3068 lines
82 KiB
JSON
3068 lines
82 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 1.1755485893416928,
|
|
"eval_steps": 500,
|
|
"global_step": 1500,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 10.74259386062622,
|
|
"epoch": 0.003918495297805642,
|
|
"grad_norm": 5.34375,
|
|
"learning_rate": 2e-06,
|
|
"loss": 10.7559,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 9174.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 10.742582988739013,
|
|
"epoch": 0.007836990595611285,
|
|
"grad_norm": 5.09375,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 10.7343,
|
|
"mean_token_accuracy": 0.00011074197245761753,
|
|
"num_tokens": 17790.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 10.742611026763916,
|
|
"epoch": 0.011755485893416929,
|
|
"grad_norm": 5.03125,
|
|
"learning_rate": 7e-06,
|
|
"loss": 10.6946,
|
|
"mean_token_accuracy": 0.0,
|
|
"num_tokens": 25850.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 10.742605590820313,
|
|
"epoch": 0.01567398119122257,
|
|
"grad_norm": 5.1875,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 10.6748,
|
|
"mean_token_accuracy": 0.001098086079582572,
|
|
"num_tokens": 35194.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 10.742543792724609,
|
|
"epoch": 0.019592476489028215,
|
|
"grad_norm": 4.53125,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 10.5631,
|
|
"mean_token_accuracy": 0.008560778619721533,
|
|
"num_tokens": 44218.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 10.74220838546753,
|
|
"epoch": 0.023510971786833857,
|
|
"grad_norm": 4.09375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 10.4525,
|
|
"mean_token_accuracy": 0.03252750001847744,
|
|
"num_tokens": 53397.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.740808391571045,
|
|
"epoch": 0.0274294670846395,
|
|
"grad_norm": 3.421875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.3079,
|
|
"mean_token_accuracy": 0.0437689334154129,
|
|
"num_tokens": 62749.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.736591911315918,
|
|
"epoch": 0.03134796238244514,
|
|
"grad_norm": 2.609375,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.1743,
|
|
"mean_token_accuracy": 0.043372325226664545,
|
|
"num_tokens": 71721.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.729655265808105,
|
|
"epoch": 0.03526645768025078,
|
|
"grad_norm": 2.4375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.0405,
|
|
"mean_token_accuracy": 0.046825164929032326,
|
|
"num_tokens": 79844.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.72356595993042,
|
|
"epoch": 0.03918495297805643,
|
|
"grad_norm": 2.09375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 9.9917,
|
|
"mean_token_accuracy": 0.04553030654788017,
|
|
"num_tokens": 88866.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.719326305389405,
|
|
"epoch": 0.04310344827586207,
|
|
"grad_norm": 1.984375,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 9.9336,
|
|
"mean_token_accuracy": 0.044035762920975684,
|
|
"num_tokens": 97918.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.715408420562744,
|
|
"epoch": 0.047021943573667714,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 9.8702,
|
|
"mean_token_accuracy": 0.04544390179216862,
|
|
"num_tokens": 107043.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.710628414154053,
|
|
"epoch": 0.050940438871473356,
|
|
"grad_norm": 2.078125,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 9.8037,
|
|
"mean_token_accuracy": 0.04611521698534489,
|
|
"num_tokens": 116000.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.704075908660888,
|
|
"epoch": 0.054858934169279,
|
|
"grad_norm": 1.9453125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.7686,
|
|
"mean_token_accuracy": 0.04174055978655815,
|
|
"num_tokens": 124559.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.700677013397216,
|
|
"epoch": 0.05877742946708464,
|
|
"grad_norm": 1.8984375,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.6568,
|
|
"mean_token_accuracy": 0.04639651067554951,
|
|
"num_tokens": 132868.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.695802783966064,
|
|
"epoch": 0.06269592476489028,
|
|
"grad_norm": 1.8671875,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.6523,
|
|
"mean_token_accuracy": 0.04210274666547775,
|
|
"num_tokens": 141286.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.688793563842774,
|
|
"epoch": 0.06661442006269593,
|
|
"grad_norm": 1.828125,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.5499,
|
|
"mean_token_accuracy": 0.05031363070011139,
|
|
"num_tokens": 150406.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.679227447509765,
|
|
"epoch": 0.07053291536050156,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.4859,
|
|
"mean_token_accuracy": 0.056176548823714256,
|
|
"num_tokens": 158770.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.663717269897461,
|
|
"epoch": 0.07445141065830721,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.4023,
|
|
"mean_token_accuracy": 0.06409536376595497,
|
|
"num_tokens": 167763.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.639647197723388,
|
|
"epoch": 0.07836990595611286,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.286,
|
|
"mean_token_accuracy": 0.06052683852612972,
|
|
"num_tokens": 177306.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.605971527099609,
|
|
"epoch": 0.0822884012539185,
|
|
"grad_norm": 1.6953125,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.1915,
|
|
"mean_token_accuracy": 0.06200857125222683,
|
|
"num_tokens": 186014.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.545482730865478,
|
|
"epoch": 0.08620689655172414,
|
|
"grad_norm": 1.6875,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.0547,
|
|
"mean_token_accuracy": 0.06552336104214192,
|
|
"num_tokens": 194122.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.471019172668457,
|
|
"epoch": 0.09012539184952978,
|
|
"grad_norm": 1.703125,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 8.9639,
|
|
"mean_token_accuracy": 0.061758537963032725,
|
|
"num_tokens": 203097.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.420811939239503,
|
|
"epoch": 0.09404388714733543,
|
|
"grad_norm": 1.59375,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 8.8529,
|
|
"mean_token_accuracy": 0.06423232220113277,
|
|
"num_tokens": 211413.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.383054542541505,
|
|
"epoch": 0.09796238244514106,
|
|
"grad_norm": 1.484375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 8.7779,
|
|
"mean_token_accuracy": 0.06270350776612758,
|
|
"num_tokens": 220550.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.31447114944458,
|
|
"epoch": 0.10188087774294671,
|
|
"grad_norm": 1.5,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 8.6079,
|
|
"mean_token_accuracy": 0.06446850374341011,
|
|
"num_tokens": 229197.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.19693603515625,
|
|
"epoch": 0.10579937304075235,
|
|
"grad_norm": 1.328125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.5715,
|
|
"mean_token_accuracy": 0.05803979113698006,
|
|
"num_tokens": 238479.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.129131984710693,
|
|
"epoch": 0.109717868338558,
|
|
"grad_norm": 1.4375,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.4073,
|
|
"mean_token_accuracy": 0.06696602776646614,
|
|
"num_tokens": 246318.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.014921092987061,
|
|
"epoch": 0.11363636363636363,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.3497,
|
|
"mean_token_accuracy": 0.06019376777112484,
|
|
"num_tokens": 254783.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.837921333312988,
|
|
"epoch": 0.11755485893416928,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.1727,
|
|
"mean_token_accuracy": 0.06220055930316448,
|
|
"num_tokens": 263416.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.704282855987548,
|
|
"epoch": 0.12147335423197492,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.1265,
|
|
"mean_token_accuracy": 0.06537272743880748,
|
|
"num_tokens": 271930.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.534114360809326,
|
|
"epoch": 0.12539184952978055,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 7.9819,
|
|
"mean_token_accuracy": 0.06310658231377601,
|
|
"num_tokens": 280546.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.289366626739502,
|
|
"epoch": 0.12931034482758622,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 7.9162,
|
|
"mean_token_accuracy": 0.0649850957095623,
|
|
"num_tokens": 288813.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.108893775939942,
|
|
"epoch": 0.13322884012539185,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 7.8379,
|
|
"mean_token_accuracy": 0.06351686343550682,
|
|
"num_tokens": 297966.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 8.892046165466308,
|
|
"epoch": 0.1371473354231975,
|
|
"grad_norm": 0.859375,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 7.8598,
|
|
"mean_token_accuracy": 0.06292634233832359,
|
|
"num_tokens": 307135.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 8.728424072265625,
|
|
"epoch": 0.14106583072100312,
|
|
"grad_norm": 0.78515625,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 7.7429,
|
|
"mean_token_accuracy": 0.06494242362678052,
|
|
"num_tokens": 315546.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 8.619221019744874,
|
|
"epoch": 0.14498432601880878,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 7.7168,
|
|
"mean_token_accuracy": 0.06593996658921242,
|
|
"num_tokens": 323930.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.409439945220948,
|
|
"epoch": 0.14890282131661442,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 7.5689,
|
|
"mean_token_accuracy": 0.0733262188732624,
|
|
"num_tokens": 332419.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.326526832580566,
|
|
"epoch": 0.15282131661442006,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 7.654,
|
|
"mean_token_accuracy": 0.0696730900555849,
|
|
"num_tokens": 341362.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.3056715965271,
|
|
"epoch": 0.15673981191222572,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 7.5645,
|
|
"mean_token_accuracy": 0.07249549068510533,
|
|
"num_tokens": 349395.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.24736156463623,
|
|
"epoch": 0.16065830721003135,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000102,
|
|
"loss": 7.5705,
|
|
"mean_token_accuracy": 0.06626113764941692,
|
|
"num_tokens": 358413.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.157195377349854,
|
|
"epoch": 0.164576802507837,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 7.6532,
|
|
"mean_token_accuracy": 0.07263443917036057,
|
|
"num_tokens": 366489.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.1927809715271,
|
|
"epoch": 0.16849529780564262,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 0.000107,
|
|
"loss": 7.5231,
|
|
"mean_token_accuracy": 0.06818853206932544,
|
|
"num_tokens": 375335.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.176179027557373,
|
|
"epoch": 0.1724137931034483,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 7.5638,
|
|
"mean_token_accuracy": 0.07268193662166596,
|
|
"num_tokens": 384276.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.140285873413086,
|
|
"epoch": 0.17633228840125392,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 7.5991,
|
|
"mean_token_accuracy": 0.06890415549278259,
|
|
"num_tokens": 393571.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.1433687210083,
|
|
"epoch": 0.18025078369905956,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 7.5228,
|
|
"mean_token_accuracy": 0.07579943463206291,
|
|
"num_tokens": 401865.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.072039985656739,
|
|
"epoch": 0.1841692789968652,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 7.4982,
|
|
"mean_token_accuracy": 0.07117367573082448,
|
|
"num_tokens": 410518.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.076621770858765,
|
|
"epoch": 0.18808777429467086,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 7.5051,
|
|
"mean_token_accuracy": 0.06969159096479416,
|
|
"num_tokens": 419769.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.030782747268677,
|
|
"epoch": 0.1920062695924765,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.000122,
|
|
"loss": 7.4391,
|
|
"mean_token_accuracy": 0.07655345052480697,
|
|
"num_tokens": 428204.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.057575464248657,
|
|
"epoch": 0.19592476489028213,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 7.5263,
|
|
"mean_token_accuracy": 0.07384133897721767,
|
|
"num_tokens": 436594.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.054854202270509,
|
|
"epoch": 0.19984326018808776,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000127,
|
|
"loss": 7.3139,
|
|
"mean_token_accuracy": 0.07670355625450612,
|
|
"num_tokens": 444645.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 7.953717613220215,
|
|
"epoch": 0.20376175548589343,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 7.3561,
|
|
"mean_token_accuracy": 0.07812742255628109,
|
|
"num_tokens": 453016.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 7.9555809020996096,
|
|
"epoch": 0.20768025078369906,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 7.4192,
|
|
"mean_token_accuracy": 0.07524299174547196,
|
|
"num_tokens": 462116.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 7.957012748718261,
|
|
"epoch": 0.2115987460815047,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 7.3694,
|
|
"mean_token_accuracy": 0.07774586156010628,
|
|
"num_tokens": 470807.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 7.9492302417755125,
|
|
"epoch": 0.21551724137931033,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 7.4561,
|
|
"mean_token_accuracy": 0.07432891242206097,
|
|
"num_tokens": 479592.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.016344738006591,
|
|
"epoch": 0.219435736677116,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 7.3803,
|
|
"mean_token_accuracy": 0.07607875801622868,
|
|
"num_tokens": 488107.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 7.870184087753296,
|
|
"epoch": 0.22335423197492163,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 7.3792,
|
|
"mean_token_accuracy": 0.07687489986419678,
|
|
"num_tokens": 496775.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 7.973387956619263,
|
|
"epoch": 0.22727272727272727,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 7.4096,
|
|
"mean_token_accuracy": 0.07644881159067154,
|
|
"num_tokens": 505415.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 7.946991682052612,
|
|
"epoch": 0.23119122257053293,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000147,
|
|
"loss": 7.3915,
|
|
"mean_token_accuracy": 0.07934824042022229,
|
|
"num_tokens": 514447.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 7.891976928710937,
|
|
"epoch": 0.23510971786833856,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 7.381,
|
|
"mean_token_accuracy": 0.07889064475893974,
|
|
"num_tokens": 522998.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 7.92128734588623,
|
|
"epoch": 0.2390282131661442,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 7.3459,
|
|
"mean_token_accuracy": 0.08099666759371757,
|
|
"num_tokens": 531542.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 7.863611841201783,
|
|
"epoch": 0.24294670846394983,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 7.2927,
|
|
"mean_token_accuracy": 0.07772218361496926,
|
|
"num_tokens": 540143.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 7.924414491653442,
|
|
"epoch": 0.2468652037617555,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000157,
|
|
"loss": 7.3679,
|
|
"mean_token_accuracy": 0.08110597431659698,
|
|
"num_tokens": 549156.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 7.803961706161499,
|
|
"epoch": 0.2507836990595611,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 7.2252,
|
|
"mean_token_accuracy": 0.08031321838498115,
|
|
"num_tokens": 557522.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 7.805857229232788,
|
|
"epoch": 0.2547021943573668,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.000162,
|
|
"loss": 7.2879,
|
|
"mean_token_accuracy": 0.08254150673747063,
|
|
"num_tokens": 566650.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 7.8912803649902346,
|
|
"epoch": 0.25862068965517243,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 7.2408,
|
|
"mean_token_accuracy": 0.08324127942323685,
|
|
"num_tokens": 576219.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 7.849798917770386,
|
|
"epoch": 0.26253918495297807,
|
|
"grad_norm": 1.4140625,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 7.3416,
|
|
"mean_token_accuracy": 0.08195492923259735,
|
|
"num_tokens": 584304.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 7.75479941368103,
|
|
"epoch": 0.2664576802507837,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 7.2574,
|
|
"mean_token_accuracy": 0.08678329288959503,
|
|
"num_tokens": 593163.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 7.7592004299163815,
|
|
"epoch": 0.27037617554858934,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 7.1773,
|
|
"mean_token_accuracy": 0.08335737958550453,
|
|
"num_tokens": 602077.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 7.792924070358277,
|
|
"epoch": 0.274294670846395,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 7.2812,
|
|
"mean_token_accuracy": 0.08297529369592667,
|
|
"num_tokens": 610009.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 7.816137742996216,
|
|
"epoch": 0.2782131661442006,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000177,
|
|
"loss": 7.3247,
|
|
"mean_token_accuracy": 0.07837995067238808,
|
|
"num_tokens": 619282.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 7.692404460906983,
|
|
"epoch": 0.28213166144200624,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 7.2457,
|
|
"mean_token_accuracy": 0.08888485580682755,
|
|
"num_tokens": 628138.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 7.819083976745605,
|
|
"epoch": 0.28605015673981193,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000182,
|
|
"loss": 7.1219,
|
|
"mean_token_accuracy": 0.09141435995697975,
|
|
"num_tokens": 637021.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 7.734615421295166,
|
|
"epoch": 0.28996865203761757,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 7.2772,
|
|
"mean_token_accuracy": 0.08298610299825668,
|
|
"num_tokens": 646703.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 7.772846841812134,
|
|
"epoch": 0.2938871473354232,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000187,
|
|
"loss": 7.1535,
|
|
"mean_token_accuracy": 0.08612042218446732,
|
|
"num_tokens": 655616.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 7.662965631484985,
|
|
"epoch": 0.29780564263322884,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 7.1548,
|
|
"mean_token_accuracy": 0.09385274946689606,
|
|
"num_tokens": 663783.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 7.72055869102478,
|
|
"epoch": 0.3017241379310345,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000192,
|
|
"loss": 7.1499,
|
|
"mean_token_accuracy": 0.09183024540543556,
|
|
"num_tokens": 671855.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 7.686996603012085,
|
|
"epoch": 0.3056426332288401,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 7.1691,
|
|
"mean_token_accuracy": 0.08594455868005753,
|
|
"num_tokens": 680981.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 7.710857486724853,
|
|
"epoch": 0.30956112852664575,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 7.1032,
|
|
"mean_token_accuracy": 0.09159721955657005,
|
|
"num_tokens": 689294.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 7.561884641647339,
|
|
"epoch": 0.31347962382445144,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 7.055,
|
|
"mean_token_accuracy": 0.09060370847582817,
|
|
"num_tokens": 697703.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 7.728968143463135,
|
|
"epoch": 0.31739811912225707,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.000202,
|
|
"loss": 7.1907,
|
|
"mean_token_accuracy": 0.08998276144266129,
|
|
"num_tokens": 706792.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 7.705010414123535,
|
|
"epoch": 0.3213166144200627,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 7.0954,
|
|
"mean_token_accuracy": 0.09226684123277665,
|
|
"num_tokens": 715864.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 7.541502141952515,
|
|
"epoch": 0.32523510971786834,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 7.1394,
|
|
"mean_token_accuracy": 0.0916991800069809,
|
|
"num_tokens": 723921.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 7.679799509048462,
|
|
"epoch": 0.329153605015674,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.1503,
|
|
"mean_token_accuracy": 0.08857458978891372,
|
|
"num_tokens": 732797.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 7.634342527389526,
|
|
"epoch": 0.3330721003134796,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000212,
|
|
"loss": 7.0126,
|
|
"mean_token_accuracy": 0.0977866668254137,
|
|
"num_tokens": 741248.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 7.6327766418457035,
|
|
"epoch": 0.33699059561128525,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 7.0961,
|
|
"mean_token_accuracy": 0.09004503935575485,
|
|
"num_tokens": 749766.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 7.5497727394104,
|
|
"epoch": 0.3409090909090909,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 7.067,
|
|
"mean_token_accuracy": 0.09081325829029083,
|
|
"num_tokens": 758695.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 7.6185346126556395,
|
|
"epoch": 0.3448275862068966,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 7.0591,
|
|
"mean_token_accuracy": 0.08728412315249442,
|
|
"num_tokens": 767624.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 7.5867267608642575,
|
|
"epoch": 0.3487460815047022,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.000222,
|
|
"loss": 7.0721,
|
|
"mean_token_accuracy": 0.09007409512996674,
|
|
"num_tokens": 776616.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 7.504705095291138,
|
|
"epoch": 0.35266457680250785,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.0826,
|
|
"mean_token_accuracy": 0.09070678725838661,
|
|
"num_tokens": 786147.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 7.628587007522583,
|
|
"epoch": 0.3565830721003135,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.1945,
|
|
"mean_token_accuracy": 0.09022908806800842,
|
|
"num_tokens": 795213.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 7.5562244892120365,
|
|
"epoch": 0.3605015673981191,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 7.0011,
|
|
"mean_token_accuracy": 0.09484408944845199,
|
|
"num_tokens": 803118.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 7.582311534881592,
|
|
"epoch": 0.36442006269592475,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 6.9727,
|
|
"mean_token_accuracy": 0.10151090547442436,
|
|
"num_tokens": 811358.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 7.470248889923096,
|
|
"epoch": 0.3683385579937304,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 6.971,
|
|
"mean_token_accuracy": 0.09536491632461548,
|
|
"num_tokens": 819653.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 7.467239570617676,
|
|
"epoch": 0.3722570532915361,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000237,
|
|
"loss": 6.9984,
|
|
"mean_token_accuracy": 0.0897679604589939,
|
|
"num_tokens": 828462.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 7.6044769287109375,
|
|
"epoch": 0.3761755485893417,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.0876,
|
|
"mean_token_accuracy": 0.09487015083432197,
|
|
"num_tokens": 836204.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 7.44920597076416,
|
|
"epoch": 0.38009404388714735,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 6.9686,
|
|
"mean_token_accuracy": 0.09797411188483238,
|
|
"num_tokens": 845032.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 7.5414040088653564,
|
|
"epoch": 0.384012539184953,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 6.9796,
|
|
"mean_token_accuracy": 0.09801916852593422,
|
|
"num_tokens": 853324.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 7.508725357055664,
|
|
"epoch": 0.3879310344827586,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000247,
|
|
"loss": 6.9449,
|
|
"mean_token_accuracy": 0.10975441187620164,
|
|
"num_tokens": 861640.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 7.553517627716064,
|
|
"epoch": 0.39184952978056425,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.116,
|
|
"mean_token_accuracy": 0.09244248941540718,
|
|
"num_tokens": 870758.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.39184952978056425,
|
|
"eval_entropy": 7.418469055678493,
|
|
"eval_loss": 7.0935492515563965,
|
|
"eval_mean_token_accuracy": 0.09673699742370798,
|
|
"eval_num_tokens": 870758.0,
|
|
"eval_runtime": 2.2092,
|
|
"eval_samples_per_second": 1865.802,
|
|
"eval_steps_per_second": 233.565,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 7.561475372314453,
|
|
"epoch": 0.3957680250783699,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.0205,
|
|
"mean_token_accuracy": 0.0910447046160698,
|
|
"num_tokens": 879968.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 7.430616950988769,
|
|
"epoch": 0.3996865203761755,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 6.9955,
|
|
"mean_token_accuracy": 0.09514781311154366,
|
|
"num_tokens": 888804.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 7.443049621582031,
|
|
"epoch": 0.4036050156739812,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 6.9991,
|
|
"mean_token_accuracy": 0.09540855661034583,
|
|
"num_tokens": 898593.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 7.458410596847534,
|
|
"epoch": 0.40752351097178685,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 6.9981,
|
|
"mean_token_accuracy": 0.093653604388237,
|
|
"num_tokens": 906873.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 7.480202293395996,
|
|
"epoch": 0.4114420062695925,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000262,
|
|
"loss": 6.9857,
|
|
"mean_token_accuracy": 0.10073858648538589,
|
|
"num_tokens": 915080.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 7.492116975784302,
|
|
"epoch": 0.4153605015673981,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 6.9989,
|
|
"mean_token_accuracy": 0.09387330785393715,
|
|
"num_tokens": 923741.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 7.394177532196045,
|
|
"epoch": 0.41927899686520376,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 6.8859,
|
|
"mean_token_accuracy": 0.09964613541960717,
|
|
"num_tokens": 932455.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 7.348754358291626,
|
|
"epoch": 0.4231974921630094,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 6.8926,
|
|
"mean_token_accuracy": 0.09797894805669785,
|
|
"num_tokens": 941064.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 7.362092924118042,
|
|
"epoch": 0.427115987460815,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.09599274471402168,
|
|
"num_tokens": 950261.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 7.476526355743408,
|
|
"epoch": 0.43103448275862066,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 6.9869,
|
|
"mean_token_accuracy": 0.09676681011915207,
|
|
"num_tokens": 959151.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 7.451156759262085,
|
|
"epoch": 0.43495297805642635,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.0111,
|
|
"mean_token_accuracy": 0.09589274004101753,
|
|
"num_tokens": 968441.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 7.319547843933106,
|
|
"epoch": 0.438871473354232,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 6.8728,
|
|
"mean_token_accuracy": 0.09684987366199493,
|
|
"num_tokens": 977058.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 7.395937252044678,
|
|
"epoch": 0.4427899686520376,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 6.8911,
|
|
"mean_token_accuracy": 0.09753472357988358,
|
|
"num_tokens": 986531.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 7.321355772018433,
|
|
"epoch": 0.44670846394984326,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 6.7391,
|
|
"mean_token_accuracy": 0.10372037217020988,
|
|
"num_tokens": 995081.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 7.329747819900513,
|
|
"epoch": 0.4506269592476489,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 6.8174,
|
|
"mean_token_accuracy": 0.09918634667992592,
|
|
"num_tokens": 1003459.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 7.2806806564331055,
|
|
"epoch": 0.45454545454545453,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 6.8646,
|
|
"mean_token_accuracy": 0.09958613365888595,
|
|
"num_tokens": 1012420.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.450819778442383,
|
|
"epoch": 0.45846394984326017,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000292,
|
|
"loss": 6.9327,
|
|
"mean_token_accuracy": 0.09935399070382119,
|
|
"num_tokens": 1021198.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 7.32666244506836,
|
|
"epoch": 0.46238244514106586,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 6.9064,
|
|
"mean_token_accuracy": 0.10413050577044487,
|
|
"num_tokens": 1030023.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 7.313800811767578,
|
|
"epoch": 0.4663009404388715,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000297,
|
|
"loss": 6.7756,
|
|
"mean_token_accuracy": 0.10411769896745682,
|
|
"num_tokens": 1038824.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 7.428035593032837,
|
|
"epoch": 0.4702194357366771,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.0354,
|
|
"mean_token_accuracy": 0.10027154311537742,
|
|
"num_tokens": 1048207.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 7.444586324691772,
|
|
"epoch": 0.47413793103448276,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.003,
|
|
"mean_token_accuracy": 0.0982631802558899,
|
|
"num_tokens": 1057354.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 7.326791095733642,
|
|
"epoch": 0.4780564263322884,
|
|
"grad_norm": 1.28125,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 6.9402,
|
|
"mean_token_accuracy": 0.09771996289491654,
|
|
"num_tokens": 1066825.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 7.27696385383606,
|
|
"epoch": 0.48197492163009403,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000307,
|
|
"loss": 6.8595,
|
|
"mean_token_accuracy": 0.09698087573051453,
|
|
"num_tokens": 1075752.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 7.259910535812378,
|
|
"epoch": 0.48589341692789967,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 6.7766,
|
|
"mean_token_accuracy": 0.10016454383730888,
|
|
"num_tokens": 1084826.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 7.3767705917358395,
|
|
"epoch": 0.4898119122257053,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000312,
|
|
"loss": 6.9169,
|
|
"mean_token_accuracy": 0.09827893227338791,
|
|
"num_tokens": 1093425.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 7.34305853843689,
|
|
"epoch": 0.493730407523511,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 6.9157,
|
|
"mean_token_accuracy": 0.09747867435216903,
|
|
"num_tokens": 1102619.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 7.221006727218628,
|
|
"epoch": 0.49764890282131663,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.000317,
|
|
"loss": 6.8191,
|
|
"mean_token_accuracy": 0.10153490155935288,
|
|
"num_tokens": 1111337.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.207894706726075,
|
|
"epoch": 0.5015673981191222,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 6.655,
|
|
"mean_token_accuracy": 0.10878187417984009,
|
|
"num_tokens": 1119381.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.210852241516113,
|
|
"epoch": 0.5054858934169278,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000322,
|
|
"loss": 6.7069,
|
|
"mean_token_accuracy": 0.10850767269730568,
|
|
"num_tokens": 1127825.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 7.236933660507202,
|
|
"epoch": 0.5094043887147336,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 6.7371,
|
|
"mean_token_accuracy": 0.10238605439662933,
|
|
"num_tokens": 1136750.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.227217006683349,
|
|
"epoch": 0.5133228840125392,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 6.783,
|
|
"mean_token_accuracy": 0.10508323460817337,
|
|
"num_tokens": 1145739.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 7.235933113098144,
|
|
"epoch": 0.5172413793103449,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 6.8375,
|
|
"mean_token_accuracy": 0.10406330898404122,
|
|
"num_tokens": 1154415.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 7.198075199127198,
|
|
"epoch": 0.5211598746081505,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 6.7247,
|
|
"mean_token_accuracy": 0.1038141705095768,
|
|
"num_tokens": 1162480.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.198743104934692,
|
|
"epoch": 0.5250783699059561,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 6.808,
|
|
"mean_token_accuracy": 0.0985616073012352,
|
|
"num_tokens": 1170774.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.186748123168945,
|
|
"epoch": 0.5289968652037618,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000337,
|
|
"loss": 6.743,
|
|
"mean_token_accuracy": 0.10074765086174012,
|
|
"num_tokens": 1179629.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.185740804672241,
|
|
"epoch": 0.5329153605015674,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 6.8976,
|
|
"mean_token_accuracy": 0.10007678046822548,
|
|
"num_tokens": 1189111.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.288842678070068,
|
|
"epoch": 0.536833855799373,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000342,
|
|
"loss": 6.8946,
|
|
"mean_token_accuracy": 0.09661665931344032,
|
|
"num_tokens": 1198827.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.352398300170899,
|
|
"epoch": 0.5407523510971787,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 6.7966,
|
|
"mean_token_accuracy": 0.10527415350079536,
|
|
"num_tokens": 1207089.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.145380878448487,
|
|
"epoch": 0.5446708463949843,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 6.7822,
|
|
"mean_token_accuracy": 0.10294175297021865,
|
|
"num_tokens": 1216509.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.118182611465454,
|
|
"epoch": 0.54858934169279,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 6.6116,
|
|
"mean_token_accuracy": 0.11104098409414291,
|
|
"num_tokens": 1225684.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 7.15866265296936,
|
|
"epoch": 0.5525078369905956,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000352,
|
|
"loss": 6.657,
|
|
"mean_token_accuracy": 0.1122810646891594,
|
|
"num_tokens": 1233675.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.050290060043335,
|
|
"epoch": 0.5564263322884012,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 6.6596,
|
|
"mean_token_accuracy": 0.10829117894172668,
|
|
"num_tokens": 1242147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.144255828857422,
|
|
"epoch": 0.5603448275862069,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000357,
|
|
"loss": 6.7579,
|
|
"mean_token_accuracy": 0.1056147575378418,
|
|
"num_tokens": 1251127.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.173990964889526,
|
|
"epoch": 0.5642633228840125,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 6.7293,
|
|
"mean_token_accuracy": 0.1077524796128273,
|
|
"num_tokens": 1260589.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.268736267089844,
|
|
"epoch": 0.5681818181818182,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.000362,
|
|
"loss": 6.7872,
|
|
"mean_token_accuracy": 0.10429466664791107,
|
|
"num_tokens": 1269281.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.0816668510437015,
|
|
"epoch": 0.5721003134796239,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 6.6605,
|
|
"mean_token_accuracy": 0.10658104047179222,
|
|
"num_tokens": 1278033.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.192685079574585,
|
|
"epoch": 0.5760188087774295,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000367,
|
|
"loss": 6.7278,
|
|
"mean_token_accuracy": 0.1095947191119194,
|
|
"num_tokens": 1285847.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.141543245315551,
|
|
"epoch": 0.5799373040752351,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 6.7272,
|
|
"mean_token_accuracy": 0.1035495713353157,
|
|
"num_tokens": 1294770.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.305365896224975,
|
|
"epoch": 0.5838557993730408,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000372,
|
|
"loss": 6.9019,
|
|
"mean_token_accuracy": 0.09915386661887168,
|
|
"num_tokens": 1304282.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.114371252059937,
|
|
"epoch": 0.5877742946708464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 6.7476,
|
|
"mean_token_accuracy": 0.10341449677944184,
|
|
"num_tokens": 1313218.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.149020004272461,
|
|
"epoch": 0.591692789968652,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.000377,
|
|
"loss": 6.7097,
|
|
"mean_token_accuracy": 0.10334419459104538,
|
|
"num_tokens": 1322093.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.016652154922485,
|
|
"epoch": 0.5956112852664577,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 6.7458,
|
|
"mean_token_accuracy": 0.10744681730866432,
|
|
"num_tokens": 1330718.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.10248818397522,
|
|
"epoch": 0.5995297805642633,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.000382,
|
|
"loss": 6.6178,
|
|
"mean_token_accuracy": 0.10776162892580032,
|
|
"num_tokens": 1339672.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.099648666381836,
|
|
"epoch": 0.603448275862069,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 6.6488,
|
|
"mean_token_accuracy": 0.1087779700756073,
|
|
"num_tokens": 1348587.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.018001747131348,
|
|
"epoch": 0.6073667711598746,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 6.6294,
|
|
"mean_token_accuracy": 0.11382819563150406,
|
|
"num_tokens": 1357597.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.176814222335816,
|
|
"epoch": 0.6112852664576802,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 6.7204,
|
|
"mean_token_accuracy": 0.10900540873408318,
|
|
"num_tokens": 1366246.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.034693956375122,
|
|
"epoch": 0.6152037617554859,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 6.6284,
|
|
"mean_token_accuracy": 0.10845897570252419,
|
|
"num_tokens": 1374664.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 6.943607664108276,
|
|
"epoch": 0.6191222570532915,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 6.642,
|
|
"mean_token_accuracy": 0.10764897763729095,
|
|
"num_tokens": 1382765.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.017871522903443,
|
|
"epoch": 0.6230407523510971,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 6.6123,
|
|
"mean_token_accuracy": 0.1094035878777504,
|
|
"num_tokens": 1391708.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.066233396530151,
|
|
"epoch": 0.6269592476489029,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 6.761,
|
|
"mean_token_accuracy": 0.09915644004940986,
|
|
"num_tokens": 1401373.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.068959426879883,
|
|
"epoch": 0.6308777429467085,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.000402,
|
|
"loss": 6.5598,
|
|
"mean_token_accuracy": 0.11302306056022644,
|
|
"num_tokens": 1409702.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.03672308921814,
|
|
"epoch": 0.6347962382445141,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 6.77,
|
|
"mean_token_accuracy": 0.10150963887572288,
|
|
"num_tokens": 1419596.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.083284759521485,
|
|
"epoch": 0.6387147335423198,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 6.7146,
|
|
"mean_token_accuracy": 0.1047218032181263,
|
|
"num_tokens": 1428437.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.074973106384277,
|
|
"epoch": 0.6426332288401254,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 6.6087,
|
|
"mean_token_accuracy": 0.10951325818896293,
|
|
"num_tokens": 1436764.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 6.933263826370239,
|
|
"epoch": 0.646551724137931,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 6.6592,
|
|
"mean_token_accuracy": 0.11087424829602241,
|
|
"num_tokens": 1445755.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.154396057128906,
|
|
"epoch": 0.6504702194357367,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 6.7092,
|
|
"mean_token_accuracy": 0.10701614245772362,
|
|
"num_tokens": 1454550.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 6.98425703048706,
|
|
"epoch": 0.6543887147335423,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.000417,
|
|
"loss": 6.5981,
|
|
"mean_token_accuracy": 0.11420469135046005,
|
|
"num_tokens": 1463039.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 6.988760995864868,
|
|
"epoch": 0.658307210031348,
|
|
"grad_norm": 1.2578125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 6.6625,
|
|
"mean_token_accuracy": 0.10574135035276414,
|
|
"num_tokens": 1472029.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.009801864624023,
|
|
"epoch": 0.6622257053291536,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 6.6865,
|
|
"mean_token_accuracy": 0.10918803215026855,
|
|
"num_tokens": 1481588.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.000622129440307,
|
|
"epoch": 0.6661442006269592,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 6.5815,
|
|
"mean_token_accuracy": 0.1104504257440567,
|
|
"num_tokens": 1489915.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.095453310012817,
|
|
"epoch": 0.6700626959247649,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.000427,
|
|
"loss": 6.6823,
|
|
"mean_token_accuracy": 0.10836535915732384,
|
|
"num_tokens": 1497920.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 6.9481316089630125,
|
|
"epoch": 0.6739811912225705,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.1121792584657669,
|
|
"num_tokens": 1506965.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 6.970028448104858,
|
|
"epoch": 0.6778996865203761,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000432,
|
|
"loss": 6.6225,
|
|
"mean_token_accuracy": 0.11289523839950562,
|
|
"num_tokens": 1516028.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.008283472061157,
|
|
"epoch": 0.6818181818181818,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 6.6409,
|
|
"mean_token_accuracy": 0.10912753567099572,
|
|
"num_tokens": 1524579.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 6.899019956588745,
|
|
"epoch": 0.6857366771159875,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000437,
|
|
"loss": 6.5487,
|
|
"mean_token_accuracy": 0.11380631253123283,
|
|
"num_tokens": 1534285.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 6.978461980819702,
|
|
"epoch": 0.6896551724137931,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 6.5658,
|
|
"mean_token_accuracy": 0.11228583827614784,
|
|
"num_tokens": 1543701.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.062817907333374,
|
|
"epoch": 0.6935736677115988,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000442,
|
|
"loss": 6.8259,
|
|
"mean_token_accuracy": 0.09782676100730896,
|
|
"num_tokens": 1552715.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 6.756741905212403,
|
|
"epoch": 0.6974921630094044,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 6.4375,
|
|
"mean_token_accuracy": 0.11271021217107773,
|
|
"num_tokens": 1561523.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.057762336730957,
|
|
"epoch": 0.70141065830721,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000447,
|
|
"loss": 6.6532,
|
|
"mean_token_accuracy": 0.11261685714125633,
|
|
"num_tokens": 1570057.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 6.8740034103393555,
|
|
"epoch": 0.7053291536050157,
|
|
"grad_norm": 1.234375,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 6.5135,
|
|
"mean_token_accuracy": 0.11441917642951012,
|
|
"num_tokens": 1578795.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 6.84917516708374,
|
|
"epoch": 0.7092476489028213,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 6.5937,
|
|
"mean_token_accuracy": 0.11051025837659836,
|
|
"num_tokens": 1588108.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.046537971496582,
|
|
"epoch": 0.713166144200627,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 6.5419,
|
|
"mean_token_accuracy": 0.1160262867808342,
|
|
"num_tokens": 1596545.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 6.86083025932312,
|
|
"epoch": 0.7170846394984326,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 6.5415,
|
|
"mean_token_accuracy": 0.11502676978707313,
|
|
"num_tokens": 1605098.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 6.83633713722229,
|
|
"epoch": 0.7210031347962382,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 6.5629,
|
|
"mean_token_accuracy": 0.11907657384872436,
|
|
"num_tokens": 1613760.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 6.9062807083129885,
|
|
"epoch": 0.7249216300940439,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.000462,
|
|
"loss": 6.5931,
|
|
"mean_token_accuracy": 0.11272144094109535,
|
|
"num_tokens": 1622345.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.0440794944763185,
|
|
"epoch": 0.7288401253918495,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 6.5803,
|
|
"mean_token_accuracy": 0.11651012524962426,
|
|
"num_tokens": 1630706.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 6.893813562393189,
|
|
"epoch": 0.7327586206896551,
|
|
"grad_norm": 1.296875,
|
|
"learning_rate": 0.000467,
|
|
"loss": 6.6478,
|
|
"mean_token_accuracy": 0.10516302213072777,
|
|
"num_tokens": 1639880.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 6.890382957458496,
|
|
"epoch": 0.7366771159874608,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 6.4892,
|
|
"mean_token_accuracy": 0.11007987186312676,
|
|
"num_tokens": 1648325.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 6.773949003219604,
|
|
"epoch": 0.7405956112852664,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000472,
|
|
"loss": 6.4465,
|
|
"mean_token_accuracy": 0.11900387927889824,
|
|
"num_tokens": 1656951.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 6.8440714359283445,
|
|
"epoch": 0.7445141065830722,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 6.5745,
|
|
"mean_token_accuracy": 0.11141012907028199,
|
|
"num_tokens": 1665832.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 6.896161079406738,
|
|
"epoch": 0.7484326018808778,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 6.5042,
|
|
"mean_token_accuracy": 0.11246282234787941,
|
|
"num_tokens": 1675116.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 6.777887010574341,
|
|
"epoch": 0.7523510971786834,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 6.5073,
|
|
"mean_token_accuracy": 0.11693843528628349,
|
|
"num_tokens": 1684195.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 6.849985933303833,
|
|
"epoch": 0.7562695924764891,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000482,
|
|
"loss": 6.4641,
|
|
"mean_token_accuracy": 0.11287508308887481,
|
|
"num_tokens": 1693025.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 6.709165954589844,
|
|
"epoch": 0.7601880877742947,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 6.4612,
|
|
"mean_token_accuracy": 0.11437803655862808,
|
|
"num_tokens": 1701210.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 6.903396844863892,
|
|
"epoch": 0.7641065830721003,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.000487,
|
|
"loss": 6.553,
|
|
"mean_token_accuracy": 0.10555337145924568,
|
|
"num_tokens": 1711026.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 6.918910503387451,
|
|
"epoch": 0.768025078369906,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 6.57,
|
|
"mean_token_accuracy": 0.11464387774467469,
|
|
"num_tokens": 1719441.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 6.7402503967285154,
|
|
"epoch": 0.7719435736677116,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000492,
|
|
"loss": 6.536,
|
|
"mean_token_accuracy": 0.11087964922189712,
|
|
"num_tokens": 1728300.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 6.8907630443573,
|
|
"epoch": 0.7758620689655172,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 6.4406,
|
|
"mean_token_accuracy": 0.11694614291191101,
|
|
"num_tokens": 1736364.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 6.68822078704834,
|
|
"epoch": 0.7797805642633229,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000497,
|
|
"loss": 6.2699,
|
|
"mean_token_accuracy": 0.1223867692053318,
|
|
"num_tokens": 1745043.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 6.7981610774993895,
|
|
"epoch": 0.7836990595611285,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.11675246804952621,
|
|
"num_tokens": 1754008.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 0.7836990595611285,
|
|
"eval_entropy": 6.654667783153149,
|
|
"eval_loss": 6.574846267700195,
|
|
"eval_mean_token_accuracy": 0.11710607061205908,
|
|
"eval_num_tokens": 1754008.0,
|
|
"eval_runtime": 2.3667,
|
|
"eval_samples_per_second": 1741.632,
|
|
"eval_steps_per_second": 218.021,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 6.649208927154541,
|
|
"epoch": 0.7876175548589341,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004999998713243189,
|
|
"loss": 6.4846,
|
|
"mean_token_accuracy": 0.11015921831130981,
|
|
"num_tokens": 1763592.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 6.839963436126709,
|
|
"epoch": 0.7915360501567398,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004999993485796164,
|
|
"loss": 6.3468,
|
|
"mean_token_accuracy": 0.11887188702821731,
|
|
"num_tokens": 1773111.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 6.739857482910156,
|
|
"epoch": 0.7954545454545454,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999984237245962,
|
|
"loss": 6.4146,
|
|
"mean_token_accuracy": 0.12405953705310821,
|
|
"num_tokens": 1781230.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 6.8473914623260494,
|
|
"epoch": 0.799373040752351,
|
|
"grad_norm": 1.3671875,
|
|
"learning_rate": 0.0004999970967609109,
|
|
"loss": 6.6375,
|
|
"mean_token_accuracy": 0.11473575830459595,
|
|
"num_tokens": 1790324.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 6.818466901779175,
|
|
"epoch": 0.8032915360501567,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.0004999953676909322,
|
|
"loss": 6.488,
|
|
"mean_token_accuracy": 0.11022145599126816,
|
|
"num_tokens": 1798849.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 6.805118274688721,
|
|
"epoch": 0.8072100313479624,
|
|
"grad_norm": 1.375,
|
|
"learning_rate": 0.0004999932365177502,
|
|
"loss": 6.3716,
|
|
"mean_token_accuracy": 0.1193474993109703,
|
|
"num_tokens": 1808019.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 6.7946457862854,
|
|
"epoch": 0.8111285266457681,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.0004999907032451735,
|
|
"loss": 6.4944,
|
|
"mean_token_accuracy": 0.11334498226642609,
|
|
"num_tokens": 1816550.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 6.711936664581299,
|
|
"epoch": 0.8150470219435737,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004999877678777296,
|
|
"loss": 6.3788,
|
|
"mean_token_accuracy": 0.12799450531601905,
|
|
"num_tokens": 1825238.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 6.681297397613525,
|
|
"epoch": 0.8189655172413793,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999844304206645,
|
|
"loss": 6.346,
|
|
"mean_token_accuracy": 0.12015214636921882,
|
|
"num_tokens": 1833645.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 6.777402448654175,
|
|
"epoch": 0.822884012539185,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004999806908799428,
|
|
"loss": 6.4284,
|
|
"mean_token_accuracy": 0.1182809017598629,
|
|
"num_tokens": 1842892.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 6.647558307647705,
|
|
"epoch": 0.8268025078369906,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004999765492622474,
|
|
"loss": 6.4544,
|
|
"mean_token_accuracy": 0.1192950263619423,
|
|
"num_tokens": 1851400.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 6.854276514053344,
|
|
"epoch": 0.8307210031347962,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999720055749804,
|
|
"loss": 6.4078,
|
|
"mean_token_accuracy": 0.1215155504643917,
|
|
"num_tokens": 1860088.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 6.762131643295288,
|
|
"epoch": 0.8346394984326019,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004999670598262619,
|
|
"loss": 6.4539,
|
|
"mean_token_accuracy": 0.12213057726621628,
|
|
"num_tokens": 1868558.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 6.676323461532593,
|
|
"epoch": 0.8385579937304075,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004999617120249308,
|
|
"loss": 6.4381,
|
|
"mean_token_accuracy": 0.11797192841768264,
|
|
"num_tokens": 1877624.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 6.867647266387939,
|
|
"epoch": 0.8424764890282131,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004999559621805445,
|
|
"loss": 6.5615,
|
|
"mean_token_accuracy": 0.11477635577321052,
|
|
"num_tokens": 1886505.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 6.589630508422852,
|
|
"epoch": 0.8463949843260188,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004999498103033788,
|
|
"loss": 6.2977,
|
|
"mean_token_accuracy": 0.12378846853971481,
|
|
"num_tokens": 1895701.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 6.666740560531617,
|
|
"epoch": 0.8503134796238244,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999432564044284,
|
|
"loss": 6.2686,
|
|
"mean_token_accuracy": 0.12306156605482102,
|
|
"num_tokens": 1904640.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 6.720833110809326,
|
|
"epoch": 0.85423197492163,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004999363004954058,
|
|
"loss": 6.481,
|
|
"mean_token_accuracy": 0.11864836812019348,
|
|
"num_tokens": 1914216.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 6.69180383682251,
|
|
"epoch": 0.8581504702194357,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999289425887425,
|
|
"loss": 6.4162,
|
|
"mean_token_accuracy": 0.12137454971671105,
|
|
"num_tokens": 1923280.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 6.79173903465271,
|
|
"epoch": 0.8620689655172413,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004999211826975884,
|
|
"loss": 6.4387,
|
|
"mean_token_accuracy": 0.12172688096761704,
|
|
"num_tokens": 1931403.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 6.808387517929077,
|
|
"epoch": 0.8659874608150471,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004999130208358114,
|
|
"loss": 6.4484,
|
|
"mean_token_accuracy": 0.1191784381866455,
|
|
"num_tokens": 1940774.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 6.769060564041138,
|
|
"epoch": 0.8699059561128527,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004999044570179983,
|
|
"loss": 6.5459,
|
|
"mean_token_accuracy": 0.1176558330655098,
|
|
"num_tokens": 1950045.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 6.803036499023437,
|
|
"epoch": 0.8738244514106583,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998954912594541,
|
|
"loss": 6.3989,
|
|
"mean_token_accuracy": 0.1201501503586769,
|
|
"num_tokens": 1959303.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 6.66001410484314,
|
|
"epoch": 0.877742946708464,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004998861235762018,
|
|
"loss": 6.382,
|
|
"mean_token_accuracy": 0.11791504472494126,
|
|
"num_tokens": 1968583.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 6.7807149410247805,
|
|
"epoch": 0.8816614420062696,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004998763539849832,
|
|
"loss": 6.4873,
|
|
"mean_token_accuracy": 0.11481102481484413,
|
|
"num_tokens": 1977489.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 6.625271320343018,
|
|
"epoch": 0.8855799373040752,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004998661825032579,
|
|
"loss": 6.2531,
|
|
"mean_token_accuracy": 0.12325010523200035,
|
|
"num_tokens": 1985308.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 6.662072563171387,
|
|
"epoch": 0.8894984326018809,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004998556091492043,
|
|
"loss": 6.4269,
|
|
"mean_token_accuracy": 0.11545633301138877,
|
|
"num_tokens": 1993930.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 6.648345518112182,
|
|
"epoch": 0.8934169278996865,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004998446339417184,
|
|
"loss": 6.4231,
|
|
"mean_token_accuracy": 0.11234267503023147,
|
|
"num_tokens": 2003802.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 6.76238489151001,
|
|
"epoch": 0.8973354231974922,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004998332569004149,
|
|
"loss": 6.3499,
|
|
"mean_token_accuracy": 0.1253760874271393,
|
|
"num_tokens": 2011957.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 6.612128019332886,
|
|
"epoch": 0.9012539184952978,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998214780456263,
|
|
"loss": 6.3359,
|
|
"mean_token_accuracy": 0.12428404837846756,
|
|
"num_tokens": 2020741.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 6.711992835998535,
|
|
"epoch": 0.9051724137931034,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004998092973984033,
|
|
"loss": 6.407,
|
|
"mean_token_accuracy": 0.11499650552868843,
|
|
"num_tokens": 2029804.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 6.695094060897827,
|
|
"epoch": 0.9090909090909091,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997967149805147,
|
|
"loss": 6.4652,
|
|
"mean_token_accuracy": 0.11959076821804046,
|
|
"num_tokens": 2038732.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 6.733592510223389,
|
|
"epoch": 0.9130094043887147,
|
|
"grad_norm": 1.25,
|
|
"learning_rate": 0.0004997837308144474,
|
|
"loss": 6.3533,
|
|
"mean_token_accuracy": 0.12914359271526338,
|
|
"num_tokens": 2046362.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 6.622471714019776,
|
|
"epoch": 0.9169278996865203,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004997703449234062,
|
|
"loss": 6.3397,
|
|
"mean_token_accuracy": 0.11818950101733208,
|
|
"num_tokens": 2056018.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 6.590902185440063,
|
|
"epoch": 0.920846394984326,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004997565573313139,
|
|
"loss": 6.3139,
|
|
"mean_token_accuracy": 0.11889233589172363,
|
|
"num_tokens": 2065077.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 6.753049659729004,
|
|
"epoch": 0.9247648902821317,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004997423680628109,
|
|
"loss": 6.3166,
|
|
"mean_token_accuracy": 0.12146413177251816,
|
|
"num_tokens": 2073405.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 6.4639513969421385,
|
|
"epoch": 0.9286833855799373,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499727777143256,
|
|
"loss": 6.2857,
|
|
"mean_token_accuracy": 0.12157903984189034,
|
|
"num_tokens": 2081891.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 6.8065098285675045,
|
|
"epoch": 0.932601880877743,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004997127845987255,
|
|
"loss": 6.3537,
|
|
"mean_token_accuracy": 0.1274058498442173,
|
|
"num_tokens": 2090732.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 6.566525173187256,
|
|
"epoch": 0.9365203761755486,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0004996973904560135,
|
|
"loss": 6.2968,
|
|
"mean_token_accuracy": 0.128861191123724,
|
|
"num_tokens": 2099352.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 6.673365211486816,
|
|
"epoch": 0.9404388714733543,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004996815947426317,
|
|
"loss": 6.3191,
|
|
"mean_token_accuracy": 0.1260666698217392,
|
|
"num_tokens": 2107365.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 6.596097421646118,
|
|
"epoch": 0.9443573667711599,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004996653974868097,
|
|
"loss": 6.1833,
|
|
"mean_token_accuracy": 0.12711638659238816,
|
|
"num_tokens": 2116192.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 6.5889317989349365,
|
|
"epoch": 0.9482758620689655,
|
|
"grad_norm": 1.140625,
|
|
"learning_rate": 0.0004996487987174946,
|
|
"loss": 6.2961,
|
|
"mean_token_accuracy": 0.11961350440979004,
|
|
"num_tokens": 2124410.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 6.608324480056763,
|
|
"epoch": 0.9521943573667712,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004996317984643512,
|
|
"loss": 6.3123,
|
|
"mean_token_accuracy": 0.12350876927375794,
|
|
"num_tokens": 2132799.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 6.701848888397217,
|
|
"epoch": 0.9561128526645768,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004996143967577615,
|
|
"loss": 6.4849,
|
|
"mean_token_accuracy": 0.11310579478740693,
|
|
"num_tokens": 2141589.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 6.725940990447998,
|
|
"epoch": 0.9600313479623824,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004995965936288254,
|
|
"loss": 6.3573,
|
|
"mean_token_accuracy": 0.12382125556468963,
|
|
"num_tokens": 2150704.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 6.522915983200074,
|
|
"epoch": 0.9639498432601881,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004995783891093597,
|
|
"loss": 6.2705,
|
|
"mean_token_accuracy": 0.1250034339725971,
|
|
"num_tokens": 2159587.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 6.510832786560059,
|
|
"epoch": 0.9678683385579937,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000499559783231899,
|
|
"loss": 6.2626,
|
|
"mean_token_accuracy": 0.11797072142362594,
|
|
"num_tokens": 2169071.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 6.600325727462769,
|
|
"epoch": 0.9717868338557993,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004995407760296952,
|
|
"loss": 6.3472,
|
|
"mean_token_accuracy": 0.11733860597014427,
|
|
"num_tokens": 2178248.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 6.62357530593872,
|
|
"epoch": 0.975705329153605,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004995213675367169,
|
|
"loss": 6.2451,
|
|
"mean_token_accuracy": 0.11946402341127396,
|
|
"num_tokens": 2187520.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 6.519882535934448,
|
|
"epoch": 0.9796238244514106,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004995015577876504,
|
|
"loss": 6.3099,
|
|
"mean_token_accuracy": 0.11913644820451737,
|
|
"num_tokens": 2196861.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 6.558454179763794,
|
|
"epoch": 0.9835423197492164,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.000499481346817899,
|
|
"loss": 6.2324,
|
|
"mean_token_accuracy": 0.12462019845843315,
|
|
"num_tokens": 2205523.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 6.541610050201416,
|
|
"epoch": 0.987460815047022,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004994607346635829,
|
|
"loss": 6.1878,
|
|
"mean_token_accuracy": 0.13304450660943984,
|
|
"num_tokens": 2214291.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 6.681296253204346,
|
|
"epoch": 0.9913793103448276,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004994397213615394,
|
|
"loss": 6.3197,
|
|
"mean_token_accuracy": 0.12296603471040726,
|
|
"num_tokens": 2223093.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 6.531461000442505,
|
|
"epoch": 0.9952978056426333,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004994183069493225,
|
|
"loss": 6.2564,
|
|
"mean_token_accuracy": 0.12638100013136863,
|
|
"num_tokens": 2231551.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 6.491727447509765,
|
|
"epoch": 0.9992163009404389,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004993964914652034,
|
|
"loss": 6.2405,
|
|
"mean_token_accuracy": 0.1313079461455345,
|
|
"num_tokens": 2240650.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 6.658728313446045,
|
|
"epoch": 1.0031347962382444,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993742749481699,
|
|
"loss": 6.186,
|
|
"mean_token_accuracy": 0.12531767934560775,
|
|
"num_tokens": 2249392.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 6.395848560333252,
|
|
"epoch": 1.0070532915360502,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004993516574379265,
|
|
"loss": 5.9971,
|
|
"mean_token_accuracy": 0.1336176484823227,
|
|
"num_tokens": 2258100.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 6.498657894134522,
|
|
"epoch": 1.0109717868338557,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004993286389748943,
|
|
"loss": 6.0709,
|
|
"mean_token_accuracy": 0.1330066442489624,
|
|
"num_tokens": 2266661.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 6.504871654510498,
|
|
"epoch": 1.0148902821316614,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004993052196002112,
|
|
"loss": 6.1585,
|
|
"mean_token_accuracy": 0.12869242057204247,
|
|
"num_tokens": 2275475.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 6.493157625198364,
|
|
"epoch": 1.0188087774294672,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004992813993557312,
|
|
"loss": 6.0818,
|
|
"mean_token_accuracy": 0.132894304394722,
|
|
"num_tokens": 2284077.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 6.4438941955566404,
|
|
"epoch": 1.0227272727272727,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.000499257178284025,
|
|
"loss": 6.0829,
|
|
"mean_token_accuracy": 0.12662639245390891,
|
|
"num_tokens": 2293029.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 6.4581663608551025,
|
|
"epoch": 1.0266457680250785,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004992325564283797,
|
|
"loss": 6.0023,
|
|
"mean_token_accuracy": 0.14118103608489035,
|
|
"num_tokens": 2301574.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 6.411208295822144,
|
|
"epoch": 1.030564263322884,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004992075338327984,
|
|
"loss": 6.0123,
|
|
"mean_token_accuracy": 0.1320878393948078,
|
|
"num_tokens": 2310366.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 6.492961502075195,
|
|
"epoch": 1.0344827586206897,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004991821105420006,
|
|
"loss": 6.1267,
|
|
"mean_token_accuracy": 0.1311240755021572,
|
|
"num_tokens": 2318943.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 6.596848249435425,
|
|
"epoch": 1.0384012539184952,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.0004991562866014219,
|
|
"loss": 6.129,
|
|
"mean_token_accuracy": 0.13712250739336013,
|
|
"num_tokens": 2327173.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 6.508216285705567,
|
|
"epoch": 1.042319749216301,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.0004991300620572138,
|
|
"loss": 6.0855,
|
|
"mean_token_accuracy": 0.1303306058049202,
|
|
"num_tokens": 2336194.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 6.420964479446411,
|
|
"epoch": 1.0462382445141065,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0004991034369562438,
|
|
"loss": 6.148,
|
|
"mean_token_accuracy": 0.13162081241607665,
|
|
"num_tokens": 2344721.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 6.457345342636108,
|
|
"epoch": 1.0501567398119123,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004990764113460953,
|
|
"loss": 6.0766,
|
|
"mean_token_accuracy": 0.13223855197429657,
|
|
"num_tokens": 2353406.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 6.553745222091675,
|
|
"epoch": 1.0540752351097178,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004990489852750675,
|
|
"loss": 6.2511,
|
|
"mean_token_accuracy": 0.12270163372159004,
|
|
"num_tokens": 2361896.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 6.4831219673156735,
|
|
"epoch": 1.0579937304075235,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004990211587921751,
|
|
"loss": 6.0737,
|
|
"mean_token_accuracy": 0.1310164101421833,
|
|
"num_tokens": 2371861.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 6.373387718200684,
|
|
"epoch": 1.061912225705329,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004989929319471487,
|
|
"loss": 5.9984,
|
|
"mean_token_accuracy": 0.13476544693112374,
|
|
"num_tokens": 2381555.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 6.396108531951905,
|
|
"epoch": 1.0658307210031348,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004989643047904341,
|
|
"loss": 6.0114,
|
|
"mean_token_accuracy": 0.1410997711122036,
|
|
"num_tokens": 2389851.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 6.388179016113281,
|
|
"epoch": 1.0697492163009406,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989352773731928,
|
|
"loss": 5.9311,
|
|
"mean_token_accuracy": 0.14064377471804618,
|
|
"num_tokens": 2397764.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 6.448402547836304,
|
|
"epoch": 1.073667711598746,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.0004989058497473014,
|
|
"loss": 6.0836,
|
|
"mean_token_accuracy": 0.1399063892662525,
|
|
"num_tokens": 2406688.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 6.525130748748779,
|
|
"epoch": 1.0775862068965518,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004988760219653518,
|
|
"loss": 6.1549,
|
|
"mean_token_accuracy": 0.131688591837883,
|
|
"num_tokens": 2415475.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 6.424018096923828,
|
|
"epoch": 1.0815047021943573,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0004988457940806513,
|
|
"loss": 6.0973,
|
|
"mean_token_accuracy": 0.1294810250401497,
|
|
"num_tokens": 2425719.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 6.418878507614136,
|
|
"epoch": 1.085423197492163,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004988151661472218,
|
|
"loss": 6.0063,
|
|
"mean_token_accuracy": 0.13849975615739823,
|
|
"num_tokens": 2435138.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 6.45744514465332,
|
|
"epoch": 1.0893416927899686,
|
|
"grad_norm": 1.3359375,
|
|
"learning_rate": 0.0004987841382198006,
|
|
"loss": 6.1368,
|
|
"mean_token_accuracy": 0.13256047889590264,
|
|
"num_tokens": 2443976.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 6.278638410568237,
|
|
"epoch": 1.0932601880877744,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.0004987527103538394,
|
|
"loss": 6.0461,
|
|
"mean_token_accuracy": 0.13556990772485733,
|
|
"num_tokens": 2453287.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 6.505991792678833,
|
|
"epoch": 1.09717868338558,
|
|
"grad_norm": 1.2890625,
|
|
"learning_rate": 0.000498720882605505,
|
|
"loss": 6.1381,
|
|
"mean_token_accuracy": 0.1309444561600685,
|
|
"num_tokens": 2462054.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 6.364629411697388,
|
|
"epoch": 1.1010971786833856,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0004986886550316788,
|
|
"loss": 6.0369,
|
|
"mean_token_accuracy": 0.13807333409786224,
|
|
"num_tokens": 2470333.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 6.3567602157592775,
|
|
"epoch": 1.1050156739811912,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004986560276899565,
|
|
"loss": 6.0596,
|
|
"mean_token_accuracy": 0.13362243846058847,
|
|
"num_tokens": 2479621.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 6.439157104492187,
|
|
"epoch": 1.108934169278997,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004986230006386488,
|
|
"loss": 6.038,
|
|
"mean_token_accuracy": 0.1371801607310772,
|
|
"num_tokens": 2488020.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 6.429386949539184,
|
|
"epoch": 1.1128526645768024,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004985895739367802,
|
|
"loss": 6.0645,
|
|
"mean_token_accuracy": 0.13344691395759584,
|
|
"num_tokens": 2497779.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 6.466728544235229,
|
|
"epoch": 1.1167711598746082,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004985557476440895,
|
|
"loss": 6.0183,
|
|
"mean_token_accuracy": 0.13338489159941674,
|
|
"num_tokens": 2506791.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 6.395694780349731,
|
|
"epoch": 1.1206896551724137,
|
|
"grad_norm": 1.1875,
|
|
"learning_rate": 0.00049852152182103,
|
|
"loss": 6.1286,
|
|
"mean_token_accuracy": 0.13055293411016464,
|
|
"num_tokens": 2515576.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 6.488246631622315,
|
|
"epoch": 1.1246081504702194,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004984868965287686,
|
|
"loss": 6.0423,
|
|
"mean_token_accuracy": 0.1341018334031105,
|
|
"num_tokens": 2523933.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 6.3448042392730715,
|
|
"epoch": 1.1285266457680252,
|
|
"grad_norm": 1.2265625,
|
|
"learning_rate": 0.0004984518718291864,
|
|
"loss": 5.983,
|
|
"mean_token_accuracy": 0.13408639580011367,
|
|
"num_tokens": 2532453.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 6.4296407222747805,
|
|
"epoch": 1.1324451410658307,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0004984164477848783,
|
|
"loss": 6.0353,
|
|
"mean_token_accuracy": 0.12955902740359307,
|
|
"num_tokens": 2541149.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 6.364090204238892,
|
|
"epoch": 1.1363636363636362,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004983806244591528,
|
|
"loss": 6.0056,
|
|
"mean_token_accuracy": 0.1397395424544811,
|
|
"num_tokens": 2550302.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 6.3148603439331055,
|
|
"epoch": 1.140282131661442,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004983444019160318,
|
|
"loss": 5.8909,
|
|
"mean_token_accuracy": 0.1413780689239502,
|
|
"num_tokens": 2558507.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 6.429317855834961,
|
|
"epoch": 1.1442006269592477,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004983077802202511,
|
|
"loss": 6.1431,
|
|
"mean_token_accuracy": 0.1290089689195156,
|
|
"num_tokens": 2567487.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 6.423747301101685,
|
|
"epoch": 1.1481191222570533,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004982707594372595,
|
|
"loss": 6.0036,
|
|
"mean_token_accuracy": 0.1317966252565384,
|
|
"num_tokens": 2576232.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 6.29382700920105,
|
|
"epoch": 1.152037617554859,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000498233339633219,
|
|
"loss": 5.9924,
|
|
"mean_token_accuracy": 0.13600054606795312,
|
|
"num_tokens": 2585081.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 6.3161386013031,
|
|
"epoch": 1.1559561128526645,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000498195520875005,
|
|
"loss": 5.9778,
|
|
"mean_token_accuracy": 0.13530001565814018,
|
|
"num_tokens": 2593890.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 6.392177295684815,
|
|
"epoch": 1.1598746081504703,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.0004981573032302059,
|
|
"loss": 6.0645,
|
|
"mean_token_accuracy": 0.13251638934016227,
|
|
"num_tokens": 2602011.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 6.3406764507293705,
|
|
"epoch": 1.1637931034482758,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004981186867671225,
|
|
"loss": 5.9532,
|
|
"mean_token_accuracy": 0.142037995159626,
|
|
"num_tokens": 2610366.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 6.236144065856934,
|
|
"epoch": 1.1677115987460815,
|
|
"grad_norm": 1.1640625,
|
|
"learning_rate": 0.0004980796715547691,
|
|
"loss": 5.958,
|
|
"mean_token_accuracy": 0.13757709935307502,
|
|
"num_tokens": 2619737.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 6.418478155136109,
|
|
"epoch": 1.171630094043887,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004980402576628717,
|
|
"loss": 5.9928,
|
|
"mean_token_accuracy": 0.13964883610606194,
|
|
"num_tokens": 2628176.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 6.250125694274902,
|
|
"epoch": 1.1755485893416928,
|
|
"grad_norm": 1.2109375,
|
|
"learning_rate": 0.0004980004451618697,
|
|
"loss": 5.9964,
|
|
"mean_token_accuracy": 0.13738328740000724,
|
|
"num_tokens": 2636195.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.1755485893416928,
|
|
"eval_entropy": 6.177109536274459,
|
|
"eval_loss": 6.257157802581787,
|
|
"eval_mean_token_accuracy": 0.1324432172574276,
|
|
"eval_num_tokens": 2636195.0,
|
|
"eval_runtime": 2.1574,
|
|
"eval_samples_per_second": 1910.589,
|
|
"eval_steps_per_second": 239.171,
|
|
"step": 1500
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 12750,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 529071051018240.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|