4035 lines
109 KiB
JSON
4035 lines
109 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.10270103728047653,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691911220550537,
|
|
"epoch": 0.00025675259320119136,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.799,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9317.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.69188060760498,
|
|
"epoch": 0.0005135051864023827,
|
|
"grad_norm": 12.8125,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7416,
|
|
"mean_token_accuracy": 0.00035366101656109095,
|
|
"num_tokens": 19116.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690849304199219,
|
|
"epoch": 0.000770257779603574,
|
|
"grad_norm": 10.3125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.4704,
|
|
"mean_token_accuracy": 0.028299430618062615,
|
|
"num_tokens": 27735.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.677557849884034,
|
|
"epoch": 0.0010270103728047654,
|
|
"grad_norm": 6.3125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1572,
|
|
"mean_token_accuracy": 0.04511106200516224,
|
|
"num_tokens": 36118.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.60137643814087,
|
|
"epoch": 0.0012837629660059566,
|
|
"grad_norm": 4.15625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9161,
|
|
"mean_token_accuracy": 0.04636201411485672,
|
|
"num_tokens": 45475.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.534404182434082,
|
|
"epoch": 0.001540515559207148,
|
|
"grad_norm": 3.375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.7714,
|
|
"mean_token_accuracy": 0.043980153650045394,
|
|
"num_tokens": 54903.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.49902982711792,
|
|
"epoch": 0.0017972681524083393,
|
|
"grad_norm": 2.9375,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7268,
|
|
"mean_token_accuracy": 0.040795256942510606,
|
|
"num_tokens": 64161.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.505692100524902,
|
|
"epoch": 0.002054020745609531,
|
|
"grad_norm": 2.71875,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.6231,
|
|
"mean_token_accuracy": 0.04894539378583431,
|
|
"num_tokens": 72522.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.48769702911377,
|
|
"epoch": 0.002310773338810722,
|
|
"grad_norm": 2.625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.5734,
|
|
"mean_token_accuracy": 0.05449059717357159,
|
|
"num_tokens": 80969.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.460338306427001,
|
|
"epoch": 0.002567525932011913,
|
|
"grad_norm": 2.40625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.5207,
|
|
"mean_token_accuracy": 0.059795507788658143,
|
|
"num_tokens": 89180.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.396354579925537,
|
|
"epoch": 0.0028242785252131047,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.489,
|
|
"mean_token_accuracy": 0.06651670336723328,
|
|
"num_tokens": 98187.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.27314863204956,
|
|
"epoch": 0.003081031118414296,
|
|
"grad_norm": 2.203125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.4142,
|
|
"mean_token_accuracy": 0.06263468749821186,
|
|
"num_tokens": 107276.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.325285243988038,
|
|
"epoch": 0.0033377837116154874,
|
|
"grad_norm": 2.296875,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.2979,
|
|
"mean_token_accuracy": 0.06050475314259529,
|
|
"num_tokens": 115585.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.303994750976562,
|
|
"epoch": 0.0035945363048166786,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3171,
|
|
"mean_token_accuracy": 0.06043886691331864,
|
|
"num_tokens": 125353.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.29463996887207,
|
|
"epoch": 0.00385128889801787,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.0823,
|
|
"mean_token_accuracy": 0.06879212968051433,
|
|
"num_tokens": 133313.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.185758209228515,
|
|
"epoch": 0.004108041491219062,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.0795,
|
|
"mean_token_accuracy": 0.06610373817384244,
|
|
"num_tokens": 141741.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.269254875183105,
|
|
"epoch": 0.004364794084420253,
|
|
"grad_norm": 1.9921875,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.0745,
|
|
"mean_token_accuracy": 0.05969331823289394,
|
|
"num_tokens": 152093.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.147056770324706,
|
|
"epoch": 0.004621546677621444,
|
|
"grad_norm": 1.9375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 8.8707,
|
|
"mean_token_accuracy": 0.06663852371275425,
|
|
"num_tokens": 161064.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.158480739593506,
|
|
"epoch": 0.004878299270822635,
|
|
"grad_norm": 1.8125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.8128,
|
|
"mean_token_accuracy": 0.05744527503848076,
|
|
"num_tokens": 169534.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.072422409057618,
|
|
"epoch": 0.005135051864023826,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.7571,
|
|
"mean_token_accuracy": 0.06482505239546299,
|
|
"num_tokens": 178090.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.049930477142334,
|
|
"epoch": 0.005391804457225018,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.6771,
|
|
"mean_token_accuracy": 0.05720922462642193,
|
|
"num_tokens": 187191.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.983645057678222,
|
|
"epoch": 0.0056485570504262094,
|
|
"grad_norm": 1.6171875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.4955,
|
|
"mean_token_accuracy": 0.06595060117542743,
|
|
"num_tokens": 196206.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.82193603515625,
|
|
"epoch": 0.005905309643627401,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.3509,
|
|
"mean_token_accuracy": 0.0692622035741806,
|
|
"num_tokens": 205427.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.779964828491211,
|
|
"epoch": 0.006162062236828592,
|
|
"grad_norm": 1.53125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.3154,
|
|
"mean_token_accuracy": 0.065452441573143,
|
|
"num_tokens": 214221.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.568517971038819,
|
|
"epoch": 0.006418814830029783,
|
|
"grad_norm": 1.84375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.131,
|
|
"mean_token_accuracy": 0.0722281850874424,
|
|
"num_tokens": 222254.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.459226322174072,
|
|
"epoch": 0.006675567423230975,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 7.9852,
|
|
"mean_token_accuracy": 0.07301772199571133,
|
|
"num_tokens": 230150.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.283203601837158,
|
|
"epoch": 0.006932320016432166,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 7.9713,
|
|
"mean_token_accuracy": 0.07257844731211663,
|
|
"num_tokens": 238453.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.199154663085938,
|
|
"epoch": 0.007189072609633357,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 7.9177,
|
|
"mean_token_accuracy": 0.06936928369104862,
|
|
"num_tokens": 246878.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 8.980141639709473,
|
|
"epoch": 0.007445825202834548,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 7.817,
|
|
"mean_token_accuracy": 0.06908788979053497,
|
|
"num_tokens": 255094.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.778532218933105,
|
|
"epoch": 0.00770257779603574,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.7953,
|
|
"mean_token_accuracy": 0.07565106488764287,
|
|
"num_tokens": 263358.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.707105541229248,
|
|
"epoch": 0.007959330389236931,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.7452,
|
|
"mean_token_accuracy": 0.07268079333007335,
|
|
"num_tokens": 272609.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.517619037628174,
|
|
"epoch": 0.008216082982438123,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.6494,
|
|
"mean_token_accuracy": 0.07468854188919068,
|
|
"num_tokens": 280983.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.38097620010376,
|
|
"epoch": 0.008472835575639314,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.5879,
|
|
"mean_token_accuracy": 0.07265269383788109,
|
|
"num_tokens": 290838.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.340629863739014,
|
|
"epoch": 0.008729588168840506,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.5351,
|
|
"mean_token_accuracy": 0.06626910269260407,
|
|
"num_tokens": 300411.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.264066886901855,
|
|
"epoch": 0.008986340762041696,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.6098,
|
|
"mean_token_accuracy": 0.07406073287129403,
|
|
"num_tokens": 308918.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.195462322235107,
|
|
"epoch": 0.009243093355242888,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.484,
|
|
"mean_token_accuracy": 0.07268239781260491,
|
|
"num_tokens": 318420.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.154305839538575,
|
|
"epoch": 0.00949984594844408,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.5116,
|
|
"mean_token_accuracy": 0.0776434913277626,
|
|
"num_tokens": 327055.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.08856725692749,
|
|
"epoch": 0.00975659854164527,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.3616,
|
|
"mean_token_accuracy": 0.07818598598241806,
|
|
"num_tokens": 335915.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.033003520965575,
|
|
"epoch": 0.010013351134846462,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.4749,
|
|
"mean_token_accuracy": 0.07401596866548062,
|
|
"num_tokens": 345106.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.030381727218629,
|
|
"epoch": 0.010270103728047653,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.4917,
|
|
"mean_token_accuracy": 0.07544633224606515,
|
|
"num_tokens": 353935.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 7.991475439071655,
|
|
"epoch": 0.010526856321248845,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.3455,
|
|
"mean_token_accuracy": 0.08079211339354515,
|
|
"num_tokens": 362605.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 7.986777877807617,
|
|
"epoch": 0.010783608914450037,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.471,
|
|
"mean_token_accuracy": 0.06931170262396336,
|
|
"num_tokens": 372168.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 7.995200538635254,
|
|
"epoch": 0.011040361507651227,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.536,
|
|
"mean_token_accuracy": 0.07623609974980354,
|
|
"num_tokens": 381468.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 7.939433813095093,
|
|
"epoch": 0.011297114100852419,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.3218,
|
|
"mean_token_accuracy": 0.08064079508185387,
|
|
"num_tokens": 389561.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 7.908961153030395,
|
|
"epoch": 0.01155386669405361,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5704,
|
|
"mean_token_accuracy": 0.0783368855714798,
|
|
"num_tokens": 397805.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 7.937159585952759,
|
|
"epoch": 0.011810619287254801,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.3989,
|
|
"mean_token_accuracy": 0.08215266242623329,
|
|
"num_tokens": 406896.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 7.882169580459594,
|
|
"epoch": 0.012067371880455993,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.2734,
|
|
"mean_token_accuracy": 0.08328061774373055,
|
|
"num_tokens": 414860.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 7.8939135551452635,
|
|
"epoch": 0.012324124473657183,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.355,
|
|
"mean_token_accuracy": 0.08430732116103172,
|
|
"num_tokens": 423443.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.78408784866333,
|
|
"epoch": 0.012580877066858375,
|
|
"grad_norm": 1.5078125,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.2922,
|
|
"mean_token_accuracy": 0.08125164359807968,
|
|
"num_tokens": 432117.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 7.77522349357605,
|
|
"epoch": 0.012837629660059566,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.3195,
|
|
"mean_token_accuracy": 0.08494714722037315,
|
|
"num_tokens": 440840.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 7.8891784191131595,
|
|
"epoch": 0.013094382253260758,
|
|
"grad_norm": 1.7109375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3992,
|
|
"mean_token_accuracy": 0.08097291365265846,
|
|
"num_tokens": 449360.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.78587646484375,
|
|
"epoch": 0.01335113484646195,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3074,
|
|
"mean_token_accuracy": 0.081997200101614,
|
|
"num_tokens": 458876.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.7375373363494875,
|
|
"epoch": 0.01360788743966314,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.1912,
|
|
"mean_token_accuracy": 0.08457952067255974,
|
|
"num_tokens": 467969.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.766811180114746,
|
|
"epoch": 0.013864640032864332,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3171,
|
|
"mean_token_accuracy": 0.08289734944701195,
|
|
"num_tokens": 476704.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.914922475814819,
|
|
"epoch": 0.014121392626065524,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3412,
|
|
"mean_token_accuracy": 0.08541710525751114,
|
|
"num_tokens": 485195.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.65853967666626,
|
|
"epoch": 0.014378145219266714,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.2956,
|
|
"mean_token_accuracy": 0.08422470390796662,
|
|
"num_tokens": 494407.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.690006875991822,
|
|
"epoch": 0.014634897812467906,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3619,
|
|
"mean_token_accuracy": 0.07757975198328496,
|
|
"num_tokens": 503531.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.814142894744873,
|
|
"epoch": 0.014891650405669097,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.2233,
|
|
"mean_token_accuracy": 0.08842376917600632,
|
|
"num_tokens": 512305.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.703822946548462,
|
|
"epoch": 0.015148402998870289,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.282,
|
|
"mean_token_accuracy": 0.08247971273958683,
|
|
"num_tokens": 520948.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.692700719833374,
|
|
"epoch": 0.01540515559207148,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.2455,
|
|
"mean_token_accuracy": 0.08176964893937111,
|
|
"num_tokens": 529977.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.672420310974121,
|
|
"epoch": 0.01566190818527267,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.2191,
|
|
"mean_token_accuracy": 0.08188581019639969,
|
|
"num_tokens": 539228.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.733105230331421,
|
|
"epoch": 0.015918660778473863,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2403,
|
|
"mean_token_accuracy": 0.08645093813538551,
|
|
"num_tokens": 548090.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.628307485580445,
|
|
"epoch": 0.016175413371675055,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.2116,
|
|
"mean_token_accuracy": 0.08736620321869851,
|
|
"num_tokens": 556418.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.579625797271729,
|
|
"epoch": 0.016432165964876247,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2116,
|
|
"mean_token_accuracy": 0.09048386812210082,
|
|
"num_tokens": 565429.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.6401910305023195,
|
|
"epoch": 0.016688918558077435,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2101,
|
|
"mean_token_accuracy": 0.08711659237742424,
|
|
"num_tokens": 573920.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.615418386459351,
|
|
"epoch": 0.016945671151278627,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2057,
|
|
"mean_token_accuracy": 0.0882373534142971,
|
|
"num_tokens": 582270.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.543057155609131,
|
|
"epoch": 0.01720242374447982,
|
|
"grad_norm": 1.6328125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2358,
|
|
"mean_token_accuracy": 0.08880284726619721,
|
|
"num_tokens": 590930.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.635246229171753,
|
|
"epoch": 0.01745917633768101,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.1773,
|
|
"mean_token_accuracy": 0.08663938567042351,
|
|
"num_tokens": 599572.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.654460287094116,
|
|
"epoch": 0.017715928930882203,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1733,
|
|
"mean_token_accuracy": 0.09477491900324822,
|
|
"num_tokens": 607911.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.545411205291748,
|
|
"epoch": 0.017972681524083392,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.1219,
|
|
"mean_token_accuracy": 0.08721931949257851,
|
|
"num_tokens": 616444.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.5561144828796385,
|
|
"epoch": 0.018229434117284584,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.0951,
|
|
"mean_token_accuracy": 0.08625085130333901,
|
|
"num_tokens": 624714.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.543462324142456,
|
|
"epoch": 0.018486186710485776,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2033,
|
|
"mean_token_accuracy": 0.08905729576945305,
|
|
"num_tokens": 633837.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.518378496170044,
|
|
"epoch": 0.018742939303686968,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.0503,
|
|
"mean_token_accuracy": 0.08793065920472146,
|
|
"num_tokens": 642651.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.5862819194793705,
|
|
"epoch": 0.01899969189688816,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1634,
|
|
"mean_token_accuracy": 0.08841664642095566,
|
|
"num_tokens": 651313.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.547570848464966,
|
|
"epoch": 0.01925644449008935,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1722,
|
|
"mean_token_accuracy": 0.08741778731346131,
|
|
"num_tokens": 659564.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.563963174819946,
|
|
"epoch": 0.01951319708329054,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1386,
|
|
"mean_token_accuracy": 0.08939097225666046,
|
|
"num_tokens": 668499.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.461315679550171,
|
|
"epoch": 0.019769949676491733,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.151,
|
|
"mean_token_accuracy": 0.08465680629014968,
|
|
"num_tokens": 677675.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.548288154602051,
|
|
"epoch": 0.020026702269692925,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1371,
|
|
"mean_token_accuracy": 0.09032455012202263,
|
|
"num_tokens": 686555.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.353896141052246,
|
|
"epoch": 0.020283454862894117,
|
|
"grad_norm": 1.5546875,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 6.9751,
|
|
"mean_token_accuracy": 0.09022012501955032,
|
|
"num_tokens": 694682.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.487498664855957,
|
|
"epoch": 0.020540207456095305,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.0423,
|
|
"mean_token_accuracy": 0.09877757728099823,
|
|
"num_tokens": 702503.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.513206958770752,
|
|
"epoch": 0.020796960049296497,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1483,
|
|
"mean_token_accuracy": 0.08845404237508774,
|
|
"num_tokens": 711877.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.403651523590088,
|
|
"epoch": 0.02105371264249769,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 6.9786,
|
|
"mean_token_accuracy": 0.09564082995057106,
|
|
"num_tokens": 720561.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.373950386047364,
|
|
"epoch": 0.02131046523569888,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 6.9904,
|
|
"mean_token_accuracy": 0.08903604373335838,
|
|
"num_tokens": 730025.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.472009801864624,
|
|
"epoch": 0.021567217828900073,
|
|
"grad_norm": 1.5625,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0876,
|
|
"mean_token_accuracy": 0.08974287211894989,
|
|
"num_tokens": 739548.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.318749809265137,
|
|
"epoch": 0.021823970422101262,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 6.9577,
|
|
"mean_token_accuracy": 0.09518374651670455,
|
|
"num_tokens": 747650.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.377895450592041,
|
|
"epoch": 0.022080723015302454,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0842,
|
|
"mean_token_accuracy": 0.0949647605419159,
|
|
"num_tokens": 756268.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.38172607421875,
|
|
"epoch": 0.022337475608503646,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.0908,
|
|
"mean_token_accuracy": 0.09451524391770363,
|
|
"num_tokens": 764547.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.389048147201538,
|
|
"epoch": 0.022594228201704838,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 6.979,
|
|
"mean_token_accuracy": 0.09728751182556153,
|
|
"num_tokens": 772770.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.495685863494873,
|
|
"epoch": 0.02285098079490603,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.033,
|
|
"mean_token_accuracy": 0.08979090973734856,
|
|
"num_tokens": 781929.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.204121541976929,
|
|
"epoch": 0.02310773338810722,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 6.9463,
|
|
"mean_token_accuracy": 0.09810084253549575,
|
|
"num_tokens": 790347.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.445066928863525,
|
|
"epoch": 0.02336448598130841,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.0875,
|
|
"mean_token_accuracy": 0.08975477069616318,
|
|
"num_tokens": 799459.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.3512537479400635,
|
|
"epoch": 0.023621238574509602,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 6.9551,
|
|
"mean_token_accuracy": 0.08982622176408768,
|
|
"num_tokens": 808794.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.38092827796936,
|
|
"epoch": 0.023877991167710794,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0366,
|
|
"mean_token_accuracy": 0.09353781640529632,
|
|
"num_tokens": 816847.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.3289604663848875,
|
|
"epoch": 0.024134743760911986,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.0519,
|
|
"mean_token_accuracy": 0.09150565192103385,
|
|
"num_tokens": 826265.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.382613706588745,
|
|
"epoch": 0.024391496354113175,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9688,
|
|
"mean_token_accuracy": 0.09861691147089005,
|
|
"num_tokens": 834946.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.206914281845092,
|
|
"epoch": 0.024648248947314367,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 6.8336,
|
|
"mean_token_accuracy": 0.097340989112854,
|
|
"num_tokens": 843314.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.18478798866272,
|
|
"epoch": 0.02490500154051556,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.8258,
|
|
"mean_token_accuracy": 0.10658634081482887,
|
|
"num_tokens": 851461.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.266985607147217,
|
|
"epoch": 0.02516175413371675,
|
|
"grad_norm": 1.390625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9496,
|
|
"mean_token_accuracy": 0.09785787239670754,
|
|
"num_tokens": 859940.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.2672138690948485,
|
|
"epoch": 0.025418506726917943,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9646,
|
|
"mean_token_accuracy": 0.09655739292502404,
|
|
"num_tokens": 868934.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.28968653678894,
|
|
"epoch": 0.02567525932011913,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.009,
|
|
"mean_token_accuracy": 0.094332055747509,
|
|
"num_tokens": 877417.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.160351991653442,
|
|
"epoch": 0.025932011913320323,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 6.8665,
|
|
"mean_token_accuracy": 0.09625929743051528,
|
|
"num_tokens": 885726.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.233226728439331,
|
|
"epoch": 0.026188764506521515,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9961,
|
|
"mean_token_accuracy": 0.09239926263689995,
|
|
"num_tokens": 895924.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.216875839233398,
|
|
"epoch": 0.026445517099722708,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9015,
|
|
"mean_token_accuracy": 0.09810723289847374,
|
|
"num_tokens": 904757.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.378665018081665,
|
|
"epoch": 0.0267022696929239,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.0072,
|
|
"mean_token_accuracy": 0.09722578376531602,
|
|
"num_tokens": 912694.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.177597379684448,
|
|
"epoch": 0.02695902228612509,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.888,
|
|
"mean_token_accuracy": 0.10232829377055168,
|
|
"num_tokens": 921690.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.1055169105529785,
|
|
"epoch": 0.02721577487932628,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.8117,
|
|
"mean_token_accuracy": 0.09707141816616058,
|
|
"num_tokens": 930496.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.26405611038208,
|
|
"epoch": 0.027472527472527472,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.9185,
|
|
"mean_token_accuracy": 0.09697732776403427,
|
|
"num_tokens": 938393.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.15372633934021,
|
|
"epoch": 0.027729280065728664,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.7994,
|
|
"mean_token_accuracy": 0.09662254527211189,
|
|
"num_tokens": 946279.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.183806324005127,
|
|
"epoch": 0.027986032658929856,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9244,
|
|
"mean_token_accuracy": 0.09983463510870934,
|
|
"num_tokens": 954395.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.1928853511810305,
|
|
"epoch": 0.028242785252131048,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.8672,
|
|
"mean_token_accuracy": 0.09900753498077393,
|
|
"num_tokens": 963694.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.128181028366089,
|
|
"epoch": 0.028499537845332237,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 6.7804,
|
|
"mean_token_accuracy": 0.10478645786643029,
|
|
"num_tokens": 971565.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.08982892036438,
|
|
"epoch": 0.02875629043853343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8344,
|
|
"mean_token_accuracy": 0.0989646278321743,
|
|
"num_tokens": 980388.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.176906585693359,
|
|
"epoch": 0.02901304303173462,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.868,
|
|
"mean_token_accuracy": 0.09915727749466896,
|
|
"num_tokens": 989043.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.148084831237793,
|
|
"epoch": 0.029269795624935813,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.8426,
|
|
"mean_token_accuracy": 0.09912259057164192,
|
|
"num_tokens": 997408.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.052320384979248,
|
|
"epoch": 0.029526548218137005,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8085,
|
|
"mean_token_accuracy": 0.0965848982334137,
|
|
"num_tokens": 1005900.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.0651140213012695,
|
|
"epoch": 0.029783300811338193,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.7154,
|
|
"mean_token_accuracy": 0.10280231162905692,
|
|
"num_tokens": 1014924.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.165651893615722,
|
|
"epoch": 0.030040053404539385,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.8296,
|
|
"mean_token_accuracy": 0.09803589880466461,
|
|
"num_tokens": 1023540.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.053730487823486,
|
|
"epoch": 0.030296805997740577,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9607,
|
|
"mean_token_accuracy": 0.09082756042480469,
|
|
"num_tokens": 1032445.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.215946769714355,
|
|
"epoch": 0.03055355859094177,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.8667,
|
|
"mean_token_accuracy": 0.09901303499937057,
|
|
"num_tokens": 1041105.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.018390941619873,
|
|
"epoch": 0.03081031118414296,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 6.8101,
|
|
"mean_token_accuracy": 0.10199690088629723,
|
|
"num_tokens": 1049639.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.048635768890381,
|
|
"epoch": 0.03106706377734415,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.000302,
|
|
"loss": 6.8004,
|
|
"mean_token_accuracy": 0.1020125299692154,
|
|
"num_tokens": 1058239.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.024094772338867,
|
|
"epoch": 0.03132381637054534,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.7958,
|
|
"mean_token_accuracy": 0.09868983328342437,
|
|
"num_tokens": 1067278.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.176386404037475,
|
|
"epoch": 0.03158056896374654,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10330700203776359,
|
|
"num_tokens": 1076650.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.066343307495117,
|
|
"epoch": 0.031837321556947726,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7426,
|
|
"mean_token_accuracy": 0.10077258050441742,
|
|
"num_tokens": 1085169.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 6.9530964374542235,
|
|
"epoch": 0.032094074150148914,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.7801,
|
|
"mean_token_accuracy": 0.10433307737112045,
|
|
"num_tokens": 1093808.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.0473321914672855,
|
|
"epoch": 0.03235082674335011,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.7775,
|
|
"mean_token_accuracy": 0.09625793769955635,
|
|
"num_tokens": 1102189.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.026541709899902,
|
|
"epoch": 0.0326075793365513,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8196,
|
|
"mean_token_accuracy": 0.10000753924250602,
|
|
"num_tokens": 1110565.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.039559555053711,
|
|
"epoch": 0.032864331929752494,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.7566,
|
|
"mean_token_accuracy": 0.10695340260863304,
|
|
"num_tokens": 1118747.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.003891038894653,
|
|
"epoch": 0.03312108452295368,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.8291,
|
|
"mean_token_accuracy": 0.10315902680158615,
|
|
"num_tokens": 1126683.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.019667053222657,
|
|
"epoch": 0.03337783711615487,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7609,
|
|
"mean_token_accuracy": 0.10408550202846527,
|
|
"num_tokens": 1134994.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.002318859100342,
|
|
"epoch": 0.033634589709356066,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.7772,
|
|
"mean_token_accuracy": 0.10078763067722321,
|
|
"num_tokens": 1144086.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.017478084564209,
|
|
"epoch": 0.033891342302557255,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.7028,
|
|
"mean_token_accuracy": 0.10772626698017121,
|
|
"num_tokens": 1151933.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 6.98622498512268,
|
|
"epoch": 0.03414809489575845,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7236,
|
|
"mean_token_accuracy": 0.09854485541582107,
|
|
"num_tokens": 1160734.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 6.965017557144165,
|
|
"epoch": 0.03440484748895964,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.684,
|
|
"mean_token_accuracy": 0.1067756362259388,
|
|
"num_tokens": 1169618.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 6.946940994262695,
|
|
"epoch": 0.03466160008216083,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.7307,
|
|
"mean_token_accuracy": 0.10008135959506034,
|
|
"num_tokens": 1178880.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 6.804220724105835,
|
|
"epoch": 0.03491835267536202,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.745,
|
|
"mean_token_accuracy": 0.10536397024989128,
|
|
"num_tokens": 1188374.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 6.966792440414428,
|
|
"epoch": 0.03517510526856321,
|
|
"grad_norm": 1.3515625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.6271,
|
|
"mean_token_accuracy": 0.10964238047599792,
|
|
"num_tokens": 1196370.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 6.972171688079834,
|
|
"epoch": 0.03543185786176441,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7921,
|
|
"mean_token_accuracy": 0.11119742840528488,
|
|
"num_tokens": 1204835.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 6.851450967788696,
|
|
"epoch": 0.035688610454965596,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.6536,
|
|
"mean_token_accuracy": 0.10988922491669655,
|
|
"num_tokens": 1213305.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 6.958383655548095,
|
|
"epoch": 0.035945363048166784,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.7254,
|
|
"mean_token_accuracy": 0.09785633236169815,
|
|
"num_tokens": 1222565.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 6.904182624816895,
|
|
"epoch": 0.03620211564136798,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.6569,
|
|
"mean_token_accuracy": 0.10749072507023812,
|
|
"num_tokens": 1230900.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 6.89570460319519,
|
|
"epoch": 0.03645886823456917,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.7108,
|
|
"mean_token_accuracy": 0.10577969029545783,
|
|
"num_tokens": 1239849.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 6.9470337390899655,
|
|
"epoch": 0.036715620827770364,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.806,
|
|
"mean_token_accuracy": 0.10210061892867088,
|
|
"num_tokens": 1248506.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 6.860368776321411,
|
|
"epoch": 0.03697237342097155,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.645,
|
|
"mean_token_accuracy": 0.099696134775877,
|
|
"num_tokens": 1257290.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 6.961266803741455,
|
|
"epoch": 0.03722912601417274,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7918,
|
|
"mean_token_accuracy": 0.0997413694858551,
|
|
"num_tokens": 1266109.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 6.942010498046875,
|
|
"epoch": 0.037485878607373936,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.747,
|
|
"mean_token_accuracy": 0.10192611664533616,
|
|
"num_tokens": 1275351.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 6.854095983505249,
|
|
"epoch": 0.037742631200575125,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7333,
|
|
"mean_token_accuracy": 0.10169029906392098,
|
|
"num_tokens": 1284012.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 6.8440882682800295,
|
|
"epoch": 0.03799938379377632,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.6419,
|
|
"mean_token_accuracy": 0.10884017422795296,
|
|
"num_tokens": 1292214.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 6.819351625442505,
|
|
"epoch": 0.03825613638697751,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.6123,
|
|
"mean_token_accuracy": 0.10305506885051727,
|
|
"num_tokens": 1302799.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 6.774753761291504,
|
|
"epoch": 0.0385128889801787,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.667,
|
|
"mean_token_accuracy": 0.10314338579773903,
|
|
"num_tokens": 1311581.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 6.904450607299805,
|
|
"epoch": 0.03876964157337989,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.6436,
|
|
"mean_token_accuracy": 0.10472412779927254,
|
|
"num_tokens": 1320219.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 6.777873897552491,
|
|
"epoch": 0.03902639416658108,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.10496818795800208,
|
|
"num_tokens": 1329714.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 6.883046340942383,
|
|
"epoch": 0.03928314675978228,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6862,
|
|
"mean_token_accuracy": 0.10567674189805984,
|
|
"num_tokens": 1338411.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.758247423171997,
|
|
"epoch": 0.039539899352983465,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.5607,
|
|
"mean_token_accuracy": 0.10631941556930542,
|
|
"num_tokens": 1347596.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 6.744019985198975,
|
|
"epoch": 0.039796651946184654,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6612,
|
|
"mean_token_accuracy": 0.10684075877070427,
|
|
"num_tokens": 1356035.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 6.807973432540893,
|
|
"epoch": 0.04005340453938585,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.6099,
|
|
"mean_token_accuracy": 0.10774841159582138,
|
|
"num_tokens": 1364164.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 6.840193223953247,
|
|
"epoch": 0.04031015713258704,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6694,
|
|
"mean_token_accuracy": 0.10652225613594055,
|
|
"num_tokens": 1372586.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.927090787887574,
|
|
"epoch": 0.04056690972578823,
|
|
"grad_norm": 1.3984375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.7943,
|
|
"mean_token_accuracy": 0.10169010534882546,
|
|
"num_tokens": 1381116.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.839020395278931,
|
|
"epoch": 0.04082366231898942,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.5962,
|
|
"mean_token_accuracy": 0.10432421043515205,
|
|
"num_tokens": 1390242.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 6.690506362915039,
|
|
"epoch": 0.04108041491219061,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.6313,
|
|
"mean_token_accuracy": 0.10073486045002937,
|
|
"num_tokens": 1398658.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.773408126831055,
|
|
"epoch": 0.041337167505391806,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.621,
|
|
"mean_token_accuracy": 0.10617334991693497,
|
|
"num_tokens": 1406967.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 6.895376968383789,
|
|
"epoch": 0.041593920098592994,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.6875,
|
|
"mean_token_accuracy": 0.10613970384001732,
|
|
"num_tokens": 1415816.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.821965837478638,
|
|
"epoch": 0.04185067269179419,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.7016,
|
|
"mean_token_accuracy": 0.10710509046912194,
|
|
"num_tokens": 1424484.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 6.823911380767822,
|
|
"epoch": 0.04210742528499538,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.5465,
|
|
"mean_token_accuracy": 0.10651697218418121,
|
|
"num_tokens": 1433495.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.732336616516113,
|
|
"epoch": 0.04236417787819657,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6606,
|
|
"mean_token_accuracy": 0.10958470776677132,
|
|
"num_tokens": 1441715.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.854647445678711,
|
|
"epoch": 0.04262093047139776,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.6268,
|
|
"mean_token_accuracy": 0.10801949054002762,
|
|
"num_tokens": 1450429.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.783870458602905,
|
|
"epoch": 0.04287768306459895,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.7265,
|
|
"mean_token_accuracy": 0.1041849248111248,
|
|
"num_tokens": 1459747.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.764624309539795,
|
|
"epoch": 0.043134435657800146,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.7442,
|
|
"mean_token_accuracy": 0.10243798047304153,
|
|
"num_tokens": 1468682.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.7156370162963865,
|
|
"epoch": 0.043391188251001335,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.5596,
|
|
"mean_token_accuracy": 0.11029603555798531,
|
|
"num_tokens": 1477724.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.7472302436828615,
|
|
"epoch": 0.043647940844202524,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5978,
|
|
"mean_token_accuracy": 0.10845020189881324,
|
|
"num_tokens": 1486117.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.772536373138427,
|
|
"epoch": 0.04390469343740372,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.5992,
|
|
"mean_token_accuracy": 0.1081738904118538,
|
|
"num_tokens": 1494565.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.7836977481842045,
|
|
"epoch": 0.04416144603060491,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.6677,
|
|
"mean_token_accuracy": 0.10965263172984123,
|
|
"num_tokens": 1503500.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.586282682418823,
|
|
"epoch": 0.0444181986238061,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.4671,
|
|
"mean_token_accuracy": 0.1118468277156353,
|
|
"num_tokens": 1512739.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.800614833831787,
|
|
"epoch": 0.04467495121700729,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6409,
|
|
"mean_token_accuracy": 0.09986573159694671,
|
|
"num_tokens": 1522122.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.722772169113159,
|
|
"epoch": 0.04493170381020848,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.6114,
|
|
"mean_token_accuracy": 0.1113222785294056,
|
|
"num_tokens": 1530703.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.568696403503418,
|
|
"epoch": 0.045188456403409676,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5627,
|
|
"mean_token_accuracy": 0.11127637922763825,
|
|
"num_tokens": 1538762.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.819272136688232,
|
|
"epoch": 0.045445208996610864,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.6229,
|
|
"mean_token_accuracy": 0.10142861902713776,
|
|
"num_tokens": 1547100.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.647491836547852,
|
|
"epoch": 0.04570196158981206,
|
|
"grad_norm": 1.3203125,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4841,
|
|
"mean_token_accuracy": 0.11103278324007988,
|
|
"num_tokens": 1555254.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.761734437942505,
|
|
"epoch": 0.04595871418301325,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6245,
|
|
"mean_token_accuracy": 0.10657144710421562,
|
|
"num_tokens": 1563671.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.675083017349243,
|
|
"epoch": 0.04621546677621444,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.6452,
|
|
"mean_token_accuracy": 0.11202742680907249,
|
|
"num_tokens": 1572145.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.782648992538452,
|
|
"epoch": 0.04647221936941563,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.6123,
|
|
"mean_token_accuracy": 0.10610926523804665,
|
|
"num_tokens": 1579855.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.7457867622375485,
|
|
"epoch": 0.04672897196261682,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.6129,
|
|
"mean_token_accuracy": 0.10743910297751427,
|
|
"num_tokens": 1588541.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.67492265701294,
|
|
"epoch": 0.046985724555818016,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5624,
|
|
"mean_token_accuracy": 0.1129060611128807,
|
|
"num_tokens": 1597627.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.641153764724732,
|
|
"epoch": 0.047242477149019205,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5613,
|
|
"mean_token_accuracy": 0.11545208096504211,
|
|
"num_tokens": 1605503.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.691807889938355,
|
|
"epoch": 0.04749922974222039,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.6224,
|
|
"mean_token_accuracy": 0.10865102782845497,
|
|
"num_tokens": 1614320.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.733487224578857,
|
|
"epoch": 0.04775598233542159,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.6323,
|
|
"mean_token_accuracy": 0.10196995958685875,
|
|
"num_tokens": 1623055.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.663532638549805,
|
|
"epoch": 0.04801273492862278,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.501,
|
|
"mean_token_accuracy": 0.10983425378799438,
|
|
"num_tokens": 1631665.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.553802251815796,
|
|
"epoch": 0.04826948752182397,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.5285,
|
|
"mean_token_accuracy": 0.1119989037513733,
|
|
"num_tokens": 1640776.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.554291391372681,
|
|
"epoch": 0.04852624011502516,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.5612,
|
|
"mean_token_accuracy": 0.11241857931017876,
|
|
"num_tokens": 1649749.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.652531671524048,
|
|
"epoch": 0.04878299270822635,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.4725,
|
|
"mean_token_accuracy": 0.11297573447227478,
|
|
"num_tokens": 1657772.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.680579757690429,
|
|
"epoch": 0.049039745301427545,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5154,
|
|
"mean_token_accuracy": 0.11011544540524483,
|
|
"num_tokens": 1666903.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.590639734268189,
|
|
"epoch": 0.049296497894628734,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5941,
|
|
"mean_token_accuracy": 0.11138836666941643,
|
|
"num_tokens": 1675348.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.5894170761108395,
|
|
"epoch": 0.04955325048782993,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.5016,
|
|
"mean_token_accuracy": 0.1073573313653469,
|
|
"num_tokens": 1683824.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.615131616592407,
|
|
"epoch": 0.04981000308103112,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.5411,
|
|
"mean_token_accuracy": 0.11030695140361786,
|
|
"num_tokens": 1692854.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.690183782577515,
|
|
"epoch": 0.050066755674232306,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.6028,
|
|
"mean_token_accuracy": 0.10664756074547768,
|
|
"num_tokens": 1701965.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.550869750976562,
|
|
"epoch": 0.0503235082674335,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.5437,
|
|
"mean_token_accuracy": 0.11219287067651748,
|
|
"num_tokens": 1710741.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.609143257141113,
|
|
"epoch": 0.05058026086063469,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.4867,
|
|
"mean_token_accuracy": 0.10923624709248543,
|
|
"num_tokens": 1719347.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.648265790939331,
|
|
"epoch": 0.050837013453835886,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.5469,
|
|
"mean_token_accuracy": 0.11248492524027824,
|
|
"num_tokens": 1727544.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.576833009719849,
|
|
"epoch": 0.051093766047037074,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.5492,
|
|
"mean_token_accuracy": 0.10793109312653541,
|
|
"num_tokens": 1735646.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.665603303909302,
|
|
"epoch": 0.05135051864023826,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.5417,
|
|
"mean_token_accuracy": 0.10529318302869797,
|
|
"num_tokens": 1744944.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.460438299179077,
|
|
"epoch": 0.05160727123343946,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.503,
|
|
"mean_token_accuracy": 0.1129407100379467,
|
|
"num_tokens": 1753905.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.596185541152954,
|
|
"epoch": 0.05186402382664065,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.5198,
|
|
"mean_token_accuracy": 0.11141006723046303,
|
|
"num_tokens": 1762556.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.49823260307312,
|
|
"epoch": 0.05212077641984184,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.4166,
|
|
"mean_token_accuracy": 0.11588752865791321,
|
|
"num_tokens": 1771709.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.593732023239136,
|
|
"epoch": 0.05237752901304303,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.4899,
|
|
"mean_token_accuracy": 0.11790797635912895,
|
|
"num_tokens": 1780527.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.609111166000366,
|
|
"epoch": 0.052634281606244226,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.6091,
|
|
"mean_token_accuracy": 0.10697147771716117,
|
|
"num_tokens": 1789963.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.605980348587036,
|
|
"epoch": 0.052891034199445415,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.5254,
|
|
"mean_token_accuracy": 0.11993336752057075,
|
|
"num_tokens": 1798328.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.540991926193238,
|
|
"epoch": 0.053147786792646604,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.6146,
|
|
"mean_token_accuracy": 0.10026314929127693,
|
|
"num_tokens": 1807755.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.696726369857788,
|
|
"epoch": 0.0534045393858478,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.5294,
|
|
"mean_token_accuracy": 0.11152229458093643,
|
|
"num_tokens": 1816277.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.571947860717773,
|
|
"epoch": 0.05366129197904899,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.4897,
|
|
"mean_token_accuracy": 0.1152253620326519,
|
|
"num_tokens": 1825315.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.63885087966919,
|
|
"epoch": 0.05391804457225018,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.5491,
|
|
"mean_token_accuracy": 0.1068581148982048,
|
|
"num_tokens": 1833774.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.478586912155151,
|
|
"epoch": 0.05417479716545137,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.498,
|
|
"mean_token_accuracy": 0.11663633137941361,
|
|
"num_tokens": 1842884.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.517606925964356,
|
|
"epoch": 0.05443154975865256,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.4028,
|
|
"mean_token_accuracy": 0.12323538959026337,
|
|
"num_tokens": 1851329.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.602098703384399,
|
|
"epoch": 0.054688302351853756,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.4638,
|
|
"mean_token_accuracy": 0.10800280347466469,
|
|
"num_tokens": 1860095.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.551811361312867,
|
|
"epoch": 0.054945054945054944,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.4557,
|
|
"mean_token_accuracy": 0.11446523144841195,
|
|
"num_tokens": 1868765.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.50655608177185,
|
|
"epoch": 0.05520180753825614,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.4522,
|
|
"mean_token_accuracy": 0.10780561491847038,
|
|
"num_tokens": 1878148.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.428852844238281,
|
|
"epoch": 0.05545856013145733,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.3975,
|
|
"mean_token_accuracy": 0.11817089915275573,
|
|
"num_tokens": 1886937.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.422956418991089,
|
|
"epoch": 0.05571531272465852,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.4018,
|
|
"mean_token_accuracy": 0.11479263305664063,
|
|
"num_tokens": 1895462.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.488995933532715,
|
|
"epoch": 0.05597206531785971,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.3837,
|
|
"mean_token_accuracy": 0.11902636885643006,
|
|
"num_tokens": 1904435.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.522593832015991,
|
|
"epoch": 0.0562288179110609,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.6353,
|
|
"mean_token_accuracy": 0.10632986500859261,
|
|
"num_tokens": 1913499.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.686061096191406,
|
|
"epoch": 0.056485570504262096,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.475,
|
|
"mean_token_accuracy": 0.11396049112081527,
|
|
"num_tokens": 1921873.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.40611891746521,
|
|
"epoch": 0.056742323097463285,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.4206,
|
|
"mean_token_accuracy": 0.11510388776659966,
|
|
"num_tokens": 1930847.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.406865930557251,
|
|
"epoch": 0.05699907569066447,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.3278,
|
|
"mean_token_accuracy": 0.11729627773165703,
|
|
"num_tokens": 1939017.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.634246063232422,
|
|
"epoch": 0.05725582828386567,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.5296,
|
|
"mean_token_accuracy": 0.11117232665419578,
|
|
"num_tokens": 1948899.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.325379371643066,
|
|
"epoch": 0.05751258087706686,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.3065,
|
|
"mean_token_accuracy": 0.12116943001747131,
|
|
"num_tokens": 1957416.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.503693151473999,
|
|
"epoch": 0.05776933347026805,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.4213,
|
|
"mean_token_accuracy": 0.1151343896985054,
|
|
"num_tokens": 1966744.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.406322431564331,
|
|
"epoch": 0.05802608606346924,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.4292,
|
|
"mean_token_accuracy": 0.11549896523356437,
|
|
"num_tokens": 1975137.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.409404754638672,
|
|
"epoch": 0.05828283865667043,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.11963799446821213,
|
|
"num_tokens": 1983714.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.475473785400391,
|
|
"epoch": 0.058539591249871625,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.4249,
|
|
"mean_token_accuracy": 0.11401411071419716,
|
|
"num_tokens": 1991684.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.383267736434936,
|
|
"epoch": 0.058796343843072814,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.4118,
|
|
"mean_token_accuracy": 0.11849818304181099,
|
|
"num_tokens": 1999565.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.512342691421509,
|
|
"epoch": 0.05905309643627401,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.4244,
|
|
"mean_token_accuracy": 0.11193430349230767,
|
|
"num_tokens": 2008249.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.468747138977051,
|
|
"epoch": 0.0593098490294752,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.4496,
|
|
"mean_token_accuracy": 0.11334523633122444,
|
|
"num_tokens": 2017563.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.578333950042724,
|
|
"epoch": 0.059566601622676386,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.5051,
|
|
"mean_token_accuracy": 0.11496832743287086,
|
|
"num_tokens": 2026433.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.4164698123931885,
|
|
"epoch": 0.05982335421587758,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.3786,
|
|
"mean_token_accuracy": 0.11880086958408356,
|
|
"num_tokens": 2034576.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.382039260864258,
|
|
"epoch": 0.06008010680907877,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.3257,
|
|
"mean_token_accuracy": 0.11798683404922486,
|
|
"num_tokens": 2042987.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.490727043151855,
|
|
"epoch": 0.060336859402279966,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.3846,
|
|
"mean_token_accuracy": 0.11852361485362053,
|
|
"num_tokens": 2051342.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.442277240753174,
|
|
"epoch": 0.060593611995481154,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.3965,
|
|
"mean_token_accuracy": 0.11326743885874749,
|
|
"num_tokens": 2060116.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.406348848342896,
|
|
"epoch": 0.06085036458868234,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.4109,
|
|
"mean_token_accuracy": 0.11511619538068771,
|
|
"num_tokens": 2067927.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.500762605667115,
|
|
"epoch": 0.06110711718188354,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.4757,
|
|
"mean_token_accuracy": 0.11053126454353332,
|
|
"num_tokens": 2076459.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.4877848625183105,
|
|
"epoch": 0.06136386977508473,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.4654,
|
|
"mean_token_accuracy": 0.11117196828126907,
|
|
"num_tokens": 2085762.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.420598840713501,
|
|
"epoch": 0.06162062236828592,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.4063,
|
|
"mean_token_accuracy": 0.11609504669904709,
|
|
"num_tokens": 2094030.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.540357208251953,
|
|
"epoch": 0.06187737496148711,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.11226271614432334,
|
|
"num_tokens": 2103282.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.325323581695557,
|
|
"epoch": 0.0621341275546883,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.2123,
|
|
"mean_token_accuracy": 0.12576676830649375,
|
|
"num_tokens": 2111630.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.393415832519532,
|
|
"epoch": 0.062390880147889495,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.3089,
|
|
"mean_token_accuracy": 0.11810488030314445,
|
|
"num_tokens": 2120817.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.341946268081665,
|
|
"epoch": 0.06264763274109068,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.2447,
|
|
"mean_token_accuracy": 0.12075709477066994,
|
|
"num_tokens": 2128646.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.490839147567749,
|
|
"epoch": 0.06290438533429188,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.4481,
|
|
"mean_token_accuracy": 0.11171429008245468,
|
|
"num_tokens": 2137507.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.387483739852906,
|
|
"epoch": 0.06316113792749307,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.3125,
|
|
"mean_token_accuracy": 0.11729855984449386,
|
|
"num_tokens": 2146970.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.364808559417725,
|
|
"epoch": 0.06341789052069426,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.2984,
|
|
"mean_token_accuracy": 0.11805230528116226,
|
|
"num_tokens": 2155342.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.433910751342774,
|
|
"epoch": 0.06367464311389545,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.3647,
|
|
"mean_token_accuracy": 0.1170346699655056,
|
|
"num_tokens": 2164507.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.456918668746948,
|
|
"epoch": 0.06393139570709665,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.1171228475868702,
|
|
"num_tokens": 2172350.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.372973251342773,
|
|
"epoch": 0.06418814830029783,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.3883,
|
|
"mean_token_accuracy": 0.11522075980901718,
|
|
"num_tokens": 2180584.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.370125436782837,
|
|
"epoch": 0.06444490089349902,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.3265,
|
|
"mean_token_accuracy": 0.11535610556602478,
|
|
"num_tokens": 2189499.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.412319755554199,
|
|
"epoch": 0.06470165348670022,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.3127,
|
|
"mean_token_accuracy": 0.12174249142408371,
|
|
"num_tokens": 2198577.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.297476816177368,
|
|
"epoch": 0.0649584060799014,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.3288,
|
|
"mean_token_accuracy": 0.1258082590997219,
|
|
"num_tokens": 2208099.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.393898057937622,
|
|
"epoch": 0.0652151586731026,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.3107,
|
|
"mean_token_accuracy": 0.12669994533061982,
|
|
"num_tokens": 2216238.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.38668007850647,
|
|
"epoch": 0.06547191126630379,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.372,
|
|
"mean_token_accuracy": 0.12261941879987717,
|
|
"num_tokens": 2224660.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.34668459892273,
|
|
"epoch": 0.06572866385950499,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.3128,
|
|
"mean_token_accuracy": 0.11945455446839333,
|
|
"num_tokens": 2233076.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.409527063369751,
|
|
"epoch": 0.06598541645270617,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.2403,
|
|
"mean_token_accuracy": 0.12162940129637718,
|
|
"num_tokens": 2241309.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.339578104019165,
|
|
"epoch": 0.06624216904590736,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.2225,
|
|
"mean_token_accuracy": 0.11572879031300545,
|
|
"num_tokens": 2250067.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.233476257324218,
|
|
"epoch": 0.06649892163910856,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.3205,
|
|
"mean_token_accuracy": 0.11559544280171394,
|
|
"num_tokens": 2258779.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.405807733535767,
|
|
"epoch": 0.06675567423230974,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.3115,
|
|
"mean_token_accuracy": 0.12557133212685584,
|
|
"num_tokens": 2267984.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.351101875305176,
|
|
"epoch": 0.06701242682551094,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.309,
|
|
"mean_token_accuracy": 0.11647475138306618,
|
|
"num_tokens": 2276470.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.331909942626953,
|
|
"epoch": 0.06726917941871213,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.2896,
|
|
"mean_token_accuracy": 0.11855655312538146,
|
|
"num_tokens": 2284708.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.454643774032593,
|
|
"epoch": 0.06752593201191331,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.3283,
|
|
"mean_token_accuracy": 0.12151680663228034,
|
|
"num_tokens": 2292735.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.204592227935791,
|
|
"epoch": 0.06778268460511451,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.163,
|
|
"mean_token_accuracy": 0.12357690259814262,
|
|
"num_tokens": 2301677.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.460263586044311,
|
|
"epoch": 0.0680394371983157,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.3441,
|
|
"mean_token_accuracy": 0.11785683333873749,
|
|
"num_tokens": 2310508.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.258358383178711,
|
|
"epoch": 0.0682961897915169,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.2408,
|
|
"mean_token_accuracy": 0.12048257812857628,
|
|
"num_tokens": 2318896.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.306467294692993,
|
|
"epoch": 0.06855294238471808,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.181,
|
|
"mean_token_accuracy": 0.12324659302830696,
|
|
"num_tokens": 2327613.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.275306177139282,
|
|
"epoch": 0.06880969497791928,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.3672,
|
|
"mean_token_accuracy": 0.11774554029107094,
|
|
"num_tokens": 2336854.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.333629369735718,
|
|
"epoch": 0.06906644757112047,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.1675,
|
|
"mean_token_accuracy": 0.12430970817804336,
|
|
"num_tokens": 2345478.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.202389335632324,
|
|
"epoch": 0.06932320016432165,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.2733,
|
|
"mean_token_accuracy": 0.12793459072709085,
|
|
"num_tokens": 2354373.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.342656087875366,
|
|
"epoch": 0.06957995275752285,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.2887,
|
|
"mean_token_accuracy": 0.11841219067573547,
|
|
"num_tokens": 2363105.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.161898565292359,
|
|
"epoch": 0.06983670535072405,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.174,
|
|
"mean_token_accuracy": 0.12459764629602432,
|
|
"num_tokens": 2371442.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.333646392822265,
|
|
"epoch": 0.07009345794392523,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.1592,
|
|
"mean_token_accuracy": 0.12238810807466508,
|
|
"num_tokens": 2379447.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.283500528335571,
|
|
"epoch": 0.07035021053712642,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.2855,
|
|
"mean_token_accuracy": 0.12455552518367767,
|
|
"num_tokens": 2387490.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.338915586471558,
|
|
"epoch": 0.07060696313032762,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.1738,
|
|
"mean_token_accuracy": 0.12791258618235588,
|
|
"num_tokens": 2396000.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.311189413070679,
|
|
"epoch": 0.07086371572352881,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.3515,
|
|
"mean_token_accuracy": 0.11353749558329582,
|
|
"num_tokens": 2404563.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.31261568069458,
|
|
"epoch": 0.07112046831673,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.2486,
|
|
"mean_token_accuracy": 0.1253374069929123,
|
|
"num_tokens": 2413268.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.326612091064453,
|
|
"epoch": 0.07137722090993119,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.2615,
|
|
"mean_token_accuracy": 0.1216346226632595,
|
|
"num_tokens": 2422043.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.290171957015991,
|
|
"epoch": 0.07163397350313239,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.3194,
|
|
"mean_token_accuracy": 0.11611472964286804,
|
|
"num_tokens": 2430538.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.288091945648193,
|
|
"epoch": 0.07189072609633357,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.1166,
|
|
"mean_token_accuracy": 0.13205000013113022,
|
|
"num_tokens": 2438785.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.248751878738403,
|
|
"epoch": 0.07214747868953476,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.2229,
|
|
"mean_token_accuracy": 0.12328648641705513,
|
|
"num_tokens": 2446849.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.309422874450684,
|
|
"epoch": 0.07240423128273596,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.2255,
|
|
"mean_token_accuracy": 0.12241806387901306,
|
|
"num_tokens": 2455903.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.227021360397339,
|
|
"epoch": 0.07266098387593714,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.2589,
|
|
"mean_token_accuracy": 0.11907726377248765,
|
|
"num_tokens": 2464398.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.342584133148193,
|
|
"epoch": 0.07291773646913834,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.316,
|
|
"mean_token_accuracy": 0.12059093490242959,
|
|
"num_tokens": 2473284.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.17990255355835,
|
|
"epoch": 0.07317448906233953,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.1969,
|
|
"mean_token_accuracy": 0.12357416152954101,
|
|
"num_tokens": 2481733.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.3772382736206055,
|
|
"epoch": 0.07343124165554073,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.2932,
|
|
"mean_token_accuracy": 0.12297726944088935,
|
|
"num_tokens": 2490706.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.213703632354736,
|
|
"epoch": 0.07368799424874191,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.2487,
|
|
"mean_token_accuracy": 0.1286434106528759,
|
|
"num_tokens": 2498984.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.365205860137939,
|
|
"epoch": 0.0739447468419431,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.2959,
|
|
"mean_token_accuracy": 0.12014952227473259,
|
|
"num_tokens": 2508129.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.198501920700073,
|
|
"epoch": 0.0742014994351443,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.1956,
|
|
"mean_token_accuracy": 0.12250755354762077,
|
|
"num_tokens": 2516535.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.3289893627166744,
|
|
"epoch": 0.07445825202834548,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.2274,
|
|
"mean_token_accuracy": 0.1266847014427185,
|
|
"num_tokens": 2525015.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.2744628429412845,
|
|
"epoch": 0.07471500462154668,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.1403,
|
|
"mean_token_accuracy": 0.133008973300457,
|
|
"num_tokens": 2532756.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.213056516647339,
|
|
"epoch": 0.07497175721474787,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.2403,
|
|
"mean_token_accuracy": 0.12591897025704385,
|
|
"num_tokens": 2541583.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.295798873901367,
|
|
"epoch": 0.07522850980794905,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.2802,
|
|
"mean_token_accuracy": 0.11927842423319816,
|
|
"num_tokens": 2549752.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.239460611343384,
|
|
"epoch": 0.07548526240115025,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.0859,
|
|
"mean_token_accuracy": 0.13107694461941719,
|
|
"num_tokens": 2557483.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.108071613311767,
|
|
"epoch": 0.07574201499435144,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.0653,
|
|
"mean_token_accuracy": 0.1297874391078949,
|
|
"num_tokens": 2565451.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.2041295051574705,
|
|
"epoch": 0.07599876758755264,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.1683,
|
|
"mean_token_accuracy": 0.1201373040676117,
|
|
"num_tokens": 2574547.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.195542526245117,
|
|
"epoch": 0.07625552018075382,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.1353,
|
|
"mean_token_accuracy": 0.12182574421167373,
|
|
"num_tokens": 2583803.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.2032914638519285,
|
|
"epoch": 0.07651227277395502,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.14,
|
|
"mean_token_accuracy": 0.13161938413977622,
|
|
"num_tokens": 2591903.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.228953456878662,
|
|
"epoch": 0.07676902536715621,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.2311,
|
|
"mean_token_accuracy": 0.12103996947407722,
|
|
"num_tokens": 2601531.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.270084095001221,
|
|
"epoch": 0.0770257779603574,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.2251,
|
|
"mean_token_accuracy": 0.12533185482025147,
|
|
"num_tokens": 2610140.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.324096918106079,
|
|
"epoch": 0.07728253055355859,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.2175,
|
|
"mean_token_accuracy": 0.12491021826863288,
|
|
"num_tokens": 2618593.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.302452754974365,
|
|
"epoch": 0.07753928314675979,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.2554,
|
|
"mean_token_accuracy": 0.12762303799390792,
|
|
"num_tokens": 2626711.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.218322849273681,
|
|
"epoch": 0.07779603573996097,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.1139,
|
|
"mean_token_accuracy": 0.1306075021624565,
|
|
"num_tokens": 2634939.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.2217262268066404,
|
|
"epoch": 0.07805278833316216,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.1224,
|
|
"mean_token_accuracy": 0.1293753780424595,
|
|
"num_tokens": 2642706.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.121373558044434,
|
|
"epoch": 0.07830954092636336,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.0573,
|
|
"mean_token_accuracy": 0.13120718002319337,
|
|
"num_tokens": 2651081.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.155263805389405,
|
|
"epoch": 0.07856629351956455,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.15,
|
|
"mean_token_accuracy": 0.12519086971879007,
|
|
"num_tokens": 2659812.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.2153678894042965,
|
|
"epoch": 0.07882304611276574,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.1721,
|
|
"mean_token_accuracy": 0.12547078728675842,
|
|
"num_tokens": 2668346.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.2948911666870115,
|
|
"epoch": 0.07907979870596693,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.2907,
|
|
"mean_token_accuracy": 0.11884583830833435,
|
|
"num_tokens": 2677228.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.2207049369812015,
|
|
"epoch": 0.07933655129916813,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.2393,
|
|
"mean_token_accuracy": 0.12552651911973953,
|
|
"num_tokens": 2686784.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.382667350769043,
|
|
"epoch": 0.07959330389236931,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.2751,
|
|
"mean_token_accuracy": 0.11902871504426002,
|
|
"num_tokens": 2694974.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.177761745452881,
|
|
"epoch": 0.0798500564855705,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.12580612376332284,
|
|
"num_tokens": 2703765.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.284360456466675,
|
|
"epoch": 0.0801068090787717,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.2204,
|
|
"mean_token_accuracy": 0.12737269848585128,
|
|
"num_tokens": 2712041.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.261340951919555,
|
|
"epoch": 0.08036356167197288,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.2617,
|
|
"mean_token_accuracy": 0.12319499254226685,
|
|
"num_tokens": 2721282.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.276726961135864,
|
|
"epoch": 0.08062031426517408,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.1825,
|
|
"mean_token_accuracy": 0.12500676587224008,
|
|
"num_tokens": 2730795.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.226504564285278,
|
|
"epoch": 0.08087706685837527,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.2397,
|
|
"mean_token_accuracy": 0.12448726296424865,
|
|
"num_tokens": 2739451.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.257420444488526,
|
|
"epoch": 0.08113381945157647,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.2542,
|
|
"mean_token_accuracy": 0.12350198328495025,
|
|
"num_tokens": 2747110.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.191071271896362,
|
|
"epoch": 0.08139057204477765,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.156,
|
|
"mean_token_accuracy": 0.12809979766607285,
|
|
"num_tokens": 2755511.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.29031810760498,
|
|
"epoch": 0.08164732463797884,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.2085,
|
|
"mean_token_accuracy": 0.12493144571781159,
|
|
"num_tokens": 2764820.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.13084306716919,
|
|
"epoch": 0.08190407723118004,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.118,
|
|
"mean_token_accuracy": 0.12518565431237222,
|
|
"num_tokens": 2773809.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.176386451721191,
|
|
"epoch": 0.08216082982438122,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.1298,
|
|
"mean_token_accuracy": 0.12971302568912507,
|
|
"num_tokens": 2781909.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.283403730392456,
|
|
"epoch": 0.08241758241758242,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.13305481374263764,
|
|
"num_tokens": 2790119.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.198763990402222,
|
|
"epoch": 0.08267433501078361,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.1821,
|
|
"mean_token_accuracy": 0.12131287008523942,
|
|
"num_tokens": 2799401.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.302814197540283,
|
|
"epoch": 0.08293108760398481,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.2436,
|
|
"mean_token_accuracy": 0.12349063828587532,
|
|
"num_tokens": 2807281.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.109517383575439,
|
|
"epoch": 0.08318784019718599,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.0774,
|
|
"mean_token_accuracy": 0.13202504813671112,
|
|
"num_tokens": 2815553.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.236009979248047,
|
|
"epoch": 0.08344459279038718,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.2218,
|
|
"mean_token_accuracy": 0.12722306177020073,
|
|
"num_tokens": 2824085.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.243375015258789,
|
|
"epoch": 0.08370134538358838,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.1691,
|
|
"mean_token_accuracy": 0.1275683157145977,
|
|
"num_tokens": 2833345.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.2176196575164795,
|
|
"epoch": 0.08395809797678956,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.1129,
|
|
"mean_token_accuracy": 0.13100848644971846,
|
|
"num_tokens": 2842273.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.184139060974121,
|
|
"epoch": 0.08421485056999076,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.2057,
|
|
"mean_token_accuracy": 0.12947914898395538,
|
|
"num_tokens": 2850749.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.292967128753662,
|
|
"epoch": 0.08447160316319195,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.1981,
|
|
"mean_token_accuracy": 0.12301181033253669,
|
|
"num_tokens": 2859028.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.167311716079712,
|
|
"epoch": 0.08472835575639313,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.1501,
|
|
"mean_token_accuracy": 0.12906862869858743,
|
|
"num_tokens": 2868361.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.265577983856201,
|
|
"epoch": 0.08498510834959433,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.1631,
|
|
"mean_token_accuracy": 0.12621149942278861,
|
|
"num_tokens": 2877177.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.2307453632354735,
|
|
"epoch": 0.08524186094279552,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.3026,
|
|
"mean_token_accuracy": 0.12135942950844765,
|
|
"num_tokens": 2884981.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.234720849990845,
|
|
"epoch": 0.08549861353599672,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.1210001491010189,
|
|
"num_tokens": 2894083.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.142803907394409,
|
|
"epoch": 0.0857553661291979,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.0632,
|
|
"mean_token_accuracy": 0.12406837791204453,
|
|
"num_tokens": 2903037.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.219666337966919,
|
|
"epoch": 0.0860121187223991,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.1582,
|
|
"mean_token_accuracy": 0.12801609188318253,
|
|
"num_tokens": 2911526.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.192300844192505,
|
|
"epoch": 0.08626887131560029,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.123,
|
|
"mean_token_accuracy": 0.12801995873451233,
|
|
"num_tokens": 2920093.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.0313129901885985,
|
|
"epoch": 0.08652562390880147,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 5.9484,
|
|
"mean_token_accuracy": 0.14002343714237214,
|
|
"num_tokens": 2928085.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.169588708877564,
|
|
"epoch": 0.08678237650200267,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.1155,
|
|
"mean_token_accuracy": 0.12732220590114593,
|
|
"num_tokens": 2936152.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.262211561203003,
|
|
"epoch": 0.08703912909520387,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.0744,
|
|
"mean_token_accuracy": 0.13020945638418197,
|
|
"num_tokens": 2944194.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.041778326034546,
|
|
"epoch": 0.08729588168840505,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.1143,
|
|
"mean_token_accuracy": 0.13238055855035782,
|
|
"num_tokens": 2952807.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.178655624389648,
|
|
"epoch": 0.08755263428160624,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.0652,
|
|
"mean_token_accuracy": 0.1316159836947918,
|
|
"num_tokens": 2961607.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 5.95141224861145,
|
|
"epoch": 0.08780938687480744,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 5.9293,
|
|
"mean_token_accuracy": 0.13691822439432144,
|
|
"num_tokens": 2969332.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.2237059593200685,
|
|
"epoch": 0.08806613946800863,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.1378,
|
|
"mean_token_accuracy": 0.1292878992855549,
|
|
"num_tokens": 2977751.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.142322206497193,
|
|
"epoch": 0.08832289206120982,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.036,
|
|
"mean_token_accuracy": 0.13863547071814536,
|
|
"num_tokens": 2985830.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.259805917739868,
|
|
"epoch": 0.08857964465441101,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.2145,
|
|
"mean_token_accuracy": 0.12537185996770858,
|
|
"num_tokens": 2994378.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.175342226028443,
|
|
"epoch": 0.0888363972476122,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.2367,
|
|
"mean_token_accuracy": 0.1263650104403496,
|
|
"num_tokens": 3003736.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.163015604019165,
|
|
"epoch": 0.08909314984081339,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.0531,
|
|
"mean_token_accuracy": 0.13277990147471427,
|
|
"num_tokens": 3011558.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.271158742904663,
|
|
"epoch": 0.08934990243401458,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.1885,
|
|
"mean_token_accuracy": 0.12793701142072678,
|
|
"num_tokens": 3019081.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.209322929382324,
|
|
"epoch": 0.08960665502721578,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.2158,
|
|
"mean_token_accuracy": 0.12215240448713302,
|
|
"num_tokens": 3028118.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.3456298351287845,
|
|
"epoch": 0.08986340762041696,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.1534,
|
|
"mean_token_accuracy": 0.12558385729789734,
|
|
"num_tokens": 3036752.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.289044666290283,
|
|
"epoch": 0.09012016021361816,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.317,
|
|
"mean_token_accuracy": 0.1205628290772438,
|
|
"num_tokens": 3045909.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.170148277282715,
|
|
"epoch": 0.09037691280681935,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.0963,
|
|
"mean_token_accuracy": 0.1319189563393593,
|
|
"num_tokens": 3054903.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.085113954544068,
|
|
"epoch": 0.09063366540002055,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.0327,
|
|
"mean_token_accuracy": 0.1328284077346325,
|
|
"num_tokens": 3063952.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.188023567199707,
|
|
"epoch": 0.09089041799322173,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 5.986,
|
|
"mean_token_accuracy": 0.14162241145968438,
|
|
"num_tokens": 3072249.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.101192378997803,
|
|
"epoch": 0.09114717058642292,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.12976548150181771,
|
|
"num_tokens": 3080947.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.135991859436035,
|
|
"epoch": 0.09140392317962412,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.0845,
|
|
"mean_token_accuracy": 0.1314239852130413,
|
|
"num_tokens": 3090021.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.107022285461426,
|
|
"epoch": 0.0916606757728253,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.0386,
|
|
"mean_token_accuracy": 0.13592078164219856,
|
|
"num_tokens": 3098233.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.061567544937134,
|
|
"epoch": 0.0919174283660265,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.09,
|
|
"mean_token_accuracy": 0.12940324023365973,
|
|
"num_tokens": 3106591.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.208729648590088,
|
|
"epoch": 0.09217418095922769,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.0886,
|
|
"mean_token_accuracy": 0.12878676876425743,
|
|
"num_tokens": 3115263.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.139681434631347,
|
|
"epoch": 0.09243093355242887,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.0626,
|
|
"mean_token_accuracy": 0.13254646211862564,
|
|
"num_tokens": 3125057.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.124689769744873,
|
|
"epoch": 0.09268768614563007,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.0195,
|
|
"mean_token_accuracy": 0.1393231302499771,
|
|
"num_tokens": 3133515.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.106488132476807,
|
|
"epoch": 0.09294443873883126,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.0244,
|
|
"mean_token_accuracy": 0.13584161549806595,
|
|
"num_tokens": 3142426.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.163878440856934,
|
|
"epoch": 0.09320119133203246,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.1113,
|
|
"mean_token_accuracy": 0.13173030391335488,
|
|
"num_tokens": 3150800.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.0394158363342285,
|
|
"epoch": 0.09345794392523364,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.0633,
|
|
"mean_token_accuracy": 0.1318834140896797,
|
|
"num_tokens": 3159830.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.2027496814727785,
|
|
"epoch": 0.09371469651843484,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.0734,
|
|
"mean_token_accuracy": 0.13092744797468187,
|
|
"num_tokens": 3168335.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.107715177536011,
|
|
"epoch": 0.09397144911163603,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.1385,
|
|
"mean_token_accuracy": 0.1300662524998188,
|
|
"num_tokens": 3177637.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.1835253715515135,
|
|
"epoch": 0.09422820170483721,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.0827,
|
|
"mean_token_accuracy": 0.1363433077931404,
|
|
"num_tokens": 3185813.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.047970199584961,
|
|
"epoch": 0.09448495429803841,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 5.9488,
|
|
"mean_token_accuracy": 0.1319119691848755,
|
|
"num_tokens": 3194502.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.201473045349121,
|
|
"epoch": 0.0947417068912396,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.1253,
|
|
"mean_token_accuracy": 0.12627599611878396,
|
|
"num_tokens": 3204066.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.141813898086548,
|
|
"epoch": 0.09499845948444079,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.1072,
|
|
"mean_token_accuracy": 0.13326140269637107,
|
|
"num_tokens": 3213995.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.162443733215332,
|
|
"epoch": 0.09525521207764198,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.1113,
|
|
"mean_token_accuracy": 0.1338210202753544,
|
|
"num_tokens": 3221715.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.183711290359497,
|
|
"epoch": 0.09551196467084318,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.0889,
|
|
"mean_token_accuracy": 0.12752361968159676,
|
|
"num_tokens": 3230365.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.042427253723145,
|
|
"epoch": 0.09576871726404437,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.0294,
|
|
"mean_token_accuracy": 0.1378671109676361,
|
|
"num_tokens": 3237893.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.10888147354126,
|
|
"epoch": 0.09602546985724555,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 5.9578,
|
|
"mean_token_accuracy": 0.13665952011942864,
|
|
"num_tokens": 3245999.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.065496253967285,
|
|
"epoch": 0.09628222245044675,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.0933,
|
|
"mean_token_accuracy": 0.13585855215787887,
|
|
"num_tokens": 3254434.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.142561149597168,
|
|
"epoch": 0.09653897504364795,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.0191,
|
|
"mean_token_accuracy": 0.13392399698495866,
|
|
"num_tokens": 3262427.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.090549612045288,
|
|
"epoch": 0.09679572763684913,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.0398,
|
|
"mean_token_accuracy": 0.1328475646674633,
|
|
"num_tokens": 3270832.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.0752910614013675,
|
|
"epoch": 0.09705248023005032,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.0441,
|
|
"mean_token_accuracy": 0.12893078327178956,
|
|
"num_tokens": 3279923.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.122604894638061,
|
|
"epoch": 0.09730923282325152,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.0172,
|
|
"mean_token_accuracy": 0.13738859295845032,
|
|
"num_tokens": 3289300.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.088231611251831,
|
|
"epoch": 0.0975659854164527,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.0629,
|
|
"mean_token_accuracy": 0.13484949991106987,
|
|
"num_tokens": 3297685.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.1511658191680905,
|
|
"epoch": 0.0978227380096539,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.0363,
|
|
"mean_token_accuracy": 0.13940526247024537,
|
|
"num_tokens": 3305621.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.108269453048706,
|
|
"epoch": 0.09807949060285509,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.0858,
|
|
"mean_token_accuracy": 0.13032101839780807,
|
|
"num_tokens": 3314342.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.149828338623047,
|
|
"epoch": 0.09833624319605629,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.0672,
|
|
"mean_token_accuracy": 0.12796618640422822,
|
|
"num_tokens": 3323377.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.012651634216309,
|
|
"epoch": 0.09859299578925747,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.0081,
|
|
"mean_token_accuracy": 0.13904084265232086,
|
|
"num_tokens": 3332521.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.131260681152344,
|
|
"epoch": 0.09884974838245866,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.1306,
|
|
"mean_token_accuracy": 0.12497629895806313,
|
|
"num_tokens": 3341377.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.23424220085144,
|
|
"epoch": 0.09910650097565986,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.1519,
|
|
"mean_token_accuracy": 0.12516835778951646,
|
|
"num_tokens": 3350220.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.0826483249664305,
|
|
"epoch": 0.09936325356886104,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 5.9715,
|
|
"mean_token_accuracy": 0.13165306746959687,
|
|
"num_tokens": 3359288.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.136231851577759,
|
|
"epoch": 0.09962000616206224,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.017,
|
|
"mean_token_accuracy": 0.13045912608504295,
|
|
"num_tokens": 3368302.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.128415536880493,
|
|
"epoch": 0.09987675875526343,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.1161,
|
|
"mean_token_accuracy": 0.1268952466547489,
|
|
"num_tokens": 3376728.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.145467138290405,
|
|
"epoch": 0.10013351134846461,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 5.9857,
|
|
"mean_token_accuracy": 0.14468327164649963,
|
|
"num_tokens": 3385390.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.0337213516235355,
|
|
"epoch": 0.10039026394166581,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 5.991,
|
|
"mean_token_accuracy": 0.13441295102238654,
|
|
"num_tokens": 3393652.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.218010187149048,
|
|
"epoch": 0.100647016534867,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.0806,
|
|
"mean_token_accuracy": 0.12356908470392228,
|
|
"num_tokens": 3403430.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 5.981335592269898,
|
|
"epoch": 0.1009037691280682,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 5.9878,
|
|
"mean_token_accuracy": 0.13102470636367797,
|
|
"num_tokens": 3412222.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.074318742752075,
|
|
"epoch": 0.10116052172126938,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 5.9248,
|
|
"mean_token_accuracy": 0.13719005659222602,
|
|
"num_tokens": 3421007.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.133344888687134,
|
|
"epoch": 0.10141727431447058,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.0251,
|
|
"mean_token_accuracy": 0.13687873035669326,
|
|
"num_tokens": 3430124.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 5.925088357925415,
|
|
"epoch": 0.10167402690767177,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 5.9342,
|
|
"mean_token_accuracy": 0.13773764222860335,
|
|
"num_tokens": 3439194.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.211205673217774,
|
|
"epoch": 0.10193077950087295,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.105,
|
|
"mean_token_accuracy": 0.12946683093905448,
|
|
"num_tokens": 3448562.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.188286209106446,
|
|
"epoch": 0.10218753209407415,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.016,
|
|
"mean_token_accuracy": 0.1287774696946144,
|
|
"num_tokens": 3457314.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.008145046234131,
|
|
"epoch": 0.10244428468727534,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.139,
|
|
"mean_token_accuracy": 0.1262603260576725,
|
|
"num_tokens": 3466566.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.213902473449707,
|
|
"epoch": 0.10270103728047653,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.0595,
|
|
"mean_token_accuracy": 0.13240915760397912,
|
|
"num_tokens": 3475404.0,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 4000,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 1,
|
|
"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": 4694960553984000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|