1035 lines
27 KiB
JSON
1035 lines
27 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 0.010720756456575576,
|
|
"eval_steps": 500,
|
|
"global_step": 500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74260492324829,
|
|
"epoch": 0.00010720756456575576,
|
|
"grad_norm": 5.4375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.6391,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 6321.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742601490020752,
|
|
"epoch": 0.00021441512913151153,
|
|
"grad_norm": 4.96875,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.6286,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 12304.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742689228057861,
|
|
"epoch": 0.0003216226936972673,
|
|
"grad_norm": 5.875,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.5323,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 17279.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742684173583985,
|
|
"epoch": 0.00042883025826302305,
|
|
"grad_norm": 5.125,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.5169,
|
|
"mean_token_accuracy": 0.00015649452107027173,
|
|
"num_tokens": 22498.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742647933959962,
|
|
"epoch": 0.0005360378228287788,
|
|
"grad_norm": 4.78125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.4956,
|
|
"mean_token_accuracy": 0.004493952146731317,
|
|
"num_tokens": 29007.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.742504405975343,
|
|
"epoch": 0.0006432453873945346,
|
|
"grad_norm": 4.625,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.3649,
|
|
"mean_token_accuracy": 0.018146800808608532,
|
|
"num_tokens": 35097.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.74176778793335,
|
|
"epoch": 0.0007504529519602903,
|
|
"grad_norm": 3.71875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.2322,
|
|
"mean_token_accuracy": 0.030227267555892466,
|
|
"num_tokens": 41259.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.740076065063477,
|
|
"epoch": 0.0008576605165260461,
|
|
"grad_norm": 3.3125,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.0467,
|
|
"mean_token_accuracy": 0.04209401421248913,
|
|
"num_tokens": 46775.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.73698902130127,
|
|
"epoch": 0.0009648680810918018,
|
|
"grad_norm": 2.640625,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 9.94,
|
|
"mean_token_accuracy": 0.04311881065368652,
|
|
"num_tokens": 52411.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.73378677368164,
|
|
"epoch": 0.0010720756456575576,
|
|
"grad_norm": 2.5625,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.867,
|
|
"mean_token_accuracy": 0.04206637293100357,
|
|
"num_tokens": 57713.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.732164478302002,
|
|
"epoch": 0.0011792832102233133,
|
|
"grad_norm": 2.171875,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.8389,
|
|
"mean_token_accuracy": 0.03966082856059074,
|
|
"num_tokens": 64001.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.731791687011718,
|
|
"epoch": 0.0012864907747890692,
|
|
"grad_norm": 3.109375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.729,
|
|
"mean_token_accuracy": 0.04017042927443981,
|
|
"num_tokens": 69383.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.730110168457031,
|
|
"epoch": 0.001393698339354825,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.7022,
|
|
"mean_token_accuracy": 0.05106379985809326,
|
|
"num_tokens": 75190.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.725842094421386,
|
|
"epoch": 0.0015009059039205806,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.6572,
|
|
"mean_token_accuracy": 0.04851874075829983,
|
|
"num_tokens": 81538.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.719818305969238,
|
|
"epoch": 0.0016081134684863363,
|
|
"grad_norm": 2.15625,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.5078,
|
|
"mean_token_accuracy": 0.047308453172445294,
|
|
"num_tokens": 86635.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.713384342193603,
|
|
"epoch": 0.0017153210330520922,
|
|
"grad_norm": 2.25,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.4751,
|
|
"mean_token_accuracy": 0.04916247427463531,
|
|
"num_tokens": 91873.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.707803440093993,
|
|
"epoch": 0.001822528597617848,
|
|
"grad_norm": 2.109375,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.4427,
|
|
"mean_token_accuracy": 0.05834655873477459,
|
|
"num_tokens": 97100.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.701716423034668,
|
|
"epoch": 0.0019297361621836036,
|
|
"grad_norm": 2.453125,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.3704,
|
|
"mean_token_accuracy": 0.06324269101023675,
|
|
"num_tokens": 103029.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.693789005279541,
|
|
"epoch": 0.0020369437267493593,
|
|
"grad_norm": 2.21875,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.2856,
|
|
"mean_token_accuracy": 0.06344883255660534,
|
|
"num_tokens": 108696.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.681031703948975,
|
|
"epoch": 0.002144151291315115,
|
|
"grad_norm": 1.9140625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.2188,
|
|
"mean_token_accuracy": 0.06291567087173462,
|
|
"num_tokens": 114384.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.65079288482666,
|
|
"epoch": 0.002251358855880871,
|
|
"grad_norm": 1.8828125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1923,
|
|
"mean_token_accuracy": 0.059306546673178674,
|
|
"num_tokens": 120985.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.60266056060791,
|
|
"epoch": 0.0023585664204466266,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 8.8874,
|
|
"mean_token_accuracy": 0.0715036116540432,
|
|
"num_tokens": 125649.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.532261848449707,
|
|
"epoch": 0.0024657739850123825,
|
|
"grad_norm": 1.8359375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9019,
|
|
"mean_token_accuracy": 0.06274469867348671,
|
|
"num_tokens": 131226.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.46213207244873,
|
|
"epoch": 0.0025729815495781384,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8396,
|
|
"mean_token_accuracy": 0.06673483438789844,
|
|
"num_tokens": 136914.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.398304557800293,
|
|
"epoch": 0.002680189114143894,
|
|
"grad_norm": 1.7890625,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.8445,
|
|
"mean_token_accuracy": 0.06518776528537273,
|
|
"num_tokens": 143746.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.363828659057617,
|
|
"epoch": 0.00278739667870965,
|
|
"grad_norm": 1.765625,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6721,
|
|
"mean_token_accuracy": 0.06042128652334213,
|
|
"num_tokens": 150145.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.312844944000243,
|
|
"epoch": 0.0028946042432754053,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5006,
|
|
"mean_token_accuracy": 0.06757240034639836,
|
|
"num_tokens": 155917.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.220444011688233,
|
|
"epoch": 0.003001811807841161,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.3604,
|
|
"mean_token_accuracy": 0.0706567782908678,
|
|
"num_tokens": 160942.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.092265033721924,
|
|
"epoch": 0.003109019372406917,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3328,
|
|
"mean_token_accuracy": 0.06713153310120105,
|
|
"num_tokens": 166473.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.969349575042724,
|
|
"epoch": 0.0032162269369726726,
|
|
"grad_norm": 1.4296875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1308,
|
|
"mean_token_accuracy": 0.07532015293836594,
|
|
"num_tokens": 171486.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.817659473419189,
|
|
"epoch": 0.0033234345015384285,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.0433,
|
|
"mean_token_accuracy": 0.0738318596035242,
|
|
"num_tokens": 176716.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.672884654998779,
|
|
"epoch": 0.0034306420661041844,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.0118,
|
|
"mean_token_accuracy": 0.07759347334504127,
|
|
"num_tokens": 181636.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.527563190460205,
|
|
"epoch": 0.00353784963066994,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9619,
|
|
"mean_token_accuracy": 0.06629271619021893,
|
|
"num_tokens": 187540.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.340129470825195,
|
|
"epoch": 0.003645057195235696,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.9015,
|
|
"mean_token_accuracy": 0.07701855972409248,
|
|
"num_tokens": 192770.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.15212345123291,
|
|
"epoch": 0.0037522647598014517,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8758,
|
|
"mean_token_accuracy": 0.0689329519867897,
|
|
"num_tokens": 198622.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.033543682098388,
|
|
"epoch": 0.003859472324367207,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7325,
|
|
"mean_token_accuracy": 0.07789348289370537,
|
|
"num_tokens": 204390.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.808248138427734,
|
|
"epoch": 0.003966679888932963,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7251,
|
|
"mean_token_accuracy": 0.07829341143369675,
|
|
"num_tokens": 210648.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.655777645111083,
|
|
"epoch": 0.004073887453498719,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5923,
|
|
"mean_token_accuracy": 0.08251099660992622,
|
|
"num_tokens": 216446.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.500771617889404,
|
|
"epoch": 0.0041810950180644745,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.7632,
|
|
"mean_token_accuracy": 0.0752104040235281,
|
|
"num_tokens": 222508.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.553161811828613,
|
|
"epoch": 0.00428830258263023,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.6422,
|
|
"mean_token_accuracy": 0.07912560775876046,
|
|
"num_tokens": 228092.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.395637798309327,
|
|
"epoch": 0.004395510147195986,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.08580457344651223,
|
|
"num_tokens": 233701.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.33106803894043,
|
|
"epoch": 0.004502717711761742,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.3889,
|
|
"mean_token_accuracy": 0.08813347965478897,
|
|
"num_tokens": 238657.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.217943000793458,
|
|
"epoch": 0.004609925276327497,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5818,
|
|
"mean_token_accuracy": 0.07985254153609275,
|
|
"num_tokens": 244179.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.332947063446046,
|
|
"epoch": 0.004717132840893253,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.6459,
|
|
"mean_token_accuracy": 0.08311917260289192,
|
|
"num_tokens": 250523.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.235616397857665,
|
|
"epoch": 0.004824340405459009,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.7159,
|
|
"mean_token_accuracy": 0.07638855129480362,
|
|
"num_tokens": 257607.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.253262996673584,
|
|
"epoch": 0.004931547970024765,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.4315,
|
|
"mean_token_accuracy": 0.0921292670071125,
|
|
"num_tokens": 263443.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.13878574371338,
|
|
"epoch": 0.005038755534590521,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.5551,
|
|
"mean_token_accuracy": 0.08070281371474267,
|
|
"num_tokens": 269558.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.104772663116455,
|
|
"epoch": 0.005145963099156277,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.4144,
|
|
"mean_token_accuracy": 0.09172740653157234,
|
|
"num_tokens": 274301.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.141210412979126,
|
|
"epoch": 0.005253170663722032,
|
|
"grad_norm": 1.46875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4997,
|
|
"mean_token_accuracy": 0.0893177330493927,
|
|
"num_tokens": 279398.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.01241855621338,
|
|
"epoch": 0.005360378228287788,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.4531,
|
|
"mean_token_accuracy": 0.09177742525935173,
|
|
"num_tokens": 284825.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.075734186172486,
|
|
"epoch": 0.005467585792853544,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.4555,
|
|
"mean_token_accuracy": 0.09476318880915642,
|
|
"num_tokens": 290916.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.005568408966065,
|
|
"epoch": 0.0055747933574193,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.5944,
|
|
"mean_token_accuracy": 0.08753971830010414,
|
|
"num_tokens": 297664.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.058106231689454,
|
|
"epoch": 0.0056820009219850556,
|
|
"grad_norm": 1.34375,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4517,
|
|
"mean_token_accuracy": 0.09504219591617584,
|
|
"num_tokens": 303353.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.011071968078614,
|
|
"epoch": 0.005789208486550811,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.4083,
|
|
"mean_token_accuracy": 0.09080644249916077,
|
|
"num_tokens": 309623.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.051470184326172,
|
|
"epoch": 0.0058964160511165665,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.3651,
|
|
"mean_token_accuracy": 0.09395760223269463,
|
|
"num_tokens": 315174.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 7.95319299697876,
|
|
"epoch": 0.006003623615682322,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.4677,
|
|
"mean_token_accuracy": 0.09074310585856438,
|
|
"num_tokens": 321338.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.950230884552002,
|
|
"epoch": 0.006110831180248078,
|
|
"grad_norm": 1.4453125,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.4856,
|
|
"mean_token_accuracy": 0.09278372302651405,
|
|
"num_tokens": 327649.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.951276445388794,
|
|
"epoch": 0.006218038744813834,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.342,
|
|
"mean_token_accuracy": 0.0961864523589611,
|
|
"num_tokens": 333289.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.04638614654541,
|
|
"epoch": 0.00632524630937959,
|
|
"grad_norm": 1.7265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.2975,
|
|
"mean_token_accuracy": 0.0981252558529377,
|
|
"num_tokens": 338467.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.920049381256104,
|
|
"epoch": 0.006432453873945345,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.413,
|
|
"mean_token_accuracy": 0.09103409945964813,
|
|
"num_tokens": 344313.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.964056444168091,
|
|
"epoch": 0.006539661438511101,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.3432,
|
|
"mean_token_accuracy": 0.09449874684214592,
|
|
"num_tokens": 351050.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.840343713760376,
|
|
"epoch": 0.006646869003076857,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2593,
|
|
"mean_token_accuracy": 0.10226496309041977,
|
|
"num_tokens": 356755.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.942127895355225,
|
|
"epoch": 0.006754076567642613,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3612,
|
|
"mean_token_accuracy": 0.10015994533896447,
|
|
"num_tokens": 362791.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.872417974472046,
|
|
"epoch": 0.006861284132208369,
|
|
"grad_norm": 1.4921875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.3319,
|
|
"mean_token_accuracy": 0.09404960200190544,
|
|
"num_tokens": 367919.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.857217264175415,
|
|
"epoch": 0.006968491696774125,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2665,
|
|
"mean_token_accuracy": 0.10184875056147576,
|
|
"num_tokens": 372896.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.820843887329102,
|
|
"epoch": 0.00707569926133988,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.1437,
|
|
"mean_token_accuracy": 0.11197337061166764,
|
|
"num_tokens": 377806.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.786612844467163,
|
|
"epoch": 0.007182906825905636,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.2726,
|
|
"mean_token_accuracy": 0.0974352166056633,
|
|
"num_tokens": 383377.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.863493776321411,
|
|
"epoch": 0.007290114390471392,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2915,
|
|
"mean_token_accuracy": 0.10096050538122654,
|
|
"num_tokens": 388851.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.8964540481567385,
|
|
"epoch": 0.0073973219550371476,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.3203,
|
|
"mean_token_accuracy": 0.09899901077151299,
|
|
"num_tokens": 394797.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.78081636428833,
|
|
"epoch": 0.0075045295196029035,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2391,
|
|
"mean_token_accuracy": 0.09974800124764442,
|
|
"num_tokens": 400189.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.784978008270263,
|
|
"epoch": 0.0076117370841686585,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3479,
|
|
"mean_token_accuracy": 0.09474938213825226,
|
|
"num_tokens": 405996.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.787995433807373,
|
|
"epoch": 0.007718944648734414,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.0812,
|
|
"mean_token_accuracy": 0.10426637381315232,
|
|
"num_tokens": 411014.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.673157167434693,
|
|
"epoch": 0.00782615221330017,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1807,
|
|
"mean_token_accuracy": 0.0934258908033371,
|
|
"num_tokens": 416452.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.71818356513977,
|
|
"epoch": 0.007933359777865925,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.1482,
|
|
"mean_token_accuracy": 0.10469824001193047,
|
|
"num_tokens": 422693.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.701494979858398,
|
|
"epoch": 0.008040567342431682,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.2076,
|
|
"mean_token_accuracy": 0.1031791903078556,
|
|
"num_tokens": 427648.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.773630094528198,
|
|
"epoch": 0.008147774906997437,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.273,
|
|
"mean_token_accuracy": 0.09940337911248207,
|
|
"num_tokens": 434504.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.668553256988526,
|
|
"epoch": 0.008254982471563194,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.2049,
|
|
"mean_token_accuracy": 0.09594230279326439,
|
|
"num_tokens": 440054.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.741257047653198,
|
|
"epoch": 0.008362190036128949,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.2417,
|
|
"mean_token_accuracy": 0.09266546368598938,
|
|
"num_tokens": 446005.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.707240200042724,
|
|
"epoch": 0.008469397600694706,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.182,
|
|
"mean_token_accuracy": 0.10206115245819092,
|
|
"num_tokens": 452249.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.619378709793091,
|
|
"epoch": 0.00857660516526046,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.1498,
|
|
"mean_token_accuracy": 0.11172568276524544,
|
|
"num_tokens": 458672.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.674522972106933,
|
|
"epoch": 0.008683812729826216,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.2479,
|
|
"mean_token_accuracy": 0.1010979875922203,
|
|
"num_tokens": 464136.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.748438692092895,
|
|
"epoch": 0.008791020294391973,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.2122,
|
|
"mean_token_accuracy": 0.10695564225316048,
|
|
"num_tokens": 469757.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.67966136932373,
|
|
"epoch": 0.008898227858957728,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.142,
|
|
"mean_token_accuracy": 0.09978308379650117,
|
|
"num_tokens": 475482.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.6171276569366455,
|
|
"epoch": 0.009005435423523485,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.0457,
|
|
"mean_token_accuracy": 0.09999699369072915,
|
|
"num_tokens": 481422.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.684847450256347,
|
|
"epoch": 0.00911264298808924,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.1056,
|
|
"mean_token_accuracy": 0.10502869784832,
|
|
"num_tokens": 486422.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.600341415405273,
|
|
"epoch": 0.009219850552654995,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.1511,
|
|
"mean_token_accuracy": 0.097078138589859,
|
|
"num_tokens": 492052.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.724478054046631,
|
|
"epoch": 0.009327058117220751,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.1779,
|
|
"mean_token_accuracy": 0.09461221843957901,
|
|
"num_tokens": 497200.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.574504089355469,
|
|
"epoch": 0.009434265681786506,
|
|
"grad_norm": 1.3828125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.1753,
|
|
"mean_token_accuracy": 0.09646964296698571,
|
|
"num_tokens": 502620.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.732940196990967,
|
|
"epoch": 0.009541473246352263,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0647,
|
|
"mean_token_accuracy": 0.10566939935088157,
|
|
"num_tokens": 508673.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.571577835083008,
|
|
"epoch": 0.009648680810918018,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.1172,
|
|
"mean_token_accuracy": 0.10332905054092408,
|
|
"num_tokens": 514607.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.5489884376525875,
|
|
"epoch": 0.009755888375483773,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1553,
|
|
"mean_token_accuracy": 0.10310770496726036,
|
|
"num_tokens": 520715.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.7070821762084964,
|
|
"epoch": 0.00986309594004953,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.1423,
|
|
"mean_token_accuracy": 0.10408871993422508,
|
|
"num_tokens": 526548.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.580347776412964,
|
|
"epoch": 0.009970303504615285,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.0557,
|
|
"mean_token_accuracy": 0.1075308345258236,
|
|
"num_tokens": 531968.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.551975536346435,
|
|
"epoch": 0.010077511069181042,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.107,
|
|
"mean_token_accuracy": 0.10239257588982582,
|
|
"num_tokens": 538009.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.61077241897583,
|
|
"epoch": 0.010184718633746797,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.2189,
|
|
"mean_token_accuracy": 0.09845631942152977,
|
|
"num_tokens": 544178.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.511112976074219,
|
|
"epoch": 0.010291926198312554,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.1712,
|
|
"mean_token_accuracy": 0.09861031472682953,
|
|
"num_tokens": 551208.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.547733783721924,
|
|
"epoch": 0.010399133762878309,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.0396,
|
|
"mean_token_accuracy": 0.10685237124562263,
|
|
"num_tokens": 557031.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.55196213722229,
|
|
"epoch": 0.010506341327444064,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.0893,
|
|
"mean_token_accuracy": 0.11042937040328979,
|
|
"num_tokens": 562666.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.52198429107666,
|
|
"epoch": 0.01061354889200982,
|
|
"grad_norm": 1.421875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.0766,
|
|
"mean_token_accuracy": 0.10476985275745392,
|
|
"num_tokens": 567754.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.596458625793457,
|
|
"epoch": 0.010720756456575576,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.1258,
|
|
"mean_token_accuracy": 0.10542995110154152,
|
|
"num_tokens": 573832.0,
|
|
"step": 500
|
|
}
|
|
],
|
|
"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": 180983125278720.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|