10145 lines
276 KiB
JSON
10145 lines
276 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 4.184100418410042,
|
|
"eval_steps": 500,
|
|
"global_step": 5000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 1.3501016451072e+16,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|