2035 lines
54 KiB
JSON
2035 lines
54 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.05135051864023826,
|
|
"eval_steps": 500,
|
|
"global_step": 1000,
|
|
"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
|
|
}
|
|
],
|
|
"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": 2344075315200000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|