4f896eeaf6
Model: fpadovani/nld-latn-100mb-after-ppt-Dp-10mb-ckpt500_seed10 Source: Original Platform
3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.3625624716038165,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"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
|
|
}
|
|
],
|
|
"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": 4496087207808000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|