4f896eeaf6
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
6101 lines
165 KiB
JSON
6101 lines
165 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.725124943207633,
|
|
"eval_steps": 500,
|
|
"global_step": 3000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 6.768713426589966,
|
|
"epoch": 0.004543389368468878,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 6.7222,
|
|
"mean_token_accuracy": 0.1095728911459446,
|
|
"num_tokens": 10131.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 6.720575475692749,
|
|
"epoch": 0.009086778736937756,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 6.6561,
|
|
"mean_token_accuracy": 0.10461715683341026,
|
|
"num_tokens": 20042.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 6.794066381454468,
|
|
"epoch": 0.013630168105406633,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 6.6777,
|
|
"mean_token_accuracy": 0.10963529199361802,
|
|
"num_tokens": 30596.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 6.780745935440064,
|
|
"epoch": 0.01817355747387551,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 6.6242,
|
|
"mean_token_accuracy": 0.11041242256760597,
|
|
"num_tokens": 40667.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 6.800935125350952,
|
|
"epoch": 0.02271694684234439,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 6.6384,
|
|
"mean_token_accuracy": 0.10581032261252403,
|
|
"num_tokens": 51027.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.829052972793579,
|
|
"epoch": 0.027260336210813266,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 6.7264,
|
|
"mean_token_accuracy": 0.10166022032499314,
|
|
"num_tokens": 61860.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 6.852202463150024,
|
|
"epoch": 0.03180372557928214,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 6.5345,
|
|
"mean_token_accuracy": 0.1121172659099102,
|
|
"num_tokens": 72136.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 6.821422100067139,
|
|
"epoch": 0.03634711494775102,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 6.624,
|
|
"mean_token_accuracy": 0.10981205776333809,
|
|
"num_tokens": 82611.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 6.870565414428711,
|
|
"epoch": 0.0408905043162199,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 6.6909,
|
|
"mean_token_accuracy": 0.10643857344985008,
|
|
"num_tokens": 92722.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 6.841821718215942,
|
|
"epoch": 0.04543389368468878,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10719444304704666,
|
|
"num_tokens": 103258.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 6.887676668167114,
|
|
"epoch": 0.049977283053157656,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 6.7029,
|
|
"mean_token_accuracy": 0.10421370416879654,
|
|
"num_tokens": 114277.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 6.867907333374023,
|
|
"epoch": 0.05452067242162653,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 6.6028,
|
|
"mean_token_accuracy": 0.10567226111888886,
|
|
"num_tokens": 124098.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 6.858053731918335,
|
|
"epoch": 0.059064061790095414,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 6.6126,
|
|
"mean_token_accuracy": 0.11433581411838531,
|
|
"num_tokens": 133775.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 6.852071142196655,
|
|
"epoch": 0.06360745115856428,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 6.6068,
|
|
"mean_token_accuracy": 0.11076592430472373,
|
|
"num_tokens": 144765.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 6.78433780670166,
|
|
"epoch": 0.06815084052703317,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 6.6368,
|
|
"mean_token_accuracy": 0.1169683001935482,
|
|
"num_tokens": 154641.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 6.829514408111573,
|
|
"epoch": 0.07269422989550205,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 6.5928,
|
|
"mean_token_accuracy": 0.11736959144473076,
|
|
"num_tokens": 164728.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 6.903993606567383,
|
|
"epoch": 0.07723761926397092,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 6.5341,
|
|
"mean_token_accuracy": 0.10781509801745415,
|
|
"num_tokens": 174739.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 6.8236547946929935,
|
|
"epoch": 0.0817810086324398,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 6.5946,
|
|
"mean_token_accuracy": 0.1138243854045868,
|
|
"num_tokens": 184580.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 6.850541257858277,
|
|
"epoch": 0.08632439800090867,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 6.6544,
|
|
"mean_token_accuracy": 0.1113918036222458,
|
|
"num_tokens": 195243.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 6.854983806610107,
|
|
"epoch": 0.09086778736937756,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 6.6038,
|
|
"mean_token_accuracy": 0.11426160708069802,
|
|
"num_tokens": 205751.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 6.885841703414917,
|
|
"epoch": 0.09541117673784644,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 6.5497,
|
|
"mean_token_accuracy": 0.11042018011212348,
|
|
"num_tokens": 217283.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 6.744452333450317,
|
|
"epoch": 0.09995456610631531,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 6.51,
|
|
"mean_token_accuracy": 0.11327873021364213,
|
|
"num_tokens": 226909.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 6.87985577583313,
|
|
"epoch": 0.10449795547478419,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 6.5855,
|
|
"mean_token_accuracy": 0.11328148767352104,
|
|
"num_tokens": 237343.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 6.860165071487427,
|
|
"epoch": 0.10904134484325306,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 6.5179,
|
|
"mean_token_accuracy": 0.1166656605899334,
|
|
"num_tokens": 246601.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 6.857304906845092,
|
|
"epoch": 0.11358473421172194,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 6.5727,
|
|
"mean_token_accuracy": 0.11443846747279167,
|
|
"num_tokens": 256818.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 6.8257914066314695,
|
|
"epoch": 0.11812812358019083,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 6.4876,
|
|
"mean_token_accuracy": 0.11914753541350365,
|
|
"num_tokens": 267157.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 6.724686956405639,
|
|
"epoch": 0.1226715129486597,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 6.4988,
|
|
"mean_token_accuracy": 0.11270106509327889,
|
|
"num_tokens": 277370.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 6.806635904312134,
|
|
"epoch": 0.12721490231712856,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 6.5163,
|
|
"mean_token_accuracy": 0.11316018253564834,
|
|
"num_tokens": 287851.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 6.965766048431396,
|
|
"epoch": 0.13175829168559747,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 6.6036,
|
|
"mean_token_accuracy": 0.11733311712741852,
|
|
"num_tokens": 299043.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 6.744367074966431,
|
|
"epoch": 0.13630168105406634,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 6.5794,
|
|
"mean_token_accuracy": 0.11460703909397126,
|
|
"num_tokens": 308525.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 6.8237261295318605,
|
|
"epoch": 0.14084507042253522,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 6.5493,
|
|
"mean_token_accuracy": 0.11844029426574706,
|
|
"num_tokens": 318301.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 6.790290117263794,
|
|
"epoch": 0.1453884597910041,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 6.6009,
|
|
"mean_token_accuracy": 0.11239130273461342,
|
|
"num_tokens": 329599.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 6.843774080276489,
|
|
"epoch": 0.14993184915947297,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 6.4774,
|
|
"mean_token_accuracy": 0.11909914687275887,
|
|
"num_tokens": 339446.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 6.9367317199707035,
|
|
"epoch": 0.15447523852794184,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 6.6333,
|
|
"mean_token_accuracy": 0.1179355539381504,
|
|
"num_tokens": 349579.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 6.8179216384887695,
|
|
"epoch": 0.15901862789641072,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 6.5631,
|
|
"mean_token_accuracy": 0.11152750179171562,
|
|
"num_tokens": 359452.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 6.830770826339721,
|
|
"epoch": 0.1635620172648796,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 6.5702,
|
|
"mean_token_accuracy": 0.11352085247635842,
|
|
"num_tokens": 369558.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 6.878524255752564,
|
|
"epoch": 0.16810540663334847,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 6.5577,
|
|
"mean_token_accuracy": 0.11322271078824997,
|
|
"num_tokens": 380458.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 6.9294568538665775,
|
|
"epoch": 0.17264879600181735,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 6.5669,
|
|
"mean_token_accuracy": 0.1112134151160717,
|
|
"num_tokens": 390051.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 6.777473735809326,
|
|
"epoch": 0.17719218537028622,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 6.5164,
|
|
"mean_token_accuracy": 0.11806451380252839,
|
|
"num_tokens": 400343.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 6.889895868301392,
|
|
"epoch": 0.18173557473875512,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 6.5619,
|
|
"mean_token_accuracy": 0.11565203443169594,
|
|
"num_tokens": 410004.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 6.7963909149169925,
|
|
"epoch": 0.186278964107224,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 6.5851,
|
|
"mean_token_accuracy": 0.10973369553685189,
|
|
"num_tokens": 421286.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 6.9212535381317135,
|
|
"epoch": 0.19082235347569287,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 6.6359,
|
|
"mean_token_accuracy": 0.10912902057170867,
|
|
"num_tokens": 432351.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 6.842854261398315,
|
|
"epoch": 0.19536574284416175,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 6.627,
|
|
"mean_token_accuracy": 0.11367111280560493,
|
|
"num_tokens": 442843.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 6.870680522918701,
|
|
"epoch": 0.19990913221263062,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 6.4973,
|
|
"mean_token_accuracy": 0.11255754232406616,
|
|
"num_tokens": 453370.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 6.800969171524048,
|
|
"epoch": 0.2044525215810995,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10846318081021308,
|
|
"num_tokens": 464715.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 6.875910711288452,
|
|
"epoch": 0.20899591094956838,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 6.4927,
|
|
"mean_token_accuracy": 0.11165988445281982,
|
|
"num_tokens": 475086.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 6.743642044067383,
|
|
"epoch": 0.21353930031803725,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 6.6147,
|
|
"mean_token_accuracy": 0.12273732423782349,
|
|
"num_tokens": 486088.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 6.893033027648926,
|
|
"epoch": 0.21808268968650613,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 6.563,
|
|
"mean_token_accuracy": 0.10623423308134079,
|
|
"num_tokens": 496978.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 6.730010509490967,
|
|
"epoch": 0.222626079054975,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 6.3713,
|
|
"mean_token_accuracy": 0.12526726275682448,
|
|
"num_tokens": 507794.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 6.739193630218506,
|
|
"epoch": 0.22716946842344388,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 6.5377,
|
|
"mean_token_accuracy": 0.10970785096287727,
|
|
"num_tokens": 517763.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 6.85530915260315,
|
|
"epoch": 0.23171285779191278,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 6.4664,
|
|
"mean_token_accuracy": 0.11921982616186141,
|
|
"num_tokens": 527374.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 6.741254997253418,
|
|
"epoch": 0.23625624716038165,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 6.5256,
|
|
"mean_token_accuracy": 0.11517209336161613,
|
|
"num_tokens": 537435.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 6.810067796707154,
|
|
"epoch": 0.24079963652885053,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 6.4963,
|
|
"mean_token_accuracy": 0.11652736142277717,
|
|
"num_tokens": 547236.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 6.788457632064819,
|
|
"epoch": 0.2453430258973194,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 6.4943,
|
|
"mean_token_accuracy": 0.11792855635285378,
|
|
"num_tokens": 557130.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 6.698097038269043,
|
|
"epoch": 0.24988641526578828,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 6.5301,
|
|
"mean_token_accuracy": 0.11694961413741112,
|
|
"num_tokens": 567482.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 6.759066724777222,
|
|
"epoch": 0.25442980463425713,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 6.3925,
|
|
"mean_token_accuracy": 0.11411905512213708,
|
|
"num_tokens": 577084.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 6.76326265335083,
|
|
"epoch": 0.25897319400272606,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 6.5115,
|
|
"mean_token_accuracy": 0.11677431538701058,
|
|
"num_tokens": 587980.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 6.667721605300903,
|
|
"epoch": 0.26351658337119493,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 6.4219,
|
|
"mean_token_accuracy": 0.12288179770112037,
|
|
"num_tokens": 598989.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 6.985043001174927,
|
|
"epoch": 0.2680599727396638,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 6.6076,
|
|
"mean_token_accuracy": 0.11418735831975937,
|
|
"num_tokens": 608599.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 6.601436614990234,
|
|
"epoch": 0.2726033621081327,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 6.4418,
|
|
"mean_token_accuracy": 0.11823201850056649,
|
|
"num_tokens": 619267.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 6.909968709945678,
|
|
"epoch": 0.27714675147660156,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000152,
|
|
"loss": 6.6386,
|
|
"mean_token_accuracy": 0.12054958641529083,
|
|
"num_tokens": 629991.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 6.833962726593017,
|
|
"epoch": 0.28169014084507044,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 6.5693,
|
|
"mean_token_accuracy": 0.11267199367284775,
|
|
"num_tokens": 641311.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 6.812391519546509,
|
|
"epoch": 0.2862335302135393,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 6.5406,
|
|
"mean_token_accuracy": 0.11235396489501,
|
|
"num_tokens": 651812.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 6.799654388427735,
|
|
"epoch": 0.2907769195820082,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 6.5081,
|
|
"mean_token_accuracy": 0.11410669162869454,
|
|
"num_tokens": 662387.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 6.780321025848389,
|
|
"epoch": 0.29532030895047706,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 6.4997,
|
|
"mean_token_accuracy": 0.11494562402367592,
|
|
"num_tokens": 673018.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 6.777340126037598,
|
|
"epoch": 0.29986369831894594,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 6.434,
|
|
"mean_token_accuracy": 0.12099529802799225,
|
|
"num_tokens": 683192.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 6.75422306060791,
|
|
"epoch": 0.3044070876874148,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 6.5675,
|
|
"mean_token_accuracy": 0.1253136858344078,
|
|
"num_tokens": 692747.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 6.676179361343384,
|
|
"epoch": 0.3089504770558837,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 6.5389,
|
|
"mean_token_accuracy": 0.11221015229821205,
|
|
"num_tokens": 705109.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 6.835248422622681,
|
|
"epoch": 0.31349386642435256,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.11584834307432175,
|
|
"num_tokens": 715658.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 6.698165464401245,
|
|
"epoch": 0.31803725579282144,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 6.4818,
|
|
"mean_token_accuracy": 0.1224296160042286,
|
|
"num_tokens": 724961.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 6.6860706329345705,
|
|
"epoch": 0.3225806451612903,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.1249639131128788,
|
|
"num_tokens": 735416.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 6.823048305511475,
|
|
"epoch": 0.3271240345297592,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 6.5408,
|
|
"mean_token_accuracy": 0.12204954698681832,
|
|
"num_tokens": 745005.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 6.874798774719238,
|
|
"epoch": 0.33166742389822806,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000182,
|
|
"loss": 6.5836,
|
|
"mean_token_accuracy": 0.11828166246414185,
|
|
"num_tokens": 754743.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 6.583785772323608,
|
|
"epoch": 0.33621081326669694,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 6.437,
|
|
"mean_token_accuracy": 0.1200392097234726,
|
|
"num_tokens": 764897.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 6.818075704574585,
|
|
"epoch": 0.3407542026351658,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000187,
|
|
"loss": 6.511,
|
|
"mean_token_accuracy": 0.11428396552801132,
|
|
"num_tokens": 775302.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 6.651220989227295,
|
|
"epoch": 0.3452975920036347,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 6.4495,
|
|
"mean_token_accuracy": 0.12070777788758277,
|
|
"num_tokens": 785694.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 6.732499074935913,
|
|
"epoch": 0.34984098137210357,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 6.3957,
|
|
"mean_token_accuracy": 0.12520513609051703,
|
|
"num_tokens": 795045.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 6.696088409423828,
|
|
"epoch": 0.35438437074057244,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 6.4921,
|
|
"mean_token_accuracy": 0.1158549316227436,
|
|
"num_tokens": 805579.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 6.654592990875244,
|
|
"epoch": 0.35892776010904137,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.3914,
|
|
"mean_token_accuracy": 0.12176291272044182,
|
|
"num_tokens": 816743.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 6.768208026885986,
|
|
"epoch": 0.36347114947751025,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 6.5313,
|
|
"mean_token_accuracy": 0.11511081382632256,
|
|
"num_tokens": 827217.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 6.615100955963134,
|
|
"epoch": 0.3680145388459791,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 6.3835,
|
|
"mean_token_accuracy": 0.1261222928762436,
|
|
"num_tokens": 837555.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 6.711054515838623,
|
|
"epoch": 0.372557928214448,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.4678,
|
|
"mean_token_accuracy": 0.11874150708317757,
|
|
"num_tokens": 846726.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 6.75947527885437,
|
|
"epoch": 0.3771013175829169,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.4571,
|
|
"mean_token_accuracy": 0.12268495261669159,
|
|
"num_tokens": 856535.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 6.683750581741333,
|
|
"epoch": 0.38164470695138575,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.1186832346022129,
|
|
"num_tokens": 866973.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 6.770697355270386,
|
|
"epoch": 0.3861880963198546,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.4716,
|
|
"mean_token_accuracy": 0.11631686612963676,
|
|
"num_tokens": 877873.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 6.69672966003418,
|
|
"epoch": 0.3907314856883235,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 6.4657,
|
|
"mean_token_accuracy": 0.11843668594956398,
|
|
"num_tokens": 887135.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 6.835273313522339,
|
|
"epoch": 0.3952748750567924,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 6.4843,
|
|
"mean_token_accuracy": 0.1205623060464859,
|
|
"num_tokens": 897206.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 6.631858015060425,
|
|
"epoch": 0.39981826442526125,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.4857,
|
|
"mean_token_accuracy": 0.1176255464553833,
|
|
"num_tokens": 906808.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 6.630966377258301,
|
|
"epoch": 0.4043616537937301,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 6.3833,
|
|
"mean_token_accuracy": 0.1273773953318596,
|
|
"num_tokens": 917177.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 6.6929981231689455,
|
|
"epoch": 0.408905043162199,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.4283,
|
|
"mean_token_accuracy": 0.11764879748225213,
|
|
"num_tokens": 927229.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 6.704322004318238,
|
|
"epoch": 0.4134484325306679,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 6.4261,
|
|
"mean_token_accuracy": 0.12415409311652184,
|
|
"num_tokens": 937832.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 6.592918109893799,
|
|
"epoch": 0.41799182189913675,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.3477,
|
|
"mean_token_accuracy": 0.12181811258196831,
|
|
"num_tokens": 948038.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 6.663812637329102,
|
|
"epoch": 0.4225352112676056,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.4591,
|
|
"mean_token_accuracy": 0.11563793793320656,
|
|
"num_tokens": 957924.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 6.706633520126343,
|
|
"epoch": 0.4270786006360745,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.4292,
|
|
"mean_token_accuracy": 0.1167376309633255,
|
|
"num_tokens": 968497.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 6.624789667129517,
|
|
"epoch": 0.4316219900045434,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.3514,
|
|
"mean_token_accuracy": 0.12173967584967613,
|
|
"num_tokens": 978878.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 6.5878393173217775,
|
|
"epoch": 0.43616537937301225,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.4639,
|
|
"mean_token_accuracy": 0.12137211859226227,
|
|
"num_tokens": 989984.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 6.6825902462005615,
|
|
"epoch": 0.4407087687414811,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.3862,
|
|
"mean_token_accuracy": 0.12439708858728409,
|
|
"num_tokens": 1000254.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 6.580802011489868,
|
|
"epoch": 0.44525215810995,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.4602,
|
|
"mean_token_accuracy": 0.11807657480239868,
|
|
"num_tokens": 1010697.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 6.8235663890838625,
|
|
"epoch": 0.4497955474784189,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.5558,
|
|
"mean_token_accuracy": 0.1186821848154068,
|
|
"num_tokens": 1020892.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 6.656906843185425,
|
|
"epoch": 0.45433893684688775,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 6.4358,
|
|
"mean_token_accuracy": 0.1271519921720028,
|
|
"num_tokens": 1030380.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.45433893684688775,
|
|
"eval_entropy": 6.70977929470151,
|
|
"eval_loss": 6.51187801361084,
|
|
"eval_mean_token_accuracy": 0.11979533588990222,
|
|
"eval_num_tokens": 1030380.0,
|
|
"eval_runtime": 2.6058,
|
|
"eval_samples_per_second": 1318.216,
|
|
"eval_steps_per_second": 165.017,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 6.737278318405151,
|
|
"epoch": 0.45888232621535663,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.4798,
|
|
"mean_token_accuracy": 0.11720329746603966,
|
|
"num_tokens": 1040301.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 6.551387023925781,
|
|
"epoch": 0.46342571558382556,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.3367,
|
|
"mean_token_accuracy": 0.12788272872567177,
|
|
"num_tokens": 1051999.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 6.841769361495972,
|
|
"epoch": 0.46796910495229443,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.11753517240285874,
|
|
"num_tokens": 1063133.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 6.560217952728271,
|
|
"epoch": 0.4725124943207633,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.3922,
|
|
"mean_token_accuracy": 0.11554303765296936,
|
|
"num_tokens": 1073972.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 6.600145530700684,
|
|
"epoch": 0.4770558836892322,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.3869,
|
|
"mean_token_accuracy": 0.11973805353045464,
|
|
"num_tokens": 1084050.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 6.673818635940552,
|
|
"epoch": 0.48159927305770106,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.4799,
|
|
"mean_token_accuracy": 0.12647488638758658,
|
|
"num_tokens": 1094415.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 6.534582090377808,
|
|
"epoch": 0.48614266242616994,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.4305,
|
|
"mean_token_accuracy": 0.11784208044409752,
|
|
"num_tokens": 1105133.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 6.662813949584961,
|
|
"epoch": 0.4906860517946388,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.3911,
|
|
"mean_token_accuracy": 0.12304398790001869,
|
|
"num_tokens": 1115266.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 6.537495183944702,
|
|
"epoch": 0.4952294411631077,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.3602,
|
|
"mean_token_accuracy": 0.12327887490391731,
|
|
"num_tokens": 1124775.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 6.56860499382019,
|
|
"epoch": 0.49977283053157656,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.3894,
|
|
"mean_token_accuracy": 0.12164504677057267,
|
|
"num_tokens": 1134498.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 6.696820211410523,
|
|
"epoch": 0.5043162199000454,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.4295,
|
|
"mean_token_accuracy": 0.11844469383358955,
|
|
"num_tokens": 1146067.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 6.667341804504394,
|
|
"epoch": 0.5088596092685143,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.4374,
|
|
"mean_token_accuracy": 0.11883379444479943,
|
|
"num_tokens": 1156349.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 6.60188570022583,
|
|
"epoch": 0.5134029986369832,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.3706,
|
|
"mean_token_accuracy": 0.1226509042084217,
|
|
"num_tokens": 1166851.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 6.5608902931213375,
|
|
"epoch": 0.5179463880054521,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.3262,
|
|
"mean_token_accuracy": 0.12422406151890755,
|
|
"num_tokens": 1176899.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 6.576656770706177,
|
|
"epoch": 0.522489777373921,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.4297,
|
|
"mean_token_accuracy": 0.1186425231397152,
|
|
"num_tokens": 1187383.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 6.718804931640625,
|
|
"epoch": 0.5270331667423899,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.4153,
|
|
"mean_token_accuracy": 0.12464964166283607,
|
|
"num_tokens": 1197836.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 6.5609543800354,
|
|
"epoch": 0.5315765561108587,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.3661,
|
|
"mean_token_accuracy": 0.1282975621521473,
|
|
"num_tokens": 1208533.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 6.60496711730957,
|
|
"epoch": 0.5361199454793276,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.4107,
|
|
"mean_token_accuracy": 0.11816166192293168,
|
|
"num_tokens": 1219397.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 6.6025289535522464,
|
|
"epoch": 0.5406633348477965,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.4063,
|
|
"mean_token_accuracy": 0.11670600771903991,
|
|
"num_tokens": 1229991.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 6.493481159210205,
|
|
"epoch": 0.5452067242162654,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.2923,
|
|
"mean_token_accuracy": 0.11882022395730019,
|
|
"num_tokens": 1240343.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 6.529299545288086,
|
|
"epoch": 0.5497501135847342,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.3761,
|
|
"mean_token_accuracy": 0.11741307675838471,
|
|
"num_tokens": 1250432.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 6.5340169906616214,
|
|
"epoch": 0.5542935029532031,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.3185,
|
|
"mean_token_accuracy": 0.12538510411977768,
|
|
"num_tokens": 1260744.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 6.600591993331909,
|
|
"epoch": 0.558836892321672,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.339,
|
|
"mean_token_accuracy": 0.12402234598994255,
|
|
"num_tokens": 1272155.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 6.398069381713867,
|
|
"epoch": 0.5633802816901409,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.2189,
|
|
"mean_token_accuracy": 0.12761275619268417,
|
|
"num_tokens": 1282240.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.573273801803589,
|
|
"epoch": 0.5679236710586097,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.4098,
|
|
"mean_token_accuracy": 0.12361594140529633,
|
|
"num_tokens": 1292169.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 6.485548496246338,
|
|
"epoch": 0.5724670604270786,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.3056,
|
|
"mean_token_accuracy": 0.12872742787003516,
|
|
"num_tokens": 1302010.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 6.528713846206665,
|
|
"epoch": 0.5770104497955475,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.365,
|
|
"mean_token_accuracy": 0.12618476673960685,
|
|
"num_tokens": 1312596.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 6.589738416671753,
|
|
"epoch": 0.5815538391640164,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.313,
|
|
"mean_token_accuracy": 0.11931094974279403,
|
|
"num_tokens": 1321753.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 6.462535858154297,
|
|
"epoch": 0.5860972285324852,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.356,
|
|
"mean_token_accuracy": 0.12229031547904015,
|
|
"num_tokens": 1332059.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 6.533907556533814,
|
|
"epoch": 0.5906406179009541,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.3321,
|
|
"mean_token_accuracy": 0.1260114513337612,
|
|
"num_tokens": 1341517.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 6.640655851364135,
|
|
"epoch": 0.595184007269423,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.4102,
|
|
"mean_token_accuracy": 0.12594819515943528,
|
|
"num_tokens": 1351439.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 6.483338737487793,
|
|
"epoch": 0.5997273966378919,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.2893,
|
|
"mean_token_accuracy": 0.12250665053725243,
|
|
"num_tokens": 1360745.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.401842403411865,
|
|
"epoch": 0.6042707860063607,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.2989,
|
|
"mean_token_accuracy": 0.1253798320889473,
|
|
"num_tokens": 1371931.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.429254722595215,
|
|
"epoch": 0.6088141753748296,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.2382,
|
|
"mean_token_accuracy": 0.12485149502754211,
|
|
"num_tokens": 1382040.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.451748323440552,
|
|
"epoch": 0.6133575647432985,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.2984,
|
|
"mean_token_accuracy": 0.12093956097960472,
|
|
"num_tokens": 1392690.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.548810720443726,
|
|
"epoch": 0.6179009541117674,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.3009,
|
|
"mean_token_accuracy": 0.12423968389630317,
|
|
"num_tokens": 1403060.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.452625370025634,
|
|
"epoch": 0.6224443434802363,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.315,
|
|
"mean_token_accuracy": 0.124802715331316,
|
|
"num_tokens": 1412895.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.463527679443359,
|
|
"epoch": 0.6269877328487051,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.2829,
|
|
"mean_token_accuracy": 0.1315432645380497,
|
|
"num_tokens": 1423024.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.438167715072632,
|
|
"epoch": 0.631531122217174,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.2903,
|
|
"mean_token_accuracy": 0.12509298026561738,
|
|
"num_tokens": 1433578.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.441053771972657,
|
|
"epoch": 0.6360745115856429,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.3409,
|
|
"mean_token_accuracy": 0.12466292232275009,
|
|
"num_tokens": 1444473.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.530716371536255,
|
|
"epoch": 0.6406179009541118,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.3658,
|
|
"mean_token_accuracy": 0.12251063138246536,
|
|
"num_tokens": 1454553.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.492323493957519,
|
|
"epoch": 0.6451612903225806,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.2283,
|
|
"mean_token_accuracy": 0.12746653333306313,
|
|
"num_tokens": 1464162.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.370441341400147,
|
|
"epoch": 0.6497046796910495,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.2923,
|
|
"mean_token_accuracy": 0.12994937747716903,
|
|
"num_tokens": 1474691.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.465943193435669,
|
|
"epoch": 0.6542480690595184,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.3715,
|
|
"mean_token_accuracy": 0.12509030997753143,
|
|
"num_tokens": 1484650.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.523843812942505,
|
|
"epoch": 0.6587914584279873,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.3244,
|
|
"mean_token_accuracy": 0.12303805649280548,
|
|
"num_tokens": 1494788.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.440337705612182,
|
|
"epoch": 0.6633348477964561,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.2751,
|
|
"mean_token_accuracy": 0.13085415214300156,
|
|
"num_tokens": 1506574.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.369408988952637,
|
|
"epoch": 0.667878237164925,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.2929,
|
|
"mean_token_accuracy": 0.12216815054416656,
|
|
"num_tokens": 1517505.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.457225704193116,
|
|
"epoch": 0.6724216265333939,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.1799,
|
|
"mean_token_accuracy": 0.1351654462516308,
|
|
"num_tokens": 1526662.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.37555251121521,
|
|
"epoch": 0.6769650159018628,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.1286969043314457,
|
|
"num_tokens": 1536798.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.40103816986084,
|
|
"epoch": 0.6815084052703316,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.13174753487110138,
|
|
"num_tokens": 1546673.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.403094959259033,
|
|
"epoch": 0.6860517946388005,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.236,
|
|
"mean_token_accuracy": 0.13387759551405906,
|
|
"num_tokens": 1556793.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.438587808609009,
|
|
"epoch": 0.6905951840072694,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.283,
|
|
"mean_token_accuracy": 0.12869395688176155,
|
|
"num_tokens": 1566450.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.352961111068725,
|
|
"epoch": 0.6951385733757383,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.3079,
|
|
"mean_token_accuracy": 0.13057589232921601,
|
|
"num_tokens": 1576353.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.418825960159301,
|
|
"epoch": 0.6996819627442071,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.192,
|
|
"mean_token_accuracy": 0.12764471173286437,
|
|
"num_tokens": 1586318.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.426858377456665,
|
|
"epoch": 0.704225352112676,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.2869,
|
|
"mean_token_accuracy": 0.12360998913645745,
|
|
"num_tokens": 1596839.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.35253381729126,
|
|
"epoch": 0.7087687414811449,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.3133,
|
|
"mean_token_accuracy": 0.12761991545557977,
|
|
"num_tokens": 1606843.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.324183559417724,
|
|
"epoch": 0.7133121308496138,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.2747,
|
|
"mean_token_accuracy": 0.12435603663325309,
|
|
"num_tokens": 1617048.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.413893175125122,
|
|
"epoch": 0.7178555202180827,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.1888,
|
|
"mean_token_accuracy": 0.13071049749851227,
|
|
"num_tokens": 1627472.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.403914165496826,
|
|
"epoch": 0.7223989095865516,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.2874,
|
|
"mean_token_accuracy": 0.12295314446091651,
|
|
"num_tokens": 1637716.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.384047985076904,
|
|
"epoch": 0.7269422989550205,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.3057,
|
|
"mean_token_accuracy": 0.12159052193164825,
|
|
"num_tokens": 1648119.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.422551822662354,
|
|
"epoch": 0.7314856883234894,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.263,
|
|
"mean_token_accuracy": 0.12927623763680457,
|
|
"num_tokens": 1658214.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.272933769226074,
|
|
"epoch": 0.7360290776919582,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.222,
|
|
"mean_token_accuracy": 0.12815281972289086,
|
|
"num_tokens": 1668725.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.38576192855835,
|
|
"epoch": 0.7405724670604271,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.291,
|
|
"mean_token_accuracy": 0.12685041427612304,
|
|
"num_tokens": 1678891.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.396528625488282,
|
|
"epoch": 0.745115856428896,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.1896,
|
|
"mean_token_accuracy": 0.12780508771538734,
|
|
"num_tokens": 1688749.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.372272872924805,
|
|
"epoch": 0.7496592457973649,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.2922,
|
|
"mean_token_accuracy": 0.12699832171201705,
|
|
"num_tokens": 1699166.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.382475662231445,
|
|
"epoch": 0.7542026351658337,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.2227,
|
|
"mean_token_accuracy": 0.12901799380779266,
|
|
"num_tokens": 1710690.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.378543329238892,
|
|
"epoch": 0.7587460245343026,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.3242,
|
|
"mean_token_accuracy": 0.12391438856720924,
|
|
"num_tokens": 1721169.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.325247430801392,
|
|
"epoch": 0.7632894139027715,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.1868,
|
|
"mean_token_accuracy": 0.12631086334586145,
|
|
"num_tokens": 1731523.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.33087682723999,
|
|
"epoch": 0.7678328032712404,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.1862,
|
|
"mean_token_accuracy": 0.12567417919635773,
|
|
"num_tokens": 1742316.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.3732359409332275,
|
|
"epoch": 0.7723761926397092,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.3318,
|
|
"mean_token_accuracy": 0.12681131660938263,
|
|
"num_tokens": 1753424.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.253160429000855,
|
|
"epoch": 0.7769195820081781,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.1569,
|
|
"mean_token_accuracy": 0.13139303475618364,
|
|
"num_tokens": 1762814.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.360691118240356,
|
|
"epoch": 0.781462971376647,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.3124,
|
|
"mean_token_accuracy": 0.1301097348332405,
|
|
"num_tokens": 1773096.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.37277398109436,
|
|
"epoch": 0.7860063607451159,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.2246,
|
|
"mean_token_accuracy": 0.1275457039475441,
|
|
"num_tokens": 1783499.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.531914615631104,
|
|
"epoch": 0.7905497501135847,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12034643739461899,
|
|
"num_tokens": 1794107.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.342921924591065,
|
|
"epoch": 0.7950931394820536,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.2924,
|
|
"mean_token_accuracy": 0.12306121364235878,
|
|
"num_tokens": 1805151.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.362142753601074,
|
|
"epoch": 0.7996365288505225,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.2709,
|
|
"mean_token_accuracy": 0.12512775808572768,
|
|
"num_tokens": 1815205.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.332495164871216,
|
|
"epoch": 0.8041799182189914,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.2403,
|
|
"mean_token_accuracy": 0.13088379427790642,
|
|
"num_tokens": 1825301.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.311787033081055,
|
|
"epoch": 0.8087233075874602,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.2579,
|
|
"mean_token_accuracy": 0.12949514910578727,
|
|
"num_tokens": 1835634.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.42405252456665,
|
|
"epoch": 0.8132666969559291,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.2745,
|
|
"mean_token_accuracy": 0.12240010797977448,
|
|
"num_tokens": 1846855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.2657325744628904,
|
|
"epoch": 0.817810086324398,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.2508,
|
|
"mean_token_accuracy": 0.13143997862935067,
|
|
"num_tokens": 1857747.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.357445240020752,
|
|
"epoch": 0.8223534756928669,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.3017,
|
|
"mean_token_accuracy": 0.12524260133504866,
|
|
"num_tokens": 1868098.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.268481779098511,
|
|
"epoch": 0.8268968650613358,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.2039,
|
|
"mean_token_accuracy": 0.12827951461076736,
|
|
"num_tokens": 1879682.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.3160333156585695,
|
|
"epoch": 0.8314402544298046,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.2286,
|
|
"mean_token_accuracy": 0.12772295102477074,
|
|
"num_tokens": 1890605.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.30132942199707,
|
|
"epoch": 0.8359836437982735,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.2498,
|
|
"mean_token_accuracy": 0.13315400406718253,
|
|
"num_tokens": 1900861.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.333718204498291,
|
|
"epoch": 0.8405270331667424,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.2322,
|
|
"mean_token_accuracy": 0.1283200278878212,
|
|
"num_tokens": 1910867.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.311747121810913,
|
|
"epoch": 0.8450704225352113,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.2479,
|
|
"mean_token_accuracy": 0.12160541415214539,
|
|
"num_tokens": 1921424.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.38282585144043,
|
|
"epoch": 0.8496138119036801,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.2312,
|
|
"mean_token_accuracy": 0.13186774998903275,
|
|
"num_tokens": 1930784.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.240966749191284,
|
|
"epoch": 0.854157201272149,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.2259,
|
|
"mean_token_accuracy": 0.1281438447535038,
|
|
"num_tokens": 1940325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.260193490982056,
|
|
"epoch": 0.8587005906406179,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.238,
|
|
"mean_token_accuracy": 0.12963936477899551,
|
|
"num_tokens": 1951799.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.3922779083251955,
|
|
"epoch": 0.8632439800090868,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.229,
|
|
"mean_token_accuracy": 0.12269769310951233,
|
|
"num_tokens": 1962311.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.225095224380493,
|
|
"epoch": 0.8677873693775556,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.2281,
|
|
"mean_token_accuracy": 0.1275744341313839,
|
|
"num_tokens": 1974128.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.168290805816651,
|
|
"epoch": 0.8723307587460245,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.1310737431049347,
|
|
"num_tokens": 1983765.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.315899515151978,
|
|
"epoch": 0.8768741481144934,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.235,
|
|
"mean_token_accuracy": 0.12682206705212593,
|
|
"num_tokens": 1994045.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.336924314498901,
|
|
"epoch": 0.8814175374829623,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.3221,
|
|
"mean_token_accuracy": 0.12170391902327538,
|
|
"num_tokens": 2005353.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.108946228027344,
|
|
"epoch": 0.8859609268514311,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.1282,
|
|
"mean_token_accuracy": 0.13348021805286409,
|
|
"num_tokens": 2015808.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.315927267074585,
|
|
"epoch": 0.8905043162199,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.1734,
|
|
"mean_token_accuracy": 0.13906338140368463,
|
|
"num_tokens": 2025824.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.163518285751342,
|
|
"epoch": 0.8950477055883689,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.197,
|
|
"mean_token_accuracy": 0.1368340127170086,
|
|
"num_tokens": 2036924.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.243646335601807,
|
|
"epoch": 0.8995910949568378,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.228,
|
|
"mean_token_accuracy": 0.13041988983750344,
|
|
"num_tokens": 2047302.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.261142539978027,
|
|
"epoch": 0.9041344843253066,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2047,
|
|
"mean_token_accuracy": 0.12715124264359473,
|
|
"num_tokens": 2057972.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.142115783691406,
|
|
"epoch": 0.9086778736937755,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.0515,
|
|
"mean_token_accuracy": 0.1383442610502243,
|
|
"num_tokens": 2067916.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.9086778736937755,
|
|
"eval_entropy": 6.119013550115186,
|
|
"eval_loss": 6.310596466064453,
|
|
"eval_mean_token_accuracy": 0.13073905674350816,
|
|
"eval_num_tokens": 2067916.0,
|
|
"eval_runtime": 2.5901,
|
|
"eval_samples_per_second": 1326.19,
|
|
"eval_steps_per_second": 166.015,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.095190238952637,
|
|
"epoch": 0.9132212630622444,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004999998223471442,
|
|
"loss": 6.2426,
|
|
"mean_token_accuracy": 0.13274624422192574,
|
|
"num_tokens": 2079119.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.302926826477051,
|
|
"epoch": 0.9177646524307133,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999991006328981,
|
|
"loss": 6.1263,
|
|
"mean_token_accuracy": 0.1312612734735012,
|
|
"num_tokens": 2090080.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.224363374710083,
|
|
"epoch": 0.9223080417991822,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999978237557377,
|
|
"loss": 6.1623,
|
|
"mean_token_accuracy": 0.13245146051049234,
|
|
"num_tokens": 2100541.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.144991064071656,
|
|
"epoch": 0.9268514311676511,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004999959917188137,
|
|
"loss": 6.1332,
|
|
"mean_token_accuracy": 0.1294662743806839,
|
|
"num_tokens": 2111213.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.290520286560058,
|
|
"epoch": 0.93139482053612,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999936045266462,
|
|
"loss": 6.2263,
|
|
"mean_token_accuracy": 0.1313753105700016,
|
|
"num_tokens": 2121714.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.139681434631347,
|
|
"epoch": 0.9359382099045889,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004999906621851255,
|
|
"loss": 6.11,
|
|
"mean_token_accuracy": 0.12752713337540628,
|
|
"num_tokens": 2131226.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.254400205612183,
|
|
"epoch": 0.9404815992730577,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004999871647015115,
|
|
"loss": 6.1907,
|
|
"mean_token_accuracy": 0.12505661994218825,
|
|
"num_tokens": 2143179.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.165806579589844,
|
|
"epoch": 0.9450249886415266,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999831120844339,
|
|
"loss": 6.1733,
|
|
"mean_token_accuracy": 0.13424054831266402,
|
|
"num_tokens": 2153646.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.0638185977935795,
|
|
"epoch": 0.9495683780099955,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999785043438922,
|
|
"loss": 6.1202,
|
|
"mean_token_accuracy": 0.1302776090800762,
|
|
"num_tokens": 2163267.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.2338563919067385,
|
|
"epoch": 0.9541117673784644,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999733414912555,
|
|
"loss": 6.024,
|
|
"mean_token_accuracy": 0.13413071036338806,
|
|
"num_tokens": 2173282.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.159117269515991,
|
|
"epoch": 0.9586551567469332,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004999676235392626,
|
|
"loss": 6.1316,
|
|
"mean_token_accuracy": 0.12614955380558968,
|
|
"num_tokens": 2184202.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.249439859390259,
|
|
"epoch": 0.9631985461154021,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999613505020219,
|
|
"loss": 6.1008,
|
|
"mean_token_accuracy": 0.13505923077464105,
|
|
"num_tokens": 2194342.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.107129049301148,
|
|
"epoch": 0.967741935483871,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999545223950116,
|
|
"loss": 6.1139,
|
|
"mean_token_accuracy": 0.13676788732409478,
|
|
"num_tokens": 2204817.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.196682405471802,
|
|
"epoch": 0.9722853248523399,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999471392350795,
|
|
"loss": 6.1875,
|
|
"mean_token_accuracy": 0.1350012630224228,
|
|
"num_tokens": 2215281.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.2491254806518555,
|
|
"epoch": 0.9768287142208087,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004999392010404426,
|
|
"loss": 6.1817,
|
|
"mean_token_accuracy": 0.13481876030564308,
|
|
"num_tokens": 2225784.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.119947338104248,
|
|
"epoch": 0.9813721035892776,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999307078306876,
|
|
"loss": 6.1209,
|
|
"mean_token_accuracy": 0.13831015974283217,
|
|
"num_tokens": 2235357.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.250767660140991,
|
|
"epoch": 0.9859154929577465,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999216596267708,
|
|
"loss": 6.1407,
|
|
"mean_token_accuracy": 0.13157120868563651,
|
|
"num_tokens": 2246438.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.163524055480957,
|
|
"epoch": 0.9904588823262154,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999120564510177,
|
|
"loss": 6.1279,
|
|
"mean_token_accuracy": 0.13285833448171616,
|
|
"num_tokens": 2257528.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.125581979751587,
|
|
"epoch": 0.9950022716946842,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499901898327123,
|
|
"loss": 6.1579,
|
|
"mean_token_accuracy": 0.13823469057679177,
|
|
"num_tokens": 2267572.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.111034965515136,
|
|
"epoch": 0.9995456610631531,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004998911852801512,
|
|
"loss": 6.0791,
|
|
"mean_token_accuracy": 0.13840872496366502,
|
|
"num_tokens": 2278520.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.206923537784153,
|
|
"epoch": 1.003634711494775,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998799173365355,
|
|
"loss": 5.9924,
|
|
"mean_token_accuracy": 0.138457167479727,
|
|
"num_tokens": 2286991.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.174563884735107,
|
|
"epoch": 1.008178100863244,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004998680945240783,
|
|
"loss": 6.0311,
|
|
"mean_token_accuracy": 0.13416522592306138,
|
|
"num_tokens": 2297857.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.178468990325928,
|
|
"epoch": 1.0127214902317128,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998557168719513,
|
|
"loss": 6.0813,
|
|
"mean_token_accuracy": 0.13441063985228538,
|
|
"num_tokens": 2308621.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.059679460525513,
|
|
"epoch": 1.0172648796001817,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004998427844106954,
|
|
"loss": 5.881,
|
|
"mean_token_accuracy": 0.14716010466217994,
|
|
"num_tokens": 2318654.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.125048398971558,
|
|
"epoch": 1.0218082689686505,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998292971722197,
|
|
"loss": 5.9387,
|
|
"mean_token_accuracy": 0.13893765211105347,
|
|
"num_tokens": 2329064.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.071296548843383,
|
|
"epoch": 1.0263516583371195,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000499815255189803,
|
|
"loss": 5.9926,
|
|
"mean_token_accuracy": 0.1351088397204876,
|
|
"num_tokens": 2339843.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.1580180644989015,
|
|
"epoch": 1.0308950477055883,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004998006584980922,
|
|
"loss": 5.9417,
|
|
"mean_token_accuracy": 0.13985907807946205,
|
|
"num_tokens": 2350715.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.097461557388305,
|
|
"epoch": 1.0354384370740572,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997855071331034,
|
|
"loss": 5.905,
|
|
"mean_token_accuracy": 0.14022022858262062,
|
|
"num_tokens": 2360635.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.042455005645752,
|
|
"epoch": 1.039981826442526,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997698011322211,
|
|
"loss": 5.9263,
|
|
"mean_token_accuracy": 0.13829426020383834,
|
|
"num_tokens": 2370627.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.115921592712402,
|
|
"epoch": 1.044525215810995,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499753540534198,
|
|
"loss": 5.9589,
|
|
"mean_token_accuracy": 0.1388851523399353,
|
|
"num_tokens": 2381790.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 5.961507940292359,
|
|
"epoch": 1.049068605179464,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499736725379156,
|
|
"loss": 5.8937,
|
|
"mean_token_accuracy": 0.14004927650094032,
|
|
"num_tokens": 2392685.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.0807301044464115,
|
|
"epoch": 1.0536119945479328,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997193557085844,
|
|
"loss": 5.9553,
|
|
"mean_token_accuracy": 0.13977984860539436,
|
|
"num_tokens": 2403722.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.059891986846924,
|
|
"epoch": 1.0581553839164017,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997014315653413,
|
|
"loss": 6.0031,
|
|
"mean_token_accuracy": 0.13041603118181228,
|
|
"num_tokens": 2415311.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.061007785797119,
|
|
"epoch": 1.0626987732848705,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004996829529936527,
|
|
"loss": 5.8757,
|
|
"mean_token_accuracy": 0.14463189542293547,
|
|
"num_tokens": 2425116.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.099240064620972,
|
|
"epoch": 1.0672421626533395,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996639200391126,
|
|
"loss": 6.0665,
|
|
"mean_token_accuracy": 0.1358228489756584,
|
|
"num_tokens": 2435916.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 5.997359848022461,
|
|
"epoch": 1.0717855520218083,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004996443327486831,
|
|
"loss": 5.9479,
|
|
"mean_token_accuracy": 0.13406080305576323,
|
|
"num_tokens": 2446920.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.106053638458252,
|
|
"epoch": 1.0763289413902772,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996241911706937,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.14392359107732772,
|
|
"num_tokens": 2456367.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 5.986075735092163,
|
|
"epoch": 1.080872330758746,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004996034953548418,
|
|
"loss": 5.8625,
|
|
"mean_token_accuracy": 0.14260815605521202,
|
|
"num_tokens": 2466790.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.0295631885528564,
|
|
"epoch": 1.085415720127215,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004995822453521924,
|
|
"loss": 5.874,
|
|
"mean_token_accuracy": 0.14739150553941727,
|
|
"num_tokens": 2476419.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.042729234695434,
|
|
"epoch": 1.0899591094956838,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004995604412151776,
|
|
"loss": 6.0248,
|
|
"mean_token_accuracy": 0.1344558708369732,
|
|
"num_tokens": 2487295.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.019842672348022,
|
|
"epoch": 1.0945024988641527,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000499538082997597,
|
|
"loss": 5.8945,
|
|
"mean_token_accuracy": 0.13653064966201783,
|
|
"num_tokens": 2497545.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.14437952041626,
|
|
"epoch": 1.0990458882326215,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004995151707546172,
|
|
"loss": 6.0565,
|
|
"mean_token_accuracy": 0.13368018791079522,
|
|
"num_tokens": 2507805.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.041342830657959,
|
|
"epoch": 1.1035892776010905,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004994917045427721,
|
|
"loss": 5.9199,
|
|
"mean_token_accuracy": 0.1381150558590889,
|
|
"num_tokens": 2518474.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.150459957122803,
|
|
"epoch": 1.1081326669695593,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004994676844199621,
|
|
"loss": 5.9511,
|
|
"mean_token_accuracy": 0.14577605128288268,
|
|
"num_tokens": 2528974.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.016877841949463,
|
|
"epoch": 1.1126760563380282,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994431104454543,
|
|
"loss": 6.0097,
|
|
"mean_token_accuracy": 0.13516863361001014,
|
|
"num_tokens": 2540675.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.028059244155884,
|
|
"epoch": 1.117219445706497,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994179826798828,
|
|
"loss": 5.8699,
|
|
"mean_token_accuracy": 0.14476859122514724,
|
|
"num_tokens": 2551367.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 5.999497795104981,
|
|
"epoch": 1.121762835074966,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004993923011852477,
|
|
"loss": 5.9653,
|
|
"mean_token_accuracy": 0.13774474412202836,
|
|
"num_tokens": 2562096.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.10791974067688,
|
|
"epoch": 1.1263062244434348,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993660660249158,
|
|
"loss": 6.0367,
|
|
"mean_token_accuracy": 0.13786499425768853,
|
|
"num_tokens": 2572917.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 5.983006000518799,
|
|
"epoch": 1.1308496138119037,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004993392772636194,
|
|
"loss": 5.8634,
|
|
"mean_token_accuracy": 0.14501798897981644,
|
|
"num_tokens": 2582971.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.108616685867309,
|
|
"epoch": 1.1353930031803725,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004993119349674573,
|
|
"loss": 6.0225,
|
|
"mean_token_accuracy": 0.13913898766040803,
|
|
"num_tokens": 2593186.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.026316690444946,
|
|
"epoch": 1.1399363925488415,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004992840392038938,
|
|
"loss": 6.018,
|
|
"mean_token_accuracy": 0.1349714659154415,
|
|
"num_tokens": 2604359.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.056495094299317,
|
|
"epoch": 1.1444797819173103,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000499255590041759,
|
|
"loss": 5.9367,
|
|
"mean_token_accuracy": 0.14222330152988433,
|
|
"num_tokens": 2614436.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 5.957977342605591,
|
|
"epoch": 1.1490231712857792,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004992265875512485,
|
|
"loss": 5.898,
|
|
"mean_token_accuracy": 0.14035152941942214,
|
|
"num_tokens": 2624175.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.0870380878448485,
|
|
"epoch": 1.153566560654248,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004991970318039228,
|
|
"loss": 5.9292,
|
|
"mean_token_accuracy": 0.14449749886989594,
|
|
"num_tokens": 2634898.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.137653446197509,
|
|
"epoch": 1.158109950022717,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004991669228727078,
|
|
"loss": 5.9626,
|
|
"mean_token_accuracy": 0.14636341333389283,
|
|
"num_tokens": 2645590.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.0029479503631595,
|
|
"epoch": 1.1626533393911858,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004991362608318947,
|
|
"loss": 5.944,
|
|
"mean_token_accuracy": 0.14702457636594773,
|
|
"num_tokens": 2656115.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.0235066413879395,
|
|
"epoch": 1.1671967287596547,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004991050457571384,
|
|
"loss": 5.9247,
|
|
"mean_token_accuracy": 0.14117783308029175,
|
|
"num_tokens": 2666760.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.054644346237183,
|
|
"epoch": 1.1717401181281235,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004990732777254594,
|
|
"loss": 5.9022,
|
|
"mean_token_accuracy": 0.14372125118970872,
|
|
"num_tokens": 2676763.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 5.955284786224365,
|
|
"epoch": 1.1762835074965925,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004990409568152422,
|
|
"loss": 5.9051,
|
|
"mean_token_accuracy": 0.1380102217197418,
|
|
"num_tokens": 2686598.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.0025769710540775,
|
|
"epoch": 1.1808268968650613,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004990080831062352,
|
|
"loss": 5.8913,
|
|
"mean_token_accuracy": 0.1458988979458809,
|
|
"num_tokens": 2697767.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 5.992935132980347,
|
|
"epoch": 1.1853702862335302,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004989746566795512,
|
|
"loss": 5.8749,
|
|
"mean_token_accuracy": 0.14183612391352654,
|
|
"num_tokens": 2707081.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 5.9335216045379635,
|
|
"epoch": 1.189913675601999,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004989406776176665,
|
|
"loss": 5.9001,
|
|
"mean_token_accuracy": 0.14268443137407302,
|
|
"num_tokens": 2717312.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.063012170791626,
|
|
"epoch": 1.194457064970468,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000498906146004421,
|
|
"loss": 5.9208,
|
|
"mean_token_accuracy": 0.14291029423475266,
|
|
"num_tokens": 2728108.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.0912096977233885,
|
|
"epoch": 1.1990004543389368,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004988710619250181,
|
|
"loss": 6.1112,
|
|
"mean_token_accuracy": 0.14074144288897514,
|
|
"num_tokens": 2738860.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.000742721557617,
|
|
"epoch": 1.2035438437074057,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004988354254660244,
|
|
"loss": 5.9563,
|
|
"mean_token_accuracy": 0.14386919811367987,
|
|
"num_tokens": 2749786.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.006845712661743,
|
|
"epoch": 1.2080872330758745,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004987992367153691,
|
|
"loss": 5.8626,
|
|
"mean_token_accuracy": 0.14512528330087662,
|
|
"num_tokens": 2760018.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 5.931078481674194,
|
|
"epoch": 1.2126306224443435,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004987624957623443,
|
|
"loss": 5.9177,
|
|
"mean_token_accuracy": 0.14297755360603331,
|
|
"num_tokens": 2770572.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 5.937653732299805,
|
|
"epoch": 1.2171740118128123,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004987252026976049,
|
|
"loss": 5.8826,
|
|
"mean_token_accuracy": 0.15135369300842286,
|
|
"num_tokens": 2780297.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.039012765884399,
|
|
"epoch": 1.2217174011812812,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004986873576131676,
|
|
"loss": 5.9992,
|
|
"mean_token_accuracy": 0.14602920040488243,
|
|
"num_tokens": 2790454.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.017839241027832,
|
|
"epoch": 1.22626079054975,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004986489606024116,
|
|
"loss": 5.85,
|
|
"mean_token_accuracy": 0.14553693532943726,
|
|
"num_tokens": 2800034.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 5.885445594787598,
|
|
"epoch": 1.230804179918219,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004986100117600776,
|
|
"loss": 5.864,
|
|
"mean_token_accuracy": 0.14794481843709945,
|
|
"num_tokens": 2810157.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 5.973232746124268,
|
|
"epoch": 1.235347569286688,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000498570511182268,
|
|
"loss": 5.8857,
|
|
"mean_token_accuracy": 0.1394151568412781,
|
|
"num_tokens": 2819404.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 5.941764545440674,
|
|
"epoch": 1.2398909586551567,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004985304589664465,
|
|
"loss": 5.8298,
|
|
"mean_token_accuracy": 0.1434129849076271,
|
|
"num_tokens": 2829551.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 5.947914123535156,
|
|
"epoch": 1.2444343480236255,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004984898552114381,
|
|
"loss": 5.8903,
|
|
"mean_token_accuracy": 0.14327759444713592,
|
|
"num_tokens": 2840381.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 5.982251596450806,
|
|
"epoch": 1.2489777373920945,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004984487000174283,
|
|
"loss": 5.8865,
|
|
"mean_token_accuracy": 0.13880339488387108,
|
|
"num_tokens": 2850279.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.027106142044067,
|
|
"epoch": 1.2535211267605635,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004984069934859637,
|
|
"loss": 5.9153,
|
|
"mean_token_accuracy": 0.150995684415102,
|
|
"num_tokens": 2860209.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 5.9328803539276125,
|
|
"epoch": 1.2580645161290323,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004983647357199509,
|
|
"loss": 5.9148,
|
|
"mean_token_accuracy": 0.14855456426739694,
|
|
"num_tokens": 2870932.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.015257740020752,
|
|
"epoch": 1.262607905497501,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004983219268236567,
|
|
"loss": 5.9053,
|
|
"mean_token_accuracy": 0.14229729101061822,
|
|
"num_tokens": 2881139.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 5.992234373092652,
|
|
"epoch": 1.26715129486597,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004982785669027079,
|
|
"loss": 5.893,
|
|
"mean_token_accuracy": 0.14636517614126204,
|
|
"num_tokens": 2890774.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 5.978844785690308,
|
|
"epoch": 1.271694684234439,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004982346560640907,
|
|
"loss": 5.8629,
|
|
"mean_token_accuracy": 0.1498524270951748,
|
|
"num_tokens": 2900259.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 5.957555437088013,
|
|
"epoch": 1.2762380736029078,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004981901944161507,
|
|
"loss": 5.8767,
|
|
"mean_token_accuracy": 0.1461343213915825,
|
|
"num_tokens": 2911365.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 5.930820655822754,
|
|
"epoch": 1.2807814629713765,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004981451820685928,
|
|
"loss": 5.8679,
|
|
"mean_token_accuracy": 0.1443587377667427,
|
|
"num_tokens": 2921189.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 5.872604942321777,
|
|
"epoch": 1.2853248523398455,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004980996191324802,
|
|
"loss": 5.8969,
|
|
"mean_token_accuracy": 0.1431989386677742,
|
|
"num_tokens": 2931271.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.04894380569458,
|
|
"epoch": 1.2898682417083145,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980535057202351,
|
|
"loss": 5.8929,
|
|
"mean_token_accuracy": 0.14426368549466134,
|
|
"num_tokens": 2941557.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 5.822301912307739,
|
|
"epoch": 1.2944116310767833,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980068419456378,
|
|
"loss": 5.8086,
|
|
"mean_token_accuracy": 0.15275876671075822,
|
|
"num_tokens": 2951305.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 5.909662580490112,
|
|
"epoch": 1.2989550204452522,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004979596279238265,
|
|
"loss": 5.8986,
|
|
"mean_token_accuracy": 0.15126647800207138,
|
|
"num_tokens": 2961948.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 5.9860834121704105,
|
|
"epoch": 1.303498409813721,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004979118637712969,
|
|
"loss": 5.9588,
|
|
"mean_token_accuracy": 0.13625321835279464,
|
|
"num_tokens": 2973300.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 5.997512292861939,
|
|
"epoch": 1.30804179918219,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004978635496059025,
|
|
"loss": 5.9007,
|
|
"mean_token_accuracy": 0.14264205247163772,
|
|
"num_tokens": 2984114.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 5.804658126831055,
|
|
"epoch": 1.3125851885506588,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004978146855468537,
|
|
"loss": 5.7665,
|
|
"mean_token_accuracy": 0.1514141008257866,
|
|
"num_tokens": 2993940.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 5.981986856460571,
|
|
"epoch": 1.3171285779191277,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004977652717147177,
|
|
"loss": 5.8833,
|
|
"mean_token_accuracy": 0.14869886934757232,
|
|
"num_tokens": 3004093.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 5.826439237594604,
|
|
"epoch": 1.3216719672875965,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004977153082314182,
|
|
"loss": 5.8068,
|
|
"mean_token_accuracy": 0.14922491908073426,
|
|
"num_tokens": 3013537.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 5.9768884658813475,
|
|
"epoch": 1.3262153566560655,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497664795220235,
|
|
"loss": 5.8106,
|
|
"mean_token_accuracy": 0.14891566783189775,
|
|
"num_tokens": 3023502.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 5.868037366867066,
|
|
"epoch": 1.3307587460245343,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004976137328058042,
|
|
"loss": 5.8706,
|
|
"mean_token_accuracy": 0.1440548375248909,
|
|
"num_tokens": 3033910.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 5.910422515869141,
|
|
"epoch": 1.3353021353930032,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000497562121114117,
|
|
"loss": 5.8994,
|
|
"mean_token_accuracy": 0.14917545020580292,
|
|
"num_tokens": 3043500.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 5.915622234344482,
|
|
"epoch": 1.339845524761472,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004975099602725202,
|
|
"loss": 5.8788,
|
|
"mean_token_accuracy": 0.14382385686039925,
|
|
"num_tokens": 3054776.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 5.952015781402588,
|
|
"epoch": 1.344388914129941,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004974572504097154,
|
|
"loss": 5.844,
|
|
"mean_token_accuracy": 0.14768940806388856,
|
|
"num_tokens": 3065403.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 5.887816143035889,
|
|
"epoch": 1.3489323034984098,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004974039916557591,
|
|
"loss": 5.8759,
|
|
"mean_token_accuracy": 0.1451573684811592,
|
|
"num_tokens": 3074661.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 5.878644323348999,
|
|
"epoch": 1.3534756928668787,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000497350184142062,
|
|
"loss": 5.8337,
|
|
"mean_token_accuracy": 0.1441040262579918,
|
|
"num_tokens": 3085319.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 5.879520559310913,
|
|
"epoch": 1.3580190822353475,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004972958280013886,
|
|
"loss": 5.8288,
|
|
"mean_token_accuracy": 0.14350272119045257,
|
|
"num_tokens": 3095115.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 5.940704250335694,
|
|
"epoch": 1.3625624716038165,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004972409233678575,
|
|
"loss": 6.0048,
|
|
"mean_token_accuracy": 0.1379089266061783,
|
|
"num_tokens": 3104841.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.3625624716038165,
|
|
"eval_entropy": 5.845436406690021,
|
|
"eval_loss": 6.132750034332275,
|
|
"eval_mean_token_accuracy": 0.14173085743258165,
|
|
"eval_num_tokens": 3104841.0,
|
|
"eval_runtime": 2.6014,
|
|
"eval_samples_per_second": 1320.451,
|
|
"eval_steps_per_second": 165.297,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 5.94781379699707,
|
|
"epoch": 1.3671058609722853,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004971854703769401,
|
|
"loss": 5.865,
|
|
"mean_token_accuracy": 0.1451248086988926,
|
|
"num_tokens": 3115098.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 5.878427028656006,
|
|
"epoch": 1.3716492503407542,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004971294691654616,
|
|
"loss": 5.7491,
|
|
"mean_token_accuracy": 0.15491948276758194,
|
|
"num_tokens": 3124831.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 5.854487037658691,
|
|
"epoch": 1.376192639709223,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000497072919871599,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.14528753682971002,
|
|
"num_tokens": 3136113.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.086994552612305,
|
|
"epoch": 1.380736029077692,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004970158226348824,
|
|
"loss": 5.839,
|
|
"mean_token_accuracy": 0.15899223536252977,
|
|
"num_tokens": 3145617.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 5.833227968215942,
|
|
"epoch": 1.3852794184461608,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004969581775961932,
|
|
"loss": 5.9398,
|
|
"mean_token_accuracy": 0.13940114825963973,
|
|
"num_tokens": 3156913.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 5.977892684936523,
|
|
"epoch": 1.3898228078146297,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004968999848977651,
|
|
"loss": 5.8803,
|
|
"mean_token_accuracy": 0.14702537804841995,
|
|
"num_tokens": 3167506.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 5.819863986968994,
|
|
"epoch": 1.3943661971830985,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004968412446831827,
|
|
"loss": 5.759,
|
|
"mean_token_accuracy": 0.14905699491500854,
|
|
"num_tokens": 3177956.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 5.9371013164520265,
|
|
"epoch": 1.3989095865515675,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004967819570973816,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.14665073454380034,
|
|
"num_tokens": 3189172.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.025431394577026,
|
|
"epoch": 1.4034529759200365,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000496722122286648,
|
|
"loss": 5.9232,
|
|
"mean_token_accuracy": 0.14157174676656722,
|
|
"num_tokens": 3200188.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 5.811288738250733,
|
|
"epoch": 1.4079963652885052,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004966617403986185,
|
|
"loss": 5.7726,
|
|
"mean_token_accuracy": 0.14368995130062104,
|
|
"num_tokens": 3210171.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.013456678390503,
|
|
"epoch": 1.412539754656974,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004966008115822793,
|
|
"loss": 5.9925,
|
|
"mean_token_accuracy": 0.14135242402553558,
|
|
"num_tokens": 3221201.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 5.860878372192383,
|
|
"epoch": 1.417083144025443,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004965393359879662,
|
|
"loss": 5.8405,
|
|
"mean_token_accuracy": 0.1457865744829178,
|
|
"num_tokens": 3231825.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.031340217590332,
|
|
"epoch": 1.421626533393912,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004964773137673641,
|
|
"loss": 6.0398,
|
|
"mean_token_accuracy": 0.13429115936160088,
|
|
"num_tokens": 3241860.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 5.798910617828369,
|
|
"epoch": 1.4261699227623807,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004964147450735067,
|
|
"loss": 5.6964,
|
|
"mean_token_accuracy": 0.1542890965938568,
|
|
"num_tokens": 3251893.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 5.867627334594727,
|
|
"epoch": 1.4307133121308495,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000496351630060776,
|
|
"loss": 5.8404,
|
|
"mean_token_accuracy": 0.14542031958699225,
|
|
"num_tokens": 3262748.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.006178951263427,
|
|
"epoch": 1.4352567014993185,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004962879688849021,
|
|
"loss": 5.9546,
|
|
"mean_token_accuracy": 0.14313093349337577,
|
|
"num_tokens": 3273035.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 5.92848391532898,
|
|
"epoch": 1.4398000908677875,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004962237617029625,
|
|
"loss": 5.8831,
|
|
"mean_token_accuracy": 0.14664052724838256,
|
|
"num_tokens": 3283413.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 5.916056871414185,
|
|
"epoch": 1.4443434802362562,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004961590086733821,
|
|
"loss": 5.8681,
|
|
"mean_token_accuracy": 0.147178016602993,
|
|
"num_tokens": 3294628.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 5.8341459274292,
|
|
"epoch": 1.448886869604725,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004960937099559326,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.15535128712654114,
|
|
"num_tokens": 3304631.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 5.913455057144165,
|
|
"epoch": 1.453430258973194,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004960278657117321,
|
|
"loss": 5.8617,
|
|
"mean_token_accuracy": 0.14451269209384918,
|
|
"num_tokens": 3315481.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 5.88686203956604,
|
|
"epoch": 1.457973648341663,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004959614761032446,
|
|
"loss": 5.8135,
|
|
"mean_token_accuracy": 0.147933691740036,
|
|
"num_tokens": 3326493.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.0231451988220215,
|
|
"epoch": 1.4625170377101318,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00049589454129428,
|
|
"loss": 5.9445,
|
|
"mean_token_accuracy": 0.1462418481707573,
|
|
"num_tokens": 3338358.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 5.90168080329895,
|
|
"epoch": 1.4670604270786005,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004958270614499932,
|
|
"loss": 5.9121,
|
|
"mean_token_accuracy": 0.14460725635290145,
|
|
"num_tokens": 3347961.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 5.902929592132568,
|
|
"epoch": 1.4716038164470695,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004957590367368841,
|
|
"loss": 5.8387,
|
|
"mean_token_accuracy": 0.14820509701967238,
|
|
"num_tokens": 3358544.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 5.889368629455566,
|
|
"epoch": 1.4761472058155385,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004956904673227968,
|
|
"loss": 5.8445,
|
|
"mean_token_accuracy": 0.14531347751617432,
|
|
"num_tokens": 3368233.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 5.820783710479736,
|
|
"epoch": 1.4806905951840073,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004956213533769195,
|
|
"loss": 5.8111,
|
|
"mean_token_accuracy": 0.1510622039437294,
|
|
"num_tokens": 3379065.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 5.984694337844848,
|
|
"epoch": 1.485233984552476,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004955516950697843,
|
|
"loss": 5.8971,
|
|
"mean_token_accuracy": 0.139473906904459,
|
|
"num_tokens": 3389251.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 5.882978105545044,
|
|
"epoch": 1.489777373920945,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004954814925732657,
|
|
"loss": 5.826,
|
|
"mean_token_accuracy": 0.15168372839689254,
|
|
"num_tokens": 3399660.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 5.889898157119751,
|
|
"epoch": 1.494320763289414,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004954107460605816,
|
|
"loss": 5.8601,
|
|
"mean_token_accuracy": 0.1534946471452713,
|
|
"num_tokens": 3409395.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 5.963047361373901,
|
|
"epoch": 1.4988641526578828,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004953394557062922,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.14099805802106857,
|
|
"num_tokens": 3419881.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 5.847076463699341,
|
|
"epoch": 1.5034075420263515,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004952676216862989,
|
|
"loss": 5.8645,
|
|
"mean_token_accuracy": 0.14429206624627114,
|
|
"num_tokens": 3430316.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 5.860261344909668,
|
|
"epoch": 1.5079509313948205,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004951952441778453,
|
|
"loss": 5.7421,
|
|
"mean_token_accuracy": 0.1556067496538162,
|
|
"num_tokens": 3440174.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 5.8579755306243895,
|
|
"epoch": 1.5124943207632895,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004951223233595158,
|
|
"loss": 5.9341,
|
|
"mean_token_accuracy": 0.14123850762844087,
|
|
"num_tokens": 3451141.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 5.965538597106933,
|
|
"epoch": 1.5170377101317583,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000495048859411235,
|
|
"loss": 5.8206,
|
|
"mean_token_accuracy": 0.14651950299739838,
|
|
"num_tokens": 3461767.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 5.793015766143799,
|
|
"epoch": 1.521581099500227,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004949748525142681,
|
|
"loss": 5.7692,
|
|
"mean_token_accuracy": 0.15745436996221543,
|
|
"num_tokens": 3470937.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 5.909920597076416,
|
|
"epoch": 1.526124488868696,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004949003028512195,
|
|
"loss": 5.8406,
|
|
"mean_token_accuracy": 0.14674706086516381,
|
|
"num_tokens": 3482130.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 5.864456653594971,
|
|
"epoch": 1.530667878237165,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004948252106060334,
|
|
"loss": 5.9126,
|
|
"mean_token_accuracy": 0.1415312610566616,
|
|
"num_tokens": 3491992.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 5.804475402832031,
|
|
"epoch": 1.5352112676056338,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004947495759639923,
|
|
"loss": 5.7755,
|
|
"mean_token_accuracy": 0.15035406947135926,
|
|
"num_tokens": 3501437.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 5.762890481948853,
|
|
"epoch": 1.5397546569741025,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004946733991117171,
|
|
"loss": 5.8417,
|
|
"mean_token_accuracy": 0.147023569047451,
|
|
"num_tokens": 3511625.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 5.87570858001709,
|
|
"epoch": 1.5442980463425715,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004945966802371668,
|
|
"loss": 5.6182,
|
|
"mean_token_accuracy": 0.15641522854566575,
|
|
"num_tokens": 3520428.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 5.840044355392456,
|
|
"epoch": 1.5488414357110405,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004945194195296374,
|
|
"loss": 5.8491,
|
|
"mean_token_accuracy": 0.14945719689130782,
|
|
"num_tokens": 3532218.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.903447675704956,
|
|
"epoch": 1.5533848250795095,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000494441617179762,
|
|
"loss": 5.8271,
|
|
"mean_token_accuracy": 0.14525461941957474,
|
|
"num_tokens": 3543177.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 5.880182600021362,
|
|
"epoch": 1.5579282144479782,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004943632733795104,
|
|
"loss": 5.7948,
|
|
"mean_token_accuracy": 0.15421123802661896,
|
|
"num_tokens": 3552833.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 5.73713264465332,
|
|
"epoch": 1.562471603816447,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004942843883221879,
|
|
"loss": 5.8598,
|
|
"mean_token_accuracy": 0.1434869036078453,
|
|
"num_tokens": 3563558.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 5.838307952880859,
|
|
"epoch": 1.567014993184916,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004942049622024357,
|
|
"loss": 5.8033,
|
|
"mean_token_accuracy": 0.15330324470996856,
|
|
"num_tokens": 3574993.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 5.911986684799194,
|
|
"epoch": 1.571558382553385,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004941249952162298,
|
|
"loss": 5.8293,
|
|
"mean_token_accuracy": 0.14535781294107436,
|
|
"num_tokens": 3584586.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 5.7583925247192385,
|
|
"epoch": 1.5761017719218537,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004940444875608807,
|
|
"loss": 5.771,
|
|
"mean_token_accuracy": 0.15141805708408357,
|
|
"num_tokens": 3595029.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 5.890980005264282,
|
|
"epoch": 1.5806451612903225,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004939634394350332,
|
|
"loss": 5.84,
|
|
"mean_token_accuracy": 0.14627694189548493,
|
|
"num_tokens": 3606034.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 5.831475687026978,
|
|
"epoch": 1.5851885506587915,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004938818510386654,
|
|
"loss": 5.7677,
|
|
"mean_token_accuracy": 0.150801981985569,
|
|
"num_tokens": 3615551.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 5.829084348678589,
|
|
"epoch": 1.5897319400272605,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004937997225730886,
|
|
"loss": 5.9243,
|
|
"mean_token_accuracy": 0.1494157448410988,
|
|
"num_tokens": 3626811.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 5.850355720520019,
|
|
"epoch": 1.5942753293957292,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004937170542409464,
|
|
"loss": 5.7805,
|
|
"mean_token_accuracy": 0.14997649937868118,
|
|
"num_tokens": 3637338.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 5.743219995498658,
|
|
"epoch": 1.598818718764198,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000493633846246215,
|
|
"loss": 5.7073,
|
|
"mean_token_accuracy": 0.15153501853346824,
|
|
"num_tokens": 3648650.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 5.89624662399292,
|
|
"epoch": 1.603362108132667,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004935500987942018,
|
|
"loss": 5.8664,
|
|
"mean_token_accuracy": 0.14928599148988725,
|
|
"num_tokens": 3659331.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 5.901072025299072,
|
|
"epoch": 1.607905497501136,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004934658120915452,
|
|
"loss": 5.93,
|
|
"mean_token_accuracy": 0.1412065401673317,
|
|
"num_tokens": 3670757.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 5.891441011428833,
|
|
"epoch": 1.6124488868696047,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004933809863462142,
|
|
"loss": 5.7722,
|
|
"mean_token_accuracy": 0.15418703854084015,
|
|
"num_tokens": 3681286.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 5.819612312316894,
|
|
"epoch": 1.6169922762380735,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004932956217675082,
|
|
"loss": 5.7627,
|
|
"mean_token_accuracy": 0.15468500554561615,
|
|
"num_tokens": 3690733.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 5.880888891220093,
|
|
"epoch": 1.6215356656065425,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004932097185660555,
|
|
"loss": 5.9492,
|
|
"mean_token_accuracy": 0.14442042708396913,
|
|
"num_tokens": 3701263.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 5.902997541427612,
|
|
"epoch": 1.6260790549750115,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004931232769538138,
|
|
"loss": 5.8574,
|
|
"mean_token_accuracy": 0.14900370091199874,
|
|
"num_tokens": 3711578.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 5.941886234283447,
|
|
"epoch": 1.6306224443434802,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004930362971440692,
|
|
"loss": 5.8372,
|
|
"mean_token_accuracy": 0.14781564697623253,
|
|
"num_tokens": 3722140.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 5.696824550628662,
|
|
"epoch": 1.635165833711949,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004929487793514358,
|
|
"loss": 5.5599,
|
|
"mean_token_accuracy": 0.15967873632907867,
|
|
"num_tokens": 3732019.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 5.805211877822876,
|
|
"epoch": 1.639709223080418,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000492860723791855,
|
|
"loss": 5.7642,
|
|
"mean_token_accuracy": 0.15093351081013678,
|
|
"num_tokens": 3741912.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 5.867282581329346,
|
|
"epoch": 1.644252612448887,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004927721306825952,
|
|
"loss": 5.7436,
|
|
"mean_token_accuracy": 0.15467035323381423,
|
|
"num_tokens": 3751365.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 5.756223678588867,
|
|
"epoch": 1.6487960018173557,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004926830002422509,
|
|
"loss": 5.7401,
|
|
"mean_token_accuracy": 0.14856993556022643,
|
|
"num_tokens": 3762160.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 5.841907358169555,
|
|
"epoch": 1.6533393911858245,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000492593332690743,
|
|
"loss": 5.7751,
|
|
"mean_token_accuracy": 0.15270450413227082,
|
|
"num_tokens": 3772608.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 5.784197378158569,
|
|
"epoch": 1.6578827805542935,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004925031282493167,
|
|
"loss": 5.8244,
|
|
"mean_token_accuracy": 0.14576919451355935,
|
|
"num_tokens": 3783344.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 5.834315538406372,
|
|
"epoch": 1.6624261699227625,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004924123871405428,
|
|
"loss": 5.8314,
|
|
"mean_token_accuracy": 0.1431602880358696,
|
|
"num_tokens": 3794103.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 5.791322803497314,
|
|
"epoch": 1.6669695592912313,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004923211095883161,
|
|
"loss": 5.7645,
|
|
"mean_token_accuracy": 0.15026065930724145,
|
|
"num_tokens": 3804537.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 5.845962476730347,
|
|
"epoch": 1.6715129486597,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004922292958178546,
|
|
"loss": 5.7828,
|
|
"mean_token_accuracy": 0.14817606657743454,
|
|
"num_tokens": 3814147.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 5.8800671100616455,
|
|
"epoch": 1.676056338028169,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004921369460557,
|
|
"loss": 5.8494,
|
|
"mean_token_accuracy": 0.1477331094443798,
|
|
"num_tokens": 3823640.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 5.768005895614624,
|
|
"epoch": 1.680599727396638,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004920440605297159,
|
|
"loss": 5.6941,
|
|
"mean_token_accuracy": 0.1621052548289299,
|
|
"num_tokens": 3833418.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 5.781481456756592,
|
|
"epoch": 1.6851431167651068,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000491950639469088,
|
|
"loss": 5.715,
|
|
"mean_token_accuracy": 0.15754078924655915,
|
|
"num_tokens": 3844998.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 5.709991407394409,
|
|
"epoch": 1.6896865061335755,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004918566831043238,
|
|
"loss": 5.6519,
|
|
"mean_token_accuracy": 0.16085057556629181,
|
|
"num_tokens": 3855549.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 5.748021173477173,
|
|
"epoch": 1.6942298955020445,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000491762191667251,
|
|
"loss": 5.77,
|
|
"mean_token_accuracy": 0.14954300299286843,
|
|
"num_tokens": 3865287.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 5.839926052093506,
|
|
"epoch": 1.6987732848705135,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000491667165391018,
|
|
"loss": 5.8156,
|
|
"mean_token_accuracy": 0.15002627968788146,
|
|
"num_tokens": 3875914.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 5.837830352783203,
|
|
"epoch": 1.7033166742389823,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004915716045100927,
|
|
"loss": 5.82,
|
|
"mean_token_accuracy": 0.14646194875240326,
|
|
"num_tokens": 3885196.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 5.858989715576172,
|
|
"epoch": 1.707860063607451,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004914755092602621,
|
|
"loss": 5.8509,
|
|
"mean_token_accuracy": 0.14010663330554962,
|
|
"num_tokens": 3895227.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 5.839261770248413,
|
|
"epoch": 1.71240345297592,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004913788798786314,
|
|
"loss": 5.7349,
|
|
"mean_token_accuracy": 0.15189915597438813,
|
|
"num_tokens": 3905926.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 5.73164644241333,
|
|
"epoch": 1.716946842344389,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004912817166036243,
|
|
"loss": 5.7775,
|
|
"mean_token_accuracy": 0.1529722660779953,
|
|
"num_tokens": 3916982.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 5.735852575302124,
|
|
"epoch": 1.7214902317128578,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004911840196749813,
|
|
"loss": 5.6888,
|
|
"mean_token_accuracy": 0.15600222796201707,
|
|
"num_tokens": 3926889.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 5.79035530090332,
|
|
"epoch": 1.7260336210813265,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004910857893337602,
|
|
"loss": 5.7173,
|
|
"mean_token_accuracy": 0.1464507170021534,
|
|
"num_tokens": 3936734.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 5.838219404220581,
|
|
"epoch": 1.7305770104497955,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004909870258223342,
|
|
"loss": 5.8571,
|
|
"mean_token_accuracy": 0.1459001362323761,
|
|
"num_tokens": 3948001.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 5.830196475982666,
|
|
"epoch": 1.7351203998182645,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004908877293843926,
|
|
"loss": 5.7328,
|
|
"mean_token_accuracy": 0.14376608580350875,
|
|
"num_tokens": 3957433.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 5.854920768737793,
|
|
"epoch": 1.7396637891867333,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004907879002649395,
|
|
"loss": 5.8053,
|
|
"mean_token_accuracy": 0.14790061712265015,
|
|
"num_tokens": 3967375.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 5.825519275665283,
|
|
"epoch": 1.744207178555202,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004906875387102935,
|
|
"loss": 5.7991,
|
|
"mean_token_accuracy": 0.1441611260175705,
|
|
"num_tokens": 3977190.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 5.819745349884033,
|
|
"epoch": 1.748750567923671,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004905866449680864,
|
|
"loss": 5.7739,
|
|
"mean_token_accuracy": 0.1543112054467201,
|
|
"num_tokens": 3987878.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 5.807843160629273,
|
|
"epoch": 1.75329395729214,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004904852192872638,
|
|
"loss": 5.7595,
|
|
"mean_token_accuracy": 0.15661360770463945,
|
|
"num_tokens": 3997166.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 5.839473009109497,
|
|
"epoch": 1.757837346660609,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004903832619180834,
|
|
"loss": 5.7745,
|
|
"mean_token_accuracy": 0.14904787093400956,
|
|
"num_tokens": 4007711.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 5.879887485504151,
|
|
"epoch": 1.7623807360290777,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004902807731121147,
|
|
"loss": 5.9244,
|
|
"mean_token_accuracy": 0.14445645958185196,
|
|
"num_tokens": 4018907.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 5.81760721206665,
|
|
"epoch": 1.7669241253975465,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004901777531222389,
|
|
"loss": 5.805,
|
|
"mean_token_accuracy": 0.15347942411899568,
|
|
"num_tokens": 4030006.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 5.913085842132569,
|
|
"epoch": 1.7714675147660155,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004900742022026474,
|
|
"loss": 5.8894,
|
|
"mean_token_accuracy": 0.15333996266126632,
|
|
"num_tokens": 4040609.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 5.780558061599732,
|
|
"epoch": 1.7760109041344845,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004899701206088419,
|
|
"loss": 5.7283,
|
|
"mean_token_accuracy": 0.14981921315193175,
|
|
"num_tokens": 4050392.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 5.786962413787842,
|
|
"epoch": 1.7805542935029532,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004898655085976333,
|
|
"loss": 5.8167,
|
|
"mean_token_accuracy": 0.14799166470766068,
|
|
"num_tokens": 4060156.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.847379684448242,
|
|
"epoch": 1.785097682871422,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004897603664271413,
|
|
"loss": 5.694,
|
|
"mean_token_accuracy": 0.1542102053761482,
|
|
"num_tokens": 4069670.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 5.715523719787598,
|
|
"epoch": 1.789641072239891,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000489654694356794,
|
|
"loss": 5.7208,
|
|
"mean_token_accuracy": 0.1497572034597397,
|
|
"num_tokens": 4080738.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 5.706692314147949,
|
|
"epoch": 1.79418446160836,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004895484926473265,
|
|
"loss": 5.7011,
|
|
"mean_token_accuracy": 0.15149858146905898,
|
|
"num_tokens": 4091122.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.886227178573608,
|
|
"epoch": 1.7987278509768287,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000489441761560781,
|
|
"loss": 5.8172,
|
|
"mean_token_accuracy": 0.15150666385889053,
|
|
"num_tokens": 4101055.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 5.796804714202881,
|
|
"epoch": 1.8032712403452975,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000489334501360506,
|
|
"loss": 5.7593,
|
|
"mean_token_accuracy": 0.15707833468914031,
|
|
"num_tokens": 4110631.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 5.802775764465332,
|
|
"epoch": 1.8078146297137665,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004892267123111551,
|
|
"loss": 5.7715,
|
|
"mean_token_accuracy": 0.15272098332643508,
|
|
"num_tokens": 4121120.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 5.787547445297241,
|
|
"epoch": 1.8123580190822355,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004891183946786873,
|
|
"loss": 5.7183,
|
|
"mean_token_accuracy": 0.15237994641065597,
|
|
"num_tokens": 4132106.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 5.80684905052185,
|
|
"epoch": 1.8169014084507042,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004890095487303655,
|
|
"loss": 5.7586,
|
|
"mean_token_accuracy": 0.1490725338459015,
|
|
"num_tokens": 4142216.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 1.8169014084507042,
|
|
"eval_entropy": 5.790170495454655,
|
|
"eval_loss": 5.993229389190674,
|
|
"eval_mean_token_accuracy": 0.14785447165716525,
|
|
"eval_num_tokens": 4142216.0,
|
|
"eval_runtime": 2.6052,
|
|
"eval_samples_per_second": 1318.527,
|
|
"eval_steps_per_second": 165.056,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 5.849746036529541,
|
|
"epoch": 1.821444797819173,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004889001747347562,
|
|
"loss": 5.8059,
|
|
"mean_token_accuracy": 0.15014866143465042,
|
|
"num_tokens": 4152168.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 5.766653680801392,
|
|
"epoch": 1.825988187187642,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000488790272961729,
|
|
"loss": 5.8032,
|
|
"mean_token_accuracy": 0.14618541896343232,
|
|
"num_tokens": 4162085.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 5.875260972976685,
|
|
"epoch": 1.830531576556111,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004886798436824556,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.1477835439145565,
|
|
"num_tokens": 4171168.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 5.769736194610596,
|
|
"epoch": 1.8350749659245797,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004885688871694091,
|
|
"loss": 5.7614,
|
|
"mean_token_accuracy": 0.15020202696323395,
|
|
"num_tokens": 4181804.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 5.821374797821045,
|
|
"epoch": 1.8396183552930485,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004884574036963639,
|
|
"loss": 5.8199,
|
|
"mean_token_accuracy": 0.1513478934764862,
|
|
"num_tokens": 4192483.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 5.816771841049194,
|
|
"epoch": 1.8441617446615175,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004883453935383941,
|
|
"loss": 5.7112,
|
|
"mean_token_accuracy": 0.1539171442389488,
|
|
"num_tokens": 4202214.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 5.823734378814697,
|
|
"epoch": 1.8487051340299865,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004882328569718738,
|
|
"loss": 5.7831,
|
|
"mean_token_accuracy": 0.15346236750483513,
|
|
"num_tokens": 4212440.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 5.768901395797729,
|
|
"epoch": 1.8532485233984552,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004881197942744758,
|
|
"loss": 5.7498,
|
|
"mean_token_accuracy": 0.1491874173283577,
|
|
"num_tokens": 4222780.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 5.76951231956482,
|
|
"epoch": 1.857791912766924,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004880062057251711,
|
|
"loss": 5.6804,
|
|
"mean_token_accuracy": 0.15827766805887222,
|
|
"num_tokens": 4232895.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 5.765989923477173,
|
|
"epoch": 1.862335302135393,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000487892091604228,
|
|
"loss": 5.7712,
|
|
"mean_token_accuracy": 0.1545240879058838,
|
|
"num_tokens": 4244287.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 5.81574444770813,
|
|
"epoch": 1.866878691503862,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048777745219321186,
|
|
"loss": 5.7374,
|
|
"mean_token_accuracy": 0.14745835065841675,
|
|
"num_tokens": 4254327.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 5.822335624694825,
|
|
"epoch": 1.8714220808723308,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048766228777498405,
|
|
"loss": 5.7705,
|
|
"mean_token_accuracy": 0.15245601162314415,
|
|
"num_tokens": 4264540.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 5.770896577835083,
|
|
"epoch": 1.8759654702407995,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048754659863370125,
|
|
"loss": 5.7368,
|
|
"mean_token_accuracy": 0.15067947059869766,
|
|
"num_tokens": 4274946.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 5.72808837890625,
|
|
"epoch": 1.8805088596092685,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00048743038505481503,
|
|
"loss": 5.6962,
|
|
"mean_token_accuracy": 0.15184763371944426,
|
|
"num_tokens": 4284395.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 5.802645063400268,
|
|
"epoch": 1.8850522489777375,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004873136473250708,
|
|
"loss": 5.7742,
|
|
"mean_token_accuracy": 0.150366872549057,
|
|
"num_tokens": 4294887.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 5.788530588150024,
|
|
"epoch": 1.8895956383462063,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048719638573250715,
|
|
"loss": 5.7803,
|
|
"mean_token_accuracy": 0.14959521740674972,
|
|
"num_tokens": 4304794.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 5.781466913223267,
|
|
"epoch": 1.894139027714675,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004870786005664555,
|
|
"loss": 5.6932,
|
|
"mean_token_accuracy": 0.15308475941419603,
|
|
"num_tokens": 4314810.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 5.787781238555908,
|
|
"epoch": 1.898682417083144,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004869602921175392,
|
|
"loss": 5.7473,
|
|
"mean_token_accuracy": 0.14809799045324326,
|
|
"num_tokens": 4325444.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 5.799683856964111,
|
|
"epoch": 1.903225806451613,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004868414606776724,
|
|
"loss": 5.8012,
|
|
"mean_token_accuracy": 0.14584705978631973,
|
|
"num_tokens": 4335620.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 5.8017576217651365,
|
|
"epoch": 1.9077691958200818,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048672210654005996,
|
|
"loss": 5.7571,
|
|
"mean_token_accuracy": 0.15560859441757202,
|
|
"num_tokens": 4345006.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 5.784935331344604,
|
|
"epoch": 1.9123125851885505,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00048660222999919616,
|
|
"loss": 5.6956,
|
|
"mean_token_accuracy": 0.15184667482972145,
|
|
"num_tokens": 4354805.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 5.795891237258911,
|
|
"epoch": 1.9168559745570195,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004864818313508647,
|
|
"loss": 5.8479,
|
|
"mean_token_accuracy": 0.1495343454182148,
|
|
"num_tokens": 4365881.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 5.868531036376953,
|
|
"epoch": 1.9213993639254885,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004863609108921372,
|
|
"loss": 5.8125,
|
|
"mean_token_accuracy": 0.1463113822042942,
|
|
"num_tokens": 4376368.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 5.78206090927124,
|
|
"epoch": 1.9259427532939573,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004862394689213728,
|
|
"loss": 5.731,
|
|
"mean_token_accuracy": 0.1575425609946251,
|
|
"num_tokens": 4386695.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 5.824512624740601,
|
|
"epoch": 1.930486142662426,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004861175057382175,
|
|
"loss": 5.764,
|
|
"mean_token_accuracy": 0.14746586680412294,
|
|
"num_tokens": 4396837.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 5.728180027008056,
|
|
"epoch": 1.935029532030895,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004859950216436034,
|
|
"loss": 5.735,
|
|
"mean_token_accuracy": 0.1538913369178772,
|
|
"num_tokens": 4407310.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 5.785648822784424,
|
|
"epoch": 1.939572921399364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004858720169397478,
|
|
"loss": 5.7763,
|
|
"mean_token_accuracy": 0.15280596613883973,
|
|
"num_tokens": 4417544.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 5.789197397232056,
|
|
"epoch": 1.9441163107678328,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004857484919301525,
|
|
"loss": 5.6953,
|
|
"mean_token_accuracy": 0.1523209258913994,
|
|
"num_tokens": 4428228.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 5.759953117370605,
|
|
"epoch": 1.9486597001363015,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004856244469196034,
|
|
"loss": 5.7348,
|
|
"mean_token_accuracy": 0.15404450446367263,
|
|
"num_tokens": 4439088.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 5.827485275268555,
|
|
"epoch": 1.9532030895047705,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000485499882214169,
|
|
"loss": 5.834,
|
|
"mean_token_accuracy": 0.14901283532381057,
|
|
"num_tokens": 4449715.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 5.700705814361572,
|
|
"epoch": 1.9577464788732395,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004853747981212003,
|
|
"loss": 5.6836,
|
|
"mean_token_accuracy": 0.155790339410305,
|
|
"num_tokens": 4460233.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 5.732491683959961,
|
|
"epoch": 1.9622898682417085,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004852491949493301,
|
|
"loss": 5.622,
|
|
"mean_token_accuracy": 0.1616998255252838,
|
|
"num_tokens": 4471401.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 5.765408706665039,
|
|
"epoch": 1.9668332576101772,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004851230730084716,
|
|
"loss": 5.6844,
|
|
"mean_token_accuracy": 0.16186238080263138,
|
|
"num_tokens": 4481185.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 5.768443298339844,
|
|
"epoch": 1.971376646978646,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00048499643260981816,
|
|
"loss": 5.7243,
|
|
"mean_token_accuracy": 0.15362609475851058,
|
|
"num_tokens": 4492342.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 5.788624048233032,
|
|
"epoch": 1.975920036347115,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004848692740658424,
|
|
"loss": 5.8147,
|
|
"mean_token_accuracy": 0.15219077914953233,
|
|
"num_tokens": 4504277.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 5.800276279449463,
|
|
"epoch": 1.980463425715584,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004847415976902954,
|
|
"loss": 5.7605,
|
|
"mean_token_accuracy": 0.15419393926858901,
|
|
"num_tokens": 4514178.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 5.759483575820923,
|
|
"epoch": 1.9850068150840527,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048461340379820587,
|
|
"loss": 5.7428,
|
|
"mean_token_accuracy": 0.1533619150519371,
|
|
"num_tokens": 4524559.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 5.756233596801758,
|
|
"epoch": 1.9895502044525215,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00048448469270587945,
|
|
"loss": 5.6855,
|
|
"mean_token_accuracy": 0.1572543576359749,
|
|
"num_tokens": 4535168.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 5.80228271484375,
|
|
"epoch": 1.9940935938209905,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00048435546473089814,
|
|
"loss": 5.7826,
|
|
"mean_token_accuracy": 0.15085486620664595,
|
|
"num_tokens": 4546105.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 5.812653303146362,
|
|
"epoch": 1.9986369831894595,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048422572019211896,
|
|
"loss": 5.708,
|
|
"mean_token_accuracy": 0.1578806847333908,
|
|
"num_tokens": 4556498.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 5.66848791970147,
|
|
"epoch": 2.0027260336210815,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00048409545940967375,
|
|
"loss": 5.5154,
|
|
"mean_token_accuracy": 0.1526057016518381,
|
|
"num_tokens": 4565029.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 5.796992635726928,
|
|
"epoch": 2.00726942298955,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000483964682704968,
|
|
"loss": 5.4685,
|
|
"mean_token_accuracy": 0.15541424453258515,
|
|
"num_tokens": 4575232.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 5.7437793731689455,
|
|
"epoch": 2.011812812358019,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048383339040068024,
|
|
"loss": 5.4532,
|
|
"mean_token_accuracy": 0.1595204696059227,
|
|
"num_tokens": 4585818.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 5.6068517684936525,
|
|
"epoch": 2.016356201726488,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004837015828207612,
|
|
"loss": 5.4012,
|
|
"mean_token_accuracy": 0.16608490496873857,
|
|
"num_tokens": 4595823.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 5.728079986572266,
|
|
"epoch": 2.020899591094957,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004835692602904329,
|
|
"loss": 5.3915,
|
|
"mean_token_accuracy": 0.1636677697300911,
|
|
"num_tokens": 4606537.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 5.746958637237549,
|
|
"epoch": 2.0254429804634255,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004834364231361882,
|
|
"loss": 5.4224,
|
|
"mean_token_accuracy": 0.1628384292125702,
|
|
"num_tokens": 4616096.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 5.710427331924438,
|
|
"epoch": 2.0299863698318945,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004833030716857894,
|
|
"loss": 5.4293,
|
|
"mean_token_accuracy": 0.16120587512850762,
|
|
"num_tokens": 4626287.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 5.759665155410767,
|
|
"epoch": 2.0345297592003635,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00048316920626826807,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.1623578727245331,
|
|
"num_tokens": 4636898.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 5.624807262420655,
|
|
"epoch": 2.0390731485688325,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048303482721392363,
|
|
"loss": 5.4013,
|
|
"mean_token_accuracy": 0.16981358975172042,
|
|
"num_tokens": 4646263.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 5.736131954193115,
|
|
"epoch": 2.043616537937301,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00048289993485432325,
|
|
"loss": 5.4431,
|
|
"mean_token_accuracy": 0.16511887609958648,
|
|
"num_tokens": 4656463.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 5.68239917755127,
|
|
"epoch": 2.04815992730577,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00048276452952230036,
|
|
"loss": 5.4688,
|
|
"mean_token_accuracy": 0.1624162033200264,
|
|
"num_tokens": 4666824.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 5.599757385253906,
|
|
"epoch": 2.052703316674239,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048262861155195407,
|
|
"loss": 5.3639,
|
|
"mean_token_accuracy": 0.16735220551490784,
|
|
"num_tokens": 4676793.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 5.742352914810181,
|
|
"epoch": 2.057246706042708,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0004824921812786486,
|
|
"loss": 5.5055,
|
|
"mean_token_accuracy": 0.16184383034706115,
|
|
"num_tokens": 4687124.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 5.761610460281372,
|
|
"epoch": 2.0617900954111765,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000482355239039012,
|
|
"loss": 5.6481,
|
|
"mean_token_accuracy": 0.1535782903432846,
|
|
"num_tokens": 4698074.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 5.754100513458252,
|
|
"epoch": 2.0663334847796455,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004822177851709356,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.15992826223373413,
|
|
"num_tokens": 4709177.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 5.782966804504395,
|
|
"epoch": 2.0708768741481145,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00048207982001357327,
|
|
"loss": 5.4517,
|
|
"mean_token_accuracy": 0.1593527875840664,
|
|
"num_tokens": 4719649.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 5.684721803665161,
|
|
"epoch": 2.0754202635165835,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004819413439073402,
|
|
"loss": 5.5078,
|
|
"mean_token_accuracy": 0.16035877615213395,
|
|
"num_tokens": 4731685.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 5.66634693145752,
|
|
"epoch": 2.079963652885052,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004818023571939125,
|
|
"loss": 5.4607,
|
|
"mean_token_accuracy": 0.16363389641046525,
|
|
"num_tokens": 4742854.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 5.704556083679199,
|
|
"epoch": 2.084507042253521,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000481662860216226,
|
|
"loss": 5.4302,
|
|
"mean_token_accuracy": 0.1609123393893242,
|
|
"num_tokens": 4754088.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 5.625769758224488,
|
|
"epoch": 2.08905043162199,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00048152285331847574,
|
|
"loss": 5.4563,
|
|
"mean_token_accuracy": 0.16070560812950135,
|
|
"num_tokens": 4764090.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 5.688674736022949,
|
|
"epoch": 2.093593820990459,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004813823368461148,
|
|
"loss": 5.4389,
|
|
"mean_token_accuracy": 0.1677153393626213,
|
|
"num_tokens": 4773704.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 5.719896936416626,
|
|
"epoch": 2.098137210358928,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004812413111458534,
|
|
"loss": 5.4889,
|
|
"mean_token_accuracy": 0.1603416234254837,
|
|
"num_tokens": 4783678.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 5.613268756866455,
|
|
"epoch": 2.1026805997273965,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004810997765656587,
|
|
"loss": 5.3904,
|
|
"mean_token_accuracy": 0.1691882386803627,
|
|
"num_tokens": 4794183.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 5.591863107681275,
|
|
"epoch": 2.1072239890958655,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048095773345475294,
|
|
"loss": 5.3512,
|
|
"mean_token_accuracy": 0.16529901921749116,
|
|
"num_tokens": 4803670.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 5.654296922683716,
|
|
"epoch": 2.1117673784643345,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00048081518216361367,
|
|
"loss": 5.3684,
|
|
"mean_token_accuracy": 0.1701149135828018,
|
|
"num_tokens": 4813611.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 5.61386194229126,
|
|
"epoch": 2.1163107678328035,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004806721230439719,
|
|
"loss": 5.4503,
|
|
"mean_token_accuracy": 0.15981992781162263,
|
|
"num_tokens": 4824356.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 5.734505891799927,
|
|
"epoch": 2.120854157201272,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004805285564488117,
|
|
"loss": 5.4841,
|
|
"mean_token_accuracy": 0.16397735029458999,
|
|
"num_tokens": 4834410.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 5.604253244400025,
|
|
"epoch": 2.125397546569741,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004803844827323696,
|
|
"loss": 5.3483,
|
|
"mean_token_accuracy": 0.16647956222295762,
|
|
"num_tokens": 4844975.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 5.64118070602417,
|
|
"epoch": 2.12994093593821,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004802399022501329,
|
|
"loss": 5.468,
|
|
"mean_token_accuracy": 0.16066430658102035,
|
|
"num_tokens": 4856419.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 5.5977051734924315,
|
|
"epoch": 2.134484325306679,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004800948153588397,
|
|
"loss": 5.3846,
|
|
"mean_token_accuracy": 0.16584357768297195,
|
|
"num_tokens": 4867115.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 5.622549533843994,
|
|
"epoch": 2.1390277146751475,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047994922241647746,
|
|
"loss": 5.4392,
|
|
"mean_token_accuracy": 0.16220083683729172,
|
|
"num_tokens": 4876675.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 5.697238397598267,
|
|
"epoch": 2.1435711040436165,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00047980312378228244,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.16455530673265456,
|
|
"num_tokens": 4887868.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 5.6144118309021,
|
|
"epoch": 2.1481144934120855,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047965651981673823,
|
|
"loss": 5.4278,
|
|
"mean_token_accuracy": 0.1708490490913391,
|
|
"num_tokens": 4896878.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 5.613104057312012,
|
|
"epoch": 2.1526578827805545,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00047950941088157576,
|
|
"loss": 5.4043,
|
|
"mean_token_accuracy": 0.16638759076595305,
|
|
"num_tokens": 4906683.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 5.624288129806518,
|
|
"epoch": 2.157201272149023,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004793617973397715,
|
|
"loss": 5.4105,
|
|
"mean_token_accuracy": 0.16810904890298844,
|
|
"num_tokens": 4917184.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 5.664835643768311,
|
|
"epoch": 2.161744661517492,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00047921367955554745,
|
|
"loss": 5.4157,
|
|
"mean_token_accuracy": 0.16286967247724532,
|
|
"num_tokens": 4927337.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 5.647590398788452,
|
|
"epoch": 2.166288050885961,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004790650578943693,
|
|
"loss": 5.4402,
|
|
"mean_token_accuracy": 0.15885256677865983,
|
|
"num_tokens": 4937661.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 5.667690324783325,
|
|
"epoch": 2.17083144025443,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00047891593272294636,
|
|
"loss": 5.4757,
|
|
"mean_token_accuracy": 0.15949209034442902,
|
|
"num_tokens": 4948068.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 5.602111148834228,
|
|
"epoch": 2.1753748296228985,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047876630440923,
|
|
"loss": 5.4359,
|
|
"mean_token_accuracy": 0.16286666691303253,
|
|
"num_tokens": 4958703.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 5.604837799072266,
|
|
"epoch": 2.1799182189913675,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004786161733224134,
|
|
"loss": 5.4303,
|
|
"mean_token_accuracy": 0.16746487617492675,
|
|
"num_tokens": 4969190.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 5.608381891250611,
|
|
"epoch": 2.1844616083598365,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00047846553983293006,
|
|
"loss": 5.3352,
|
|
"mean_token_accuracy": 0.17012607008218766,
|
|
"num_tokens": 4978663.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 5.718901681900024,
|
|
"epoch": 2.1890049977283055,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047831440431245307,
|
|
"loss": 5.5016,
|
|
"mean_token_accuracy": 0.1635862797498703,
|
|
"num_tokens": 4989839.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 5.570841836929321,
|
|
"epoch": 2.193548387096774,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047816276713389424,
|
|
"loss": 5.3359,
|
|
"mean_token_accuracy": 0.16816314607858657,
|
|
"num_tokens": 5000030.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 5.614529418945312,
|
|
"epoch": 2.198091776465243,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004780106286714034,
|
|
"loss": 5.5246,
|
|
"mean_token_accuracy": 0.1619194909930229,
|
|
"num_tokens": 5010229.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 5.72342209815979,
|
|
"epoch": 2.202635165833712,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000477857989300367,
|
|
"loss": 5.4352,
|
|
"mean_token_accuracy": 0.16469539403915406,
|
|
"num_tokens": 5021000.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 5.556600046157837,
|
|
"epoch": 2.207178555202181,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047770484939740757,
|
|
"loss": 5.3528,
|
|
"mean_token_accuracy": 0.16867441087961196,
|
|
"num_tokens": 5031319.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 5.554885625839233,
|
|
"epoch": 2.2117219445706495,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004775512093403824,
|
|
"loss": 5.3877,
|
|
"mean_token_accuracy": 0.16844506412744523,
|
|
"num_tokens": 5041892.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 5.641312980651856,
|
|
"epoch": 2.2162653339391185,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00047739706950838327,
|
|
"loss": 5.4625,
|
|
"mean_token_accuracy": 0.16928391605615617,
|
|
"num_tokens": 5052363.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 5.721763467788696,
|
|
"epoch": 2.2208087233075875,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047724243028173484,
|
|
"loss": 5.4828,
|
|
"mean_token_accuracy": 0.16696733087301255,
|
|
"num_tokens": 5063342.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 5.595815992355346,
|
|
"epoch": 2.2253521126760565,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047708729204199406,
|
|
"loss": 5.3949,
|
|
"mean_token_accuracy": 0.15966860353946685,
|
|
"num_tokens": 5073393.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 5.556210947036743,
|
|
"epoch": 2.229895502044525,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000476931655171949,
|
|
"loss": 5.3653,
|
|
"mean_token_accuracy": 0.17017862498760222,
|
|
"num_tokens": 5082417.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 5.705835771560669,
|
|
"epoch": 2.234438891412994,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004767755200556183,
|
|
"loss": 5.5362,
|
|
"mean_token_accuracy": 0.15762273222208023,
|
|
"num_tokens": 5093383.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 5.540210199356079,
|
|
"epoch": 2.238982280781463,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0004766188870782497,
|
|
"loss": 5.3225,
|
|
"mean_token_accuracy": 0.17020710408687592,
|
|
"num_tokens": 5104346.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 5.558584880828858,
|
|
"epoch": 2.243525670149932,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047646175662631967,
|
|
"loss": 5.493,
|
|
"mean_token_accuracy": 0.1641610309481621,
|
|
"num_tokens": 5114375.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 5.656518793106079,
|
|
"epoch": 2.2480690595184005,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00047630412908753183,
|
|
"loss": 5.4652,
|
|
"mean_token_accuracy": 0.16726542860269547,
|
|
"num_tokens": 5125035.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 5.6312737464904785,
|
|
"epoch": 2.2526124488868695,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047614600485081657,
|
|
"loss": 5.4692,
|
|
"mean_token_accuracy": 0.1580390676856041,
|
|
"num_tokens": 5135869.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 5.73362922668457,
|
|
"epoch": 2.2571558382553385,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047598738430632963,
|
|
"loss": 5.4279,
|
|
"mean_token_accuracy": 0.16724849492311478,
|
|
"num_tokens": 5146615.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 5.522592401504516,
|
|
"epoch": 2.2616992276238075,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047582826784545146,
|
|
"loss": 5.3366,
|
|
"mean_token_accuracy": 0.16956065744161605,
|
|
"num_tokens": 5157490.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 5.586135482788086,
|
|
"epoch": 2.266242616992276,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00047566865586078615,
|
|
"loss": 5.4806,
|
|
"mean_token_accuracy": 0.1588055059313774,
|
|
"num_tokens": 5168693.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 5.658676815032959,
|
|
"epoch": 2.270786006360745,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00047550854874616047,
|
|
"loss": 5.4435,
|
|
"mean_token_accuracy": 0.1633205607533455,
|
|
"num_tokens": 5179232.0,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"epoch": 2.270786006360745,
|
|
"eval_entropy": 5.410215771475504,
|
|
"eval_loss": 5.926918983459473,
|
|
"eval_mean_token_accuracy": 0.15270435387311979,
|
|
"eval_num_tokens": 5179232.0,
|
|
"eval_runtime": 2.6005,
|
|
"eval_samples_per_second": 1320.907,
|
|
"eval_steps_per_second": 165.354,
|
|
"step": 2500
|
|
},
|
|
{
|
|
"entropy": 5.575441551208496,
|
|
"epoch": 2.275329395729214,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047534794689662257,
|
|
"loss": 5.4851,
|
|
"mean_token_accuracy": 0.16839685291051865,
|
|
"num_tokens": 5188613.0,
|
|
"step": 2505
|
|
},
|
|
{
|
|
"entropy": 5.622188568115234,
|
|
"epoch": 2.279872785097683,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004751868507084418,
|
|
"loss": 5.3639,
|
|
"mean_token_accuracy": 0.17478009462356567,
|
|
"num_tokens": 5198441.0,
|
|
"step": 2510
|
|
},
|
|
{
|
|
"entropy": 5.584142398834229,
|
|
"epoch": 2.284416174466152,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004750252605791068,
|
|
"loss": 5.4392,
|
|
"mean_token_accuracy": 0.16842508912086487,
|
|
"num_tokens": 5208466.0,
|
|
"step": 2515
|
|
},
|
|
{
|
|
"entropy": 5.495612478256225,
|
|
"epoch": 2.2889595638346205,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0004748631769073254,
|
|
"loss": 5.4106,
|
|
"mean_token_accuracy": 0.16598602533340454,
|
|
"num_tokens": 5218847.0,
|
|
"step": 2520
|
|
},
|
|
{
|
|
"entropy": 5.6372777938842775,
|
|
"epoch": 2.2935029532030895,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004747006000930227,
|
|
"loss": 5.4535,
|
|
"mean_token_accuracy": 0.1546911284327507,
|
|
"num_tokens": 5228870.0,
|
|
"step": 2525
|
|
},
|
|
{
|
|
"entropy": 5.6832057476043705,
|
|
"epoch": 2.2980463425715585,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004745375305373409,
|
|
"loss": 5.3881,
|
|
"mean_token_accuracy": 0.1696573704481125,
|
|
"num_tokens": 5238186.0,
|
|
"step": 2530
|
|
},
|
|
{
|
|
"entropy": 5.5625715255737305,
|
|
"epoch": 2.302589731940027,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00047437396864263804,
|
|
"loss": 5.4582,
|
|
"mean_token_accuracy": 0.16980416923761368,
|
|
"num_tokens": 5247478.0,
|
|
"step": 2535
|
|
},
|
|
{
|
|
"entropy": 5.65769944190979,
|
|
"epoch": 2.307133121308496,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004742099148124866,
|
|
"loss": 5.4044,
|
|
"mean_token_accuracy": 0.1642364203929901,
|
|
"num_tokens": 5258519.0,
|
|
"step": 2540
|
|
},
|
|
{
|
|
"entropy": 5.6592381477355955,
|
|
"epoch": 2.311676510676965,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004740453694516734,
|
|
"loss": 5.4296,
|
|
"mean_token_accuracy": 0.16748272031545638,
|
|
"num_tokens": 5268218.0,
|
|
"step": 2545
|
|
},
|
|
{
|
|
"entropy": 5.493631792068482,
|
|
"epoch": 2.316219900045434,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00047388033296619746,
|
|
"loss": 5.3937,
|
|
"mean_token_accuracy": 0.16241876482963563,
|
|
"num_tokens": 5277926.0,
|
|
"step": 2550
|
|
},
|
|
{
|
|
"entropy": 5.5903431415557865,
|
|
"epoch": 2.320763289413903,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0004737148057632701,
|
|
"loss": 5.4205,
|
|
"mean_token_accuracy": 0.1694105461239815,
|
|
"num_tokens": 5287735.0,
|
|
"step": 2555
|
|
},
|
|
{
|
|
"entropy": 5.570258474349975,
|
|
"epoch": 2.3253066787823715,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004735487882513131,
|
|
"loss": 5.3496,
|
|
"mean_token_accuracy": 0.17214408963918687,
|
|
"num_tokens": 5297120.0,
|
|
"step": 2560
|
|
},
|
|
{
|
|
"entropy": 5.45852484703064,
|
|
"epoch": 2.3298500681508405,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0004733822808399583,
|
|
"loss": 5.3662,
|
|
"mean_token_accuracy": 0.17136418521404267,
|
|
"num_tokens": 5307448.0,
|
|
"step": 2565
|
|
},
|
|
{
|
|
"entropy": 5.61703200340271,
|
|
"epoch": 2.3343934575193095,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00047321528394004615,
|
|
"loss": 5.383,
|
|
"mean_token_accuracy": 0.17348483204841614,
|
|
"num_tokens": 5317914.0,
|
|
"step": 2570
|
|
},
|
|
{
|
|
"entropy": 5.543197679519653,
|
|
"epoch": 2.3389368468877785,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00047304779796362485,
|
|
"loss": 5.3766,
|
|
"mean_token_accuracy": 0.17093151062726974,
|
|
"num_tokens": 5328471.0,
|
|
"step": 2575
|
|
},
|
|
{
|
|
"entropy": 5.516396713256836,
|
|
"epoch": 2.343480236256247,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00047287982332394943,
|
|
"loss": 5.4029,
|
|
"mean_token_accuracy": 0.16134533137083054,
|
|
"num_tokens": 5339118.0,
|
|
"step": 2580
|
|
},
|
|
{
|
|
"entropy": 5.67015905380249,
|
|
"epoch": 2.348023625624716,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047271136043548057,
|
|
"loss": 5.5006,
|
|
"mean_token_accuracy": 0.15954657047986984,
|
|
"num_tokens": 5350183.0,
|
|
"step": 2585
|
|
},
|
|
{
|
|
"entropy": 5.617479610443115,
|
|
"epoch": 2.352567014993185,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004725424097138838,
|
|
"loss": 5.4171,
|
|
"mean_token_accuracy": 0.16176852732896804,
|
|
"num_tokens": 5361317.0,
|
|
"step": 2590
|
|
},
|
|
{
|
|
"entropy": 5.568935775756836,
|
|
"epoch": 2.357110404361654,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00047237297157602816,
|
|
"loss": 5.423,
|
|
"mean_token_accuracy": 0.1682720050215721,
|
|
"num_tokens": 5371338.0,
|
|
"step": 2595
|
|
},
|
|
{
|
|
"entropy": 5.66186933517456,
|
|
"epoch": 2.3616537937301225,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0004722030464399854,
|
|
"loss": 5.5323,
|
|
"mean_token_accuracy": 0.16112422049045563,
|
|
"num_tokens": 5382043.0,
|
|
"step": 2600
|
|
},
|
|
{
|
|
"entropy": 5.6375768184661865,
|
|
"epoch": 2.3661971830985915,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000472032634725029,
|
|
"loss": 5.4346,
|
|
"mean_token_accuracy": 0.16537451148033142,
|
|
"num_tokens": 5392547.0,
|
|
"step": 2605
|
|
},
|
|
{
|
|
"entropy": 5.46543173789978,
|
|
"epoch": 2.3707405724670605,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00047186173685163273,
|
|
"loss": 5.3103,
|
|
"mean_token_accuracy": 0.17888594716787337,
|
|
"num_tokens": 5402455.0,
|
|
"step": 2610
|
|
},
|
|
{
|
|
"entropy": 5.549544048309326,
|
|
"epoch": 2.3752839618355295,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047169035324147023,
|
|
"loss": 5.4883,
|
|
"mean_token_accuracy": 0.16207679510116577,
|
|
"num_tokens": 5413925.0,
|
|
"step": 2615
|
|
},
|
|
{
|
|
"entropy": 5.649377870559692,
|
|
"epoch": 2.379827351203998,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00047151848431741363,
|
|
"loss": 5.5142,
|
|
"mean_token_accuracy": 0.15564770698547364,
|
|
"num_tokens": 5425344.0,
|
|
"step": 2620
|
|
},
|
|
{
|
|
"entropy": 5.551960802078247,
|
|
"epoch": 2.384370740572467,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004713461305035322,
|
|
"loss": 5.4531,
|
|
"mean_token_accuracy": 0.16062026470899582,
|
|
"num_tokens": 5435197.0,
|
|
"step": 2625
|
|
},
|
|
{
|
|
"entropy": 5.683890295028687,
|
|
"epoch": 2.388914129940936,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000471173292225092,
|
|
"loss": 5.441,
|
|
"mean_token_accuracy": 0.166593936085701,
|
|
"num_tokens": 5445683.0,
|
|
"step": 2630
|
|
},
|
|
{
|
|
"entropy": 5.553751707077026,
|
|
"epoch": 2.393457519309405,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00047099996990855424,
|
|
"loss": 5.3168,
|
|
"mean_token_accuracy": 0.17248447388410568,
|
|
"num_tokens": 5456846.0,
|
|
"step": 2635
|
|
},
|
|
{
|
|
"entropy": 5.486712694168091,
|
|
"epoch": 2.3980009086778735,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00047082616398157456,
|
|
"loss": 5.3966,
|
|
"mean_token_accuracy": 0.16698428243398666,
|
|
"num_tokens": 5467701.0,
|
|
"step": 2640
|
|
},
|
|
{
|
|
"entropy": 5.618938636779785,
|
|
"epoch": 2.4025442980463425,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004706518748730017,
|
|
"loss": 5.4277,
|
|
"mean_token_accuracy": 0.1629345163702965,
|
|
"num_tokens": 5478246.0,
|
|
"step": 2645
|
|
},
|
|
{
|
|
"entropy": 5.579886436462402,
|
|
"epoch": 2.4070876874148115,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004704771030128768,
|
|
"loss": 5.4166,
|
|
"mean_token_accuracy": 0.17258816808462143,
|
|
"num_tokens": 5488107.0,
|
|
"step": 2650
|
|
},
|
|
{
|
|
"entropy": 5.611078310012817,
|
|
"epoch": 2.4116310767832805,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 0.00047030184883243203,
|
|
"loss": 5.4146,
|
|
"mean_token_accuracy": 0.16715776920318604,
|
|
"num_tokens": 5498521.0,
|
|
"step": 2655
|
|
},
|
|
{
|
|
"entropy": 5.542270088195801,
|
|
"epoch": 2.416174466151749,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00047012611276408964,
|
|
"loss": 5.4674,
|
|
"mean_token_accuracy": 0.15941065400838852,
|
|
"num_tokens": 5509450.0,
|
|
"step": 2660
|
|
},
|
|
{
|
|
"entropy": 5.610725355148316,
|
|
"epoch": 2.420717855520218,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004699498952414609,
|
|
"loss": 5.3711,
|
|
"mean_token_accuracy": 0.16671083718538285,
|
|
"num_tokens": 5519662.0,
|
|
"step": 2665
|
|
},
|
|
{
|
|
"entropy": 5.5634044170379635,
|
|
"epoch": 2.425261244888687,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00046977319669934506,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.16818245649337768,
|
|
"num_tokens": 5529499.0,
|
|
"step": 2670
|
|
},
|
|
{
|
|
"entropy": 5.544408178329467,
|
|
"epoch": 2.429804634257156,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00046959601757372814,
|
|
"loss": 5.5263,
|
|
"mean_token_accuracy": 0.1580025553703308,
|
|
"num_tokens": 5540938.0,
|
|
"step": 2675
|
|
},
|
|
{
|
|
"entropy": 5.617356300354004,
|
|
"epoch": 2.4343480236256245,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00046941835830178216,
|
|
"loss": 5.3945,
|
|
"mean_token_accuracy": 0.16811649203300477,
|
|
"num_tokens": 5552120.0,
|
|
"step": 2680
|
|
},
|
|
{
|
|
"entropy": 5.520823812484741,
|
|
"epoch": 2.4388914129940935,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00046924021932186357,
|
|
"loss": 5.3612,
|
|
"mean_token_accuracy": 0.16833633482456206,
|
|
"num_tokens": 5561590.0,
|
|
"step": 2685
|
|
},
|
|
{
|
|
"entropy": 5.538807725906372,
|
|
"epoch": 2.4434348023625625,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00046906160107351263,
|
|
"loss": 5.3543,
|
|
"mean_token_accuracy": 0.17522021383047104,
|
|
"num_tokens": 5571682.0,
|
|
"step": 2690
|
|
},
|
|
{
|
|
"entropy": 5.5401527881622314,
|
|
"epoch": 2.4479781917310315,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00046888250399745217,
|
|
"loss": 5.3823,
|
|
"mean_token_accuracy": 0.17362736463546752,
|
|
"num_tokens": 5581299.0,
|
|
"step": 2695
|
|
},
|
|
{
|
|
"entropy": 5.500322532653809,
|
|
"epoch": 2.4525215810995,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00046870292853558646,
|
|
"loss": 5.3322,
|
|
"mean_token_accuracy": 0.17054419070482255,
|
|
"num_tokens": 5590766.0,
|
|
"step": 2700
|
|
},
|
|
{
|
|
"entropy": 5.596356201171875,
|
|
"epoch": 2.457064970467969,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046852287513099997,
|
|
"loss": 5.4528,
|
|
"mean_token_accuracy": 0.1649368464946747,
|
|
"num_tokens": 5600898.0,
|
|
"step": 2705
|
|
},
|
|
{
|
|
"entropy": 5.5878630638122555,
|
|
"epoch": 2.461608359836438,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004683423442279566,
|
|
"loss": 5.4471,
|
|
"mean_token_accuracy": 0.1655479371547699,
|
|
"num_tokens": 5612052.0,
|
|
"step": 2710
|
|
},
|
|
{
|
|
"entropy": 5.576379442214966,
|
|
"epoch": 2.466151749204907,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004681613362718985,
|
|
"loss": 5.5255,
|
|
"mean_token_accuracy": 0.16940135806798934,
|
|
"num_tokens": 5622847.0,
|
|
"step": 2715
|
|
},
|
|
{
|
|
"entropy": 5.519678974151612,
|
|
"epoch": 2.470695138573376,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0004679798517094447,
|
|
"loss": 5.3434,
|
|
"mean_token_accuracy": 0.17103047668933868,
|
|
"num_tokens": 5633134.0,
|
|
"step": 2720
|
|
},
|
|
{
|
|
"entropy": 5.61676058769226,
|
|
"epoch": 2.4752385279418445,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00046779789098839043,
|
|
"loss": 5.5131,
|
|
"mean_token_accuracy": 0.16538699269294738,
|
|
"num_tokens": 5644223.0,
|
|
"step": 2725
|
|
},
|
|
{
|
|
"entropy": 5.535487318038941,
|
|
"epoch": 2.4797819173103135,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00046761545455770555,
|
|
"loss": 5.441,
|
|
"mean_token_accuracy": 0.16498867720365523,
|
|
"num_tokens": 5653620.0,
|
|
"step": 2730
|
|
},
|
|
{
|
|
"entropy": 5.589802694320679,
|
|
"epoch": 2.4843253066787825,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004674325428675339,
|
|
"loss": 5.3628,
|
|
"mean_token_accuracy": 0.16851337254047394,
|
|
"num_tokens": 5663800.0,
|
|
"step": 2735
|
|
},
|
|
{
|
|
"entropy": 5.5757472038269045,
|
|
"epoch": 2.488868696047251,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004672491563691919,
|
|
"loss": 5.4142,
|
|
"mean_token_accuracy": 0.17137184292078017,
|
|
"num_tokens": 5674606.0,
|
|
"step": 2740
|
|
},
|
|
{
|
|
"entropy": 5.553486776351929,
|
|
"epoch": 2.49341208541572,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046706529551516753,
|
|
"loss": 5.5256,
|
|
"mean_token_accuracy": 0.16617924422025682,
|
|
"num_tokens": 5684975.0,
|
|
"step": 2745
|
|
},
|
|
{
|
|
"entropy": 5.5582334995269775,
|
|
"epoch": 2.497955474784189,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046688096075911905,
|
|
"loss": 5.3629,
|
|
"mean_token_accuracy": 0.17054101526737214,
|
|
"num_tokens": 5695336.0,
|
|
"step": 2750
|
|
},
|
|
{
|
|
"entropy": 5.629152488708496,
|
|
"epoch": 2.502498864152658,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004666961525558743,
|
|
"loss": 5.3853,
|
|
"mean_token_accuracy": 0.17582341730594636,
|
|
"num_tokens": 5706027.0,
|
|
"step": 2755
|
|
},
|
|
{
|
|
"entropy": 5.53211817741394,
|
|
"epoch": 2.507042253521127,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000466510871361429,
|
|
"loss": 5.4212,
|
|
"mean_token_accuracy": 0.16578415632247925,
|
|
"num_tokens": 5716026.0,
|
|
"step": 2760
|
|
},
|
|
{
|
|
"entropy": 5.543093681335449,
|
|
"epoch": 2.5115856428895955,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046632511763294615,
|
|
"loss": 5.4318,
|
|
"mean_token_accuracy": 0.1688849925994873,
|
|
"num_tokens": 5726055.0,
|
|
"step": 2765
|
|
},
|
|
{
|
|
"entropy": 5.624571752548218,
|
|
"epoch": 2.5161290322580645,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004661388918287546,
|
|
"loss": 5.4575,
|
|
"mean_token_accuracy": 0.1618290051817894,
|
|
"num_tokens": 5736562.0,
|
|
"step": 2770
|
|
},
|
|
{
|
|
"entropy": 5.6554388999938965,
|
|
"epoch": 2.5206724216265335,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004659521944083481,
|
|
"loss": 5.5098,
|
|
"mean_token_accuracy": 0.1590184301137924,
|
|
"num_tokens": 5747297.0,
|
|
"step": 2775
|
|
},
|
|
{
|
|
"entropy": 5.538739776611328,
|
|
"epoch": 2.525215810995002,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004657650258323838,
|
|
"loss": 5.478,
|
|
"mean_token_accuracy": 0.16293535381555557,
|
|
"num_tokens": 5756801.0,
|
|
"step": 2780
|
|
},
|
|
{
|
|
"entropy": 5.576935529708862,
|
|
"epoch": 2.529759200363471,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00046557738656268166,
|
|
"loss": 5.4731,
|
|
"mean_token_accuracy": 0.1629885956645012,
|
|
"num_tokens": 5767493.0,
|
|
"step": 2785
|
|
},
|
|
{
|
|
"entropy": 5.555518960952758,
|
|
"epoch": 2.53430258973194,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000465389277062223,
|
|
"loss": 5.3594,
|
|
"mean_token_accuracy": 0.17765380591154098,
|
|
"num_tokens": 5777539.0,
|
|
"step": 2790
|
|
},
|
|
{
|
|
"entropy": 5.553316736221314,
|
|
"epoch": 2.538845979100409,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00046520069779514917,
|
|
"loss": 5.3735,
|
|
"mean_token_accuracy": 0.17463183254003525,
|
|
"num_tokens": 5787202.0,
|
|
"step": 2795
|
|
},
|
|
{
|
|
"entropy": 5.628433799743652,
|
|
"epoch": 2.543389368468878,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004650116492267609,
|
|
"loss": 5.4325,
|
|
"mean_token_accuracy": 0.1649623319506645,
|
|
"num_tokens": 5798023.0,
|
|
"step": 2800
|
|
},
|
|
{
|
|
"entropy": 5.5028839111328125,
|
|
"epoch": 2.5479327578373465,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0004648221318235165,
|
|
"loss": 5.33,
|
|
"mean_token_accuracy": 0.173333603143692,
|
|
"num_tokens": 5807711.0,
|
|
"step": 2805
|
|
},
|
|
{
|
|
"entropy": 5.53934383392334,
|
|
"epoch": 2.5524761472058155,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.0004646321460530316,
|
|
"loss": 5.3492,
|
|
"mean_token_accuracy": 0.17637757062911988,
|
|
"num_tokens": 5817053.0,
|
|
"step": 2810
|
|
},
|
|
{
|
|
"entropy": 5.51912112236023,
|
|
"epoch": 2.5570195365742845,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000464441692384077,
|
|
"loss": 5.4475,
|
|
"mean_token_accuracy": 0.1696236550807953,
|
|
"num_tokens": 5827792.0,
|
|
"step": 2815
|
|
},
|
|
{
|
|
"entropy": 5.547548723220825,
|
|
"epoch": 2.561562925942753,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004642507712865784,
|
|
"loss": 5.3467,
|
|
"mean_token_accuracy": 0.17662595361471176,
|
|
"num_tokens": 5837702.0,
|
|
"step": 2820
|
|
},
|
|
{
|
|
"entropy": 5.595866680145264,
|
|
"epoch": 2.566106315311222,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.00046405938323161453,
|
|
"loss": 5.4854,
|
|
"mean_token_accuracy": 0.16242269724607467,
|
|
"num_tokens": 5847973.0,
|
|
"step": 2825
|
|
},
|
|
{
|
|
"entropy": 5.641831970214843,
|
|
"epoch": 2.570649704679691,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00046386752869141624,
|
|
"loss": 5.4681,
|
|
"mean_token_accuracy": 0.16168300807476044,
|
|
"num_tokens": 5857591.0,
|
|
"step": 2830
|
|
},
|
|
{
|
|
"entropy": 5.531976175308228,
|
|
"epoch": 2.57519309404816,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004636752081393658,
|
|
"loss": 5.427,
|
|
"mean_token_accuracy": 0.16628410071134567,
|
|
"num_tokens": 5868616.0,
|
|
"step": 2835
|
|
},
|
|
{
|
|
"entropy": 5.584429121017456,
|
|
"epoch": 2.579736483416629,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.00046348242204999487,
|
|
"loss": 5.3655,
|
|
"mean_token_accuracy": 0.16841525882482528,
|
|
"num_tokens": 5878526.0,
|
|
"step": 2840
|
|
},
|
|
{
|
|
"entropy": 5.5044105529785154,
|
|
"epoch": 2.5842798727850975,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00046328917089898396,
|
|
"loss": 5.3646,
|
|
"mean_token_accuracy": 0.17235694974660873,
|
|
"num_tokens": 5887844.0,
|
|
"step": 2845
|
|
},
|
|
{
|
|
"entropy": 5.548652219772339,
|
|
"epoch": 2.5888232621535665,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004630954551631612,
|
|
"loss": 5.4672,
|
|
"mean_token_accuracy": 0.16923740059137343,
|
|
"num_tokens": 5898523.0,
|
|
"step": 2850
|
|
},
|
|
{
|
|
"entropy": 5.582967376708984,
|
|
"epoch": 2.5933666515220355,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00046290127532050087,
|
|
"loss": 5.4856,
|
|
"mean_token_accuracy": 0.16219591945409775,
|
|
"num_tokens": 5909671.0,
|
|
"step": 2855
|
|
},
|
|
{
|
|
"entropy": 5.532243728637695,
|
|
"epoch": 2.5979100408905045,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004627066318501223,
|
|
"loss": 5.3943,
|
|
"mean_token_accuracy": 0.1625651866197586,
|
|
"num_tokens": 5920451.0,
|
|
"step": 2860
|
|
},
|
|
{
|
|
"entropy": 5.5454424858093265,
|
|
"epoch": 2.602453430258973,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00046251152523228906,
|
|
"loss": 5.4,
|
|
"mean_token_accuracy": 0.17224705815315247,
|
|
"num_tokens": 5931218.0,
|
|
"step": 2865
|
|
},
|
|
{
|
|
"entropy": 5.466064500808716,
|
|
"epoch": 2.606996819627442,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004623159559484074,
|
|
"loss": 5.409,
|
|
"mean_token_accuracy": 0.16545435637235642,
|
|
"num_tokens": 5941380.0,
|
|
"step": 2870
|
|
},
|
|
{
|
|
"entropy": 5.5833117961883545,
|
|
"epoch": 2.611540208995911,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004621199244810248,
|
|
"loss": 5.3604,
|
|
"mean_token_accuracy": 0.1685359075665474,
|
|
"num_tokens": 5951989.0,
|
|
"step": 2875
|
|
},
|
|
{
|
|
"entropy": 5.5274125099182125,
|
|
"epoch": 2.61608359836438,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00046192343131382964,
|
|
"loss": 5.4266,
|
|
"mean_token_accuracy": 0.16473983377218246,
|
|
"num_tokens": 5961982.0,
|
|
"step": 2880
|
|
},
|
|
{
|
|
"entropy": 5.592204570770264,
|
|
"epoch": 2.620626987732849,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.00046172647693164926,
|
|
"loss": 5.4702,
|
|
"mean_token_accuracy": 0.16245893388986588,
|
|
"num_tokens": 5972184.0,
|
|
"step": 2885
|
|
},
|
|
{
|
|
"entropy": 5.5910967826843265,
|
|
"epoch": 2.6251703771013175,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004615290618204489,
|
|
"loss": 5.3963,
|
|
"mean_token_accuracy": 0.16848855316638947,
|
|
"num_tokens": 5981912.0,
|
|
"step": 2890
|
|
},
|
|
{
|
|
"entropy": 5.533260202407837,
|
|
"epoch": 2.6297137664697865,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004613311864673309,
|
|
"loss": 5.3817,
|
|
"mean_token_accuracy": 0.17015166729688644,
|
|
"num_tokens": 5992460.0,
|
|
"step": 2895
|
|
},
|
|
{
|
|
"entropy": 5.5702805519104,
|
|
"epoch": 2.6342571558382555,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.00046113285136053286,
|
|
"loss": 5.4398,
|
|
"mean_token_accuracy": 0.16693648248910903,
|
|
"num_tokens": 6002170.0,
|
|
"step": 2900
|
|
},
|
|
{
|
|
"entropy": 5.510522031784058,
|
|
"epoch": 2.638800545206724,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000460934056989427,
|
|
"loss": 5.349,
|
|
"mean_token_accuracy": 0.16113705635070802,
|
|
"num_tokens": 6012240.0,
|
|
"step": 2905
|
|
},
|
|
{
|
|
"entropy": 5.564363622665406,
|
|
"epoch": 2.643343934575193,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0004607348038445187,
|
|
"loss": 5.3199,
|
|
"mean_token_accuracy": 0.17848723530769348,
|
|
"num_tokens": 6022289.0,
|
|
"step": 2910
|
|
},
|
|
{
|
|
"entropy": 5.595773458480835,
|
|
"epoch": 2.647887323943662,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00046053509241744526,
|
|
"loss": 5.4483,
|
|
"mean_token_accuracy": 0.1646754503250122,
|
|
"num_tokens": 6032268.0,
|
|
"step": 2915
|
|
},
|
|
{
|
|
"entropy": 5.666926097869873,
|
|
"epoch": 2.652430713312131,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046033492320097483,
|
|
"loss": 5.5214,
|
|
"mean_token_accuracy": 0.16039399057626724,
|
|
"num_tokens": 6042574.0,
|
|
"step": 2920
|
|
},
|
|
{
|
|
"entropy": 5.6914349555969235,
|
|
"epoch": 2.6569741026806,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004601342966890049,
|
|
"loss": 5.4947,
|
|
"mean_token_accuracy": 0.1591247946023941,
|
|
"num_tokens": 6052508.0,
|
|
"step": 2925
|
|
},
|
|
{
|
|
"entropy": 5.463538599014282,
|
|
"epoch": 2.6615174920490685,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00045993321337656163,
|
|
"loss": 5.3321,
|
|
"mean_token_accuracy": 0.1798750251531601,
|
|
"num_tokens": 6062541.0,
|
|
"step": 2930
|
|
},
|
|
{
|
|
"entropy": 5.529683589935303,
|
|
"epoch": 2.6660608814175375,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000459731673759798,
|
|
"loss": 5.501,
|
|
"mean_token_accuracy": 0.1589680328965187,
|
|
"num_tokens": 6072532.0,
|
|
"step": 2935
|
|
},
|
|
{
|
|
"entropy": 5.580250597000122,
|
|
"epoch": 2.6706042707860065,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00045952967833599303,
|
|
"loss": 5.3838,
|
|
"mean_token_accuracy": 0.17602769434452056,
|
|
"num_tokens": 6082518.0,
|
|
"step": 2940
|
|
},
|
|
{
|
|
"entropy": 5.56395001411438,
|
|
"epoch": 2.675147660154475,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004593272276035502,
|
|
"loss": 5.3736,
|
|
"mean_token_accuracy": 0.17306891083717346,
|
|
"num_tokens": 6092356.0,
|
|
"step": 2945
|
|
},
|
|
{
|
|
"entropy": 5.564646244049072,
|
|
"epoch": 2.679691049522944,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004591243220619969,
|
|
"loss": 5.4849,
|
|
"mean_token_accuracy": 0.16785520017147065,
|
|
"num_tokens": 6103450.0,
|
|
"step": 2950
|
|
},
|
|
{
|
|
"entropy": 5.5815471649169925,
|
|
"epoch": 2.684234438891413,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000458920962211982,
|
|
"loss": 5.3605,
|
|
"mean_token_accuracy": 0.17186060249805452,
|
|
"num_tokens": 6113708.0,
|
|
"step": 2955
|
|
},
|
|
{
|
|
"entropy": 5.501417446136474,
|
|
"epoch": 2.688777828259882,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004587171485552759,
|
|
"loss": 5.4345,
|
|
"mean_token_accuracy": 0.16605052649974822,
|
|
"num_tokens": 6124508.0,
|
|
"step": 2960
|
|
},
|
|
{
|
|
"entropy": 5.609457397460938,
|
|
"epoch": 2.693321217628351,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00045851288159476854,
|
|
"loss": 5.4159,
|
|
"mean_token_accuracy": 0.17805173248052597,
|
|
"num_tokens": 6135516.0,
|
|
"step": 2965
|
|
},
|
|
{
|
|
"entropy": 5.531106567382812,
|
|
"epoch": 2.6978646069968195,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004583081618344684,
|
|
"loss": 5.3405,
|
|
"mean_token_accuracy": 0.16844869256019593,
|
|
"num_tokens": 6146206.0,
|
|
"step": 2970
|
|
},
|
|
{
|
|
"entropy": 5.461429405212402,
|
|
"epoch": 2.7024079963652885,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000458102989779501,
|
|
"loss": 5.3038,
|
|
"mean_token_accuracy": 0.17313728034496306,
|
|
"num_tokens": 6156244.0,
|
|
"step": 2975
|
|
},
|
|
{
|
|
"entropy": 5.415205717086792,
|
|
"epoch": 2.7069513857337575,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004578973659361081,
|
|
"loss": 5.3411,
|
|
"mean_token_accuracy": 0.1671074226498604,
|
|
"num_tokens": 6167091.0,
|
|
"step": 2980
|
|
},
|
|
{
|
|
"entropy": 5.516201972961426,
|
|
"epoch": 2.711494775102226,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00045769129081164596,
|
|
"loss": 5.4108,
|
|
"mean_token_accuracy": 0.17164330929517746,
|
|
"num_tokens": 6177993.0,
|
|
"step": 2985
|
|
},
|
|
{
|
|
"entropy": 5.499716949462891,
|
|
"epoch": 2.716038164470695,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.0004574847649145845,
|
|
"loss": 5.4394,
|
|
"mean_token_accuracy": 0.17122387140989304,
|
|
"num_tokens": 6187963.0,
|
|
"step": 2990
|
|
},
|
|
{
|
|
"entropy": 5.5332225322723385,
|
|
"epoch": 2.720581553839164,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000457277788754506,
|
|
"loss": 5.419,
|
|
"mean_token_accuracy": 0.16892076134681702,
|
|
"num_tokens": 6198442.0,
|
|
"step": 2995
|
|
},
|
|
{
|
|
"entropy": 5.560413360595703,
|
|
"epoch": 2.725124943207633,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.00045707036284210344,
|
|
"loss": 5.4295,
|
|
"mean_token_accuracy": 0.1676975205540657,
|
|
"num_tokens": 6209332.0,
|
|
"step": 3000
|
|
},
|
|
{
|
|
"epoch": 2.725124943207633,
|
|
"eval_entropy": 5.441976602687392,
|
|
"eval_loss": 5.870466232299805,
|
|
"eval_mean_token_accuracy": 0.15629731280685857,
|
|
"eval_num_tokens": 6209332.0,
|
|
"eval_runtime": 2.6005,
|
|
"eval_samples_per_second": 1320.919,
|
|
"eval_steps_per_second": 165.355,
|
|
"step": 3000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 11000,
|
|
"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": 9004511278080000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|