5035 lines
136 KiB
JSON
5035 lines
136 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.14612192413349698,
|
|
"eval_steps": 500,
|
|
"global_step": 2500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.691908264160157,
|
|
"epoch": 0.000292243848266994,
|
|
"grad_norm": 12.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.8406,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 11114.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.691956710815429,
|
|
"epoch": 0.000584487696533988,
|
|
"grad_norm": 13.375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7353,
|
|
"mean_token_accuracy": 0.00011737089371308685,
|
|
"num_tokens": 20935.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.690714645385743,
|
|
"epoch": 0.000876731544800982,
|
|
"grad_norm": 9.4375,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5144,
|
|
"mean_token_accuracy": 0.03050256703281775,
|
|
"num_tokens": 31930.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.654311275482177,
|
|
"epoch": 0.001168975393067976,
|
|
"grad_norm": 5.96875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.1606,
|
|
"mean_token_accuracy": 0.05743609480559826,
|
|
"num_tokens": 41672.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.475521564483643,
|
|
"epoch": 0.00146121924133497,
|
|
"grad_norm": 3.90625,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 9.9607,
|
|
"mean_token_accuracy": 0.05290710777044296,
|
|
"num_tokens": 52510.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.543301010131836,
|
|
"epoch": 0.001753463089601964,
|
|
"grad_norm": 3.75,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 9.8154,
|
|
"mean_token_accuracy": 0.045297696441411975,
|
|
"num_tokens": 63214.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.49622220993042,
|
|
"epoch": 0.002045706937868958,
|
|
"grad_norm": 2.5,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 9.7008,
|
|
"mean_token_accuracy": 0.05302377603948116,
|
|
"num_tokens": 73942.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.475418663024902,
|
|
"epoch": 0.002337950786135952,
|
|
"grad_norm": 2.375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 9.7051,
|
|
"mean_token_accuracy": 0.05076396390795708,
|
|
"num_tokens": 84526.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.52062005996704,
|
|
"epoch": 0.0026301946344029457,
|
|
"grad_norm": 2.234375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.7003,
|
|
"mean_token_accuracy": 0.04803669825196266,
|
|
"num_tokens": 95178.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.508524227142335,
|
|
"epoch": 0.00292243848266994,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.6047,
|
|
"mean_token_accuracy": 0.05312369242310524,
|
|
"num_tokens": 106653.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.445098209381104,
|
|
"epoch": 0.0032146823309369336,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.4922,
|
|
"mean_token_accuracy": 0.058677663654088975,
|
|
"num_tokens": 115969.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.33490447998047,
|
|
"epoch": 0.003506926179203928,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.469,
|
|
"mean_token_accuracy": 0.06182386502623558,
|
|
"num_tokens": 125826.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.331936168670655,
|
|
"epoch": 0.0037991700274709215,
|
|
"grad_norm": 2.3125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.4398,
|
|
"mean_token_accuracy": 0.06020851917564869,
|
|
"num_tokens": 136677.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.381442165374756,
|
|
"epoch": 0.004091413875737916,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.3696,
|
|
"mean_token_accuracy": 0.05836572460830212,
|
|
"num_tokens": 147535.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.336608695983887,
|
|
"epoch": 0.00438365772400491,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.2591,
|
|
"mean_token_accuracy": 0.06145334914326668,
|
|
"num_tokens": 157524.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.270106220245362,
|
|
"epoch": 0.004675901572271904,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.1922,
|
|
"mean_token_accuracy": 0.063629350066185,
|
|
"num_tokens": 168344.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.35278034210205,
|
|
"epoch": 0.004968145420538897,
|
|
"grad_norm": 1.953125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.14,
|
|
"mean_token_accuracy": 0.05551427491009235,
|
|
"num_tokens": 178976.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.23039789199829,
|
|
"epoch": 0.0052603892688058915,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.0429,
|
|
"mean_token_accuracy": 0.06621964909136295,
|
|
"num_tokens": 190567.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.196823787689208,
|
|
"epoch": 0.005552633117072886,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 8.9449,
|
|
"mean_token_accuracy": 0.07107578478753566,
|
|
"num_tokens": 201587.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.113740062713623,
|
|
"epoch": 0.00584487696533988,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 8.8501,
|
|
"mean_token_accuracy": 0.06472622714936734,
|
|
"num_tokens": 211913.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.149368000030517,
|
|
"epoch": 0.006137120813606874,
|
|
"grad_norm": 1.8046875,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 8.7762,
|
|
"mean_token_accuracy": 0.06379772424697876,
|
|
"num_tokens": 222247.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 9.989063167572022,
|
|
"epoch": 0.006429364661873867,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.6212,
|
|
"mean_token_accuracy": 0.07087173387408256,
|
|
"num_tokens": 232361.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 9.965903949737548,
|
|
"epoch": 0.006721608510140861,
|
|
"grad_norm": 1.578125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.5651,
|
|
"mean_token_accuracy": 0.06980551891028881,
|
|
"num_tokens": 242885.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 9.803161907196046,
|
|
"epoch": 0.007013852358407856,
|
|
"grad_norm": 1.453125,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.4687,
|
|
"mean_token_accuracy": 0.06787880472838878,
|
|
"num_tokens": 254482.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 9.760642433166504,
|
|
"epoch": 0.00730609620667485,
|
|
"grad_norm": 1.40625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.3483,
|
|
"mean_token_accuracy": 0.06577648222446442,
|
|
"num_tokens": 265744.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 9.514495658874512,
|
|
"epoch": 0.007598340054941843,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.2621,
|
|
"mean_token_accuracy": 0.0672954872250557,
|
|
"num_tokens": 276924.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 9.37683687210083,
|
|
"epoch": 0.007890583903208837,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.1773,
|
|
"mean_token_accuracy": 0.07101362608373166,
|
|
"num_tokens": 287779.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 9.18982000350952,
|
|
"epoch": 0.008182827751475831,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.1052,
|
|
"mean_token_accuracy": 0.07210481017827988,
|
|
"num_tokens": 298807.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 9.113053703308106,
|
|
"epoch": 0.008475071599742826,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.1007,
|
|
"mean_token_accuracy": 0.061491196230053904,
|
|
"num_tokens": 309760.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 8.824873638153075,
|
|
"epoch": 0.00876731544800982,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 7.9698,
|
|
"mean_token_accuracy": 0.0695835679769516,
|
|
"num_tokens": 319750.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 8.6828182220459,
|
|
"epoch": 0.009059559296276814,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 7.8822,
|
|
"mean_token_accuracy": 0.07002829425036908,
|
|
"num_tokens": 329931.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 8.598028564453125,
|
|
"epoch": 0.009351803144543808,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9106,
|
|
"mean_token_accuracy": 0.06758572831749916,
|
|
"num_tokens": 340457.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 8.501685905456544,
|
|
"epoch": 0.009644046992810802,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.8105,
|
|
"mean_token_accuracy": 0.07629426941275597,
|
|
"num_tokens": 352450.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 8.418387317657471,
|
|
"epoch": 0.009936290841077795,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8297,
|
|
"mean_token_accuracy": 0.07713140919804573,
|
|
"num_tokens": 363204.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.363510417938233,
|
|
"epoch": 0.010228534689344789,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8784,
|
|
"mean_token_accuracy": 0.06442252397537232,
|
|
"num_tokens": 373577.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.313191223144532,
|
|
"epoch": 0.010520778537611783,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7833,
|
|
"mean_token_accuracy": 0.07531154453754425,
|
|
"num_tokens": 384267.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.214359951019286,
|
|
"epoch": 0.010813022385878777,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7491,
|
|
"mean_token_accuracy": 0.06811888329684734,
|
|
"num_tokens": 394767.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.22794075012207,
|
|
"epoch": 0.011105266234145771,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.7665,
|
|
"mean_token_accuracy": 0.07282254844903946,
|
|
"num_tokens": 405807.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.217875862121582,
|
|
"epoch": 0.011397510082412765,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7191,
|
|
"mean_token_accuracy": 0.06757043041288853,
|
|
"num_tokens": 416381.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.152586936950684,
|
|
"epoch": 0.01168975393067976,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.8287,
|
|
"mean_token_accuracy": 0.07003305219113827,
|
|
"num_tokens": 427262.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.141828060150146,
|
|
"epoch": 0.011981997778946754,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.7651,
|
|
"mean_token_accuracy": 0.0717297799885273,
|
|
"num_tokens": 438504.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.196744251251221,
|
|
"epoch": 0.012274241627213748,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.7513,
|
|
"mean_token_accuracy": 0.07213160172104835,
|
|
"num_tokens": 449923.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.13697328567505,
|
|
"epoch": 0.01256648547548074,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.7046,
|
|
"mean_token_accuracy": 0.07478658258914947,
|
|
"num_tokens": 460736.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.1306658744812,
|
|
"epoch": 0.012858729323747735,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.8163,
|
|
"mean_token_accuracy": 0.06586915515363216,
|
|
"num_tokens": 471513.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.089940357208253,
|
|
"epoch": 0.013150973172014729,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.6608,
|
|
"mean_token_accuracy": 0.07503279075026512,
|
|
"num_tokens": 482115.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.06955795288086,
|
|
"epoch": 0.013443217020281723,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.6789,
|
|
"mean_token_accuracy": 0.07244161255657673,
|
|
"num_tokens": 493273.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.088439989089967,
|
|
"epoch": 0.013735460868548717,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.7067,
|
|
"mean_token_accuracy": 0.07398871332406998,
|
|
"num_tokens": 503939.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.001152181625367,
|
|
"epoch": 0.014027704716815711,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.6139,
|
|
"mean_token_accuracy": 0.07592850960791112,
|
|
"num_tokens": 515215.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 7.98775544166565,
|
|
"epoch": 0.014319948565082705,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.7051,
|
|
"mean_token_accuracy": 0.07671321779489518,
|
|
"num_tokens": 526485.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.02779312133789,
|
|
"epoch": 0.0146121924133497,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.6616,
|
|
"mean_token_accuracy": 0.07241038978099823,
|
|
"num_tokens": 536735.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.066807174682618,
|
|
"epoch": 0.014904436261616694,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.7528,
|
|
"mean_token_accuracy": 0.06985807269811631,
|
|
"num_tokens": 547326.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.9721503257751465,
|
|
"epoch": 0.015196680109883686,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.6367,
|
|
"mean_token_accuracy": 0.07485130280256272,
|
|
"num_tokens": 558275.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.969989395141601,
|
|
"epoch": 0.01548892395815068,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.6104,
|
|
"mean_token_accuracy": 0.0731944501399994,
|
|
"num_tokens": 569711.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.970643997192383,
|
|
"epoch": 0.015781167806417674,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.5828,
|
|
"mean_token_accuracy": 0.07749090045690536,
|
|
"num_tokens": 579602.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.9638073444366455,
|
|
"epoch": 0.01607341165468467,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.6231,
|
|
"mean_token_accuracy": 0.07638676390051842,
|
|
"num_tokens": 590899.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.931737279891967,
|
|
"epoch": 0.016365655502951663,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.6258,
|
|
"mean_token_accuracy": 0.07663728073239326,
|
|
"num_tokens": 602080.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.949186515808106,
|
|
"epoch": 0.016657899351218655,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.6619,
|
|
"mean_token_accuracy": 0.07214237600564957,
|
|
"num_tokens": 613692.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.928168773651123,
|
|
"epoch": 0.01695014319948565,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.6255,
|
|
"mean_token_accuracy": 0.07801526971161366,
|
|
"num_tokens": 625019.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.858082962036133,
|
|
"epoch": 0.017242387047752643,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.5324,
|
|
"mean_token_accuracy": 0.07193354479968547,
|
|
"num_tokens": 636453.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.900658082962036,
|
|
"epoch": 0.01753463089601964,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.5805,
|
|
"mean_token_accuracy": 0.07368646003305912,
|
|
"num_tokens": 647272.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.980489492416382,
|
|
"epoch": 0.017826874744286632,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.5929,
|
|
"mean_token_accuracy": 0.07068270668387414,
|
|
"num_tokens": 657724.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.874118614196777,
|
|
"epoch": 0.018119118592553628,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.5913,
|
|
"mean_token_accuracy": 0.07631154656410218,
|
|
"num_tokens": 669183.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.852222537994384,
|
|
"epoch": 0.01841136244082062,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.5557,
|
|
"mean_token_accuracy": 0.07173253633081914,
|
|
"num_tokens": 680439.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.902084875106811,
|
|
"epoch": 0.018703606289087616,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.6115,
|
|
"mean_token_accuracy": 0.07198056317865849,
|
|
"num_tokens": 690789.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.909448766708374,
|
|
"epoch": 0.01899585013735461,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.5829,
|
|
"mean_token_accuracy": 0.07492739222943783,
|
|
"num_tokens": 701818.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.837027454376221,
|
|
"epoch": 0.019288093985621604,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.5366,
|
|
"mean_token_accuracy": 0.07418245449662209,
|
|
"num_tokens": 713344.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.862662744522095,
|
|
"epoch": 0.019580337833888597,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.5289,
|
|
"mean_token_accuracy": 0.07505915723741055,
|
|
"num_tokens": 723279.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.862085199356079,
|
|
"epoch": 0.01987258168215559,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.5211,
|
|
"mean_token_accuracy": 0.07510280013084411,
|
|
"num_tokens": 733875.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.86326003074646,
|
|
"epoch": 0.020164825530422585,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.5595,
|
|
"mean_token_accuracy": 0.073441531509161,
|
|
"num_tokens": 745449.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.847935771942138,
|
|
"epoch": 0.020457069378689578,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.5717,
|
|
"mean_token_accuracy": 0.07788249254226684,
|
|
"num_tokens": 757570.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.815750980377198,
|
|
"epoch": 0.020749313226956573,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.4513,
|
|
"mean_token_accuracy": 0.08608488291501999,
|
|
"num_tokens": 767601.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.804518222808838,
|
|
"epoch": 0.021041557075223566,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.5134,
|
|
"mean_token_accuracy": 0.07662104293704033,
|
|
"num_tokens": 777852.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.8234796047210695,
|
|
"epoch": 0.021333800923490562,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.4728,
|
|
"mean_token_accuracy": 0.07879027798771858,
|
|
"num_tokens": 788339.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.763150691986084,
|
|
"epoch": 0.021626044771757554,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.5509,
|
|
"mean_token_accuracy": 0.07561626769602299,
|
|
"num_tokens": 799791.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.7915912628173825,
|
|
"epoch": 0.02191828862002455,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.4559,
|
|
"mean_token_accuracy": 0.07913151681423188,
|
|
"num_tokens": 810141.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.82445068359375,
|
|
"epoch": 0.022210532468291543,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.4938,
|
|
"mean_token_accuracy": 0.07601816579699516,
|
|
"num_tokens": 820929.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.806576347351074,
|
|
"epoch": 0.022502776316558535,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.4415,
|
|
"mean_token_accuracy": 0.07921079695224761,
|
|
"num_tokens": 831340.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.650366687774659,
|
|
"epoch": 0.02279502016482553,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.4444,
|
|
"mean_token_accuracy": 0.07921848297119141,
|
|
"num_tokens": 841846.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.847888851165772,
|
|
"epoch": 0.023087264013092523,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.4322,
|
|
"mean_token_accuracy": 0.0797240749001503,
|
|
"num_tokens": 852089.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.713356971740723,
|
|
"epoch": 0.02337950786135952,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.4116,
|
|
"mean_token_accuracy": 0.08028053492307663,
|
|
"num_tokens": 862568.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.74327917098999,
|
|
"epoch": 0.02367175170962651,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.4684,
|
|
"mean_token_accuracy": 0.07664722427725792,
|
|
"num_tokens": 874676.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.674819469451904,
|
|
"epoch": 0.023963995557893508,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.3829,
|
|
"mean_token_accuracy": 0.08288070000708103,
|
|
"num_tokens": 886338.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.71968150138855,
|
|
"epoch": 0.0242562394061605,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.4114,
|
|
"mean_token_accuracy": 0.07737804017961025,
|
|
"num_tokens": 896671.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.6655017852783205,
|
|
"epoch": 0.024548483254427496,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.3914,
|
|
"mean_token_accuracy": 0.07870943620800971,
|
|
"num_tokens": 908265.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.6367028713226315,
|
|
"epoch": 0.02484072710269449,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.3495,
|
|
"mean_token_accuracy": 0.0839125782251358,
|
|
"num_tokens": 919024.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.721334648132324,
|
|
"epoch": 0.02513297095096148,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.3314,
|
|
"mean_token_accuracy": 0.08483307361602783,
|
|
"num_tokens": 930837.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.614313220977783,
|
|
"epoch": 0.025425214799228477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.3096,
|
|
"mean_token_accuracy": 0.08256931453943253,
|
|
"num_tokens": 941374.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.603088331222534,
|
|
"epoch": 0.02571745864749547,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.2577,
|
|
"mean_token_accuracy": 0.08504624664783478,
|
|
"num_tokens": 953273.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.580397319793701,
|
|
"epoch": 0.026009702495762465,
|
|
"grad_norm": 1.75,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.2375,
|
|
"mean_token_accuracy": 0.0916403166949749,
|
|
"num_tokens": 964882.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.614603757858276,
|
|
"epoch": 0.026301946344029457,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.4103,
|
|
"mean_token_accuracy": 0.08159535191953182,
|
|
"num_tokens": 974732.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.595666790008545,
|
|
"epoch": 0.026594190192296453,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.3392,
|
|
"mean_token_accuracy": 0.07731116712093353,
|
|
"num_tokens": 985709.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.649027681350708,
|
|
"epoch": 0.026886434040563446,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.3048,
|
|
"mean_token_accuracy": 0.08609704971313477,
|
|
"num_tokens": 998177.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.5427803039550785,
|
|
"epoch": 0.02717867788883044,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.2407,
|
|
"mean_token_accuracy": 0.08310350701212883,
|
|
"num_tokens": 1008676.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.5991309642791744,
|
|
"epoch": 0.027470921737097434,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.357,
|
|
"mean_token_accuracy": 0.08094775602221489,
|
|
"num_tokens": 1019951.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.504101514816284,
|
|
"epoch": 0.027763165585364426,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.2908,
|
|
"mean_token_accuracy": 0.07796959653496742,
|
|
"num_tokens": 1030219.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.5679099559783936,
|
|
"epoch": 0.028055409433631422,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.3201,
|
|
"mean_token_accuracy": 0.08201361894607544,
|
|
"num_tokens": 1041911.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.6739061832427975,
|
|
"epoch": 0.028347653281898415,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.3237,
|
|
"mean_token_accuracy": 0.08364914655685425,
|
|
"num_tokens": 1052638.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5074201107025145,
|
|
"epoch": 0.02863989713016541,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.2604,
|
|
"mean_token_accuracy": 0.08207505084574222,
|
|
"num_tokens": 1063465.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.632791471481323,
|
|
"epoch": 0.028932140978432403,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.2625,
|
|
"mean_token_accuracy": 0.08835876137018203,
|
|
"num_tokens": 1073583.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.605600214004516,
|
|
"epoch": 0.0292243848266994,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.327,
|
|
"mean_token_accuracy": 0.08249956965446473,
|
|
"num_tokens": 1083961.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.432604742050171,
|
|
"epoch": 0.02951662867496639,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.1818,
|
|
"mean_token_accuracy": 0.08461034670472145,
|
|
"num_tokens": 1095339.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.534429121017456,
|
|
"epoch": 0.029808872523233387,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.2615,
|
|
"mean_token_accuracy": 0.07991603463888168,
|
|
"num_tokens": 1105518.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.572704219818116,
|
|
"epoch": 0.03010111637150038,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.1485,
|
|
"mean_token_accuracy": 0.08869004771113395,
|
|
"num_tokens": 1115666.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.499062871932983,
|
|
"epoch": 0.030393360219767372,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.2854,
|
|
"mean_token_accuracy": 0.08312782868742943,
|
|
"num_tokens": 1125827.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.49239444732666,
|
|
"epoch": 0.030685604068034368,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.2312,
|
|
"mean_token_accuracy": 0.08309667333960533,
|
|
"num_tokens": 1136154.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.477995872497559,
|
|
"epoch": 0.03097784791630136,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.2247,
|
|
"mean_token_accuracy": 0.09041514843702317,
|
|
"num_tokens": 1146392.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.515807485580444,
|
|
"epoch": 0.03127009176456835,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.2327,
|
|
"mean_token_accuracy": 0.08347614929080009,
|
|
"num_tokens": 1157566.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.408181762695312,
|
|
"epoch": 0.03156233561283535,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.2244,
|
|
"mean_token_accuracy": 0.08636801540851594,
|
|
"num_tokens": 1167795.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.55600643157959,
|
|
"epoch": 0.031854579461102345,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.1614,
|
|
"mean_token_accuracy": 0.09404403418302536,
|
|
"num_tokens": 1178894.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.379703378677368,
|
|
"epoch": 0.03214682330936934,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.1863,
|
|
"mean_token_accuracy": 0.08515683636069298,
|
|
"num_tokens": 1189361.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.44665265083313,
|
|
"epoch": 0.03243906715763633,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.2063,
|
|
"mean_token_accuracy": 0.08360610157251358,
|
|
"num_tokens": 1198992.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.450517177581787,
|
|
"epoch": 0.032731311005903325,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.1858,
|
|
"mean_token_accuracy": 0.08214344829320908,
|
|
"num_tokens": 1208732.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.416987085342408,
|
|
"epoch": 0.03302355485417032,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.1961,
|
|
"mean_token_accuracy": 0.08448583483695984,
|
|
"num_tokens": 1219015.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.383893156051636,
|
|
"epoch": 0.03331579870243731,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.0907,
|
|
"mean_token_accuracy": 0.10338108763098716,
|
|
"num_tokens": 1229227.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.4993305683135985,
|
|
"epoch": 0.033608042550704306,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.2226,
|
|
"mean_token_accuracy": 0.08231785595417022,
|
|
"num_tokens": 1239654.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.332780599594116,
|
|
"epoch": 0.0339002863989713,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.1941,
|
|
"mean_token_accuracy": 0.08100020363926888,
|
|
"num_tokens": 1250955.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.445628499984741,
|
|
"epoch": 0.0341925302472383,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.1521,
|
|
"mean_token_accuracy": 0.09030932411551476,
|
|
"num_tokens": 1262574.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.327143812179566,
|
|
"epoch": 0.03448477409550529,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.1163,
|
|
"mean_token_accuracy": 0.09456900656223297,
|
|
"num_tokens": 1273272.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.291459846496582,
|
|
"epoch": 0.03477701794377228,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.1011,
|
|
"mean_token_accuracy": 0.09272942468523979,
|
|
"num_tokens": 1284001.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.443050575256348,
|
|
"epoch": 0.03506926179203928,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.1401,
|
|
"mean_token_accuracy": 0.08779146820306778,
|
|
"num_tokens": 1294671.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.328179550170899,
|
|
"epoch": 0.035361505640306275,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.1451,
|
|
"mean_token_accuracy": 0.09261744394898415,
|
|
"num_tokens": 1305397.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.390170431137085,
|
|
"epoch": 0.035653749488573264,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.1696,
|
|
"mean_token_accuracy": 0.08689776062965393,
|
|
"num_tokens": 1317449.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.28363561630249,
|
|
"epoch": 0.03594599333684026,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.0866,
|
|
"mean_token_accuracy": 0.08486035652458668,
|
|
"num_tokens": 1328645.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.2835465431213375,
|
|
"epoch": 0.036238237185107255,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.091,
|
|
"mean_token_accuracy": 0.0892544224858284,
|
|
"num_tokens": 1339815.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.290801572799682,
|
|
"epoch": 0.036530481033374244,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.0954,
|
|
"mean_token_accuracy": 0.08872693330049515,
|
|
"num_tokens": 1350301.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.280572080612183,
|
|
"epoch": 0.03682272488164124,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.0829,
|
|
"mean_token_accuracy": 0.09459017589688301,
|
|
"num_tokens": 1360928.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.34552264213562,
|
|
"epoch": 0.037114968729908236,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.1047,
|
|
"mean_token_accuracy": 0.08707956597208977,
|
|
"num_tokens": 1371618.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.211892175674438,
|
|
"epoch": 0.03740721257817523,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.0702,
|
|
"mean_token_accuracy": 0.0876346081495285,
|
|
"num_tokens": 1382824.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.2274785995483395,
|
|
"epoch": 0.03769945642644222,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.0401,
|
|
"mean_token_accuracy": 0.09252982586622238,
|
|
"num_tokens": 1393244.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.224736738204956,
|
|
"epoch": 0.03799170027470922,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.0136,
|
|
"mean_token_accuracy": 0.09895449355244637,
|
|
"num_tokens": 1403469.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.358276271820069,
|
|
"epoch": 0.03828394412297621,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.086,
|
|
"mean_token_accuracy": 0.09135140627622604,
|
|
"num_tokens": 1413860.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.256574583053589,
|
|
"epoch": 0.03857618797124321,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.1348,
|
|
"mean_token_accuracy": 0.08090706467628479,
|
|
"num_tokens": 1423985.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.278745222091675,
|
|
"epoch": 0.0388684318195102,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.087,
|
|
"mean_token_accuracy": 0.09359103664755822,
|
|
"num_tokens": 1434838.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.2903660297393795,
|
|
"epoch": 0.039160675667777194,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.007,
|
|
"mean_token_accuracy": 0.08955025300383568,
|
|
"num_tokens": 1446429.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.332035541534424,
|
|
"epoch": 0.03945291951604419,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.1183,
|
|
"mean_token_accuracy": 0.08762314394116402,
|
|
"num_tokens": 1456726.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.197989416122437,
|
|
"epoch": 0.03974516336431118,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.0058,
|
|
"mean_token_accuracy": 0.09932290315628052,
|
|
"num_tokens": 1466378.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.104801225662231,
|
|
"epoch": 0.040037407212578174,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.9835,
|
|
"mean_token_accuracy": 0.09476312026381492,
|
|
"num_tokens": 1477755.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.237457656860352,
|
|
"epoch": 0.04032965106084517,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.036,
|
|
"mean_token_accuracy": 0.08923429250717163,
|
|
"num_tokens": 1488209.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.137836837768555,
|
|
"epoch": 0.040621894909112166,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.0728,
|
|
"mean_token_accuracy": 0.08890949785709382,
|
|
"num_tokens": 1498658.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.236037731170654,
|
|
"epoch": 0.040914138757379155,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.0789,
|
|
"mean_token_accuracy": 0.0860823355615139,
|
|
"num_tokens": 1509766.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.218784284591675,
|
|
"epoch": 0.04120638260564615,
|
|
"grad_norm": 1.671875,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.975,
|
|
"mean_token_accuracy": 0.0937421977519989,
|
|
"num_tokens": 1521139.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.174445295333863,
|
|
"epoch": 0.04149862645391315,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.0357,
|
|
"mean_token_accuracy": 0.0900140330195427,
|
|
"num_tokens": 1531875.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.153319358825684,
|
|
"epoch": 0.041790870302180136,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.08912704512476921,
|
|
"num_tokens": 1542678.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.196989250183106,
|
|
"epoch": 0.04208311415044713,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.9703,
|
|
"mean_token_accuracy": 0.09389843866229057,
|
|
"num_tokens": 1553882.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.12842378616333,
|
|
"epoch": 0.04237535799871413,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.0568,
|
|
"mean_token_accuracy": 0.0915483608841896,
|
|
"num_tokens": 1564854.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.176506567001343,
|
|
"epoch": 0.042667601846981124,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.9303,
|
|
"mean_token_accuracy": 0.09940993562340736,
|
|
"num_tokens": 1576112.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.137564849853516,
|
|
"epoch": 0.04295984569524811,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.9771,
|
|
"mean_token_accuracy": 0.0963865041732788,
|
|
"num_tokens": 1586235.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.083181285858155,
|
|
"epoch": 0.04325208954351511,
|
|
"grad_norm": 2.046875,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.9769,
|
|
"mean_token_accuracy": 0.09416570290923118,
|
|
"num_tokens": 1596727.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.099421977996826,
|
|
"epoch": 0.043544333391782104,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.8926,
|
|
"mean_token_accuracy": 0.09487500786781311,
|
|
"num_tokens": 1607008.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.178588104248047,
|
|
"epoch": 0.0438365772400491,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.9975,
|
|
"mean_token_accuracy": 0.0891700565814972,
|
|
"num_tokens": 1618081.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.064407539367676,
|
|
"epoch": 0.04412882108831609,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.9569,
|
|
"mean_token_accuracy": 0.08645836561918259,
|
|
"num_tokens": 1629095.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.101617288589478,
|
|
"epoch": 0.044421064936583085,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.9537,
|
|
"mean_token_accuracy": 0.0951661191880703,
|
|
"num_tokens": 1639861.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.189848804473877,
|
|
"epoch": 0.04471330878485008,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.0003,
|
|
"mean_token_accuracy": 0.09473721534013749,
|
|
"num_tokens": 1649169.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 6.957789611816406,
|
|
"epoch": 0.04500555263311707,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.8833,
|
|
"mean_token_accuracy": 0.10007616281509399,
|
|
"num_tokens": 1658528.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.049975156784058,
|
|
"epoch": 0.045297796481384066,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.9657,
|
|
"mean_token_accuracy": 0.09249654635787011,
|
|
"num_tokens": 1669231.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.0429846286773685,
|
|
"epoch": 0.04559004032965106,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.8987,
|
|
"mean_token_accuracy": 0.09256255626678467,
|
|
"num_tokens": 1680503.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.069688749313355,
|
|
"epoch": 0.04588228417791806,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.9239,
|
|
"mean_token_accuracy": 0.09164713248610497,
|
|
"num_tokens": 1690827.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.1588634014129635,
|
|
"epoch": 0.04617452802618505,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.992,
|
|
"mean_token_accuracy": 0.09126574769616128,
|
|
"num_tokens": 1702931.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 6.960534286499024,
|
|
"epoch": 0.04646677187445204,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.8832,
|
|
"mean_token_accuracy": 0.09681101739406586,
|
|
"num_tokens": 1712834.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.059118413925171,
|
|
"epoch": 0.04675901572271904,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.9505,
|
|
"mean_token_accuracy": 0.09173840507864953,
|
|
"num_tokens": 1723289.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 6.98992805480957,
|
|
"epoch": 0.04705125957098603,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.9111,
|
|
"mean_token_accuracy": 0.09471877589821816,
|
|
"num_tokens": 1734477.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.110216426849365,
|
|
"epoch": 0.04734350341925302,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.9201,
|
|
"mean_token_accuracy": 0.09181275144219399,
|
|
"num_tokens": 1745187.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 6.9768541812896725,
|
|
"epoch": 0.04763574726752002,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.8563,
|
|
"mean_token_accuracy": 0.09723289534449578,
|
|
"num_tokens": 1756687.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.012206411361694,
|
|
"epoch": 0.047927991115787015,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.9795,
|
|
"mean_token_accuracy": 0.09605197682976722,
|
|
"num_tokens": 1767928.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.063567161560059,
|
|
"epoch": 0.048220234964054004,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.0963786244392395,
|
|
"num_tokens": 1777951.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 6.963260746002197,
|
|
"epoch": 0.048512478812321,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.9047,
|
|
"mean_token_accuracy": 0.09518637284636497,
|
|
"num_tokens": 1788598.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.022843551635742,
|
|
"epoch": 0.048804722660587996,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.8708,
|
|
"mean_token_accuracy": 0.0998458556830883,
|
|
"num_tokens": 1799446.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.028498888015747,
|
|
"epoch": 0.04909696650885499,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.8757,
|
|
"mean_token_accuracy": 0.0957573838531971,
|
|
"num_tokens": 1809687.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 6.976888608932495,
|
|
"epoch": 0.04938921035712198,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.9107,
|
|
"mean_token_accuracy": 0.09788398742675782,
|
|
"num_tokens": 1820535.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 6.943860197067261,
|
|
"epoch": 0.04968145420538898,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.8376,
|
|
"mean_token_accuracy": 0.09310686662793159,
|
|
"num_tokens": 1830834.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 6.961981105804443,
|
|
"epoch": 0.04997369805365597,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.8257,
|
|
"mean_token_accuracy": 0.0944649763405323,
|
|
"num_tokens": 1842615.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.919750738143921,
|
|
"epoch": 0.05026594190192296,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.8328,
|
|
"mean_token_accuracy": 0.09815948903560638,
|
|
"num_tokens": 1852844.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.876930141448975,
|
|
"epoch": 0.05055818575018996,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.8918,
|
|
"mean_token_accuracy": 0.09418935924768448,
|
|
"num_tokens": 1862528.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 6.934880113601684,
|
|
"epoch": 0.05085042959845695,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.8469,
|
|
"mean_token_accuracy": 0.10216829851269722,
|
|
"num_tokens": 1873275.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.956448936462403,
|
|
"epoch": 0.05114267344672395,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.8847,
|
|
"mean_token_accuracy": 0.09507919698953629,
|
|
"num_tokens": 1884724.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.051344537734986,
|
|
"epoch": 0.05143491729499094,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.8598,
|
|
"mean_token_accuracy": 0.09828044772148133,
|
|
"num_tokens": 1894705.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 6.862812566757202,
|
|
"epoch": 0.051727161143257934,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.9184,
|
|
"mean_token_accuracy": 0.09541104733943939,
|
|
"num_tokens": 1906553.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.020113182067871,
|
|
"epoch": 0.05201940499152493,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.8544,
|
|
"mean_token_accuracy": 0.10240027904510499,
|
|
"num_tokens": 1916628.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 6.766639471054077,
|
|
"epoch": 0.052311648839791926,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6849,
|
|
"mean_token_accuracy": 0.11086282283067703,
|
|
"num_tokens": 1926686.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.947674942016602,
|
|
"epoch": 0.052603892688058915,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.8638,
|
|
"mean_token_accuracy": 0.09778930097818375,
|
|
"num_tokens": 1938150.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.891585969924927,
|
|
"epoch": 0.05289613653632591,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.873,
|
|
"mean_token_accuracy": 0.09291322231292724,
|
|
"num_tokens": 1948617.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 6.900164985656739,
|
|
"epoch": 0.053188380384592907,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.8565,
|
|
"mean_token_accuracy": 0.09954984486103058,
|
|
"num_tokens": 1958943.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.856646156311035,
|
|
"epoch": 0.053480624232859895,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.7881,
|
|
"mean_token_accuracy": 0.10126558542251587,
|
|
"num_tokens": 1968988.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.895283889770508,
|
|
"epoch": 0.05377286808112689,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.7692,
|
|
"mean_token_accuracy": 0.10075787082314491,
|
|
"num_tokens": 1979490.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.938627767562866,
|
|
"epoch": 0.05406511192939389,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.8187,
|
|
"mean_token_accuracy": 0.10156044065952301,
|
|
"num_tokens": 1990377.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 6.881887578964234,
|
|
"epoch": 0.05435735577766088,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.7914,
|
|
"mean_token_accuracy": 0.09942576885223389,
|
|
"num_tokens": 2001678.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.809711265563965,
|
|
"epoch": 0.05464959962592787,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.7898,
|
|
"mean_token_accuracy": 0.11002987027168273,
|
|
"num_tokens": 2011550.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.853000497817993,
|
|
"epoch": 0.05494184347419487,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.8297,
|
|
"mean_token_accuracy": 0.10066121965646743,
|
|
"num_tokens": 2021607.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.862749242782593,
|
|
"epoch": 0.055234087322461864,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.8211,
|
|
"mean_token_accuracy": 0.09814092963933944,
|
|
"num_tokens": 2030975.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.836636781692505,
|
|
"epoch": 0.05552633117072885,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.8844,
|
|
"mean_token_accuracy": 0.09624653086066245,
|
|
"num_tokens": 2041799.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.891287755966187,
|
|
"epoch": 0.05581857501899585,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.7935,
|
|
"mean_token_accuracy": 0.10220010504126549,
|
|
"num_tokens": 2052960.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.868294525146484,
|
|
"epoch": 0.056110818867262845,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.8472,
|
|
"mean_token_accuracy": 0.09924568086862565,
|
|
"num_tokens": 2064156.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.914956092834473,
|
|
"epoch": 0.05640306271552984,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.8016,
|
|
"mean_token_accuracy": 0.100481665879488,
|
|
"num_tokens": 2074774.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.793342113494873,
|
|
"epoch": 0.05669530656379683,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.8288,
|
|
"mean_token_accuracy": 0.09989185482263566,
|
|
"num_tokens": 2086164.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.78997688293457,
|
|
"epoch": 0.056987550412063825,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.6708,
|
|
"mean_token_accuracy": 0.1248696506023407,
|
|
"num_tokens": 2096268.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.888183069229126,
|
|
"epoch": 0.05727979426033082,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.8985,
|
|
"mean_token_accuracy": 0.09069314152002335,
|
|
"num_tokens": 2108511.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.923859405517578,
|
|
"epoch": 0.05757203810859782,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.8155,
|
|
"mean_token_accuracy": 0.09430013298988342,
|
|
"num_tokens": 2119622.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.742640447616577,
|
|
"epoch": 0.057864281956864806,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.8703,
|
|
"mean_token_accuracy": 0.10323725119233132,
|
|
"num_tokens": 2131021.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.867950057983398,
|
|
"epoch": 0.0581565258051318,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.8448,
|
|
"mean_token_accuracy": 0.10064574256539345,
|
|
"num_tokens": 2141797.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.782402372360229,
|
|
"epoch": 0.0584487696533988,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.7141,
|
|
"mean_token_accuracy": 0.09859876409173012,
|
|
"num_tokens": 2153609.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.873832368850708,
|
|
"epoch": 0.05874101350166579,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499998026082006,
|
|
"loss": 6.8556,
|
|
"mean_token_accuracy": 0.09399589225649833,
|
|
"num_tokens": 2164237.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.912446355819702,
|
|
"epoch": 0.05903325734993278,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999900070995136,
|
|
"loss": 6.8018,
|
|
"mean_token_accuracy": 0.10092566832900048,
|
|
"num_tokens": 2175035.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.6948404788970945,
|
|
"epoch": 0.05932550119819978,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004999758199023239,
|
|
"loss": 6.7217,
|
|
"mean_token_accuracy": 0.10070772990584373,
|
|
"num_tokens": 2186217.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.849718952178955,
|
|
"epoch": 0.059617745046466775,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004999554648793858,
|
|
"loss": 6.7685,
|
|
"mean_token_accuracy": 0.1027240552008152,
|
|
"num_tokens": 2197290.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.755612897872925,
|
|
"epoch": 0.059909988894733764,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.8215,
|
|
"mean_token_accuracy": 0.09566072076559066,
|
|
"num_tokens": 2208835.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.851420831680298,
|
|
"epoch": 0.06020223274300076,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004998962537575161,
|
|
"loss": 6.7081,
|
|
"mean_token_accuracy": 0.10895594730973243,
|
|
"num_tokens": 2219153.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.780207109451294,
|
|
"epoch": 0.060494476591267755,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004998573992818874,
|
|
"loss": 6.7401,
|
|
"mean_token_accuracy": 0.1018265351653099,
|
|
"num_tokens": 2230289.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.748766994476318,
|
|
"epoch": 0.060786720439534744,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004998123802270715,
|
|
"loss": 6.8198,
|
|
"mean_token_accuracy": 0.10137188360095024,
|
|
"num_tokens": 2241768.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.8524377822875975,
|
|
"epoch": 0.06107896428780174,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004997611978272886,
|
|
"loss": 6.7065,
|
|
"mean_token_accuracy": 0.10613882467150688,
|
|
"num_tokens": 2253249.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.758801507949829,
|
|
"epoch": 0.061371208136068736,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004997038534857298,
|
|
"loss": 6.7723,
|
|
"mean_token_accuracy": 0.09954821765422821,
|
|
"num_tokens": 2263961.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.748946809768677,
|
|
"epoch": 0.06166345198433573,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996403487745194,
|
|
"loss": 6.6831,
|
|
"mean_token_accuracy": 0.10762475430965424,
|
|
"num_tokens": 2274016.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.716818904876709,
|
|
"epoch": 0.06195569583260272,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499570685434671,
|
|
"loss": 6.7368,
|
|
"mean_token_accuracy": 0.10165347680449485,
|
|
"num_tokens": 2285164.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.843360662460327,
|
|
"epoch": 0.06224793968086972,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004994948653760405,
|
|
"loss": 6.7192,
|
|
"mean_token_accuracy": 0.10079561397433281,
|
|
"num_tokens": 2294887.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.701195096969604,
|
|
"epoch": 0.0625401835291367,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004994128906772729,
|
|
"loss": 6.7315,
|
|
"mean_token_accuracy": 0.10079780966043472,
|
|
"num_tokens": 2305787.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.763786745071411,
|
|
"epoch": 0.0628324273774037,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000499324763585746,
|
|
"loss": 6.7659,
|
|
"mean_token_accuracy": 0.1027215838432312,
|
|
"num_tokens": 2316756.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.758350419998169,
|
|
"epoch": 0.0631246712256707,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004992304865175085,
|
|
"loss": 6.7782,
|
|
"mean_token_accuracy": 0.10599919855594635,
|
|
"num_tokens": 2327335.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.76016902923584,
|
|
"epoch": 0.0634169150739377,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.7328,
|
|
"mean_token_accuracy": 0.10450217202305793,
|
|
"num_tokens": 2337182.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.7664146900177,
|
|
"epoch": 0.06370915892220469,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004990234929580494,
|
|
"loss": 6.7124,
|
|
"mean_token_accuracy": 0.10350674986839295,
|
|
"num_tokens": 2347843.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.714575910568238,
|
|
"epoch": 0.06400140277047169,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004989107821416609,
|
|
"loss": 6.7235,
|
|
"mean_token_accuracy": 0.10555968135595321,
|
|
"num_tokens": 2358159.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.721023511886597,
|
|
"epoch": 0.06429364661873868,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004987919326980723,
|
|
"loss": 6.6924,
|
|
"mean_token_accuracy": 0.10607861652970314,
|
|
"num_tokens": 2368191.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.734277248382568,
|
|
"epoch": 0.06458589046700566,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004986669478856011,
|
|
"loss": 6.6338,
|
|
"mean_token_accuracy": 0.11026455089449883,
|
|
"num_tokens": 2377672.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.6623273372650145,
|
|
"epoch": 0.06487813431527266,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004985358311307688,
|
|
"loss": 6.7138,
|
|
"mean_token_accuracy": 0.10811490938067436,
|
|
"num_tokens": 2387612.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.6812506198883055,
|
|
"epoch": 0.06517037816353966,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004983985860282081,
|
|
"loss": 6.6572,
|
|
"mean_token_accuracy": 0.11455306559801101,
|
|
"num_tokens": 2397198.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.769713687896728,
|
|
"epoch": 0.06546262201180665,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004982552163405623,
|
|
"loss": 6.7003,
|
|
"mean_token_accuracy": 0.10592205822467804,
|
|
"num_tokens": 2407816.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.6080625534057615,
|
|
"epoch": 0.06575486586007365,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004981057259983839,
|
|
"loss": 6.6798,
|
|
"mean_token_accuracy": 0.10187085941433907,
|
|
"num_tokens": 2419537.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.763877916336059,
|
|
"epoch": 0.06604710970834064,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004979501191000262,
|
|
"loss": 6.6267,
|
|
"mean_token_accuracy": 0.10548629984259605,
|
|
"num_tokens": 2429043.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.634230661392212,
|
|
"epoch": 0.06633935355660764,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004977883999115311,
|
|
"loss": 6.6642,
|
|
"mean_token_accuracy": 0.10568090230226516,
|
|
"num_tokens": 2439469.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.669453048706055,
|
|
"epoch": 0.06663159740487462,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004976205728665113,
|
|
"loss": 6.659,
|
|
"mean_token_accuracy": 0.11069119200110436,
|
|
"num_tokens": 2449831.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.6411519050598145,
|
|
"epoch": 0.06692384125314162,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004974466425660307,
|
|
"loss": 6.6127,
|
|
"mean_token_accuracy": 0.11009648814797401,
|
|
"num_tokens": 2460709.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.681810665130615,
|
|
"epoch": 0.06721608510140861,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004972666137784759,
|
|
"loss": 6.6562,
|
|
"mean_token_accuracy": 0.11087686195969582,
|
|
"num_tokens": 2472246.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.661732769012451,
|
|
"epoch": 0.06750832894967561,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004970804914394271,
|
|
"loss": 6.6214,
|
|
"mean_token_accuracy": 0.10803418233990669,
|
|
"num_tokens": 2482793.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.574620342254638,
|
|
"epoch": 0.0678005727979426,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004968882806515225,
|
|
"loss": 6.6042,
|
|
"mean_token_accuracy": 0.11461034417152405,
|
|
"num_tokens": 2493186.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.723605728149414,
|
|
"epoch": 0.0680928166462096,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004966899866843177,
|
|
"loss": 6.6715,
|
|
"mean_token_accuracy": 0.10363015085458756,
|
|
"num_tokens": 2505100.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.638859796524048,
|
|
"epoch": 0.0683850604944766,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496485614974142,
|
|
"loss": 6.6775,
|
|
"mean_token_accuracy": 0.10529874190688134,
|
|
"num_tokens": 2515373.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.724088048934936,
|
|
"epoch": 0.06867730434274359,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004962751711239492,
|
|
"loss": 6.6687,
|
|
"mean_token_accuracy": 0.10807336121797562,
|
|
"num_tokens": 2526645.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.642391300201416,
|
|
"epoch": 0.06896954819101057,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004960586609031636,
|
|
"loss": 6.6747,
|
|
"mean_token_accuracy": 0.10391177088022233,
|
|
"num_tokens": 2537109.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.675477027893066,
|
|
"epoch": 0.06926179203927757,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004958360902475224,
|
|
"loss": 6.6186,
|
|
"mean_token_accuracy": 0.11234116032719613,
|
|
"num_tokens": 2548709.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.690750455856323,
|
|
"epoch": 0.06955403588754457,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004956074652589125,
|
|
"loss": 6.6621,
|
|
"mean_token_accuracy": 0.10677864402532578,
|
|
"num_tokens": 2559802.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.617762088775635,
|
|
"epoch": 0.06984627973581156,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004953727922052035,
|
|
"loss": 6.6607,
|
|
"mean_token_accuracy": 0.1037447288632393,
|
|
"num_tokens": 2570343.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.693934535980224,
|
|
"epoch": 0.07013852358407856,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004951320775200756,
|
|
"loss": 6.6457,
|
|
"mean_token_accuracy": 0.1090537242591381,
|
|
"num_tokens": 2580772.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.620019578933716,
|
|
"epoch": 0.07043076743234555,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004948853278028436,
|
|
"loss": 6.643,
|
|
"mean_token_accuracy": 0.10731385722756386,
|
|
"num_tokens": 2592383.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.674355459213257,
|
|
"epoch": 0.07072301128061255,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004946325498182755,
|
|
"loss": 6.6392,
|
|
"mean_token_accuracy": 0.10924243628978729,
|
|
"num_tokens": 2602807.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.649422836303711,
|
|
"epoch": 0.07101525512887953,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004943737504964076,
|
|
"loss": 6.6336,
|
|
"mean_token_accuracy": 0.10555684268474579,
|
|
"num_tokens": 2613978.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.645706224441528,
|
|
"epoch": 0.07130749897714653,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000494108936932354,
|
|
"loss": 6.6105,
|
|
"mean_token_accuracy": 0.10822624936699868,
|
|
"num_tokens": 2624044.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.6873753547668455,
|
|
"epoch": 0.07159974282541352,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004938381163861124,
|
|
"loss": 6.6315,
|
|
"mean_token_accuracy": 0.11078074425458909,
|
|
"num_tokens": 2635550.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.588435077667237,
|
|
"epoch": 0.07189198667368052,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004935612962823645,
|
|
"loss": 6.5788,
|
|
"mean_token_accuracy": 0.11432232186198235,
|
|
"num_tokens": 2645343.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.617160987854004,
|
|
"epoch": 0.07218423052194751,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004932784842102739,
|
|
"loss": 6.646,
|
|
"mean_token_accuracy": 0.11204688772559165,
|
|
"num_tokens": 2655366.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.6750807762146,
|
|
"epoch": 0.07247647437021451,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004929896879232758,
|
|
"loss": 6.6115,
|
|
"mean_token_accuracy": 0.1087801031768322,
|
|
"num_tokens": 2666438.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.569998693466187,
|
|
"epoch": 0.0727687182184815,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004926949153388668,
|
|
"loss": 6.5661,
|
|
"mean_token_accuracy": 0.1062084048986435,
|
|
"num_tokens": 2677800.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.579834270477295,
|
|
"epoch": 0.07306096206674849,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004923941745383859,
|
|
"loss": 6.5762,
|
|
"mean_token_accuracy": 0.10903873145580292,
|
|
"num_tokens": 2688876.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.6498009204864506,
|
|
"epoch": 0.07335320591501548,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000492087473766794,
|
|
"loss": 6.541,
|
|
"mean_token_accuracy": 0.12037940993905068,
|
|
"num_tokens": 2700265.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.541846752166748,
|
|
"epoch": 0.07364544976328248,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000491774821432448,
|
|
"loss": 6.5526,
|
|
"mean_token_accuracy": 0.11863013431429863,
|
|
"num_tokens": 2711336.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.63809118270874,
|
|
"epoch": 0.07393769361154948,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004914562261068693,
|
|
"loss": 6.6417,
|
|
"mean_token_accuracy": 0.10962062701582909,
|
|
"num_tokens": 2722138.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.627244329452514,
|
|
"epoch": 0.07422993745981647,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004911316965245098,
|
|
"loss": 6.5637,
|
|
"mean_token_accuracy": 0.1153975710272789,
|
|
"num_tokens": 2733016.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.589581537246704,
|
|
"epoch": 0.07452218130808347,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000490801241582512,
|
|
"loss": 6.589,
|
|
"mean_token_accuracy": 0.11180200576782226,
|
|
"num_tokens": 2743424.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.650690221786499,
|
|
"epoch": 0.07481442515635046,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000490464870340465,
|
|
"loss": 6.6218,
|
|
"mean_token_accuracy": 0.11065411493182183,
|
|
"num_tokens": 2753867.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.561734628677368,
|
|
"epoch": 0.07510666900461745,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004901225920201563,
|
|
"loss": 6.5209,
|
|
"mean_token_accuracy": 0.1133801057934761,
|
|
"num_tokens": 2764602.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.577672100067138,
|
|
"epoch": 0.07539891285288444,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.000489774416005319,
|
|
"loss": 6.6399,
|
|
"mean_token_accuracy": 0.1088312640786171,
|
|
"num_tokens": 2775907.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.631254863739014,
|
|
"epoch": 0.07569115670115144,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004894203518413742,
|
|
"loss": 6.6094,
|
|
"mean_token_accuracy": 0.10955363363027573,
|
|
"num_tokens": 2787455.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.5455607891082765,
|
|
"epoch": 0.07598340054941843,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004890604092351701,
|
|
"loss": 6.5188,
|
|
"mean_token_accuracy": 0.12422384396195411,
|
|
"num_tokens": 2797600.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.591296243667602,
|
|
"epoch": 0.07627564439768543,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000488694598054715,
|
|
"loss": 6.4734,
|
|
"mean_token_accuracy": 0.120032849162817,
|
|
"num_tokens": 2808056.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.517116546630859,
|
|
"epoch": 0.07656788824595243,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004883229283289071,
|
|
"loss": 6.5297,
|
|
"mean_token_accuracy": 0.11552904546260834,
|
|
"num_tokens": 2817962.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.5853719234466555,
|
|
"epoch": 0.07686013209421942,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00048794541024725993,
|
|
"loss": 6.5356,
|
|
"mean_token_accuracy": 0.11282036751508713,
|
|
"num_tokens": 2828769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.507770824432373,
|
|
"epoch": 0.07715237594248642,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004875620541596221,
|
|
"loss": 6.4596,
|
|
"mean_token_accuracy": 0.12131244838237762,
|
|
"num_tokens": 2838983.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.5859967231750485,
|
|
"epoch": 0.0774446197907534,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048717287057589454,
|
|
"loss": 6.5421,
|
|
"mean_token_accuracy": 0.1098010502755642,
|
|
"num_tokens": 2848892.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.555781459808349,
|
|
"epoch": 0.0777368636390204,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004867778701657417,
|
|
"loss": 6.4932,
|
|
"mean_token_accuracy": 0.11479394286870956,
|
|
"num_tokens": 2859303.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.587380409240723,
|
|
"epoch": 0.07802910748728739,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00048637706375829955,
|
|
"loss": 6.7083,
|
|
"mean_token_accuracy": 0.10509744510054589,
|
|
"num_tokens": 2871318.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.610728073120117,
|
|
"epoch": 0.07832135133555439,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000485970462341878,
|
|
"loss": 6.5321,
|
|
"mean_token_accuracy": 0.1154100090265274,
|
|
"num_tokens": 2882150.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.5057868480682375,
|
|
"epoch": 0.07861359518382138,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048555807706366044,
|
|
"loss": 6.5277,
|
|
"mean_token_accuracy": 0.11850375607609749,
|
|
"num_tokens": 2892657.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.593202018737793,
|
|
"epoch": 0.07890583903208838,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00048513991922939756,
|
|
"loss": 6.4756,
|
|
"mean_token_accuracy": 0.12083587348461151,
|
|
"num_tokens": 2903298.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.473386001586914,
|
|
"epoch": 0.07919808288035537,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00048471600030309744,
|
|
"loss": 6.5195,
|
|
"mean_token_accuracy": 0.11845917701721191,
|
|
"num_tokens": 2914684.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.5273338794708256,
|
|
"epoch": 0.07949032672862236,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048428633190671186,
|
|
"loss": 6.4561,
|
|
"mean_token_accuracy": 0.11816495433449745,
|
|
"num_tokens": 2925191.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.580375480651855,
|
|
"epoch": 0.07978257057688935,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004838509258198167,
|
|
"loss": 6.5132,
|
|
"mean_token_accuracy": 0.11696040779352188,
|
|
"num_tokens": 2935498.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.557421731948852,
|
|
"epoch": 0.08007481442515635,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00048340979397929,
|
|
"loss": 6.4569,
|
|
"mean_token_accuracy": 0.119329334795475,
|
|
"num_tokens": 2945590.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.543434095382691,
|
|
"epoch": 0.08036705827342334,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00048296294847898386,
|
|
"loss": 6.4328,
|
|
"mean_token_accuracy": 0.12298492044210434,
|
|
"num_tokens": 2954959.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.470787572860718,
|
|
"epoch": 0.08065930212169034,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004825104015693934,
|
|
"loss": 6.5096,
|
|
"mean_token_accuracy": 0.1182025596499443,
|
|
"num_tokens": 2964808.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.538543033599853,
|
|
"epoch": 0.08095154596995734,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004820521656573208,
|
|
"loss": 6.4936,
|
|
"mean_token_accuracy": 0.12162516564130783,
|
|
"num_tokens": 2975032.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.520711994171142,
|
|
"epoch": 0.08124378981822433,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00048158825330553505,
|
|
"loss": 6.455,
|
|
"mean_token_accuracy": 0.12000622674822807,
|
|
"num_tokens": 2984426.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.461547470092773,
|
|
"epoch": 0.08153603366649131,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00048111867723242763,
|
|
"loss": 6.531,
|
|
"mean_token_accuracy": 0.11707211509346963,
|
|
"num_tokens": 2994871.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.565800571441651,
|
|
"epoch": 0.08182827751475831,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004806434503116637,
|
|
"loss": 6.4464,
|
|
"mean_token_accuracy": 0.12056399285793304,
|
|
"num_tokens": 3004800.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.511224269866943,
|
|
"epoch": 0.0821205213630253,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004801625855718296,
|
|
"loss": 6.533,
|
|
"mean_token_accuracy": 0.1121494509279728,
|
|
"num_tokens": 3015561.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.4862833499908445,
|
|
"epoch": 0.0824127652112923,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00047967609619607477,
|
|
"loss": 6.4833,
|
|
"mean_token_accuracy": 0.1104965016245842,
|
|
"num_tokens": 3025885.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.561811447143555,
|
|
"epoch": 0.0827050090595593,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004791839955217513,
|
|
"loss": 6.4301,
|
|
"mean_token_accuracy": 0.12150253728032112,
|
|
"num_tokens": 3035524.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.486034107208252,
|
|
"epoch": 0.0829972529078263,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00047868629704004786,
|
|
"loss": 6.5348,
|
|
"mean_token_accuracy": 0.11588730812072753,
|
|
"num_tokens": 3046957.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.556643438339234,
|
|
"epoch": 0.08328949675609329,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00047818301439561965,
|
|
"loss": 6.4536,
|
|
"mean_token_accuracy": 0.11856252178549767,
|
|
"num_tokens": 3056713.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.448641347885132,
|
|
"epoch": 0.08358174060436027,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00047767416138621454,
|
|
"loss": 6.4515,
|
|
"mean_token_accuracy": 0.11449124291539192,
|
|
"num_tokens": 3068369.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.454101324081421,
|
|
"epoch": 0.08387398445262727,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.000477159751962295,
|
|
"loss": 6.5033,
|
|
"mean_token_accuracy": 0.11825905069708824,
|
|
"num_tokens": 3079083.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.496670722961426,
|
|
"epoch": 0.08416622830089426,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00047663980022665507,
|
|
"loss": 6.5362,
|
|
"mean_token_accuracy": 0.11073331981897354,
|
|
"num_tokens": 3089150.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.540896987915039,
|
|
"epoch": 0.08445847214916126,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047611432043403437,
|
|
"loss": 6.4313,
|
|
"mean_token_accuracy": 0.12387260645627976,
|
|
"num_tokens": 3098696.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.554162788391113,
|
|
"epoch": 0.08475071599742826,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004755833269907267,
|
|
"loss": 6.4593,
|
|
"mean_token_accuracy": 0.11571898385882377,
|
|
"num_tokens": 3108879.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.521293210983276,
|
|
"epoch": 0.08504295984569525,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004750468344541857,
|
|
"loss": 6.4213,
|
|
"mean_token_accuracy": 0.12200254499912262,
|
|
"num_tokens": 3118756.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.462752103805542,
|
|
"epoch": 0.08533520369396225,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00047450485753262525,
|
|
"loss": 6.472,
|
|
"mean_token_accuracy": 0.11719053238630295,
|
|
"num_tokens": 3129130.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.469192171096802,
|
|
"epoch": 0.08562744754222923,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00047395741108461633,
|
|
"loss": 6.4329,
|
|
"mean_token_accuracy": 0.12074794396758079,
|
|
"num_tokens": 3139528.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.510004281997681,
|
|
"epoch": 0.08591969139049623,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00047340451011867985,
|
|
"loss": 6.5121,
|
|
"mean_token_accuracy": 0.11626194566488265,
|
|
"num_tokens": 3149456.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.424893188476562,
|
|
"epoch": 0.08621193523876322,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00047284616979287515,
|
|
"loss": 6.4404,
|
|
"mean_token_accuracy": 0.1183363862335682,
|
|
"num_tokens": 3159851.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.492526721954346,
|
|
"epoch": 0.08650417908703022,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00047228240541438433,
|
|
"loss": 6.4013,
|
|
"mean_token_accuracy": 0.12410363852977753,
|
|
"num_tokens": 3169897.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.505418395996093,
|
|
"epoch": 0.08679642293529721,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00047171323243909257,
|
|
"loss": 6.5236,
|
|
"mean_token_accuracy": 0.12042340859770775,
|
|
"num_tokens": 3180696.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.4698486328125,
|
|
"epoch": 0.08708866678356421,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00047113866647116457,
|
|
"loss": 6.4374,
|
|
"mean_token_accuracy": 0.11517155095934868,
|
|
"num_tokens": 3191836.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.432517099380493,
|
|
"epoch": 0.0873809106318312,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004705587232626164,
|
|
"loss": 6.3758,
|
|
"mean_token_accuracy": 0.12214858755469322,
|
|
"num_tokens": 3202009.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.38523383140564,
|
|
"epoch": 0.0876731544800982,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00046997341871288424,
|
|
"loss": 6.3545,
|
|
"mean_token_accuracy": 0.12869924455881118,
|
|
"num_tokens": 3212707.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 6.449147129058838,
|
|
"epoch": 0.08796539832836518,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004693827688683879,
|
|
"loss": 6.4027,
|
|
"mean_token_accuracy": 0.11799835041165352,
|
|
"num_tokens": 3223007.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 6.5049091339111325,
|
|
"epoch": 0.08825764217663218,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004687867899220914,
|
|
"loss": 6.4222,
|
|
"mean_token_accuracy": 0.12286672741174698,
|
|
"num_tokens": 3234634.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 6.402359294891357,
|
|
"epoch": 0.08854988602489917,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046818549821305846,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.12333594635128975,
|
|
"num_tokens": 3245338.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 6.5087133884429935,
|
|
"epoch": 0.08884212987316617,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00046757891022600494,
|
|
"loss": 6.4985,
|
|
"mean_token_accuracy": 0.12035001292824746,
|
|
"num_tokens": 3255776.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 6.436366891860962,
|
|
"epoch": 0.08913437372143317,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004669670425908471,
|
|
"loss": 6.3967,
|
|
"mean_token_accuracy": 0.12574007660150527,
|
|
"num_tokens": 3265908.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 6.488069295883179,
|
|
"epoch": 0.08942661756970016,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004663499120822451,
|
|
"loss": 6.4717,
|
|
"mean_token_accuracy": 0.11899737641215324,
|
|
"num_tokens": 3277623.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 6.431543445587158,
|
|
"epoch": 0.08971886141796716,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004657275356191437,
|
|
"loss": 6.4053,
|
|
"mean_token_accuracy": 0.12159285694360733,
|
|
"num_tokens": 3288804.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 6.3821464538574215,
|
|
"epoch": 0.09001110526623414,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00046509993026430804,
|
|
"loss": 6.4055,
|
|
"mean_token_accuracy": 0.12484046667814255,
|
|
"num_tokens": 3299162.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 6.473441171646118,
|
|
"epoch": 0.09030334911450114,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004644671132238558,
|
|
"loss": 6.4532,
|
|
"mean_token_accuracy": 0.1200271911919117,
|
|
"num_tokens": 3311425.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 6.41570463180542,
|
|
"epoch": 0.09059559296276813,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00046382910184678585,
|
|
"loss": 6.392,
|
|
"mean_token_accuracy": 0.1237281620502472,
|
|
"num_tokens": 3322623.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 6.559175443649292,
|
|
"epoch": 0.09088783681103513,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004631859136245025,
|
|
"loss": 6.4016,
|
|
"mean_token_accuracy": 0.119428800791502,
|
|
"num_tokens": 3333234.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 6.382933664321899,
|
|
"epoch": 0.09118008065930212,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004625375661903357,
|
|
"loss": 6.4188,
|
|
"mean_token_accuracy": 0.12374623641371726,
|
|
"num_tokens": 3343302.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 6.434831190109253,
|
|
"epoch": 0.09147232450756912,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00046188407731905787,
|
|
"loss": 6.4081,
|
|
"mean_token_accuracy": 0.11539566516876221,
|
|
"num_tokens": 3355107.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 6.464893102645874,
|
|
"epoch": 0.09176456835583612,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00046122546492639643,
|
|
"loss": 6.3832,
|
|
"mean_token_accuracy": 0.12429433092474937,
|
|
"num_tokens": 3366459.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 6.394007062911987,
|
|
"epoch": 0.0920568122041031,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000460561747068543,
|
|
"loss": 6.3437,
|
|
"mean_token_accuracy": 0.12754202634096146,
|
|
"num_tokens": 3377050.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 6.351640415191651,
|
|
"epoch": 0.0923490560523701,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004598929419416578,
|
|
"loss": 6.3578,
|
|
"mean_token_accuracy": 0.12364773005247116,
|
|
"num_tokens": 3387810.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 6.372348213195801,
|
|
"epoch": 0.09264129990063709,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00045921906788137123,
|
|
"loss": 6.3087,
|
|
"mean_token_accuracy": 0.1251809149980545,
|
|
"num_tokens": 3398038.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 6.48484992980957,
|
|
"epoch": 0.09293354374890408,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045854014336228115,
|
|
"loss": 6.3971,
|
|
"mean_token_accuracy": 0.12183109745383262,
|
|
"num_tokens": 3408426.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 6.383055734634399,
|
|
"epoch": 0.09322578759717108,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00045785618699744615,
|
|
"loss": 6.3343,
|
|
"mean_token_accuracy": 0.12766708433628082,
|
|
"num_tokens": 3419268.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 6.44209623336792,
|
|
"epoch": 0.09351803144543808,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.00045716721753787543,
|
|
"loss": 6.3343,
|
|
"mean_token_accuracy": 0.1299479439854622,
|
|
"num_tokens": 3429813.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 6.311538505554199,
|
|
"epoch": 0.09381027529370507,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004564732538720148,
|
|
"loss": 6.3186,
|
|
"mean_token_accuracy": 0.13134126588702202,
|
|
"num_tokens": 3440062.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 6.4730775356292725,
|
|
"epoch": 0.09410251914197205,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00045577431502522877,
|
|
"loss": 6.3151,
|
|
"mean_token_accuracy": 0.119265066832304,
|
|
"num_tokens": 3450204.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 6.293524074554443,
|
|
"epoch": 0.09439476299023905,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004550704201592787,
|
|
"loss": 6.3152,
|
|
"mean_token_accuracy": 0.12793571650981903,
|
|
"num_tokens": 3460593.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 6.456835985183716,
|
|
"epoch": 0.09468700683850605,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004543615885717981,
|
|
"loss": 6.3474,
|
|
"mean_token_accuracy": 0.1260027453303337,
|
|
"num_tokens": 3471617.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 6.301732873916626,
|
|
"epoch": 0.09497925068677304,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045364783969576296,
|
|
"loss": 6.3462,
|
|
"mean_token_accuracy": 0.1214242309331894,
|
|
"num_tokens": 3481701.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 6.368519926071167,
|
|
"epoch": 0.09527149453504004,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004529291930989592,
|
|
"loss": 6.3516,
|
|
"mean_token_accuracy": 0.12556039541959763,
|
|
"num_tokens": 3491148.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 6.446704864501953,
|
|
"epoch": 0.09556373838330703,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004522056684834464,
|
|
"loss": 6.408,
|
|
"mean_token_accuracy": 0.12289013266563416,
|
|
"num_tokens": 3501675.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 6.411620616912842,
|
|
"epoch": 0.09585598223157403,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004514772856850173,
|
|
"loss": 6.3322,
|
|
"mean_token_accuracy": 0.12715326100587845,
|
|
"num_tokens": 3512988.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 6.336518287658691,
|
|
"epoch": 0.09614822607984103,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004507440646726542,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.12428070306777954,
|
|
"num_tokens": 3524004.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 6.390318155288696,
|
|
"epoch": 0.09644046992810801,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004500060255479818,
|
|
"loss": 6.3273,
|
|
"mean_token_accuracy": 0.1296905018389225,
|
|
"num_tokens": 3533627.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 6.385443687438965,
|
|
"epoch": 0.096732713776375,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004492631885447151,
|
|
"loss": 6.3199,
|
|
"mean_token_accuracy": 0.12533507943153382,
|
|
"num_tokens": 3544798.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 6.4146082401275635,
|
|
"epoch": 0.097024957624642,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00044851557402810616,
|
|
"loss": 6.4395,
|
|
"mean_token_accuracy": 0.12418477311730385,
|
|
"num_tokens": 3556379.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 6.486456441879272,
|
|
"epoch": 0.097317201472909,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00044776320249438444,
|
|
"loss": 6.3223,
|
|
"mean_token_accuracy": 0.12915950194001197,
|
|
"num_tokens": 3567032.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 6.362237405776978,
|
|
"epoch": 0.09760944532117599,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00044700609457019565,
|
|
"loss": 6.3932,
|
|
"mean_token_accuracy": 0.1188717357814312,
|
|
"num_tokens": 3578419.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 6.366336917877197,
|
|
"epoch": 0.09790168916944299,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004462442710120359,
|
|
"loss": 6.3035,
|
|
"mean_token_accuracy": 0.12955877259373666,
|
|
"num_tokens": 3589501.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 6.345707559585572,
|
|
"epoch": 0.09819393301770998,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000445477752705683,
|
|
"loss": 6.2906,
|
|
"mean_token_accuracy": 0.1341054067015648,
|
|
"num_tokens": 3600162.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 6.401959180831909,
|
|
"epoch": 0.09848617686597697,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00044470656066562336,
|
|
"loss": 6.3893,
|
|
"mean_token_accuracy": 0.12404495999217033,
|
|
"num_tokens": 3610860.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 6.393278408050537,
|
|
"epoch": 0.09877842071424396,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004439307160344765,
|
|
"loss": 6.3654,
|
|
"mean_token_accuracy": 0.118367550522089,
|
|
"num_tokens": 3620989.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 6.366997861862183,
|
|
"epoch": 0.09907066456251096,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00044315024008241473,
|
|
"loss": 6.2943,
|
|
"mean_token_accuracy": 0.12480385452508927,
|
|
"num_tokens": 3631602.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 6.428997039794922,
|
|
"epoch": 0.09936290841077795,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004423651542065806,
|
|
"loss": 6.3501,
|
|
"mean_token_accuracy": 0.12229208797216415,
|
|
"num_tokens": 3642326.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 6.342395067214966,
|
|
"epoch": 0.09965515225904495,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00044157547993050006,
|
|
"loss": 6.303,
|
|
"mean_token_accuracy": 0.1299465276300907,
|
|
"num_tokens": 3652036.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 6.422306537628174,
|
|
"epoch": 0.09994739610731194,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00044078123890349227,
|
|
"loss": 6.3607,
|
|
"mean_token_accuracy": 0.1249079592525959,
|
|
"num_tokens": 3661557.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 6.4041478633880615,
|
|
"epoch": 0.10023963995557894,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00043998245290007606,
|
|
"loss": 6.3382,
|
|
"mean_token_accuracy": 0.12813913747668265,
|
|
"num_tokens": 3671487.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 6.363600063323974,
|
|
"epoch": 0.10053188380384592,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043917914381937323,
|
|
"loss": 6.3671,
|
|
"mean_token_accuracy": 0.12218450903892517,
|
|
"num_tokens": 3682873.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 6.3568596839904785,
|
|
"epoch": 0.10082412765211292,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.00043837133368450815,
|
|
"loss": 6.2743,
|
|
"mean_token_accuracy": 0.12958545982837677,
|
|
"num_tokens": 3693635.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 6.338172101974488,
|
|
"epoch": 0.10111637150037991,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004375590446420037,
|
|
"loss": 6.2803,
|
|
"mean_token_accuracy": 0.129776468873024,
|
|
"num_tokens": 3703777.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 6.341785001754761,
|
|
"epoch": 0.10140861534864691,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004367422989611743,
|
|
"loss": 6.3153,
|
|
"mean_token_accuracy": 0.13043999075889587,
|
|
"num_tokens": 3716370.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 6.41265835762024,
|
|
"epoch": 0.1017008591969139,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004359211190335153,
|
|
"loss": 6.3648,
|
|
"mean_token_accuracy": 0.12569364979863168,
|
|
"num_tokens": 3726473.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 6.404611396789551,
|
|
"epoch": 0.1019931030451809,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00043509552737208923,
|
|
"loss": 6.3738,
|
|
"mean_token_accuracy": 0.1196278490126133,
|
|
"num_tokens": 3737645.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 6.358111000061035,
|
|
"epoch": 0.1022853468934479,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00043426554661090853,
|
|
"loss": 6.3765,
|
|
"mean_token_accuracy": 0.12168645784258843,
|
|
"num_tokens": 3749347.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 6.338151216506958,
|
|
"epoch": 0.10257759074171488,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00043343119950431516,
|
|
"loss": 6.3223,
|
|
"mean_token_accuracy": 0.12965482100844383,
|
|
"num_tokens": 3760131.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 6.3946410655975345,
|
|
"epoch": 0.10286983458998188,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00043259250892635644,
|
|
"loss": 6.3055,
|
|
"mean_token_accuracy": 0.12608301565051078,
|
|
"num_tokens": 3770722.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 6.347224617004395,
|
|
"epoch": 0.10316207843824887,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004317494978701582,
|
|
"loss": 6.2764,
|
|
"mean_token_accuracy": 0.1338924378156662,
|
|
"num_tokens": 3781562.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 6.364816331863404,
|
|
"epoch": 0.10345432228651587,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004309021894472943,
|
|
"loss": 6.2817,
|
|
"mean_token_accuracy": 0.12620307505130768,
|
|
"num_tokens": 3791029.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 6.393705987930298,
|
|
"epoch": 0.10374656613478286,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004300506068871534,
|
|
"loss": 6.289,
|
|
"mean_token_accuracy": 0.13021805733442307,
|
|
"num_tokens": 3802526.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 6.30857138633728,
|
|
"epoch": 0.10403880998304986,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00042919477353630135,
|
|
"loss": 6.2654,
|
|
"mean_token_accuracy": 0.13339556902647018,
|
|
"num_tokens": 3813394.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 6.312510919570923,
|
|
"epoch": 0.10433105383131686,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000428334712857842,
|
|
"loss": 6.2581,
|
|
"mean_token_accuracy": 0.1305323101580143,
|
|
"num_tokens": 3824852.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 6.29648756980896,
|
|
"epoch": 0.10462329767958385,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00042747044843077304,
|
|
"loss": 6.2738,
|
|
"mean_token_accuracy": 0.12493606880307198,
|
|
"num_tokens": 3835743.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 6.343502330780029,
|
|
"epoch": 0.10491554152785083,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00042660200394934047,
|
|
"loss": 6.2605,
|
|
"mean_token_accuracy": 0.12992578819394113,
|
|
"num_tokens": 3846897.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.3801350593566895,
|
|
"epoch": 0.10520778537611783,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.00042572940322238844,
|
|
"loss": 6.2724,
|
|
"mean_token_accuracy": 0.13262551203370093,
|
|
"num_tokens": 3857351.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 6.275812339782715,
|
|
"epoch": 0.10550002922438483,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00042485267017270664,
|
|
"loss": 6.2366,
|
|
"mean_token_accuracy": 0.13671592473983765,
|
|
"num_tokens": 3867696.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 6.325400447845459,
|
|
"epoch": 0.10579227307265182,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004239718288363745,
|
|
"loss": 6.3275,
|
|
"mean_token_accuracy": 0.1320234626531601,
|
|
"num_tokens": 3878473.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 6.3584887981414795,
|
|
"epoch": 0.10608451692091882,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004230869033621023,
|
|
"loss": 6.3161,
|
|
"mean_token_accuracy": 0.13206197172403336,
|
|
"num_tokens": 3888750.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 6.311602640151977,
|
|
"epoch": 0.10637676076918581,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004221979180105688,
|
|
"loss": 6.3215,
|
|
"mean_token_accuracy": 0.12421507909893989,
|
|
"num_tokens": 3898867.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 6.32457480430603,
|
|
"epoch": 0.10666900461745281,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00042130489715375645,
|
|
"loss": 6.258,
|
|
"mean_token_accuracy": 0.13351302370429038,
|
|
"num_tokens": 3909103.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 6.318645715713501,
|
|
"epoch": 0.10696124846571979,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00042040786527428335,
|
|
"loss": 6.1914,
|
|
"mean_token_accuracy": 0.13429306745529174,
|
|
"num_tokens": 3919592.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 6.346998023986816,
|
|
"epoch": 0.10725349231398679,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004195068469647315,
|
|
"loss": 6.2732,
|
|
"mean_token_accuracy": 0.12874911576509476,
|
|
"num_tokens": 3930730.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 6.2924802780151365,
|
|
"epoch": 0.10754573616225378,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00041860186692697297,
|
|
"loss": 6.2902,
|
|
"mean_token_accuracy": 0.12385930791497231,
|
|
"num_tokens": 3942535.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 6.329216051101684,
|
|
"epoch": 0.10783798001052078,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00041769294997149264,
|
|
"loss": 6.2765,
|
|
"mean_token_accuracy": 0.12904328033328055,
|
|
"num_tokens": 3953651.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 6.392510271072387,
|
|
"epoch": 0.10813022385878777,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004167801210167081,
|
|
"loss": 6.2396,
|
|
"mean_token_accuracy": 0.13275438472628592,
|
|
"num_tokens": 3963501.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 6.2914917945861815,
|
|
"epoch": 0.10842246770705477,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004158634050882861,
|
|
"loss": 6.2307,
|
|
"mean_token_accuracy": 0.12977104857563973,
|
|
"num_tokens": 3973617.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 6.336210870742798,
|
|
"epoch": 0.10871471155532177,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004149428273184569,
|
|
"loss": 6.2988,
|
|
"mean_token_accuracy": 0.12246305719017983,
|
|
"num_tokens": 3985313.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 6.2909324169158936,
|
|
"epoch": 0.10900695540358875,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004140184129453253,
|
|
"loss": 6.24,
|
|
"mean_token_accuracy": 0.12916646972298623,
|
|
"num_tokens": 3995136.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 6.251706933975219,
|
|
"epoch": 0.10929919925185574,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 0.000413090187312178,
|
|
"loss": 6.1416,
|
|
"mean_token_accuracy": 0.14623225331306458,
|
|
"num_tokens": 4004633.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.329136228561401,
|
|
"epoch": 0.10959144310012274,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004121581758667898,
|
|
"loss": 6.2446,
|
|
"mean_token_accuracy": 0.1265272855758667,
|
|
"num_tokens": 4013949.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 6.2680024147033695,
|
|
"epoch": 0.10988368694838974,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00041122240416072533,
|
|
"loss": 6.2237,
|
|
"mean_token_accuracy": 0.13091572299599646,
|
|
"num_tokens": 4024456.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.260085582733154,
|
|
"epoch": 0.11017593079665673,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004102828978486385,
|
|
"loss": 6.2148,
|
|
"mean_token_accuracy": 0.14045012667775153,
|
|
"num_tokens": 4035271.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 6.353982257843017,
|
|
"epoch": 0.11046817464492373,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004093396826875695,
|
|
"loss": 6.2946,
|
|
"mean_token_accuracy": 0.13420966342091561,
|
|
"num_tokens": 4045861.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.334712219238281,
|
|
"epoch": 0.11076041849319072,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00040839278453623837,
|
|
"loss": 6.1899,
|
|
"mean_token_accuracy": 0.13066370338201522,
|
|
"num_tokens": 4056060.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 6.219831371307373,
|
|
"epoch": 0.1110526623414577,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004074422293543363,
|
|
"loss": 6.2573,
|
|
"mean_token_accuracy": 0.13225416764616965,
|
|
"num_tokens": 4068098.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.317618751525879,
|
|
"epoch": 0.1113449061897247,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004064880432018137,
|
|
"loss": 6.1764,
|
|
"mean_token_accuracy": 0.14364669099450111,
|
|
"num_tokens": 4079186.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 6.2356115818023685,
|
|
"epoch": 0.1116371500379917,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00040553025223816615,
|
|
"loss": 6.1873,
|
|
"mean_token_accuracy": 0.1371534802019596,
|
|
"num_tokens": 4089430.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 6.309043121337891,
|
|
"epoch": 0.1119293938862587,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00040456888272171653,
|
|
"loss": 6.189,
|
|
"mean_token_accuracy": 0.13867145776748657,
|
|
"num_tokens": 4099702.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.352001428604126,
|
|
"epoch": 0.11222163773452569,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00040360396100889577,
|
|
"loss": 6.2926,
|
|
"mean_token_accuracy": 0.12570124194025994,
|
|
"num_tokens": 4111108.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 6.280226993560791,
|
|
"epoch": 0.11251388158279269,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004026355135535202,
|
|
"loss": 6.2632,
|
|
"mean_token_accuracy": 0.12929408848285676,
|
|
"num_tokens": 4122470.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.2651409149169925,
|
|
"epoch": 0.11280612543105968,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000401663566906066,
|
|
"loss": 6.1811,
|
|
"mean_token_accuracy": 0.13483544066548347,
|
|
"num_tokens": 4132652.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.323620223999024,
|
|
"epoch": 0.11309836927932668,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00040068814771294134,
|
|
"loss": 6.2881,
|
|
"mean_token_accuracy": 0.12270993292331696,
|
|
"num_tokens": 4143449.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.342724847793579,
|
|
"epoch": 0.11339061312759366,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003997092827157562,
|
|
"loss": 6.2089,
|
|
"mean_token_accuracy": 0.136571054905653,
|
|
"num_tokens": 4153536.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.257393646240234,
|
|
"epoch": 0.11368285697586065,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000398726998750589,
|
|
"loss": 6.2617,
|
|
"mean_token_accuracy": 0.1267533928155899,
|
|
"num_tokens": 4165339.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.361189365386963,
|
|
"epoch": 0.11397510082412765,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00039774132274725076,
|
|
"loss": 6.3298,
|
|
"mean_token_accuracy": 0.12190727666020393,
|
|
"num_tokens": 4177171.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.362505912780762,
|
|
"epoch": 0.11426734467239465,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00039675228172854707,
|
|
"loss": 6.209,
|
|
"mean_token_accuracy": 0.13935441598296167,
|
|
"num_tokens": 4186577.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 6.115205717086792,
|
|
"epoch": 0.11455958852066164,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003957599028095371,
|
|
"loss": 6.2431,
|
|
"mean_token_accuracy": 0.13071082010865212,
|
|
"num_tokens": 4196962.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.336116027832031,
|
|
"epoch": 0.11485183236892864,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00039476421319679017,
|
|
"loss": 6.2418,
|
|
"mean_token_accuracy": 0.13397978022694587,
|
|
"num_tokens": 4207544.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.38735728263855,
|
|
"epoch": 0.11514407621719563,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00039376524018764,
|
|
"loss": 6.2176,
|
|
"mean_token_accuracy": 0.134936810284853,
|
|
"num_tokens": 4217359.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.221807336807251,
|
|
"epoch": 0.11543632006546262,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00039276301116943616,
|
|
"loss": 6.2375,
|
|
"mean_token_accuracy": 0.1313203439116478,
|
|
"num_tokens": 4227826.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.366882801055908,
|
|
"epoch": 0.11572856391372961,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0003917575536187936,
|
|
"loss": 6.1952,
|
|
"mean_token_accuracy": 0.12822012454271317,
|
|
"num_tokens": 4238591.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.311385679244995,
|
|
"epoch": 0.11602080776199661,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00039074889510083894,
|
|
"loss": 6.2126,
|
|
"mean_token_accuracy": 0.1306764416396618,
|
|
"num_tokens": 4248771.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.303131580352783,
|
|
"epoch": 0.1163130516102636,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00038973706326845495,
|
|
"loss": 6.2769,
|
|
"mean_token_accuracy": 0.1265501409769058,
|
|
"num_tokens": 4259234.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.277475070953369,
|
|
"epoch": 0.1166052954585306,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0003887220858615225,
|
|
"loss": 6.2888,
|
|
"mean_token_accuracy": 0.12363493666052819,
|
|
"num_tokens": 4270891.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.377652549743653,
|
|
"epoch": 0.1168975393067976,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0003877039907061597,
|
|
"loss": 6.2989,
|
|
"mean_token_accuracy": 0.1283785842359066,
|
|
"num_tokens": 4282893.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"entropy": 6.249635171890259,
|
|
"epoch": 0.11718978315506459,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003866828057139598,
|
|
"loss": 6.1859,
|
|
"mean_token_accuracy": 0.1322122819721699,
|
|
"num_tokens": 4292630.0,
|
|
"step": 2005
|
|
},
|
|
{
|
|
"entropy": 6.261950254440308,
|
|
"epoch": 0.11748202700333157,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00038565855888122503,
|
|
"loss": 6.2306,
|
|
"mean_token_accuracy": 0.13367020040750505,
|
|
"num_tokens": 4302849.0,
|
|
"step": 2010
|
|
},
|
|
{
|
|
"entropy": 6.2926130294799805,
|
|
"epoch": 0.11777427085159857,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00038463127828819975,
|
|
"loss": 6.2355,
|
|
"mean_token_accuracy": 0.13290671557188033,
|
|
"num_tokens": 4313178.0,
|
|
"step": 2015
|
|
},
|
|
{
|
|
"entropy": 6.382750606536865,
|
|
"epoch": 0.11806651469986557,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00038360099209830043,
|
|
"loss": 6.2729,
|
|
"mean_token_accuracy": 0.13396400362253189,
|
|
"num_tokens": 4323205.0,
|
|
"step": 2020
|
|
},
|
|
{
|
|
"entropy": 6.220796394348144,
|
|
"epoch": 0.11835875854813256,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003825677285573433,
|
|
"loss": 6.2158,
|
|
"mean_token_accuracy": 0.13015230521559715,
|
|
"num_tokens": 4333840.0,
|
|
"step": 2025
|
|
},
|
|
{
|
|
"entropy": 6.344536399841308,
|
|
"epoch": 0.11865100239639956,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00038153151599277027,
|
|
"loss": 6.2155,
|
|
"mean_token_accuracy": 0.13493136763572694,
|
|
"num_tokens": 4344489.0,
|
|
"step": 2030
|
|
},
|
|
{
|
|
"entropy": 6.278243160247802,
|
|
"epoch": 0.11894324624466655,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003804923828128723,
|
|
"loss": 6.1856,
|
|
"mean_token_accuracy": 0.13395058140158653,
|
|
"num_tokens": 4355008.0,
|
|
"step": 2035
|
|
},
|
|
{
|
|
"entropy": 6.238671875,
|
|
"epoch": 0.11923549009293355,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003794503575060104,
|
|
"loss": 6.2135,
|
|
"mean_token_accuracy": 0.1329597570002079,
|
|
"num_tokens": 4365944.0,
|
|
"step": 2040
|
|
},
|
|
{
|
|
"entropy": 6.319523668289184,
|
|
"epoch": 0.11952773394120053,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00037840546863983484,
|
|
"loss": 6.2223,
|
|
"mean_token_accuracy": 0.13217944651842117,
|
|
"num_tokens": 4376448.0,
|
|
"step": 2045
|
|
},
|
|
{
|
|
"entropy": 6.299238443374634,
|
|
"epoch": 0.11981997778946753,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003773577448605015,
|
|
"loss": 6.1553,
|
|
"mean_token_accuracy": 0.13440817669034005,
|
|
"num_tokens": 4386748.0,
|
|
"step": 2050
|
|
},
|
|
{
|
|
"entropy": 6.270139694213867,
|
|
"epoch": 0.12011222163773452,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003763072148918872,
|
|
"loss": 6.1454,
|
|
"mean_token_accuracy": 0.14509789347648622,
|
|
"num_tokens": 4396956.0,
|
|
"step": 2055
|
|
},
|
|
{
|
|
"entropy": 6.2123939037323,
|
|
"epoch": 0.12040446548600152,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0003752539075348017,
|
|
"loss": 6.1742,
|
|
"mean_token_accuracy": 0.1364543430507183,
|
|
"num_tokens": 4406551.0,
|
|
"step": 2060
|
|
},
|
|
{
|
|
"entropy": 6.310663366317749,
|
|
"epoch": 0.12069670933426851,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00037419785166619817,
|
|
"loss": 6.1642,
|
|
"mean_token_accuracy": 0.13440534472465515,
|
|
"num_tokens": 4416847.0,
|
|
"step": 2065
|
|
},
|
|
{
|
|
"entropy": 6.242231559753418,
|
|
"epoch": 0.12098895318253551,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003731390762383818,
|
|
"loss": 6.1556,
|
|
"mean_token_accuracy": 0.1388556867837906,
|
|
"num_tokens": 4427579.0,
|
|
"step": 2070
|
|
},
|
|
{
|
|
"entropy": 6.2008545875549315,
|
|
"epoch": 0.1212811970308025,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003720776102782158,
|
|
"loss": 6.1474,
|
|
"mean_token_accuracy": 0.13912375494837761,
|
|
"num_tokens": 4438473.0,
|
|
"step": 2075
|
|
},
|
|
{
|
|
"entropy": 6.234990024566651,
|
|
"epoch": 0.12157344087906949,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00037101348288632555,
|
|
"loss": 6.1402,
|
|
"mean_token_accuracy": 0.1427859090268612,
|
|
"num_tokens": 4448851.0,
|
|
"step": 2080
|
|
},
|
|
{
|
|
"entropy": 6.150416326522827,
|
|
"epoch": 0.12186568472733648,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003699467232363012,
|
|
"loss": 6.1615,
|
|
"mean_token_accuracy": 0.13463753834366798,
|
|
"num_tokens": 4459612.0,
|
|
"step": 2085
|
|
},
|
|
{
|
|
"entropy": 6.215969181060791,
|
|
"epoch": 0.12215792857560348,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0003688773605738973,
|
|
"loss": 6.112,
|
|
"mean_token_accuracy": 0.13399859592318536,
|
|
"num_tokens": 4470365.0,
|
|
"step": 2090
|
|
},
|
|
{
|
|
"entropy": 6.249767017364502,
|
|
"epoch": 0.12245017242387048,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00036780542421623134,
|
|
"loss": 6.1956,
|
|
"mean_token_accuracy": 0.1419701375067234,
|
|
"num_tokens": 4481329.0,
|
|
"step": 2095
|
|
},
|
|
{
|
|
"entropy": 6.248475027084351,
|
|
"epoch": 0.12274241627213747,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003667309435509802,
|
|
"loss": 6.1503,
|
|
"mean_token_accuracy": 0.12978014945983887,
|
|
"num_tokens": 4491904.0,
|
|
"step": 2100
|
|
},
|
|
{
|
|
"entropy": 6.251853132247925,
|
|
"epoch": 0.12303466012040447,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003656539480355741,
|
|
"loss": 6.0943,
|
|
"mean_token_accuracy": 0.1463254511356354,
|
|
"num_tokens": 4501432.0,
|
|
"step": 2105
|
|
},
|
|
{
|
|
"entropy": 6.2275115013122555,
|
|
"epoch": 0.12332690396867146,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003645744671963891,
|
|
"loss": 6.096,
|
|
"mean_token_accuracy": 0.13819159865379332,
|
|
"num_tokens": 4511512.0,
|
|
"step": 2110
|
|
},
|
|
{
|
|
"entropy": 6.2278444290161135,
|
|
"epoch": 0.12361914781693846,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0003634925306279376,
|
|
"loss": 6.207,
|
|
"mean_token_accuracy": 0.13566816449165345,
|
|
"num_tokens": 4521701.0,
|
|
"step": 2115
|
|
},
|
|
{
|
|
"entropy": 6.278735494613647,
|
|
"epoch": 0.12391139166520544,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003624081679920574,
|
|
"loss": 6.1613,
|
|
"mean_token_accuracy": 0.14083656966686248,
|
|
"num_tokens": 4531512.0,
|
|
"step": 2120
|
|
},
|
|
{
|
|
"entropy": 6.239475536346435,
|
|
"epoch": 0.12420363551347244,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003613214090170977,
|
|
"loss": 6.1391,
|
|
"mean_token_accuracy": 0.1424753561615944,
|
|
"num_tokens": 4542083.0,
|
|
"step": 2125
|
|
},
|
|
{
|
|
"entropy": 6.218933725357056,
|
|
"epoch": 0.12449587936173943,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0003602322834971048,
|
|
"loss": 6.1574,
|
|
"mean_token_accuracy": 0.13780356496572493,
|
|
"num_tokens": 4553207.0,
|
|
"step": 2130
|
|
},
|
|
{
|
|
"entropy": 6.344869232177734,
|
|
"epoch": 0.12478812321000643,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003591408212910051,
|
|
"loss": 6.1931,
|
|
"mean_token_accuracy": 0.13503697365522385,
|
|
"num_tokens": 4564349.0,
|
|
"step": 2135
|
|
},
|
|
{
|
|
"entropy": 6.253999805450439,
|
|
"epoch": 0.1250803670582734,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003580470523217863,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.1333679050207138,
|
|
"num_tokens": 4574935.0,
|
|
"step": 2140
|
|
},
|
|
{
|
|
"entropy": 6.1710728168487545,
|
|
"epoch": 0.12537261090654042,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003569510065756771,
|
|
"loss": 6.1987,
|
|
"mean_token_accuracy": 0.1306650847196579,
|
|
"num_tokens": 4586439.0,
|
|
"step": 2145
|
|
},
|
|
{
|
|
"entropy": 6.208186054229737,
|
|
"epoch": 0.1256648547548074,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0003558527141013254,
|
|
"loss": 6.056,
|
|
"mean_token_accuracy": 0.14841998964548112,
|
|
"num_tokens": 4596439.0,
|
|
"step": 2150
|
|
},
|
|
{
|
|
"entropy": 6.174758529663086,
|
|
"epoch": 0.1259570986030744,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003547522050089742,
|
|
"loss": 6.1309,
|
|
"mean_token_accuracy": 0.1388734348118305,
|
|
"num_tokens": 4606920.0,
|
|
"step": 2155
|
|
},
|
|
{
|
|
"entropy": 6.334303569793701,
|
|
"epoch": 0.1262493424513414,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00035364950946963606,
|
|
"loss": 6.1975,
|
|
"mean_token_accuracy": 0.13557516783475876,
|
|
"num_tokens": 4618234.0,
|
|
"step": 2160
|
|
},
|
|
{
|
|
"entropy": 6.131727790832519,
|
|
"epoch": 0.1265415862996084,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003525446577142663,
|
|
"loss": 6.006,
|
|
"mean_token_accuracy": 0.15448770821094512,
|
|
"num_tokens": 4629201.0,
|
|
"step": 2165
|
|
},
|
|
{
|
|
"entropy": 6.234293460845947,
|
|
"epoch": 0.1268338301478754,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00035143768003293395,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.1372879721224308,
|
|
"num_tokens": 4641008.0,
|
|
"step": 2170
|
|
},
|
|
{
|
|
"entropy": 6.19882378578186,
|
|
"epoch": 0.12712607399614237,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0003503286067739913,
|
|
"loss": 6.1479,
|
|
"mean_token_accuracy": 0.1357593283057213,
|
|
"num_tokens": 4651969.0,
|
|
"step": 2175
|
|
},
|
|
{
|
|
"entropy": 6.329754066467285,
|
|
"epoch": 0.12741831784440938,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00034921746834324193,
|
|
"loss": 6.1915,
|
|
"mean_token_accuracy": 0.12660348191857337,
|
|
"num_tokens": 4663081.0,
|
|
"step": 2180
|
|
},
|
|
{
|
|
"entropy": 6.133418941497803,
|
|
"epoch": 0.12771056169267636,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003481042952031072,
|
|
"loss": 5.9656,
|
|
"mean_token_accuracy": 0.15549228340387344,
|
|
"num_tokens": 4673147.0,
|
|
"step": 2185
|
|
},
|
|
{
|
|
"entropy": 6.121980333328247,
|
|
"epoch": 0.12800280554094337,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003469891178717911,
|
|
"loss": 6.1111,
|
|
"mean_token_accuracy": 0.138456729054451,
|
|
"num_tokens": 4683237.0,
|
|
"step": 2190
|
|
},
|
|
{
|
|
"entropy": 6.280539798736572,
|
|
"epoch": 0.12829504938921035,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0003458719669224436,
|
|
"loss": 6.0805,
|
|
"mean_token_accuracy": 0.1401494786143303,
|
|
"num_tokens": 4694541.0,
|
|
"step": 2195
|
|
},
|
|
{
|
|
"entropy": 6.218093156814575,
|
|
"epoch": 0.12858729323747736,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0003447528729823221,
|
|
"loss": 6.1361,
|
|
"mean_token_accuracy": 0.13514513298869132,
|
|
"num_tokens": 4704583.0,
|
|
"step": 2200
|
|
},
|
|
{
|
|
"entropy": 6.1446513652801515,
|
|
"epoch": 0.12887953708574434,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003436318667319525,
|
|
"loss": 6.1118,
|
|
"mean_token_accuracy": 0.14716241881251335,
|
|
"num_tokens": 4715880.0,
|
|
"step": 2205
|
|
},
|
|
{
|
|
"entropy": 6.24896502494812,
|
|
"epoch": 0.12917178093401133,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00034250897890428716,
|
|
"loss": 6.12,
|
|
"mean_token_accuracy": 0.13427569568157197,
|
|
"num_tokens": 4726428.0,
|
|
"step": 2210
|
|
},
|
|
{
|
|
"entropy": 6.245270919799805,
|
|
"epoch": 0.12946402478227834,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0003413842402838633,
|
|
"loss": 6.0904,
|
|
"mean_token_accuracy": 0.13762594237923623,
|
|
"num_tokens": 4736782.0,
|
|
"step": 2215
|
|
},
|
|
{
|
|
"entropy": 6.168338775634766,
|
|
"epoch": 0.12975626863054532,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.00034025768170595834,
|
|
"loss": 6.1812,
|
|
"mean_token_accuracy": 0.14207304567098616,
|
|
"num_tokens": 4747084.0,
|
|
"step": 2220
|
|
},
|
|
{
|
|
"entropy": 6.260279273986816,
|
|
"epoch": 0.13004851247881233,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0003391293340557446,
|
|
"loss": 6.1211,
|
|
"mean_token_accuracy": 0.13106030598282814,
|
|
"num_tokens": 4758499.0,
|
|
"step": 2225
|
|
},
|
|
{
|
|
"entropy": 6.233145523071289,
|
|
"epoch": 0.1303407563270793,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003379992282674431,
|
|
"loss": 6.1241,
|
|
"mean_token_accuracy": 0.1361076258122921,
|
|
"num_tokens": 4768553.0,
|
|
"step": 2230
|
|
},
|
|
{
|
|
"entropy": 6.198996543884277,
|
|
"epoch": 0.13063300017534632,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003368673953234749,
|
|
"loss": 6.0605,
|
|
"mean_token_accuracy": 0.14450395554304124,
|
|
"num_tokens": 4778214.0,
|
|
"step": 2235
|
|
},
|
|
{
|
|
"entropy": 6.198760271072388,
|
|
"epoch": 0.1309252440236133,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00033573386625361176,
|
|
"loss": 6.133,
|
|
"mean_token_accuracy": 0.137807933986187,
|
|
"num_tokens": 4789870.0,
|
|
"step": 2240
|
|
},
|
|
{
|
|
"entropy": 6.0669878959655765,
|
|
"epoch": 0.13121748787188028,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00033459867213412567,
|
|
"loss": 5.8788,
|
|
"mean_token_accuracy": 0.16402744203805925,
|
|
"num_tokens": 4801320.0,
|
|
"step": 2245
|
|
},
|
|
{
|
|
"entropy": 6.114277029037476,
|
|
"epoch": 0.1315097317201473,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000333461844086937,
|
|
"loss": 6.1554,
|
|
"mean_token_accuracy": 0.14035945236682892,
|
|
"num_tokens": 4812170.0,
|
|
"step": 2250
|
|
},
|
|
{
|
|
"entropy": 6.273629903793335,
|
|
"epoch": 0.13180197556841428,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033232341327876097,
|
|
"loss": 6.1522,
|
|
"mean_token_accuracy": 0.13791727647185326,
|
|
"num_tokens": 4823279.0,
|
|
"step": 2255
|
|
},
|
|
{
|
|
"entropy": 6.267527294158936,
|
|
"epoch": 0.13209421941668129,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0003311834109202531,
|
|
"loss": 6.0096,
|
|
"mean_token_accuracy": 0.14894347488880158,
|
|
"num_tokens": 4833590.0,
|
|
"step": 2260
|
|
},
|
|
{
|
|
"entropy": 6.066952228546143,
|
|
"epoch": 0.13238646326494827,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00033004186826515416,
|
|
"loss": 6.0902,
|
|
"mean_token_accuracy": 0.13956383019685745,
|
|
"num_tokens": 4844364.0,
|
|
"step": 2265
|
|
},
|
|
{
|
|
"entropy": 6.272472238540649,
|
|
"epoch": 0.13267870711321528,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003288988166094324,
|
|
"loss": 6.0771,
|
|
"mean_token_accuracy": 0.13904936313629152,
|
|
"num_tokens": 4854375.0,
|
|
"step": 2270
|
|
},
|
|
{
|
|
"entropy": 6.233609294891357,
|
|
"epoch": 0.13297095096148226,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00032775428729042656,
|
|
"loss": 6.1009,
|
|
"mean_token_accuracy": 0.1401657611131668,
|
|
"num_tokens": 4864704.0,
|
|
"step": 2275
|
|
},
|
|
{
|
|
"entropy": 6.175923871994018,
|
|
"epoch": 0.13326319480974924,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000326608311685986,
|
|
"loss": 6.0895,
|
|
"mean_token_accuracy": 0.1436429314315319,
|
|
"num_tokens": 4874774.0,
|
|
"step": 2280
|
|
},
|
|
{
|
|
"entropy": 6.119866847991943,
|
|
"epoch": 0.13355543865801625,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003254609212136108,
|
|
"loss": 6.1121,
|
|
"mean_token_accuracy": 0.14237722456455232,
|
|
"num_tokens": 4886263.0,
|
|
"step": 2285
|
|
},
|
|
{
|
|
"entropy": 6.14636082649231,
|
|
"epoch": 0.13384768250628323,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00032431214732959036,
|
|
"loss": 6.139,
|
|
"mean_token_accuracy": 0.13580704405903815,
|
|
"num_tokens": 4897416.0,
|
|
"step": 2290
|
|
},
|
|
{
|
|
"entropy": 6.2751532077789305,
|
|
"epoch": 0.13413992635455024,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000323162021528141,
|
|
"loss": 6.0936,
|
|
"mean_token_accuracy": 0.144794699549675,
|
|
"num_tokens": 4907710.0,
|
|
"step": 2295
|
|
},
|
|
{
|
|
"entropy": 6.2298330783844,
|
|
"epoch": 0.13443217020281722,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032201057534054264,
|
|
"loss": 6.155,
|
|
"mean_token_accuracy": 0.13874071165919305,
|
|
"num_tokens": 4919190.0,
|
|
"step": 2300
|
|
},
|
|
{
|
|
"entropy": 6.189581823348999,
|
|
"epoch": 0.13472441405108423,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.00032085784033427414,
|
|
"loss": 6.1164,
|
|
"mean_token_accuracy": 0.1413622170686722,
|
|
"num_tokens": 4929892.0,
|
|
"step": 2305
|
|
},
|
|
{
|
|
"entropy": 6.261451244354248,
|
|
"epoch": 0.13501665789935122,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003197038481121478,
|
|
"loss": 6.1305,
|
|
"mean_token_accuracy": 0.1402155190706253,
|
|
"num_tokens": 4940649.0,
|
|
"step": 2310
|
|
},
|
|
{
|
|
"entropy": 6.201195812225341,
|
|
"epoch": 0.1353089017476182,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0003185486303114436,
|
|
"loss": 6.1781,
|
|
"mean_token_accuracy": 0.1384969875216484,
|
|
"num_tokens": 4950952.0,
|
|
"step": 2315
|
|
},
|
|
{
|
|
"entropy": 6.243939065933228,
|
|
"epoch": 0.1356011455958852,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003173922186030409,
|
|
"loss": 6.0969,
|
|
"mean_token_accuracy": 0.1391613446176052,
|
|
"num_tokens": 4961473.0,
|
|
"step": 2320
|
|
},
|
|
{
|
|
"entropy": 6.227145767211914,
|
|
"epoch": 0.1358933894441522,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000316234644690551,
|
|
"loss": 6.158,
|
|
"mean_token_accuracy": 0.13428287506103515,
|
|
"num_tokens": 4972702.0,
|
|
"step": 2325
|
|
},
|
|
{
|
|
"entropy": 6.225094270706177,
|
|
"epoch": 0.1361856332924192,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003150759403094473,
|
|
"loss": 6.1394,
|
|
"mean_token_accuracy": 0.13642262294888496,
|
|
"num_tokens": 4983003.0,
|
|
"step": 2330
|
|
},
|
|
{
|
|
"entropy": 6.138577270507812,
|
|
"epoch": 0.13647787714068618,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00031391613722619587,
|
|
"loss": 6.0828,
|
|
"mean_token_accuracy": 0.14404839053750038,
|
|
"num_tokens": 4993378.0,
|
|
"step": 2335
|
|
},
|
|
{
|
|
"entropy": 6.229124355316162,
|
|
"epoch": 0.1367701209889532,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000312755267237384,
|
|
"loss": 6.1788,
|
|
"mean_token_accuracy": 0.1317272149026394,
|
|
"num_tokens": 5004076.0,
|
|
"step": 2340
|
|
},
|
|
{
|
|
"entropy": 6.341947317123413,
|
|
"epoch": 0.13706236483722017,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003115933621688488,
|
|
"loss": 6.1481,
|
|
"mean_token_accuracy": 0.1339823916554451,
|
|
"num_tokens": 5014718.0,
|
|
"step": 2345
|
|
},
|
|
{
|
|
"entropy": 6.240772914886475,
|
|
"epoch": 0.13735460868548718,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.00031043045387480487,
|
|
"loss": 6.1432,
|
|
"mean_token_accuracy": 0.13380517512559892,
|
|
"num_tokens": 5026102.0,
|
|
"step": 2350
|
|
},
|
|
{
|
|
"entropy": 6.144141101837159,
|
|
"epoch": 0.13764685253375417,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003092665742369703,
|
|
"loss": 6.0743,
|
|
"mean_token_accuracy": 0.13829942494630815,
|
|
"num_tokens": 5036847.0,
|
|
"step": 2355
|
|
},
|
|
{
|
|
"entropy": 6.189305591583252,
|
|
"epoch": 0.13793909638202115,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00030810175516369343,
|
|
"loss": 6.0016,
|
|
"mean_token_accuracy": 0.15031686201691627,
|
|
"num_tokens": 5047751.0,
|
|
"step": 2360
|
|
},
|
|
{
|
|
"entropy": 6.242601823806763,
|
|
"epoch": 0.13823134023028816,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0003069360285890775,
|
|
"loss": 6.0581,
|
|
"mean_token_accuracy": 0.1428899347782135,
|
|
"num_tokens": 5057914.0,
|
|
"step": 2365
|
|
},
|
|
{
|
|
"entropy": 6.1711660861969,
|
|
"epoch": 0.13852358407855514,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00030576942647210547,
|
|
"loss": 6.123,
|
|
"mean_token_accuracy": 0.13385452926158906,
|
|
"num_tokens": 5068465.0,
|
|
"step": 2370
|
|
},
|
|
{
|
|
"entropy": 6.139814281463623,
|
|
"epoch": 0.13881582792682215,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00030460198079576355,
|
|
"loss": 6.0607,
|
|
"mean_token_accuracy": 0.14090710505843163,
|
|
"num_tokens": 5079162.0,
|
|
"step": 2375
|
|
},
|
|
{
|
|
"entropy": 6.18707013130188,
|
|
"epoch": 0.13910807177508913,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003034337235661648,
|
|
"loss": 6.1177,
|
|
"mean_token_accuracy": 0.13520955443382263,
|
|
"num_tokens": 5089409.0,
|
|
"step": 2380
|
|
},
|
|
{
|
|
"entropy": 6.231440830230713,
|
|
"epoch": 0.13940031562335614,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0003022646868116714,
|
|
"loss": 6.0331,
|
|
"mean_token_accuracy": 0.14986666589975356,
|
|
"num_tokens": 5100150.0,
|
|
"step": 2385
|
|
},
|
|
{
|
|
"entropy": 6.166530799865723,
|
|
"epoch": 0.13969255947162312,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0003010949025820163,
|
|
"loss": 6.006,
|
|
"mean_token_accuracy": 0.14616978093981742,
|
|
"num_tokens": 5110601.0,
|
|
"step": 2390
|
|
},
|
|
{
|
|
"entropy": 6.099657964706421,
|
|
"epoch": 0.1399848033198901,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002999244029474252,
|
|
"loss": 6.097,
|
|
"mean_token_accuracy": 0.1380453534424305,
|
|
"num_tokens": 5119665.0,
|
|
"step": 2395
|
|
},
|
|
{
|
|
"entropy": 6.14389591217041,
|
|
"epoch": 0.14027704716815712,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00029875321999773684,
|
|
"loss": 6.0084,
|
|
"mean_token_accuracy": 0.1483282908797264,
|
|
"num_tokens": 5129712.0,
|
|
"step": 2400
|
|
},
|
|
{
|
|
"entropy": 6.248686361312866,
|
|
"epoch": 0.1405692910164241,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.00029758138584152333,
|
|
"loss": 6.1013,
|
|
"mean_token_accuracy": 0.14273960143327713,
|
|
"num_tokens": 5139858.0,
|
|
"step": 2405
|
|
},
|
|
{
|
|
"entropy": 6.252632999420166,
|
|
"epoch": 0.1408615348646911,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0002964089326052102,
|
|
"loss": 6.0796,
|
|
"mean_token_accuracy": 0.13942125216126441,
|
|
"num_tokens": 5149330.0,
|
|
"step": 2410
|
|
},
|
|
{
|
|
"entropy": 6.133875226974487,
|
|
"epoch": 0.1411537787129581,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002952358924321949,
|
|
"loss": 6.0689,
|
|
"mean_token_accuracy": 0.14331426620483398,
|
|
"num_tokens": 5159458.0,
|
|
"step": 2415
|
|
},
|
|
{
|
|
"entropy": 6.215303516387939,
|
|
"epoch": 0.1414460225612251,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00029406229748196657,
|
|
"loss": 6.0775,
|
|
"mean_token_accuracy": 0.14083075076341628,
|
|
"num_tokens": 5170439.0,
|
|
"step": 2420
|
|
},
|
|
{
|
|
"entropy": 6.1326995372772215,
|
|
"epoch": 0.14173826640949208,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0002928881799292235,
|
|
"loss": 6.0245,
|
|
"mean_token_accuracy": 0.14517416507005693,
|
|
"num_tokens": 5181345.0,
|
|
"step": 2425
|
|
},
|
|
{
|
|
"entropy": 6.190633773803711,
|
|
"epoch": 0.14203051025775906,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00029171357196299154,
|
|
"loss": 6.0229,
|
|
"mean_token_accuracy": 0.1475578874349594,
|
|
"num_tokens": 5191271.0,
|
|
"step": 2430
|
|
},
|
|
{
|
|
"entropy": 6.119750928878784,
|
|
"epoch": 0.14232275410602607,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0002905385057857414,
|
|
"loss": 6.0537,
|
|
"mean_token_accuracy": 0.1410386562347412,
|
|
"num_tokens": 5201868.0,
|
|
"step": 2435
|
|
},
|
|
{
|
|
"entropy": 6.17066764831543,
|
|
"epoch": 0.14261499795429305,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002893630136125058,
|
|
"loss": 6.0477,
|
|
"mean_token_accuracy": 0.14192107766866685,
|
|
"num_tokens": 5212463.0,
|
|
"step": 2440
|
|
},
|
|
{
|
|
"entropy": 6.21774115562439,
|
|
"epoch": 0.14290724180256006,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0002881871276699967,
|
|
"loss": 6.0601,
|
|
"mean_token_accuracy": 0.14529359042644502,
|
|
"num_tokens": 5223485.0,
|
|
"step": 2445
|
|
},
|
|
{
|
|
"entropy": 6.125836896896362,
|
|
"epoch": 0.14319948565082705,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00028701088019572114,
|
|
"loss": 6.0074,
|
|
"mean_token_accuracy": 0.15054480880498886,
|
|
"num_tokens": 5233818.0,
|
|
"step": 2450
|
|
},
|
|
{
|
|
"entropy": 6.130956363677979,
|
|
"epoch": 0.14349172949909406,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0002858343034370977,
|
|
"loss": 6.0491,
|
|
"mean_token_accuracy": 0.15077575743198396,
|
|
"num_tokens": 5244008.0,
|
|
"step": 2455
|
|
},
|
|
{
|
|
"entropy": 6.12300705909729,
|
|
"epoch": 0.14378397334736104,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.00028465742965057267,
|
|
"loss": 5.9782,
|
|
"mean_token_accuracy": 0.15135489031672478,
|
|
"num_tokens": 5254263.0,
|
|
"step": 2460
|
|
},
|
|
{
|
|
"entropy": 6.200008344650269,
|
|
"epoch": 0.14407621719562802,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028348029110073533,
|
|
"loss": 6.0761,
|
|
"mean_token_accuracy": 0.1395764470100403,
|
|
"num_tokens": 5264635.0,
|
|
"step": 2465
|
|
},
|
|
{
|
|
"entropy": 6.163009548187256,
|
|
"epoch": 0.14436846104389503,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.00028230292005943365,
|
|
"loss": 6.0824,
|
|
"mean_token_accuracy": 0.14634474888443946,
|
|
"num_tokens": 5275795.0,
|
|
"step": 2470
|
|
},
|
|
{
|
|
"entropy": 6.134585332870484,
|
|
"epoch": 0.144660704892162,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028112534880488945,
|
|
"loss": 6.018,
|
|
"mean_token_accuracy": 0.1454907976090908,
|
|
"num_tokens": 5286188.0,
|
|
"step": 2475
|
|
},
|
|
{
|
|
"entropy": 6.1603742122650145,
|
|
"epoch": 0.14495294874042902,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002799476096208137,
|
|
"loss": 6.03,
|
|
"mean_token_accuracy": 0.14633230417966842,
|
|
"num_tokens": 5296224.0,
|
|
"step": 2480
|
|
},
|
|
{
|
|
"entropy": 6.133947277069092,
|
|
"epoch": 0.145245192588696,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00027876973479552087,
|
|
"loss": 6.0368,
|
|
"mean_token_accuracy": 0.1396023951470852,
|
|
"num_tokens": 5308199.0,
|
|
"step": 2485
|
|
},
|
|
{
|
|
"entropy": 6.183510255813599,
|
|
"epoch": 0.145537436436963,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.00027759175662104424,
|
|
"loss": 6.0316,
|
|
"mean_token_accuracy": 0.144721619784832,
|
|
"num_tokens": 5319141.0,
|
|
"step": 2490
|
|
},
|
|
{
|
|
"entropy": 6.197761201858521,
|
|
"epoch": 0.14582968028523,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0002764137073922508,
|
|
"loss": 6.1171,
|
|
"mean_token_accuracy": 0.142631995677948,
|
|
"num_tokens": 5331387.0,
|
|
"step": 2495
|
|
},
|
|
{
|
|
"entropy": 6.18598370552063,
|
|
"epoch": 0.14612192413349698,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00027523561940595505,
|
|
"loss": 6.0048,
|
|
"mean_token_accuracy": 0.14339563697576524,
|
|
"num_tokens": 5341374.0,
|
|
"step": 2500
|
|
}
|
|
],
|
|
"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": 7946503096320000.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|