
"num_tokens": 10022897.0,
|
|
"step": 5420
|
|
},
|
|
{
|
|
"entropy": 5.011214256286621,
|
|
"epoch": 4.539748953974895,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003418220278183507,
|
|
"loss": 4.6452,
|
|
"mean_token_accuracy": 0.21748925894498825,
|
|
"num_tokens": 10032281.0,
|
|
"step": 5425
|
|
},
|
|
{
|
|
"entropy": 4.897187662124634,
|
|
"epoch": 4.543933054393305,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00034151375550715753,
|
|
"loss": 4.4958,
|
|
"mean_token_accuracy": 0.23996804356575013,
|
|
"num_tokens": 10041194.0,
|
|
"step": 5430
|
|
},
|
|
{
|
|
"entropy": 4.824381732940674,
|
|
"epoch": 4.548117154811716,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00034120534632134917,
|
|
"loss": 4.5819,
|
|
"mean_token_accuracy": 0.2189012959599495,
|
|
"num_tokens": 10050062.0,
|
|
"step": 5435
|
|
},
|
|
{
|
|
"entropy": 4.925784158706665,
|
|
"epoch": 4.552301255230126,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00034089680089558226,
|
|
"loss": 4.6704,
|
|
"mean_token_accuracy": 0.2268177554011345,
|
|
"num_tokens": 10059747.0,
|
|
"step": 5440
|
|
},
|
|
{
|
|
"entropy": 5.006157302856446,
|
|
"epoch": 4.556485355648536,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00034058811986479367,
|
|
"loss": 4.6758,
|
|
"mean_token_accuracy": 0.21756935268640518,
|
|
"num_tokens": 10069351.0,
|
|
"step": 5445
|
|
},
|
|
{
|
|
"entropy": 4.99604344367981,
|
|
"epoch": 4.560669456066946,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00034027930386419935,
|
|
"loss": 4.6033,
|
|
"mean_token_accuracy": 0.22192914485931398,
|
|
"num_tokens": 10078727.0,
|
|
"step": 5450
|
|
},
|
|
{
|
|
"entropy": 4.893542909622193,
|
|
"epoch": 4.564853556485356,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033997035352929285,
|
|
"loss": 4.566,
|
|
"mean_token_accuracy": 0.22029880732297896,
|
|
"num_tokens": 10088151.0,
|
|
"step": 5455
|
|
},
|
|
{
|
|
"entropy": 4.9014698505401615,
|
|
"epoch": 4.569037656903766,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003396612694958444,
|
|
"loss": 4.5119,
|
|
"mean_token_accuracy": 0.22707427740097047,
|
|
"num_tokens": 10097311.0,
|
|
"step": 5460
|
|
},
|
|
{
|
|
"entropy": 4.953253602981567,
|
|
"epoch": 4.573221757322176,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00033935205239989927,
|
|
"loss": 4.6545,
|
|
"mean_token_accuracy": 0.21970983892679213,
|
|
"num_tokens": 10107474.0,
|
|
"step": 5465
|
|
},
|
|
{
|
|
"entropy": 4.941572809219361,
|
|
"epoch": 4.577405857740586,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00033904270287777645,
|
|
"loss": 4.5788,
|
|
"mean_token_accuracy": 0.23626233637332916,
|
|
"num_tokens": 10116305.0,
|
|
"step": 5470
|
|
},
|
|
{
|
|
"entropy": 4.974158048629761,
|
|
"epoch": 4.581589958158996,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003387332215660675,
|
|
"loss": 4.6254,
|
|
"mean_token_accuracy": 0.22385287880897523,
|
|
"num_tokens": 10125947.0,
|
|
"step": 5475
|
|
},
|
|
{
|
|
"entropy": 4.920748901367188,
|
|
"epoch": 4.585774058577406,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00033842360910163536,
|
|
"loss": 4.6043,
|
|
"mean_token_accuracy": 0.22699405997991562,
|
|
"num_tokens": 10134510.0,
|
|
"step": 5480
|
|
},
|
|
{
|
|
"entropy": 5.012820720672607,
|
|
"epoch": 4.589958158995816,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003381138661216125,
|
|
"loss": 4.6569,
|
|
"mean_token_accuracy": 0.2172386333346367,
|
|
"num_tokens": 10143429.0,
|
|
"step": 5485
|
|
},
|
|
{
|
|
"entropy": 4.997296905517578,
|
|
"epoch": 4.594142259414226,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00033780399326340026,
|
|
"loss": 4.6534,
|
|
"mean_token_accuracy": 0.21922401934862137,
|
|
"num_tokens": 10153405.0,
|
|
"step": 5490
|
|
},
|
|
{
|
|
"entropy": 4.978293132781983,
|
|
"epoch": 4.598326359832636,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00033749399116466723,
|
|
"loss": 4.5939,
|
|
"mean_token_accuracy": 0.2325430616736412,
|
|
"num_tokens": 10162334.0,
|
|
"step": 5495
|
|
},
|
|
{
|
|
"entropy": 4.909941673278809,
|
|
"epoch": 4.602510460251046,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003371838604633479,
|
|
"loss": 4.5989,
|
|
"mean_token_accuracy": 0.22104209065437316,
|
|
"num_tokens": 10171604.0,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"epoch": 4.602510460251046,
|
|
"eval_entropy": 4.740977853841333,
|
|
"eval_loss": 5.372191429138184,
|
|
"eval_mean_token_accuracy": 0.1966732437960453,
|
|
"eval_num_tokens": 10171604.0,
|
|
"eval_runtime": 2.7422,
|
|
"eval_samples_per_second": 1424.75,
|
|
"eval_steps_per_second": 178.322,
|
|
"step": 5500
|
|
},
|
|
{
|
|
"entropy": 4.887475299835205,
|
|
"epoch": 4.606694560669456,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00033687360179764125,
|
|
"loss": 4.5793,
|
|
"mean_token_accuracy": 0.22717299312353134,
|
|
"num_tokens": 10180659.0,
|
|
"step": 5505
|
|
},
|
|
{
|
|
"entropy": 4.983587646484375,
|
|
"epoch": 4.610878661087866,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003365632158060098,
|
|
"loss": 4.6308,
|
|
"mean_token_accuracy": 0.22400773018598558,
|
|
"num_tokens": 10190226.0,
|
|
"step": 5510
|
|
},
|
|
{
|
|
"entropy": 4.970303010940552,
|
|
"epoch": 4.615062761506276,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033625270312717805,
|
|
"loss": 4.5205,
|
|
"mean_token_accuracy": 0.2294809028506279,
|
|
"num_tokens": 10198806.0,
|
|
"step": 5515
|
|
},
|
|
{
|
|
"entropy": 4.902194929122925,
|
|
"epoch": 4.619246861924686,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003359420644001311,
|
|
"loss": 4.5349,
|
|
"mean_token_accuracy": 0.22978270202875137,
|
|
"num_tokens": 10207927.0,
|
|
"step": 5520
|
|
},
|
|
{
|
|
"entropy": 4.841322231292724,
|
|
"epoch": 4.623430962343097,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00033563130026411357,
|
|
"loss": 4.4779,
|
|
"mean_token_accuracy": 0.23885904997587204,
|
|
"num_tokens": 10216241.0,
|
|
"step": 5525
|
|
},
|
|
{
|
|
"entropy": 4.939700937271118,
|
|
"epoch": 4.627615062761507,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00033532041135862816,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.21889656335115432,
|
|
"num_tokens": 10225241.0,
|
|
"step": 5530
|
|
},
|
|
{
|
|
"entropy": 4.953241205215454,
|
|
"epoch": 4.631799163179917,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00033500939832343407,
|
|
"loss": 4.6266,
|
|
"mean_token_accuracy": 0.22123186588287352,
|
|
"num_tokens": 10233968.0,
|
|
"step": 5535
|
|
},
|
|
{
|
|
"entropy": 5.0703329086303714,
|
|
"epoch": 4.635983263598327,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003346982617985463,
|
|
"loss": 4.7037,
|
|
"mean_token_accuracy": 0.21673406958580016,
|
|
"num_tokens": 10243589.0,
|
|
"step": 5540
|
|
},
|
|
{
|
|
"entropy": 4.869228839874268,
|
|
"epoch": 4.640167364016737,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003343870024242337,
|
|
"loss": 4.5312,
|
|
"mean_token_accuracy": 0.230489881336689,
|
|
"num_tokens": 10252416.0,
|
|
"step": 5545
|
|
},
|
|
{
|
|
"entropy": 4.973748064041137,
|
|
"epoch": 4.644351464435147,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00033407562084101795,
|
|
"loss": 4.6636,
|
|
"mean_token_accuracy": 0.21605071276426316,
|
|
"num_tokens": 10262213.0,
|
|
"step": 5550
|
|
},
|
|
{
|
|
"entropy": 5.023733329772949,
|
|
"epoch": 4.648535564853557,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033376411768967247,
|
|
"loss": 4.7218,
|
|
"mean_token_accuracy": 0.2145490512251854,
|
|
"num_tokens": 10272774.0,
|
|
"step": 5555
|
|
},
|
|
{
|
|
"entropy": 4.9784989833831785,
|
|
"epoch": 4.652719665271967,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033345249361122056,
|
|
"loss": 4.5597,
|
|
"mean_token_accuracy": 0.22975188195705415,
|
|
"num_tokens": 10281286.0,
|
|
"step": 5560
|
|
},
|
|
{
|
|
"entropy": 4.877393484115601,
|
|
"epoch": 4.656903765690377,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033314074924693446,
|
|
"loss": 4.5676,
|
|
"mean_token_accuracy": 0.225330650806427,
|
|
"num_tokens": 10290101.0,
|
|
"step": 5565
|
|
},
|
|
{
|
|
"entropy": 4.889590644836426,
|
|
"epoch": 4.661087866108787,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00033282888523833394,
|
|
"loss": 4.5945,
|
|
"mean_token_accuracy": 0.22907819598913193,
|
|
"num_tokens": 10299580.0,
|
|
"step": 5570
|
|
},
|
|
{
|
|
"entropy": 4.993817472457886,
|
|
"epoch": 4.665271966527197,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00033251690222718504,
|
|
"loss": 4.5886,
|
|
"mean_token_accuracy": 0.22838464975357056,
|
|
"num_tokens": 10307860.0,
|
|
"step": 5575
|
|
},
|
|
{
|
|
"entropy": 4.92207088470459,
|
|
"epoch": 4.669456066945607,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003322048008554986,
|
|
"loss": 4.6053,
|
|
"mean_token_accuracy": 0.23097976744174958,
|
|
"num_tokens": 10316546.0,
|
|
"step": 5580
|
|
},
|
|
{
|
|
"entropy": 4.986917877197266,
|
|
"epoch": 4.673640167364017,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003318925817655291,
|
|
"loss": 4.7004,
|
|
"mean_token_accuracy": 0.22112769931554793,
|
|
"num_tokens": 10326012.0,
|
|
"step": 5585
|
|
},
|
|
{
|
|
"entropy": 5.066364097595215,
|
|
"epoch": 4.677824267782427,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003315802455997732,
|
|
"loss": 4.6276,
|
|
"mean_token_accuracy": 0.22098480463027953,
|
|
"num_tokens": 10334771.0,
|
|
"step": 5590
|
|
},
|
|
{
|
|
"entropy": 5.036166429519653,
|
|
"epoch": 4.682008368200837,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00033126779300096855,
|
|
"loss": 4.6326,
|
|
"mean_token_accuracy": 0.22032282501459122,
|
|
"num_tokens": 10343416.0,
|
|
"step": 5595
|
|
},
|
|
{
|
|
"entropy": 5.0165605545043945,
|
|
"epoch": 4.686192468619247,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003309552246120924,
|
|
"loss": 4.7252,
|
|
"mean_token_accuracy": 0.21418508887290955,
|
|
"num_tokens": 10353539.0,
|
|
"step": 5600
|
|
},
|
|
{
|
|
"entropy": 4.919169235229492,
|
|
"epoch": 4.690376569037657,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0003306425410763601,
|
|
"loss": 4.544,
|
|
"mean_token_accuracy": 0.22654212713241578,
|
|
"num_tokens": 10362020.0,
|
|
"step": 5605
|
|
},
|
|
{
|
|
"entropy": 4.922832107543945,
|
|
"epoch": 4.694560669456067,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003303297430372241,
|
|
"loss": 4.4898,
|
|
"mean_token_accuracy": 0.23462637960910798,
|
|
"num_tokens": 10371277.0,
|
|
"step": 5610
|
|
},
|
|
{
|
|
"entropy": 4.900220251083374,
|
|
"epoch": 4.698744769874477,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003300168311383727,
|
|
"loss": 4.5952,
|
|
"mean_token_accuracy": 0.22010781317949296,
|
|
"num_tokens": 10380680.0,
|
|
"step": 5615
|
|
},
|
|
{
|
|
"entropy": 4.950555086135864,
|
|
"epoch": 4.702928870292887,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032970380602372805,
|
|
"loss": 4.5366,
|
|
"mean_token_accuracy": 0.23083105981349944,
|
|
"num_tokens": 10389576.0,
|
|
"step": 5620
|
|
},
|
|
{
|
|
"entropy": 4.957669401168824,
|
|
"epoch": 4.707112970711297,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003293906683374456,
|
|
"loss": 4.6401,
|
|
"mean_token_accuracy": 0.22613726258277894,
|
|
"num_tokens": 10399182.0,
|
|
"step": 5625
|
|
},
|
|
{
|
|
"entropy": 5.064533948898315,
|
|
"epoch": 4.711297071129707,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003290774187239124,
|
|
"loss": 4.7436,
|
|
"mean_token_accuracy": 0.21829744875431062,
|
|
"num_tokens": 10409595.0,
|
|
"step": 5630
|
|
},
|
|
{
|
|
"entropy": 5.046321868896484,
|
|
"epoch": 4.7154811715481175,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032876405782774574,
|
|
"loss": 4.718,
|
|
"mean_token_accuracy": 0.20706165879964827,
|
|
"num_tokens": 10419057.0,
|
|
"step": 5635
|
|
},
|
|
{
|
|
"entropy": 4.861374807357788,
|
|
"epoch": 4.7196652719665275,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003284505862937919,
|
|
"loss": 4.4911,
|
|
"mean_token_accuracy": 0.236003278195858,
|
|
"num_tokens": 10428090.0,
|
|
"step": 5640
|
|
},
|
|
{
|
|
"entropy": 4.956670045852661,
|
|
"epoch": 4.7238493723849375,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.000328137004767125,
|
|
"loss": 4.6555,
|
|
"mean_token_accuracy": 0.21835391521453856,
|
|
"num_tokens": 10438088.0,
|
|
"step": 5645
|
|
},
|
|
{
|
|
"entropy": 4.942261695861816,
|
|
"epoch": 4.7280334728033475,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003278233138930454,
|
|
"loss": 4.6475,
|
|
"mean_token_accuracy": 0.2236612021923065,
|
|
"num_tokens": 10447934.0,
|
|
"step": 5650
|
|
},
|
|
{
|
|
"entropy": 4.938869476318359,
|
|
"epoch": 4.7322175732217575,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003275095143170784,
|
|
"loss": 4.5081,
|
|
"mean_token_accuracy": 0.23616990745067595,
|
|
"num_tokens": 10456088.0,
|
|
"step": 5655
|
|
},
|
|
{
|
|
"entropy": 4.891930294036865,
|
|
"epoch": 4.7364016736401675,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032719560668497316,
|
|
"loss": 4.512,
|
|
"mean_token_accuracy": 0.23732430785894393,
|
|
"num_tokens": 10465096.0,
|
|
"step": 5660
|
|
},
|
|
{
|
|
"entropy": 4.837410831451416,
|
|
"epoch": 4.7405857740585775,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00032688159164270114,
|
|
"loss": 4.6032,
|
|
"mean_token_accuracy": 0.23223818242549896,
|
|
"num_tokens": 10473516.0,
|
|
"step": 5665
|
|
},
|
|
{
|
|
"entropy": 4.911863327026367,
|
|
"epoch": 4.7447698744769875,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00032656746983645473,
|
|
"loss": 4.5885,
|
|
"mean_token_accuracy": 0.23326796442270278,
|
|
"num_tokens": 10482651.0,
|
|
"step": 5670
|
|
},
|
|
{
|
|
"entropy": 4.991975736618042,
|
|
"epoch": 4.7489539748953975,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003262532419126461,
|
|
"loss": 4.6048,
|
|
"mean_token_accuracy": 0.21997690051794053,
|
|
"num_tokens": 10491831.0,
|
|
"step": 5675
|
|
},
|
|
{
|
|
"entropy": 4.922548532485962,
|
|
"epoch": 4.7531380753138075,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00032593890851790593,
|
|
"loss": 4.6121,
|
|
"mean_token_accuracy": 0.22672457247972488,
|
|
"num_tokens": 10502070.0,
|
|
"step": 5680
|
|
},
|
|
{
|
|
"entropy": 4.909999513626099,
|
|
"epoch": 4.7573221757322175,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003256244702990817,
|
|
"loss": 4.5931,
|
|
"mean_token_accuracy": 0.22097909450531006,
|
|
"num_tokens": 10510638.0,
|
|
"step": 5685
|
|
},
|
|
{
|
|
"entropy": 4.8751520156860355,
|
|
"epoch": 4.7615062761506275,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032530992790323664,
|
|
"loss": 4.5393,
|
|
"mean_token_accuracy": 0.2271535113453865,
|
|
"num_tokens": 10519305.0,
|
|
"step": 5690
|
|
},
|
|
{
|
|
"entropy": 4.947201585769653,
|
|
"epoch": 4.7656903765690375,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00032499528197764855,
|
|
"loss": 4.5824,
|
|
"mean_token_accuracy": 0.22470725029706956,
|
|
"num_tokens": 10528844.0,
|
|
"step": 5695
|
|
},
|
|
{
|
|
"entropy": 4.997877025604248,
|
|
"epoch": 4.7698744769874475,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00032468053316980815,
|
|
"loss": 4.6914,
|
|
"mean_token_accuracy": 0.2203351765871048,
|
|
"num_tokens": 10538260.0,
|
|
"step": 5700
|
|
},
|
|
{
|
|
"entropy": 4.947293424606324,
|
|
"epoch": 4.7740585774058575,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003243656821274178,
|
|
"loss": 4.6144,
|
|
"mean_token_accuracy": 0.22794263809919357,
|
|
"num_tokens": 10547733.0,
|
|
"step": 5705
|
|
},
|
|
{
|
|
"entropy": 4.9433341979980465,
|
|
"epoch": 4.7782426778242675,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003240507294983905,
|
|
"loss": 4.5895,
|
|
"mean_token_accuracy": 0.23046046495437622,
|
|
"num_tokens": 10558337.0,
|
|
"step": 5710
|
|
},
|
|
{
|
|
"entropy": 4.946340084075928,
|
|
"epoch": 4.7824267782426775,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00032373567593084804,
|
|
"loss": 4.5879,
|
|
"mean_token_accuracy": 0.22187992036342621,
|
|
"num_tokens": 10568374.0,
|
|
"step": 5715
|
|
},
|
|
{
|
|
"entropy": 4.895954942703247,
|
|
"epoch": 4.786610878661088,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032342052207311986,
|
|
"loss": 4.6648,
|
|
"mean_token_accuracy": 0.22497455179691314,
|
|
"num_tokens": 10578448.0,
|
|
"step": 5720
|
|
},
|
|
{
|
|
"entropy": 5.00159969329834,
|
|
"epoch": 4.790794979079498,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00032310526857374204,
|
|
"loss": 4.6788,
|
|
"mean_token_accuracy": 0.2173873543739319,
|
|
"num_tokens": 10586803.0,
|
|
"step": 5725
|
|
},
|
|
{
|
|
"entropy": 5.052724123001099,
|
|
"epoch": 4.794979079497908,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003227899160814556,
|
|
"loss": 4.665,
|
|
"mean_token_accuracy": 0.22732132971286773,
|
|
"num_tokens": 10596596.0,
|
|
"step": 5730
|
|
},
|
|
{
|
|
"entropy": 4.942400026321411,
|
|
"epoch": 4.799163179916318,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0003224744652452054,
|
|
"loss": 4.5447,
|
|
"mean_token_accuracy": 0.22640974670648575,
|
|
"num_tokens": 10605462.0,
|
|
"step": 5735
|
|
},
|
|
{
|
|
"entropy": 5.037791347503662,
|
|
"epoch": 4.803347280334728,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003221589167141382,
|
|
"loss": 4.6961,
|
|
"mean_token_accuracy": 0.2149311438202858,
|
|
"num_tokens": 10614908.0,
|
|
"step": 5740
|
|
},
|
|
{
|
|
"entropy": 4.993225717544556,
|
|
"epoch": 4.807531380753138,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003218432711376025,
|
|
"loss": 4.6659,
|
|
"mean_token_accuracy": 0.223326176404953,
|
|
"num_tokens": 10624389.0,
|
|
"step": 5745
|
|
},
|
|
{
|
|
"entropy": 4.934679269790649,
|
|
"epoch": 4.811715481171548,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00032152752916514605,
|
|
"loss": 4.6163,
|
|
"mean_token_accuracy": 0.22118015885353087,
|
|
"num_tokens": 10633228.0,
|
|
"step": 5750
|
|
},
|
|
{
|
|
"entropy": 4.92312331199646,
|
|
"epoch": 4.815899581589958,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00032121169144651496,
|
|
"loss": 4.6684,
|
|
"mean_token_accuracy": 0.22240606248378753,
|
|
"num_tokens": 10641996.0,
|
|
"step": 5755
|
|
},
|
|
{
|
|
"entropy": 4.911015844345092,
|
|
"epoch": 4.820083682008368,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003208957586316525,
|
|
"loss": 4.5766,
|
|
"mean_token_accuracy": 0.23004275560379028,
|
|
"num_tokens": 10651724.0,
|
|
"step": 5760
|
|
},
|
|
{
|
|
"entropy": 4.993270492553711,
|
|
"epoch": 4.824267782426778,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003205797313706977,
|
|
"loss": 4.6387,
|
|
"mean_token_accuracy": 0.22105501741170883,
|
|
"num_tokens": 10661150.0,
|
|
"step": 5765
|
|
},
|
|
{
|
|
"entropy": 4.958125162124634,
|
|
"epoch": 4.828451882845188,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00032026361031398386,
|
|
"loss": 4.6501,
|
|
"mean_token_accuracy": 0.22121685445308686,
|
|
"num_tokens": 10669890.0,
|
|
"step": 5770
|
|
},
|
|
{
|
|
"entropy": 4.986698198318481,
|
|
"epoch": 4.832635983263598,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00031994739611203716,
|
|
"loss": 4.5767,
|
|
"mean_token_accuracy": 0.22113114446401597,
|
|
"num_tokens": 10678939.0,
|
|
"step": 5775
|
|
},
|
|
{
|
|
"entropy": 4.895010232925415,
|
|
"epoch": 4.836820083682008,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031963108941557576,
|
|
"loss": 4.5316,
|
|
"mean_token_accuracy": 0.2235198900103569,
|
|
"num_tokens": 10687465.0,
|
|
"step": 5780
|
|
},
|
|
{
|
|
"entropy": 4.952823734283447,
|
|
"epoch": 4.841004184100418,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003193146908755078,
|
|
"loss": 4.6273,
|
|
"mean_token_accuracy": 0.22349171638488768,
|
|
"num_tokens": 10696930.0,
|
|
"step": 5785
|
|
},
|
|
{
|
|
"entropy": 4.933053779602051,
|
|
"epoch": 4.845188284518828,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00031899820114293073,
|
|
"loss": 4.5579,
|
|
"mean_token_accuracy": 0.23062346279621124,
|
|
"num_tokens": 10706820.0,
|
|
"step": 5790
|
|
},
|
|
{
|
|
"entropy": 4.900616836547852,
|
|
"epoch": 4.849372384937238,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031868162086912946,
|
|
"loss": 4.5426,
|
|
"mean_token_accuracy": 0.23274454176425935,
|
|
"num_tokens": 10715528.0,
|
|
"step": 5795
|
|
},
|
|
{
|
|
"entropy": 4.883484935760498,
|
|
"epoch": 4.853556485355648,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0003183649507055754,
|
|
"loss": 4.5456,
|
|
"mean_token_accuracy": 0.22708174139261245,
|
|
"num_tokens": 10724807.0,
|
|
"step": 5800
|
|
},
|
|
{
|
|
"entropy": 4.981265640258789,
|
|
"epoch": 4.857740585774058,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00031804819130392477,
|
|
"loss": 4.6253,
|
|
"mean_token_accuracy": 0.2187033474445343,
|
|
"num_tokens": 10733754.0,
|
|
"step": 5805
|
|
},
|
|
{
|
|
"entropy": 4.909184551239013,
|
|
"epoch": 4.861924686192468,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003177313433160175,
|
|
"loss": 4.6272,
|
|
"mean_token_accuracy": 0.2266688421368599,
|
|
"num_tokens": 10742256.0,
|
|
"step": 5810
|
|
},
|
|
{
|
|
"entropy": 4.913168096542359,
|
|
"epoch": 4.866108786610878,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0003174144073938757,
|
|
"loss": 4.5663,
|
|
"mean_token_accuracy": 0.24062089920043944,
|
|
"num_tokens": 10751025.0,
|
|
"step": 5815
|
|
},
|
|
{
|
|
"entropy": 4.970402002334595,
|
|
"epoch": 4.870292887029288,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00031709738418970275,
|
|
"loss": 4.5441,
|
|
"mean_token_accuracy": 0.226358999311924,
|
|
"num_tokens": 10759970.0,
|
|
"step": 5820
|
|
},
|
|
{
|
|
"entropy": 4.928719329833984,
|
|
"epoch": 4.874476987447698,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003167802743558813,
|
|
"loss": 4.5738,
|
|
"mean_token_accuracy": 0.23167754709720612,
|
|
"num_tokens": 10768531.0,
|
|
"step": 5825
|
|
},
|
|
{
|
|
"entropy": 4.982670783996582,
|
|
"epoch": 4.878661087866108,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003164630785449724,
|
|
"loss": 4.6675,
|
|
"mean_token_accuracy": 0.22024451047182084,
|
|
"num_tokens": 10778802.0,
|
|
"step": 5830
|
|
},
|
|
{
|
|
"entropy": 4.975974416732788,
|
|
"epoch": 4.882845188284519,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00031614579740971415,
|
|
"loss": 4.5869,
|
|
"mean_token_accuracy": 0.2243398532271385,
|
|
"num_tokens": 10788121.0,
|
|
"step": 5835
|
|
},
|
|
{
|
|
"entropy": 5.031265211105347,
|
|
"epoch": 4.887029288702929,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00031582843160302,
|
|
"loss": 4.6973,
|
|
"mean_token_accuracy": 0.22460703402757645,
|
|
"num_tokens": 10797652.0,
|
|
"step": 5840
|
|
},
|
|
{
|
|
"entropy": 4.952078771591187,
|
|
"epoch": 4.891213389121339,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00031551098177797784,
|
|
"loss": 4.6155,
|
|
"mean_token_accuracy": 0.21827297359704972,
|
|
"num_tokens": 10807564.0,
|
|
"step": 5845
|
|
},
|
|
{
|
|
"entropy": 4.956915235519409,
|
|
"epoch": 4.895397489539749,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0003151934485878482,
|
|
"loss": 4.5822,
|
|
"mean_token_accuracy": 0.22356773763895035,
|
|
"num_tokens": 10816666.0,
|
|
"step": 5850
|
|
},
|
|
{
|
|
"entropy": 4.910748243331909,
|
|
"epoch": 4.899581589958159,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031487583268606354,
|
|
"loss": 4.5951,
|
|
"mean_token_accuracy": 0.23078953623771667,
|
|
"num_tokens": 10825639.0,
|
|
"step": 5855
|
|
},
|
|
{
|
|
"entropy": 4.929580736160278,
|
|
"epoch": 4.903765690376569,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003145581347262262,
|
|
"loss": 4.6361,
|
|
"mean_token_accuracy": 0.2225961908698082,
|
|
"num_tokens": 10834681.0,
|
|
"step": 5860
|
|
},
|
|
{
|
|
"entropy": 4.958722543716431,
|
|
"epoch": 4.907949790794979,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031424035536210755,
|
|
"loss": 4.6336,
|
|
"mean_token_accuracy": 0.22361528426408767,
|
|
"num_tokens": 10844465.0,
|
|
"step": 5865
|
|
},
|
|
{
|
|
"entropy": 4.956167221069336,
|
|
"epoch": 4.912133891213389,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00031392249524764645,
|
|
"loss": 4.6491,
|
|
"mean_token_accuracy": 0.2189427360892296,
|
|
"num_tokens": 10853327.0,
|
|
"step": 5870
|
|
},
|
|
{
|
|
"entropy": 4.902080535888672,
|
|
"epoch": 4.916317991631799,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00031360455503694787,
|
|
"loss": 4.5535,
|
|
"mean_token_accuracy": 0.22752024382352828,
|
|
"num_tokens": 10861865.0,
|
|
"step": 5875
|
|
},
|
|
{
|
|
"entropy": 4.999043941497803,
|
|
"epoch": 4.920502092050209,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003132865353842816,
|
|
"loss": 4.6404,
|
|
"mean_token_accuracy": 0.22591783106327057,
|
|
"num_tokens": 10870794.0,
|
|
"step": 5880
|
|
},
|
|
{
|
|
"entropy": 4.985387754440308,
|
|
"epoch": 4.924686192468619,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00031296843694408107,
|
|
"loss": 4.5296,
|
|
"mean_token_accuracy": 0.23055646121501921,
|
|
"num_tokens": 10879626.0,
|
|
"step": 5885
|
|
},
|
|
{
|
|
"entropy": 4.8676410675048825,
|
|
"epoch": 4.928870292887029,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0003126502603709417,
|
|
"loss": 4.4968,
|
|
"mean_token_accuracy": 0.23666098713874817,
|
|
"num_tokens": 10888359.0,
|
|
"step": 5890
|
|
},
|
|
{
|
|
"entropy": 4.9276433944702145,
|
|
"epoch": 4.933054393305439,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003123320063196196,
|
|
"loss": 4.5712,
|
|
"mean_token_accuracy": 0.22536735236644745,
|
|
"num_tokens": 10897174.0,
|
|
"step": 5895
|
|
},
|
|
{
|
|
"entropy": 4.897340297698975,
|
|
"epoch": 4.937238493723849,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003120136754450306,
|
|
"loss": 4.6372,
|
|
"mean_token_accuracy": 0.21138296276330948,
|
|
"num_tokens": 10906660.0,
|
|
"step": 5900
|
|
},
|
|
{
|
|
"entropy": 4.96869912147522,
|
|
"epoch": 4.941422594142259,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00031169526840224836,
|
|
"loss": 4.6517,
|
|
"mean_token_accuracy": 0.22115846276283263,
|
|
"num_tokens": 10915990.0,
|
|
"step": 5905
|
|
},
|
|
{
|
|
"entropy": 5.087694931030273,
|
|
"epoch": 4.945606694560669,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003113767858465034,
|
|
"loss": 4.7703,
|
|
"mean_token_accuracy": 0.20570726841688156,
|
|
"num_tokens": 10926355.0,
|
|
"step": 5910
|
|
},
|
|
{
|
|
"entropy": 4.9310167789459225,
|
|
"epoch": 4.949790794979079,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00031105822843318173,
|
|
"loss": 4.6186,
|
|
"mean_token_accuracy": 0.22633125633001328,
|
|
"num_tokens": 10935854.0,
|
|
"step": 5915
|
|
},
|
|
{
|
|
"entropy": 4.902202558517456,
|
|
"epoch": 4.95397489539749,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003107395968178232,
|
|
"loss": 4.5621,
|
|
"mean_token_accuracy": 0.23685746043920516,
|
|
"num_tokens": 10944903.0,
|
|
"step": 5920
|
|
},
|
|
{
|
|
"entropy": 4.9367992877960205,
|
|
"epoch": 4.9581589958159,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003104208916561203,
|
|
"loss": 4.5808,
|
|
"mean_token_accuracy": 0.22444928884506227,
|
|
"num_tokens": 10954009.0,
|
|
"step": 5925
|
|
},
|
|
{
|
|
"entropy": 5.0358720302581785,
|
|
"epoch": 4.96234309623431,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00031010211360391727,
|
|
"loss": 4.7337,
|
|
"mean_token_accuracy": 0.21483712196350097,
|
|
"num_tokens": 10963174.0,
|
|
"step": 5930
|
|
},
|
|
{
|
|
"entropy": 5.040430402755737,
|
|
"epoch": 4.96652719665272,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000309783263317208,
|
|
"loss": 4.6536,
|
|
"mean_token_accuracy": 0.22425227463245392,
|
|
"num_tokens": 10972822.0,
|
|
"step": 5935
|
|
},
|
|
{
|
|
"entropy": 4.925096464157105,
|
|
"epoch": 4.97071129707113,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0003094643414521349,
|
|
"loss": 4.5443,
|
|
"mean_token_accuracy": 0.2292025253176689,
|
|
"num_tokens": 10981758.0,
|
|
"step": 5940
|
|
},
|
|
{
|
|
"entropy": 4.987518978118897,
|
|
"epoch": 4.97489539748954,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003091453486649883,
|
|
"loss": 4.615,
|
|
"mean_token_accuracy": 0.22280609756708145,
|
|
"num_tokens": 10991730.0,
|
|
"step": 5945
|
|
},
|
|
{
|
|
"entropy": 5.006465482711792,
|
|
"epoch": 4.97907949790795,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030882628561220376,
|
|
"loss": 4.6297,
|
|
"mean_token_accuracy": 0.2240285411477089,
|
|
"num_tokens": 11000827.0,
|
|
"step": 5950
|
|
},
|
|
{
|
|
"entropy": 4.955653953552246,
|
|
"epoch": 4.98326359832636,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00030850715295036187,
|
|
"loss": 4.5832,
|
|
"mean_token_accuracy": 0.23452527970075607,
|
|
"num_tokens": 11010544.0,
|
|
"step": 5955
|
|
},
|
|
{
|
|
"entropy": 4.943860340118408,
|
|
"epoch": 4.98744769874477,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0003081879513361863,
|
|
"loss": 4.6284,
|
|
"mean_token_accuracy": 0.22700540870428085,
|
|
"num_tokens": 11019826.0,
|
|
"step": 5960
|
|
},
|
|
{
|
|
"entropy": 4.988158321380615,
|
|
"epoch": 4.99163179916318,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0003078686814265426,
|
|
"loss": 4.6034,
|
|
"mean_token_accuracy": 0.23255193382501602,
|
|
"num_tokens": 11028408.0,
|
|
"step": 5965
|
|
},
|
|
{
|
|
"entropy": 5.002375984191895,
|
|
"epoch": 4.99581589958159,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000307549343878437,
|
|
"loss": 4.6565,
|
|
"mean_token_accuracy": 0.22523596584796907,
|
|
"num_tokens": 11038564.0,
|
|
"step": 5970
|
|
},
|
|
{
|
|
"entropy": 4.932086610794068,
|
|
"epoch": 5.0,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00030722993934901487,
|
|
"loss": 4.5085,
|
|
"mean_token_accuracy": 0.2353028506040573,
|
|
"num_tokens": 11046610.0,
|
|
"step": 5975
|
|
},
|
|
{
|
|
"entropy": 4.824899530410766,
|
|
"epoch": 5.00418410041841,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00030691046849555927,
|
|
"loss": 4.2443,
|
|
"mean_token_accuracy": 0.2535263478755951,
|
|
"num_tokens": 11055929.0,
|
|
"step": 5980
|
|
},
|
|
{
|
|
"entropy": 4.956370878219604,
|
|
"epoch": 5.00836820083682,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00030659093197549,
|
|
"loss": 4.3197,
|
|
"mean_token_accuracy": 0.2461391270160675,
|
|
"num_tokens": 11064748.0,
|
|
"step": 5985
|
|
},
|
|
{
|
|
"entropy": 4.857566976547242,
|
|
"epoch": 5.01255230125523,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003062713304463616,
|
|
"loss": 4.2977,
|
|
"mean_token_accuracy": 0.2529305458068848,
|
|
"num_tokens": 11073707.0,
|
|
"step": 5990
|
|
},
|
|
{
|
|
"entropy": 4.909223031997681,
|
|
"epoch": 5.01673640167364,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003059516645658629,
|
|
"loss": 4.3224,
|
|
"mean_token_accuracy": 0.2479591742157936,
|
|
"num_tokens": 11082967.0,
|
|
"step": 5995
|
|
},
|
|
{
|
|
"entropy": 4.86334342956543,
|
|
"epoch": 5.02092050209205,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003056319349918146,
|
|
"loss": 4.3189,
|
|
"mean_token_accuracy": 0.24600037038326264,
|
|
"num_tokens": 11092105.0,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"epoch": 5.02092050209205,
|
|
"eval_entropy": 4.730428561598733,
|
|
"eval_loss": 5.358774185180664,
|
|
"eval_mean_token_accuracy": 0.19859819039550053,
|
|
"eval_num_tokens": 11092105.0,
|
|
"eval_runtime": 2.7257,
|
|
"eval_samples_per_second": 1433.383,
|
|
"eval_steps_per_second": 179.402,
|
|
"step": 6000
|
|
},
|
|
{
|
|
"entropy": 4.9469725608825685,
|
|
"epoch": 5.02510460251046,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00030531214238216906,
|
|
"loss": 4.4013,
|
|
"mean_token_accuracy": 0.23444748818874359,
|
|
"num_tokens": 11100806.0,
|
|
"step": 6005
|
|
},
|
|
{
|
|
"entropy": 4.852447938919068,
|
|
"epoch": 5.02928870292887,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030499228739500793,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.25530788451433184,
|
|
"num_tokens": 11109779.0,
|
|
"step": 6010
|
|
},
|
|
{
|
|
"entropy": 4.888129472732544,
|
|
"epoch": 5.03347280334728,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00030467237068854136,
|
|
"loss": 4.3577,
|
|
"mean_token_accuracy": 0.23798221200704575,
|
|
"num_tokens": 11118838.0,
|
|
"step": 6015
|
|
},
|
|
{
|
|
"entropy": 4.842834329605102,
|
|
"epoch": 5.03765690376569,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003043523929211064,
|
|
"loss": 4.2808,
|
|
"mean_token_accuracy": 0.2523635119199753,
|
|
"num_tokens": 11127029.0,
|
|
"step": 6020
|
|
},
|
|
{
|
|
"entropy": 4.868286514282227,
|
|
"epoch": 5.0418410041841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000304032354751166,
|
|
"loss": 4.3359,
|
|
"mean_token_accuracy": 0.24998421221971512,
|
|
"num_tokens": 11136073.0,
|
|
"step": 6025
|
|
},
|
|
{
|
|
"entropy": 4.886257982254028,
|
|
"epoch": 5.046025104602511,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003037122568373071,
|
|
"loss": 4.3749,
|
|
"mean_token_accuracy": 0.24850379228591918,
|
|
"num_tokens": 11144807.0,
|
|
"step": 6030
|
|
},
|
|
{
|
|
"entropy": 4.9287115097045895,
|
|
"epoch": 5.050209205020921,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003033920998382399,
|
|
"loss": 4.3561,
|
|
"mean_token_accuracy": 0.23847561180591584,
|
|
"num_tokens": 11154308.0,
|
|
"step": 6035
|
|
},
|
|
{
|
|
"entropy": 4.931772375106812,
|
|
"epoch": 5.054393305439331,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003030718844127959,
|
|
"loss": 4.3909,
|
|
"mean_token_accuracy": 0.23861492425203323,
|
|
"num_tokens": 11163752.0,
|
|
"step": 6040
|
|
},
|
|
{
|
|
"entropy": 4.8643440246582035,
|
|
"epoch": 5.058577405857741,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003027516112199271,
|
|
"loss": 4.3598,
|
|
"mean_token_accuracy": 0.24447393268346787,
|
|
"num_tokens": 11173763.0,
|
|
"step": 6045
|
|
},
|
|
{
|
|
"entropy": 4.937761688232422,
|
|
"epoch": 5.062761506276151,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030243128091870413,
|
|
"loss": 4.4036,
|
|
"mean_token_accuracy": 0.2328161895275116,
|
|
"num_tokens": 11183577.0,
|
|
"step": 6050
|
|
},
|
|
{
|
|
"entropy": 4.910134410858154,
|
|
"epoch": 5.066945606694561,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0003021108941683152,
|
|
"loss": 4.3611,
|
|
"mean_token_accuracy": 0.2410369798541069,
|
|
"num_tokens": 11192708.0,
|
|
"step": 6055
|
|
},
|
|
{
|
|
"entropy": 4.890791559219361,
|
|
"epoch": 5.071129707112971,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00030179045162806494,
|
|
"loss": 4.419,
|
|
"mean_token_accuracy": 0.23849352747201918,
|
|
"num_tokens": 11201803.0,
|
|
"step": 6060
|
|
},
|
|
{
|
|
"entropy": 4.833604001998902,
|
|
"epoch": 5.075313807531381,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00030146995395737247,
|
|
"loss": 4.2861,
|
|
"mean_token_accuracy": 0.24582685977220536,
|
|
"num_tokens": 11211208.0,
|
|
"step": 6065
|
|
},
|
|
{
|
|
"entropy": 4.920818328857422,
|
|
"epoch": 5.079497907949791,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003011494018157704,
|
|
"loss": 4.3464,
|
|
"mean_token_accuracy": 0.24972191751003264,
|
|
"num_tokens": 11221091.0,
|
|
"step": 6070
|
|
},
|
|
{
|
|
"entropy": 4.935697078704834,
|
|
"epoch": 5.083682008368201,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003008287958629036,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.24234738796949387,
|
|
"num_tokens": 11230084.0,
|
|
"step": 6075
|
|
},
|
|
{
|
|
"entropy": 4.912802219390869,
|
|
"epoch": 5.087866108786611,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00030050813675852747,
|
|
"loss": 4.4416,
|
|
"mean_token_accuracy": 0.23867365717887878,
|
|
"num_tokens": 11239310.0,
|
|
"step": 6080
|
|
},
|
|
{
|
|
"entropy": 4.881706619262696,
|
|
"epoch": 5.092050209205021,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00030018742516250697,
|
|
"loss": 4.3102,
|
|
"mean_token_accuracy": 0.2551098048686981,
|
|
"num_tokens": 11248738.0,
|
|
"step": 6085
|
|
},
|
|
{
|
|
"entropy": 4.937249279022216,
|
|
"epoch": 5.096234309623431,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000299866661734815,
|
|
"loss": 4.3332,
|
|
"mean_token_accuracy": 0.24525767117738723,
|
|
"num_tokens": 11257813.0,
|
|
"step": 6090
|
|
},
|
|
{
|
|
"entropy": 4.953934383392334,
|
|
"epoch": 5.100418410041841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029954584713553107,
|
|
"loss": 4.3872,
|
|
"mean_token_accuracy": 0.23423128426074982,
|
|
"num_tokens": 11266475.0,
|
|
"step": 6095
|
|
},
|
|
{
|
|
"entropy": 4.8846697330474855,
|
|
"epoch": 5.104602510460251,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029922498202484014,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.24647708982229233,
|
|
"num_tokens": 11276011.0,
|
|
"step": 6100
|
|
},
|
|
{
|
|
"entropy": 4.93492956161499,
|
|
"epoch": 5.108786610878661,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000298904067063031,
|
|
"loss": 4.4491,
|
|
"mean_token_accuracy": 0.24073699563741685,
|
|
"num_tokens": 11284619.0,
|
|
"step": 6105
|
|
},
|
|
{
|
|
"entropy": 4.890891790390015,
|
|
"epoch": 5.112970711297071,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002985831029104949,
|
|
"loss": 4.3504,
|
|
"mean_token_accuracy": 0.2564302638173103,
|
|
"num_tokens": 11293105.0,
|
|
"step": 6110
|
|
},
|
|
{
|
|
"entropy": 4.885581159591675,
|
|
"epoch": 5.117154811715481,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002982620902277246,
|
|
"loss": 4.3453,
|
|
"mean_token_accuracy": 0.24282574653625488,
|
|
"num_tokens": 11302717.0,
|
|
"step": 6115
|
|
},
|
|
{
|
|
"entropy": 4.98964295387268,
|
|
"epoch": 5.121338912133891,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00029794102967531263,
|
|
"loss": 4.456,
|
|
"mean_token_accuracy": 0.24496397823095323,
|
|
"num_tokens": 11312709.0,
|
|
"step": 6120
|
|
},
|
|
{
|
|
"entropy": 4.864888763427734,
|
|
"epoch": 5.125523012552302,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00029761992191394985,
|
|
"loss": 4.3768,
|
|
"mean_token_accuracy": 0.24371286928653718,
|
|
"num_tokens": 11322040.0,
|
|
"step": 6125
|
|
},
|
|
{
|
|
"entropy": 4.912308931350708,
|
|
"epoch": 5.129707112970712,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00029729876760442456,
|
|
"loss": 4.4339,
|
|
"mean_token_accuracy": 0.2408338651061058,
|
|
"num_tokens": 11332029.0,
|
|
"step": 6130
|
|
},
|
|
{
|
|
"entropy": 4.922824239730835,
|
|
"epoch": 5.133891213389122,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002969775674076206,
|
|
"loss": 4.4189,
|
|
"mean_token_accuracy": 0.23722399175167083,
|
|
"num_tokens": 11340960.0,
|
|
"step": 6135
|
|
},
|
|
{
|
|
"entropy": 4.935235834121704,
|
|
"epoch": 5.138075313807532,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002966563219845164,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.24024122655391694,
|
|
"num_tokens": 11349497.0,
|
|
"step": 6140
|
|
},
|
|
{
|
|
"entropy": 4.951579809188843,
|
|
"epoch": 5.142259414225942,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002963350319961832,
|
|
"loss": 4.3651,
|
|
"mean_token_accuracy": 0.2448259934782982,
|
|
"num_tokens": 11358477.0,
|
|
"step": 6145
|
|
},
|
|
{
|
|
"entropy": 4.91556396484375,
|
|
"epoch": 5.146443514644352,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002960136981037844,
|
|
"loss": 4.3648,
|
|
"mean_token_accuracy": 0.24073498398065568,
|
|
"num_tokens": 11367635.0,
|
|
"step": 6150
|
|
},
|
|
{
|
|
"entropy": 4.8824268817901615,
|
|
"epoch": 5.150627615062762,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002956923209685733,
|
|
"loss": 4.3295,
|
|
"mean_token_accuracy": 0.24819556325674058,
|
|
"num_tokens": 11376737.0,
|
|
"step": 6155
|
|
},
|
|
{
|
|
"entropy": 4.871239376068115,
|
|
"epoch": 5.154811715481172,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002953709012518925,
|
|
"loss": 4.4157,
|
|
"mean_token_accuracy": 0.23524584025144576,
|
|
"num_tokens": 11386756.0,
|
|
"step": 6160
|
|
},
|
|
{
|
|
"entropy": 4.836379814147949,
|
|
"epoch": 5.158995815899582,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00029504943961517204,
|
|
"loss": 4.3021,
|
|
"mean_token_accuracy": 0.24969165474176408,
|
|
"num_tokens": 11394889.0,
|
|
"step": 6165
|
|
},
|
|
{
|
|
"entropy": 4.9464160919189455,
|
|
"epoch": 5.163179916317992,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00029472793671992824,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.24341897964477538,
|
|
"num_tokens": 11404759.0,
|
|
"step": 6170
|
|
},
|
|
{
|
|
"entropy": 4.969911384582519,
|
|
"epoch": 5.167364016736402,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00029440639322776246,
|
|
"loss": 4.4029,
|
|
"mean_token_accuracy": 0.2379152297973633,
|
|
"num_tokens": 11413739.0,
|
|
"step": 6175
|
|
},
|
|
{
|
|
"entropy": 4.8698570251464846,
|
|
"epoch": 5.171548117154812,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00029408480980035943,
|
|
"loss": 4.4226,
|
|
"mean_token_accuracy": 0.24212104678153992,
|
|
"num_tokens": 11422835.0,
|
|
"step": 6180
|
|
},
|
|
{
|
|
"entropy": 4.93922529220581,
|
|
"epoch": 5.175732217573222,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002937631870994863,
|
|
"loss": 4.4024,
|
|
"mean_token_accuracy": 0.2507467150688171,
|
|
"num_tokens": 11432613.0,
|
|
"step": 6185
|
|
},
|
|
{
|
|
"entropy": 4.851590585708618,
|
|
"epoch": 5.179916317991632,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029344152578699074,
|
|
"loss": 4.3675,
|
|
"mean_token_accuracy": 0.24312460720539092,
|
|
"num_tokens": 11441297.0,
|
|
"step": 6190
|
|
},
|
|
{
|
|
"entropy": 4.837188386917115,
|
|
"epoch": 5.184100418410042,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00029311982652479993,
|
|
"loss": 4.3261,
|
|
"mean_token_accuracy": 0.24020966440439223,
|
|
"num_tokens": 11450409.0,
|
|
"step": 6195
|
|
},
|
|
{
|
|
"entropy": 4.9418659687042235,
|
|
"epoch": 5.188284518828452,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002927980899749195,
|
|
"loss": 4.4558,
|
|
"mean_token_accuracy": 0.23206801414489747,
|
|
"num_tokens": 11460132.0,
|
|
"step": 6200
|
|
},
|
|
{
|
|
"entropy": 4.923004770278931,
|
|
"epoch": 5.192468619246862,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002924763167994312,
|
|
"loss": 4.4186,
|
|
"mean_token_accuracy": 0.2366531416773796,
|
|
"num_tokens": 11468984.0,
|
|
"step": 6205
|
|
},
|
|
{
|
|
"entropy": 4.964491844177246,
|
|
"epoch": 5.196652719665272,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002921545076604928,
|
|
"loss": 4.4383,
|
|
"mean_token_accuracy": 0.24058830887079238,
|
|
"num_tokens": 11478371.0,
|
|
"step": 6210
|
|
},
|
|
{
|
|
"entropy": 4.895745515823364,
|
|
"epoch": 5.200836820083682,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00029183266322033565,
|
|
"loss": 4.436,
|
|
"mean_token_accuracy": 0.24233003258705138,
|
|
"num_tokens": 11487924.0,
|
|
"step": 6215
|
|
},
|
|
{
|
|
"entropy": 4.94669017791748,
|
|
"epoch": 5.205020920502092,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0002915107841412638,
|
|
"loss": 4.4073,
|
|
"mean_token_accuracy": 0.23667048960924147,
|
|
"num_tokens": 11497892.0,
|
|
"step": 6220
|
|
},
|
|
{
|
|
"entropy": 4.891891193389893,
|
|
"epoch": 5.209205020920502,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002911888710856528,
|
|
"loss": 4.3054,
|
|
"mean_token_accuracy": 0.24660219103097916,
|
|
"num_tokens": 11507266.0,
|
|
"step": 6225
|
|
},
|
|
{
|
|
"entropy": 4.856154441833496,
|
|
"epoch": 5.2133891213389125,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002908669247159477,
|
|
"loss": 4.2876,
|
|
"mean_token_accuracy": 0.2508368656039238,
|
|
"num_tokens": 11516672.0,
|
|
"step": 6230
|
|
},
|
|
{
|
|
"entropy": 4.865242290496826,
|
|
"epoch": 5.2175732217573225,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00029054494569466266,
|
|
"loss": 4.4119,
|
|
"mean_token_accuracy": 0.23385217636823655,
|
|
"num_tokens": 11525865.0,
|
|
"step": 6235
|
|
},
|
|
{
|
|
"entropy": 4.955391788482666,
|
|
"epoch": 5.2217573221757325,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00029022293468437853,
|
|
"loss": 4.4324,
|
|
"mean_token_accuracy": 0.24337540715932846,
|
|
"num_tokens": 11534910.0,
|
|
"step": 6240
|
|
},
|
|
{
|
|
"entropy": 4.821013593673706,
|
|
"epoch": 5.2259414225941425,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028990089234774236,
|
|
"loss": 4.3031,
|
|
"mean_token_accuracy": 0.2577949285507202,
|
|
"num_tokens": 11543507.0,
|
|
"step": 6245
|
|
},
|
|
{
|
|
"entropy": 4.826885509490967,
|
|
"epoch": 5.2301255230125525,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00028957881934746544,
|
|
"loss": 4.3305,
|
|
"mean_token_accuracy": 0.24868865460157394,
|
|
"num_tokens": 11552562.0,
|
|
"step": 6250
|
|
},
|
|
{
|
|
"entropy": 4.936593437194825,
|
|
"epoch": 5.2343096234309625,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028925671634632216,
|
|
"loss": 4.3848,
|
|
"mean_token_accuracy": 0.2429310768842697,
|
|
"num_tokens": 11561931.0,
|
|
"step": 6255
|
|
},
|
|
{
|
|
"entropy": 4.925588846206665,
|
|
"epoch": 5.2384937238493725,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00028893458400714876,
|
|
"loss": 4.4057,
|
|
"mean_token_accuracy": 0.24676982909440995,
|
|
"num_tokens": 11570639.0,
|
|
"step": 6260
|
|
},
|
|
{
|
|
"entropy": 4.884219694137573,
|
|
"epoch": 5.2426778242677825,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002886124229928418,
|
|
"loss": 4.3379,
|
|
"mean_token_accuracy": 0.2486286222934723,
|
|
"num_tokens": 11579796.0,
|
|
"step": 6265
|
|
},
|
|
{
|
|
"entropy": 4.917233896255493,
|
|
"epoch": 5.2468619246861925,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028829023396635683,
|
|
"loss": 4.434,
|
|
"mean_token_accuracy": 0.2443879649043083,
|
|
"num_tokens": 11589343.0,
|
|
"step": 6270
|
|
},
|
|
{
|
|
"entropy": 4.936890506744385,
|
|
"epoch": 5.2510460251046025,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002879680175907071,
|
|
"loss": 4.4257,
|
|
"mean_token_accuracy": 0.2432400703430176,
|
|
"num_tokens": 11598633.0,
|
|
"step": 6275
|
|
},
|
|
{
|
|
"entropy": 4.874241065979004,
|
|
"epoch": 5.2552301255230125,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002876457745289621,
|
|
"loss": 4.3026,
|
|
"mean_token_accuracy": 0.24992417842149733,
|
|
"num_tokens": 11607746.0,
|
|
"step": 6280
|
|
},
|
|
{
|
|
"entropy": 4.9070806980133055,
|
|
"epoch": 5.2594142259414225,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002873235054442463,
|
|
"loss": 4.4119,
|
|
"mean_token_accuracy": 0.23910572826862336,
|
|
"num_tokens": 11616574.0,
|
|
"step": 6285
|
|
},
|
|
{
|
|
"entropy": 4.892051458358765,
|
|
"epoch": 5.2635983263598325,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028700121099973765,
|
|
"loss": 4.417,
|
|
"mean_token_accuracy": 0.23886230140924453,
|
|
"num_tokens": 11625918.0,
|
|
"step": 6290
|
|
},
|
|
{
|
|
"entropy": 4.921296739578247,
|
|
"epoch": 5.2677824267782425,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002866788918586661,
|
|
"loss": 4.3986,
|
|
"mean_token_accuracy": 0.24384705722332,
|
|
"num_tokens": 11634255.0,
|
|
"step": 6295
|
|
},
|
|
{
|
|
"entropy": 4.90167007446289,
|
|
"epoch": 5.2719665271966525,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028635654868431286,
|
|
"loss": 4.3874,
|
|
"mean_token_accuracy": 0.23205001056194305,
|
|
"num_tokens": 11643532.0,
|
|
"step": 6300
|
|
},
|
|
{
|
|
"entropy": 4.876720142364502,
|
|
"epoch": 5.2761506276150625,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00028603418214000827,
|
|
"loss": 4.3534,
|
|
"mean_token_accuracy": 0.23860932141542435,
|
|
"num_tokens": 11653104.0,
|
|
"step": 6305
|
|
},
|
|
{
|
|
"entropy": 4.9180694103240965,
|
|
"epoch": 5.2803347280334725,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002857117928891308,
|
|
"loss": 4.4426,
|
|
"mean_token_accuracy": 0.24005795866250992,
|
|
"num_tokens": 11662752.0,
|
|
"step": 6310
|
|
},
|
|
{
|
|
"entropy": 4.806065893173217,
|
|
"epoch": 5.2845188284518825,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00028538938159510587,
|
|
"loss": 4.3197,
|
|
"mean_token_accuracy": 0.2503700599074364,
|
|
"num_tokens": 11671566.0,
|
|
"step": 6315
|
|
},
|
|
{
|
|
"entropy": 4.911614084243775,
|
|
"epoch": 5.2887029288702925,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00028506694892140385,
|
|
"loss": 4.3957,
|
|
"mean_token_accuracy": 0.2486814796924591,
|
|
"num_tokens": 11681770.0,
|
|
"step": 6320
|
|
},
|
|
{
|
|
"entropy": 4.821918916702271,
|
|
"epoch": 5.292887029288703,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00028474449553153944,
|
|
"loss": 4.4108,
|
|
"mean_token_accuracy": 0.23857569843530654,
|
|
"num_tokens": 11691177.0,
|
|
"step": 6325
|
|
},
|
|
{
|
|
"entropy": 4.8456258296966555,
|
|
"epoch": 5.297071129707113,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00028442202208906993,
|
|
"loss": 4.334,
|
|
"mean_token_accuracy": 0.24584351629018783,
|
|
"num_tokens": 11700461.0,
|
|
"step": 6330
|
|
},
|
|
{
|
|
"entropy": 4.862869501113892,
|
|
"epoch": 5.301255230125523,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002840995292575938,
|
|
"loss": 4.3763,
|
|
"mean_token_accuracy": 0.23797542005777358,
|
|
"num_tokens": 11708783.0,
|
|
"step": 6335
|
|
},
|
|
{
|
|
"entropy": 4.884965276718139,
|
|
"epoch": 5.305439330543933,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0002837770177007494,
|
|
"loss": 4.4053,
|
|
"mean_token_accuracy": 0.24028726667165756,
|
|
"num_tokens": 11717419.0,
|
|
"step": 6340
|
|
},
|
|
{
|
|
"entropy": 4.920804977416992,
|
|
"epoch": 5.309623430962343,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00028345448808221374,
|
|
"loss": 4.4509,
|
|
"mean_token_accuracy": 0.23954248279333115,
|
|
"num_tokens": 11727014.0,
|
|
"step": 6345
|
|
},
|
|
{
|
|
"entropy": 4.933978462219239,
|
|
"epoch": 5.313807531380753,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00028313194106570095,
|
|
"loss": 4.386,
|
|
"mean_token_accuracy": 0.24088611900806428,
|
|
"num_tokens": 11736718.0,
|
|
"step": 6350
|
|
},
|
|
{
|
|
"entropy": 4.987976169586181,
|
|
"epoch": 5.317991631799163,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002828093773149607,
|
|
"loss": 4.4776,
|
|
"mean_token_accuracy": 0.23227674067020415,
|
|
"num_tokens": 11747236.0,
|
|
"step": 6355
|
|
},
|
|
{
|
|
"entropy": 4.888700819015503,
|
|
"epoch": 5.322175732217573,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028248679749377754,
|
|
"loss": 4.3937,
|
|
"mean_token_accuracy": 0.24064283818006516,
|
|
"num_tokens": 11756912.0,
|
|
"step": 6360
|
|
},
|
|
{
|
|
"entropy": 4.837135171890258,
|
|
"epoch": 5.326359832635983,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028216420226596884,
|
|
"loss": 4.3496,
|
|
"mean_token_accuracy": 0.24567523896694182,
|
|
"num_tokens": 11766035.0,
|
|
"step": 6365
|
|
},
|
|
{
|
|
"entropy": 4.807533073425293,
|
|
"epoch": 5.330543933054393,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002818415922953837,
|
|
"loss": 4.2953,
|
|
"mean_token_accuracy": 0.25234455764293673,
|
|
"num_tokens": 11775934.0,
|
|
"step": 6370
|
|
},
|
|
{
|
|
"entropy": 4.870753574371338,
|
|
"epoch": 5.334728033472803,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00028151896824590156,
|
|
"loss": 4.3822,
|
|
"mean_token_accuracy": 0.23798278272151946,
|
|
"num_tokens": 11784549.0,
|
|
"step": 6375
|
|
},
|
|
{
|
|
"entropy": 4.905602025985718,
|
|
"epoch": 5.338912133891213,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002811963307814309,
|
|
"loss": 4.4354,
|
|
"mean_token_accuracy": 0.23337181359529496,
|
|
"num_tokens": 11794369.0,
|
|
"step": 6380
|
|
},
|
|
{
|
|
"entropy": 4.875904369354248,
|
|
"epoch": 5.343096234309623,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00028087368056590767,
|
|
"loss": 4.405,
|
|
"mean_token_accuracy": 0.24161886274814606,
|
|
"num_tokens": 11803221.0,
|
|
"step": 6385
|
|
},
|
|
{
|
|
"entropy": 4.920868635177612,
|
|
"epoch": 5.347280334728033,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0002805510182632942,
|
|
"loss": 4.4091,
|
|
"mean_token_accuracy": 0.2412685349583626,
|
|
"num_tokens": 11812811.0,
|
|
"step": 6390
|
|
},
|
|
{
|
|
"entropy": 4.92840986251831,
|
|
"epoch": 5.351464435146443,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002802283445375775,
|
|
"loss": 4.4217,
|
|
"mean_token_accuracy": 0.2422350376844406,
|
|
"num_tokens": 11821522.0,
|
|
"step": 6395
|
|
},
|
|
{
|
|
"entropy": 4.861080026626587,
|
|
"epoch": 5.355648535564853,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027990566005276846,
|
|
"loss": 4.37,
|
|
"mean_token_accuracy": 0.2505931928753853,
|
|
"num_tokens": 11830021.0,
|
|
"step": 6400
|
|
},
|
|
{
|
|
"entropy": 4.865070152282715,
|
|
"epoch": 5.359832635983263,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027958296547289966,
|
|
"loss": 4.3547,
|
|
"mean_token_accuracy": 0.24615051001310348,
|
|
"num_tokens": 11838612.0,
|
|
"step": 6405
|
|
},
|
|
{
|
|
"entropy": 4.842417907714844,
|
|
"epoch": 5.364016736401673,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00027926026146202475,
|
|
"loss": 4.3029,
|
|
"mean_token_accuracy": 0.254127049446106,
|
|
"num_tokens": 11847648.0,
|
|
"step": 6410
|
|
},
|
|
{
|
|
"entropy": 4.816158103942871,
|
|
"epoch": 5.368200836820083,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00027893754868421665,
|
|
"loss": 4.3358,
|
|
"mean_token_accuracy": 0.25257784128189087,
|
|
"num_tokens": 11857008.0,
|
|
"step": 6415
|
|
},
|
|
{
|
|
"entropy": 4.899079608917236,
|
|
"epoch": 5.372384937238493,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002786148278035665,
|
|
"loss": 4.3653,
|
|
"mean_token_accuracy": 0.2519300639629364,
|
|
"num_tokens": 11866055.0,
|
|
"step": 6420
|
|
},
|
|
{
|
|
"entropy": 4.946145343780517,
|
|
"epoch": 5.376569037656903,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002782920994841819,
|
|
"loss": 4.4342,
|
|
"mean_token_accuracy": 0.23677102029323577,
|
|
"num_tokens": 11875496.0,
|
|
"step": 6425
|
|
},
|
|
{
|
|
"entropy": 4.869200277328491,
|
|
"epoch": 5.380753138075314,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002779693643901858,
|
|
"loss": 4.3646,
|
|
"mean_token_accuracy": 0.24217727333307265,
|
|
"num_tokens": 11884115.0,
|
|
"step": 6430
|
|
},
|
|
{
|
|
"entropy": 4.890128469467163,
|
|
"epoch": 5.384937238493724,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00027764662318571526,
|
|
"loss": 4.5225,
|
|
"mean_token_accuracy": 0.23514971137046814,
|
|
"num_tokens": 11893930.0,
|
|
"step": 6435
|
|
},
|
|
{
|
|
"entropy": 4.903466892242432,
|
|
"epoch": 5.389121338912134,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002773238765349196,
|
|
"loss": 4.3914,
|
|
"mean_token_accuracy": 0.23717236667871475,
|
|
"num_tokens": 11902702.0,
|
|
"step": 6440
|
|
},
|
|
{
|
|
"entropy": 4.951507997512818,
|
|
"epoch": 5.393305439330544,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002770011251019598,
|
|
"loss": 4.4184,
|
|
"mean_token_accuracy": 0.2410748615860939,
|
|
"num_tokens": 11911915.0,
|
|
"step": 6445
|
|
},
|
|
{
|
|
"entropy": 4.858889484405518,
|
|
"epoch": 5.397489539748954,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00027667836955100615,
|
|
"loss": 4.3586,
|
|
"mean_token_accuracy": 0.24552163928747178,
|
|
"num_tokens": 11920737.0,
|
|
"step": 6450
|
|
},
|
|
{
|
|
"entropy": 4.887974643707276,
|
|
"epoch": 5.401673640167364,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002763556105462378,
|
|
"loss": 4.398,
|
|
"mean_token_accuracy": 0.2450038343667984,
|
|
"num_tokens": 11929548.0,
|
|
"step": 6455
|
|
},
|
|
{
|
|
"entropy": 4.87781457901001,
|
|
"epoch": 5.405857740585774,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027603284875184095,
|
|
"loss": 4.3652,
|
|
"mean_token_accuracy": 0.23706325739622117,
|
|
"num_tokens": 11938459.0,
|
|
"step": 6460
|
|
},
|
|
{
|
|
"entropy": 4.942236518859863,
|
|
"epoch": 5.410041841004184,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027571008483200744,
|
|
"loss": 4.3698,
|
|
"mean_token_accuracy": 0.24951007962226868,
|
|
"num_tokens": 11947282.0,
|
|
"step": 6465
|
|
},
|
|
{
|
|
"entropy": 4.879883861541748,
|
|
"epoch": 5.414225941422594,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00027538731945093346,
|
|
"loss": 4.3547,
|
|
"mean_token_accuracy": 0.2438443586230278,
|
|
"num_tokens": 11956209.0,
|
|
"step": 6470
|
|
},
|
|
{
|
|
"entropy": 4.8672380447387695,
|
|
"epoch": 5.418410041841004,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00027506455327281834,
|
|
"loss": 4.4043,
|
|
"mean_token_accuracy": 0.24089341908693312,
|
|
"num_tokens": 11965689.0,
|
|
"step": 6475
|
|
},
|
|
{
|
|
"entropy": 4.891218662261963,
|
|
"epoch": 5.422594142259414,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000274741786961863,
|
|
"loss": 4.4121,
|
|
"mean_token_accuracy": 0.2368292212486267,
|
|
"num_tokens": 11975622.0,
|
|
"step": 6480
|
|
},
|
|
{
|
|
"entropy": 4.909551525115967,
|
|
"epoch": 5.426778242677824,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002744190211822685,
|
|
"loss": 4.4758,
|
|
"mean_token_accuracy": 0.23757023215293885,
|
|
"num_tokens": 11984904.0,
|
|
"step": 6485
|
|
},
|
|
{
|
|
"entropy": 4.955643892288208,
|
|
"epoch": 5.430962343096234,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000274096256598235,
|
|
"loss": 4.4972,
|
|
"mean_token_accuracy": 0.23711335510015488,
|
|
"num_tokens": 11995456.0,
|
|
"step": 6490
|
|
},
|
|
{
|
|
"entropy": 4.891779470443725,
|
|
"epoch": 5.435146443514644,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00027377349387396017,
|
|
"loss": 4.3662,
|
|
"mean_token_accuracy": 0.23640024811029434,
|
|
"num_tokens": 12004566.0,
|
|
"step": 6495
|
|
},
|
|
{
|
|
"entropy": 4.855230283737183,
|
|
"epoch": 5.439330543933054,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00027345073367363775,
|
|
"loss": 4.3015,
|
|
"mean_token_accuracy": 0.25092219561338425,
|
|
"num_tokens": 12013304.0,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"epoch": 5.439330543933054,
|
|
"eval_entropy": 4.764241058158484,
|
|
"eval_loss": 5.350369453430176,
|
|
"eval_mean_token_accuracy": 0.1999506031321114,
|
|
"eval_num_tokens": 12013304.0,
|
|
"eval_runtime": 2.7276,
|
|
"eval_samples_per_second": 1432.388,
|
|
"eval_steps_per_second": 179.278,
|
|
"step": 6500
|
|
},
|
|
{
|
|
"entropy": 4.829323768615723,
|
|
"epoch": 5.443514644351464,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002731279766614563,
|
|
"loss": 4.3584,
|
|
"mean_token_accuracy": 0.244656702876091,
|
|
"num_tokens": 12022007.0,
|
|
"step": 6505
|
|
},
|
|
{
|
|
"entropy": 4.833226108551026,
|
|
"epoch": 5.447698744769874,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000272805223501598,
|
|
"loss": 4.4135,
|
|
"mean_token_accuracy": 0.2419069916009903,
|
|
"num_tokens": 12031493.0,
|
|
"step": 6510
|
|
},
|
|
{
|
|
"entropy": 4.875311851501465,
|
|
"epoch": 5.451882845188284,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002724824748582369,
|
|
"loss": 4.3733,
|
|
"mean_token_accuracy": 0.2510312333703041,
|
|
"num_tokens": 12040143.0,
|
|
"step": 6515
|
|
},
|
|
{
|
|
"entropy": 5.030745220184326,
|
|
"epoch": 5.456066945606695,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002721597313955376,
|
|
"loss": 4.5574,
|
|
"mean_token_accuracy": 0.23192945271730422,
|
|
"num_tokens": 12050136.0,
|
|
"step": 6520
|
|
},
|
|
{
|
|
"entropy": 4.887574338912964,
|
|
"epoch": 5.460251046025105,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002718369937776546,
|
|
"loss": 4.4266,
|
|
"mean_token_accuracy": 0.23636944741010665,
|
|
"num_tokens": 12060065.0,
|
|
"step": 6525
|
|
},
|
|
{
|
|
"entropy": 4.953693914413452,
|
|
"epoch": 5.464435146443515,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002715142626687298,
|
|
"loss": 4.452,
|
|
"mean_token_accuracy": 0.2278196170926094,
|
|
"num_tokens": 12069175.0,
|
|
"step": 6530
|
|
},
|
|
{
|
|
"entropy": 4.873070955276489,
|
|
"epoch": 5.468619246861925,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.000271191538732892,
|
|
"loss": 4.3605,
|
|
"mean_token_accuracy": 0.25032857805490494,
|
|
"num_tokens": 12077308.0,
|
|
"step": 6535
|
|
},
|
|
{
|
|
"entropy": 4.838415670394897,
|
|
"epoch": 5.472803347280335,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00027086882263425527,
|
|
"loss": 4.3298,
|
|
"mean_token_accuracy": 0.24296893626451493,
|
|
"num_tokens": 12085678.0,
|
|
"step": 6540
|
|
},
|
|
{
|
|
"entropy": 4.86780104637146,
|
|
"epoch": 5.476987447698745,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00027054611503691736,
|
|
"loss": 4.4241,
|
|
"mean_token_accuracy": 0.23863587528467178,
|
|
"num_tokens": 12095163.0,
|
|
"step": 6545
|
|
},
|
|
{
|
|
"entropy": 4.936135911941529,
|
|
"epoch": 5.481171548117155,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00027022341660495853,
|
|
"loss": 4.4154,
|
|
"mean_token_accuracy": 0.23589290380477906,
|
|
"num_tokens": 12104185.0,
|
|
"step": 6550
|
|
},
|
|
{
|
|
"entropy": 4.9144854068756105,
|
|
"epoch": 5.485355648535565,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026990072800244045,
|
|
"loss": 4.4574,
|
|
"mean_token_accuracy": 0.2319075033068657,
|
|
"num_tokens": 12114298.0,
|
|
"step": 6555
|
|
},
|
|
{
|
|
"entropy": 4.882907438278198,
|
|
"epoch": 5.489539748953975,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00026957804989340427,
|
|
"loss": 4.3909,
|
|
"mean_token_accuracy": 0.24646756201982498,
|
|
"num_tokens": 12122512.0,
|
|
"step": 6560
|
|
},
|
|
{
|
|
"entropy": 4.845264339447022,
|
|
"epoch": 5.493723849372385,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002692553829418697,
|
|
"loss": 4.3533,
|
|
"mean_token_accuracy": 0.24628245383501052,
|
|
"num_tokens": 12132133.0,
|
|
"step": 6565
|
|
},
|
|
{
|
|
"entropy": 4.8802161693573,
|
|
"epoch": 5.497907949790795,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002689327278118334,
|
|
"loss": 4.4031,
|
|
"mean_token_accuracy": 0.241229847073555,
|
|
"num_tokens": 12141299.0,
|
|
"step": 6570
|
|
},
|
|
{
|
|
"entropy": 4.9944110870361325,
|
|
"epoch": 5.502092050209205,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000268610085167268,
|
|
"loss": 4.4907,
|
|
"mean_token_accuracy": 0.2368161529302597,
|
|
"num_tokens": 12150412.0,
|
|
"step": 6575
|
|
},
|
|
{
|
|
"entropy": 4.915564966201782,
|
|
"epoch": 5.506276150627615,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002682874556721198,
|
|
"loss": 4.4396,
|
|
"mean_token_accuracy": 0.23802562654018403,
|
|
"num_tokens": 12159754.0,
|
|
"step": 6580
|
|
},
|
|
{
|
|
"entropy": 4.912384700775147,
|
|
"epoch": 5.510460251046025,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002679648399903087,
|
|
"loss": 4.3909,
|
|
"mean_token_accuracy": 0.24707238525152206,
|
|
"num_tokens": 12169098.0,
|
|
"step": 6585
|
|
},
|
|
{
|
|
"entropy": 4.84827995300293,
|
|
"epoch": 5.514644351464435,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002676422387857258,
|
|
"loss": 4.3674,
|
|
"mean_token_accuracy": 0.24549924731254577,
|
|
"num_tokens": 12178304.0,
|
|
"step": 6590
|
|
},
|
|
{
|
|
"entropy": 4.79505295753479,
|
|
"epoch": 5.518828451882845,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00026731965272223246,
|
|
"loss": 4.3604,
|
|
"mean_token_accuracy": 0.24779249727725983,
|
|
"num_tokens": 12188029.0,
|
|
"step": 6595
|
|
},
|
|
{
|
|
"entropy": 4.844727516174316,
|
|
"epoch": 5.523012552301255,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002669970824636591,
|
|
"loss": 4.3501,
|
|
"mean_token_accuracy": 0.24134178161621095,
|
|
"num_tokens": 12196663.0,
|
|
"step": 6600
|
|
},
|
|
{
|
|
"entropy": 4.923648118972778,
|
|
"epoch": 5.527196652719665,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002666745286738033,
|
|
"loss": 4.4685,
|
|
"mean_token_accuracy": 0.23181113302707673,
|
|
"num_tokens": 12206496.0,
|
|
"step": 6605
|
|
},
|
|
{
|
|
"entropy": 4.961730861663819,
|
|
"epoch": 5.531380753138075,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00026635199201642914,
|
|
"loss": 4.4643,
|
|
"mean_token_accuracy": 0.24141950011253357,
|
|
"num_tokens": 12216597.0,
|
|
"step": 6610
|
|
},
|
|
{
|
|
"entropy": 4.856242704391479,
|
|
"epoch": 5.535564853556485,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002660294731552649,
|
|
"loss": 4.3882,
|
|
"mean_token_accuracy": 0.23969252556562423,
|
|
"num_tokens": 12225125.0,
|
|
"step": 6615
|
|
},
|
|
{
|
|
"entropy": 4.827517986297607,
|
|
"epoch": 5.539748953974895,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00026570697275400267,
|
|
"loss": 4.3645,
|
|
"mean_token_accuracy": 0.2407393276691437,
|
|
"num_tokens": 12234402.0,
|
|
"step": 6620
|
|
},
|
|
{
|
|
"entropy": 4.839079427719116,
|
|
"epoch": 5.543933054393305,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002653844914762964,
|
|
"loss": 4.339,
|
|
"mean_token_accuracy": 0.24318913966417313,
|
|
"num_tokens": 12244136.0,
|
|
"step": 6625
|
|
},
|
|
{
|
|
"entropy": 4.906605863571167,
|
|
"epoch": 5.548117154811716,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002650620299857607,
|
|
"loss": 4.3943,
|
|
"mean_token_accuracy": 0.23823688477277755,
|
|
"num_tokens": 12253039.0,
|
|
"step": 6630
|
|
},
|
|
{
|
|
"entropy": 4.928784465789795,
|
|
"epoch": 5.552301255230126,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026473958894596963,
|
|
"loss": 4.4133,
|
|
"mean_token_accuracy": 0.24667657762765885,
|
|
"num_tokens": 12263172.0,
|
|
"step": 6635
|
|
},
|
|
{
|
|
"entropy": 4.888383579254151,
|
|
"epoch": 5.556485355648536,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000264417169020455,
|
|
"loss": 4.461,
|
|
"mean_token_accuracy": 0.2373802348971367,
|
|
"num_tokens": 12273324.0,
|
|
"step": 6640
|
|
},
|
|
{
|
|
"entropy": 4.935473203659058,
|
|
"epoch": 5.560669456066946,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002640947708727049,
|
|
"loss": 4.5209,
|
|
"mean_token_accuracy": 0.23320891261100768,
|
|
"num_tokens": 12283719.0,
|
|
"step": 6645
|
|
},
|
|
{
|
|
"entropy": 4.852397632598877,
|
|
"epoch": 5.564853556485356,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002637723951661631,
|
|
"loss": 4.447,
|
|
"mean_token_accuracy": 0.23743972778320313,
|
|
"num_tokens": 12293451.0,
|
|
"step": 6650
|
|
},
|
|
{
|
|
"entropy": 4.973465347290039,
|
|
"epoch": 5.569037656903766,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002634500425642268,
|
|
"loss": 4.4948,
|
|
"mean_token_accuracy": 0.2366950511932373,
|
|
"num_tokens": 12302560.0,
|
|
"step": 6655
|
|
},
|
|
{
|
|
"entropy": 4.955617666244507,
|
|
"epoch": 5.573221757322176,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002631277137302461,
|
|
"loss": 4.4364,
|
|
"mean_token_accuracy": 0.2341605931520462,
|
|
"num_tokens": 12312115.0,
|
|
"step": 6660
|
|
},
|
|
{
|
|
"entropy": 4.820385932922363,
|
|
"epoch": 5.577405857740586,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00026280540932752176,
|
|
"loss": 4.3348,
|
|
"mean_token_accuracy": 0.25350783616304395,
|
|
"num_tokens": 12320689.0,
|
|
"step": 6665
|
|
},
|
|
{
|
|
"entropy": 4.793215227127075,
|
|
"epoch": 5.581589958158996,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026248313001930445,
|
|
"loss": 4.3289,
|
|
"mean_token_accuracy": 0.2471508651971817,
|
|
"num_tokens": 12329643.0,
|
|
"step": 6670
|
|
},
|
|
{
|
|
"entropy": 4.908515357971192,
|
|
"epoch": 5.585774058577406,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026216087646879326,
|
|
"loss": 4.4296,
|
|
"mean_token_accuracy": 0.24640488028526306,
|
|
"num_tokens": 12339267.0,
|
|
"step": 6675
|
|
},
|
|
{
|
|
"entropy": 4.968141746520996,
|
|
"epoch": 5.589958158995816,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002618386493391339,
|
|
"loss": 4.5112,
|
|
"mean_token_accuracy": 0.23495193868875502,
|
|
"num_tokens": 12348515.0,
|
|
"step": 6680
|
|
},
|
|
{
|
|
"entropy": 4.900367498397827,
|
|
"epoch": 5.594142259414226,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00026151644929341833,
|
|
"loss": 4.4762,
|
|
"mean_token_accuracy": 0.23157283067703247,
|
|
"num_tokens": 12358023.0,
|
|
"step": 6685
|
|
},
|
|
{
|
|
"entropy": 4.846280813217163,
|
|
"epoch": 5.598326359832636,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026119427699468226,
|
|
"loss": 4.378,
|
|
"mean_token_accuracy": 0.23762039691209794,
|
|
"num_tokens": 12367129.0,
|
|
"step": 6690
|
|
},
|
|
{
|
|
"entropy": 4.864776754379273,
|
|
"epoch": 5.602510460251046,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00026087213310590455,
|
|
"loss": 4.4009,
|
|
"mean_token_accuracy": 0.24618713110685347,
|
|
"num_tokens": 12375933.0,
|
|
"step": 6695
|
|
},
|
|
{
|
|
"entropy": 4.817289400100708,
|
|
"epoch": 5.606694560669456,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00026055001829000543,
|
|
"loss": 4.3883,
|
|
"mean_token_accuracy": 0.24649878740310668,
|
|
"num_tokens": 12384784.0,
|
|
"step": 6700
|
|
},
|
|
{
|
|
"entropy": 4.887301969528198,
|
|
"epoch": 5.610878661087866,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026022793320984553,
|
|
"loss": 4.423,
|
|
"mean_token_accuracy": 0.24790360033512115,
|
|
"num_tokens": 12393933.0,
|
|
"step": 6705
|
|
},
|
|
{
|
|
"entropy": 4.909762144088745,
|
|
"epoch": 5.615062761506276,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002599058785282241,
|
|
"loss": 4.4352,
|
|
"mean_token_accuracy": 0.2354859232902527,
|
|
"num_tokens": 12403442.0,
|
|
"step": 6710
|
|
},
|
|
{
|
|
"entropy": 4.8850054264068605,
|
|
"epoch": 5.619246861924686,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025958385490787797,
|
|
"loss": 4.3282,
|
|
"mean_token_accuracy": 0.2514948144555092,
|
|
"num_tokens": 12412317.0,
|
|
"step": 6715
|
|
},
|
|
{
|
|
"entropy": 4.862397241592407,
|
|
"epoch": 5.623430962343097,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002592618630114798,
|
|
"loss": 4.3788,
|
|
"mean_token_accuracy": 0.24227384030818938,
|
|
"num_tokens": 12421459.0,
|
|
"step": 6720
|
|
},
|
|
{
|
|
"entropy": 4.883806610107422,
|
|
"epoch": 5.627615062761507,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00025893990350163724,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.24089335203170775,
|
|
"num_tokens": 12431224.0,
|
|
"step": 6725
|
|
},
|
|
{
|
|
"entropy": 4.806319046020508,
|
|
"epoch": 5.631799163179917,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025861797704089133,
|
|
"loss": 4.329,
|
|
"mean_token_accuracy": 0.24332554936408995,
|
|
"num_tokens": 12439891.0,
|
|
"step": 6730
|
|
},
|
|
{
|
|
"entropy": 4.823120641708374,
|
|
"epoch": 5.635983263598327,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002582960842917148,
|
|
"loss": 4.3167,
|
|
"mean_token_accuracy": 0.24289154559373854,
|
|
"num_tokens": 12449852.0,
|
|
"step": 6735
|
|
},
|
|
{
|
|
"entropy": 4.936335182189941,
|
|
"epoch": 5.640167364016737,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002579742259165113,
|
|
"loss": 4.4272,
|
|
"mean_token_accuracy": 0.2495082288980484,
|
|
"num_tokens": 12459058.0,
|
|
"step": 6740
|
|
},
|
|
{
|
|
"entropy": 4.9523485660552975,
|
|
"epoch": 5.644351464435147,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002576524025776135,
|
|
"loss": 4.448,
|
|
"mean_token_accuracy": 0.2418569415807724,
|
|
"num_tokens": 12468684.0,
|
|
"step": 6745
|
|
},
|
|
{
|
|
"entropy": 4.855965709686279,
|
|
"epoch": 5.648535564853557,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002573306149372821,
|
|
"loss": 4.4225,
|
|
"mean_token_accuracy": 0.2413572311401367,
|
|
"num_tokens": 12478102.0,
|
|
"step": 6750
|
|
},
|
|
{
|
|
"entropy": 4.896460628509521,
|
|
"epoch": 5.652719665271967,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002570088636577044,
|
|
"loss": 4.4284,
|
|
"mean_token_accuracy": 0.235513137280941,
|
|
"num_tokens": 12488036.0,
|
|
"step": 6755
|
|
},
|
|
{
|
|
"entropy": 4.941737794876099,
|
|
"epoch": 5.656903765690377,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002566871494009928,
|
|
"loss": 4.3926,
|
|
"mean_token_accuracy": 0.24554055482149123,
|
|
"num_tokens": 12496593.0,
|
|
"step": 6760
|
|
},
|
|
{
|
|
"entropy": 4.935305070877075,
|
|
"epoch": 5.661087866108787,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002563654728291835,
|
|
"loss": 4.4986,
|
|
"mean_token_accuracy": 0.2304642006754875,
|
|
"num_tokens": 12505752.0,
|
|
"step": 6765
|
|
},
|
|
{
|
|
"entropy": 4.843280935287476,
|
|
"epoch": 5.665271966527197,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002560438346042351,
|
|
"loss": 4.3791,
|
|
"mean_token_accuracy": 0.24640879184007644,
|
|
"num_tokens": 12515192.0,
|
|
"step": 6770
|
|
},
|
|
{
|
|
"entropy": 4.937770128250122,
|
|
"epoch": 5.669456066945607,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002557222353880275,
|
|
"loss": 4.4627,
|
|
"mean_token_accuracy": 0.23305097222328186,
|
|
"num_tokens": 12525613.0,
|
|
"step": 6775
|
|
},
|
|
{
|
|
"entropy": 4.953578281402588,
|
|
"epoch": 5.673640167364017,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00025540067584235997,
|
|
"loss": 4.4226,
|
|
"mean_token_accuracy": 0.24094831496477126,
|
|
"num_tokens": 12535076.0,
|
|
"step": 6780
|
|
},
|
|
{
|
|
"entropy": 4.860519313812256,
|
|
"epoch": 5.677824267782427,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025507915662895055,
|
|
"loss": 4.4363,
|
|
"mean_token_accuracy": 0.23784058392047883,
|
|
"num_tokens": 12544333.0,
|
|
"step": 6785
|
|
},
|
|
{
|
|
"entropy": 4.887219667434692,
|
|
"epoch": 5.682008368200837,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00025475767840943403,
|
|
"loss": 4.3999,
|
|
"mean_token_accuracy": 0.2391606479883194,
|
|
"num_tokens": 12554292.0,
|
|
"step": 6790
|
|
},
|
|
{
|
|
"entropy": 4.8962180614471436,
|
|
"epoch": 5.686192468619247,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002544362418453608,
|
|
"loss": 4.4673,
|
|
"mean_token_accuracy": 0.23454530537128448,
|
|
"num_tokens": 12564240.0,
|
|
"step": 6795
|
|
},
|
|
{
|
|
"entropy": 4.871890211105347,
|
|
"epoch": 5.690376569037657,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00025411484759819584,
|
|
"loss": 4.3726,
|
|
"mean_token_accuracy": 0.25466072410345075,
|
|
"num_tokens": 12573280.0,
|
|
"step": 6800
|
|
},
|
|
{
|
|
"entropy": 4.894176578521728,
|
|
"epoch": 5.694560669456067,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00025379349632931666,
|
|
"loss": 4.492,
|
|
"mean_token_accuracy": 0.2352101758122444,
|
|
"num_tokens": 12583133.0,
|
|
"step": 6805
|
|
},
|
|
{
|
|
"entropy": 4.956999349594116,
|
|
"epoch": 5.698744769874477,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002534721887000127,
|
|
"loss": 4.4339,
|
|
"mean_token_accuracy": 0.24088352769613267,
|
|
"num_tokens": 12591709.0,
|
|
"step": 6810
|
|
},
|
|
{
|
|
"entropy": 4.924673700332642,
|
|
"epoch": 5.702928870292887,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002531509253714831,
|
|
"loss": 4.3692,
|
|
"mean_token_accuracy": 0.24559981226921082,
|
|
"num_tokens": 12600187.0,
|
|
"step": 6815
|
|
},
|
|
{
|
|
"entropy": 4.768226480484008,
|
|
"epoch": 5.707112970711297,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002528297070048364,
|
|
"loss": 4.353,
|
|
"mean_token_accuracy": 0.24972946792840958,
|
|
"num_tokens": 12608910.0,
|
|
"step": 6820
|
|
},
|
|
{
|
|
"entropy": 4.8285926342010494,
|
|
"epoch": 5.711297071129707,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00025250853426108814,
|
|
"loss": 4.3387,
|
|
"mean_token_accuracy": 0.24860597997903824,
|
|
"num_tokens": 12617370.0,
|
|
"step": 6825
|
|
},
|
|
{
|
|
"entropy": 4.935642051696777,
|
|
"epoch": 5.7154811715481175,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00025218740780116037,
|
|
"loss": 4.4128,
|
|
"mean_token_accuracy": 0.24011216610670089,
|
|
"num_tokens": 12626146.0,
|
|
"step": 6830
|
|
},
|
|
{
|
|
"entropy": 4.880072355270386,
|
|
"epoch": 5.7196652719665275,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00025186632828587977,
|
|
"loss": 4.3677,
|
|
"mean_token_accuracy": 0.2451144427061081,
|
|
"num_tokens": 12635565.0,
|
|
"step": 6835
|
|
},
|
|
{
|
|
"entropy": 4.918350553512573,
|
|
"epoch": 5.7238493723849375,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002515452963759761,
|
|
"loss": 4.4551,
|
|
"mean_token_accuracy": 0.23449336290359496,
|
|
"num_tokens": 12644869.0,
|
|
"step": 6840
|
|
},
|
|
{
|
|
"entropy": 4.921226978302002,
|
|
"epoch": 5.7280334728033475,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002512243127320817,
|
|
"loss": 4.3741,
|
|
"mean_token_accuracy": 0.24369516521692275,
|
|
"num_tokens": 12655215.0,
|
|
"step": 6845
|
|
},
|
|
{
|
|
"entropy": 4.871899938583374,
|
|
"epoch": 5.7322175732217575,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002509033780147291,
|
|
"loss": 4.3299,
|
|
"mean_token_accuracy": 0.24529674649238586,
|
|
"num_tokens": 12663684.0,
|
|
"step": 6850
|
|
},
|
|
{
|
|
"entropy": 4.881000614166259,
|
|
"epoch": 5.7364016736401675,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00025058249288435037,
|
|
"loss": 4.4246,
|
|
"mean_token_accuracy": 0.2378972962498665,
|
|
"num_tokens": 12673367.0,
|
|
"step": 6855
|
|
},
|
|
{
|
|
"entropy": 4.854715728759766,
|
|
"epoch": 5.7405857740585775,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002502616580012756,
|
|
"loss": 4.3724,
|
|
"mean_token_accuracy": 0.24865390509366989,
|
|
"num_tokens": 12682472.0,
|
|
"step": 6860
|
|
},
|
|
{
|
|
"entropy": 4.8440618991851805,
|
|
"epoch": 5.7447698744769875,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024994087402573133,
|
|
"loss": 4.3567,
|
|
"mean_token_accuracy": 0.25143353193998336,
|
|
"num_tokens": 12691215.0,
|
|
"step": 6865
|
|
},
|
|
{
|
|
"entropy": 4.800286722183228,
|
|
"epoch": 5.7489539748953975,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024962014161783946,
|
|
"loss": 4.3344,
|
|
"mean_token_accuracy": 0.2383082091808319,
|
|
"num_tokens": 12700332.0,
|
|
"step": 6870
|
|
},
|
|
{
|
|
"entropy": 4.862757205963135,
|
|
"epoch": 5.7531380753138075,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024929946143761585,
|
|
"loss": 4.3862,
|
|
"mean_token_accuracy": 0.23991287797689437,
|
|
"num_tokens": 12708886.0,
|
|
"step": 6875
|
|
},
|
|
{
|
|
"entropy": 4.8845991611480715,
|
|
"epoch": 5.7573221757322175,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024897883414496844,
|
|
"loss": 4.3686,
|
|
"mean_token_accuracy": 0.24966248124837875,
|
|
"num_tokens": 12717476.0,
|
|
"step": 6880
|
|
},
|
|
{
|
|
"entropy": 5.000989770889282,
|
|
"epoch": 5.7615062761506275,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024865826039969683,
|
|
"loss": 4.5121,
|
|
"mean_token_accuracy": 0.23742893040180207,
|
|
"num_tokens": 12726737.0,
|
|
"step": 6885
|
|
},
|
|
{
|
|
"entropy": 4.865860891342163,
|
|
"epoch": 5.7656903765690375,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002483377408614903,
|
|
"loss": 4.3989,
|
|
"mean_token_accuracy": 0.2390950232744217,
|
|
"num_tokens": 12735750.0,
|
|
"step": 6890
|
|
},
|
|
{
|
|
"entropy": 4.841947078704834,
|
|
"epoch": 5.7698744769874475,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00024801727618992635,
|
|
"loss": 4.3738,
|
|
"mean_token_accuracy": 0.24022731184959412,
|
|
"num_tokens": 12745143.0,
|
|
"step": 6895
|
|
},
|
|
{
|
|
"entropy": 4.854324102401733,
|
|
"epoch": 5.7740585774058575,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024769686704446996,
|
|
"loss": 4.3829,
|
|
"mean_token_accuracy": 0.23483601957559586,
|
|
"num_tokens": 12754623.0,
|
|
"step": 6900
|
|
},
|
|
{
|
|
"entropy": 4.9409972667694095,
|
|
"epoch": 5.7782426778242675,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00024737651408447155,
|
|
"loss": 4.4121,
|
|
"mean_token_accuracy": 0.24663819819688798,
|
|
"num_tokens": 12762921.0,
|
|
"step": 6905
|
|
},
|
|
{
|
|
"entropy": 4.933979558944702,
|
|
"epoch": 5.7824267782426775,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000247056217969166,
|
|
"loss": 4.4492,
|
|
"mean_token_accuracy": 0.24589634090662002,
|
|
"num_tokens": 12772159.0,
|
|
"step": 6910
|
|
},
|
|
{
|
|
"entropy": 4.861163711547851,
|
|
"epoch": 5.786610878661088,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00024673597935767123,
|
|
"loss": 4.4411,
|
|
"mean_token_accuracy": 0.23663627356290817,
|
|
"num_tokens": 12781916.0,
|
|
"step": 6915
|
|
},
|
|
{
|
|
"entropy": 4.869154024124145,
|
|
"epoch": 5.790794979079498,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0002464157989089867,
|
|
"loss": 4.4306,
|
|
"mean_token_accuracy": 0.23866749405860901,
|
|
"num_tokens": 12791275.0,
|
|
"step": 6920
|
|
},
|
|
{
|
|
"entropy": 4.888151550292969,
|
|
"epoch": 5.794979079497908,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002460956772819925,
|
|
"loss": 4.3864,
|
|
"mean_token_accuracy": 0.2475417360663414,
|
|
"num_tokens": 12800667.0,
|
|
"step": 6925
|
|
},
|
|
{
|
|
"entropy": 4.863349771499633,
|
|
"epoch": 5.799163179916318,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00024577561513544724,
|
|
"loss": 4.3997,
|
|
"mean_token_accuracy": 0.23956514596939088,
|
|
"num_tokens": 12808986.0,
|
|
"step": 6930
|
|
},
|
|
{
|
|
"entropy": 4.897569179534912,
|
|
"epoch": 5.803347280334728,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00024545561312798756,
|
|
"loss": 4.4306,
|
|
"mean_token_accuracy": 0.24270670115947723,
|
|
"num_tokens": 12817658.0,
|
|
"step": 6935
|
|
},
|
|
{
|
|
"entropy": 4.931417036056518,
|
|
"epoch": 5.807531380753138,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024513567191812607,
|
|
"loss": 4.4202,
|
|
"mean_token_accuracy": 0.24022019207477568,
|
|
"num_tokens": 12826301.0,
|
|
"step": 6940
|
|
},
|
|
{
|
|
"entropy": 4.9625977039337155,
|
|
"epoch": 5.811715481171548,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002448157921642505,
|
|
"loss": 4.4587,
|
|
"mean_token_accuracy": 0.2328162297606468,
|
|
"num_tokens": 12836295.0,
|
|
"step": 6945
|
|
},
|
|
{
|
|
"entropy": 4.852469491958618,
|
|
"epoch": 5.815899581589958,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00024449597452462165,
|
|
"loss": 4.413,
|
|
"mean_token_accuracy": 0.241861554980278,
|
|
"num_tokens": 12847128.0,
|
|
"step": 6950
|
|
},
|
|
{
|
|
"entropy": 4.853127145767212,
|
|
"epoch": 5.820083682008368,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002441762196573729,
|
|
"loss": 4.4324,
|
|
"mean_token_accuracy": 0.23614680916070938,
|
|
"num_tokens": 12856198.0,
|
|
"step": 6955
|
|
},
|
|
{
|
|
"entropy": 4.924079656600952,
|
|
"epoch": 5.824267782426778,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00024385652822050846,
|
|
"loss": 4.3545,
|
|
"mean_token_accuracy": 0.2484133332967758,
|
|
"num_tokens": 12865513.0,
|
|
"step": 6960
|
|
},
|
|
{
|
|
"entropy": 4.912899112701416,
|
|
"epoch": 5.828451882845188,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00024353690087190188,
|
|
"loss": 4.3963,
|
|
"mean_token_accuracy": 0.23833778500556946,
|
|
"num_tokens": 12874086.0,
|
|
"step": 6965
|
|
},
|
|
{
|
|
"entropy": 4.8451258659362795,
|
|
"epoch": 5.832635983263598,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024321733826929487,
|
|
"loss": 4.4086,
|
|
"mean_token_accuracy": 0.24610654562711715,
|
|
"num_tokens": 12883382.0,
|
|
"step": 6970
|
|
},
|
|
{
|
|
"entropy": 4.950758457183838,
|
|
"epoch": 5.836820083682008,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00024289784107029588,
|
|
"loss": 4.4388,
|
|
"mean_token_accuracy": 0.24292005598545074,
|
|
"num_tokens": 12893510.0,
|
|
"step": 6975
|
|
},
|
|
{
|
|
"entropy": 4.958487892150879,
|
|
"epoch": 5.841004184100418,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00024257840993237862,
|
|
"loss": 4.4609,
|
|
"mean_token_accuracy": 0.2353873997926712,
|
|
"num_tokens": 12902719.0,
|
|
"step": 6980
|
|
},
|
|
{
|
|
"entropy": 4.905917358398438,
|
|
"epoch": 5.845188284518828,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00024225904551288124,
|
|
"loss": 4.5004,
|
|
"mean_token_accuracy": 0.23385030180215835,
|
|
"num_tokens": 12912889.0,
|
|
"step": 6985
|
|
},
|
|
{
|
|
"entropy": 4.880865669250488,
|
|
"epoch": 5.849372384937238,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00024193974846900424,
|
|
"loss": 4.3288,
|
|
"mean_token_accuracy": 0.2484860509634018,
|
|
"num_tokens": 12922454.0,
|
|
"step": 6990
|
|
},
|
|
{
|
|
"entropy": 4.877858018875122,
|
|
"epoch": 5.853556485355648,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00024162051945780967,
|
|
"loss": 4.3745,
|
|
"mean_token_accuracy": 0.24683092385530472,
|
|
"num_tokens": 12931038.0,
|
|
"step": 6995
|
|
},
|
|
{
|
|
"entropy": 4.855751323699951,
|
|
"epoch": 5.857740585774058,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002413013591362195,
|
|
"loss": 4.4466,
|
|
"mean_token_accuracy": 0.2359599232673645,
|
|
"num_tokens": 12940830.0,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"epoch": 5.857740585774058,
|
|
"eval_entropy": 4.726174839191398,
|
|
"eval_loss": 5.321946620941162,
|
|
"eval_mean_token_accuracy": 0.20184762519927113,
|
|
"eval_num_tokens": 12940830.0,
|
|
"eval_runtime": 2.7419,
|
|
"eval_samples_per_second": 1424.913,
|
|
"eval_steps_per_second": 178.342,
|
|
"step": 7000
|
|
},
|
|
{
|
|
"entropy": 4.901564311981201,
|
|
"epoch": 5.861924686192468,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00024098226816101427,
|
|
"loss": 4.4436,
|
|
"mean_token_accuracy": 0.2320139601826668,
|
|
"num_tokens": 12949159.0,
|
|
"step": 7005
|
|
},
|
|
{
|
|
"entropy": 5.00155463218689,
|
|
"epoch": 5.866108786610878,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00024066324718883205,
|
|
"loss": 4.4891,
|
|
"mean_token_accuracy": 0.23273158818483353,
|
|
"num_tokens": 12957818.0,
|
|
"step": 7010
|
|
},
|
|
{
|
|
"entropy": 4.945521306991577,
|
|
"epoch": 5.870292887029288,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002403442968761666,
|
|
"loss": 4.4956,
|
|
"mean_token_accuracy": 0.2360580235719681,
|
|
"num_tokens": 12966667.0,
|
|
"step": 7015
|
|
},
|
|
{
|
|
"entropy": 4.8954854011535645,
|
|
"epoch": 5.874476987447698,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00024002541787936644,
|
|
"loss": 4.3708,
|
|
"mean_token_accuracy": 0.23737565726041793,
|
|
"num_tokens": 12976081.0,
|
|
"step": 7020
|
|
},
|
|
{
|
|
"entropy": 4.882282972335815,
|
|
"epoch": 5.878661087866108,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002397066108546332,
|
|
"loss": 4.4655,
|
|
"mean_token_accuracy": 0.23689265549182892,
|
|
"num_tokens": 12985006.0,
|
|
"step": 7025
|
|
},
|
|
{
|
|
"entropy": 4.889105892181396,
|
|
"epoch": 5.882845188284519,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00023938787645802056,
|
|
"loss": 4.3981,
|
|
"mean_token_accuracy": 0.2426225483417511,
|
|
"num_tokens": 12993862.0,
|
|
"step": 7030
|
|
},
|
|
{
|
|
"entropy": 4.861910438537597,
|
|
"epoch": 5.887029288702929,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00023906921534543269,
|
|
"loss": 4.412,
|
|
"mean_token_accuracy": 0.2424500435590744,
|
|
"num_tokens": 13003637.0,
|
|
"step": 7035
|
|
},
|
|
{
|
|
"entropy": 4.928217363357544,
|
|
"epoch": 5.891213389121339,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023875062817262284,
|
|
"loss": 4.455,
|
|
"mean_token_accuracy": 0.2439105287194252,
|
|
"num_tokens": 13013364.0,
|
|
"step": 7040
|
|
},
|
|
{
|
|
"entropy": 4.872283983230591,
|
|
"epoch": 5.895397489539749,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00023843211559519228,
|
|
"loss": 4.4176,
|
|
"mean_token_accuracy": 0.2410770446062088,
|
|
"num_tokens": 13023432.0,
|
|
"step": 7045
|
|
},
|
|
{
|
|
"entropy": 4.930894899368286,
|
|
"epoch": 5.899581589958159,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00023811367826858854,
|
|
"loss": 4.443,
|
|
"mean_token_accuracy": 0.24367202520370485,
|
|
"num_tokens": 13034293.0,
|
|
"step": 7050
|
|
},
|
|
{
|
|
"entropy": 4.947891569137573,
|
|
"epoch": 5.903765690376569,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023779531684810462,
|
|
"loss": 4.4442,
|
|
"mean_token_accuracy": 0.23850242048501968,
|
|
"num_tokens": 13043985.0,
|
|
"step": 7055
|
|
},
|
|
{
|
|
"entropy": 4.803210592269897,
|
|
"epoch": 5.907949790794979,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00023747703198887698,
|
|
"loss": 4.2867,
|
|
"mean_token_accuracy": 0.2544312566518784,
|
|
"num_tokens": 13052339.0,
|
|
"step": 7060
|
|
},
|
|
{
|
|
"entropy": 4.898479413986206,
|
|
"epoch": 5.912133891213389,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00023715882434588483,
|
|
"loss": 4.367,
|
|
"mean_token_accuracy": 0.24173324555158615,
|
|
"num_tokens": 13060858.0,
|
|
"step": 7065
|
|
},
|
|
{
|
|
"entropy": 4.89065523147583,
|
|
"epoch": 5.916317991631799,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00023684069457394826,
|
|
"loss": 4.33,
|
|
"mean_token_accuracy": 0.2578976571559906,
|
|
"num_tokens": 13069542.0,
|
|
"step": 7070
|
|
},
|
|
{
|
|
"entropy": 4.818481731414795,
|
|
"epoch": 5.920502092050209,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023652264332772721,
|
|
"loss": 4.302,
|
|
"mean_token_accuracy": 0.25420833826065065,
|
|
"num_tokens": 13078868.0,
|
|
"step": 7075
|
|
},
|
|
{
|
|
"entropy": 4.907024478912353,
|
|
"epoch": 5.924686192468619,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00023620467126172001,
|
|
"loss": 4.4721,
|
|
"mean_token_accuracy": 0.2343473866581917,
|
|
"num_tokens": 13088573.0,
|
|
"step": 7080
|
|
},
|
|
{
|
|
"entropy": 4.896252918243408,
|
|
"epoch": 5.928870292887029,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00023588677903026202,
|
|
"loss": 4.3784,
|
|
"mean_token_accuracy": 0.24064815491437913,
|
|
"num_tokens": 13097723.0,
|
|
"step": 7085
|
|
},
|
|
{
|
|
"entropy": 4.957651519775391,
|
|
"epoch": 5.933054393305439,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00023556896728752435,
|
|
"loss": 4.3568,
|
|
"mean_token_accuracy": 0.2474076807498932,
|
|
"num_tokens": 13106071.0,
|
|
"step": 7090
|
|
},
|
|
{
|
|
"entropy": 4.847598934173584,
|
|
"epoch": 5.937238493723849,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002352512366875125,
|
|
"loss": 4.3913,
|
|
"mean_token_accuracy": 0.2439265489578247,
|
|
"num_tokens": 13115150.0,
|
|
"step": 7095
|
|
},
|
|
{
|
|
"entropy": 4.8773551940917965,
|
|
"epoch": 5.941422594142259,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000234933587884065,
|
|
"loss": 4.45,
|
|
"mean_token_accuracy": 0.24270433336496353,
|
|
"num_tokens": 13124833.0,
|
|
"step": 7100
|
|
},
|
|
{
|
|
"entropy": 4.919125699996949,
|
|
"epoch": 5.945606694560669,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000234616021530852,
|
|
"loss": 4.468,
|
|
"mean_token_accuracy": 0.23678543120622636,
|
|
"num_tokens": 13135751.0,
|
|
"step": 7105
|
|
},
|
|
{
|
|
"entropy": 4.809729719161988,
|
|
"epoch": 5.949790794979079,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00023429853828137393,
|
|
"loss": 4.3091,
|
|
"mean_token_accuracy": 0.25612871050834657,
|
|
"num_tokens": 13144285.0,
|
|
"step": 7110
|
|
},
|
|
{
|
|
"entropy": 4.858006381988526,
|
|
"epoch": 5.95397489539749,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00023398113878896033,
|
|
"loss": 4.4079,
|
|
"mean_token_accuracy": 0.24151684939861298,
|
|
"num_tokens": 13152856.0,
|
|
"step": 7115
|
|
},
|
|
{
|
|
"entropy": 4.933739137649536,
|
|
"epoch": 5.9581589958159,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023366382370676824,
|
|
"loss": 4.43,
|
|
"mean_token_accuracy": 0.24213735312223433,
|
|
"num_tokens": 13162665.0,
|
|
"step": 7120
|
|
},
|
|
{
|
|
"entropy": 4.895508050918579,
|
|
"epoch": 5.96234309623431,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00023334659368778117,
|
|
"loss": 4.3441,
|
|
"mean_token_accuracy": 0.2532011032104492,
|
|
"num_tokens": 13171110.0,
|
|
"step": 7125
|
|
},
|
|
{
|
|
"entropy": 4.880380630493164,
|
|
"epoch": 5.96652719665272,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00023302944938480752,
|
|
"loss": 4.4366,
|
|
"mean_token_accuracy": 0.23820230215787888,
|
|
"num_tokens": 13179948.0,
|
|
"step": 7130
|
|
},
|
|
{
|
|
"entropy": 4.937228775024414,
|
|
"epoch": 5.97071129707113,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00023271239145047917,
|
|
"loss": 4.4537,
|
|
"mean_token_accuracy": 0.24071571975946426,
|
|
"num_tokens": 13190214.0,
|
|
"step": 7135
|
|
},
|
|
{
|
|
"entropy": 4.799185705184937,
|
|
"epoch": 5.97489539748954,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00023239542053725044,
|
|
"loss": 4.3021,
|
|
"mean_token_accuracy": 0.2552438423037529,
|
|
"num_tokens": 13199048.0,
|
|
"step": 7140
|
|
},
|
|
{
|
|
"entropy": 4.859217119216919,
|
|
"epoch": 5.97907949790795,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00023207853729739663,
|
|
"loss": 4.3981,
|
|
"mean_token_accuracy": 0.23969229757785798,
|
|
"num_tokens": 13208553.0,
|
|
"step": 7145
|
|
},
|
|
{
|
|
"entropy": 4.946369552612305,
|
|
"epoch": 5.98326359832636,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00023176174238301228,
|
|
"loss": 4.4777,
|
|
"mean_token_accuracy": 0.23841027021408082,
|
|
"num_tokens": 13218151.0,
|
|
"step": 7150
|
|
},
|
|
{
|
|
"entropy": 4.885543632507324,
|
|
"epoch": 5.98744769874477,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002314450364460105,
|
|
"loss": 4.3762,
|
|
"mean_token_accuracy": 0.2484045907855034,
|
|
"num_tokens": 13227730.0,
|
|
"step": 7155
|
|
},
|
|
{
|
|
"entropy": 4.897786617279053,
|
|
"epoch": 5.99163179916318,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002311284201381212,
|
|
"loss": 4.4785,
|
|
"mean_token_accuracy": 0.24020008593797684,
|
|
"num_tokens": 13237449.0,
|
|
"step": 7160
|
|
},
|
|
{
|
|
"entropy": 4.8804174900054935,
|
|
"epoch": 5.99581589958159,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00023081189411088972,
|
|
"loss": 4.3758,
|
|
"mean_token_accuracy": 0.24060661494731903,
|
|
"num_tokens": 13246662.0,
|
|
"step": 7165
|
|
},
|
|
{
|
|
"entropy": 4.905799436569214,
|
|
"epoch": 6.0,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00023049545901567586,
|
|
"loss": 4.4318,
|
|
"mean_token_accuracy": 0.240387524664402,
|
|
"num_tokens": 13255932.0,
|
|
"step": 7170
|
|
},
|
|
{
|
|
"entropy": 4.836054801940918,
|
|
"epoch": 6.00418410041841,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00023017911550365233,
|
|
"loss": 4.1922,
|
|
"mean_token_accuracy": 0.2580294132232666,
|
|
"num_tokens": 13264600.0,
|
|
"step": 7175
|
|
},
|
|
{
|
|
"entropy": 4.936648845672607,
|
|
"epoch": 6.00836820083682,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022986286422580288,
|
|
"loss": 4.2819,
|
|
"mean_token_accuracy": 0.25435069054365156,
|
|
"num_tokens": 13274261.0,
|
|
"step": 7180
|
|
},
|
|
{
|
|
"entropy": 4.916930150985718,
|
|
"epoch": 6.01255230125523,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00022954670583292196,
|
|
"loss": 4.2317,
|
|
"mean_token_accuracy": 0.26068459153175355,
|
|
"num_tokens": 13284150.0,
|
|
"step": 7185
|
|
},
|
|
{
|
|
"entropy": 4.951982879638672,
|
|
"epoch": 6.01673640167364,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022923064097561265,
|
|
"loss": 4.2026,
|
|
"mean_token_accuracy": 0.24831557124853135,
|
|
"num_tokens": 13293114.0,
|
|
"step": 7190
|
|
},
|
|
{
|
|
"entropy": 4.854368877410889,
|
|
"epoch": 6.02092050209205,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002289146703042857,
|
|
"loss": 4.2658,
|
|
"mean_token_accuracy": 0.2571833238005638,
|
|
"num_tokens": 13303391.0,
|
|
"step": 7195
|
|
},
|
|
{
|
|
"entropy": 4.906893301010132,
|
|
"epoch": 6.02510460251046,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022859879446915788,
|
|
"loss": 4.3137,
|
|
"mean_token_accuracy": 0.24538363069295882,
|
|
"num_tokens": 13312372.0,
|
|
"step": 7200
|
|
},
|
|
{
|
|
"entropy": 4.941884517669678,
|
|
"epoch": 6.02928870292887,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00022828301412025094,
|
|
"loss": 4.2623,
|
|
"mean_token_accuracy": 0.25973704010248183,
|
|
"num_tokens": 13321543.0,
|
|
"step": 7205
|
|
},
|
|
{
|
|
"entropy": 4.857606506347656,
|
|
"epoch": 6.03347280334728,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002279673299073901,
|
|
"loss": 4.2018,
|
|
"mean_token_accuracy": 0.26342098563909533,
|
|
"num_tokens": 13330179.0,
|
|
"step": 7210
|
|
},
|
|
{
|
|
"entropy": 4.849253129959107,
|
|
"epoch": 6.03765690376569,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00022765174248020258,
|
|
"loss": 4.1798,
|
|
"mean_token_accuracy": 0.26064721345901487,
|
|
"num_tokens": 13339116.0,
|
|
"step": 7215
|
|
},
|
|
{
|
|
"entropy": 4.817509937286377,
|
|
"epoch": 6.0418410041841,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022733625248811656,
|
|
"loss": 4.2101,
|
|
"mean_token_accuracy": 0.26393435448408126,
|
|
"num_tokens": 13347908.0,
|
|
"step": 7220
|
|
},
|
|
{
|
|
"entropy": 4.814939737319946,
|
|
"epoch": 6.046025104602511,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022702086058035985,
|
|
"loss": 4.1644,
|
|
"mean_token_accuracy": 0.26378460228443146,
|
|
"num_tokens": 13356840.0,
|
|
"step": 7225
|
|
},
|
|
{
|
|
"entropy": 4.832661247253418,
|
|
"epoch": 6.050209205020921,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00022670556740595827,
|
|
"loss": 4.2019,
|
|
"mean_token_accuracy": 0.2560404047369957,
|
|
"num_tokens": 13365288.0,
|
|
"step": 7230
|
|
},
|
|
{
|
|
"entropy": 4.822348690032959,
|
|
"epoch": 6.054393305439331,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00022639037361373444,
|
|
"loss": 4.1858,
|
|
"mean_token_accuracy": 0.2688261762261391,
|
|
"num_tokens": 13374612.0,
|
|
"step": 7235
|
|
},
|
|
{
|
|
"entropy": 4.889249801635742,
|
|
"epoch": 6.058577405857741,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00022607527985230648,
|
|
"loss": 4.2063,
|
|
"mean_token_accuracy": 0.2606476664543152,
|
|
"num_tokens": 13384307.0,
|
|
"step": 7240
|
|
},
|
|
{
|
|
"entropy": 4.838326787948608,
|
|
"epoch": 6.062761506276151,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00022576028677008688,
|
|
"loss": 4.217,
|
|
"mean_token_accuracy": 0.2519887939095497,
|
|
"num_tokens": 13392626.0,
|
|
"step": 7245
|
|
},
|
|
{
|
|
"entropy": 4.869049549102783,
|
|
"epoch": 6.066945606694561,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0002254453950152804,
|
|
"loss": 4.21,
|
|
"mean_token_accuracy": 0.26383565962314603,
|
|
"num_tokens": 13401694.0,
|
|
"step": 7250
|
|
},
|
|
{
|
|
"entropy": 4.882580900192261,
|
|
"epoch": 6.071129707112971,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002251306052358839,
|
|
"loss": 4.2794,
|
|
"mean_token_accuracy": 0.2609379544854164,
|
|
"num_tokens": 13410330.0,
|
|
"step": 7255
|
|
},
|
|
{
|
|
"entropy": 4.846832132339477,
|
|
"epoch": 6.075313807531381,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00022481591807968417,
|
|
"loss": 4.1675,
|
|
"mean_token_accuracy": 0.2616812899708748,
|
|
"num_tokens": 13418988.0,
|
|
"step": 7260
|
|
},
|
|
{
|
|
"entropy": 4.858353805541992,
|
|
"epoch": 6.079497907949791,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002245013341942566,
|
|
"loss": 4.2842,
|
|
"mean_token_accuracy": 0.25137888044118883,
|
|
"num_tokens": 13427487.0,
|
|
"step": 7265
|
|
},
|
|
{
|
|
"entropy": 4.855068922042847,
|
|
"epoch": 6.083682008368201,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022418685422696444,
|
|
"loss": 4.2131,
|
|
"mean_token_accuracy": 0.26196681410074235,
|
|
"num_tokens": 13436450.0,
|
|
"step": 7270
|
|
},
|
|
{
|
|
"entropy": 4.841048860549927,
|
|
"epoch": 6.087866108786611,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0002238724788249568,
|
|
"loss": 4.2037,
|
|
"mean_token_accuracy": 0.2570369094610214,
|
|
"num_tokens": 13445363.0,
|
|
"step": 7275
|
|
},
|
|
{
|
|
"entropy": 4.8380491733551025,
|
|
"epoch": 6.092050209205021,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00022355820863516774,
|
|
"loss": 4.1302,
|
|
"mean_token_accuracy": 0.2694198861718178,
|
|
"num_tokens": 13453960.0,
|
|
"step": 7280
|
|
},
|
|
{
|
|
"entropy": 4.898212862014771,
|
|
"epoch": 6.096234309623431,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002232440443043147,
|
|
"loss": 4.2807,
|
|
"mean_token_accuracy": 0.2595510885119438,
|
|
"num_tokens": 13463619.0,
|
|
"step": 7285
|
|
},
|
|
{
|
|
"entropy": 4.88677453994751,
|
|
"epoch": 6.100418410041841,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002229299864788975,
|
|
"loss": 4.2314,
|
|
"mean_token_accuracy": 0.25874114632606504,
|
|
"num_tokens": 13473566.0,
|
|
"step": 7290
|
|
},
|
|
{
|
|
"entropy": 4.819874954223633,
|
|
"epoch": 6.104602510460251,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00022261603580519656,
|
|
"loss": 4.1642,
|
|
"mean_token_accuracy": 0.2670183524489403,
|
|
"num_tokens": 13481848.0,
|
|
"step": 7295
|
|
},
|
|
{
|
|
"entropy": 4.928783655166626,
|
|
"epoch": 6.108786610878661,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00022230219292927188,
|
|
"loss": 4.3382,
|
|
"mean_token_accuracy": 0.2511828035116196,
|
|
"num_tokens": 13491600.0,
|
|
"step": 7300
|
|
},
|
|
{
|
|
"entropy": 4.923250484466553,
|
|
"epoch": 6.112970711297071,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00022198845849696175,
|
|
"loss": 4.2923,
|
|
"mean_token_accuracy": 0.2563703447580338,
|
|
"num_tokens": 13500888.0,
|
|
"step": 7305
|
|
},
|
|
{
|
|
"entropy": 4.883869171142578,
|
|
"epoch": 6.117154811715481,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00022167483315388108,
|
|
"loss": 4.2449,
|
|
"mean_token_accuracy": 0.26028769463300705,
|
|
"num_tokens": 13511031.0,
|
|
"step": 7310
|
|
},
|
|
{
|
|
"entropy": 4.841921043395996,
|
|
"epoch": 6.121338912133891,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00022136131754542048,
|
|
"loss": 4.2877,
|
|
"mean_token_accuracy": 0.24345141053199768,
|
|
"num_tokens": 13520704.0,
|
|
"step": 7315
|
|
},
|
|
{
|
|
"entropy": 4.823957920074463,
|
|
"epoch": 6.125523012552302,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00022104791231674465,
|
|
"loss": 4.1658,
|
|
"mean_token_accuracy": 0.26567095071077346,
|
|
"num_tokens": 13530297.0,
|
|
"step": 7320
|
|
},
|
|
{
|
|
"entropy": 4.872324419021607,
|
|
"epoch": 6.129707112970712,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00022073461811279106,
|
|
"loss": 4.1916,
|
|
"mean_token_accuracy": 0.2590560108423233,
|
|
"num_tokens": 13538796.0,
|
|
"step": 7325
|
|
},
|
|
{
|
|
"entropy": 4.76206226348877,
|
|
"epoch": 6.133891213389122,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00022042143557826892,
|
|
"loss": 4.1336,
|
|
"mean_token_accuracy": 0.2697451084852219,
|
|
"num_tokens": 13548145.0,
|
|
"step": 7330
|
|
},
|
|
{
|
|
"entropy": 4.726474285125732,
|
|
"epoch": 6.138075313807532,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00022010836535765749,
|
|
"loss": 4.1499,
|
|
"mean_token_accuracy": 0.260024194419384,
|
|
"num_tokens": 13557025.0,
|
|
"step": 7335
|
|
},
|
|
{
|
|
"entropy": 4.793763923645019,
|
|
"epoch": 6.142259414225942,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00021979540809520504,
|
|
"loss": 4.2286,
|
|
"mean_token_accuracy": 0.26041017323732374,
|
|
"num_tokens": 13566485.0,
|
|
"step": 7340
|
|
},
|
|
{
|
|
"entropy": 4.903836107254028,
|
|
"epoch": 6.146443514644352,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0002194825644349272,
|
|
"loss": 4.236,
|
|
"mean_token_accuracy": 0.2586864173412323,
|
|
"num_tokens": 13575167.0,
|
|
"step": 7345
|
|
},
|
|
{
|
|
"entropy": 4.934427547454834,
|
|
"epoch": 6.150627615062762,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021916983502060595,
|
|
"loss": 4.2515,
|
|
"mean_token_accuracy": 0.2542705968022346,
|
|
"num_tokens": 13585757.0,
|
|
"step": 7350
|
|
},
|
|
{
|
|
"entropy": 4.8368745803833,
|
|
"epoch": 6.154811715481172,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00021885722049578817,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.24855349957942963,
|
|
"num_tokens": 13595862.0,
|
|
"step": 7355
|
|
},
|
|
{
|
|
"entropy": 4.888582181930542,
|
|
"epoch": 6.158995815899582,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002185447215037842,
|
|
"loss": 4.2607,
|
|
"mean_token_accuracy": 0.2548852160573006,
|
|
"num_tokens": 13605106.0,
|
|
"step": 7360
|
|
},
|
|
{
|
|
"entropy": 4.8575514793396,
|
|
"epoch": 6.163179916317992,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021823233868766684,
|
|
"loss": 4.2037,
|
|
"mean_token_accuracy": 0.2603456795215607,
|
|
"num_tokens": 13614317.0,
|
|
"step": 7365
|
|
},
|
|
{
|
|
"entropy": 4.818218994140625,
|
|
"epoch": 6.167364016736402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002179200726902697,
|
|
"loss": 4.1684,
|
|
"mean_token_accuracy": 0.2628387942910194,
|
|
"num_tokens": 13623196.0,
|
|
"step": 7370
|
|
},
|
|
{
|
|
"entropy": 4.80979290008545,
|
|
"epoch": 6.171548117154812,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002176079241541859,
|
|
"loss": 4.1865,
|
|
"mean_token_accuracy": 0.2569532886147499,
|
|
"num_tokens": 13631872.0,
|
|
"step": 7375
|
|
},
|
|
{
|
|
"entropy": 4.866264533996582,
|
|
"epoch": 6.175732217573222,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000217295893721767,
|
|
"loss": 4.2196,
|
|
"mean_token_accuracy": 0.25725708454847335,
|
|
"num_tokens": 13641240.0,
|
|
"step": 7380
|
|
},
|
|
{
|
|
"entropy": 4.810574102401733,
|
|
"epoch": 6.179916317991632,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021698398203512148,
|
|
"loss": 4.1908,
|
|
"mean_token_accuracy": 0.2603297963738441,
|
|
"num_tokens": 13650495.0,
|
|
"step": 7385
|
|
},
|
|
{
|
|
"entropy": 4.779106378555298,
|
|
"epoch": 6.184100418410042,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021667218973611346,
|
|
"loss": 4.156,
|
|
"mean_token_accuracy": 0.26774022579193113,
|
|
"num_tokens": 13660234.0,
|
|
"step": 7390
|
|
},
|
|
{
|
|
"entropy": 4.868104743957519,
|
|
"epoch": 6.188284518828452,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00021636051746636126,
|
|
"loss": 4.3194,
|
|
"mean_token_accuracy": 0.2498449355363846,
|
|
"num_tokens": 13670060.0,
|
|
"step": 7395
|
|
},
|
|
{
|
|
"entropy": 4.879341983795166,
|
|
"epoch": 6.192468619246862,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00021604896586723644,
|
|
"loss": 4.2638,
|
|
"mean_token_accuracy": 0.2550901651382446,
|
|
"num_tokens": 13679551.0,
|
|
"step": 7400
|
|
},
|
|
{
|
|
"entropy": 4.92070460319519,
|
|
"epoch": 6.196652719665272,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00021573753557986198,
|
|
"loss": 4.2613,
|
|
"mean_token_accuracy": 0.2546157330274582,
|
|
"num_tokens": 13688772.0,
|
|
"step": 7405
|
|
},
|
|
{
|
|
"entropy": 4.9432690143585205,
|
|
"epoch": 6.200836820083682,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00021542622724511146,
|
|
"loss": 4.3138,
|
|
"mean_token_accuracy": 0.25216483175754545,
|
|
"num_tokens": 13698110.0,
|
|
"step": 7410
|
|
},
|
|
{
|
|
"entropy": 4.901742029190063,
|
|
"epoch": 6.205020920502092,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00021511504150360724,
|
|
"loss": 4.2117,
|
|
"mean_token_accuracy": 0.2544358655810356,
|
|
"num_tokens": 13707459.0,
|
|
"step": 7415
|
|
},
|
|
{
|
|
"entropy": 4.875245571136475,
|
|
"epoch": 6.209205020920502,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00021480397899571963,
|
|
"loss": 4.3079,
|
|
"mean_token_accuracy": 0.25624307692050935,
|
|
"num_tokens": 13717596.0,
|
|
"step": 7420
|
|
},
|
|
{
|
|
"entropy": 4.817593383789062,
|
|
"epoch": 6.2133891213389125,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00021449304036156525,
|
|
"loss": 4.2101,
|
|
"mean_token_accuracy": 0.26217239797115327,
|
|
"num_tokens": 13726871.0,
|
|
"step": 7425
|
|
},
|
|
{
|
|
"entropy": 4.859941577911377,
|
|
"epoch": 6.2175732217573225,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00021418222624100587,
|
|
"loss": 4.1875,
|
|
"mean_token_accuracy": 0.2594889059662819,
|
|
"num_tokens": 13734960.0,
|
|
"step": 7430
|
|
},
|
|
{
|
|
"entropy": 4.865563774108887,
|
|
"epoch": 6.2217573221757325,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00021387153727364677,
|
|
"loss": 4.2112,
|
|
"mean_token_accuracy": 0.2646074861288071,
|
|
"num_tokens": 13743771.0,
|
|
"step": 7435
|
|
},
|
|
{
|
|
"entropy": 4.828087425231933,
|
|
"epoch": 6.2259414225941425,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0002135609740988362,
|
|
"loss": 4.1737,
|
|
"mean_token_accuracy": 0.2630107939243317,
|
|
"num_tokens": 13753080.0,
|
|
"step": 7440
|
|
},
|
|
{
|
|
"entropy": 4.879051256179809,
|
|
"epoch": 6.2301255230125525,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0002132505373556632,
|
|
"loss": 4.307,
|
|
"mean_token_accuracy": 0.2550677314400673,
|
|
"num_tokens": 13762698.0,
|
|
"step": 7445
|
|
},
|
|
{
|
|
"entropy": 4.814902400970459,
|
|
"epoch": 6.2343096234309625,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002129402276829564,
|
|
"loss": 4.2107,
|
|
"mean_token_accuracy": 0.25703648179769517,
|
|
"num_tokens": 13771693.0,
|
|
"step": 7450
|
|
},
|
|
{
|
|
"entropy": 4.9024067401885985,
|
|
"epoch": 6.2384937238493725,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00021263004571928345,
|
|
"loss": 4.3063,
|
|
"mean_token_accuracy": 0.2564240172505379,
|
|
"num_tokens": 13780939.0,
|
|
"step": 7455
|
|
},
|
|
{
|
|
"entropy": 4.821919012069702,
|
|
"epoch": 6.2426778242677825,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00021231999210294882,
|
|
"loss": 4.1642,
|
|
"mean_token_accuracy": 0.26761078387498854,
|
|
"num_tokens": 13790001.0,
|
|
"step": 7460
|
|
},
|
|
{
|
|
"entropy": 4.895594882965088,
|
|
"epoch": 6.2468619246861925,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00021201006747199313,
|
|
"loss": 4.2463,
|
|
"mean_token_accuracy": 0.2626880303025246,
|
|
"num_tokens": 13799360.0,
|
|
"step": 7465
|
|
},
|
|
{
|
|
"entropy": 4.840899038314819,
|
|
"epoch": 6.2510460251046025,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00021170027246419137,
|
|
"loss": 4.2402,
|
|
"mean_token_accuracy": 0.26169245690107346,
|
|
"num_tokens": 13808518.0,
|
|
"step": 7470
|
|
},
|
|
{
|
|
"entropy": 4.927405977249146,
|
|
"epoch": 6.2552301255230125,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00021139060771705205,
|
|
"loss": 4.3078,
|
|
"mean_token_accuracy": 0.2464145839214325,
|
|
"num_tokens": 13818144.0,
|
|
"step": 7475
|
|
},
|
|
{
|
|
"entropy": 4.8714135646820065,
|
|
"epoch": 6.2594142259414225,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00021108107386781517,
|
|
"loss": 4.214,
|
|
"mean_token_accuracy": 0.26198332458734513,
|
|
"num_tokens": 13828063.0,
|
|
"step": 7480
|
|
},
|
|
{
|
|
"entropy": 4.788683557510376,
|
|
"epoch": 6.2635983263598325,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00021077167155345166,
|
|
"loss": 4.176,
|
|
"mean_token_accuracy": 0.2565083563327789,
|
|
"num_tokens": 13837014.0,
|
|
"step": 7485
|
|
},
|
|
{
|
|
"entropy": 4.816254949569702,
|
|
"epoch": 6.2677824267782425,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002104624014106618,
|
|
"loss": 4.2658,
|
|
"mean_token_accuracy": 0.26418200731277464,
|
|
"num_tokens": 13845518.0,
|
|
"step": 7490
|
|
},
|
|
{
|
|
"entropy": 4.867457914352417,
|
|
"epoch": 6.2719665271966525,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00021015326407587389,
|
|
"loss": 4.262,
|
|
"mean_token_accuracy": 0.25674390196800234,
|
|
"num_tokens": 13854154.0,
|
|
"step": 7495
|
|
},
|
|
{
|
|
"entropy": 4.873294973373413,
|
|
"epoch": 6.2761506276150625,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00020984426018524278,
|
|
"loss": 4.2413,
|
|
"mean_token_accuracy": 0.2664510190486908,
|
|
"num_tokens": 13863404.0,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"epoch": 6.2761506276150625,
|
|
"eval_entropy": 4.720188296401915,
|
|
"eval_loss": 5.3274149894714355,
|
|
"eval_mean_token_accuracy": 0.2022889997878689,
|
|
"eval_num_tokens": 13863404.0,
|
|
"eval_runtime": 2.7447,
|
|
"eval_samples_per_second": 1423.479,
|
|
"eval_steps_per_second": 178.163,
|
|
"step": 7500
|
|
},
|
|
{
|
|
"entropy": 4.776959753036499,
|
|
"epoch": 6.2803347280334725,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020953539037464882,
|
|
"loss": 4.1131,
|
|
"mean_token_accuracy": 0.270844492316246,
|
|
"num_tokens": 13872554.0,
|
|
"step": 7505
|
|
},
|
|
{
|
|
"entropy": 4.81703748703003,
|
|
"epoch": 6.2845188284518825,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020922665527969646,
|
|
"loss": 4.2127,
|
|
"mean_token_accuracy": 0.26202008724212644,
|
|
"num_tokens": 13881740.0,
|
|
"step": 7510
|
|
},
|
|
{
|
|
"entropy": 4.893031311035156,
|
|
"epoch": 6.2887029288702925,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002089180555357128,
|
|
"loss": 4.2891,
|
|
"mean_token_accuracy": 0.26249132603406905,
|
|
"num_tokens": 13890114.0,
|
|
"step": 7515
|
|
},
|
|
{
|
|
"entropy": 4.8745018482208256,
|
|
"epoch": 6.292887029288703,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00020860959177774652,
|
|
"loss": 4.3672,
|
|
"mean_token_accuracy": 0.24493516534566878,
|
|
"num_tokens": 13899744.0,
|
|
"step": 7520
|
|
},
|
|
{
|
|
"entropy": 4.912326145172119,
|
|
"epoch": 6.297071129707113,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00020830126464056647,
|
|
"loss": 4.2485,
|
|
"mean_token_accuracy": 0.25126638263463974,
|
|
"num_tokens": 13909011.0,
|
|
"step": 7525
|
|
},
|
|
{
|
|
"entropy": 4.866432142257691,
|
|
"epoch": 6.301255230125523,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020799307475866035,
|
|
"loss": 4.2399,
|
|
"mean_token_accuracy": 0.2555455192923546,
|
|
"num_tokens": 13918435.0,
|
|
"step": 7530
|
|
},
|
|
{
|
|
"entropy": 4.853250122070312,
|
|
"epoch": 6.305439330543933,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020768502276623336,
|
|
"loss": 4.2535,
|
|
"mean_token_accuracy": 0.25673849880695343,
|
|
"num_tokens": 13927164.0,
|
|
"step": 7535
|
|
},
|
|
{
|
|
"entropy": 4.8357329845428465,
|
|
"epoch": 6.309623430962343,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020737710929720694,
|
|
"loss": 4.1538,
|
|
"mean_token_accuracy": 0.2714319363236427,
|
|
"num_tokens": 13935694.0,
|
|
"step": 7540
|
|
},
|
|
{
|
|
"entropy": 4.940909957885742,
|
|
"epoch": 6.313807531380753,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00020706933498521767,
|
|
"loss": 4.3424,
|
|
"mean_token_accuracy": 0.24053982198238372,
|
|
"num_tokens": 13946068.0,
|
|
"step": 7545
|
|
},
|
|
{
|
|
"entropy": 4.948669195175171,
|
|
"epoch": 6.317991631799163,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0002067617004636153,
|
|
"loss": 4.3033,
|
|
"mean_token_accuracy": 0.2581422358751297,
|
|
"num_tokens": 13955238.0,
|
|
"step": 7550
|
|
},
|
|
{
|
|
"entropy": 4.84863314628601,
|
|
"epoch": 6.322175732217573,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00020645420636546246,
|
|
"loss": 4.1913,
|
|
"mean_token_accuracy": 0.2703102007508278,
|
|
"num_tokens": 13964251.0,
|
|
"step": 7555
|
|
},
|
|
{
|
|
"entropy": 4.907941913604736,
|
|
"epoch": 6.326359832635983,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00020614685332353246,
|
|
"loss": 4.3523,
|
|
"mean_token_accuracy": 0.2410745769739151,
|
|
"num_tokens": 13974111.0,
|
|
"step": 7560
|
|
},
|
|
{
|
|
"entropy": 4.8534918308258055,
|
|
"epoch": 6.330543933054393,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020583964197030852,
|
|
"loss": 4.2033,
|
|
"mean_token_accuracy": 0.26038936525583267,
|
|
"num_tokens": 13982842.0,
|
|
"step": 7565
|
|
},
|
|
{
|
|
"entropy": 4.775629758834839,
|
|
"epoch": 6.334728033472803,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00020553257293798216,
|
|
"loss": 4.1742,
|
|
"mean_token_accuracy": 0.2640007361769676,
|
|
"num_tokens": 13992010.0,
|
|
"step": 7570
|
|
},
|
|
{
|
|
"entropy": 4.827420043945312,
|
|
"epoch": 6.338912133891213,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020522564685845207,
|
|
"loss": 4.1556,
|
|
"mean_token_accuracy": 0.27125497460365294,
|
|
"num_tokens": 14000746.0,
|
|
"step": 7575
|
|
},
|
|
{
|
|
"entropy": 4.868718433380127,
|
|
"epoch": 6.343096234309623,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00020491886436332287,
|
|
"loss": 4.2196,
|
|
"mean_token_accuracy": 0.2634586736559868,
|
|
"num_tokens": 14010664.0,
|
|
"step": 7580
|
|
},
|
|
{
|
|
"entropy": 4.897809267044067,
|
|
"epoch": 6.347280334728033,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00020461222608390346,
|
|
"loss": 4.2562,
|
|
"mean_token_accuracy": 0.25114535689353945,
|
|
"num_tokens": 14019473.0,
|
|
"step": 7585
|
|
},
|
|
{
|
|
"entropy": 4.8412168502807615,
|
|
"epoch": 6.351464435146443,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00020430573265120623,
|
|
"loss": 4.2512,
|
|
"mean_token_accuracy": 0.260942716896534,
|
|
"num_tokens": 14027971.0,
|
|
"step": 7590
|
|
},
|
|
{
|
|
"entropy": 4.741074085235596,
|
|
"epoch": 6.355648535564853,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0002039993846959453,
|
|
"loss": 4.0896,
|
|
"mean_token_accuracy": 0.2700889647006989,
|
|
"num_tokens": 14036972.0,
|
|
"step": 7595
|
|
},
|
|
{
|
|
"entropy": 4.854313230514526,
|
|
"epoch": 6.359832635983263,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020369318284853562,
|
|
"loss": 4.2138,
|
|
"mean_token_accuracy": 0.25523861199617387,
|
|
"num_tokens": 14046486.0,
|
|
"step": 7600
|
|
},
|
|
{
|
|
"entropy": 4.814856433868409,
|
|
"epoch": 6.364016736401673,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00020338712773909125,
|
|
"loss": 4.2094,
|
|
"mean_token_accuracy": 0.25918586552143097,
|
|
"num_tokens": 14056398.0,
|
|
"step": 7605
|
|
},
|
|
{
|
|
"entropy": 4.995163345336914,
|
|
"epoch": 6.368200836820083,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00020308121999742458,
|
|
"loss": 4.3579,
|
|
"mean_token_accuracy": 0.2454006165266037,
|
|
"num_tokens": 14065517.0,
|
|
"step": 7610
|
|
},
|
|
{
|
|
"entropy": 4.879616689682007,
|
|
"epoch": 6.372384937238493,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00020277546025304432,
|
|
"loss": 4.2398,
|
|
"mean_token_accuracy": 0.2534463942050934,
|
|
"num_tokens": 14074563.0,
|
|
"step": 7615
|
|
},
|
|
{
|
|
"entropy": 4.837642621994019,
|
|
"epoch": 6.376569037656903,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020246984913515503,
|
|
"loss": 4.2691,
|
|
"mean_token_accuracy": 0.25693084895610807,
|
|
"num_tokens": 14083125.0,
|
|
"step": 7620
|
|
},
|
|
{
|
|
"entropy": 4.8278741359710695,
|
|
"epoch": 6.380753138075314,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00020216438727265517,
|
|
"loss": 4.2125,
|
|
"mean_token_accuracy": 0.26259455680847166,
|
|
"num_tokens": 14091965.0,
|
|
"step": 7625
|
|
},
|
|
{
|
|
"entropy": 4.793202447891235,
|
|
"epoch": 6.384937238493724,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00020185907529413622,
|
|
"loss": 4.2042,
|
|
"mean_token_accuracy": 0.2579145535826683,
|
|
"num_tokens": 14100872.0,
|
|
"step": 7630
|
|
},
|
|
{
|
|
"entropy": 4.874279451370239,
|
|
"epoch": 6.389121338912134,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00020155391382788108,
|
|
"loss": 4.2476,
|
|
"mean_token_accuracy": 0.25333658456802366,
|
|
"num_tokens": 14109860.0,
|
|
"step": 7635
|
|
},
|
|
{
|
|
"entropy": 4.920699501037598,
|
|
"epoch": 6.393305439330544,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00020124890350186304,
|
|
"loss": 4.3026,
|
|
"mean_token_accuracy": 0.25044397711753846,
|
|
"num_tokens": 14119256.0,
|
|
"step": 7640
|
|
},
|
|
{
|
|
"entropy": 4.9620238780975345,
|
|
"epoch": 6.397489539748954,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002009440449437443,
|
|
"loss": 4.3825,
|
|
"mean_token_accuracy": 0.2503238096833229,
|
|
"num_tokens": 14128160.0,
|
|
"step": 7645
|
|
},
|
|
{
|
|
"entropy": 4.869436025619507,
|
|
"epoch": 6.401673640167364,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002006393387808747,
|
|
"loss": 4.3162,
|
|
"mean_token_accuracy": 0.2507800593972206,
|
|
"num_tokens": 14137810.0,
|
|
"step": 7650
|
|
},
|
|
{
|
|
"entropy": 4.894627952575684,
|
|
"epoch": 6.405857740585774,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00020033478564029063,
|
|
"loss": 4.158,
|
|
"mean_token_accuracy": 0.26564100086688996,
|
|
"num_tokens": 14145914.0,
|
|
"step": 7655
|
|
},
|
|
{
|
|
"entropy": 4.968897485733033,
|
|
"epoch": 6.410041841004184,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002000303861487134,
|
|
"loss": 4.3465,
|
|
"mean_token_accuracy": 0.24638063162565232,
|
|
"num_tokens": 14156348.0,
|
|
"step": 7660
|
|
},
|
|
{
|
|
"entropy": 4.957503986358643,
|
|
"epoch": 6.414225941422594,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00019972614093254822,
|
|
"loss": 4.3608,
|
|
"mean_token_accuracy": 0.25741453617811205,
|
|
"num_tokens": 14166079.0,
|
|
"step": 7665
|
|
},
|
|
{
|
|
"entropy": 4.852921485900879,
|
|
"epoch": 6.418410041841004,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019942205061788292,
|
|
"loss": 4.2437,
|
|
"mean_token_accuracy": 0.2569070249795914,
|
|
"num_tokens": 14175732.0,
|
|
"step": 7670
|
|
},
|
|
{
|
|
"entropy": 4.87563533782959,
|
|
"epoch": 6.422594142259414,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019911811583048645,
|
|
"loss": 4.2637,
|
|
"mean_token_accuracy": 0.2580623462796211,
|
|
"num_tokens": 14184935.0,
|
|
"step": 7675
|
|
},
|
|
{
|
|
"entropy": 4.88238754272461,
|
|
"epoch": 6.426778242677824,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00019881433719580766,
|
|
"loss": 4.3357,
|
|
"mean_token_accuracy": 0.24908436536788942,
|
|
"num_tokens": 14193693.0,
|
|
"step": 7680
|
|
},
|
|
{
|
|
"entropy": 4.825200510025025,
|
|
"epoch": 6.430962343096234,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00019851071533897425,
|
|
"loss": 4.2282,
|
|
"mean_token_accuracy": 0.27025439590215683,
|
|
"num_tokens": 14203323.0,
|
|
"step": 7685
|
|
},
|
|
{
|
|
"entropy": 4.85184268951416,
|
|
"epoch": 6.435146443514644,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00019820725088479107,
|
|
"loss": 4.205,
|
|
"mean_token_accuracy": 0.25644595474004744,
|
|
"num_tokens": 14211992.0,
|
|
"step": 7690
|
|
},
|
|
{
|
|
"entropy": 4.847141170501709,
|
|
"epoch": 6.439330543933054,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001979039444577393,
|
|
"loss": 4.2315,
|
|
"mean_token_accuracy": 0.261556901037693,
|
|
"num_tokens": 14221578.0,
|
|
"step": 7695
|
|
},
|
|
{
|
|
"entropy": 4.843842792510986,
|
|
"epoch": 6.443514644351464,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019760079668197474,
|
|
"loss": 4.2594,
|
|
"mean_token_accuracy": 0.25760510712862017,
|
|
"num_tokens": 14231411.0,
|
|
"step": 7700
|
|
},
|
|
{
|
|
"entropy": 4.848341560363769,
|
|
"epoch": 6.447698744769874,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00019729780818132664,
|
|
"loss": 4.2159,
|
|
"mean_token_accuracy": 0.26392988711595533,
|
|
"num_tokens": 14241317.0,
|
|
"step": 7705
|
|
},
|
|
{
|
|
"entropy": 4.876695680618286,
|
|
"epoch": 6.451882845188284,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019699497957929702,
|
|
"loss": 4.2906,
|
|
"mean_token_accuracy": 0.26330185532569883,
|
|
"num_tokens": 14251248.0,
|
|
"step": 7710
|
|
},
|
|
{
|
|
"entropy": 4.831506061553955,
|
|
"epoch": 6.456066945606695,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019669231149905804,
|
|
"loss": 4.2262,
|
|
"mean_token_accuracy": 0.25645307153463365,
|
|
"num_tokens": 14261487.0,
|
|
"step": 7715
|
|
},
|
|
{
|
|
"entropy": 4.887025928497314,
|
|
"epoch": 6.460251046025105,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019638980456345207,
|
|
"loss": 4.3018,
|
|
"mean_token_accuracy": 0.24319240003824233,
|
|
"num_tokens": 14271438.0,
|
|
"step": 7720
|
|
},
|
|
{
|
|
"entropy": 4.855402708053589,
|
|
"epoch": 6.464435146443515,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00019608745939498983,
|
|
"loss": 4.2438,
|
|
"mean_token_accuracy": 0.2543949201703072,
|
|
"num_tokens": 14280683.0,
|
|
"step": 7725
|
|
},
|
|
{
|
|
"entropy": 4.825745582580566,
|
|
"epoch": 6.468619246861925,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00019578527661584898,
|
|
"loss": 4.293,
|
|
"mean_token_accuracy": 0.2554644048213959,
|
|
"num_tokens": 14290041.0,
|
|
"step": 7730
|
|
},
|
|
{
|
|
"entropy": 4.891612863540649,
|
|
"epoch": 6.472803347280335,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00019548325684787311,
|
|
"loss": 4.2968,
|
|
"mean_token_accuracy": 0.2470787987112999,
|
|
"num_tokens": 14298914.0,
|
|
"step": 7735
|
|
},
|
|
{
|
|
"entropy": 4.8621903419494625,
|
|
"epoch": 6.476987447698745,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001951814007125704,
|
|
"loss": 4.2443,
|
|
"mean_token_accuracy": 0.2460426762700081,
|
|
"num_tokens": 14308347.0,
|
|
"step": 7740
|
|
},
|
|
{
|
|
"entropy": 4.819801473617554,
|
|
"epoch": 6.481171548117155,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019487970883111234,
|
|
"loss": 4.2393,
|
|
"mean_token_accuracy": 0.25540131628513335,
|
|
"num_tokens": 14317839.0,
|
|
"step": 7745
|
|
},
|
|
{
|
|
"entropy": 4.864144563674927,
|
|
"epoch": 6.485355648535565,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00019457818182433206,
|
|
"loss": 4.251,
|
|
"mean_token_accuracy": 0.2531994119286537,
|
|
"num_tokens": 14327037.0,
|
|
"step": 7750
|
|
},
|
|
{
|
|
"entropy": 4.840308284759521,
|
|
"epoch": 6.489539748953975,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00019427682031272382,
|
|
"loss": 4.2305,
|
|
"mean_token_accuracy": 0.2636267080903053,
|
|
"num_tokens": 14337417.0,
|
|
"step": 7755
|
|
},
|
|
{
|
|
"entropy": 4.965086507797241,
|
|
"epoch": 6.493723849372385,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019397562491644112,
|
|
"loss": 4.3605,
|
|
"mean_token_accuracy": 0.2530719816684723,
|
|
"num_tokens": 14347037.0,
|
|
"step": 7760
|
|
},
|
|
{
|
|
"entropy": 4.867241144180298,
|
|
"epoch": 6.497907949790795,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00019367459625529564,
|
|
"loss": 4.303,
|
|
"mean_token_accuracy": 0.25217505246400834,
|
|
"num_tokens": 14356509.0,
|
|
"step": 7765
|
|
},
|
|
{
|
|
"entropy": 4.852929401397705,
|
|
"epoch": 6.502092050209205,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019337373494875604,
|
|
"loss": 4.1666,
|
|
"mean_token_accuracy": 0.26403731256723406,
|
|
"num_tokens": 14365276.0,
|
|
"step": 7770
|
|
},
|
|
{
|
|
"entropy": 4.87110447883606,
|
|
"epoch": 6.506276150627615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00019307304161594657,
|
|
"loss": 4.2757,
|
|
"mean_token_accuracy": 0.255099730193615,
|
|
"num_tokens": 14373546.0,
|
|
"step": 7775
|
|
},
|
|
{
|
|
"entropy": 4.832482767105103,
|
|
"epoch": 6.510460251046025,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019277251687564557,
|
|
"loss": 4.239,
|
|
"mean_token_accuracy": 0.2599311962723732,
|
|
"num_tokens": 14382767.0,
|
|
"step": 7780
|
|
},
|
|
{
|
|
"entropy": 4.798986339569092,
|
|
"epoch": 6.514644351464435,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019247216134628476,
|
|
"loss": 4.2389,
|
|
"mean_token_accuracy": 0.25119818150997164,
|
|
"num_tokens": 14392602.0,
|
|
"step": 7785
|
|
},
|
|
{
|
|
"entropy": 4.823914813995361,
|
|
"epoch": 6.518828451882845,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019217197564594746,
|
|
"loss": 4.3134,
|
|
"mean_token_accuracy": 0.24683130979537965,
|
|
"num_tokens": 14402191.0,
|
|
"step": 7790
|
|
},
|
|
{
|
|
"entropy": 4.884454870223999,
|
|
"epoch": 6.523012552301255,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019187196039236772,
|
|
"loss": 4.2592,
|
|
"mean_token_accuracy": 0.2584839388728142,
|
|
"num_tokens": 14411116.0,
|
|
"step": 7795
|
|
},
|
|
{
|
|
"entropy": 4.813151216506958,
|
|
"epoch": 6.527196652719665,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00019157211620292853,
|
|
"loss": 4.1782,
|
|
"mean_token_accuracy": 0.26361566185951235,
|
|
"num_tokens": 14419725.0,
|
|
"step": 7800
|
|
},
|
|
{
|
|
"entropy": 4.86748046875,
|
|
"epoch": 6.531380753138075,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019127244369466118,
|
|
"loss": 4.3381,
|
|
"mean_token_accuracy": 0.25306046158075335,
|
|
"num_tokens": 14429904.0,
|
|
"step": 7805
|
|
},
|
|
{
|
|
"entropy": 4.820752763748169,
|
|
"epoch": 6.535564853556485,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00019097294348424348,
|
|
"loss": 4.1977,
|
|
"mean_token_accuracy": 0.2561549603939056,
|
|
"num_tokens": 14439384.0,
|
|
"step": 7810
|
|
},
|
|
{
|
|
"entropy": 4.854028034210205,
|
|
"epoch": 6.539748953974895,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00019067361618799856,
|
|
"loss": 4.1889,
|
|
"mean_token_accuracy": 0.2659531384706497,
|
|
"num_tokens": 14448002.0,
|
|
"step": 7815
|
|
},
|
|
{
|
|
"entropy": 4.841810989379883,
|
|
"epoch": 6.543933054393305,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00019037446242189398,
|
|
"loss": 4.2285,
|
|
"mean_token_accuracy": 0.2554969906806946,
|
|
"num_tokens": 14456887.0,
|
|
"step": 7820
|
|
},
|
|
{
|
|
"entropy": 4.930584192276001,
|
|
"epoch": 6.548117154811716,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00019007548280154007,
|
|
"loss": 4.3769,
|
|
"mean_token_accuracy": 0.24620434045791625,
|
|
"num_tokens": 14466615.0,
|
|
"step": 7825
|
|
},
|
|
{
|
|
"entropy": 4.835114812850952,
|
|
"epoch": 6.552301255230126,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00018977667794218894,
|
|
"loss": 4.1942,
|
|
"mean_token_accuracy": 0.2565800815820694,
|
|
"num_tokens": 14476684.0,
|
|
"step": 7830
|
|
},
|
|
{
|
|
"entropy": 4.849806690216065,
|
|
"epoch": 6.556485355648536,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001894780484587328,
|
|
"loss": 4.1986,
|
|
"mean_token_accuracy": 0.2611277252435684,
|
|
"num_tokens": 14485364.0,
|
|
"step": 7835
|
|
},
|
|
{
|
|
"entropy": 4.842734146118164,
|
|
"epoch": 6.560669456066946,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00018917959496570318,
|
|
"loss": 4.3047,
|
|
"mean_token_accuracy": 0.2552759230136871,
|
|
"num_tokens": 14495099.0,
|
|
"step": 7840
|
|
},
|
|
{
|
|
"entropy": 4.862345933914185,
|
|
"epoch": 6.564853556485356,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001888813180772694,
|
|
"loss": 4.2229,
|
|
"mean_token_accuracy": 0.25729253739118574,
|
|
"num_tokens": 14504051.0,
|
|
"step": 7845
|
|
},
|
|
{
|
|
"entropy": 4.872078514099121,
|
|
"epoch": 6.569037656903766,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018858321840723724,
|
|
"loss": 4.2264,
|
|
"mean_token_accuracy": 0.2517424374818802,
|
|
"num_tokens": 14513517.0,
|
|
"step": 7850
|
|
},
|
|
{
|
|
"entropy": 4.803602313995361,
|
|
"epoch": 6.573221757322176,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018828529656904802,
|
|
"loss": 4.2847,
|
|
"mean_token_accuracy": 0.258138407766819,
|
|
"num_tokens": 14522504.0,
|
|
"step": 7855
|
|
},
|
|
{
|
|
"entropy": 4.898182964324951,
|
|
"epoch": 6.577405857740586,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00018798755317577691,
|
|
"loss": 4.2778,
|
|
"mean_token_accuracy": 0.26196613907814026,
|
|
"num_tokens": 14532215.0,
|
|
"step": 7860
|
|
},
|
|
{
|
|
"entropy": 4.814784574508667,
|
|
"epoch": 6.581589958158996,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000187689988840132,
|
|
"loss": 4.1294,
|
|
"mean_token_accuracy": 0.27090332955121993,
|
|
"num_tokens": 14540489.0,
|
|
"step": 7865
|
|
},
|
|
{
|
|
"entropy": 4.816546487808227,
|
|
"epoch": 6.585774058577406,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001873926041744528,
|
|
"loss": 4.2386,
|
|
"mean_token_accuracy": 0.26186566352844237,
|
|
"num_tokens": 14550476.0,
|
|
"step": 7870
|
|
},
|
|
{
|
|
"entropy": 4.860863256454468,
|
|
"epoch": 6.589958158995816,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00018709539979070922,
|
|
"loss": 4.1998,
|
|
"mean_token_accuracy": 0.26297075897455213,
|
|
"num_tokens": 14559349.0,
|
|
"step": 7875
|
|
},
|
|
{
|
|
"entropy": 4.813492107391357,
|
|
"epoch": 6.594142259414226,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00018679837630049994,
|
|
"loss": 4.2578,
|
|
"mean_token_accuracy": 0.25920507311820984,
|
|
"num_tokens": 14568480.0,
|
|
"step": 7880
|
|
},
|
|
{
|
|
"entropy": 4.8453480243682865,
|
|
"epoch": 6.598326359832636,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018650153431505167,
|
|
"loss": 4.2263,
|
|
"mean_token_accuracy": 0.26624853909015656,
|
|
"num_tokens": 14578071.0,
|
|
"step": 7885
|
|
},
|
|
{
|
|
"entropy": 4.845380449295044,
|
|
"epoch": 6.602510460251046,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001862048744452175,
|
|
"loss": 4.2266,
|
|
"mean_token_accuracy": 0.2620606511831284,
|
|
"num_tokens": 14586211.0,
|
|
"step": 7890
|
|
},
|
|
{
|
|
"entropy": 4.845580625534057,
|
|
"epoch": 6.606694560669456,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001859083973014757,
|
|
"loss": 4.2617,
|
|
"mean_token_accuracy": 0.2570053443312645,
|
|
"num_tokens": 14595206.0,
|
|
"step": 7895
|
|
},
|
|
{
|
|
"entropy": 4.846336555480957,
|
|
"epoch": 6.610878661087866,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018561210349392855,
|
|
"loss": 4.2239,
|
|
"mean_token_accuracy": 0.25742473006248473,
|
|
"num_tokens": 14604569.0,
|
|
"step": 7900
|
|
},
|
|
{
|
|
"entropy": 4.886371898651123,
|
|
"epoch": 6.615062761506276,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001853159936323011,
|
|
"loss": 4.2405,
|
|
"mean_token_accuracy": 0.2562516048550606,
|
|
"num_tokens": 14614040.0,
|
|
"step": 7905
|
|
},
|
|
{
|
|
"entropy": 4.884872102737427,
|
|
"epoch": 6.619246861924686,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00018502006832593992,
|
|
"loss": 4.2373,
|
|
"mean_token_accuracy": 0.25881728529930115,
|
|
"num_tokens": 14624461.0,
|
|
"step": 7910
|
|
},
|
|
{
|
|
"entropy": 4.850576400756836,
|
|
"epoch": 6.623430962343097,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00018472432818381158,
|
|
"loss": 4.2781,
|
|
"mean_token_accuracy": 0.26179194897413255,
|
|
"num_tokens": 14633417.0,
|
|
"step": 7915
|
|
},
|
|
{
|
|
"entropy": 4.783723592758179,
|
|
"epoch": 6.627615062761507,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00018442877381450175,
|
|
"loss": 4.1779,
|
|
"mean_token_accuracy": 0.26252846270799635,
|
|
"num_tokens": 14642387.0,
|
|
"step": 7920
|
|
},
|
|
{
|
|
"entropy": 4.831194972991943,
|
|
"epoch": 6.631799163179917,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001841334058262139,
|
|
"loss": 4.1712,
|
|
"mean_token_accuracy": 0.2622673466801643,
|
|
"num_tokens": 14650613.0,
|
|
"step": 7925
|
|
},
|
|
{
|
|
"entropy": 4.827280569076538,
|
|
"epoch": 6.635983263598327,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00018383822482676777,
|
|
"loss": 4.2774,
|
|
"mean_token_accuracy": 0.2710778802633286,
|
|
"num_tokens": 14659411.0,
|
|
"step": 7930
|
|
},
|
|
{
|
|
"entropy": 4.800174856185913,
|
|
"epoch": 6.640167364016737,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018354323142359848,
|
|
"loss": 4.2144,
|
|
"mean_token_accuracy": 0.262070968747139,
|
|
"num_tokens": 14668375.0,
|
|
"step": 7935
|
|
},
|
|
{
|
|
"entropy": 4.865789747238159,
|
|
"epoch": 6.644351464435147,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00018324842622375497,
|
|
"loss": 4.2189,
|
|
"mean_token_accuracy": 0.25676227658987044,
|
|
"num_tokens": 14677608.0,
|
|
"step": 7940
|
|
},
|
|
{
|
|
"entropy": 4.89156813621521,
|
|
"epoch": 6.648535564853557,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00018295380983389905,
|
|
"loss": 4.2627,
|
|
"mean_token_accuracy": 0.2535777434706688,
|
|
"num_tokens": 14686646.0,
|
|
"step": 7945
|
|
},
|
|
{
|
|
"entropy": 4.788813686370849,
|
|
"epoch": 6.652719665271967,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001826593828603037,
|
|
"loss": 4.2056,
|
|
"mean_token_accuracy": 0.2588700696825981,
|
|
"num_tokens": 14695703.0,
|
|
"step": 7950
|
|
},
|
|
{
|
|
"entropy": 4.825063896179199,
|
|
"epoch": 6.656903765690377,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001823651459088524,
|
|
"loss": 4.2184,
|
|
"mean_token_accuracy": 0.25830917805433273,
|
|
"num_tokens": 14704493.0,
|
|
"step": 7955
|
|
},
|
|
{
|
|
"entropy": 4.86861400604248,
|
|
"epoch": 6.661087866108787,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00018207109958503747,
|
|
"loss": 4.2746,
|
|
"mean_token_accuracy": 0.2611465469002724,
|
|
"num_tokens": 14713119.0,
|
|
"step": 7960
|
|
},
|
|
{
|
|
"entropy": 4.879072427749634,
|
|
"epoch": 6.665271966527197,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000181777244493959,
|
|
"loss": 4.2506,
|
|
"mean_token_accuracy": 0.2564949855208397,
|
|
"num_tokens": 14722360.0,
|
|
"step": 7965
|
|
},
|
|
{
|
|
"entropy": 4.881319808959961,
|
|
"epoch": 6.669456066945607,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001814835812403235,
|
|
"loss": 4.2817,
|
|
"mean_token_accuracy": 0.2552459880709648,
|
|
"num_tokens": 14732066.0,
|
|
"step": 7970
|
|
},
|
|
{
|
|
"entropy": 4.839188528060913,
|
|
"epoch": 6.673640167364017,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00018119011042844286,
|
|
"loss": 4.2063,
|
|
"mean_token_accuracy": 0.2549559146165848,
|
|
"num_tokens": 14740968.0,
|
|
"step": 7975
|
|
},
|
|
{
|
|
"entropy": 4.8186040878295895,
|
|
"epoch": 6.677824267782427,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001808968326622325,
|
|
"loss": 4.2516,
|
|
"mean_token_accuracy": 0.2527588501572609,
|
|
"num_tokens": 14750574.0,
|
|
"step": 7980
|
|
},
|
|
{
|
|
"entropy": 4.828572750091553,
|
|
"epoch": 6.682008368200837,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00018060374854521116,
|
|
"loss": 4.2095,
|
|
"mean_token_accuracy": 0.2546260476112366,
|
|
"num_tokens": 14760064.0,
|
|
"step": 7985
|
|
},
|
|
{
|
|
"entropy": 4.844476890563965,
|
|
"epoch": 6.686192468619247,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00018031085868049884,
|
|
"loss": 4.2874,
|
|
"mean_token_accuracy": 0.2560105562210083,
|
|
"num_tokens": 14769721.0,
|
|
"step": 7990
|
|
},
|
|
{
|
|
"entropy": 4.897967433929443,
|
|
"epoch": 6.690376569037657,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001800181636708157,
|
|
"loss": 4.2862,
|
|
"mean_token_accuracy": 0.25470000654459,
|
|
"num_tokens": 14778895.0,
|
|
"step": 7995
|
|
},
|
|
{
|
|
"entropy": 4.871638298034668,
|
|
"epoch": 6.694560669456067,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017972566411848112,
|
|
"loss": 4.2683,
|
|
"mean_token_accuracy": 0.2533519804477692,
|
|
"num_tokens": 14787541.0,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"epoch": 6.694560669456067,
|
|
"eval_entropy": 4.742505435807086,
|
|
"eval_loss": 5.319558620452881,
|
|
"eval_mean_token_accuracy": 0.20353617866529278,
|
|
"eval_num_tokens": 14787541.0,
|
|
"eval_runtime": 2.9156,
|
|
"eval_samples_per_second": 1340.045,
|
|
"eval_steps_per_second": 167.72,
|
|
"step": 8000
|
|
},
|
|
{
|
|
"entropy": 4.887615394592285,
|
|
"epoch": 6.698744769874477,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000179433360625412,
|
|
"loss": 4.2464,
|
|
"mean_token_accuracy": 0.2638427570462227,
|
|
"num_tokens": 14795669.0,
|
|
"step": 8005
|
|
},
|
|
{
|
|
"entropy": 4.83154034614563,
|
|
"epoch": 6.702928870292887,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00017914125379312208,
|
|
"loss": 4.2162,
|
|
"mean_token_accuracy": 0.2605195254087448,
|
|
"num_tokens": 14804587.0,
|
|
"step": 8010
|
|
},
|
|
{
|
|
"entropy": 4.823407936096191,
|
|
"epoch": 6.707112970711297,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00017884934422272026,
|
|
"loss": 4.2183,
|
|
"mean_token_accuracy": 0.25581967532634736,
|
|
"num_tokens": 14813746.0,
|
|
"step": 8015
|
|
},
|
|
{
|
|
"entropy": 4.8555797100067135,
|
|
"epoch": 6.711297071129707,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017855763251490924,
|
|
"loss": 4.2665,
|
|
"mean_token_accuracy": 0.25045024454593656,
|
|
"num_tokens": 14822906.0,
|
|
"step": 8020
|
|
},
|
|
{
|
|
"entropy": 4.843943071365357,
|
|
"epoch": 6.7154811715481175,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017826611926998514,
|
|
"loss": 4.2545,
|
|
"mean_token_accuracy": 0.2562082067131996,
|
|
"num_tokens": 14832212.0,
|
|
"step": 8025
|
|
},
|
|
{
|
|
"entropy": 4.876161003112793,
|
|
"epoch": 6.7196652719665275,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001779748050878352,
|
|
"loss": 4.2478,
|
|
"mean_token_accuracy": 0.25309942811727526,
|
|
"num_tokens": 14841381.0,
|
|
"step": 8030
|
|
},
|
|
{
|
|
"entropy": 4.749048471450806,
|
|
"epoch": 6.7238493723849375,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00017768369056793727,
|
|
"loss": 4.1183,
|
|
"mean_token_accuracy": 0.2651283651590347,
|
|
"num_tokens": 14850090.0,
|
|
"step": 8035
|
|
},
|
|
{
|
|
"entropy": 4.828725385665893,
|
|
"epoch": 6.7280334728033475,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000177392776309358,
|
|
"loss": 4.2564,
|
|
"mean_token_accuracy": 0.2553395673632622,
|
|
"num_tokens": 14858885.0,
|
|
"step": 8040
|
|
},
|
|
{
|
|
"entropy": 4.8099684715271,
|
|
"epoch": 6.7322175732217575,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017710206291075248,
|
|
"loss": 4.2131,
|
|
"mean_token_accuracy": 0.25365455001592635,
|
|
"num_tokens": 14868173.0,
|
|
"step": 8045
|
|
},
|
|
{
|
|
"entropy": 4.811608123779297,
|
|
"epoch": 6.7364016736401675,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00017681155097036194,
|
|
"loss": 4.2215,
|
|
"mean_token_accuracy": 0.25941952168941496,
|
|
"num_tokens": 14877164.0,
|
|
"step": 8050
|
|
},
|
|
{
|
|
"entropy": 4.834946346282959,
|
|
"epoch": 6.7405857740585775,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001765212410860132,
|
|
"loss": 4.2531,
|
|
"mean_token_accuracy": 0.2632778599858284,
|
|
"num_tokens": 14886285.0,
|
|
"step": 8055
|
|
},
|
|
{
|
|
"entropy": 4.79612250328064,
|
|
"epoch": 6.7447698744769875,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017623113385511753,
|
|
"loss": 4.2214,
|
|
"mean_token_accuracy": 0.25318200439214705,
|
|
"num_tokens": 14895079.0,
|
|
"step": 8060
|
|
},
|
|
{
|
|
"entropy": 4.838102626800537,
|
|
"epoch": 6.7489539748953975,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001759412298746687,
|
|
"loss": 4.2446,
|
|
"mean_token_accuracy": 0.2533893495798111,
|
|
"num_tokens": 14904418.0,
|
|
"step": 8065
|
|
},
|
|
{
|
|
"entropy": 4.814104413986206,
|
|
"epoch": 6.7531380753138075,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017565152974124276,
|
|
"loss": 4.2823,
|
|
"mean_token_accuracy": 0.25026304721832277,
|
|
"num_tokens": 14913621.0,
|
|
"step": 8070
|
|
},
|
|
{
|
|
"entropy": 4.799177551269532,
|
|
"epoch": 6.7573221757322175,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.0001753620340509959,
|
|
"loss": 4.204,
|
|
"mean_token_accuracy": 0.26247245669364927,
|
|
"num_tokens": 14921621.0,
|
|
"step": 8075
|
|
},
|
|
{
|
|
"entropy": 4.760947608947754,
|
|
"epoch": 6.7615062761506275,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017507274339966368,
|
|
"loss": 4.1885,
|
|
"mean_token_accuracy": 0.26194957792758944,
|
|
"num_tokens": 14930732.0,
|
|
"step": 8080
|
|
},
|
|
{
|
|
"entropy": 4.803826761245728,
|
|
"epoch": 6.7656903765690375,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017478365838255978,
|
|
"loss": 4.2608,
|
|
"mean_token_accuracy": 0.2580926641821861,
|
|
"num_tokens": 14939626.0,
|
|
"step": 8085
|
|
},
|
|
{
|
|
"entropy": 4.879623126983643,
|
|
"epoch": 6.7698744769874475,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001744947795945746,
|
|
"loss": 4.2536,
|
|
"mean_token_accuracy": 0.26261085718870164,
|
|
"num_tokens": 14948718.0,
|
|
"step": 8090
|
|
},
|
|
{
|
|
"entropy": 4.800691747665406,
|
|
"epoch": 6.7740585774058575,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001742061076301744,
|
|
"loss": 4.0757,
|
|
"mean_token_accuracy": 0.2711473420262337,
|
|
"num_tokens": 14957703.0,
|
|
"step": 8095
|
|
},
|
|
{
|
|
"entropy": 4.778258037567139,
|
|
"epoch": 6.7782426778242675,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00017391764308339957,
|
|
"loss": 4.2021,
|
|
"mean_token_accuracy": 0.2605176389217377,
|
|
"num_tokens": 14966693.0,
|
|
"step": 8100
|
|
},
|
|
{
|
|
"entropy": 4.830504322052002,
|
|
"epoch": 6.7824267782426775,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00017362938654786383,
|
|
"loss": 4.2931,
|
|
"mean_token_accuracy": 0.25367347747087476,
|
|
"num_tokens": 14977267.0,
|
|
"step": 8105
|
|
},
|
|
{
|
|
"entropy": 4.906477403640747,
|
|
"epoch": 6.786610878661088,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00017334133861675278,
|
|
"loss": 4.2655,
|
|
"mean_token_accuracy": 0.2541155770421028,
|
|
"num_tokens": 14985648.0,
|
|
"step": 8110
|
|
},
|
|
{
|
|
"entropy": 4.832819652557373,
|
|
"epoch": 6.790794979079498,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001730534998828226,
|
|
"loss": 4.2354,
|
|
"mean_token_accuracy": 0.25536262392997744,
|
|
"num_tokens": 14994915.0,
|
|
"step": 8115
|
|
},
|
|
{
|
|
"entropy": 4.83640079498291,
|
|
"epoch": 6.794979079497908,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001727658709383993,
|
|
"loss": 4.2117,
|
|
"mean_token_accuracy": 0.2655081436038017,
|
|
"num_tokens": 15004288.0,
|
|
"step": 8120
|
|
},
|
|
{
|
|
"entropy": 4.872209692001343,
|
|
"epoch": 6.799163179916318,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00017247845237537693,
|
|
"loss": 4.271,
|
|
"mean_token_accuracy": 0.25907942801713946,
|
|
"num_tokens": 15012970.0,
|
|
"step": 8125
|
|
},
|
|
{
|
|
"entropy": 4.825591993331909,
|
|
"epoch": 6.803347280334728,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00017219124478521676,
|
|
"loss": 4.3111,
|
|
"mean_token_accuracy": 0.24774395674467087,
|
|
"num_tokens": 15022989.0,
|
|
"step": 8130
|
|
},
|
|
{
|
|
"entropy": 4.8946826457977295,
|
|
"epoch": 6.807531380753138,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00017190424875894567,
|
|
"loss": 4.278,
|
|
"mean_token_accuracy": 0.2608008742332458,
|
|
"num_tokens": 15032665.0,
|
|
"step": 8135
|
|
},
|
|
{
|
|
"entropy": 4.8771737575531,
|
|
"epoch": 6.811715481171548,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00017161746488715546,
|
|
"loss": 4.2226,
|
|
"mean_token_accuracy": 0.2609778597950935,
|
|
"num_tokens": 15041109.0,
|
|
"step": 8140
|
|
},
|
|
{
|
|
"entropy": 4.765309810638428,
|
|
"epoch": 6.815899581589958,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00017133089376000142,
|
|
"loss": 4.189,
|
|
"mean_token_accuracy": 0.26934668719768523,
|
|
"num_tokens": 15050080.0,
|
|
"step": 8145
|
|
},
|
|
{
|
|
"entropy": 4.796897125244141,
|
|
"epoch": 6.820083682008368,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001710445359672005,
|
|
"loss": 4.2321,
|
|
"mean_token_accuracy": 0.26099548786878585,
|
|
"num_tokens": 15058429.0,
|
|
"step": 8150
|
|
},
|
|
{
|
|
"entropy": 4.828517484664917,
|
|
"epoch": 6.824267782426778,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00017075839209803117,
|
|
"loss": 4.1867,
|
|
"mean_token_accuracy": 0.25331434011459353,
|
|
"num_tokens": 15068652.0,
|
|
"step": 8155
|
|
},
|
|
{
|
|
"entropy": 4.831222295761108,
|
|
"epoch": 6.828451882845188,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017047246274133144,
|
|
"loss": 4.1818,
|
|
"mean_token_accuracy": 0.2603465422987938,
|
|
"num_tokens": 15077643.0,
|
|
"step": 8160
|
|
},
|
|
{
|
|
"entropy": 4.85469913482666,
|
|
"epoch": 6.832635983263598,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001701867484854981,
|
|
"loss": 4.2486,
|
|
"mean_token_accuracy": 0.25973550975322723,
|
|
"num_tokens": 15087471.0,
|
|
"step": 8165
|
|
},
|
|
{
|
|
"entropy": 4.916524934768677,
|
|
"epoch": 6.836820083682008,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016990124991848497,
|
|
"loss": 4.3466,
|
|
"mean_token_accuracy": 0.25505239963531495,
|
|
"num_tokens": 15096357.0,
|
|
"step": 8170
|
|
},
|
|
{
|
|
"entropy": 4.847906112670898,
|
|
"epoch": 6.841004184100418,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001696159676278023,
|
|
"loss": 4.2233,
|
|
"mean_token_accuracy": 0.25698807686567304,
|
|
"num_tokens": 15105424.0,
|
|
"step": 8175
|
|
},
|
|
{
|
|
"entropy": 4.848576307296753,
|
|
"epoch": 6.845188284518828,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001693309022005153,
|
|
"loss": 4.2497,
|
|
"mean_token_accuracy": 0.2619160905480385,
|
|
"num_tokens": 15115641.0,
|
|
"step": 8180
|
|
},
|
|
{
|
|
"entropy": 4.839803695678711,
|
|
"epoch": 6.849372384937238,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00016904605422324255,
|
|
"loss": 4.203,
|
|
"mean_token_accuracy": 0.26394710689783096,
|
|
"num_tokens": 15124765.0,
|
|
"step": 8185
|
|
},
|
|
{
|
|
"entropy": 4.769663000106812,
|
|
"epoch": 6.853556485355648,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016876142428215552,
|
|
"loss": 4.1839,
|
|
"mean_token_accuracy": 0.2659631952643394,
|
|
"num_tokens": 15133591.0,
|
|
"step": 8190
|
|
},
|
|
{
|
|
"entropy": 4.821732234954834,
|
|
"epoch": 6.857740585774058,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001684770129629768,
|
|
"loss": 4.2737,
|
|
"mean_token_accuracy": 0.2502298727631569,
|
|
"num_tokens": 15143291.0,
|
|
"step": 8195
|
|
},
|
|
{
|
|
"entropy": 4.8864587306976315,
|
|
"epoch": 6.861924686192468,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016819282085097918,
|
|
"loss": 4.3323,
|
|
"mean_token_accuracy": 0.2472022533416748,
|
|
"num_tokens": 15152806.0,
|
|
"step": 8200
|
|
},
|
|
{
|
|
"entropy": 4.863205289840698,
|
|
"epoch": 6.866108786610878,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00016790884853098441,
|
|
"loss": 4.2536,
|
|
"mean_token_accuracy": 0.25089748948812485,
|
|
"num_tokens": 15163116.0,
|
|
"step": 8205
|
|
},
|
|
{
|
|
"entropy": 4.8248027801513675,
|
|
"epoch": 6.870292887029288,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001676250965873619,
|
|
"loss": 4.2393,
|
|
"mean_token_accuracy": 0.25998661667108536,
|
|
"num_tokens": 15172916.0,
|
|
"step": 8210
|
|
},
|
|
{
|
|
"entropy": 4.84943585395813,
|
|
"epoch": 6.874476987447698,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016734156560402742,
|
|
"loss": 4.2431,
|
|
"mean_token_accuracy": 0.2559011414647102,
|
|
"num_tokens": 15181715.0,
|
|
"step": 8215
|
|
},
|
|
{
|
|
"entropy": 4.827383327484131,
|
|
"epoch": 6.878661087866108,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001670582561644421,
|
|
"loss": 4.2083,
|
|
"mean_token_accuracy": 0.2605092242360115,
|
|
"num_tokens": 15191290.0,
|
|
"step": 8220
|
|
},
|
|
{
|
|
"entropy": 4.828218650817871,
|
|
"epoch": 6.882845188284519,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00016677516885161127,
|
|
"loss": 4.2229,
|
|
"mean_token_accuracy": 0.2648338571190834,
|
|
"num_tokens": 15200614.0,
|
|
"step": 8225
|
|
},
|
|
{
|
|
"entropy": 4.94260835647583,
|
|
"epoch": 6.887029288702929,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001664923042480832,
|
|
"loss": 4.3623,
|
|
"mean_token_accuracy": 0.24058161526918412,
|
|
"num_tokens": 15210793.0,
|
|
"step": 8230
|
|
},
|
|
{
|
|
"entropy": 4.811166715621948,
|
|
"epoch": 6.891213389121339,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016620966293594763,
|
|
"loss": 4.1804,
|
|
"mean_token_accuracy": 0.2617438793182373,
|
|
"num_tokens": 15219968.0,
|
|
"step": 8235
|
|
},
|
|
{
|
|
"entropy": 4.840092039108276,
|
|
"epoch": 6.895397489539749,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00016592724549683514,
|
|
"loss": 4.2374,
|
|
"mean_token_accuracy": 0.2593460738658905,
|
|
"num_tokens": 15229434.0,
|
|
"step": 8240
|
|
},
|
|
{
|
|
"entropy": 4.810783004760742,
|
|
"epoch": 6.899581589958159,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001656450525119152,
|
|
"loss": 4.17,
|
|
"mean_token_accuracy": 0.26430466026067734,
|
|
"num_tokens": 15238784.0,
|
|
"step": 8245
|
|
},
|
|
{
|
|
"entropy": 4.859014701843262,
|
|
"epoch": 6.903765690376569,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00016536308456189568,
|
|
"loss": 4.2967,
|
|
"mean_token_accuracy": 0.24801349639892578,
|
|
"num_tokens": 15247926.0,
|
|
"step": 8250
|
|
},
|
|
{
|
|
"entropy": 4.903424167633057,
|
|
"epoch": 6.907949790794979,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016508134222702133,
|
|
"loss": 4.2896,
|
|
"mean_token_accuracy": 0.2522130191326141,
|
|
"num_tokens": 15257158.0,
|
|
"step": 8255
|
|
},
|
|
{
|
|
"entropy": 4.901652097702026,
|
|
"epoch": 6.912133891213389,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016479982608707245,
|
|
"loss": 4.2766,
|
|
"mean_token_accuracy": 0.2580004781484604,
|
|
"num_tokens": 15266639.0,
|
|
"step": 8260
|
|
},
|
|
{
|
|
"entropy": 4.823661804199219,
|
|
"epoch": 6.916317991631799,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016451853672136418,
|
|
"loss": 4.2135,
|
|
"mean_token_accuracy": 0.26091996431350706,
|
|
"num_tokens": 15275446.0,
|
|
"step": 8265
|
|
},
|
|
{
|
|
"entropy": 4.831355190277099,
|
|
"epoch": 6.920502092050209,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001642374747087446,
|
|
"loss": 4.3118,
|
|
"mean_token_accuracy": 0.2499162271618843,
|
|
"num_tokens": 15285011.0,
|
|
"step": 8270
|
|
},
|
|
{
|
|
"entropy": 4.8741161823272705,
|
|
"epoch": 6.924686192468619,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00016395664062759442,
|
|
"loss": 4.4161,
|
|
"mean_token_accuracy": 0.2459487333893776,
|
|
"num_tokens": 15295066.0,
|
|
"step": 8275
|
|
},
|
|
{
|
|
"entropy": 4.93277382850647,
|
|
"epoch": 6.928870292887029,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016367603505582483,
|
|
"loss": 4.2485,
|
|
"mean_token_accuracy": 0.25511693954467773,
|
|
"num_tokens": 15304660.0,
|
|
"step": 8280
|
|
},
|
|
{
|
|
"entropy": 4.854059267044067,
|
|
"epoch": 6.933054393305439,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0001633956585708769,
|
|
"loss": 4.2258,
|
|
"mean_token_accuracy": 0.2570783257484436,
|
|
"num_tokens": 15314376.0,
|
|
"step": 8285
|
|
},
|
|
{
|
|
"entropy": 4.736228895187378,
|
|
"epoch": 6.937238493723849,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00016311551174972062,
|
|
"loss": 4.1716,
|
|
"mean_token_accuracy": 0.26641361117362977,
|
|
"num_tokens": 15323414.0,
|
|
"step": 8290
|
|
},
|
|
{
|
|
"entropy": 4.83124532699585,
|
|
"epoch": 6.941422594142259,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016283559516885285,
|
|
"loss": 4.2459,
|
|
"mean_token_accuracy": 0.25388423055410386,
|
|
"num_tokens": 15332266.0,
|
|
"step": 8295
|
|
},
|
|
{
|
|
"entropy": 4.872973203659058,
|
|
"epoch": 6.945606694560669,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00016255590940429717,
|
|
"loss": 4.2441,
|
|
"mean_token_accuracy": 0.2548660904169083,
|
|
"num_tokens": 15342238.0,
|
|
"step": 8300
|
|
},
|
|
{
|
|
"entropy": 4.945631790161133,
|
|
"epoch": 6.949790794979079,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001622764550316016,
|
|
"loss": 4.3168,
|
|
"mean_token_accuracy": 0.25005537569522857,
|
|
"num_tokens": 15351565.0,
|
|
"step": 8305
|
|
},
|
|
{
|
|
"entropy": 4.8963117599487305,
|
|
"epoch": 6.95397489539749,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001619972326258386,
|
|
"loss": 4.2664,
|
|
"mean_token_accuracy": 0.25251236855983733,
|
|
"num_tokens": 15361064.0,
|
|
"step": 8310
|
|
},
|
|
{
|
|
"entropy": 4.8093935489654545,
|
|
"epoch": 6.9581589958159,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00016171824276160288,
|
|
"loss": 4.2784,
|
|
"mean_token_accuracy": 0.2507546439766884,
|
|
"num_tokens": 15371152.0,
|
|
"step": 8315
|
|
},
|
|
{
|
|
"entropy": 4.851338720321655,
|
|
"epoch": 6.96234309623431,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00016143948601301064,
|
|
"loss": 4.2791,
|
|
"mean_token_accuracy": 0.250583416223526,
|
|
"num_tokens": 15381546.0,
|
|
"step": 8320
|
|
},
|
|
{
|
|
"entropy": 4.795394325256348,
|
|
"epoch": 6.96652719665272,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001611609629536986,
|
|
"loss": 4.2143,
|
|
"mean_token_accuracy": 0.26262331306934356,
|
|
"num_tokens": 15390581.0,
|
|
"step": 8325
|
|
},
|
|
{
|
|
"entropy": 4.849160099029541,
|
|
"epoch": 6.97071129707113,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001608826741568223,
|
|
"loss": 4.2357,
|
|
"mean_token_accuracy": 0.2629994034767151,
|
|
"num_tokens": 15400035.0,
|
|
"step": 8330
|
|
},
|
|
{
|
|
"entropy": 4.853904867172242,
|
|
"epoch": 6.97489539748954,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001606046201950555,
|
|
"loss": 4.2739,
|
|
"mean_token_accuracy": 0.25484455078840257,
|
|
"num_tokens": 15409199.0,
|
|
"step": 8335
|
|
},
|
|
{
|
|
"entropy": 4.78905725479126,
|
|
"epoch": 6.97907949790795,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00016032680164058832,
|
|
"loss": 4.1682,
|
|
"mean_token_accuracy": 0.27196655571460726,
|
|
"num_tokens": 15417851.0,
|
|
"step": 8340
|
|
},
|
|
{
|
|
"entropy": 4.8359057903289795,
|
|
"epoch": 6.98326359832636,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016004921906512699,
|
|
"loss": 4.1876,
|
|
"mean_token_accuracy": 0.2623885691165924,
|
|
"num_tokens": 15427072.0,
|
|
"step": 8345
|
|
},
|
|
{
|
|
"entropy": 4.8682746410369875,
|
|
"epoch": 6.98744769874477,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001597718730398916,
|
|
"loss": 4.2947,
|
|
"mean_token_accuracy": 0.2529307246208191,
|
|
"num_tokens": 15437486.0,
|
|
"step": 8350
|
|
},
|
|
{
|
|
"entropy": 4.846360301971435,
|
|
"epoch": 6.99163179916318,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00015949476413561564,
|
|
"loss": 4.2406,
|
|
"mean_token_accuracy": 0.2589369937777519,
|
|
"num_tokens": 15447650.0,
|
|
"step": 8355
|
|
},
|
|
{
|
|
"entropy": 4.77945032119751,
|
|
"epoch": 6.99581589958159,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00015921789292254484,
|
|
"loss": 4.1693,
|
|
"mean_token_accuracy": 0.2651693731546402,
|
|
"num_tokens": 15456650.0,
|
|
"step": 8360
|
|
},
|
|
{
|
|
"entropy": 4.791766881942749,
|
|
"epoch": 7.0,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00015894125997043551,
|
|
"loss": 4.2059,
|
|
"mean_token_accuracy": 0.2635926902294159,
|
|
"num_tokens": 15465254.0,
|
|
"step": 8365
|
|
},
|
|
{
|
|
"entropy": 4.822571039199829,
|
|
"epoch": 7.00418410041841,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00015866486584855397,
|
|
"loss": 4.0745,
|
|
"mean_token_accuracy": 0.27917979657649994,
|
|
"num_tokens": 15474018.0,
|
|
"step": 8370
|
|
},
|
|
{
|
|
"entropy": 4.874614953994751,
|
|
"epoch": 7.00836820083682,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015838871112567465,
|
|
"loss": 4.1601,
|
|
"mean_token_accuracy": 0.2659838944673538,
|
|
"num_tokens": 15483871.0,
|
|
"step": 8375
|
|
},
|
|
{
|
|
"entropy": 4.779301166534424,
|
|
"epoch": 7.01255230125523,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0001581127963700798,
|
|
"loss": 4.0868,
|
|
"mean_token_accuracy": 0.2756116956472397,
|
|
"num_tokens": 15492996.0,
|
|
"step": 8380
|
|
},
|
|
{
|
|
"entropy": 4.90011510848999,
|
|
"epoch": 7.01673640167364,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.0001578371221495575,
|
|
"loss": 4.1733,
|
|
"mean_token_accuracy": 0.2669140830636024,
|
|
"num_tokens": 15501703.0,
|
|
"step": 8385
|
|
},
|
|
{
|
|
"entropy": 4.8575090885162355,
|
|
"epoch": 7.02092050209205,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001575616890314009,
|
|
"loss": 4.1214,
|
|
"mean_token_accuracy": 0.27430482655763627,
|
|
"num_tokens": 15510110.0,
|
|
"step": 8390
|
|
},
|
|
{
|
|
"entropy": 4.754076814651489,
|
|
"epoch": 7.02510460251046,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00015728649758240727,
|
|
"loss": 4.0543,
|
|
"mean_token_accuracy": 0.2765975594520569,
|
|
"num_tokens": 15518352.0,
|
|
"step": 8395
|
|
},
|
|
{
|
|
"entropy": 4.769154357910156,
|
|
"epoch": 7.02928870292887,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001570115483688762,
|
|
"loss": 4.0594,
|
|
"mean_token_accuracy": 0.2778519704937935,
|
|
"num_tokens": 15527133.0,
|
|
"step": 8400
|
|
},
|
|
{
|
|
"entropy": 4.765915250778198,
|
|
"epoch": 7.03347280334728,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001567368419566091,
|
|
"loss": 4.0357,
|
|
"mean_token_accuracy": 0.27917644679546355,
|
|
"num_tokens": 15536326.0,
|
|
"step": 8405
|
|
},
|
|
{
|
|
"entropy": 4.9052910804748535,
|
|
"epoch": 7.03765690376569,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00015646237891090749,
|
|
"loss": 4.1772,
|
|
"mean_token_accuracy": 0.26176621615886686,
|
|
"num_tokens": 15546408.0,
|
|
"step": 8410
|
|
},
|
|
{
|
|
"entropy": 4.831066465377807,
|
|
"epoch": 7.0418410041841,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.0001561881597965722,
|
|
"loss": 4.167,
|
|
"mean_token_accuracy": 0.26272565573453904,
|
|
"num_tokens": 15554988.0,
|
|
"step": 8415
|
|
},
|
|
{
|
|
"entropy": 4.817286968231201,
|
|
"epoch": 7.046025104602511,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00015591418517790217,
|
|
"loss": 4.0233,
|
|
"mean_token_accuracy": 0.28832119703292847,
|
|
"num_tokens": 15564127.0,
|
|
"step": 8420
|
|
},
|
|
{
|
|
"entropy": 4.85369348526001,
|
|
"epoch": 7.050209205020921,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.000155640455618693,
|
|
"loss": 4.1418,
|
|
"mean_token_accuracy": 0.2658167392015457,
|
|
"num_tokens": 15573658.0,
|
|
"step": 8425
|
|
},
|
|
{
|
|
"entropy": 4.882144212722778,
|
|
"epoch": 7.054393305439331,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015536697168223626,
|
|
"loss": 4.1705,
|
|
"mean_token_accuracy": 0.26629233211278913,
|
|
"num_tokens": 15583728.0,
|
|
"step": 8430
|
|
},
|
|
{
|
|
"entropy": 4.753903436660766,
|
|
"epoch": 7.058577405857741,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001550937339313177,
|
|
"loss": 4.0776,
|
|
"mean_token_accuracy": 0.27828299403190615,
|
|
"num_tokens": 15593335.0,
|
|
"step": 8435
|
|
},
|
|
{
|
|
"entropy": 4.754533910751343,
|
|
"epoch": 7.062761506276151,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00015482074292821687,
|
|
"loss": 3.9999,
|
|
"mean_token_accuracy": 0.2760852426290512,
|
|
"num_tokens": 15603133.0,
|
|
"step": 8440
|
|
},
|
|
{
|
|
"entropy": 4.824836301803589,
|
|
"epoch": 7.066945606694561,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001545479992347055,
|
|
"loss": 4.1238,
|
|
"mean_token_accuracy": 0.2693237617611885,
|
|
"num_tokens": 15612509.0,
|
|
"step": 8445
|
|
},
|
|
{
|
|
"entropy": 4.807280683517456,
|
|
"epoch": 7.071129707112971,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00015427550341204585,
|
|
"loss": 4.097,
|
|
"mean_token_accuracy": 0.2769905000925064,
|
|
"num_tokens": 15622160.0,
|
|
"step": 8450
|
|
},
|
|
{
|
|
"entropy": 4.94478816986084,
|
|
"epoch": 7.075313807531381,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00015400325602099086,
|
|
"loss": 4.2239,
|
|
"mean_token_accuracy": 0.25962463617324827,
|
|
"num_tokens": 15631268.0,
|
|
"step": 8455
|
|
},
|
|
{
|
|
"entropy": 4.884337139129639,
|
|
"epoch": 7.079497907949791,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00015373125762178168,
|
|
"loss": 4.1657,
|
|
"mean_token_accuracy": 0.2623539239168167,
|
|
"num_tokens": 15640153.0,
|
|
"step": 8460
|
|
},
|
|
{
|
|
"entropy": 4.7956067562103275,
|
|
"epoch": 7.083682008368201,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015345950877414732,
|
|
"loss": 4.0173,
|
|
"mean_token_accuracy": 0.27455361783504484,
|
|
"num_tokens": 15647752.0,
|
|
"step": 8465
|
|
},
|
|
{
|
|
"entropy": 4.885242700576782,
|
|
"epoch": 7.087866108786611,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00015318801003730344,
|
|
"loss": 4.1677,
|
|
"mean_token_accuracy": 0.2595323011279106,
|
|
"num_tokens": 15656822.0,
|
|
"step": 8470
|
|
},
|
|
{
|
|
"entropy": 4.788377189636231,
|
|
"epoch": 7.092050209205021,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00015291676196995047,
|
|
"loss": 4.0356,
|
|
"mean_token_accuracy": 0.279031029343605,
|
|
"num_tokens": 15665045.0,
|
|
"step": 8475
|
|
},
|
|
{
|
|
"entropy": 4.823301649093628,
|
|
"epoch": 7.096234309623431,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00015264576513027367,
|
|
"loss": 4.1195,
|
|
"mean_token_accuracy": 0.2664502367377281,
|
|
"num_tokens": 15674206.0,
|
|
"step": 8480
|
|
},
|
|
{
|
|
"entropy": 4.903131628036499,
|
|
"epoch": 7.100418410041841,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 0.0001523750200759406,
|
|
"loss": 4.1984,
|
|
"mean_token_accuracy": 0.2568995729088783,
|
|
"num_tokens": 15683075.0,
|
|
"step": 8485
|
|
},
|
|
{
|
|
"entropy": 4.8078266143798825,
|
|
"epoch": 7.104602510460251,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001521045273641013,
|
|
"loss": 4.015,
|
|
"mean_token_accuracy": 0.28059446066617966,
|
|
"num_tokens": 15691698.0,
|
|
"step": 8490
|
|
},
|
|
{
|
|
"entropy": 4.869126987457276,
|
|
"epoch": 7.108786610878661,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015183428755138613,
|
|
"loss": 4.2011,
|
|
"mean_token_accuracy": 0.26358516961336137,
|
|
"num_tokens": 15701927.0,
|
|
"step": 8495
|
|
},
|
|
{
|
|
"entropy": 4.837640380859375,
|
|
"epoch": 7.112970711297071,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00015156430119390524,
|
|
"loss": 4.1448,
|
|
"mean_token_accuracy": 0.26466852575540545,
|
|
"num_tokens": 15710813.0,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"epoch": 7.112970711297071,
|
|
"eval_entropy": 4.737250124749962,
|
|
"eval_loss": 5.315697193145752,
|
|
"eval_mean_token_accuracy": 0.2044193222356964,
|
|
"eval_num_tokens": 15710813.0,
|
|
"eval_runtime": 2.7386,
|
|
"eval_samples_per_second": 1426.628,
|
|
"eval_steps_per_second": 178.557,
|
|
"step": 8500
|
|
},
|
|
{
|
|
"entropy": 4.936899328231812,
|
|
"epoch": 7.117154811715481,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00015129456884724723,
|
|
"loss": 4.1389,
|
|
"mean_token_accuracy": 0.2623409733176231,
|
|
"num_tokens": 15720559.0,
|
|
"step": 8505
|
|
},
|
|
{
|
|
"entropy": 4.827795743942261,
|
|
"epoch": 7.121338912133891,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015102509106647782,
|
|
"loss": 4.1294,
|
|
"mean_token_accuracy": 0.267277392745018,
|
|
"num_tokens": 15730649.0,
|
|
"step": 8510
|
|
},
|
|
{
|
|
"entropy": 4.811471939086914,
|
|
"epoch": 7.125523012552302,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00015075586840613902,
|
|
"loss": 4.0641,
|
|
"mean_token_accuracy": 0.28056363612413404,
|
|
"num_tokens": 15739619.0,
|
|
"step": 8515
|
|
},
|
|
{
|
|
"entropy": 4.818192338943481,
|
|
"epoch": 7.129707112970712,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00015048690142024763,
|
|
"loss": 4.1473,
|
|
"mean_token_accuracy": 0.2705888569355011,
|
|
"num_tokens": 15748457.0,
|
|
"step": 8520
|
|
},
|
|
{
|
|
"entropy": 4.782347106933594,
|
|
"epoch": 7.133891213389122,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00015021819066229452,
|
|
"loss": 4.0677,
|
|
"mean_token_accuracy": 0.27303474247455595,
|
|
"num_tokens": 15758527.0,
|
|
"step": 8525
|
|
},
|
|
{
|
|
"entropy": 4.842038249969482,
|
|
"epoch": 7.138075313807532,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00014994973668524343,
|
|
"loss": 4.1912,
|
|
"mean_token_accuracy": 0.26760219484567643,
|
|
"num_tokens": 15767980.0,
|
|
"step": 8530
|
|
},
|
|
{
|
|
"entropy": 4.767838430404663,
|
|
"epoch": 7.142259414225942,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014968154004152922,
|
|
"loss": 4.0909,
|
|
"mean_token_accuracy": 0.26861087679862977,
|
|
"num_tokens": 15777323.0,
|
|
"step": 8535
|
|
},
|
|
{
|
|
"entropy": 4.770531558990479,
|
|
"epoch": 7.146443514644352,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00014941360128305765,
|
|
"loss": 4.0962,
|
|
"mean_token_accuracy": 0.27746211737394333,
|
|
"num_tokens": 15787589.0,
|
|
"step": 8540
|
|
},
|
|
{
|
|
"entropy": 4.8456223487854,
|
|
"epoch": 7.150627615062762,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014914592096120348,
|
|
"loss": 4.1473,
|
|
"mean_token_accuracy": 0.2677858754992485,
|
|
"num_tokens": 15797248.0,
|
|
"step": 8545
|
|
},
|
|
{
|
|
"entropy": 4.837526512145996,
|
|
"epoch": 7.154811715481172,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014887849962680978,
|
|
"loss": 4.1681,
|
|
"mean_token_accuracy": 0.25567179173231125,
|
|
"num_tokens": 15806389.0,
|
|
"step": 8550
|
|
},
|
|
{
|
|
"entropy": 4.885165786743164,
|
|
"epoch": 7.158995815899582,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00014861133783018678,
|
|
"loss": 4.1093,
|
|
"mean_token_accuracy": 0.27599428296089173,
|
|
"num_tokens": 15814719.0,
|
|
"step": 8555
|
|
},
|
|
{
|
|
"entropy": 4.879810571670532,
|
|
"epoch": 7.163179916317992,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014834443612111035,
|
|
"loss": 4.2009,
|
|
"mean_token_accuracy": 0.2606581479310989,
|
|
"num_tokens": 15823751.0,
|
|
"step": 8560
|
|
},
|
|
{
|
|
"entropy": 4.869597244262695,
|
|
"epoch": 7.167364016736402,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00014807779504882147,
|
|
"loss": 4.17,
|
|
"mean_token_accuracy": 0.2654101178050041,
|
|
"num_tokens": 15832717.0,
|
|
"step": 8565
|
|
},
|
|
{
|
|
"entropy": 4.796422290802002,
|
|
"epoch": 7.171548117154812,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014781141516202439,
|
|
"loss": 4.1019,
|
|
"mean_token_accuracy": 0.27154178023338316,
|
|
"num_tokens": 15841494.0,
|
|
"step": 8570
|
|
},
|
|
{
|
|
"entropy": 4.856682395935058,
|
|
"epoch": 7.175732217573222,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001475452970088863,
|
|
"loss": 4.1811,
|
|
"mean_token_accuracy": 0.2622469291090965,
|
|
"num_tokens": 15851619.0,
|
|
"step": 8575
|
|
},
|
|
{
|
|
"entropy": 4.841749382019043,
|
|
"epoch": 7.179916317991632,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001472794411370355,
|
|
"loss": 4.1695,
|
|
"mean_token_accuracy": 0.2660231560468674,
|
|
"num_tokens": 15861563.0,
|
|
"step": 8580
|
|
},
|
|
{
|
|
"entropy": 4.8286243915557865,
|
|
"epoch": 7.184100418410042,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014701384809356053,
|
|
"loss": 4.1456,
|
|
"mean_token_accuracy": 0.27059703022241594,
|
|
"num_tokens": 15870554.0,
|
|
"step": 8585
|
|
},
|
|
{
|
|
"entropy": 4.827089500427246,
|
|
"epoch": 7.188284518828452,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.00014674851842500931,
|
|
"loss": 4.104,
|
|
"mean_token_accuracy": 0.27260050773620603,
|
|
"num_tokens": 15879679.0,
|
|
"step": 8590
|
|
},
|
|
{
|
|
"entropy": 4.772777986526489,
|
|
"epoch": 7.192468619246862,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001464834526773875,
|
|
"loss": 4.098,
|
|
"mean_token_accuracy": 0.26630769670009613,
|
|
"num_tokens": 15888533.0,
|
|
"step": 8595
|
|
},
|
|
{
|
|
"entropy": 4.840709161758423,
|
|
"epoch": 7.196652719665272,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00014621865139615795,
|
|
"loss": 4.0981,
|
|
"mean_token_accuracy": 0.2772808879613876,
|
|
"num_tokens": 15896888.0,
|
|
"step": 8600
|
|
},
|
|
{
|
|
"entropy": 4.880133533477784,
|
|
"epoch": 7.200836820083682,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014595411512623895,
|
|
"loss": 4.1644,
|
|
"mean_token_accuracy": 0.27118918895721433,
|
|
"num_tokens": 15906843.0,
|
|
"step": 8605
|
|
},
|
|
{
|
|
"entropy": 4.743094730377197,
|
|
"epoch": 7.205020920502092,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00014568984441200384,
|
|
"loss": 4.0571,
|
|
"mean_token_accuracy": 0.27775878757238387,
|
|
"num_tokens": 15915333.0,
|
|
"step": 8610
|
|
},
|
|
{
|
|
"entropy": 4.872111177444458,
|
|
"epoch": 7.209205020920502,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014542583979727908,
|
|
"loss": 4.1623,
|
|
"mean_token_accuracy": 0.2638191878795624,
|
|
"num_tokens": 15924187.0,
|
|
"step": 8615
|
|
},
|
|
{
|
|
"entropy": 4.870446348190308,
|
|
"epoch": 7.2133891213389125,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014516210182534373,
|
|
"loss": 4.1347,
|
|
"mean_token_accuracy": 0.2740634992718697,
|
|
"num_tokens": 15932763.0,
|
|
"step": 8620
|
|
},
|
|
{
|
|
"entropy": 4.8782435894012455,
|
|
"epoch": 7.2175732217573225,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00014489863103892825,
|
|
"loss": 4.072,
|
|
"mean_token_accuracy": 0.27504298388957976,
|
|
"num_tokens": 15942272.0,
|
|
"step": 8625
|
|
},
|
|
{
|
|
"entropy": 4.8433913230896,
|
|
"epoch": 7.2217573221757325,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000144635427980213,
|
|
"loss": 4.1787,
|
|
"mean_token_accuracy": 0.2600004404783249,
|
|
"num_tokens": 15952099.0,
|
|
"step": 8630
|
|
},
|
|
{
|
|
"entropy": 4.798773860931396,
|
|
"epoch": 7.2259414225941425,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00014437249319082773,
|
|
"loss": 4.0709,
|
|
"mean_token_accuracy": 0.26991727352142336,
|
|
"num_tokens": 15961206.0,
|
|
"step": 8635
|
|
},
|
|
{
|
|
"entropy": 4.785668468475341,
|
|
"epoch": 7.2301255230125525,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00014410982721184984,
|
|
"loss": 4.1063,
|
|
"mean_token_accuracy": 0.277280168235302,
|
|
"num_tokens": 15970694.0,
|
|
"step": 8640
|
|
},
|
|
{
|
|
"entropy": 4.815045595169067,
|
|
"epoch": 7.2343096234309625,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014384743058380373,
|
|
"loss": 4.0982,
|
|
"mean_token_accuracy": 0.2740779250860214,
|
|
"num_tokens": 15979271.0,
|
|
"step": 8645
|
|
},
|
|
{
|
|
"entropy": 4.835446214675903,
|
|
"epoch": 7.2384937238493725,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00014358530384665953,
|
|
"loss": 4.1314,
|
|
"mean_token_accuracy": 0.272321417927742,
|
|
"num_tokens": 15988124.0,
|
|
"step": 8650
|
|
},
|
|
{
|
|
"entropy": 4.873277378082276,
|
|
"epoch": 7.2426778242677825,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00014332344753983174,
|
|
"loss": 4.1732,
|
|
"mean_token_accuracy": 0.26711160093545916,
|
|
"num_tokens": 15998912.0,
|
|
"step": 8655
|
|
},
|
|
{
|
|
"entropy": 4.832841205596924,
|
|
"epoch": 7.2468619246861925,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00014306186220217873,
|
|
"loss": 4.1236,
|
|
"mean_token_accuracy": 0.26601000130176544,
|
|
"num_tokens": 16008295.0,
|
|
"step": 8660
|
|
},
|
|
{
|
|
"entropy": 4.8436235904693605,
|
|
"epoch": 7.2510460251046025,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014280054837200094,
|
|
"loss": 4.1107,
|
|
"mean_token_accuracy": 0.2764942929148674,
|
|
"num_tokens": 16017120.0,
|
|
"step": 8665
|
|
},
|
|
{
|
|
"entropy": 4.919704294204712,
|
|
"epoch": 7.2552301255230125,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00014253950658704034,
|
|
"loss": 4.2172,
|
|
"mean_token_accuracy": 0.2600611999630928,
|
|
"num_tokens": 16026630.0,
|
|
"step": 8670
|
|
},
|
|
{
|
|
"entropy": 4.8698629379272464,
|
|
"epoch": 7.2594142259414225,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00014227873738447883,
|
|
"loss": 4.1444,
|
|
"mean_token_accuracy": 0.27093573808670046,
|
|
"num_tokens": 16034897.0,
|
|
"step": 8675
|
|
},
|
|
{
|
|
"entropy": 4.773536825180054,
|
|
"epoch": 7.2635983263598325,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014201824130093764,
|
|
"loss": 4.0759,
|
|
"mean_token_accuracy": 0.27372115254402163,
|
|
"num_tokens": 16044387.0,
|
|
"step": 8680
|
|
},
|
|
{
|
|
"entropy": 4.789889097213745,
|
|
"epoch": 7.2677824267782425,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00014175801887247576,
|
|
"loss": 4.1043,
|
|
"mean_token_accuracy": 0.27188908159732816,
|
|
"num_tokens": 16053659.0,
|
|
"step": 8685
|
|
},
|
|
{
|
|
"entropy": 4.8129674911499025,
|
|
"epoch": 7.2719665271966525,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00014149807063458896,
|
|
"loss": 4.0959,
|
|
"mean_token_accuracy": 0.27668505609035493,
|
|
"num_tokens": 16062997.0,
|
|
"step": 8690
|
|
},
|
|
{
|
|
"entropy": 4.852210187911988,
|
|
"epoch": 7.2761506276150625,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00014123839712220921,
|
|
"loss": 4.115,
|
|
"mean_token_accuracy": 0.26653716862201693,
|
|
"num_tokens": 16071491.0,
|
|
"step": 8695
|
|
},
|
|
{
|
|
"entropy": 4.853640842437744,
|
|
"epoch": 7.2803347280334725,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001409789988697026,
|
|
"loss": 4.1927,
|
|
"mean_token_accuracy": 0.2674920499324799,
|
|
"num_tokens": 16081008.0,
|
|
"step": 8700
|
|
},
|
|
{
|
|
"entropy": 4.814996290206909,
|
|
"epoch": 7.2845188284518825,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00014071987641086926,
|
|
"loss": 4.0979,
|
|
"mean_token_accuracy": 0.2704083263874054,
|
|
"num_tokens": 16090516.0,
|
|
"step": 8705
|
|
},
|
|
{
|
|
"entropy": 4.830596590042115,
|
|
"epoch": 7.2887029288702925,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0001404610302789414,
|
|
"loss": 4.2098,
|
|
"mean_token_accuracy": 0.2660282164812088,
|
|
"num_tokens": 16100952.0,
|
|
"step": 8710
|
|
},
|
|
{
|
|
"entropy": 4.7945719242095945,
|
|
"epoch": 7.292887029288703,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00014020246100658268,
|
|
"loss": 4.1356,
|
|
"mean_token_accuracy": 0.27069662511348724,
|
|
"num_tokens": 16110493.0,
|
|
"step": 8715
|
|
},
|
|
{
|
|
"entropy": 4.818515300750732,
|
|
"epoch": 7.297071129707113,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013994416912588726,
|
|
"loss": 4.1028,
|
|
"mean_token_accuracy": 0.27514911592006686,
|
|
"num_tokens": 16119227.0,
|
|
"step": 8720
|
|
},
|
|
{
|
|
"entropy": 4.864872550964355,
|
|
"epoch": 7.301255230125523,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001396861551683781,
|
|
"loss": 4.2272,
|
|
"mean_token_accuracy": 0.2616916835308075,
|
|
"num_tokens": 16129268.0,
|
|
"step": 8725
|
|
},
|
|
{
|
|
"entropy": 4.869310808181763,
|
|
"epoch": 7.305439330543933,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001394284196650067,
|
|
"loss": 4.1651,
|
|
"mean_token_accuracy": 0.269468592107296,
|
|
"num_tokens": 16137523.0,
|
|
"step": 8730
|
|
},
|
|
{
|
|
"entropy": 4.848382949829102,
|
|
"epoch": 7.309623430962343,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001391709631461511,
|
|
"loss": 4.1832,
|
|
"mean_token_accuracy": 0.26226443350315093,
|
|
"num_tokens": 16146535.0,
|
|
"step": 8735
|
|
},
|
|
{
|
|
"entropy": 4.833006238937378,
|
|
"epoch": 7.313807531380753,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00013891378614161538,
|
|
"loss": 4.1287,
|
|
"mean_token_accuracy": 0.2732284665107727,
|
|
"num_tokens": 16155892.0,
|
|
"step": 8740
|
|
},
|
|
{
|
|
"entropy": 4.799647617340088,
|
|
"epoch": 7.317991631799163,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00013865688918062875,
|
|
"loss": 4.1174,
|
|
"mean_token_accuracy": 0.2645187646150589,
|
|
"num_tokens": 16164697.0,
|
|
"step": 8745
|
|
},
|
|
{
|
|
"entropy": 4.8323486328125,
|
|
"epoch": 7.322175732217573,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00013840027279184353,
|
|
"loss": 4.1256,
|
|
"mean_token_accuracy": 0.2690981552004814,
|
|
"num_tokens": 16173387.0,
|
|
"step": 8750
|
|
},
|
|
{
|
|
"entropy": 4.831083345413208,
|
|
"epoch": 7.326359832635983,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013814393750333524,
|
|
"loss": 4.0784,
|
|
"mean_token_accuracy": 0.2754525512456894,
|
|
"num_tokens": 16182438.0,
|
|
"step": 8755
|
|
},
|
|
{
|
|
"entropy": 4.79350700378418,
|
|
"epoch": 7.330543933054393,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00013788788384260043,
|
|
"loss": 4.0543,
|
|
"mean_token_accuracy": 0.2726194202899933,
|
|
"num_tokens": 16193149.0,
|
|
"step": 8760
|
|
},
|
|
{
|
|
"entropy": 4.837381076812744,
|
|
"epoch": 7.334728033472803,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013763211233655656,
|
|
"loss": 4.2033,
|
|
"mean_token_accuracy": 0.2669885277748108,
|
|
"num_tokens": 16202354.0,
|
|
"step": 8765
|
|
},
|
|
{
|
|
"entropy": 4.806185340881347,
|
|
"epoch": 7.338912133891213,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00013737662351154028,
|
|
"loss": 4.1382,
|
|
"mean_token_accuracy": 0.26164352148771286,
|
|
"num_tokens": 16210608.0,
|
|
"step": 8770
|
|
},
|
|
{
|
|
"entropy": 4.885645055770874,
|
|
"epoch": 7.343096234309623,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013712141789330642,
|
|
"loss": 4.182,
|
|
"mean_token_accuracy": 0.2716671824455261,
|
|
"num_tokens": 16219544.0,
|
|
"step": 8775
|
|
},
|
|
{
|
|
"entropy": 4.830056810379029,
|
|
"epoch": 7.347280334728033,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013686649600702728,
|
|
"loss": 4.1017,
|
|
"mean_token_accuracy": 0.27059135437011717,
|
|
"num_tokens": 16228820.0,
|
|
"step": 8780
|
|
},
|
|
{
|
|
"entropy": 4.764642333984375,
|
|
"epoch": 7.351464435146443,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00013661185837729085,
|
|
"loss": 4.0053,
|
|
"mean_token_accuracy": 0.28363204300403594,
|
|
"num_tokens": 16237751.0,
|
|
"step": 8785
|
|
},
|
|
{
|
|
"entropy": 4.8326116561889645,
|
|
"epoch": 7.355648535564853,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00013635750552810065,
|
|
"loss": 4.1212,
|
|
"mean_token_accuracy": 0.26770464777946473,
|
|
"num_tokens": 16246944.0,
|
|
"step": 8790
|
|
},
|
|
{
|
|
"entropy": 4.8570832252502445,
|
|
"epoch": 7.359832635983263,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013610343798287376,
|
|
"loss": 4.2051,
|
|
"mean_token_accuracy": 0.2649522989988327,
|
|
"num_tokens": 16256565.0,
|
|
"step": 8795
|
|
},
|
|
{
|
|
"entropy": 4.791439628601074,
|
|
"epoch": 7.364016736401673,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00013584965626444045,
|
|
"loss": 4.0613,
|
|
"mean_token_accuracy": 0.28343687355518343,
|
|
"num_tokens": 16264886.0,
|
|
"step": 8800
|
|
},
|
|
{
|
|
"entropy": 4.922781562805175,
|
|
"epoch": 7.368200836820083,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00013559616089504285,
|
|
"loss": 4.2424,
|
|
"mean_token_accuracy": 0.25674278289079666,
|
|
"num_tokens": 16274039.0,
|
|
"step": 8805
|
|
},
|
|
{
|
|
"entropy": 4.773569774627686,
|
|
"epoch": 7.372384937238493,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013534295239633342,
|
|
"loss": 4.1296,
|
|
"mean_token_accuracy": 0.27022925317287444,
|
|
"num_tokens": 16283274.0,
|
|
"step": 8810
|
|
},
|
|
{
|
|
"entropy": 4.802807807922363,
|
|
"epoch": 7.376569037656903,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001350900312893747,
|
|
"loss": 4.0894,
|
|
"mean_token_accuracy": 0.274614529311657,
|
|
"num_tokens": 16292048.0,
|
|
"step": 8815
|
|
},
|
|
{
|
|
"entropy": 4.8299750804901125,
|
|
"epoch": 7.380753138075314,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00013483739809463753,
|
|
"loss": 4.1027,
|
|
"mean_token_accuracy": 0.2684297293424606,
|
|
"num_tokens": 16301910.0,
|
|
"step": 8820
|
|
},
|
|
{
|
|
"entropy": 4.839323091506958,
|
|
"epoch": 7.384937238493724,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00013458505333200044,
|
|
"loss": 4.1166,
|
|
"mean_token_accuracy": 0.2716024026274681,
|
|
"num_tokens": 16311221.0,
|
|
"step": 8825
|
|
},
|
|
{
|
|
"entropy": 4.779283666610718,
|
|
"epoch": 7.389121338912134,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013433299752074857,
|
|
"loss": 4.0588,
|
|
"mean_token_accuracy": 0.28303671777248385,
|
|
"num_tokens": 16319530.0,
|
|
"step": 8830
|
|
},
|
|
{
|
|
"entropy": 4.7277318954467775,
|
|
"epoch": 7.393305439330544,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.000134081231179572,
|
|
"loss": 4.0174,
|
|
"mean_token_accuracy": 0.2853615999221802,
|
|
"num_tokens": 16327927.0,
|
|
"step": 8835
|
|
},
|
|
{
|
|
"entropy": 4.867428970336914,
|
|
"epoch": 7.397489539748954,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00013382975482656556,
|
|
"loss": 4.1672,
|
|
"mean_token_accuracy": 0.26182259172201156,
|
|
"num_tokens": 16337551.0,
|
|
"step": 8840
|
|
},
|
|
{
|
|
"entropy": 4.787972402572632,
|
|
"epoch": 7.401673640167364,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.0001335785689792271,
|
|
"loss": 4.1128,
|
|
"mean_token_accuracy": 0.273408779501915,
|
|
"num_tokens": 16345937.0,
|
|
"step": 8845
|
|
},
|
|
{
|
|
"entropy": 4.845895481109619,
|
|
"epoch": 7.405857740585774,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001333276741544566,
|
|
"loss": 4.1835,
|
|
"mean_token_accuracy": 0.260826975107193,
|
|
"num_tokens": 16355797.0,
|
|
"step": 8850
|
|
},
|
|
{
|
|
"entropy": 4.836893701553345,
|
|
"epoch": 7.410041841004184,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00013307707086855546,
|
|
"loss": 4.1171,
|
|
"mean_token_accuracy": 0.2721149086952209,
|
|
"num_tokens": 16364547.0,
|
|
"step": 8855
|
|
},
|
|
{
|
|
"entropy": 4.7586305141448975,
|
|
"epoch": 7.414225941422594,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001328267596372248,
|
|
"loss": 4.0213,
|
|
"mean_token_accuracy": 0.2917042404413223,
|
|
"num_tokens": 16373506.0,
|
|
"step": 8860
|
|
},
|
|
{
|
|
"entropy": 4.825217390060425,
|
|
"epoch": 7.418410041841004,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000132576740975565,
|
|
"loss": 4.0927,
|
|
"mean_token_accuracy": 0.2783514425158501,
|
|
"num_tokens": 16383006.0,
|
|
"step": 8865
|
|
},
|
|
{
|
|
"entropy": 4.841627359390259,
|
|
"epoch": 7.422594142259414,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013232701539807415,
|
|
"loss": 4.1683,
|
|
"mean_token_accuracy": 0.26749606877565385,
|
|
"num_tokens": 16392721.0,
|
|
"step": 8870
|
|
},
|
|
{
|
|
"entropy": 4.776940774917603,
|
|
"epoch": 7.426778242677824,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.0001320775834186475,
|
|
"loss": 4.1052,
|
|
"mean_token_accuracy": 0.2708070531487465,
|
|
"num_tokens": 16402404.0,
|
|
"step": 8875
|
|
},
|
|
{
|
|
"entropy": 4.8672322750091555,
|
|
"epoch": 7.430962343096234,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013182844555057583,
|
|
"loss": 4.1501,
|
|
"mean_token_accuracy": 0.26621402651071546,
|
|
"num_tokens": 16411386.0,
|
|
"step": 8880
|
|
},
|
|
{
|
|
"entropy": 4.793548059463501,
|
|
"epoch": 7.435146443514644,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001315796023065448,
|
|
"loss": 4.0973,
|
|
"mean_token_accuracy": 0.2676581710577011,
|
|
"num_tokens": 16420407.0,
|
|
"step": 8885
|
|
},
|
|
{
|
|
"entropy": 4.771454763412476,
|
|
"epoch": 7.439330543933054,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00013133105419863394,
|
|
"loss": 4.0231,
|
|
"mean_token_accuracy": 0.28709601759910586,
|
|
"num_tokens": 16429197.0,
|
|
"step": 8890
|
|
},
|
|
{
|
|
"entropy": 4.828998041152954,
|
|
"epoch": 7.443514644351464,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00013108280173831517,
|
|
"loss": 4.1646,
|
|
"mean_token_accuracy": 0.2649181753396988,
|
|
"num_tokens": 16438144.0,
|
|
"step": 8895
|
|
},
|
|
{
|
|
"entropy": 4.80730504989624,
|
|
"epoch": 7.447698744769874,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.0001308348454364522,
|
|
"loss": 4.1626,
|
|
"mean_token_accuracy": 0.2705792337656021,
|
|
"num_tokens": 16447861.0,
|
|
"step": 8900
|
|
},
|
|
{
|
|
"entropy": 4.7793128490448,
|
|
"epoch": 7.451882845188284,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00013058718580329917,
|
|
"loss": 3.9903,
|
|
"mean_token_accuracy": 0.2890268087387085,
|
|
"num_tokens": 16456937.0,
|
|
"step": 8905
|
|
},
|
|
{
|
|
"entropy": 4.8487780570983885,
|
|
"epoch": 7.456066945606695,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0001303398233484999,
|
|
"loss": 4.1455,
|
|
"mean_token_accuracy": 0.26934240758419037,
|
|
"num_tokens": 16466714.0,
|
|
"step": 8910
|
|
},
|
|
{
|
|
"entropy": 4.817425537109375,
|
|
"epoch": 7.460251046025105,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00013009275858108648,
|
|
"loss": 4.0861,
|
|
"mean_token_accuracy": 0.2757225424051285,
|
|
"num_tokens": 16475219.0,
|
|
"step": 8915
|
|
},
|
|
{
|
|
"entropy": 4.766094112396241,
|
|
"epoch": 7.464435146443515,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0001298459920094784,
|
|
"loss": 4.0273,
|
|
"mean_token_accuracy": 0.2786060556769371,
|
|
"num_tokens": 16484169.0,
|
|
"step": 8920
|
|
},
|
|
{
|
|
"entropy": 4.8142975807189945,
|
|
"epoch": 7.468619246861925,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.0001295995241414818,
|
|
"loss": 4.1308,
|
|
"mean_token_accuracy": 0.2729588970541954,
|
|
"num_tokens": 16493413.0,
|
|
"step": 8925
|
|
},
|
|
{
|
|
"entropy": 4.787752389907837,
|
|
"epoch": 7.472803347280335,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00012935335548428767,
|
|
"loss": 4.0887,
|
|
"mean_token_accuracy": 0.27368318289518356,
|
|
"num_tokens": 16503195.0,
|
|
"step": 8930
|
|
},
|
|
{
|
|
"entropy": 4.773417139053345,
|
|
"epoch": 7.476987447698745,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0001291074865444718,
|
|
"loss": 4.0183,
|
|
"mean_token_accuracy": 0.2713374882936478,
|
|
"num_tokens": 16511168.0,
|
|
"step": 8935
|
|
},
|
|
{
|
|
"entropy": 4.785832452774048,
|
|
"epoch": 7.481171548117155,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00012886191782799267,
|
|
"loss": 4.102,
|
|
"mean_token_accuracy": 0.27533206790685655,
|
|
"num_tokens": 16520125.0,
|
|
"step": 8940
|
|
},
|
|
{
|
|
"entropy": 4.750186729431152,
|
|
"epoch": 7.485355648535565,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00012861664984019144,
|
|
"loss": 4.0222,
|
|
"mean_token_accuracy": 0.2755730390548706,
|
|
"num_tokens": 16528121.0,
|
|
"step": 8945
|
|
},
|
|
{
|
|
"entropy": 4.7566142082214355,
|
|
"epoch": 7.489539748953975,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00012837168308579012,
|
|
"loss": 4.0717,
|
|
"mean_token_accuracy": 0.2763488918542862,
|
|
"num_tokens": 16537741.0,
|
|
"step": 8950
|
|
},
|
|
{
|
|
"entropy": 4.832406759262085,
|
|
"epoch": 7.493723849372385,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012812701806889077,
|
|
"loss": 4.1539,
|
|
"mean_token_accuracy": 0.2622687414288521,
|
|
"num_tokens": 16548091.0,
|
|
"step": 8955
|
|
},
|
|
{
|
|
"entropy": 4.846547842025757,
|
|
"epoch": 7.497907949790795,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001278826552929749,
|
|
"loss": 4.1245,
|
|
"mean_token_accuracy": 0.27143164426088334,
|
|
"num_tokens": 16557011.0,
|
|
"step": 8960
|
|
},
|
|
{
|
|
"entropy": 4.857215690612793,
|
|
"epoch": 7.502092050209205,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.0001276385952609016,
|
|
"loss": 4.1692,
|
|
"mean_token_accuracy": 0.26139678359031676,
|
|
"num_tokens": 16565868.0,
|
|
"step": 8965
|
|
},
|
|
{
|
|
"entropy": 4.841455507278442,
|
|
"epoch": 7.506276150627615,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00012739483847490734,
|
|
"loss": 4.1414,
|
|
"mean_token_accuracy": 0.27098162919282914,
|
|
"num_tokens": 16575131.0,
|
|
"step": 8970
|
|
},
|
|
{
|
|
"entropy": 4.8561851501464846,
|
|
"epoch": 7.510460251046025,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012715138543660425,
|
|
"loss": 4.175,
|
|
"mean_token_accuracy": 0.26232859641313555,
|
|
"num_tokens": 16584789.0,
|
|
"step": 8975
|
|
},
|
|
{
|
|
"entropy": 4.864765548706055,
|
|
"epoch": 7.514644351464435,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0001269082366469797,
|
|
"loss": 4.1621,
|
|
"mean_token_accuracy": 0.2656904399394989,
|
|
"num_tokens": 16594488.0,
|
|
"step": 8980
|
|
},
|
|
{
|
|
"entropy": 4.785166931152344,
|
|
"epoch": 7.518828451882845,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00012666539260639475,
|
|
"loss": 4.0453,
|
|
"mean_token_accuracy": 0.2732027381658554,
|
|
"num_tokens": 16602691.0,
|
|
"step": 8985
|
|
},
|
|
{
|
|
"entropy": 4.887872362136841,
|
|
"epoch": 7.523012552301255,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0001264228538145833,
|
|
"loss": 4.1611,
|
|
"mean_token_accuracy": 0.27232592552900314,
|
|
"num_tokens": 16611843.0,
|
|
"step": 8990
|
|
},
|
|
{
|
|
"entropy": 4.886546516418457,
|
|
"epoch": 7.527196652719665,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 0.00012618062077065126,
|
|
"loss": 4.168,
|
|
"mean_token_accuracy": 0.2587636739015579,
|
|
"num_tokens": 16620587.0,
|
|
"step": 8995
|
|
},
|
|
{
|
|
"entropy": 4.859817123413086,
|
|
"epoch": 7.531380753138075,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00012593869397307527,
|
|
"loss": 4.2384,
|
|
"mean_token_accuracy": 0.25846696048974993,
|
|
"num_tokens": 16631113.0,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"epoch": 7.531380753138075,
|
|
"eval_entropy": 4.711305785520433,
|
|
"eval_loss": 5.310866355895996,
|
|
"eval_mean_token_accuracy": 0.2047013683065559,
|
|
"eval_num_tokens": 16631113.0,
|
|
"eval_runtime": 2.7476,
|
|
"eval_samples_per_second": 1421.956,
|
|
"eval_steps_per_second": 177.972,
|
|
"step": 9000
|
|
},
|
|
{
|
|
"entropy": 4.852446603775024,
|
|
"epoch": 7.535564853556485,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00012569707391970187,
|
|
"loss": 4.034,
|
|
"mean_token_accuracy": 0.28020896315574645,
|
|
"num_tokens": 16639747.0,
|
|
"step": 9005
|
|
},
|
|
{
|
|
"entropy": 4.871560001373291,
|
|
"epoch": 7.539748953974895,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012545576110774617,
|
|
"loss": 4.1016,
|
|
"mean_token_accuracy": 0.2760698705911636,
|
|
"num_tokens": 16648619.0,
|
|
"step": 9010
|
|
},
|
|
{
|
|
"entropy": 4.834973287582398,
|
|
"epoch": 7.543933054393305,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00012521475603379117,
|
|
"loss": 4.1836,
|
|
"mean_token_accuracy": 0.27270149886608125,
|
|
"num_tokens": 16656860.0,
|
|
"step": 9015
|
|
},
|
|
{
|
|
"entropy": 4.887981605529785,
|
|
"epoch": 7.548117154811716,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00012497405919378662,
|
|
"loss": 4.2207,
|
|
"mean_token_accuracy": 0.26337409764528275,
|
|
"num_tokens": 16666121.0,
|
|
"step": 9020
|
|
},
|
|
{
|
|
"entropy": 4.832073307037353,
|
|
"epoch": 7.552301255230126,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001247336710830478,
|
|
"loss": 4.0921,
|
|
"mean_token_accuracy": 0.2733010768890381,
|
|
"num_tokens": 16675296.0,
|
|
"step": 9025
|
|
},
|
|
{
|
|
"entropy": 4.783435869216919,
|
|
"epoch": 7.556485355648536,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 0.00012449359219625494,
|
|
"loss": 4.0617,
|
|
"mean_token_accuracy": 0.27845894992351533,
|
|
"num_tokens": 16683554.0,
|
|
"step": 9030
|
|
},
|
|
{
|
|
"entropy": 4.872682380676269,
|
|
"epoch": 7.560669456066946,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00012425382302745167,
|
|
"loss": 4.166,
|
|
"mean_token_accuracy": 0.2649672508239746,
|
|
"num_tokens": 16692081.0,
|
|
"step": 9035
|
|
},
|
|
{
|
|
"entropy": 4.882773113250733,
|
|
"epoch": 7.564853556485356,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00012401436407004444,
|
|
"loss": 4.204,
|
|
"mean_token_accuracy": 0.2655919104814529,
|
|
"num_tokens": 16701543.0,
|
|
"step": 9040
|
|
},
|
|
{
|
|
"entropy": 4.82576355934143,
|
|
"epoch": 7.569037656903766,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012377521581680146,
|
|
"loss": 4.1372,
|
|
"mean_token_accuracy": 0.2711882680654526,
|
|
"num_tokens": 16710044.0,
|
|
"step": 9045
|
|
},
|
|
{
|
|
"entropy": 4.839276742935181,
|
|
"epoch": 7.573221757322176,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000123536378759851,
|
|
"loss": 4.1309,
|
|
"mean_token_accuracy": 0.2781183421611786,
|
|
"num_tokens": 16718765.0,
|
|
"step": 9050
|
|
},
|
|
{
|
|
"entropy": 4.777922821044922,
|
|
"epoch": 7.577405857740586,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012329785339068163,
|
|
"loss": 4.0425,
|
|
"mean_token_accuracy": 0.27285882979631426,
|
|
"num_tokens": 16727470.0,
|
|
"step": 9055
|
|
},
|
|
{
|
|
"entropy": 4.893827819824219,
|
|
"epoch": 7.581589958158996,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001230596402001401,
|
|
"loss": 4.1867,
|
|
"mean_token_accuracy": 0.26303688287734983,
|
|
"num_tokens": 16736557.0,
|
|
"step": 9060
|
|
},
|
|
{
|
|
"entropy": 4.840340852737427,
|
|
"epoch": 7.585774058577406,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00012282173967843085,
|
|
"loss": 4.1207,
|
|
"mean_token_accuracy": 0.2702972024679184,
|
|
"num_tokens": 16745601.0,
|
|
"step": 9065
|
|
},
|
|
{
|
|
"entropy": 4.858548593521118,
|
|
"epoch": 7.589958158995816,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00012258415231511497,
|
|
"loss": 4.17,
|
|
"mean_token_accuracy": 0.266457599401474,
|
|
"num_tokens": 16756141.0,
|
|
"step": 9070
|
|
},
|
|
{
|
|
"entropy": 4.797425746917725,
|
|
"epoch": 7.594142259414226,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00012234687859910895,
|
|
"loss": 4.1639,
|
|
"mean_token_accuracy": 0.2731698274612427,
|
|
"num_tokens": 16766080.0,
|
|
"step": 9075
|
|
},
|
|
{
|
|
"entropy": 4.82517671585083,
|
|
"epoch": 7.598326359832636,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001221099190186842,
|
|
"loss": 4.122,
|
|
"mean_token_accuracy": 0.26734019815921783,
|
|
"num_tokens": 16775826.0,
|
|
"step": 9080
|
|
},
|
|
{
|
|
"entropy": 4.790913772583008,
|
|
"epoch": 7.602510460251046,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001218732740614651,
|
|
"loss": 4.1498,
|
|
"mean_token_accuracy": 0.2691832110285759,
|
|
"num_tokens": 16785139.0,
|
|
"step": 9085
|
|
},
|
|
{
|
|
"entropy": 4.798343992233276,
|
|
"epoch": 7.606694560669456,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00012163694421442905,
|
|
"loss": 4.1211,
|
|
"mean_token_accuracy": 0.2758825957775116,
|
|
"num_tokens": 16793935.0,
|
|
"step": 9090
|
|
},
|
|
{
|
|
"entropy": 4.827640676498413,
|
|
"epoch": 7.610878661087866,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00012140092996390502,
|
|
"loss": 4.0819,
|
|
"mean_token_accuracy": 0.27120884954929353,
|
|
"num_tokens": 16803451.0,
|
|
"step": 9095
|
|
},
|
|
{
|
|
"entropy": 4.897553491592407,
|
|
"epoch": 7.615062761506276,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00012116523179557221,
|
|
"loss": 4.1037,
|
|
"mean_token_accuracy": 0.2760483235120773,
|
|
"num_tokens": 16813229.0,
|
|
"step": 9100
|
|
},
|
|
{
|
|
"entropy": 4.817846727371216,
|
|
"epoch": 7.619246861924686,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00012092985019445968,
|
|
"loss": 4.0858,
|
|
"mean_token_accuracy": 0.2741702914237976,
|
|
"num_tokens": 16822088.0,
|
|
"step": 9105
|
|
},
|
|
{
|
|
"entropy": 4.8244061946868895,
|
|
"epoch": 7.623430962343097,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.0001206947856449448,
|
|
"loss": 4.1302,
|
|
"mean_token_accuracy": 0.27048187553882597,
|
|
"num_tokens": 16831172.0,
|
|
"step": 9110
|
|
},
|
|
{
|
|
"entropy": 4.854541873931884,
|
|
"epoch": 7.627615062761507,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00012046003863075285,
|
|
"loss": 4.1842,
|
|
"mean_token_accuracy": 0.2627058506011963,
|
|
"num_tokens": 16840482.0,
|
|
"step": 9115
|
|
},
|
|
{
|
|
"entropy": 4.8424522399902346,
|
|
"epoch": 7.631799163179917,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00012022560963495516,
|
|
"loss": 4.1383,
|
|
"mean_token_accuracy": 0.2578867718577385,
|
|
"num_tokens": 16849205.0,
|
|
"step": 9120
|
|
},
|
|
{
|
|
"entropy": 4.844172811508178,
|
|
"epoch": 7.635983263598327,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00011999149913996902,
|
|
"loss": 4.1548,
|
|
"mean_token_accuracy": 0.27086285799741744,
|
|
"num_tokens": 16857500.0,
|
|
"step": 9125
|
|
},
|
|
{
|
|
"entropy": 4.77755913734436,
|
|
"epoch": 7.640167364016737,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00011975770762755627,
|
|
"loss": 4.0377,
|
|
"mean_token_accuracy": 0.2824198707938194,
|
|
"num_tokens": 16866329.0,
|
|
"step": 9130
|
|
},
|
|
{
|
|
"entropy": 4.835915899276733,
|
|
"epoch": 7.644351464435147,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001195242355788221,
|
|
"loss": 4.1379,
|
|
"mean_token_accuracy": 0.2698509320616722,
|
|
"num_tokens": 16875408.0,
|
|
"step": 9135
|
|
},
|
|
{
|
|
"entropy": 4.868906450271607,
|
|
"epoch": 7.648535564853557,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011929108347421468,
|
|
"loss": 4.1676,
|
|
"mean_token_accuracy": 0.25796899795532224,
|
|
"num_tokens": 16885272.0,
|
|
"step": 9140
|
|
},
|
|
{
|
|
"entropy": 4.804734468460083,
|
|
"epoch": 7.652719665271967,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00011905825179352337,
|
|
"loss": 4.121,
|
|
"mean_token_accuracy": 0.2679247334599495,
|
|
"num_tokens": 16894375.0,
|
|
"step": 9145
|
|
},
|
|
{
|
|
"entropy": 4.7810886859893795,
|
|
"epoch": 7.656903765690377,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00011882574101587834,
|
|
"loss": 4.0808,
|
|
"mean_token_accuracy": 0.2704143926501274,
|
|
"num_tokens": 16903790.0,
|
|
"step": 9150
|
|
},
|
|
{
|
|
"entropy": 4.814520025253296,
|
|
"epoch": 7.661087866108787,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011859355161974948,
|
|
"loss": 4.1079,
|
|
"mean_token_accuracy": 0.26891871094703673,
|
|
"num_tokens": 16913453.0,
|
|
"step": 9155
|
|
},
|
|
{
|
|
"entropy": 4.769123935699463,
|
|
"epoch": 7.665271966527197,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00011836168408294513,
|
|
"loss": 4.1038,
|
|
"mean_token_accuracy": 0.27280661314725874,
|
|
"num_tokens": 16922392.0,
|
|
"step": 9160
|
|
},
|
|
{
|
|
"entropy": 4.839308357238769,
|
|
"epoch": 7.669456066945607,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011813013888261154,
|
|
"loss": 4.1977,
|
|
"mean_token_accuracy": 0.25938019454479216,
|
|
"num_tokens": 16931945.0,
|
|
"step": 9165
|
|
},
|
|
{
|
|
"entropy": 4.824350261688233,
|
|
"epoch": 7.673640167364017,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011789891649523137,
|
|
"loss": 4.1234,
|
|
"mean_token_accuracy": 0.26331140100955963,
|
|
"num_tokens": 16941285.0,
|
|
"step": 9170
|
|
},
|
|
{
|
|
"entropy": 4.852743005752563,
|
|
"epoch": 7.677824267782427,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001176680173966232,
|
|
"loss": 4.146,
|
|
"mean_token_accuracy": 0.2628300815820694,
|
|
"num_tokens": 16950155.0,
|
|
"step": 9175
|
|
},
|
|
{
|
|
"entropy": 4.857287073135376,
|
|
"epoch": 7.682008368200837,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011743744206194028,
|
|
"loss": 4.2494,
|
|
"mean_token_accuracy": 0.26586121022701265,
|
|
"num_tokens": 16959994.0,
|
|
"step": 9180
|
|
},
|
|
{
|
|
"entropy": 4.837646532058716,
|
|
"epoch": 7.686192468619247,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00011720719096566946,
|
|
"loss": 4.1296,
|
|
"mean_token_accuracy": 0.2697991028428078,
|
|
"num_tokens": 16969095.0,
|
|
"step": 9185
|
|
},
|
|
{
|
|
"entropy": 4.870945024490356,
|
|
"epoch": 7.690376569037657,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00011697726458163056,
|
|
"loss": 4.2019,
|
|
"mean_token_accuracy": 0.2576313838362694,
|
|
"num_tokens": 16978574.0,
|
|
"step": 9190
|
|
},
|
|
{
|
|
"entropy": 4.852782106399536,
|
|
"epoch": 7.694560669456067,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011674766338297508,
|
|
"loss": 4.1971,
|
|
"mean_token_accuracy": 0.2700987532734871,
|
|
"num_tokens": 16988165.0,
|
|
"step": 9195
|
|
},
|
|
{
|
|
"entropy": 4.7451318264007565,
|
|
"epoch": 7.698744769874477,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00011651838784218545,
|
|
"loss": 4.0991,
|
|
"mean_token_accuracy": 0.2801639050245285,
|
|
"num_tokens": 16998419.0,
|
|
"step": 9200
|
|
},
|
|
{
|
|
"entropy": 4.795413208007813,
|
|
"epoch": 7.702928870292887,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00011628943843107372,
|
|
"loss": 4.134,
|
|
"mean_token_accuracy": 0.26194210052490235,
|
|
"num_tokens": 17008276.0,
|
|
"step": 9205
|
|
},
|
|
{
|
|
"entropy": 4.835207319259643,
|
|
"epoch": 7.707112970711297,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00011606081562078113,
|
|
"loss": 4.1706,
|
|
"mean_token_accuracy": 0.2654357239603996,
|
|
"num_tokens": 17017157.0,
|
|
"step": 9210
|
|
},
|
|
{
|
|
"entropy": 4.829309511184692,
|
|
"epoch": 7.711297071129707,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00011583251988177664,
|
|
"loss": 4.1162,
|
|
"mean_token_accuracy": 0.2760755270719528,
|
|
"num_tokens": 17027100.0,
|
|
"step": 9215
|
|
},
|
|
{
|
|
"entropy": 4.76090440750122,
|
|
"epoch": 7.7154811715481175,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011560455168385601,
|
|
"loss": 4.0872,
|
|
"mean_token_accuracy": 0.2663909733295441,
|
|
"num_tokens": 17036094.0,
|
|
"step": 9220
|
|
},
|
|
{
|
|
"entropy": 4.790591812133789,
|
|
"epoch": 7.7196652719665275,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011537691149614136,
|
|
"loss": 4.0813,
|
|
"mean_token_accuracy": 0.28508140444755553,
|
|
"num_tokens": 17044720.0,
|
|
"step": 9225
|
|
},
|
|
{
|
|
"entropy": 4.802257251739502,
|
|
"epoch": 7.7238493723849375,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011514959978707944,
|
|
"loss": 4.151,
|
|
"mean_token_accuracy": 0.26882929503917696,
|
|
"num_tokens": 17053074.0,
|
|
"step": 9230
|
|
},
|
|
{
|
|
"entropy": 4.855953931808472,
|
|
"epoch": 7.7280334728033475,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0001149226170244413,
|
|
"loss": 4.1707,
|
|
"mean_token_accuracy": 0.26459203511476515,
|
|
"num_tokens": 17063036.0,
|
|
"step": 9235
|
|
},
|
|
{
|
|
"entropy": 4.844877815246582,
|
|
"epoch": 7.7322175732217575,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.0001146959636753209,
|
|
"loss": 4.1266,
|
|
"mean_token_accuracy": 0.27187849283218385,
|
|
"num_tokens": 17072722.0,
|
|
"step": 9240
|
|
},
|
|
{
|
|
"entropy": 4.865857839584351,
|
|
"epoch": 7.7364016736401675,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011446964020613448,
|
|
"loss": 4.1482,
|
|
"mean_token_accuracy": 0.27105664908885957,
|
|
"num_tokens": 17081620.0,
|
|
"step": 9245
|
|
},
|
|
{
|
|
"entropy": 4.745229482650757,
|
|
"epoch": 7.7405857740585775,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00011424364708261931,
|
|
"loss": 4.0337,
|
|
"mean_token_accuracy": 0.2878252357244492,
|
|
"num_tokens": 17090628.0,
|
|
"step": 9250
|
|
},
|
|
{
|
|
"entropy": 4.854003667831421,
|
|
"epoch": 7.7447698744769875,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00011401798476983281,
|
|
"loss": 4.1897,
|
|
"mean_token_accuracy": 0.2654903903603554,
|
|
"num_tokens": 17100089.0,
|
|
"step": 9255
|
|
},
|
|
{
|
|
"entropy": 4.818533086776734,
|
|
"epoch": 7.7489539748953975,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00011379265373215189,
|
|
"loss": 4.1053,
|
|
"mean_token_accuracy": 0.26808108687400817,
|
|
"num_tokens": 17108455.0,
|
|
"step": 9260
|
|
},
|
|
{
|
|
"entropy": 4.835537719726562,
|
|
"epoch": 7.7531380753138075,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011356765443327146,
|
|
"loss": 4.1377,
|
|
"mean_token_accuracy": 0.2707476645708084,
|
|
"num_tokens": 17117803.0,
|
|
"step": 9265
|
|
},
|
|
{
|
|
"entropy": 4.819061660766602,
|
|
"epoch": 7.7573221757322175,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 0.00011334298733620405,
|
|
"loss": 4.1044,
|
|
"mean_token_accuracy": 0.2797745794057846,
|
|
"num_tokens": 17126436.0,
|
|
"step": 9270
|
|
},
|
|
{
|
|
"entropy": 4.859604120254517,
|
|
"epoch": 7.7615062761506275,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00011311865290327827,
|
|
"loss": 4.2206,
|
|
"mean_token_accuracy": 0.25634242296218873,
|
|
"num_tokens": 17135467.0,
|
|
"step": 9275
|
|
},
|
|
{
|
|
"entropy": 4.852531480789184,
|
|
"epoch": 7.7656903765690375,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 0.00011289465159613849,
|
|
"loss": 4.1784,
|
|
"mean_token_accuracy": 0.2683362811803818,
|
|
"num_tokens": 17145282.0,
|
|
"step": 9280
|
|
},
|
|
{
|
|
"entropy": 4.843582057952881,
|
|
"epoch": 7.7698744769874475,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00011267098387574328,
|
|
"loss": 4.205,
|
|
"mean_token_accuracy": 0.26879130154848097,
|
|
"num_tokens": 17155003.0,
|
|
"step": 9285
|
|
},
|
|
{
|
|
"entropy": 4.832408380508423,
|
|
"epoch": 7.7740585774058575,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011244765020236477,
|
|
"loss": 4.1185,
|
|
"mean_token_accuracy": 0.269975021481514,
|
|
"num_tokens": 17163784.0,
|
|
"step": 9290
|
|
},
|
|
{
|
|
"entropy": 4.793617677688599,
|
|
"epoch": 7.7782426778242675,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00011222465103558794,
|
|
"loss": 4.127,
|
|
"mean_token_accuracy": 0.2720103919506073,
|
|
"num_tokens": 17172991.0,
|
|
"step": 9295
|
|
},
|
|
{
|
|
"entropy": 4.810043382644653,
|
|
"epoch": 7.7824267782426775,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.00011200198683430904,
|
|
"loss": 4.1071,
|
|
"mean_token_accuracy": 0.2708059087395668,
|
|
"num_tokens": 17182035.0,
|
|
"step": 9300
|
|
},
|
|
{
|
|
"entropy": 4.780710458755493,
|
|
"epoch": 7.786610878661088,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00011177965805673528,
|
|
"loss": 4.0847,
|
|
"mean_token_accuracy": 0.27438362389802934,
|
|
"num_tokens": 17191257.0,
|
|
"step": 9305
|
|
},
|
|
{
|
|
"entropy": 4.84466700553894,
|
|
"epoch": 7.790794979079498,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00011155766516038363,
|
|
"loss": 4.1087,
|
|
"mean_token_accuracy": 0.2684356600046158,
|
|
"num_tokens": 17200792.0,
|
|
"step": 9310
|
|
},
|
|
{
|
|
"entropy": 4.790325355529785,
|
|
"epoch": 7.794979079497908,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.00011133600860207965,
|
|
"loss": 4.1522,
|
|
"mean_token_accuracy": 0.2578818708658218,
|
|
"num_tokens": 17209833.0,
|
|
"step": 9315
|
|
},
|
|
{
|
|
"entropy": 4.809247636795044,
|
|
"epoch": 7.799163179916318,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00011111468883795698,
|
|
"loss": 4.1066,
|
|
"mean_token_accuracy": 0.26896059364080427,
|
|
"num_tokens": 17218982.0,
|
|
"step": 9320
|
|
},
|
|
{
|
|
"entropy": 4.856968593597412,
|
|
"epoch": 7.803347280334728,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00011089370632345599,
|
|
"loss": 4.1664,
|
|
"mean_token_accuracy": 0.2646897003054619,
|
|
"num_tokens": 17228557.0,
|
|
"step": 9325
|
|
},
|
|
{
|
|
"entropy": 4.835558843612671,
|
|
"epoch": 7.807531380753138,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00011067306151332331,
|
|
"loss": 4.1389,
|
|
"mean_token_accuracy": 0.27084655612707137,
|
|
"num_tokens": 17238081.0,
|
|
"step": 9330
|
|
},
|
|
{
|
|
"entropy": 4.755098295211792,
|
|
"epoch": 7.811715481171548,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00011045275486161039,
|
|
"loss": 4.1174,
|
|
"mean_token_accuracy": 0.2707542613148689,
|
|
"num_tokens": 17249000.0,
|
|
"step": 9335
|
|
},
|
|
{
|
|
"entropy": 4.811800289154053,
|
|
"epoch": 7.815899581589958,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001102327868216729,
|
|
"loss": 4.1646,
|
|
"mean_token_accuracy": 0.26927409768104554,
|
|
"num_tokens": 17259374.0,
|
|
"step": 9340
|
|
},
|
|
{
|
|
"entropy": 4.766207075119018,
|
|
"epoch": 7.820083682008368,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011001315784616988,
|
|
"loss": 4.103,
|
|
"mean_token_accuracy": 0.2754746347665787,
|
|
"num_tokens": 17268334.0,
|
|
"step": 9345
|
|
},
|
|
{
|
|
"entropy": 4.827593040466309,
|
|
"epoch": 7.824267782426778,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0001097938683870622,
|
|
"loss": 4.1059,
|
|
"mean_token_accuracy": 0.2708084523677826,
|
|
"num_tokens": 17277364.0,
|
|
"step": 9350
|
|
},
|
|
{
|
|
"entropy": 4.817640066146851,
|
|
"epoch": 7.828451882845188,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010957491889561244,
|
|
"loss": 4.1043,
|
|
"mean_token_accuracy": 0.2635605946183205,
|
|
"num_tokens": 17286552.0,
|
|
"step": 9355
|
|
},
|
|
{
|
|
"entropy": 4.827054166793824,
|
|
"epoch": 7.832635983263598,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001093563098223834,
|
|
"loss": 4.1482,
|
|
"mean_token_accuracy": 0.2710980400443077,
|
|
"num_tokens": 17297114.0,
|
|
"step": 9360
|
|
},
|
|
{
|
|
"entropy": 4.845240783691406,
|
|
"epoch": 7.836820083682008,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010913804161723742,
|
|
"loss": 4.162,
|
|
"mean_token_accuracy": 0.267504420876503,
|
|
"num_tokens": 17306469.0,
|
|
"step": 9365
|
|
},
|
|
{
|
|
"entropy": 4.856093454360962,
|
|
"epoch": 7.841004184100418,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00010892011472933547,
|
|
"loss": 4.17,
|
|
"mean_token_accuracy": 0.2755287751555443,
|
|
"num_tokens": 17315790.0,
|
|
"step": 9370
|
|
},
|
|
{
|
|
"entropy": 4.837453889846802,
|
|
"epoch": 7.845188284518828,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00010870252960713587,
|
|
"loss": 4.2157,
|
|
"mean_token_accuracy": 0.2560058355331421,
|
|
"num_tokens": 17324803.0,
|
|
"step": 9375
|
|
},
|
|
{
|
|
"entropy": 4.835926580429077,
|
|
"epoch": 7.849372384937238,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00010848528669839409,
|
|
"loss": 4.1911,
|
|
"mean_token_accuracy": 0.2658363178372383,
|
|
"num_tokens": 17333856.0,
|
|
"step": 9380
|
|
},
|
|
{
|
|
"entropy": 4.791576242446899,
|
|
"epoch": 7.853556485355648,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010826838645016077,
|
|
"loss": 4.0492,
|
|
"mean_token_accuracy": 0.2775577142834663,
|
|
"num_tokens": 17343399.0,
|
|
"step": 9385
|
|
},
|
|
{
|
|
"entropy": 4.805856657028198,
|
|
"epoch": 7.857740585774058,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.00010805182930878191,
|
|
"loss": 4.0476,
|
|
"mean_token_accuracy": 0.274207229912281,
|
|
"num_tokens": 17352547.0,
|
|
"step": 9390
|
|
},
|
|
{
|
|
"entropy": 4.84045991897583,
|
|
"epoch": 7.861924686192468,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010783561571989739,
|
|
"loss": 4.1583,
|
|
"mean_token_accuracy": 0.2659938231110573,
|
|
"num_tokens": 17362507.0,
|
|
"step": 9395
|
|
},
|
|
{
|
|
"entropy": 4.782192325592041,
|
|
"epoch": 7.866108786610878,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00010761974612843984,
|
|
"loss": 4.0637,
|
|
"mean_token_accuracy": 0.2749798148870468,
|
|
"num_tokens": 17372133.0,
|
|
"step": 9400
|
|
},
|
|
{
|
|
"entropy": 4.8532952785491945,
|
|
"epoch": 7.870292887029288,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010740422097863431,
|
|
"loss": 4.1925,
|
|
"mean_token_accuracy": 0.25674797147512435,
|
|
"num_tokens": 17383178.0,
|
|
"step": 9405
|
|
},
|
|
{
|
|
"entropy": 4.814047479629517,
|
|
"epoch": 7.874476987447698,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.0001071890407139968,
|
|
"loss": 4.1827,
|
|
"mean_token_accuracy": 0.26482645720243453,
|
|
"num_tokens": 17392357.0,
|
|
"step": 9410
|
|
},
|
|
{
|
|
"entropy": 4.87243390083313,
|
|
"epoch": 7.878661087866108,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010697420577733385,
|
|
"loss": 4.1687,
|
|
"mean_token_accuracy": 0.27126303762197496,
|
|
"num_tokens": 17401101.0,
|
|
"step": 9415
|
|
},
|
|
{
|
|
"entropy": 4.785106945037842,
|
|
"epoch": 7.882845188284519,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010675971661074102,
|
|
"loss": 4.0165,
|
|
"mean_token_accuracy": 0.27872402369976046,
|
|
"num_tokens": 17409909.0,
|
|
"step": 9420
|
|
},
|
|
{
|
|
"entropy": 4.871726608276367,
|
|
"epoch": 7.887029288702929,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010654557365560261,
|
|
"loss": 4.2021,
|
|
"mean_token_accuracy": 0.2644994705915451,
|
|
"num_tokens": 17419631.0,
|
|
"step": 9425
|
|
},
|
|
{
|
|
"entropy": 4.813071823120117,
|
|
"epoch": 7.891213389121339,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 0.00010633177735259047,
|
|
"loss": 4.1579,
|
|
"mean_token_accuracy": 0.26130197644233705,
|
|
"num_tokens": 17428557.0,
|
|
"step": 9430
|
|
},
|
|
{
|
|
"entropy": 4.808069038391113,
|
|
"epoch": 7.895397489539749,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010611832814166292,
|
|
"loss": 4.1405,
|
|
"mean_token_accuracy": 0.2709905907511711,
|
|
"num_tokens": 17438442.0,
|
|
"step": 9435
|
|
},
|
|
{
|
|
"entropy": 4.83789587020874,
|
|
"epoch": 7.899581589958159,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010590522646206419,
|
|
"loss": 4.1547,
|
|
"mean_token_accuracy": 0.2669744551181793,
|
|
"num_tokens": 17448116.0,
|
|
"step": 9440
|
|
},
|
|
{
|
|
"entropy": 4.792590522766114,
|
|
"epoch": 7.903765690376569,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00010569247275232328,
|
|
"loss": 4.0686,
|
|
"mean_token_accuracy": 0.27617967277765276,
|
|
"num_tokens": 17456823.0,
|
|
"step": 9445
|
|
},
|
|
{
|
|
"entropy": 4.842328405380249,
|
|
"epoch": 7.907949790794979,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010548006745025298,
|
|
"loss": 4.1371,
|
|
"mean_token_accuracy": 0.27206799387931824,
|
|
"num_tokens": 17466313.0,
|
|
"step": 9450
|
|
},
|
|
{
|
|
"entropy": 4.846059560775757,
|
|
"epoch": 7.912133891213389,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010526801099294947,
|
|
"loss": 4.2363,
|
|
"mean_token_accuracy": 0.2616952046751976,
|
|
"num_tokens": 17477448.0,
|
|
"step": 9455
|
|
},
|
|
{
|
|
"entropy": 4.893253517150879,
|
|
"epoch": 7.916317991631799,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.00010505630381679063,
|
|
"loss": 4.2092,
|
|
"mean_token_accuracy": 0.26588337272405627,
|
|
"num_tokens": 17487587.0,
|
|
"step": 9460
|
|
},
|
|
{
|
|
"entropy": 4.846000051498413,
|
|
"epoch": 7.920502092050209,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.00010484494635743598,
|
|
"loss": 4.1123,
|
|
"mean_token_accuracy": 0.2735391080379486,
|
|
"num_tokens": 17496282.0,
|
|
"step": 9465
|
|
},
|
|
{
|
|
"entropy": 4.857050275802612,
|
|
"epoch": 7.924686192468619,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000104633939049825,
|
|
"loss": 4.1967,
|
|
"mean_token_accuracy": 0.2582026347517967,
|
|
"num_tokens": 17505468.0,
|
|
"step": 9470
|
|
},
|
|
{
|
|
"entropy": 4.844735908508301,
|
|
"epoch": 7.928870292887029,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00010442328232817702,
|
|
"loss": 4.1254,
|
|
"mean_token_accuracy": 0.2690133213996887,
|
|
"num_tokens": 17515743.0,
|
|
"step": 9475
|
|
},
|
|
{
|
|
"entropy": 4.849387788772583,
|
|
"epoch": 7.933054393305439,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.00010421297662598955,
|
|
"loss": 4.0942,
|
|
"mean_token_accuracy": 0.27485825270414355,
|
|
"num_tokens": 17524350.0,
|
|
"step": 9480
|
|
},
|
|
{
|
|
"entropy": 4.8091527938842775,
|
|
"epoch": 7.937238493723849,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00010400302237603793,
|
|
"loss": 4.0855,
|
|
"mean_token_accuracy": 0.26840528696775434,
|
|
"num_tokens": 17533439.0,
|
|
"step": 9485
|
|
},
|
|
{
|
|
"entropy": 4.83451828956604,
|
|
"epoch": 7.941422594142259,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00010379342001037431,
|
|
"loss": 4.1179,
|
|
"mean_token_accuracy": 0.2664906978607178,
|
|
"num_tokens": 17542782.0,
|
|
"step": 9490
|
|
},
|
|
{
|
|
"entropy": 4.834074306488037,
|
|
"epoch": 7.945606694560669,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00010358416996032658,
|
|
"loss": 4.1356,
|
|
"mean_token_accuracy": 0.27035591900348666,
|
|
"num_tokens": 17552495.0,
|
|
"step": 9495
|
|
},
|
|
{
|
|
"entropy": 4.820766878128052,
|
|
"epoch": 7.949790794979079,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00010337527265649782,
|
|
"loss": 4.1329,
|
|
"mean_token_accuracy": 0.265675987303257,
|
|
"num_tokens": 17560933.0,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"epoch": 7.949790794979079,
|
|
"eval_entropy": 4.695549250136612,
|
|
"eval_loss": 5.3067402839660645,
|
|
"eval_mean_token_accuracy": 0.20541439138368106,
|
|
"eval_num_tokens": 17560933.0,
|
|
"eval_runtime": 2.7439,
|
|
"eval_samples_per_second": 1423.883,
|
|
"eval_steps_per_second": 178.213,
|
|
"step": 9500
|
|
},
|
|
{
|
|
"entropy": 4.8355261325836185,
|
|
"epoch": 7.95397489539749,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00010316672852876494,
|
|
"loss": 4.1853,
|
|
"mean_token_accuracy": 0.26423315703868866,
|
|
"num_tokens": 17571077.0,
|
|
"step": 9505
|
|
},
|
|
{
|
|
"entropy": 4.8459460735321045,
|
|
"epoch": 7.9581589958159,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010295853800627836,
|
|
"loss": 4.1226,
|
|
"mean_token_accuracy": 0.268256901204586,
|
|
"num_tokens": 17580128.0,
|
|
"step": 9510
|
|
},
|
|
{
|
|
"entropy": 4.8846320629119875,
|
|
"epoch": 7.96234309623431,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.00010275070151746069,
|
|
"loss": 4.1755,
|
|
"mean_token_accuracy": 0.2666592955589294,
|
|
"num_tokens": 17589811.0,
|
|
"step": 9515
|
|
},
|
|
{
|
|
"entropy": 4.855812931060791,
|
|
"epoch": 7.96652719665272,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010254321949000587,
|
|
"loss": 4.1611,
|
|
"mean_token_accuracy": 0.26461099088191986,
|
|
"num_tokens": 17599018.0,
|
|
"step": 9520
|
|
},
|
|
{
|
|
"entropy": 4.810562515258789,
|
|
"epoch": 7.97071129707113,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010233609235087874,
|
|
"loss": 4.082,
|
|
"mean_token_accuracy": 0.2644841566681862,
|
|
"num_tokens": 17607416.0,
|
|
"step": 9525
|
|
},
|
|
{
|
|
"entropy": 4.822035646438598,
|
|
"epoch": 7.97489539748954,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00010212932052631351,
|
|
"loss": 4.1749,
|
|
"mean_token_accuracy": 0.2686312973499298,
|
|
"num_tokens": 17616674.0,
|
|
"step": 9530
|
|
},
|
|
{
|
|
"entropy": 4.824155473709107,
|
|
"epoch": 7.97907949790795,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010192290444181347,
|
|
"loss": 4.1302,
|
|
"mean_token_accuracy": 0.2692561775445938,
|
|
"num_tokens": 17626099.0,
|
|
"step": 9535
|
|
},
|
|
{
|
|
"entropy": 4.795995998382568,
|
|
"epoch": 7.98326359832636,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00010171684452214959,
|
|
"loss": 4.1172,
|
|
"mean_token_accuracy": 0.2651543542742729,
|
|
"num_tokens": 17636013.0,
|
|
"step": 9540
|
|
},
|
|
{
|
|
"entropy": 4.87143759727478,
|
|
"epoch": 7.98744769874477,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.00010151114119136024,
|
|
"loss": 4.1449,
|
|
"mean_token_accuracy": 0.2678538501262665,
|
|
"num_tokens": 17645849.0,
|
|
"step": 9545
|
|
},
|
|
{
|
|
"entropy": 4.8627862453460695,
|
|
"epoch": 7.99163179916318,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010130579487274969,
|
|
"loss": 4.1992,
|
|
"mean_token_accuracy": 0.26687067598104475,
|
|
"num_tokens": 17655557.0,
|
|
"step": 9550
|
|
},
|
|
{
|
|
"entropy": 4.840360498428344,
|
|
"epoch": 7.99581589958159,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.0001011008059888876,
|
|
"loss": 4.2061,
|
|
"mean_token_accuracy": 0.2634638145565987,
|
|
"num_tokens": 17665717.0,
|
|
"step": 9555
|
|
},
|
|
{
|
|
"entropy": 4.766755867004394,
|
|
"epoch": 8.0,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.00010089617496160825,
|
|
"loss": 4.0548,
|
|
"mean_token_accuracy": 0.27672168910503386,
|
|
"num_tokens": 17674576.0,
|
|
"step": 9560
|
|
},
|
|
{
|
|
"entropy": 4.839120483398437,
|
|
"epoch": 8.00418410041841,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00010069190221200929,
|
|
"loss": 4.0171,
|
|
"mean_token_accuracy": 0.285295295715332,
|
|
"num_tokens": 17684147.0,
|
|
"step": 9565
|
|
},
|
|
{
|
|
"entropy": 4.744625186920166,
|
|
"epoch": 8.00836820083682,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00010048798816045132,
|
|
"loss": 3.9613,
|
|
"mean_token_accuracy": 0.27588112056255343,
|
|
"num_tokens": 17693082.0,
|
|
"step": 9570
|
|
},
|
|
{
|
|
"entropy": 4.759832048416138,
|
|
"epoch": 8.01255230125523,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.00010028443322655652,
|
|
"loss": 4.0256,
|
|
"mean_token_accuracy": 0.2788006514310837,
|
|
"num_tokens": 17702149.0,
|
|
"step": 9575
|
|
},
|
|
{
|
|
"entropy": 4.802634286880493,
|
|
"epoch": 8.01673640167364,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00010008123782920836,
|
|
"loss": 4.0126,
|
|
"mean_token_accuracy": 0.28438694179058077,
|
|
"num_tokens": 17710666.0,
|
|
"step": 9580
|
|
},
|
|
{
|
|
"entropy": 4.822308969497681,
|
|
"epoch": 8.02092050209205,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 9.987840238655026e-05,
|
|
"loss": 4.0691,
|
|
"mean_token_accuracy": 0.28215044140815737,
|
|
"num_tokens": 17719643.0,
|
|
"step": 9585
|
|
},
|
|
{
|
|
"entropy": 4.865143489837647,
|
|
"epoch": 8.02510460251046,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.967592731598487e-05,
|
|
"loss": 4.1118,
|
|
"mean_token_accuracy": 0.27615293562412263,
|
|
"num_tokens": 17728714.0,
|
|
"step": 9590
|
|
},
|
|
{
|
|
"entropy": 4.905548238754273,
|
|
"epoch": 8.02928870292887,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.947381303417346e-05,
|
|
"loss": 4.0699,
|
|
"mean_token_accuracy": 0.2787525713443756,
|
|
"num_tokens": 17736496.0,
|
|
"step": 9595
|
|
},
|
|
{
|
|
"entropy": 4.864214181900024,
|
|
"epoch": 8.03347280334728,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.927205995703456e-05,
|
|
"loss": 4.1366,
|
|
"mean_token_accuracy": 0.2699583351612091,
|
|
"num_tokens": 17745909.0,
|
|
"step": 9600
|
|
},
|
|
{
|
|
"entropy": 4.842680358886719,
|
|
"epoch": 8.03765690376569,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 9.907066849974372e-05,
|
|
"loss": 4.0121,
|
|
"mean_token_accuracy": 0.28604514598846437,
|
|
"num_tokens": 17755125.0,
|
|
"step": 9605
|
|
},
|
|
{
|
|
"entropy": 4.819202136993408,
|
|
"epoch": 8.0418410041841,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.886963907673214e-05,
|
|
"loss": 4.0261,
|
|
"mean_token_accuracy": 0.2827119633555412,
|
|
"num_tokens": 17764253.0,
|
|
"step": 9610
|
|
},
|
|
{
|
|
"entropy": 4.789328622817993,
|
|
"epoch": 8.04602510460251,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.866897210168608e-05,
|
|
"loss": 4.0631,
|
|
"mean_token_accuracy": 0.27191328406333926,
|
|
"num_tokens": 17773485.0,
|
|
"step": 9615
|
|
},
|
|
{
|
|
"entropy": 4.854859256744385,
|
|
"epoch": 8.05020920502092,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.846866798754583e-05,
|
|
"loss": 4.0871,
|
|
"mean_token_accuracy": 0.2720204874873161,
|
|
"num_tokens": 17781942.0,
|
|
"step": 9620
|
|
},
|
|
{
|
|
"entropy": 4.841558647155762,
|
|
"epoch": 8.05439330543933,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.826872714650509e-05,
|
|
"loss": 4.0413,
|
|
"mean_token_accuracy": 0.28018817156553266,
|
|
"num_tokens": 17791576.0,
|
|
"step": 9625
|
|
},
|
|
{
|
|
"entropy": 4.863581037521362,
|
|
"epoch": 8.05857740585774,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.806914999000998e-05,
|
|
"loss": 4.0811,
|
|
"mean_token_accuracy": 0.27329387068748473,
|
|
"num_tokens": 17801439.0,
|
|
"step": 9630
|
|
},
|
|
{
|
|
"entropy": 4.877225255966186,
|
|
"epoch": 8.06276150627615,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.786993692875814e-05,
|
|
"loss": 4.1068,
|
|
"mean_token_accuracy": 0.27641271352767943,
|
|
"num_tokens": 17810523.0,
|
|
"step": 9635
|
|
},
|
|
{
|
|
"entropy": 4.786565160751342,
|
|
"epoch": 8.06694560669456,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.767108837269808e-05,
|
|
"loss": 4.007,
|
|
"mean_token_accuracy": 0.2821986317634583,
|
|
"num_tokens": 17819050.0,
|
|
"step": 9640
|
|
},
|
|
{
|
|
"entropy": 4.838639736175537,
|
|
"epoch": 8.07112970711297,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.747260473102829e-05,
|
|
"loss": 4.0886,
|
|
"mean_token_accuracy": 0.2711678594350815,
|
|
"num_tokens": 17828215.0,
|
|
"step": 9645
|
|
},
|
|
{
|
|
"entropy": 4.778063821792602,
|
|
"epoch": 8.07531380753138,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.727448641219591e-05,
|
|
"loss": 4.0642,
|
|
"mean_token_accuracy": 0.2768439590930939,
|
|
"num_tokens": 17837091.0,
|
|
"step": 9650
|
|
},
|
|
{
|
|
"entropy": 4.789417839050293,
|
|
"epoch": 8.07949790794979,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.707673382389678e-05,
|
|
"loss": 4.0808,
|
|
"mean_token_accuracy": 0.2835092097520828,
|
|
"num_tokens": 17845647.0,
|
|
"step": 9655
|
|
},
|
|
{
|
|
"entropy": 4.856493759155273,
|
|
"epoch": 8.0836820083682,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.687934737307386e-05,
|
|
"loss": 4.0418,
|
|
"mean_token_accuracy": 0.2817514419555664,
|
|
"num_tokens": 17854096.0,
|
|
"step": 9660
|
|
},
|
|
{
|
|
"entropy": 4.843071317672729,
|
|
"epoch": 8.087866108786612,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 9.668232746591671e-05,
|
|
"loss": 4.0322,
|
|
"mean_token_accuracy": 0.2812614440917969,
|
|
"num_tokens": 17863239.0,
|
|
"step": 9665
|
|
},
|
|
{
|
|
"entropy": 4.8168000221252445,
|
|
"epoch": 8.092050209205022,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.648567450786077e-05,
|
|
"loss": 4.0319,
|
|
"mean_token_accuracy": 0.27752512097358706,
|
|
"num_tokens": 17872447.0,
|
|
"step": 9670
|
|
},
|
|
{
|
|
"entropy": 4.778050470352173,
|
|
"epoch": 8.096234309623432,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 9.628938890358604e-05,
|
|
"loss": 4.0104,
|
|
"mean_token_accuracy": 0.28107163310050964,
|
|
"num_tokens": 17881862.0,
|
|
"step": 9675
|
|
},
|
|
{
|
|
"entropy": 4.76653642654419,
|
|
"epoch": 8.100418410041842,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.609347105701691e-05,
|
|
"loss": 4.0569,
|
|
"mean_token_accuracy": 0.27385979145765305,
|
|
"num_tokens": 17891683.0,
|
|
"step": 9680
|
|
},
|
|
{
|
|
"entropy": 4.842598009109497,
|
|
"epoch": 8.104602510460252,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.589792137132061e-05,
|
|
"loss": 4.0811,
|
|
"mean_token_accuracy": 0.2744866073131561,
|
|
"num_tokens": 17901105.0,
|
|
"step": 9685
|
|
},
|
|
{
|
|
"entropy": 4.752630233764648,
|
|
"epoch": 8.108786610878662,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.570274024890704e-05,
|
|
"loss": 4.0316,
|
|
"mean_token_accuracy": 0.28273406326770784,
|
|
"num_tokens": 17910761.0,
|
|
"step": 9690
|
|
},
|
|
{
|
|
"entropy": 4.831161308288574,
|
|
"epoch": 8.112970711297072,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 9.550792809142764e-05,
|
|
"loss": 4.0641,
|
|
"mean_token_accuracy": 0.27813163995742796,
|
|
"num_tokens": 17919917.0,
|
|
"step": 9695
|
|
},
|
|
{
|
|
"entropy": 4.869170999526977,
|
|
"epoch": 8.117154811715482,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.531348529977439e-05,
|
|
"loss": 4.1222,
|
|
"mean_token_accuracy": 0.2731568232178688,
|
|
"num_tokens": 17928588.0,
|
|
"step": 9700
|
|
},
|
|
{
|
|
"entropy": 4.84056715965271,
|
|
"epoch": 8.121338912133892,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 9.51194122740794e-05,
|
|
"loss": 4.0572,
|
|
"mean_token_accuracy": 0.2846394762396812,
|
|
"num_tokens": 17938142.0,
|
|
"step": 9705
|
|
},
|
|
{
|
|
"entropy": 4.794088745117188,
|
|
"epoch": 8.125523012552302,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.492570941371365e-05,
|
|
"loss": 4.001,
|
|
"mean_token_accuracy": 0.27924694418907164,
|
|
"num_tokens": 17947676.0,
|
|
"step": 9710
|
|
},
|
|
{
|
|
"entropy": 4.784554290771484,
|
|
"epoch": 8.129707112970712,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 9.47323771172866e-05,
|
|
"loss": 4.0522,
|
|
"mean_token_accuracy": 0.27874593436717987,
|
|
"num_tokens": 17957166.0,
|
|
"step": 9715
|
|
},
|
|
{
|
|
"entropy": 4.842915439605713,
|
|
"epoch": 8.133891213389122,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.453941578264499e-05,
|
|
"loss": 4.0783,
|
|
"mean_token_accuracy": 0.2734838381409645,
|
|
"num_tokens": 17965592.0,
|
|
"step": 9720
|
|
},
|
|
{
|
|
"entropy": 4.773156070709229,
|
|
"epoch": 8.138075313807532,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 9.434682580687215e-05,
|
|
"loss": 4.0257,
|
|
"mean_token_accuracy": 0.2774977579712868,
|
|
"num_tokens": 17973949.0,
|
|
"step": 9725
|
|
},
|
|
{
|
|
"entropy": 4.767611742019653,
|
|
"epoch": 8.142259414225942,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.415460758628739e-05,
|
|
"loss": 4.0159,
|
|
"mean_token_accuracy": 0.27419193685054777,
|
|
"num_tokens": 17982540.0,
|
|
"step": 9730
|
|
},
|
|
{
|
|
"entropy": 4.794537734985352,
|
|
"epoch": 8.146443514644352,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 9.396276151644478e-05,
|
|
"loss": 4.0375,
|
|
"mean_token_accuracy": 0.27893260419368743,
|
|
"num_tokens": 17991611.0,
|
|
"step": 9735
|
|
},
|
|
{
|
|
"entropy": 4.830135726928711,
|
|
"epoch": 8.150627615062762,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 9.377128799213286e-05,
|
|
"loss": 4.0055,
|
|
"mean_token_accuracy": 0.2844906359910965,
|
|
"num_tokens": 17999462.0,
|
|
"step": 9740
|
|
},
|
|
{
|
|
"entropy": 4.851837825775147,
|
|
"epoch": 8.154811715481172,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.358018740737325e-05,
|
|
"loss": 4.1423,
|
|
"mean_token_accuracy": 0.2806723669171333,
|
|
"num_tokens": 18009693.0,
|
|
"step": 9745
|
|
},
|
|
{
|
|
"entropy": 4.769418001174927,
|
|
"epoch": 8.158995815899582,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.338946015542018e-05,
|
|
"loss": 3.9397,
|
|
"mean_token_accuracy": 0.28969680666923525,
|
|
"num_tokens": 18019101.0,
|
|
"step": 9750
|
|
},
|
|
{
|
|
"entropy": 4.8186970233917235,
|
|
"epoch": 8.163179916317992,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 9.31991066287598e-05,
|
|
"loss": 4.1657,
|
|
"mean_token_accuracy": 0.26260460168123245,
|
|
"num_tokens": 18028198.0,
|
|
"step": 9755
|
|
},
|
|
{
|
|
"entropy": 4.877345371246338,
|
|
"epoch": 8.167364016736402,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 9.300912721910894e-05,
|
|
"loss": 4.1137,
|
|
"mean_token_accuracy": 0.2747832417488098,
|
|
"num_tokens": 18037592.0,
|
|
"step": 9760
|
|
},
|
|
{
|
|
"entropy": 4.845637035369873,
|
|
"epoch": 8.171548117154812,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.28195223174148e-05,
|
|
"loss": 4.0912,
|
|
"mean_token_accuracy": 0.26816652417182923,
|
|
"num_tokens": 18046988.0,
|
|
"step": 9765
|
|
},
|
|
{
|
|
"entropy": 4.74591326713562,
|
|
"epoch": 8.175732217573222,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 9.263029231385366e-05,
|
|
"loss": 3.9876,
|
|
"mean_token_accuracy": 0.29165747165679934,
|
|
"num_tokens": 18056322.0,
|
|
"step": 9770
|
|
},
|
|
{
|
|
"entropy": 4.756015396118164,
|
|
"epoch": 8.179916317991632,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 9.244143759783058e-05,
|
|
"loss": 4.0124,
|
|
"mean_token_accuracy": 0.2830548495054245,
|
|
"num_tokens": 18065157.0,
|
|
"step": 9775
|
|
},
|
|
{
|
|
"entropy": 4.878354263305664,
|
|
"epoch": 8.184100418410042,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 9.225295855797812e-05,
|
|
"loss": 4.1287,
|
|
"mean_token_accuracy": 0.27897276282310485,
|
|
"num_tokens": 18074389.0,
|
|
"step": 9780
|
|
},
|
|
{
|
|
"entropy": 4.861226987838745,
|
|
"epoch": 8.188284518828452,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 9.20648555821558e-05,
|
|
"loss": 4.0365,
|
|
"mean_token_accuracy": 0.2846009433269501,
|
|
"num_tokens": 18083434.0,
|
|
"step": 9785
|
|
},
|
|
{
|
|
"entropy": 4.7950582027435305,
|
|
"epoch": 8.192468619246862,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 9.18771290574494e-05,
|
|
"loss": 4.0193,
|
|
"mean_token_accuracy": 0.27717192769050597,
|
|
"num_tokens": 18094145.0,
|
|
"step": 9790
|
|
},
|
|
{
|
|
"entropy": 4.824673843383789,
|
|
"epoch": 8.196652719665272,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 9.168977937016981e-05,
|
|
"loss": 4.0845,
|
|
"mean_token_accuracy": 0.2820479109883308,
|
|
"num_tokens": 18103707.0,
|
|
"step": 9795
|
|
},
|
|
{
|
|
"entropy": 4.782747554779053,
|
|
"epoch": 8.200836820083682,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 9.150280690585271e-05,
|
|
"loss": 4.0563,
|
|
"mean_token_accuracy": 0.2760318383574486,
|
|
"num_tokens": 18113416.0,
|
|
"step": 9800
|
|
},
|
|
{
|
|
"entropy": 4.7660027027130125,
|
|
"epoch": 8.205020920502092,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 9.131621204925724e-05,
|
|
"loss": 4.0045,
|
|
"mean_token_accuracy": 0.27658907473087313,
|
|
"num_tokens": 18122809.0,
|
|
"step": 9805
|
|
},
|
|
{
|
|
"entropy": 4.863335084915161,
|
|
"epoch": 8.209205020920502,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 9.112999518436574e-05,
|
|
"loss": 4.1284,
|
|
"mean_token_accuracy": 0.2629508450627327,
|
|
"num_tokens": 18132518.0,
|
|
"step": 9810
|
|
},
|
|
{
|
|
"entropy": 4.838024663925171,
|
|
"epoch": 8.213389121338912,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 9.09441566943825e-05,
|
|
"loss": 4.0953,
|
|
"mean_token_accuracy": 0.2701710909605026,
|
|
"num_tokens": 18141499.0,
|
|
"step": 9815
|
|
},
|
|
{
|
|
"entropy": 4.8498186588287355,
|
|
"epoch": 8.217573221757322,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 9.075869696173321e-05,
|
|
"loss": 4.0634,
|
|
"mean_token_accuracy": 0.27597665786743164,
|
|
"num_tokens": 18150330.0,
|
|
"step": 9820
|
|
},
|
|
{
|
|
"entropy": 4.821561813354492,
|
|
"epoch": 8.221757322175732,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 9.057361636806428e-05,
|
|
"loss": 4.0459,
|
|
"mean_token_accuracy": 0.2782487630844116,
|
|
"num_tokens": 18158925.0,
|
|
"step": 9825
|
|
},
|
|
{
|
|
"entropy": 4.80013337135315,
|
|
"epoch": 8.225941422594142,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 9.038891529424173e-05,
|
|
"loss": 4.0242,
|
|
"mean_token_accuracy": 0.2828985512256622,
|
|
"num_tokens": 18167612.0,
|
|
"step": 9830
|
|
},
|
|
{
|
|
"entropy": 4.845247840881347,
|
|
"epoch": 8.230125523012552,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 9.020459412035075e-05,
|
|
"loss": 4.0458,
|
|
"mean_token_accuracy": 0.2800912976264954,
|
|
"num_tokens": 18176727.0,
|
|
"step": 9835
|
|
},
|
|
{
|
|
"entropy": 4.843050289154053,
|
|
"epoch": 8.234309623430962,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 9.002065322569464e-05,
|
|
"loss": 4.053,
|
|
"mean_token_accuracy": 0.2799303039908409,
|
|
"num_tokens": 18186542.0,
|
|
"step": 9840
|
|
},
|
|
{
|
|
"entropy": 4.809372901916504,
|
|
"epoch": 8.238493723849372,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.983709298879424e-05,
|
|
"loss": 4.0388,
|
|
"mean_token_accuracy": 0.2774491563439369,
|
|
"num_tokens": 18195514.0,
|
|
"step": 9845
|
|
},
|
|
{
|
|
"entropy": 4.7864138126373295,
|
|
"epoch": 8.242677824267782,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.965391378738692e-05,
|
|
"loss": 4.0243,
|
|
"mean_token_accuracy": 0.27670903205871583,
|
|
"num_tokens": 18204481.0,
|
|
"step": 9850
|
|
},
|
|
{
|
|
"entropy": 4.8761755466461185,
|
|
"epoch": 8.246861924686192,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.947111599842606e-05,
|
|
"loss": 4.0486,
|
|
"mean_token_accuracy": 0.2775596708059311,
|
|
"num_tokens": 18213342.0,
|
|
"step": 9855
|
|
},
|
|
{
|
|
"entropy": 4.860723495483398,
|
|
"epoch": 8.251046025104603,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 8.928869999808017e-05,
|
|
"loss": 4.1076,
|
|
"mean_token_accuracy": 0.27648585438728335,
|
|
"num_tokens": 18223083.0,
|
|
"step": 9860
|
|
},
|
|
{
|
|
"entropy": 4.777573966979981,
|
|
"epoch": 8.255230125523013,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.910666616173196e-05,
|
|
"loss": 4.0226,
|
|
"mean_token_accuracy": 0.2822031408548355,
|
|
"num_tokens": 18231792.0,
|
|
"step": 9865
|
|
},
|
|
{
|
|
"entropy": 4.798950099945069,
|
|
"epoch": 8.259414225941423,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.89250148639779e-05,
|
|
"loss": 4.0561,
|
|
"mean_token_accuracy": 0.28065574169158936,
|
|
"num_tokens": 18240099.0,
|
|
"step": 9870
|
|
},
|
|
{
|
|
"entropy": 4.782928419113159,
|
|
"epoch": 8.263598326359833,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.874374647862704e-05,
|
|
"loss": 4.0758,
|
|
"mean_token_accuracy": 0.2649018943309784,
|
|
"num_tokens": 18249237.0,
|
|
"step": 9875
|
|
},
|
|
{
|
|
"entropy": 4.836393022537232,
|
|
"epoch": 8.267782426778243,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 8.856286137870068e-05,
|
|
"loss": 4.0847,
|
|
"mean_token_accuracy": 0.2737008959054947,
|
|
"num_tokens": 18257896.0,
|
|
"step": 9880
|
|
},
|
|
{
|
|
"entropy": 4.819747543334961,
|
|
"epoch": 8.271966527196653,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 8.838235993643125e-05,
|
|
"loss": 4.0168,
|
|
"mean_token_accuracy": 0.2835097461938858,
|
|
"num_tokens": 18266251.0,
|
|
"step": 9885
|
|
},
|
|
{
|
|
"entropy": 4.810356950759887,
|
|
"epoch": 8.276150627615063,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 8.820224252326161e-05,
|
|
"loss": 4.0338,
|
|
"mean_token_accuracy": 0.28094419538974763,
|
|
"num_tokens": 18275729.0,
|
|
"step": 9890
|
|
},
|
|
{
|
|
"entropy": 4.87390365600586,
|
|
"epoch": 8.280334728033473,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.802250950984461e-05,
|
|
"loss": 4.1108,
|
|
"mean_token_accuracy": 0.2735268399119377,
|
|
"num_tokens": 18284625.0,
|
|
"step": 9895
|
|
},
|
|
{
|
|
"entropy": 4.7600573062896725,
|
|
"epoch": 8.284518828451883,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.784316126604169e-05,
|
|
"loss": 3.9913,
|
|
"mean_token_accuracy": 0.2869440346956253,
|
|
"num_tokens": 18293390.0,
|
|
"step": 9900
|
|
},
|
|
{
|
|
"entropy": 4.767490816116333,
|
|
"epoch": 8.288702928870293,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 8.766419816092284e-05,
|
|
"loss": 4.0052,
|
|
"mean_token_accuracy": 0.2851350575685501,
|
|
"num_tokens": 18302955.0,
|
|
"step": 9905
|
|
},
|
|
{
|
|
"entropy": 4.815600252151489,
|
|
"epoch": 8.292887029288703,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 8.748562056276535e-05,
|
|
"loss": 4.0907,
|
|
"mean_token_accuracy": 0.27388441264629365,
|
|
"num_tokens": 18311614.0,
|
|
"step": 9910
|
|
},
|
|
{
|
|
"entropy": 4.8289577007293705,
|
|
"epoch": 8.297071129707113,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 8.730742883905321e-05,
|
|
"loss": 4.074,
|
|
"mean_token_accuracy": 0.27131131291389465,
|
|
"num_tokens": 18320588.0,
|
|
"step": 9915
|
|
},
|
|
{
|
|
"entropy": 4.757870531082153,
|
|
"epoch": 8.301255230125523,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 8.712962335647629e-05,
|
|
"loss": 3.9645,
|
|
"mean_token_accuracy": 0.2914938449859619,
|
|
"num_tokens": 18329976.0,
|
|
"step": 9920
|
|
},
|
|
{
|
|
"entropy": 4.824270725250244,
|
|
"epoch": 8.305439330543933,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.695220448092964e-05,
|
|
"loss": 4.0763,
|
|
"mean_token_accuracy": 0.2757848545908928,
|
|
"num_tokens": 18338495.0,
|
|
"step": 9925
|
|
},
|
|
{
|
|
"entropy": 4.8270515441894535,
|
|
"epoch": 8.309623430962343,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.677517257751286e-05,
|
|
"loss": 4.061,
|
|
"mean_token_accuracy": 0.2806748867034912,
|
|
"num_tokens": 18348212.0,
|
|
"step": 9930
|
|
},
|
|
{
|
|
"entropy": 4.73794379234314,
|
|
"epoch": 8.313807531380753,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.659852801052915e-05,
|
|
"loss": 3.9446,
|
|
"mean_token_accuracy": 0.28820323646068574,
|
|
"num_tokens": 18356803.0,
|
|
"step": 9935
|
|
},
|
|
{
|
|
"entropy": 4.897261333465576,
|
|
"epoch": 8.317991631799163,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.64222711434846e-05,
|
|
"loss": 4.1772,
|
|
"mean_token_accuracy": 0.2701303869485855,
|
|
"num_tokens": 18366822.0,
|
|
"step": 9940
|
|
},
|
|
{
|
|
"entropy": 4.8429491996765135,
|
|
"epoch": 8.322175732217573,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.62464023390876e-05,
|
|
"loss": 4.0783,
|
|
"mean_token_accuracy": 0.27329236418008807,
|
|
"num_tokens": 18376495.0,
|
|
"step": 9945
|
|
},
|
|
{
|
|
"entropy": 4.767585754394531,
|
|
"epoch": 8.326359832635983,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 8.607092195924768e-05,
|
|
"loss": 3.969,
|
|
"mean_token_accuracy": 0.28751012980937957,
|
|
"num_tokens": 18385652.0,
|
|
"step": 9950
|
|
},
|
|
{
|
|
"entropy": 4.80076904296875,
|
|
"epoch": 8.330543933054393,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 8.58958303650754e-05,
|
|
"loss": 4.0104,
|
|
"mean_token_accuracy": 0.2864374428987503,
|
|
"num_tokens": 18394604.0,
|
|
"step": 9955
|
|
},
|
|
{
|
|
"entropy": 4.8469977378845215,
|
|
"epoch": 8.334728033472803,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.572112791688105e-05,
|
|
"loss": 4.0221,
|
|
"mean_token_accuracy": 0.28502110540866854,
|
|
"num_tokens": 18403477.0,
|
|
"step": 9960
|
|
},
|
|
{
|
|
"entropy": 4.817348575592041,
|
|
"epoch": 8.338912133891213,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.554681497417416e-05,
|
|
"loss": 4.0274,
|
|
"mean_token_accuracy": 0.2812277853488922,
|
|
"num_tokens": 18412866.0,
|
|
"step": 9965
|
|
},
|
|
{
|
|
"entropy": 4.822745513916016,
|
|
"epoch": 8.343096234309623,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 8.537289189566284e-05,
|
|
"loss": 4.0103,
|
|
"mean_token_accuracy": 0.2800925374031067,
|
|
"num_tokens": 18421742.0,
|
|
"step": 9970
|
|
},
|
|
{
|
|
"entropy": 4.868430519104004,
|
|
"epoch": 8.347280334728033,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.519935903925271e-05,
|
|
"loss": 4.0894,
|
|
"mean_token_accuracy": 0.2630730226635933,
|
|
"num_tokens": 18431331.0,
|
|
"step": 9975
|
|
},
|
|
{
|
|
"entropy": 4.8344118118286135,
|
|
"epoch": 8.351464435146443,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.502621676204666e-05,
|
|
"loss": 4.1403,
|
|
"mean_token_accuracy": 0.27154895216226577,
|
|
"num_tokens": 18440906.0,
|
|
"step": 9980
|
|
},
|
|
{
|
|
"entropy": 4.817474460601806,
|
|
"epoch": 8.355648535564853,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.485346542034343e-05,
|
|
"loss": 4.0766,
|
|
"mean_token_accuracy": 0.26904295980930326,
|
|
"num_tokens": 18450093.0,
|
|
"step": 9985
|
|
},
|
|
{
|
|
"entropy": 4.748283004760742,
|
|
"epoch": 8.359832635983263,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.468110536963763e-05,
|
|
"loss": 4.0005,
|
|
"mean_token_accuracy": 0.28088130354881286,
|
|
"num_tokens": 18458536.0,
|
|
"step": 9990
|
|
},
|
|
{
|
|
"entropy": 4.832397985458374,
|
|
"epoch": 8.364016736401673,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.450913696461856e-05,
|
|
"loss": 4.0282,
|
|
"mean_token_accuracy": 0.2791893795132637,
|
|
"num_tokens": 18466910.0,
|
|
"step": 9995
|
|
},
|
|
{
|
|
"entropy": 4.845706224441528,
|
|
"epoch": 8.368200836820083,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 8.433756055916944e-05,
|
|
"loss": 4.1009,
|
|
"mean_token_accuracy": 0.27481754422187804,
|
|
"num_tokens": 18476098.0,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"epoch": 8.368200836820083,
|
|
"eval_entropy": 4.677452039133551,
|
|
"eval_loss": 5.312139987945557,
|
|
"eval_mean_token_accuracy": 0.20509978069910487,
|
|
"eval_num_tokens": 18476098.0,
|
|
"eval_runtime": 2.9188,
|
|
"eval_samples_per_second": 1338.58,
|
|
"eval_steps_per_second": 167.537,
|
|
"step": 10000
|
|
},
|
|
{
|
|
"entropy": 4.754444360733032,
|
|
"epoch": 8.372384937238493,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.416637650636708e-05,
|
|
"loss": 4.0007,
|
|
"mean_token_accuracy": 0.28317738622426986,
|
|
"num_tokens": 18486059.0,
|
|
"step": 10005
|
|
},
|
|
{
|
|
"entropy": 4.8514151096344,
|
|
"epoch": 8.376569037656903,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 8.399558515848059e-05,
|
|
"loss": 4.1328,
|
|
"mean_token_accuracy": 0.2734020322561264,
|
|
"num_tokens": 18495373.0,
|
|
"step": 10010
|
|
},
|
|
{
|
|
"entropy": 4.794278526306153,
|
|
"epoch": 8.380753138075313,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.382518686697122e-05,
|
|
"loss": 4.0342,
|
|
"mean_token_accuracy": 0.28334894180297854,
|
|
"num_tokens": 18504559.0,
|
|
"step": 10015
|
|
},
|
|
{
|
|
"entropy": 4.836968278884887,
|
|
"epoch": 8.384937238493723,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 8.365518198249122e-05,
|
|
"loss": 4.0566,
|
|
"mean_token_accuracy": 0.27975860238075256,
|
|
"num_tokens": 18514267.0,
|
|
"step": 10020
|
|
},
|
|
{
|
|
"entropy": 4.672913360595703,
|
|
"epoch": 8.389121338912133,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 8.348557085488323e-05,
|
|
"loss": 3.9373,
|
|
"mean_token_accuracy": 0.29653652012348175,
|
|
"num_tokens": 18523664.0,
|
|
"step": 10025
|
|
},
|
|
{
|
|
"entropy": 4.82189302444458,
|
|
"epoch": 8.393305439330543,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.331635383317981e-05,
|
|
"loss": 4.1223,
|
|
"mean_token_accuracy": 0.2762631595134735,
|
|
"num_tokens": 18532731.0,
|
|
"step": 10030
|
|
},
|
|
{
|
|
"entropy": 4.793521404266357,
|
|
"epoch": 8.397489539748953,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.314753126560229e-05,
|
|
"loss": 4.0426,
|
|
"mean_token_accuracy": 0.27650155276060107,
|
|
"num_tokens": 18542041.0,
|
|
"step": 10035
|
|
},
|
|
{
|
|
"entropy": 4.826600885391235,
|
|
"epoch": 8.401673640167363,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.297910349956045e-05,
|
|
"loss": 4.0531,
|
|
"mean_token_accuracy": 0.2778463214635849,
|
|
"num_tokens": 18551006.0,
|
|
"step": 10040
|
|
},
|
|
{
|
|
"entropy": 4.798341131210327,
|
|
"epoch": 8.405857740585773,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.28110708816515e-05,
|
|
"loss": 4.0803,
|
|
"mean_token_accuracy": 0.27366496324539186,
|
|
"num_tokens": 18560822.0,
|
|
"step": 10045
|
|
},
|
|
{
|
|
"entropy": 4.7535888671875,
|
|
"epoch": 8.410041841004183,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 8.264343375765951e-05,
|
|
"loss": 4.0313,
|
|
"mean_token_accuracy": 0.27690701484680175,
|
|
"num_tokens": 18570993.0,
|
|
"step": 10050
|
|
},
|
|
{
|
|
"entropy": 4.810549259185791,
|
|
"epoch": 8.414225941422593,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 8.247619247255486e-05,
|
|
"loss": 4.0877,
|
|
"mean_token_accuracy": 0.2766179770231247,
|
|
"num_tokens": 18580424.0,
|
|
"step": 10055
|
|
},
|
|
{
|
|
"entropy": 4.785850763320923,
|
|
"epoch": 8.418410041841003,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.230934737049309e-05,
|
|
"loss": 4.074,
|
|
"mean_token_accuracy": 0.26954350918531417,
|
|
"num_tokens": 18588789.0,
|
|
"step": 10060
|
|
},
|
|
{
|
|
"entropy": 4.848625040054321,
|
|
"epoch": 8.422594142259415,
|
|
"grad_norm": 1.5859375,
|
|
"learning_rate": 8.21428987948147e-05,
|
|
"loss": 4.0637,
|
|
"mean_token_accuracy": 0.26828747391700747,
|
|
"num_tokens": 18597553.0,
|
|
"step": 10065
|
|
},
|
|
{
|
|
"entropy": 4.84667387008667,
|
|
"epoch": 8.426778242677825,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 8.197684708804399e-05,
|
|
"loss": 4.0853,
|
|
"mean_token_accuracy": 0.283654060959816,
|
|
"num_tokens": 18606997.0,
|
|
"step": 10070
|
|
},
|
|
{
|
|
"entropy": 4.811169242858886,
|
|
"epoch": 8.430962343096235,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.18111925918888e-05,
|
|
"loss": 4.0556,
|
|
"mean_token_accuracy": 0.27860024124383925,
|
|
"num_tokens": 18615807.0,
|
|
"step": 10075
|
|
},
|
|
{
|
|
"entropy": 4.87128643989563,
|
|
"epoch": 8.435146443514645,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 8.164593564723938e-05,
|
|
"loss": 4.0931,
|
|
"mean_token_accuracy": 0.27292597442865374,
|
|
"num_tokens": 18625276.0,
|
|
"step": 10080
|
|
},
|
|
{
|
|
"entropy": 4.832700395584107,
|
|
"epoch": 8.439330543933055,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 8.148107659416793e-05,
|
|
"loss": 4.0993,
|
|
"mean_token_accuracy": 0.2774732351303101,
|
|
"num_tokens": 18634269.0,
|
|
"step": 10085
|
|
},
|
|
{
|
|
"entropy": 4.917843389511108,
|
|
"epoch": 8.443514644351465,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 8.131661577192792e-05,
|
|
"loss": 4.1455,
|
|
"mean_token_accuracy": 0.27100523114204406,
|
|
"num_tokens": 18644506.0,
|
|
"step": 10090
|
|
},
|
|
{
|
|
"entropy": 4.812201261520386,
|
|
"epoch": 8.447698744769875,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.115255351895325e-05,
|
|
"loss": 4.0202,
|
|
"mean_token_accuracy": 0.280697637796402,
|
|
"num_tokens": 18654250.0,
|
|
"step": 10095
|
|
},
|
|
{
|
|
"entropy": 4.8669922828674315,
|
|
"epoch": 8.451882845188285,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.098889017285771e-05,
|
|
"loss": 4.0896,
|
|
"mean_token_accuracy": 0.27393464744091034,
|
|
"num_tokens": 18664089.0,
|
|
"step": 10100
|
|
},
|
|
{
|
|
"entropy": 4.89521861076355,
|
|
"epoch": 8.456066945606695,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.082562607043407e-05,
|
|
"loss": 4.1523,
|
|
"mean_token_accuracy": 0.2668469071388245,
|
|
"num_tokens": 18674233.0,
|
|
"step": 10105
|
|
},
|
|
{
|
|
"entropy": 4.850672817230224,
|
|
"epoch": 8.460251046025105,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 8.066276154765374e-05,
|
|
"loss": 4.0959,
|
|
"mean_token_accuracy": 0.27475244402885435,
|
|
"num_tokens": 18683960.0,
|
|
"step": 10110
|
|
},
|
|
{
|
|
"entropy": 4.80786771774292,
|
|
"epoch": 8.464435146443515,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.050029693966561e-05,
|
|
"loss": 4.107,
|
|
"mean_token_accuracy": 0.26953665912151337,
|
|
"num_tokens": 18694418.0,
|
|
"step": 10115
|
|
},
|
|
{
|
|
"entropy": 4.830559587478637,
|
|
"epoch": 8.468619246861925,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 8.033823258079572e-05,
|
|
"loss": 4.0627,
|
|
"mean_token_accuracy": 0.27534993886947634,
|
|
"num_tokens": 18703986.0,
|
|
"step": 10120
|
|
},
|
|
{
|
|
"entropy": 4.852240324020386,
|
|
"epoch": 8.472803347280335,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 8.017656880454651e-05,
|
|
"loss": 4.095,
|
|
"mean_token_accuracy": 0.2743106558918953,
|
|
"num_tokens": 18712856.0,
|
|
"step": 10125
|
|
},
|
|
{
|
|
"entropy": 4.812576103210449,
|
|
"epoch": 8.476987447698745,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 8.0015305943596e-05,
|
|
"loss": 4.0946,
|
|
"mean_token_accuracy": 0.2723763927817345,
|
|
"num_tokens": 18721795.0,
|
|
"step": 10130
|
|
},
|
|
{
|
|
"entropy": 4.802620124816895,
|
|
"epoch": 8.481171548117155,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.985444432979733e-05,
|
|
"loss": 4.104,
|
|
"mean_token_accuracy": 0.27691680788993833,
|
|
"num_tokens": 18730475.0,
|
|
"step": 10135
|
|
},
|
|
{
|
|
"entropy": 4.821155261993408,
|
|
"epoch": 8.485355648535565,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.969398429417768e-05,
|
|
"loss": 4.0125,
|
|
"mean_token_accuracy": 0.28785293102264403,
|
|
"num_tokens": 18740017.0,
|
|
"step": 10140
|
|
},
|
|
{
|
|
"entropy": 4.879079437255859,
|
|
"epoch": 8.489539748953975,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 7.953392616693814e-05,
|
|
"loss": 4.0961,
|
|
"mean_token_accuracy": 0.2772545456886292,
|
|
"num_tokens": 18749018.0,
|
|
"step": 10145
|
|
},
|
|
{
|
|
"entropy": 4.917897701263428,
|
|
"epoch": 8.493723849372385,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.937427027745259e-05,
|
|
"loss": 4.1168,
|
|
"mean_token_accuracy": 0.2755123436450958,
|
|
"num_tokens": 18758661.0,
|
|
"step": 10150
|
|
},
|
|
{
|
|
"entropy": 4.790317010879517,
|
|
"epoch": 8.497907949790795,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 7.921501695426707e-05,
|
|
"loss": 4.0378,
|
|
"mean_token_accuracy": 0.27724991738796234,
|
|
"num_tokens": 18767769.0,
|
|
"step": 10155
|
|
},
|
|
{
|
|
"entropy": 4.828195810317993,
|
|
"epoch": 8.502092050209205,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.905616652509945e-05,
|
|
"loss": 4.1409,
|
|
"mean_token_accuracy": 0.27320107221603396,
|
|
"num_tokens": 18777069.0,
|
|
"step": 10160
|
|
},
|
|
{
|
|
"entropy": 4.8315245628356935,
|
|
"epoch": 8.506276150627615,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.889771931683826e-05,
|
|
"loss": 4.1056,
|
|
"mean_token_accuracy": 0.2708635300397873,
|
|
"num_tokens": 18788123.0,
|
|
"step": 10165
|
|
},
|
|
{
|
|
"entropy": 4.823835802078247,
|
|
"epoch": 8.510460251046025,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.873967565554247e-05,
|
|
"loss": 4.072,
|
|
"mean_token_accuracy": 0.2769174337387085,
|
|
"num_tokens": 18797421.0,
|
|
"step": 10170
|
|
},
|
|
{
|
|
"entropy": 4.775837373733521,
|
|
"epoch": 8.514644351464435,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.858203586644047e-05,
|
|
"loss": 4.0296,
|
|
"mean_token_accuracy": 0.2836910426616669,
|
|
"num_tokens": 18806542.0,
|
|
"step": 10175
|
|
},
|
|
{
|
|
"entropy": 4.8729935646057125,
|
|
"epoch": 8.518828451882845,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.842480027392966e-05,
|
|
"loss": 4.1134,
|
|
"mean_token_accuracy": 0.27465411424636843,
|
|
"num_tokens": 18815498.0,
|
|
"step": 10180
|
|
},
|
|
{
|
|
"entropy": 4.812866497039795,
|
|
"epoch": 8.523012552301255,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.826796920157558e-05,
|
|
"loss": 4.0433,
|
|
"mean_token_accuracy": 0.2765009880065918,
|
|
"num_tokens": 18824388.0,
|
|
"step": 10185
|
|
},
|
|
{
|
|
"entropy": 4.811968660354614,
|
|
"epoch": 8.527196652719665,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 7.81115429721113e-05,
|
|
"loss": 4.0422,
|
|
"mean_token_accuracy": 0.2715842217206955,
|
|
"num_tokens": 18833180.0,
|
|
"step": 10190
|
|
},
|
|
{
|
|
"entropy": 4.799845170974732,
|
|
"epoch": 8.531380753138075,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.795552190743693e-05,
|
|
"loss": 4.0633,
|
|
"mean_token_accuracy": 0.2782320737838745,
|
|
"num_tokens": 18842306.0,
|
|
"step": 10195
|
|
},
|
|
{
|
|
"entropy": 4.782264137268067,
|
|
"epoch": 8.535564853556485,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.779990632861869e-05,
|
|
"loss": 4.0443,
|
|
"mean_token_accuracy": 0.2785312205553055,
|
|
"num_tokens": 18851843.0,
|
|
"step": 10200
|
|
},
|
|
{
|
|
"entropy": 4.826796436309815,
|
|
"epoch": 8.539748953974895,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.764469655588845e-05,
|
|
"loss": 4.1143,
|
|
"mean_token_accuracy": 0.27062701135873796,
|
|
"num_tokens": 18860943.0,
|
|
"step": 10205
|
|
},
|
|
{
|
|
"entropy": 4.781267547607422,
|
|
"epoch": 8.543933054393305,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.748989290864301e-05,
|
|
"loss": 3.9852,
|
|
"mean_token_accuracy": 0.2901459500193596,
|
|
"num_tokens": 18869925.0,
|
|
"step": 10210
|
|
},
|
|
{
|
|
"entropy": 4.743795680999756,
|
|
"epoch": 8.548117154811715,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.733549570544334e-05,
|
|
"loss": 4.1339,
|
|
"mean_token_accuracy": 0.27910166382789614,
|
|
"num_tokens": 18880022.0,
|
|
"step": 10215
|
|
},
|
|
{
|
|
"entropy": 4.749596071243286,
|
|
"epoch": 8.552301255230125,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 7.718150526401405e-05,
|
|
"loss": 3.9824,
|
|
"mean_token_accuracy": 0.2822796702384949,
|
|
"num_tokens": 18889756.0,
|
|
"step": 10220
|
|
},
|
|
{
|
|
"entropy": 4.797429800033569,
|
|
"epoch": 8.556485355648535,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.702792190124268e-05,
|
|
"loss": 4.0703,
|
|
"mean_token_accuracy": 0.2732760474085808,
|
|
"num_tokens": 18899688.0,
|
|
"step": 10225
|
|
},
|
|
{
|
|
"entropy": 4.784254455566407,
|
|
"epoch": 8.560669456066945,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.687474593317915e-05,
|
|
"loss": 4.0388,
|
|
"mean_token_accuracy": 0.28504807949066163,
|
|
"num_tokens": 18908862.0,
|
|
"step": 10230
|
|
},
|
|
{
|
|
"entropy": 4.850057077407837,
|
|
"epoch": 8.564853556485355,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.6721977675035e-05,
|
|
"loss": 4.0725,
|
|
"mean_token_accuracy": 0.27079208493232726,
|
|
"num_tokens": 18918292.0,
|
|
"step": 10235
|
|
},
|
|
{
|
|
"entropy": 4.728587865829468,
|
|
"epoch": 8.569037656903765,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.656961744118266e-05,
|
|
"loss": 3.9555,
|
|
"mean_token_accuracy": 0.2832908481359482,
|
|
"num_tokens": 18927181.0,
|
|
"step": 10240
|
|
},
|
|
{
|
|
"entropy": 4.761023712158203,
|
|
"epoch": 8.573221757322175,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.641766554515512e-05,
|
|
"loss": 4.0449,
|
|
"mean_token_accuracy": 0.2850276529788971,
|
|
"num_tokens": 18936259.0,
|
|
"step": 10245
|
|
},
|
|
{
|
|
"entropy": 4.849989938735962,
|
|
"epoch": 8.577405857740585,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.626612229964486e-05,
|
|
"loss": 4.1233,
|
|
"mean_token_accuracy": 0.2739103451371193,
|
|
"num_tokens": 18946073.0,
|
|
"step": 10250
|
|
},
|
|
{
|
|
"entropy": 4.812704992294312,
|
|
"epoch": 8.581589958158997,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.611498801650353e-05,
|
|
"loss": 3.9936,
|
|
"mean_token_accuracy": 0.2875778526067734,
|
|
"num_tokens": 18955472.0,
|
|
"step": 10255
|
|
},
|
|
{
|
|
"entropy": 4.805995035171509,
|
|
"epoch": 8.585774058577407,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 7.596426300674113e-05,
|
|
"loss": 4.0492,
|
|
"mean_token_accuracy": 0.27312630116939546,
|
|
"num_tokens": 18965407.0,
|
|
"step": 10260
|
|
},
|
|
{
|
|
"entropy": 4.787683820724487,
|
|
"epoch": 8.589958158995817,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 7.581394758052559e-05,
|
|
"loss": 4.0172,
|
|
"mean_token_accuracy": 0.27489573061466216,
|
|
"num_tokens": 18974505.0,
|
|
"step": 10265
|
|
},
|
|
{
|
|
"entropy": 4.85532774925232,
|
|
"epoch": 8.594142259414227,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.566404204718188e-05,
|
|
"loss": 4.1031,
|
|
"mean_token_accuracy": 0.2702629804611206,
|
|
"num_tokens": 18983439.0,
|
|
"step": 10270
|
|
},
|
|
{
|
|
"entropy": 4.775296640396118,
|
|
"epoch": 8.598326359832637,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.551454671519142e-05,
|
|
"loss": 4.0389,
|
|
"mean_token_accuracy": 0.2738132536411285,
|
|
"num_tokens": 18992548.0,
|
|
"step": 10275
|
|
},
|
|
{
|
|
"entropy": 4.7787919521331785,
|
|
"epoch": 8.602510460251047,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.536546189219171e-05,
|
|
"loss": 4.0461,
|
|
"mean_token_accuracy": 0.28804281949996946,
|
|
"num_tokens": 19000846.0,
|
|
"step": 10280
|
|
},
|
|
{
|
|
"entropy": 4.761201620101929,
|
|
"epoch": 8.606694560669457,
|
|
"grad_norm": 1.609375,
|
|
"learning_rate": 7.521678788497515e-05,
|
|
"loss": 3.9568,
|
|
"mean_token_accuracy": 0.29866126775741575,
|
|
"num_tokens": 19009387.0,
|
|
"step": 10285
|
|
},
|
|
{
|
|
"entropy": 4.782753753662109,
|
|
"epoch": 8.610878661087867,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.506852499948905e-05,
|
|
"loss": 4.0526,
|
|
"mean_token_accuracy": 0.277069254219532,
|
|
"num_tokens": 19018039.0,
|
|
"step": 10290
|
|
},
|
|
{
|
|
"entropy": 4.776949501037597,
|
|
"epoch": 8.615062761506277,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.492067354083462e-05,
|
|
"loss": 4.0007,
|
|
"mean_token_accuracy": 0.2779260754585266,
|
|
"num_tokens": 19026769.0,
|
|
"step": 10295
|
|
},
|
|
{
|
|
"entropy": 4.858229112625122,
|
|
"epoch": 8.619246861924687,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 7.47732338132663e-05,
|
|
"loss": 4.147,
|
|
"mean_token_accuracy": 0.27093727588653566,
|
|
"num_tokens": 19037030.0,
|
|
"step": 10300
|
|
},
|
|
{
|
|
"entropy": 4.895801067352295,
|
|
"epoch": 8.623430962343097,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 7.462620612019146e-05,
|
|
"loss": 4.1442,
|
|
"mean_token_accuracy": 0.2652414575219154,
|
|
"num_tokens": 19046891.0,
|
|
"step": 10305
|
|
},
|
|
{
|
|
"entropy": 4.804647541046142,
|
|
"epoch": 8.627615062761507,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.447959076416934e-05,
|
|
"loss": 4.0433,
|
|
"mean_token_accuracy": 0.27573239207267763,
|
|
"num_tokens": 19055863.0,
|
|
"step": 10310
|
|
},
|
|
{
|
|
"entropy": 4.826656103134155,
|
|
"epoch": 8.631799163179917,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 7.433338804691087e-05,
|
|
"loss": 4.095,
|
|
"mean_token_accuracy": 0.27579325437545776,
|
|
"num_tokens": 19065564.0,
|
|
"step": 10315
|
|
},
|
|
{
|
|
"entropy": 4.892518806457519,
|
|
"epoch": 8.635983263598327,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.41875982692777e-05,
|
|
"loss": 4.156,
|
|
"mean_token_accuracy": 0.26417338848114014,
|
|
"num_tokens": 19075532.0,
|
|
"step": 10320
|
|
},
|
|
{
|
|
"entropy": 4.798560619354248,
|
|
"epoch": 8.640167364016737,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 7.40422217312817e-05,
|
|
"loss": 3.9703,
|
|
"mean_token_accuracy": 0.28827973902225495,
|
|
"num_tokens": 19083915.0,
|
|
"step": 10325
|
|
},
|
|
{
|
|
"entropy": 4.7648152828216555,
|
|
"epoch": 8.644351464435147,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.38972587320845e-05,
|
|
"loss": 4.0252,
|
|
"mean_token_accuracy": 0.27711681425571444,
|
|
"num_tokens": 19093094.0,
|
|
"step": 10330
|
|
},
|
|
{
|
|
"entropy": 4.853082323074341,
|
|
"epoch": 8.648535564853557,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 7.375270956999652e-05,
|
|
"loss": 4.0624,
|
|
"mean_token_accuracy": 0.283171983063221,
|
|
"num_tokens": 19103124.0,
|
|
"step": 10335
|
|
},
|
|
{
|
|
"entropy": 4.793401288986206,
|
|
"epoch": 8.652719665271967,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 7.360857454247686e-05,
|
|
"loss": 4.0184,
|
|
"mean_token_accuracy": 0.28055262863636016,
|
|
"num_tokens": 19111307.0,
|
|
"step": 10340
|
|
},
|
|
{
|
|
"entropy": 4.815118598937988,
|
|
"epoch": 8.656903765690377,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.34648539461321e-05,
|
|
"loss": 4.0751,
|
|
"mean_token_accuracy": 0.27517028599977494,
|
|
"num_tokens": 19120756.0,
|
|
"step": 10345
|
|
},
|
|
{
|
|
"entropy": 4.804310274124146,
|
|
"epoch": 8.661087866108787,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 7.332154807671623e-05,
|
|
"loss": 4.0118,
|
|
"mean_token_accuracy": 0.28492148518562316,
|
|
"num_tokens": 19129374.0,
|
|
"step": 10350
|
|
},
|
|
{
|
|
"entropy": 4.802014636993408,
|
|
"epoch": 8.665271966527197,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 7.317865722912958e-05,
|
|
"loss": 4.0505,
|
|
"mean_token_accuracy": 0.2761240839958191,
|
|
"num_tokens": 19138384.0,
|
|
"step": 10355
|
|
},
|
|
{
|
|
"entropy": 4.780322408676147,
|
|
"epoch": 8.669456066945607,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.30361816974186e-05,
|
|
"loss": 3.9967,
|
|
"mean_token_accuracy": 0.28542815446853637,
|
|
"num_tokens": 19146926.0,
|
|
"step": 10360
|
|
},
|
|
{
|
|
"entropy": 4.84246826171875,
|
|
"epoch": 8.673640167364017,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 7.289412177477505e-05,
|
|
"loss": 4.1198,
|
|
"mean_token_accuracy": 0.2748185858130455,
|
|
"num_tokens": 19156181.0,
|
|
"step": 10365
|
|
},
|
|
{
|
|
"entropy": 4.848178052902222,
|
|
"epoch": 8.677824267782427,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 7.275247775353531e-05,
|
|
"loss": 4.1602,
|
|
"mean_token_accuracy": 0.26341464221477506,
|
|
"num_tokens": 19165743.0,
|
|
"step": 10370
|
|
},
|
|
{
|
|
"entropy": 4.827004528045654,
|
|
"epoch": 8.682008368200837,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 7.261124992518016e-05,
|
|
"loss": 4.1099,
|
|
"mean_token_accuracy": 0.26661452651023865,
|
|
"num_tokens": 19174498.0,
|
|
"step": 10375
|
|
},
|
|
{
|
|
"entropy": 4.783060836791992,
|
|
"epoch": 8.686192468619247,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.247043858033366e-05,
|
|
"loss": 4.0372,
|
|
"mean_token_accuracy": 0.28643020838499067,
|
|
"num_tokens": 19183326.0,
|
|
"step": 10380
|
|
},
|
|
{
|
|
"entropy": 4.841526126861572,
|
|
"epoch": 8.690376569037657,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.233004400876287e-05,
|
|
"loss": 4.0426,
|
|
"mean_token_accuracy": 0.27708722800016405,
|
|
"num_tokens": 19192593.0,
|
|
"step": 10385
|
|
},
|
|
{
|
|
"entropy": 4.7986907958984375,
|
|
"epoch": 8.694560669456067,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.21900664993774e-05,
|
|
"loss": 4.0881,
|
|
"mean_token_accuracy": 0.28490635454654695,
|
|
"num_tokens": 19201751.0,
|
|
"step": 10390
|
|
},
|
|
{
|
|
"entropy": 4.76883978843689,
|
|
"epoch": 8.698744769874477,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 7.20505063402283e-05,
|
|
"loss": 4.0248,
|
|
"mean_token_accuracy": 0.28200308978557587,
|
|
"num_tokens": 19211293.0,
|
|
"step": 10395
|
|
},
|
|
{
|
|
"entropy": 4.768481302261352,
|
|
"epoch": 8.702928870292887,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 7.191136381850806e-05,
|
|
"loss": 4.0792,
|
|
"mean_token_accuracy": 0.2711384892463684,
|
|
"num_tokens": 19220555.0,
|
|
"step": 10400
|
|
},
|
|
{
|
|
"entropy": 4.781294107437134,
|
|
"epoch": 8.707112970711297,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.177263922054956e-05,
|
|
"loss": 4.0612,
|
|
"mean_token_accuracy": 0.26924450397491456,
|
|
"num_tokens": 19230149.0,
|
|
"step": 10405
|
|
},
|
|
{
|
|
"entropy": 4.7708948135375975,
|
|
"epoch": 8.711297071129707,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 7.163433283182571e-05,
|
|
"loss": 4.0822,
|
|
"mean_token_accuracy": 0.2758555680513382,
|
|
"num_tokens": 19238575.0,
|
|
"step": 10410
|
|
},
|
|
{
|
|
"entropy": 4.75157732963562,
|
|
"epoch": 8.715481171548117,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.14964449369489e-05,
|
|
"loss": 3.9115,
|
|
"mean_token_accuracy": 0.2899089068174362,
|
|
"num_tokens": 19247540.0,
|
|
"step": 10415
|
|
},
|
|
{
|
|
"entropy": 4.824244260787964,
|
|
"epoch": 8.719665271966527,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.135897581967013e-05,
|
|
"loss": 4.0493,
|
|
"mean_token_accuracy": 0.28137223720550536,
|
|
"num_tokens": 19257245.0,
|
|
"step": 10420
|
|
},
|
|
{
|
|
"entropy": 4.785420894622803,
|
|
"epoch": 8.723849372384937,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.122192576287885e-05,
|
|
"loss": 4.0519,
|
|
"mean_token_accuracy": 0.2826233893632889,
|
|
"num_tokens": 19267580.0,
|
|
"step": 10425
|
|
},
|
|
{
|
|
"entropy": 4.793396711349487,
|
|
"epoch": 8.728033472803347,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 7.108529504860195e-05,
|
|
"loss": 4.045,
|
|
"mean_token_accuracy": 0.28436287194490434,
|
|
"num_tokens": 19276461.0,
|
|
"step": 10430
|
|
},
|
|
{
|
|
"entropy": 4.8677449226379395,
|
|
"epoch": 8.732217573221757,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 7.094908395800357e-05,
|
|
"loss": 4.079,
|
|
"mean_token_accuracy": 0.2761824935674667,
|
|
"num_tokens": 19285184.0,
|
|
"step": 10435
|
|
},
|
|
{
|
|
"entropy": 4.827707147598266,
|
|
"epoch": 8.736401673640167,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 7.081329277138412e-05,
|
|
"loss": 4.0457,
|
|
"mean_token_accuracy": 0.27814962565898893,
|
|
"num_tokens": 19294170.0,
|
|
"step": 10440
|
|
},
|
|
{
|
|
"entropy": 4.864910316467285,
|
|
"epoch": 8.740585774058577,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 7.067792176818011e-05,
|
|
"loss": 4.1325,
|
|
"mean_token_accuracy": 0.2691267132759094,
|
|
"num_tokens": 19303786.0,
|
|
"step": 10445
|
|
},
|
|
{
|
|
"entropy": 4.772932529449463,
|
|
"epoch": 8.744769874476987,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.054297122696336e-05,
|
|
"loss": 4.0621,
|
|
"mean_token_accuracy": 0.27990056574344635,
|
|
"num_tokens": 19313759.0,
|
|
"step": 10450
|
|
},
|
|
{
|
|
"entropy": 4.857951593399048,
|
|
"epoch": 8.748953974895397,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 7.040844142544026e-05,
|
|
"loss": 4.1487,
|
|
"mean_token_accuracy": 0.2698888435959816,
|
|
"num_tokens": 19322462.0,
|
|
"step": 10455
|
|
},
|
|
{
|
|
"entropy": 4.839073753356933,
|
|
"epoch": 8.753138075313807,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.027433264045156e-05,
|
|
"loss": 4.0974,
|
|
"mean_token_accuracy": 0.28137233257293703,
|
|
"num_tokens": 19330860.0,
|
|
"step": 10460
|
|
},
|
|
{
|
|
"entropy": 4.831358861923218,
|
|
"epoch": 8.757322175732218,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 7.014064514797157e-05,
|
|
"loss": 4.1038,
|
|
"mean_token_accuracy": 0.2730175107717514,
|
|
"num_tokens": 19340226.0,
|
|
"step": 10465
|
|
},
|
|
{
|
|
"entropy": 4.787745761871338,
|
|
"epoch": 8.761506276150628,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 7.000737922310772e-05,
|
|
"loss": 4.0135,
|
|
"mean_token_accuracy": 0.28209146559238435,
|
|
"num_tokens": 19349238.0,
|
|
"step": 10470
|
|
},
|
|
{
|
|
"entropy": 4.858227777481079,
|
|
"epoch": 8.765690376569038,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.987453514009977e-05,
|
|
"loss": 4.0997,
|
|
"mean_token_accuracy": 0.27717286348342896,
|
|
"num_tokens": 19358887.0,
|
|
"step": 10475
|
|
},
|
|
{
|
|
"entropy": 4.8094377517700195,
|
|
"epoch": 8.769874476987448,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.974211317231954e-05,
|
|
"loss": 4.0575,
|
|
"mean_token_accuracy": 0.2787767291069031,
|
|
"num_tokens": 19368800.0,
|
|
"step": 10480
|
|
},
|
|
{
|
|
"entropy": 4.760870504379272,
|
|
"epoch": 8.774058577405858,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.961011359227014e-05,
|
|
"loss": 3.9827,
|
|
"mean_token_accuracy": 0.28313044607639315,
|
|
"num_tokens": 19377162.0,
|
|
"step": 10485
|
|
},
|
|
{
|
|
"entropy": 4.88700909614563,
|
|
"epoch": 8.778242677824268,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.947853667158544e-05,
|
|
"loss": 4.1233,
|
|
"mean_token_accuracy": 0.2662218615412712,
|
|
"num_tokens": 19387236.0,
|
|
"step": 10490
|
|
},
|
|
{
|
|
"entropy": 4.804559659957886,
|
|
"epoch": 8.782426778242678,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.934738268102963e-05,
|
|
"loss": 4.0916,
|
|
"mean_token_accuracy": 0.275602313876152,
|
|
"num_tokens": 19396880.0,
|
|
"step": 10495
|
|
},
|
|
{
|
|
"entropy": 4.84183521270752,
|
|
"epoch": 8.786610878661088,
|
|
"grad_norm": 1.6484375,
|
|
"learning_rate": 6.921665189049654e-05,
|
|
"loss": 4.0777,
|
|
"mean_token_accuracy": 0.2720334678888321,
|
|
"num_tokens": 19405796.0,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"epoch": 8.786610878661088,
|
|
"eval_entropy": 4.686278187667909,
|
|
"eval_loss": 5.3094329833984375,
|
|
"eval_mean_token_accuracy": 0.20575361775047696,
|
|
"eval_num_tokens": 19405796.0,
|
|
"eval_runtime": 2.73,
|
|
"eval_samples_per_second": 1431.147,
|
|
"eval_steps_per_second": 179.122,
|
|
"step": 10500
|
|
},
|
|
{
|
|
"entropy": 4.767242383956909,
|
|
"epoch": 8.790794979079498,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.908634456900908e-05,
|
|
"loss": 4.0053,
|
|
"mean_token_accuracy": 0.28747248351573945,
|
|
"num_tokens": 19414911.0,
|
|
"step": 10505
|
|
},
|
|
{
|
|
"entropy": 4.846585035324097,
|
|
"epoch": 8.794979079497908,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.895646098471883e-05,
|
|
"loss": 4.0991,
|
|
"mean_token_accuracy": 0.27171247005462645,
|
|
"num_tokens": 19424660.0,
|
|
"step": 10510
|
|
},
|
|
{
|
|
"entropy": 4.878277111053467,
|
|
"epoch": 8.799163179916318,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.882700140490531e-05,
|
|
"loss": 4.119,
|
|
"mean_token_accuracy": 0.269419951736927,
|
|
"num_tokens": 19434106.0,
|
|
"step": 10515
|
|
},
|
|
{
|
|
"entropy": 4.833722639083862,
|
|
"epoch": 8.803347280334728,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.869796609597547e-05,
|
|
"loss": 4.1162,
|
|
"mean_token_accuracy": 0.2777192324399948,
|
|
"num_tokens": 19443756.0,
|
|
"step": 10520
|
|
},
|
|
{
|
|
"entropy": 4.806939744949341,
|
|
"epoch": 8.807531380753138,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.856935532346328e-05,
|
|
"loss": 4.0317,
|
|
"mean_token_accuracy": 0.27892889380455016,
|
|
"num_tokens": 19452361.0,
|
|
"step": 10525
|
|
},
|
|
{
|
|
"entropy": 4.8370085716247555,
|
|
"epoch": 8.811715481171548,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.844116935202896e-05,
|
|
"loss": 4.1442,
|
|
"mean_token_accuracy": 0.269351652264595,
|
|
"num_tokens": 19461718.0,
|
|
"step": 10530
|
|
},
|
|
{
|
|
"entropy": 4.776788520812988,
|
|
"epoch": 8.815899581589958,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 6.831340844545875e-05,
|
|
"loss": 4.0831,
|
|
"mean_token_accuracy": 0.27124001681804655,
|
|
"num_tokens": 19471048.0,
|
|
"step": 10535
|
|
},
|
|
{
|
|
"entropy": 4.811686992645264,
|
|
"epoch": 8.820083682008368,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.818607286666396e-05,
|
|
"loss": 4.0602,
|
|
"mean_token_accuracy": 0.27922221422195437,
|
|
"num_tokens": 19479489.0,
|
|
"step": 10540
|
|
},
|
|
{
|
|
"entropy": 4.835172128677368,
|
|
"epoch": 8.824267782426778,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.805916287768076e-05,
|
|
"loss": 4.0453,
|
|
"mean_token_accuracy": 0.286376890540123,
|
|
"num_tokens": 19488853.0,
|
|
"step": 10545
|
|
},
|
|
{
|
|
"entropy": 4.80435357093811,
|
|
"epoch": 8.828451882845188,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 6.793267873966957e-05,
|
|
"loss": 4.1361,
|
|
"mean_token_accuracy": 0.2684462696313858,
|
|
"num_tokens": 19499224.0,
|
|
"step": 10550
|
|
},
|
|
{
|
|
"entropy": 4.863946008682251,
|
|
"epoch": 8.832635983263598,
|
|
"grad_norm": 1.6015625,
|
|
"learning_rate": 6.78066207129143e-05,
|
|
"loss": 4.1148,
|
|
"mean_token_accuracy": 0.27350460588932035,
|
|
"num_tokens": 19509412.0,
|
|
"step": 10555
|
|
},
|
|
{
|
|
"entropy": 4.839503192901612,
|
|
"epoch": 8.836820083682008,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 6.768098905682218e-05,
|
|
"loss": 4.053,
|
|
"mean_token_accuracy": 0.2722203403711319,
|
|
"num_tokens": 19517945.0,
|
|
"step": 10560
|
|
},
|
|
{
|
|
"entropy": 4.801218509674072,
|
|
"epoch": 8.841004184100418,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.755578402992289e-05,
|
|
"loss": 4.0244,
|
|
"mean_token_accuracy": 0.2783090442419052,
|
|
"num_tokens": 19526513.0,
|
|
"step": 10565
|
|
},
|
|
{
|
|
"entropy": 4.829199600219726,
|
|
"epoch": 8.845188284518828,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.743100588986833e-05,
|
|
"loss": 4.114,
|
|
"mean_token_accuracy": 0.27859323620796206,
|
|
"num_tokens": 19536904.0,
|
|
"step": 10570
|
|
},
|
|
{
|
|
"entropy": 4.8433760643005375,
|
|
"epoch": 8.849372384937238,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.730665489343178e-05,
|
|
"loss": 4.1649,
|
|
"mean_token_accuracy": 0.2678702473640442,
|
|
"num_tokens": 19545440.0,
|
|
"step": 10575
|
|
},
|
|
{
|
|
"entropy": 4.793996572494507,
|
|
"epoch": 8.853556485355648,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.71827312965077e-05,
|
|
"loss": 4.0448,
|
|
"mean_token_accuracy": 0.2794655919075012,
|
|
"num_tokens": 19554574.0,
|
|
"step": 10580
|
|
},
|
|
{
|
|
"entropy": 4.880933618545532,
|
|
"epoch": 8.857740585774058,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 6.70592353541108e-05,
|
|
"loss": 4.0656,
|
|
"mean_token_accuracy": 0.2781416535377502,
|
|
"num_tokens": 19564073.0,
|
|
"step": 10585
|
|
},
|
|
{
|
|
"entropy": 4.791238021850586,
|
|
"epoch": 8.861924686192468,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.693616732037592e-05,
|
|
"loss": 4.0694,
|
|
"mean_token_accuracy": 0.27903437316417695,
|
|
"num_tokens": 19573597.0,
|
|
"step": 10590
|
|
},
|
|
{
|
|
"entropy": 4.8482402801513675,
|
|
"epoch": 8.866108786610878,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.681352744855735e-05,
|
|
"loss": 4.055,
|
|
"mean_token_accuracy": 0.2818945974111557,
|
|
"num_tokens": 19583336.0,
|
|
"step": 10595
|
|
},
|
|
{
|
|
"entropy": 4.812443590164184,
|
|
"epoch": 8.870292887029288,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 6.66913159910281e-05,
|
|
"loss": 4.0767,
|
|
"mean_token_accuracy": 0.27774510979652406,
|
|
"num_tokens": 19592780.0,
|
|
"step": 10600
|
|
},
|
|
{
|
|
"entropy": 4.795892190933228,
|
|
"epoch": 8.874476987447698,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.656953319927987e-05,
|
|
"loss": 4.0634,
|
|
"mean_token_accuracy": 0.2758260816335678,
|
|
"num_tokens": 19601631.0,
|
|
"step": 10605
|
|
},
|
|
{
|
|
"entropy": 4.836074018478394,
|
|
"epoch": 8.878661087866108,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 6.644817932392192e-05,
|
|
"loss": 4.1051,
|
|
"mean_token_accuracy": 0.2758748009800911,
|
|
"num_tokens": 19611703.0,
|
|
"step": 10610
|
|
},
|
|
{
|
|
"entropy": 4.8196042537689205,
|
|
"epoch": 8.882845188284518,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.632725461468108e-05,
|
|
"loss": 4.0724,
|
|
"mean_token_accuracy": 0.2760209619998932,
|
|
"num_tokens": 19621639.0,
|
|
"step": 10615
|
|
},
|
|
{
|
|
"entropy": 4.813204336166382,
|
|
"epoch": 8.887029288702928,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.620675932040099e-05,
|
|
"loss": 4.0371,
|
|
"mean_token_accuracy": 0.27946056723594664,
|
|
"num_tokens": 19630128.0,
|
|
"step": 10620
|
|
},
|
|
{
|
|
"entropy": 4.819009065628052,
|
|
"epoch": 8.891213389121338,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.608669368904147e-05,
|
|
"loss": 4.071,
|
|
"mean_token_accuracy": 0.27776946127414703,
|
|
"num_tokens": 19639420.0,
|
|
"step": 10625
|
|
},
|
|
{
|
|
"entropy": 4.8688013553619385,
|
|
"epoch": 8.895397489539748,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.596705796767844e-05,
|
|
"loss": 4.0516,
|
|
"mean_token_accuracy": 0.27618888914585116,
|
|
"num_tokens": 19649625.0,
|
|
"step": 10630
|
|
},
|
|
{
|
|
"entropy": 4.848150587081909,
|
|
"epoch": 8.899581589958158,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.584785240250289e-05,
|
|
"loss": 4.1163,
|
|
"mean_token_accuracy": 0.2729650974273682,
|
|
"num_tokens": 19659054.0,
|
|
"step": 10635
|
|
},
|
|
{
|
|
"entropy": 4.788381671905517,
|
|
"epoch": 8.903765690376568,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.572907723882073e-05,
|
|
"loss": 4.0228,
|
|
"mean_token_accuracy": 0.27885275334119797,
|
|
"num_tokens": 19667966.0,
|
|
"step": 10640
|
|
},
|
|
{
|
|
"entropy": 4.844439268112183,
|
|
"epoch": 8.907949790794978,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.561073272105213e-05,
|
|
"loss": 4.1475,
|
|
"mean_token_accuracy": 0.2567688822746277,
|
|
"num_tokens": 19678055.0,
|
|
"step": 10645
|
|
},
|
|
{
|
|
"entropy": 4.843932056427002,
|
|
"epoch": 8.91213389121339,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 6.549281909273113e-05,
|
|
"loss": 4.0833,
|
|
"mean_token_accuracy": 0.276842825114727,
|
|
"num_tokens": 19687540.0,
|
|
"step": 10650
|
|
},
|
|
{
|
|
"entropy": 4.8460324764251705,
|
|
"epoch": 8.9163179916318,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.537533659650497e-05,
|
|
"loss": 4.063,
|
|
"mean_token_accuracy": 0.2794926047325134,
|
|
"num_tokens": 19697688.0,
|
|
"step": 10655
|
|
},
|
|
{
|
|
"entropy": 4.7900238037109375,
|
|
"epoch": 8.92050209205021,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.525828547413374e-05,
|
|
"loss": 4.0454,
|
|
"mean_token_accuracy": 0.28314405083656313,
|
|
"num_tokens": 19707031.0,
|
|
"step": 10660
|
|
},
|
|
{
|
|
"entropy": 4.873225545883178,
|
|
"epoch": 8.92468619246862,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 6.514166596648987e-05,
|
|
"loss": 4.1038,
|
|
"mean_token_accuracy": 0.27146571278572085,
|
|
"num_tokens": 19716285.0,
|
|
"step": 10665
|
|
},
|
|
{
|
|
"entropy": 4.840128326416016,
|
|
"epoch": 8.92887029288703,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.502547831355753e-05,
|
|
"loss": 4.0979,
|
|
"mean_token_accuracy": 0.2723117917776108,
|
|
"num_tokens": 19725766.0,
|
|
"step": 10670
|
|
},
|
|
{
|
|
"entropy": 4.851867818832398,
|
|
"epoch": 8.93305439330544,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.490972275443226e-05,
|
|
"loss": 4.1065,
|
|
"mean_token_accuracy": 0.26163311302661896,
|
|
"num_tokens": 19734908.0,
|
|
"step": 10675
|
|
},
|
|
{
|
|
"entropy": 4.78533296585083,
|
|
"epoch": 8.93723849372385,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.479439952732036e-05,
|
|
"loss": 4.0732,
|
|
"mean_token_accuracy": 0.2755788594484329,
|
|
"num_tokens": 19743698.0,
|
|
"step": 10680
|
|
},
|
|
{
|
|
"entropy": 4.768015718460083,
|
|
"epoch": 8.94142259414226,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.467950886953845e-05,
|
|
"loss": 3.9727,
|
|
"mean_token_accuracy": 0.2803987398743629,
|
|
"num_tokens": 19753667.0,
|
|
"step": 10685
|
|
},
|
|
{
|
|
"entropy": 4.814316368103027,
|
|
"epoch": 8.94560669456067,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 6.456505101751312e-05,
|
|
"loss": 4.0397,
|
|
"mean_token_accuracy": 0.28699344098567964,
|
|
"num_tokens": 19762829.0,
|
|
"step": 10690
|
|
},
|
|
{
|
|
"entropy": 4.827935838699341,
|
|
"epoch": 8.94979079497908,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.445102620678015e-05,
|
|
"loss": 4.1004,
|
|
"mean_token_accuracy": 0.2764452427625656,
|
|
"num_tokens": 19771782.0,
|
|
"step": 10695
|
|
},
|
|
{
|
|
"entropy": 4.782147789001465,
|
|
"epoch": 8.95397489539749,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 6.433743467198432e-05,
|
|
"loss": 4.0712,
|
|
"mean_token_accuracy": 0.2824526995420456,
|
|
"num_tokens": 19781395.0,
|
|
"step": 10700
|
|
},
|
|
{
|
|
"entropy": 4.871569061279297,
|
|
"epoch": 8.9581589958159,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 6.422427664687875e-05,
|
|
"loss": 4.1481,
|
|
"mean_token_accuracy": 0.2688703387975693,
|
|
"num_tokens": 19790329.0,
|
|
"step": 10705
|
|
},
|
|
{
|
|
"entropy": 4.817228937149048,
|
|
"epoch": 8.96234309623431,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 6.411155236432442e-05,
|
|
"loss": 4.111,
|
|
"mean_token_accuracy": 0.2720699071884155,
|
|
"num_tokens": 19799453.0,
|
|
"step": 10710
|
|
},
|
|
{
|
|
"entropy": 4.824449157714843,
|
|
"epoch": 8.96652719665272,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.399926205628983e-05,
|
|
"loss": 4.0555,
|
|
"mean_token_accuracy": 0.27237731516361235,
|
|
"num_tokens": 19808100.0,
|
|
"step": 10715
|
|
},
|
|
{
|
|
"entropy": 4.873416328430176,
|
|
"epoch": 8.97071129707113,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.388740595385031e-05,
|
|
"loss": 4.1264,
|
|
"mean_token_accuracy": 0.2709594964981079,
|
|
"num_tokens": 19817105.0,
|
|
"step": 10720
|
|
},
|
|
{
|
|
"entropy": 4.763794231414795,
|
|
"epoch": 8.97489539748954,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.377598428718783e-05,
|
|
"loss": 4.0522,
|
|
"mean_token_accuracy": 0.2716932699084282,
|
|
"num_tokens": 19826501.0,
|
|
"step": 10725
|
|
},
|
|
{
|
|
"entropy": 4.890101671218872,
|
|
"epoch": 8.97907949790795,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 6.366499728559015e-05,
|
|
"loss": 4.1079,
|
|
"mean_token_accuracy": 0.26831778287887575,
|
|
"num_tokens": 19836169.0,
|
|
"step": 10730
|
|
},
|
|
{
|
|
"entropy": 4.803574800491333,
|
|
"epoch": 8.98326359832636,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.355444517745079e-05,
|
|
"loss": 4.0567,
|
|
"mean_token_accuracy": 0.27670869082212446,
|
|
"num_tokens": 19845403.0,
|
|
"step": 10735
|
|
},
|
|
{
|
|
"entropy": 4.803315830230713,
|
|
"epoch": 8.98744769874477,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.344432819026813e-05,
|
|
"loss": 4.0518,
|
|
"mean_token_accuracy": 0.2766092747449875,
|
|
"num_tokens": 19854280.0,
|
|
"step": 10740
|
|
},
|
|
{
|
|
"entropy": 4.800844192504883,
|
|
"epoch": 8.99163179916318,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.333464655064527e-05,
|
|
"loss": 4.0449,
|
|
"mean_token_accuracy": 0.281219582259655,
|
|
"num_tokens": 19863889.0,
|
|
"step": 10745
|
|
},
|
|
{
|
|
"entropy": 4.819407224655151,
|
|
"epoch": 8.99581589958159,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.322540048428944e-05,
|
|
"loss": 4.0872,
|
|
"mean_token_accuracy": 0.2701942265033722,
|
|
"num_tokens": 19874114.0,
|
|
"step": 10750
|
|
},
|
|
{
|
|
"entropy": 4.895250558853149,
|
|
"epoch": 9.0,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.311659021601137e-05,
|
|
"loss": 4.1169,
|
|
"mean_token_accuracy": 0.27526594400405885,
|
|
"num_tokens": 19883898.0,
|
|
"step": 10755
|
|
},
|
|
{
|
|
"entropy": 4.8633805274963375,
|
|
"epoch": 9.00418410041841,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.300821596972517e-05,
|
|
"loss": 4.0212,
|
|
"mean_token_accuracy": 0.2839258432388306,
|
|
"num_tokens": 19892623.0,
|
|
"step": 10760
|
|
},
|
|
{
|
|
"entropy": 4.785554599761963,
|
|
"epoch": 9.00836820083682,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.290027796844753e-05,
|
|
"loss": 4.0309,
|
|
"mean_token_accuracy": 0.28394087255001066,
|
|
"num_tokens": 19902299.0,
|
|
"step": 10765
|
|
},
|
|
{
|
|
"entropy": 4.7978143215179445,
|
|
"epoch": 9.01255230125523,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.279277643429761e-05,
|
|
"loss": 4.1025,
|
|
"mean_token_accuracy": 0.26973499953746793,
|
|
"num_tokens": 19911554.0,
|
|
"step": 10770
|
|
},
|
|
{
|
|
"entropy": 4.738802862167359,
|
|
"epoch": 9.01673640167364,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 6.268571158849624e-05,
|
|
"loss": 3.9426,
|
|
"mean_token_accuracy": 0.2883836627006531,
|
|
"num_tokens": 19920207.0,
|
|
"step": 10775
|
|
},
|
|
{
|
|
"entropy": 4.791349744796753,
|
|
"epoch": 9.02092050209205,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 6.257908365136571e-05,
|
|
"loss": 4.0197,
|
|
"mean_token_accuracy": 0.2778959333896637,
|
|
"num_tokens": 19929507.0,
|
|
"step": 10780
|
|
},
|
|
{
|
|
"entropy": 4.842765426635742,
|
|
"epoch": 9.02510460251046,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.247289284232908e-05,
|
|
"loss": 4.0749,
|
|
"mean_token_accuracy": 0.2702461123466492,
|
|
"num_tokens": 19938159.0,
|
|
"step": 10785
|
|
},
|
|
{
|
|
"entropy": 4.823717880249023,
|
|
"epoch": 9.02928870292887,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.236713937990998e-05,
|
|
"loss": 4.0067,
|
|
"mean_token_accuracy": 0.28211130797863004,
|
|
"num_tokens": 19947212.0,
|
|
"step": 10790
|
|
},
|
|
{
|
|
"entropy": 4.8320025444030765,
|
|
"epoch": 9.03347280334728,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.22618234817321e-05,
|
|
"loss": 4.0844,
|
|
"mean_token_accuracy": 0.27379921823740005,
|
|
"num_tokens": 19955770.0,
|
|
"step": 10795
|
|
},
|
|
{
|
|
"entropy": 4.815884256362915,
|
|
"epoch": 9.03765690376569,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 6.215694536451855e-05,
|
|
"loss": 4.0893,
|
|
"mean_token_accuracy": 0.27252263724803927,
|
|
"num_tokens": 19965038.0,
|
|
"step": 10800
|
|
},
|
|
{
|
|
"entropy": 4.783570957183838,
|
|
"epoch": 9.0418410041841,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.205250524409166e-05,
|
|
"loss": 4.0015,
|
|
"mean_token_accuracy": 0.2852965176105499,
|
|
"num_tokens": 19974410.0,
|
|
"step": 10805
|
|
},
|
|
{
|
|
"entropy": 4.767372798919678,
|
|
"epoch": 9.04602510460251,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.194850333537244e-05,
|
|
"loss": 3.9827,
|
|
"mean_token_accuracy": 0.28792745172977446,
|
|
"num_tokens": 19983845.0,
|
|
"step": 10810
|
|
},
|
|
{
|
|
"entropy": 4.767030048370361,
|
|
"epoch": 9.05020920502092,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.184493985238008e-05,
|
|
"loss": 3.904,
|
|
"mean_token_accuracy": 0.2919979810714722,
|
|
"num_tokens": 19992438.0,
|
|
"step": 10815
|
|
},
|
|
{
|
|
"entropy": 4.865966176986694,
|
|
"epoch": 9.05439330543933,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 6.174181500823156e-05,
|
|
"loss": 4.0803,
|
|
"mean_token_accuracy": 0.2713268846273422,
|
|
"num_tokens": 20001926.0,
|
|
"step": 10820
|
|
},
|
|
{
|
|
"entropy": 4.85145673751831,
|
|
"epoch": 9.05857740585774,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.16391290151412e-05,
|
|
"loss": 4.1447,
|
|
"mean_token_accuracy": 0.2741641864180565,
|
|
"num_tokens": 20011518.0,
|
|
"step": 10825
|
|
},
|
|
{
|
|
"entropy": 4.820166683197021,
|
|
"epoch": 9.06276150627615,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 6.153688208442032e-05,
|
|
"loss": 4.0082,
|
|
"mean_token_accuracy": 0.2761981725692749,
|
|
"num_tokens": 20020923.0,
|
|
"step": 10830
|
|
},
|
|
{
|
|
"entropy": 4.857508563995362,
|
|
"epoch": 9.06694560669456,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 6.143507442647668e-05,
|
|
"loss": 4.0836,
|
|
"mean_token_accuracy": 0.26819988191127775,
|
|
"num_tokens": 20031174.0,
|
|
"step": 10835
|
|
},
|
|
{
|
|
"entropy": 4.802803230285645,
|
|
"epoch": 9.07112970711297,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.133370625081399e-05,
|
|
"loss": 4.0606,
|
|
"mean_token_accuracy": 0.2732583612203598,
|
|
"num_tokens": 20040392.0,
|
|
"step": 10840
|
|
},
|
|
{
|
|
"entropy": 4.8468687534332275,
|
|
"epoch": 9.07531380753138,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.123277776603175e-05,
|
|
"loss": 4.0164,
|
|
"mean_token_accuracy": 0.2809222102165222,
|
|
"num_tokens": 20050027.0,
|
|
"step": 10845
|
|
},
|
|
{
|
|
"entropy": 4.889333820343017,
|
|
"epoch": 9.07949790794979,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.11322891798245e-05,
|
|
"loss": 4.0662,
|
|
"mean_token_accuracy": 0.2839016497135162,
|
|
"num_tokens": 20060652.0,
|
|
"step": 10850
|
|
},
|
|
{
|
|
"entropy": 4.681153869628906,
|
|
"epoch": 9.0836820083682,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.103224069898161e-05,
|
|
"loss": 3.8358,
|
|
"mean_token_accuracy": 0.29734522849321365,
|
|
"num_tokens": 20069193.0,
|
|
"step": 10855
|
|
},
|
|
{
|
|
"entropy": 4.789216327667236,
|
|
"epoch": 9.087866108786612,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.093263252938674e-05,
|
|
"loss": 4.0288,
|
|
"mean_token_accuracy": 0.2784813463687897,
|
|
"num_tokens": 20078524.0,
|
|
"step": 10860
|
|
},
|
|
{
|
|
"entropy": 4.8343981266021725,
|
|
"epoch": 9.092050209205022,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.0833464876017535e-05,
|
|
"loss": 4.0724,
|
|
"mean_token_accuracy": 0.27119586765766146,
|
|
"num_tokens": 20087746.0,
|
|
"step": 10865
|
|
},
|
|
{
|
|
"entropy": 4.810818719863891,
|
|
"epoch": 9.096234309623432,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 6.073473794294505e-05,
|
|
"loss": 4.0547,
|
|
"mean_token_accuracy": 0.2870690137147903,
|
|
"num_tokens": 20096958.0,
|
|
"step": 10870
|
|
},
|
|
{
|
|
"entropy": 4.863609790802002,
|
|
"epoch": 9.100418410041842,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 6.063645193333345e-05,
|
|
"loss": 4.0602,
|
|
"mean_token_accuracy": 0.27343924939632414,
|
|
"num_tokens": 20106648.0,
|
|
"step": 10875
|
|
},
|
|
{
|
|
"entropy": 4.806730604171753,
|
|
"epoch": 9.104602510460252,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 6.053860704943956e-05,
|
|
"loss": 3.9759,
|
|
"mean_token_accuracy": 0.2846586793661118,
|
|
"num_tokens": 20116136.0,
|
|
"step": 10880
|
|
},
|
|
{
|
|
"entropy": 4.810272645950318,
|
|
"epoch": 9.108786610878662,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 6.044120349261242e-05,
|
|
"loss": 4.0712,
|
|
"mean_token_accuracy": 0.28050764203071593,
|
|
"num_tokens": 20125652.0,
|
|
"step": 10885
|
|
},
|
|
{
|
|
"entropy": 4.842195510864258,
|
|
"epoch": 9.112970711297072,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 6.0344241463292805e-05,
|
|
"loss": 4.0046,
|
|
"mean_token_accuracy": 0.28512563109397887,
|
|
"num_tokens": 20134684.0,
|
|
"step": 10890
|
|
},
|
|
{
|
|
"entropy": 4.768399810791015,
|
|
"epoch": 9.117154811715482,
|
|
"grad_norm": 1.515625,
|
|
"learning_rate": 6.0247721161013096e-05,
|
|
"loss": 3.9784,
|
|
"mean_token_accuracy": 0.2850962907075882,
|
|
"num_tokens": 20143213.0,
|
|
"step": 10895
|
|
},
|
|
{
|
|
"entropy": 4.819011068344116,
|
|
"epoch": 9.121338912133892,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 6.015164278439649e-05,
|
|
"loss": 4.0074,
|
|
"mean_token_accuracy": 0.29256229996681216,
|
|
"num_tokens": 20151420.0,
|
|
"step": 10900
|
|
},
|
|
{
|
|
"entropy": 4.780377197265625,
|
|
"epoch": 9.125523012552302,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 6.005600653115688e-05,
|
|
"loss": 3.9697,
|
|
"mean_token_accuracy": 0.2880295068025589,
|
|
"num_tokens": 20161114.0,
|
|
"step": 10905
|
|
},
|
|
{
|
|
"entropy": 4.808141660690308,
|
|
"epoch": 9.129707112970712,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.996081259809824e-05,
|
|
"loss": 3.9765,
|
|
"mean_token_accuracy": 0.2799105793237686,
|
|
"num_tokens": 20169291.0,
|
|
"step": 10910
|
|
},
|
|
{
|
|
"entropy": 4.811607408523559,
|
|
"epoch": 9.133891213389122,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 5.986606118111446e-05,
|
|
"loss": 4.066,
|
|
"mean_token_accuracy": 0.2726957738399506,
|
|
"num_tokens": 20179345.0,
|
|
"step": 10915
|
|
},
|
|
{
|
|
"entropy": 4.776149129867553,
|
|
"epoch": 9.138075313807532,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.977175247518868e-05,
|
|
"loss": 3.9756,
|
|
"mean_token_accuracy": 0.2899958550930023,
|
|
"num_tokens": 20188157.0,
|
|
"step": 10920
|
|
},
|
|
{
|
|
"entropy": 4.824333333969117,
|
|
"epoch": 9.142259414225942,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 5.967788667439304e-05,
|
|
"loss": 4.105,
|
|
"mean_token_accuracy": 0.2707114741206169,
|
|
"num_tokens": 20197764.0,
|
|
"step": 10925
|
|
},
|
|
{
|
|
"entropy": 4.797021055221558,
|
|
"epoch": 9.146443514644352,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 5.958446397188828e-05,
|
|
"loss": 4.077,
|
|
"mean_token_accuracy": 0.2750277310609818,
|
|
"num_tokens": 20206332.0,
|
|
"step": 10930
|
|
},
|
|
{
|
|
"entropy": 4.775074291229248,
|
|
"epoch": 9.150627615062762,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.949148455992329e-05,
|
|
"loss": 3.9975,
|
|
"mean_token_accuracy": 0.28161196410655975,
|
|
"num_tokens": 20215344.0,
|
|
"step": 10935
|
|
},
|
|
{
|
|
"entropy": 4.873343801498413,
|
|
"epoch": 9.154811715481172,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 5.939894862983478e-05,
|
|
"loss": 4.113,
|
|
"mean_token_accuracy": 0.27435722947120667,
|
|
"num_tokens": 20225264.0,
|
|
"step": 10940
|
|
},
|
|
{
|
|
"entropy": 4.732141065597534,
|
|
"epoch": 9.158995815899582,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 5.9306856372046794e-05,
|
|
"loss": 3.9315,
|
|
"mean_token_accuracy": 0.3092048346996307,
|
|
"num_tokens": 20234253.0,
|
|
"step": 10945
|
|
},
|
|
{
|
|
"entropy": 4.744606590270996,
|
|
"epoch": 9.163179916317992,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.9215207976070415e-05,
|
|
"loss": 3.9654,
|
|
"mean_token_accuracy": 0.28284602165222167,
|
|
"num_tokens": 20243387.0,
|
|
"step": 10950
|
|
},
|
|
{
|
|
"entropy": 4.843107652664185,
|
|
"epoch": 9.167364016736402,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 5.91240036305033e-05,
|
|
"loss": 4.0348,
|
|
"mean_token_accuracy": 0.2718674123287201,
|
|
"num_tokens": 20252716.0,
|
|
"step": 10955
|
|
},
|
|
{
|
|
"entropy": 4.815390396118164,
|
|
"epoch": 9.171548117154812,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.903324352302927e-05,
|
|
"loss": 4.0443,
|
|
"mean_token_accuracy": 0.2781895101070404,
|
|
"num_tokens": 20261065.0,
|
|
"step": 10960
|
|
},
|
|
{
|
|
"entropy": 4.86800103187561,
|
|
"epoch": 9.175732217573222,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 5.894292784041813e-05,
|
|
"loss": 3.9957,
|
|
"mean_token_accuracy": 0.2897786259651184,
|
|
"num_tokens": 20269183.0,
|
|
"step": 10965
|
|
},
|
|
{
|
|
"entropy": 4.758119916915893,
|
|
"epoch": 9.179916317991632,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 5.885305676852498e-05,
|
|
"loss": 3.936,
|
|
"mean_token_accuracy": 0.292240709066391,
|
|
"num_tokens": 20276898.0,
|
|
"step": 10970
|
|
},
|
|
{
|
|
"entropy": 4.834212446212769,
|
|
"epoch": 9.184100418410042,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.8763630492290036e-05,
|
|
"loss": 4.0793,
|
|
"mean_token_accuracy": 0.27045844197273256,
|
|
"num_tokens": 20286808.0,
|
|
"step": 10975
|
|
},
|
|
{
|
|
"entropy": 4.777300453186035,
|
|
"epoch": 9.188284518828452,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 5.867464919573821e-05,
|
|
"loss": 4.0211,
|
|
"mean_token_accuracy": 0.2754934191703796,
|
|
"num_tokens": 20295783.0,
|
|
"step": 10980
|
|
},
|
|
{
|
|
"entropy": 4.850827217102051,
|
|
"epoch": 9.192468619246862,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 5.85861130619787e-05,
|
|
"loss": 4.0507,
|
|
"mean_token_accuracy": 0.28360946774482726,
|
|
"num_tokens": 20305201.0,
|
|
"step": 10985
|
|
},
|
|
{
|
|
"entropy": 4.796186113357544,
|
|
"epoch": 9.196652719665272,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.849802227320465e-05,
|
|
"loss": 4.0364,
|
|
"mean_token_accuracy": 0.2752275615930557,
|
|
"num_tokens": 20313541.0,
|
|
"step": 10990
|
|
},
|
|
{
|
|
"entropy": 4.77450385093689,
|
|
"epoch": 9.200836820083682,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 5.841037701069271e-05,
|
|
"loss": 3.9715,
|
|
"mean_token_accuracy": 0.2830368548631668,
|
|
"num_tokens": 20322822.0,
|
|
"step": 10995
|
|
},
|
|
{
|
|
"entropy": 4.813671064376831,
|
|
"epoch": 9.205020920502092,
|
|
"grad_norm": 1.546875,
|
|
"learning_rate": 5.8323177454802775e-05,
|
|
"loss": 4.0833,
|
|
"mean_token_accuracy": 0.2801369845867157,
|
|
"num_tokens": 20331966.0,
|
|
"step": 11000
|
|
},
|
|
{
|
|
"epoch": 9.205020920502092,
|
|
"eval_entropy": 4.69530665850103,
|
|
"eval_loss": 5.310837745666504,
|
|
"eval_mean_token_accuracy": 0.20525817600495977,
|
|
"eval_num_tokens": 20331966.0,
|
|
"eval_runtime": 2.7254,
|
|
"eval_samples_per_second": 1433.529,
|
|
"eval_steps_per_second": 179.421,
|
|
"step": 11000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11950,
|
|
"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": 2.9638257205248e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|