pythia-2_8b_sft-gpt-turbo / trainer_state.json
DatPySci's picture
Model save
881d74f verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 100,
"global_step": 1312,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.007621951219512195,
"grad_norm": 4.6875,
"learning_rate": 1.5151515151515152e-06,
"loss": 1.6918,
"step": 10
},
{
"epoch": 0.01524390243902439,
"grad_norm": 5.0625,
"learning_rate": 3.0303030303030305e-06,
"loss": 1.6227,
"step": 20
},
{
"epoch": 0.022865853658536585,
"grad_norm": 4.34375,
"learning_rate": 4.5454545454545455e-06,
"loss": 1.6367,
"step": 30
},
{
"epoch": 0.03048780487804878,
"grad_norm": 4.21875,
"learning_rate": 6.060606060606061e-06,
"loss": 1.493,
"step": 40
},
{
"epoch": 0.038109756097560975,
"grad_norm": 2.921875,
"learning_rate": 7.5757575757575764e-06,
"loss": 1.441,
"step": 50
},
{
"epoch": 0.04573170731707317,
"grad_norm": 3.078125,
"learning_rate": 9.090909090909091e-06,
"loss": 1.3488,
"step": 60
},
{
"epoch": 0.053353658536585365,
"grad_norm": 2.828125,
"learning_rate": 1.0606060606060606e-05,
"loss": 1.3289,
"step": 70
},
{
"epoch": 0.06097560975609756,
"grad_norm": 2.765625,
"learning_rate": 1.2121212121212122e-05,
"loss": 1.2742,
"step": 80
},
{
"epoch": 0.06859756097560976,
"grad_norm": 2.890625,
"learning_rate": 1.3636363636363637e-05,
"loss": 1.2758,
"step": 90
},
{
"epoch": 0.07621951219512195,
"grad_norm": 2.859375,
"learning_rate": 1.5151515151515153e-05,
"loss": 1.2707,
"step": 100
},
{
"epoch": 0.08384146341463415,
"grad_norm": 2.890625,
"learning_rate": 1.6666666666666667e-05,
"loss": 1.273,
"step": 110
},
{
"epoch": 0.09146341463414634,
"grad_norm": 2.765625,
"learning_rate": 1.8181818181818182e-05,
"loss": 1.2414,
"step": 120
},
{
"epoch": 0.09908536585365854,
"grad_norm": 2.75,
"learning_rate": 1.96969696969697e-05,
"loss": 1.275,
"step": 130
},
{
"epoch": 0.10670731707317073,
"grad_norm": 2.78125,
"learning_rate": 1.986440677966102e-05,
"loss": 1.2547,
"step": 140
},
{
"epoch": 0.11432926829268293,
"grad_norm": 2.578125,
"learning_rate": 1.969491525423729e-05,
"loss": 1.2391,
"step": 150
},
{
"epoch": 0.12195121951219512,
"grad_norm": 2.75,
"learning_rate": 1.9525423728813562e-05,
"loss": 1.2379,
"step": 160
},
{
"epoch": 0.12957317073170732,
"grad_norm": 2.65625,
"learning_rate": 1.9355932203389832e-05,
"loss": 1.2105,
"step": 170
},
{
"epoch": 0.13719512195121952,
"grad_norm": 2.53125,
"learning_rate": 1.9186440677966102e-05,
"loss": 1.2359,
"step": 180
},
{
"epoch": 0.1448170731707317,
"grad_norm": 2.75,
"learning_rate": 1.9016949152542375e-05,
"loss": 1.2867,
"step": 190
},
{
"epoch": 0.1524390243902439,
"grad_norm": 2.8125,
"learning_rate": 1.8847457627118645e-05,
"loss": 1.1762,
"step": 200
},
{
"epoch": 0.1600609756097561,
"grad_norm": 2.765625,
"learning_rate": 1.8677966101694918e-05,
"loss": 1.1992,
"step": 210
},
{
"epoch": 0.1676829268292683,
"grad_norm": 2.75,
"learning_rate": 1.8508474576271188e-05,
"loss": 1.232,
"step": 220
},
{
"epoch": 0.17530487804878048,
"grad_norm": 2.65625,
"learning_rate": 1.8338983050847458e-05,
"loss": 1.2172,
"step": 230
},
{
"epoch": 0.18292682926829268,
"grad_norm": 2.734375,
"learning_rate": 1.816949152542373e-05,
"loss": 1.182,
"step": 240
},
{
"epoch": 0.19054878048780488,
"grad_norm": 2.875,
"learning_rate": 1.8e-05,
"loss": 1.2492,
"step": 250
},
{
"epoch": 0.19817073170731708,
"grad_norm": 2.71875,
"learning_rate": 1.7830508474576274e-05,
"loss": 1.2312,
"step": 260
},
{
"epoch": 0.20579268292682926,
"grad_norm": 2.65625,
"learning_rate": 1.7661016949152543e-05,
"loss": 1.241,
"step": 270
},
{
"epoch": 0.21341463414634146,
"grad_norm": 2.78125,
"learning_rate": 1.7491525423728813e-05,
"loss": 1.1879,
"step": 280
},
{
"epoch": 0.22103658536585366,
"grad_norm": 2.734375,
"learning_rate": 1.7322033898305086e-05,
"loss": 1.2211,
"step": 290
},
{
"epoch": 0.22865853658536586,
"grad_norm": 2.734375,
"learning_rate": 1.715254237288136e-05,
"loss": 1.2152,
"step": 300
},
{
"epoch": 0.23628048780487804,
"grad_norm": 2.8125,
"learning_rate": 1.698305084745763e-05,
"loss": 1.1938,
"step": 310
},
{
"epoch": 0.24390243902439024,
"grad_norm": 2.71875,
"learning_rate": 1.68135593220339e-05,
"loss": 1.2266,
"step": 320
},
{
"epoch": 0.25152439024390244,
"grad_norm": 2.84375,
"learning_rate": 1.6644067796610172e-05,
"loss": 1.1672,
"step": 330
},
{
"epoch": 0.25914634146341464,
"grad_norm": 2.71875,
"learning_rate": 1.6474576271186442e-05,
"loss": 1.2104,
"step": 340
},
{
"epoch": 0.26676829268292684,
"grad_norm": 2.640625,
"learning_rate": 1.6305084745762715e-05,
"loss": 1.2098,
"step": 350
},
{
"epoch": 0.27439024390243905,
"grad_norm": 2.703125,
"learning_rate": 1.6135593220338985e-05,
"loss": 1.1684,
"step": 360
},
{
"epoch": 0.2820121951219512,
"grad_norm": 2.765625,
"learning_rate": 1.5966101694915255e-05,
"loss": 1.1734,
"step": 370
},
{
"epoch": 0.2896341463414634,
"grad_norm": 2.546875,
"learning_rate": 1.5796610169491528e-05,
"loss": 1.2008,
"step": 380
},
{
"epoch": 0.2972560975609756,
"grad_norm": 2.78125,
"learning_rate": 1.5627118644067798e-05,
"loss": 1.1848,
"step": 390
},
{
"epoch": 0.3048780487804878,
"grad_norm": 2.625,
"learning_rate": 1.545762711864407e-05,
"loss": 1.2033,
"step": 400
},
{
"epoch": 0.3125,
"grad_norm": 2.625,
"learning_rate": 1.528813559322034e-05,
"loss": 1.1781,
"step": 410
},
{
"epoch": 0.3201219512195122,
"grad_norm": 2.96875,
"learning_rate": 1.511864406779661e-05,
"loss": 1.2043,
"step": 420
},
{
"epoch": 0.3277439024390244,
"grad_norm": 2.453125,
"learning_rate": 1.4949152542372882e-05,
"loss": 1.1699,
"step": 430
},
{
"epoch": 0.3353658536585366,
"grad_norm": 2.5625,
"learning_rate": 1.4779661016949153e-05,
"loss": 1.1871,
"step": 440
},
{
"epoch": 0.3429878048780488,
"grad_norm": 2.578125,
"learning_rate": 1.4610169491525426e-05,
"loss": 1.1418,
"step": 450
},
{
"epoch": 0.35060975609756095,
"grad_norm": 2.640625,
"learning_rate": 1.4440677966101698e-05,
"loss": 1.1824,
"step": 460
},
{
"epoch": 0.35823170731707316,
"grad_norm": 2.828125,
"learning_rate": 1.4271186440677966e-05,
"loss": 1.176,
"step": 470
},
{
"epoch": 0.36585365853658536,
"grad_norm": 2.96875,
"learning_rate": 1.4101694915254239e-05,
"loss": 1.1785,
"step": 480
},
{
"epoch": 0.37347560975609756,
"grad_norm": 2.796875,
"learning_rate": 1.393220338983051e-05,
"loss": 1.1758,
"step": 490
},
{
"epoch": 0.38109756097560976,
"grad_norm": 2.75,
"learning_rate": 1.3762711864406782e-05,
"loss": 1.1988,
"step": 500
},
{
"epoch": 0.38871951219512196,
"grad_norm": 2.578125,
"learning_rate": 1.3593220338983053e-05,
"loss": 1.1822,
"step": 510
},
{
"epoch": 0.39634146341463417,
"grad_norm": 2.953125,
"learning_rate": 1.3423728813559323e-05,
"loss": 1.1801,
"step": 520
},
{
"epoch": 0.40396341463414637,
"grad_norm": 2.640625,
"learning_rate": 1.3254237288135595e-05,
"loss": 1.2043,
"step": 530
},
{
"epoch": 0.4115853658536585,
"grad_norm": 2.546875,
"learning_rate": 1.3084745762711866e-05,
"loss": 1.2098,
"step": 540
},
{
"epoch": 0.4192073170731707,
"grad_norm": 2.625,
"learning_rate": 1.2915254237288137e-05,
"loss": 1.1973,
"step": 550
},
{
"epoch": 0.4268292682926829,
"grad_norm": 2.546875,
"learning_rate": 1.2745762711864407e-05,
"loss": 1.2066,
"step": 560
},
{
"epoch": 0.4344512195121951,
"grad_norm": 2.4375,
"learning_rate": 1.2576271186440679e-05,
"loss": 1.1543,
"step": 570
},
{
"epoch": 0.4420731707317073,
"grad_norm": 2.640625,
"learning_rate": 1.240677966101695e-05,
"loss": 1.1703,
"step": 580
},
{
"epoch": 0.4496951219512195,
"grad_norm": 3.09375,
"learning_rate": 1.2237288135593222e-05,
"loss": 1.1566,
"step": 590
},
{
"epoch": 0.4573170731707317,
"grad_norm": 2.578125,
"learning_rate": 1.2067796610169493e-05,
"loss": 1.1623,
"step": 600
},
{
"epoch": 0.4649390243902439,
"grad_norm": 2.734375,
"learning_rate": 1.1898305084745763e-05,
"loss": 1.198,
"step": 610
},
{
"epoch": 0.4725609756097561,
"grad_norm": 2.78125,
"learning_rate": 1.1728813559322034e-05,
"loss": 1.1971,
"step": 620
},
{
"epoch": 0.4801829268292683,
"grad_norm": 2.546875,
"learning_rate": 1.1559322033898306e-05,
"loss": 1.159,
"step": 630
},
{
"epoch": 0.4878048780487805,
"grad_norm": 2.84375,
"learning_rate": 1.1389830508474577e-05,
"loss": 1.184,
"step": 640
},
{
"epoch": 0.4954268292682927,
"grad_norm": 2.578125,
"learning_rate": 1.1220338983050849e-05,
"loss": 1.1859,
"step": 650
},
{
"epoch": 0.5030487804878049,
"grad_norm": 2.609375,
"learning_rate": 1.1050847457627118e-05,
"loss": 1.1434,
"step": 660
},
{
"epoch": 0.510670731707317,
"grad_norm": 2.671875,
"learning_rate": 1.088135593220339e-05,
"loss": 1.1775,
"step": 670
},
{
"epoch": 0.5182926829268293,
"grad_norm": 2.515625,
"learning_rate": 1.0711864406779661e-05,
"loss": 1.15,
"step": 680
},
{
"epoch": 0.5259146341463414,
"grad_norm": 2.609375,
"learning_rate": 1.0542372881355933e-05,
"loss": 1.1613,
"step": 690
},
{
"epoch": 0.5335365853658537,
"grad_norm": 2.5625,
"learning_rate": 1.0372881355932204e-05,
"loss": 1.1602,
"step": 700
},
{
"epoch": 0.5411585365853658,
"grad_norm": 2.671875,
"learning_rate": 1.0203389830508474e-05,
"loss": 1.1879,
"step": 710
},
{
"epoch": 0.5487804878048781,
"grad_norm": 2.625,
"learning_rate": 1.0033898305084746e-05,
"loss": 1.1941,
"step": 720
},
{
"epoch": 0.5564024390243902,
"grad_norm": 2.8125,
"learning_rate": 9.864406779661017e-06,
"loss": 1.1574,
"step": 730
},
{
"epoch": 0.5640243902439024,
"grad_norm": 2.75,
"learning_rate": 9.69491525423729e-06,
"loss": 1.1793,
"step": 740
},
{
"epoch": 0.5716463414634146,
"grad_norm": 2.703125,
"learning_rate": 9.52542372881356e-06,
"loss": 1.1711,
"step": 750
},
{
"epoch": 0.5792682926829268,
"grad_norm": 2.703125,
"learning_rate": 9.355932203389831e-06,
"loss": 1.1875,
"step": 760
},
{
"epoch": 0.586890243902439,
"grad_norm": 2.8125,
"learning_rate": 9.186440677966101e-06,
"loss": 1.19,
"step": 770
},
{
"epoch": 0.5945121951219512,
"grad_norm": 2.9375,
"learning_rate": 9.016949152542374e-06,
"loss": 1.141,
"step": 780
},
{
"epoch": 0.6021341463414634,
"grad_norm": 2.75,
"learning_rate": 8.847457627118646e-06,
"loss": 1.1699,
"step": 790
},
{
"epoch": 0.6097560975609756,
"grad_norm": 2.65625,
"learning_rate": 8.677966101694915e-06,
"loss": 1.1809,
"step": 800
},
{
"epoch": 0.6173780487804879,
"grad_norm": 2.703125,
"learning_rate": 8.508474576271187e-06,
"loss": 1.1824,
"step": 810
},
{
"epoch": 0.625,
"grad_norm": 2.796875,
"learning_rate": 8.338983050847458e-06,
"loss": 1.1729,
"step": 820
},
{
"epoch": 0.6326219512195121,
"grad_norm": 2.765625,
"learning_rate": 8.16949152542373e-06,
"loss": 1.1113,
"step": 830
},
{
"epoch": 0.6402439024390244,
"grad_norm": 2.5,
"learning_rate": 8.000000000000001e-06,
"loss": 1.1434,
"step": 840
},
{
"epoch": 0.6478658536585366,
"grad_norm": 2.515625,
"learning_rate": 7.830508474576271e-06,
"loss": 1.1643,
"step": 850
},
{
"epoch": 0.6554878048780488,
"grad_norm": 2.65625,
"learning_rate": 7.661016949152543e-06,
"loss": 1.1738,
"step": 860
},
{
"epoch": 0.663109756097561,
"grad_norm": 2.609375,
"learning_rate": 7.491525423728814e-06,
"loss": 1.125,
"step": 870
},
{
"epoch": 0.6707317073170732,
"grad_norm": 2.5625,
"learning_rate": 7.3220338983050855e-06,
"loss": 1.1723,
"step": 880
},
{
"epoch": 0.6783536585365854,
"grad_norm": 2.546875,
"learning_rate": 7.152542372881357e-06,
"loss": 1.1617,
"step": 890
},
{
"epoch": 0.6859756097560976,
"grad_norm": 2.6875,
"learning_rate": 6.9830508474576275e-06,
"loss": 1.1445,
"step": 900
},
{
"epoch": 0.6935975609756098,
"grad_norm": 2.546875,
"learning_rate": 6.813559322033899e-06,
"loss": 1.1807,
"step": 910
},
{
"epoch": 0.7012195121951219,
"grad_norm": 2.65625,
"learning_rate": 6.64406779661017e-06,
"loss": 1.1865,
"step": 920
},
{
"epoch": 0.7088414634146342,
"grad_norm": 2.734375,
"learning_rate": 6.474576271186441e-06,
"loss": 1.198,
"step": 930
},
{
"epoch": 0.7164634146341463,
"grad_norm": 2.765625,
"learning_rate": 6.3050847457627125e-06,
"loss": 1.1734,
"step": 940
},
{
"epoch": 0.7240853658536586,
"grad_norm": 2.609375,
"learning_rate": 6.135593220338983e-06,
"loss": 1.1633,
"step": 950
},
{
"epoch": 0.7317073170731707,
"grad_norm": 2.578125,
"learning_rate": 5.9661016949152555e-06,
"loss": 1.1953,
"step": 960
},
{
"epoch": 0.739329268292683,
"grad_norm": 2.5,
"learning_rate": 5.796610169491525e-06,
"loss": 1.1879,
"step": 970
},
{
"epoch": 0.7469512195121951,
"grad_norm": 2.609375,
"learning_rate": 5.6271186440677975e-06,
"loss": 1.1664,
"step": 980
},
{
"epoch": 0.7545731707317073,
"grad_norm": 2.765625,
"learning_rate": 5.457627118644067e-06,
"loss": 1.1789,
"step": 990
},
{
"epoch": 0.7621951219512195,
"grad_norm": 2.890625,
"learning_rate": 5.28813559322034e-06,
"loss": 1.1738,
"step": 1000
},
{
"epoch": 0.7698170731707317,
"grad_norm": 2.6875,
"learning_rate": 5.118644067796611e-06,
"loss": 1.1582,
"step": 1010
},
{
"epoch": 0.7774390243902439,
"grad_norm": 2.9375,
"learning_rate": 4.949152542372882e-06,
"loss": 1.1391,
"step": 1020
},
{
"epoch": 0.7850609756097561,
"grad_norm": 2.578125,
"learning_rate": 4.779661016949153e-06,
"loss": 1.1652,
"step": 1030
},
{
"epoch": 0.7926829268292683,
"grad_norm": 2.703125,
"learning_rate": 4.610169491525424e-06,
"loss": 1.1426,
"step": 1040
},
{
"epoch": 0.8003048780487805,
"grad_norm": 2.71875,
"learning_rate": 4.440677966101695e-06,
"loss": 1.1713,
"step": 1050
},
{
"epoch": 0.8079268292682927,
"grad_norm": 2.515625,
"learning_rate": 4.271186440677967e-06,
"loss": 1.1727,
"step": 1060
},
{
"epoch": 0.8155487804878049,
"grad_norm": 2.59375,
"learning_rate": 4.101694915254237e-06,
"loss": 1.1695,
"step": 1070
},
{
"epoch": 0.823170731707317,
"grad_norm": 2.671875,
"learning_rate": 3.932203389830509e-06,
"loss": 1.1672,
"step": 1080
},
{
"epoch": 0.8307926829268293,
"grad_norm": 2.734375,
"learning_rate": 3.76271186440678e-06,
"loss": 1.2016,
"step": 1090
},
{
"epoch": 0.8384146341463414,
"grad_norm": 2.90625,
"learning_rate": 3.5932203389830512e-06,
"loss": 1.1727,
"step": 1100
},
{
"epoch": 0.8460365853658537,
"grad_norm": 2.84375,
"learning_rate": 3.4237288135593223e-06,
"loss": 1.1799,
"step": 1110
},
{
"epoch": 0.8536585365853658,
"grad_norm": 2.8125,
"learning_rate": 3.2542372881355933e-06,
"loss": 1.1555,
"step": 1120
},
{
"epoch": 0.8612804878048781,
"grad_norm": 2.6875,
"learning_rate": 3.0847457627118648e-06,
"loss": 1.1678,
"step": 1130
},
{
"epoch": 0.8689024390243902,
"grad_norm": 2.59375,
"learning_rate": 2.915254237288136e-06,
"loss": 1.1578,
"step": 1140
},
{
"epoch": 0.8765243902439024,
"grad_norm": 2.375,
"learning_rate": 2.745762711864407e-06,
"loss": 1.1535,
"step": 1150
},
{
"epoch": 0.8841463414634146,
"grad_norm": 2.703125,
"learning_rate": 2.576271186440678e-06,
"loss": 1.1602,
"step": 1160
},
{
"epoch": 0.8917682926829268,
"grad_norm": 2.78125,
"learning_rate": 2.4067796610169493e-06,
"loss": 1.1852,
"step": 1170
},
{
"epoch": 0.899390243902439,
"grad_norm": 2.59375,
"learning_rate": 2.2372881355932204e-06,
"loss": 1.1605,
"step": 1180
},
{
"epoch": 0.9070121951219512,
"grad_norm": 2.59375,
"learning_rate": 2.0677966101694914e-06,
"loss": 1.1451,
"step": 1190
},
{
"epoch": 0.9146341463414634,
"grad_norm": 2.5625,
"learning_rate": 1.8983050847457629e-06,
"loss": 1.1496,
"step": 1200
},
{
"epoch": 0.9222560975609756,
"grad_norm": 2.859375,
"learning_rate": 1.728813559322034e-06,
"loss": 1.1582,
"step": 1210
},
{
"epoch": 0.9298780487804879,
"grad_norm": 2.65625,
"learning_rate": 1.5593220338983054e-06,
"loss": 1.227,
"step": 1220
},
{
"epoch": 0.9375,
"grad_norm": 2.703125,
"learning_rate": 1.3898305084745764e-06,
"loss": 1.1977,
"step": 1230
},
{
"epoch": 0.9451219512195121,
"grad_norm": 2.671875,
"learning_rate": 1.2203389830508477e-06,
"loss": 1.1885,
"step": 1240
},
{
"epoch": 0.9527439024390244,
"grad_norm": 2.6875,
"learning_rate": 1.0508474576271187e-06,
"loss": 1.1549,
"step": 1250
},
{
"epoch": 0.9603658536585366,
"grad_norm": 2.53125,
"learning_rate": 8.813559322033899e-07,
"loss": 1.1781,
"step": 1260
},
{
"epoch": 0.9679878048780488,
"grad_norm": 2.5625,
"learning_rate": 7.118644067796611e-07,
"loss": 1.1957,
"step": 1270
},
{
"epoch": 0.975609756097561,
"grad_norm": 2.640625,
"learning_rate": 5.423728813559322e-07,
"loss": 1.1898,
"step": 1280
},
{
"epoch": 0.9832317073170732,
"grad_norm": 2.625,
"learning_rate": 3.7288135593220347e-07,
"loss": 1.1531,
"step": 1290
},
{
"epoch": 0.9908536585365854,
"grad_norm": 2.6875,
"learning_rate": 2.0338983050847458e-07,
"loss": 1.1773,
"step": 1300
},
{
"epoch": 0.9984756097560976,
"grad_norm": 2.640625,
"learning_rate": 3.3898305084745764e-08,
"loss": 1.0754,
"step": 1310
},
{
"epoch": 1.0,
"step": 1312,
"total_flos": 4.7710598211200614e+17,
"train_loss": 1.2026635146722562,
"train_runtime": 2875.3707,
"train_samples_per_second": 14.591,
"train_steps_per_second": 0.456
}
],
"logging_steps": 10,
"max_steps": 1312,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 4.7710598211200614e+17,
"train_batch_size": 16,
"trial_name": null,
"trial_params": null
}