Files
qwen3-4b-safechain/trainer_state.json

1794 lines
40 KiB
JSON
Raw Permalink Normal View History

{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.0,
"eval_steps": 500,
"global_step": 5000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.008,
"grad_norm": 18.832239151000977,
"learning_rate": 3.8e-07,
"loss": 1.3021,
"step": 20
},
{
"epoch": 0.016,
"grad_norm": 7.525812149047852,
"learning_rate": 7.8e-07,
"loss": 1.2112,
"step": 40
},
{
"epoch": 0.024,
"grad_norm": 2.6391570568084717,
"learning_rate": 1.1800000000000001e-06,
"loss": 1.0085,
"step": 60
},
{
"epoch": 0.032,
"grad_norm": 1.94918692111969,
"learning_rate": 1.5800000000000001e-06,
"loss": 0.9484,
"step": 80
},
{
"epoch": 0.04,
"grad_norm": 1.8315504789352417,
"learning_rate": 1.98e-06,
"loss": 0.9203,
"step": 100
},
{
"epoch": 0.048,
"grad_norm": 2.217707633972168,
"learning_rate": 2.38e-06,
"loss": 0.9135,
"step": 120
},
{
"epoch": 0.056,
"grad_norm": 1.7724204063415527,
"learning_rate": 2.7800000000000005e-06,
"loss": 0.9045,
"step": 140
},
{
"epoch": 0.064,
"grad_norm": 1.609054446220398,
"learning_rate": 3.1800000000000005e-06,
"loss": 0.9032,
"step": 160
},
{
"epoch": 0.072,
"grad_norm": 1.861961007118225,
"learning_rate": 3.58e-06,
"loss": 0.881,
"step": 180
},
{
"epoch": 0.08,
"grad_norm": 1.7415540218353271,
"learning_rate": 3.980000000000001e-06,
"loss": 0.89,
"step": 200
},
{
"epoch": 0.088,
"grad_norm": 1.7969343662261963,
"learning_rate": 4.38e-06,
"loss": 0.8808,
"step": 220
},
{
"epoch": 0.096,
"grad_norm": 1.7322224378585815,
"learning_rate": 4.78e-06,
"loss": 0.8928,
"step": 240
},
{
"epoch": 0.104,
"grad_norm": 1.7264833450317383,
"learning_rate": 5.18e-06,
"loss": 0.8817,
"step": 260
},
{
"epoch": 0.112,
"grad_norm": 1.7735146284103394,
"learning_rate": 5.580000000000001e-06,
"loss": 0.8859,
"step": 280
},
{
"epoch": 0.12,
"grad_norm": 1.9881502389907837,
"learning_rate": 5.98e-06,
"loss": 0.8658,
"step": 300
},
{
"epoch": 0.128,
"grad_norm": 1.8438222408294678,
"learning_rate": 6.380000000000001e-06,
"loss": 0.8688,
"step": 320
},
{
"epoch": 0.136,
"grad_norm": 1.8867048025131226,
"learning_rate": 6.780000000000001e-06,
"loss": 0.8642,
"step": 340
},
{
"epoch": 0.144,
"grad_norm": 1.7361332178115845,
"learning_rate": 7.180000000000001e-06,
"loss": 0.8694,
"step": 360
},
{
"epoch": 0.152,
"grad_norm": 1.804467797279358,
"learning_rate": 7.58e-06,
"loss": 0.8823,
"step": 380
},
{
"epoch": 0.16,
"grad_norm": 1.7290053367614746,
"learning_rate": 7.980000000000002e-06,
"loss": 0.8616,
"step": 400
},
{
"epoch": 0.168,
"grad_norm": 1.7055102586746216,
"learning_rate": 8.380000000000001e-06,
"loss": 0.879,
"step": 420
},
{
"epoch": 0.176,
"grad_norm": 1.5449683666229248,
"learning_rate": 8.78e-06,
"loss": 0.8707,
"step": 440
},
{
"epoch": 0.184,
"grad_norm": 1.8237775564193726,
"learning_rate": 9.180000000000002e-06,
"loss": 0.8734,
"step": 460
},
{
"epoch": 0.192,
"grad_norm": 1.7599401473999023,
"learning_rate": 9.58e-06,
"loss": 0.8724,
"step": 480
},
{
"epoch": 0.2,
"grad_norm": 1.7165312767028809,
"learning_rate": 9.980000000000001e-06,
"loss": 0.8805,
"step": 500
},
{
"epoch": 0.208,
"grad_norm": 1.744134783744812,
"learning_rate": 9.999560138895238e-06,
"loss": 0.8781,
"step": 520
},
{
"epoch": 0.216,
"grad_norm": 1.716552495956421,
"learning_rate": 9.998146822104943e-06,
"loss": 0.8495,
"step": 540
},
{
"epoch": 0.224,
"grad_norm": 1.8554117679595947,
"learning_rate": 9.995759106688394e-06,
"loss": 0.8514,
"step": 560
},
{
"epoch": 0.232,
"grad_norm": 1.7680516242980957,
"learning_rate": 9.992397458135438e-06,
"loss": 0.8619,
"step": 580
},
{
"epoch": 0.24,
"grad_norm": 2.0474295616149902,
"learning_rate": 9.988062531806127e-06,
"loss": 0.8689,
"step": 600
},
{
"epoch": 0.248,
"grad_norm": 1.8329510688781738,
"learning_rate": 9.982755172802933e-06,
"loss": 0.8694,
"step": 620
},
{
"epoch": 0.256,
"grad_norm": 1.7524733543395996,
"learning_rate": 9.976476415806013e-06,
"loss": 0.8604,
"step": 640
},
{
"epoch": 0.264,
"grad_norm": 1.7632076740264893,
"learning_rate": 9.969227484871485e-06,
"loss": 0.8483,
"step": 660
},
{
"epoch": 0.272,
"grad_norm": 1.8328086137771606,
"learning_rate": 9.961009793192793e-06,
"loss": 0.8356,
"step": 680
},
{
"epoch": 0.28,
"grad_norm": 1.5859813690185547,
"learning_rate": 9.951824942825215e-06,
"loss": 0.8524,
"step": 700
},
{
"epoch": 0.288,
"grad_norm": 1.6783335208892822,
"learning_rate": 9.94167472437353e-06,
"loss": 0.8541,
"step": 720
},
{
"epoch": 0.296,
"grad_norm": 1.3135913610458374,
"learning_rate": 9.930561116642936e-06,
"loss": 0.8303,
"step": 740
},
{
"epoch": 0.304,
"grad_norm": 1.8271369934082031,
"learning_rate": 9.918486286253279e-06,
"loss": 0.876,
"step": 760
},
{
"epoch": 0.312,
"grad_norm": 1.706070899963379,
"learning_rate": 9.90545258721667e-06,
"loss": 0.8503,
"step": 780
},
{
"epoch": 0.32,
"grad_norm": 1.7325022220611572,
"learning_rate": 9.891462560478562e-06,
"loss": 0.8481,
"step": 800
},
{
"epoch": 0.328,
"grad_norm": 1.3082691431045532,
"learning_rate": 9.876518933422385e-06,
"loss": 0.837,
"step": 820
},
{
"epoch": 0.336,
"grad_norm": 1.729883074760437,
"learning_rate": 9.860624619337844e-06,
"loss": 0.8564,
"step": 840
},
{
"epoch": 0.344,
"grad_norm": 1.729577898979187,
"learning_rate": 9.843782716852963e-06,
"loss": 0.852,
"step": 860
},
{
"epoch": 0.352,
"grad_norm": 1.7094006538391113,
"learning_rate": 9.825996509330001e-06,
"loss": 0.8368,
"step": 880
},
{
"epoch": 0.36,
"grad_norm": 1.7552564144134521,
"learning_rate": 9.807269464225355e-06,
"loss": 0.842,
"step": 900
},
{
"epoch": 0.368,
"grad_norm": 1.7738618850708008,
"learning_rate": 9.787605232413575e-06,
"loss": 0.8434,
"step": 920
},
{
"epoch": 0.376,
"grad_norm": 1.6761345863342285,
"learning_rate": 9.767007647475618e-06,
"loss": 0.8458,
"step": 940
},
{
"epoch": 0.384,
"grad_norm": 1.6989866495132446,
"learning_rate": 9.745480724951473e-06,
"loss": 0.8429,
"step": 960
},
{
"epoch": 0.392,
"grad_norm": 1.668524980545044,
"learning_rate": 9.723028661557345e-06,
"loss": 0.827,
"step": 980
},
{
"epoch": 0.4,
"grad_norm": 1.8168995380401611,
"learning_rate": 9.699655834367479e-06,
"loss": 0.8311,
"step": 1000
},
{
"epoch": 0.408,
"grad_norm": 1.6532375812530518,
"learning_rate": 9.675366799960842e-06,
"loss": 0.8218,
"step": 1020
},
{
"epoch": 0.416,
"grad_norm": 1.556692361831665,
"learning_rate": 9.650166293532822e-06,
"loss": 0.832,
"step": 1040
},
{
"epoch": 0.424,
"grad_norm": 1.3959892988204956,
"learning_rate": 9.624059227972077e-06,
"loss": 0.8501,
"step": 1060
},
{
"epoch": 0.432,
"grad_norm": 1.5428262948989868,
"learning_rate": 9.597050692902765e-06,
"loss": 0.8258,
"step": 1080
},
{
"epoch": 0.44,
"grad_norm": 1.657296895980835,
"learning_rate": 9.569145953692316e-06,
"loss": 0.8516,
"step": 1100
},
{
"epoch": 0.448,
"grad_norm": 1.4994111061096191,
"learning_rate": 9.540350450424927e-06,
"loss": 0.8351,
"step": 1120
},
{
"epoch": 0.456,
"grad_norm": 1.7212347984313965,
"learning_rate": 9.510669796841014e-06,
"loss": 0.8346,
"step": 1140
},
{
"epoch": 0.464,
"grad_norm": 1.5698916912078857,
"learning_rate": 9.480109779242805e-06,
"loss": 0.8399,
"step": 1160
},
{
"epoch": 0.472,
"grad_norm": 1.6550339460372925,
"learning_rate": 9.448676355366282e-06,
"loss": 0.8477,
"step": 1180
},
{
"epoch": 0.48,
"grad_norm": 1.4831129312515259,
"learning_rate": 9.41637565321971e-06,
"loss": 0.8382,
"step": 1200
},
{
"epoch": 0.488,
"grad_norm": 1.7176513671875,
"learning_rate": 9.383213969888972e-06,
"loss": 0.8182,
"step": 1220
},
{
"epoch": 0.496,
"grad_norm": 1.6853461265563965,
"learning_rate": 9.349197770309942e-06,
"loss": 0.8161,
"step": 1240
},
{
"epoch": 0.504,
"grad_norm": 1.6701298952102661,
"learning_rate": 9.314333686008125e-06,
"loss": 0.8179,
"step": 1260
},
{
"epoch": 0.512,
"grad_norm": 1.5644961595535278,
"learning_rate": 9.278628513805838e-06,
"loss": 0.8348,
"step": 1280
},
{
"epoch": 0.52,
"grad_norm": 1.7242437601089478,
"learning_rate": 9.242089214497146e-06,
"loss": 0.8421,
"step": 1300
},
{
"epoch": 0.528,
"grad_norm": 1.5268667936325073,
"learning_rate": 9.204722911490847e-06,
"loss": 0.8266,
"step": 1320
},
{
"epoch": 0.536,
"grad_norm": 1.8252116441726685,
"learning_rate": 9.16653688942175e-06,
"loss": 0.8465,
"step": 1340
},
{
"epoch": 0.544,
"grad_norm": 1.5816881656646729,
"learning_rate": 9.12753859273052e-06,
"loss": 0.8253,
"step": 1360
},
{
"epoch": 0.552,
"grad_norm": 1.7188431024551392,
"learning_rate": 9.087735624212365e-06,
"loss": 0.827,
"step": 1380
},
{
"epoch": 0.56,
"grad_norm": 1.3610879182815552,
"learning_rate": 9.047135743534866e-06,
"loss": 0.83,
"step": 1400
},
{
"epoch": 0.568,
"grad_norm": 1.542251706123352,
"learning_rate": 9.005746865725206e-06,
"loss": 0.8083,
"step": 1420
},
{
"epoch": 0.576,
"grad_norm": 1.7079123258590698,
"learning_rate": 8.963577059627117e-06,
"loss": 0.8223,
"step": 1440
},
{
"epoch": 0.584,
"grad_norm": 1.7531414031982422,
"learning_rate": 8.920634546327857e-06,
"loss": 0.8317,
"step": 1460
},
{
"epoch": 0.592,
"grad_norm": 1.6270157098770142,
"learning_rate": 8.87692769755548e-06,
"loss": 0.8232,
"step": 1480
},
{
"epoch": 0.6,
"grad_norm": 1.5250872373580933,
"learning_rate": 8.83246503404675e-06,
"loss": 0.8322,
"step": 1500
},
{
"epoch": 0.608,
"grad_norm": 1.6226458549499512,
"learning_rate": 8.78725522388602e-06,
"loss": 0.8306,
"step": 1520
},
{
"epoch": 0.616,
"grad_norm": 1.5596694946289062,
"learning_rate": 8.741307080815357e-06,
"loss": 0.816,
"step": 1540
},
{
"epoch": 0.624,
"grad_norm": 1.5898826122283936,
"learning_rate": 8.694629562516295e-06,
"loss": 0.8267,
"step": 1560
},
{
"epoch": 0.632,
"grad_norm": 1.604196548461914,
"learning_rate": 8.647231768863513e-06,
"loss": 0.8065,
"step": 1580
},
{
"epoch": 0.64,
"grad_norm": 1.615822196006775,
"learning_rate": 8.599122940150795e-06,
"loss": 0.8195,
"step": 1600
},
{
"epoch": 0.648,
"grad_norm": 1.4879181385040283,
"learning_rate": 8.550312455289624e-06,
"loss": 0.8092,
"step": 1620
},
{
"epoch": 0.656,
"grad_norm": 1.5847649574279785,
"learning_rate": 8.500809829980734e-06,
"loss": 0.8105,
"step": 1640
},
{
"epoch": 0.664,
"grad_norm": 1.6388778686523438,
"learning_rate": 8.450624714859016e-06,
"loss": 0.8035,
"step": 1660
},
{
"epoch": 0.672,
"grad_norm": 1.6703004837036133,
"learning_rate": 8.399766893612096e-06,
"loss": 0.8179,
"step": 1680
},
{
"epoch": 0.68,
"grad_norm": 1.573598027229309,
"learning_rate": 8.348246281072998e-06,
"loss": 0.7876,
"step": 1700
},
{
"epoch": 0.688,
"grad_norm": 1.6595710515975952,
"learning_rate": 8.296072921287217e-06,
"loss": 0.8193,
"step": 1720
},
{
"epoch": 0.696,
"grad_norm": 1.6333134174346924,
"learning_rate": 8.243256985554622e-06,
"loss": 0.8192,
"step": 1740
},
{
"epoch": 0.704,
"grad_norm": 1.6130130290985107,
"learning_rate": 8.189808770446528e-06,
"loss": 0.8128,
"step": 1760
},
{
"epoch": 0.712,
"grad_norm": 1.5962647199630737,
"learning_rate": 8.135738695798377e-06,
"loss": 0.8118,
"step": 1780
},
{
"epoch": 0.72,
"grad_norm": 1.5578866004943848,
"learning_rate": 8.081057302678352e-06,
"loss": 0.8081,
"step": 1800
},
{
"epoch": 0.728,
"grad_norm": 1.6597858667373657,
"learning_rate": 8.02577525133239e-06,
"loss": 0.8222,
"step": 1820
},
{
"epoch": 0.736,
"grad_norm": 1.7409942150115967,
"learning_rate": 7.969903319105935e-06,
"loss": 0.8263,
"step": 1840
},
{
"epoch": 0.744,
"grad_norm": 1.7179594039916992,
"learning_rate": 7.913452398342882e-06,
"loss": 0.8154,
"step": 1860
},
{
"epoch": 0.752,
"grad_norm": 1.6091378927230835,
"learning_rate": 7.856433494262078e-06,
"loss": 0.8073,
"step": 1880
},
{
"epoch": 0.76,
"grad_norm": 1.6636652946472168,
"learning_rate": 7.798857722811857e-06,
"loss": 0.8127,
"step": 1900
},
{
"epoch": 0.768,
"grad_norm": 1.6128122806549072,
"learning_rate": 7.740736308502939e-06,
"loss": 0.7946,
"step": 1920
},
{
"epoch": 0.776,
"grad_norm": 1.8580641746520996,
"learning_rate": 7.682080582220206e-06,
"loss": 0.8081,
"step": 1940
},
{
"epoch": 0.784,
"grad_norm": 1.536949872970581,
"learning_rate": 7.622901979013717e-06,
"loss": 0.7946,
"step": 1960
},
{
"epoch": 0.792,
"grad_norm": 1.6701791286468506,
"learning_rate": 7.563212035869426e-06,
"loss": 0.8156,
"step": 1980
},
{
"epoch": 0.8,
"grad_norm": 1.5435974597930908,
"learning_rate": 7.503022389460014e-06,
"loss": 0.8137,
"step": 2000
},
{
"epoch": 0.808,
"grad_norm": 1.547313928604126,
"learning_rate": 7.44234477387631e-06,
"loss": 0.8045,
"step": 2020
},
{
"epoch": 0.816,
"grad_norm": 1.7339191436767578,
"learning_rate": 7.381191018339697e-06,
"loss": 0.7921,
"step": 2040
},
{
"epoch": 0.824,
"grad_norm": 1.7117959260940552,
"learning_rate": 7.319573044895986e-06,
"loss": 0.7882,
"step": 2060
},
{
"epoch": 0.832,
"grad_norm": 1.6268044710159302,
"learning_rate": 7.257502866091192e-06,
"loss": 0.8121,
"step": 2080
},
{
"epoch": 0.84,
"grad_norm": 1.5561449527740479,
"learning_rate": 7.194992582629654e-06,
"loss": 0.8066,
"step": 2100
},
{
"epoch": 0.848,
"grad_norm": 1.6417181491851807,
"learning_rate": 7.1320543810149945e-06,
"loss": 0.7999,
"step": 2120
},
{
"epoch": 0.856,
"grad_norm": 1.4862862825393677,
"learning_rate": 7.0687005311743195e-06,
"loss": 0.8009,
"step": 2140
},
{
"epoch": 0.864,
"grad_norm": 1.6011143922805786,
"learning_rate": 7.0049433840661875e-06,
"loss": 0.8059,
"step": 2160
},
{
"epoch": 0.872,
"grad_norm": 1.5853888988494873,
"learning_rate": 6.940795369272754e-06,
"loss": 0.8054,
"step": 2180
},
{
"epoch": 0.88,
"grad_norm": 1.4722031354904175,
"learning_rate": 6.876268992576605e-06,
"loss": 0.7843,
"step": 2200
},
{
"epoch": 0.888,
"grad_norm": 1.6270049810409546,
"learning_rate": 6.811376833522729e-06,
"loss": 0.7797,
"step": 2220
},
{
"epoch": 0.896,
"grad_norm": 1.553013801574707,
"learning_rate": 6.746131542966112e-06,
"loss": 0.8049,
"step": 2240
},
{
"epoch": 0.904,
"grad_norm": 1.593279480934143,
"learning_rate": 6.680545840605423e-06,
"loss": 0.8001,
"step": 2260
},
{
"epoch": 0.912,
"grad_norm": 1.6089750528335571,
"learning_rate": 6.614632512503289e-06,
"loss": 0.7969,
"step": 2280
},
{
"epoch": 0.92,
"grad_norm": 1.635054349899292,
"learning_rate": 6.548404408593622e-06,
"loss": 0.807,
"step": 2300
},
{
"epoch": 0.928,
"grad_norm": 1.584202766418457,
"learning_rate": 6.481874440176506e-06,
"loss": 0.7973,
"step": 2320
},
{
"epoch": 0.936,
"grad_norm": 1.6124290227890015,
"learning_rate": 6.415055577401101e-06,
"loss": 0.7843,
"step": 2340
},
{
"epoch": 0.944,
"grad_norm": 1.6321603059768677,
"learning_rate": 6.3479608467371055e-06,
"loss": 0.783,
"step": 2360
},
{
"epoch": 0.952,
"grad_norm": 1.6267094612121582,
"learning_rate": 6.280603328435199e-06,
"loss": 0.7784,
"step": 2380
},
{
"epoch": 0.96,
"grad_norm": 1.692747712135315,
"learning_rate": 6.212996153977038e-06,
"loss": 0.797,
"step": 2400
},
{
"epoch": 0.968,
"grad_norm": 1.5670863389968872,
"learning_rate": 6.145152503515239e-06,
"loss": 0.7918,
"step": 2420
},
{
"epoch": 0.976,
"grad_norm": 1.630425214767456,
"learning_rate": 6.077085603303883e-06,
"loss": 0.7972,
"step": 2440
},
{
"epoch": 0.984,
"grad_norm": 1.1841243505477905,
"learning_rate": 6.008808723120035e-06,
"loss": 0.7931,
"step": 2460
},
{
"epoch": 0.992,
"grad_norm": 1.5890851020812988,
"learning_rate": 5.94033517367677e-06,
"loss": 0.7893,
"step": 2480
},
{
"epoch": 1.0,
"grad_norm": 1.592110514640808,
"learning_rate": 5.871678304028224e-06,
"loss": 0.7937,
"step": 2500
},
{
"epoch": 1.008,
"grad_norm": 1.6044059991836548,
"learning_rate": 5.802851498967173e-06,
"loss": 0.6091,
"step": 2520
},
{
"epoch": 1.016,
"grad_norm": 1.825309157371521,
"learning_rate": 5.733868176415633e-06,
"loss": 0.6033,
"step": 2540
},
{
"epoch": 1.024,
"grad_norm": 1.6619549989700317,
"learning_rate": 5.6647417848090225e-06,
"loss": 0.6056,
"step": 2560
},
{
"epoch": 1.032,
"grad_norm": 1.6869597434997559,
"learning_rate": 5.59548580047435e-06,
"loss": 0.6035,
"step": 2580
},
{
"epoch": 1.04,
"grad_norm": 1.6975550651550293,
"learning_rate": 5.526113725002984e-06,
"loss": 0.6038,
"step": 2600
},
{
"epoch": 1.048,
"grad_norm": 1.7238929271697998,
"learning_rate": 5.456639082618489e-06,
"loss": 0.6032,
"step": 2620
},
{
"epoch": 1.056,
"grad_norm": 1.625978946685791,
"learning_rate": 5.3870754175400595e-06,
"loss": 0.593,
"step": 2640
},
{
"epoch": 1.064,
"grad_norm": 1.6855403184890747,
"learning_rate": 5.317436291342031e-06,
"loss": 0.5998,
"step": 2660
},
{
"epoch": 1.072,
"grad_norm": 1.610180139541626,
"learning_rate": 5.247735280310041e-06,
"loss": 0.6009,
"step": 2680
},
{
"epoch": 1.08,
"grad_norm": 1.8414939641952515,
"learning_rate": 5.177985972794293e-06,
"loss": 0.6035,
"step": 2700
},
{
"epoch": 1.088,
"grad_norm": 1.8488266468048096,
"learning_rate": 5.1082019665604895e-06,
"loss": 0.5954,
"step": 2720
},
{
"epoch": 1.096,
"grad_norm": 1.7618485689163208,
"learning_rate": 5.038396866138915e-06,
"loss": 0.6085,
"step": 2740
},
{
"epoch": 1.104,
"grad_norm": 1.967399001121521,
"learning_rate": 4.968584280172206e-06,
"loss": 0.5875,
"step": 2760
},
{
"epoch": 1.112,
"grad_norm": 1.7717429399490356,
"learning_rate": 4.898777818762325e-06,
"loss": 0.599,
"step": 2780
},
{
"epoch": 1.12,
"grad_norm": 1.3948484659194946,
"learning_rate": 4.828991090817238e-06,
"loss": 0.602,
"step": 2800
},
{
"epoch": 1.1280000000000001,
"grad_norm": 1.8133684396743774,
"learning_rate": 4.759237701397831e-06,
"loss": 0.6088,
"step": 2820
},
{
"epoch": 1.1360000000000001,
"grad_norm": 1.8852509260177612,
"learning_rate": 4.689531249065581e-06,
"loss": 0.5926,
"step": 2840
},
{
"epoch": 1.144,
"grad_norm": 1.783538579940796,
"learning_rate": 4.619885323231484e-06,
"loss": 0.616,
"step": 2860
},
{
"epoch": 1.152,
"grad_norm": 1.8637661933898926,
"learning_rate": 4.5503135015067815e-06,
"loss": 0.603,
"step": 2880
},
{
"epoch": 1.16,
"grad_norm": 1.6997549533843994,
"learning_rate": 4.4808293470559645e-06,
"loss": 0.6091,
"step": 2900
},
{
"epoch": 1.168,
"grad_norm": 1.7715973854064941,
"learning_rate": 4.4114464059526185e-06,
"loss": 0.5878,
"step": 2920
},
{
"epoch": 1.176,
"grad_norm": 1.7021915912628174,
"learning_rate": 4.342178204538588e-06,
"loss": 0.6124,
"step": 2940
},
{
"epoch": 1.184,
"grad_norm": 1.739148497581482,
"learning_rate": 4.273038246786986e-06,
"loss": 0.6061,
"step": 2960
},
{
"epoch": 1.192,
"grad_norm": 1.7212711572647095,
"learning_rate": 4.204040011669567e-06,
"loss": 0.5879,
"step": 2980
},
{
"epoch": 1.2,
"grad_norm": 1.6656508445739746,
"learning_rate": 4.135196950528982e-06,
"loss": 0.6107,
"step": 3000
},
{
"epoch": 1.208,
"grad_norm": 1.894425392150879,
"learning_rate": 4.066522484456406e-06,
"loss": 0.5969,
"step": 3020
},
{
"epoch": 1.216,
"grad_norm": 1.7288280725479126,
"learning_rate": 3.9980300016750696e-06,
"loss": 0.6048,
"step": 3040
},
{
"epoch": 1.224,
"grad_norm": 1.5493868589401245,
"learning_rate": 3.9297328549302e-06,
"loss": 0.5958,
"step": 3060
},
{
"epoch": 1.232,
"grad_norm": 1.8579810857772827,
"learning_rate": 3.86164435888588e-06,
"loss": 0.5977,
"step": 3080
},
{
"epoch": 1.24,
"grad_norm": 1.703183650970459,
"learning_rate": 3.793777787529325e-06,
"loss": 0.5965,
"step": 3100
},
{
"epoch": 1.248,
"grad_norm": 1.7419129610061646,
"learning_rate": 3.7261463715830902e-06,
"loss": 0.5999,
"step": 3120
},
{
"epoch": 1.256,
"grad_norm": 1.8387292623519897,
"learning_rate": 3.6587632959257168e-06,
"loss": 0.603,
"step": 3140
},
{
"epoch": 1.264,
"grad_norm": 1.6103179454803467,
"learning_rate": 3.5916416970213173e-06,
"loss": 0.6023,
"step": 3160
},
{
"epoch": 1.272,
"grad_norm": 1.7035914659500122,
"learning_rate": 3.524794660358593e-06,
"loss": 0.5961,
"step": 3180
},
{
"epoch": 1.28,
"grad_norm": 1.745127558708191,
"learning_rate": 3.4582352178997937e-06,
"loss": 0.6055,
"step": 3200
},
{
"epoch": 1.288,
"grad_norm": 1.7260342836380005,
"learning_rate": 3.3919763455401016e-06,
"loss": 0.5978,
"step": 3220
},
{
"epoch": 1.296,
"grad_norm": 1.7544012069702148,
"learning_rate": 3.3260309605779717e-06,
"loss": 0.5982,
"step": 3240
},
{
"epoch": 1.304,
"grad_norm": 1.7547940015792847,
"learning_rate": 3.260411919196866e-06,
"loss": 0.5912,
"step": 3260
},
{
"epoch": 1.312,
"grad_norm": 1.6964139938354492,
"learning_rate": 3.195132013958918e-06,
"loss": 0.6008,
"step": 3280
},
{
"epoch": 1.32,
"grad_norm": 1.7409098148345947,
"learning_rate": 3.130203971310999e-06,
"loss": 0.587,
"step": 3300
},
{
"epoch": 1.328,
"grad_norm": 1.9034337997436523,
"learning_rate": 3.0656404491036696e-06,
"loss": 0.6021,
"step": 3320
},
{
"epoch": 1.336,
"grad_norm": 1.7063469886779785,
"learning_rate": 3.001454034123512e-06,
"loss": 0.5959,
"step": 3340
},
{
"epoch": 1.3439999999999999,
"grad_norm": 1.6503005027770996,
"learning_rate": 2.9376572396393047e-06,
"loss": 0.5882,
"step": 3360
},
{
"epoch": 1.3519999999999999,
"grad_norm": 1.6911360025405884,
"learning_rate": 2.874262502962537e-06,
"loss": 0.5898,
"step": 3380
},
{
"epoch": 1.3599999999999999,
"grad_norm": 1.6424927711486816,
"learning_rate": 2.811282183022736e-06,
"loss": 0.5947,
"step": 3400
},
{
"epoch": 1.3679999999999999,
"grad_norm": 1.7646067142486572,
"learning_rate": 2.7487285579580635e-06,
"loss": 0.5936,
"step": 3420
},
{
"epoch": 1.376,
"grad_norm": 1.7227609157562256,
"learning_rate": 2.686613822721666e-06,
"loss": 0.5905,
"step": 3440
},
{
"epoch": 1.384,
"grad_norm": 1.89302396774292,
"learning_rate": 2.6249500867042523e-06,
"loss": 0.5845,
"step": 3460
},
{
"epoch": 1.392,
"grad_norm": 1.620714783668518,
"learning_rate": 2.5637493713733376e-06,
"loss": 0.5971,
"step": 3480
},
{
"epoch": 1.4,
"grad_norm": 1.671188235282898,
"learning_rate": 2.5030236079296443e-06,
"loss": 0.5861,
"step": 3500
},
{
"epoch": 1.408,
"grad_norm": 1.7739337682724,
"learning_rate": 2.442784634981071e-06,
"loss": 0.6021,
"step": 3520
},
{
"epoch": 1.416,
"grad_norm": 1.6397454738616943,
"learning_rate": 2.3830441962347528e-06,
"loss": 0.6008,
"step": 3540
},
{
"epoch": 1.424,
"grad_norm": 1.6463392972946167,
"learning_rate": 2.323813938207593e-06,
"loss": 0.5791,
"step": 3560
},
{
"epoch": 1.432,
"grad_norm": 1.6889275312423706,
"learning_rate": 2.265105407955752e-06,
"loss": 0.5907,
"step": 3580
},
{
"epoch": 1.44,
"grad_norm": 1.8393819332122803,
"learning_rate": 2.2069300508235273e-06,
"loss": 0.6024,
"step": 3600
},
{
"epoch": 1.448,
"grad_norm": 1.7592689990997314,
"learning_rate": 2.14929920821206e-06,
"loss": 0.6016,
"step": 3620
},
{
"epoch": 1.456,
"grad_norm": 1.5433863401412964,
"learning_rate": 2.0922241153683064e-06,
"loss": 0.5918,
"step": 3640
},
{
"epoch": 1.464,
"grad_norm": 1.6745505332946777,
"learning_rate": 2.035715899194704e-06,
"loss": 0.5887,
"step": 3660
},
{
"epoch": 1.472,
"grad_norm": 1.7321064472198486,
"learning_rate": 1.979785576079961e-06,
"loss": 0.5823,
"step": 3680
},
{
"epoch": 1.48,
"grad_norm": 1.8335254192352295,
"learning_rate": 1.9244440497513895e-06,
"loss": 0.5923,
"step": 3700
},
{
"epoch": 1.488,
"grad_norm": 1.6677194833755493,
"learning_rate": 1.8697021091491991e-06,
"loss": 0.5828,
"step": 3720
},
{
"epoch": 1.496,
"grad_norm": 1.756002426147461,
"learning_rate": 1.8155704263231777e-06,
"loss": 0.605,
"step": 3740
},
{
"epoch": 1.504,
"grad_norm": 1.7666428089141846,
"learning_rate": 1.762059554352143e-06,
"loss": 0.5957,
"step": 3760
},
{
"epoch": 1.512,
"grad_norm": 1.7919585704803467,
"learning_rate": 1.709179925286617e-06,
"loss": 0.5861,
"step": 3780
},
{
"epoch": 1.52,
"grad_norm": 1.6821610927581787,
"learning_rate": 1.6569418481150596e-06,
"loss": 0.5832,
"step": 3800
},
{
"epoch": 1.528,
"grad_norm": 1.69204580783844,
"learning_rate": 1.605355506754121e-06,
"loss": 0.5985,
"step": 3820
},
{
"epoch": 1.536,
"grad_norm": 1.9063982963562012,
"learning_rate": 1.554430958063259e-06,
"loss": 0.5742,
"step": 3840
},
{
"epoch": 1.544,
"grad_norm": 1.7453293800354004,
"learning_rate": 1.5041781298841424e-06,
"loss": 0.5982,
"step": 3860
},
{
"epoch": 1.552,
"grad_norm": 1.663316249847412,
"learning_rate": 1.4546068191051988e-06,
"loss": 0.5991,
"step": 3880
},
{
"epoch": 1.56,
"grad_norm": 1.6827841997146606,
"learning_rate": 1.4057266897516842e-06,
"loss": 0.5861,
"step": 3900
},
{
"epoch": 1.568,
"grad_norm": 1.8167757987976074,
"learning_rate": 1.3575472711016634e-06,
"loss": 0.5835,
"step": 3920
},
{
"epoch": 1.576,
"grad_norm": 1.6353213787078857,
"learning_rate": 1.3100779558282673e-06,
"loss": 0.5843,
"step": 3940
},
{
"epoch": 1.584,
"grad_norm": 1.572274923324585,
"learning_rate": 1.2633279981685608e-06,
"loss": 0.5723,
"step": 3960
},
{
"epoch": 1.592,
"grad_norm": 1.7011008262634277,
"learning_rate": 1.217306512119425e-06,
"loss": 0.5701,
"step": 3980
},
{
"epoch": 1.6,
"grad_norm": 1.8143210411071777,
"learning_rate": 1.1720224696607474e-06,
"loss": 0.5952,
"step": 4000
},
{
"epoch": 1.608,
"grad_norm": 2.021284818649292,
"learning_rate": 1.1274846990063314e-06,
"loss": 0.5852,
"step": 4020
},
{
"epoch": 1.616,
"grad_norm": 1.6640852689743042,
"learning_rate": 1.0837018828828133e-06,
"loss": 0.6007,
"step": 4040
},
{
"epoch": 1.624,
"grad_norm": 1.7076753377914429,
"learning_rate": 1.0406825568369478e-06,
"loss": 0.5916,
"step": 4060
},
{
"epoch": 1.6320000000000001,
"grad_norm": 1.653029441833496,
"learning_rate": 9.984351075715848e-07,
"loss": 0.5828,
"step": 4080
},
{
"epoch": 1.6400000000000001,
"grad_norm": 1.7047526836395264,
"learning_rate": 9.569677713106673e-07,
"loss": 0.5955,
"step": 4100
},
{
"epoch": 1.6480000000000001,
"grad_norm": 1.8480520248413086,
"learning_rate": 9.162886321935632e-07,
"loss": 0.5891,
"step": 4120
},
{
"epoch": 1.6560000000000001,
"grad_norm": 1.7753719091415405,
"learning_rate": 8.764056206990446e-07,
"loss": 0.5837,
"step": 4140
},
{
"epoch": 1.6640000000000001,
"grad_norm": 1.2343868017196655,
"learning_rate": 8.373265120992252e-07,
"loss": 0.5786,
"step": 4160
},
{
"epoch": 1.6720000000000002,
"grad_norm": 1.7852405309677124,
"learning_rate": 7.990589249437591e-07,
"loss": 0.585,
"step": 4180
},
{
"epoch": 1.6800000000000002,
"grad_norm": 1.5402637720108032,
"learning_rate": 7.61610319574585e-07,
"loss": 0.5842,
"step": 4200
},
{
"epoch": 1.688,
"grad_norm": 1.6950104236602783,
"learning_rate": 7.249879966715174e-07,
"loss": 0.6006,
"step": 4220
},
{
"epoch": 1.696,
"grad_norm": 1.5620394945144653,
"learning_rate": 6.891990958289724e-07,
"loss": 0.5933,
"step": 4240
},
{
"epoch": 1.704,
"grad_norm": 1.7696399688720703,
"learning_rate": 6.542505941640803e-07,
"loss": 0.5783,
"step": 4260
},
{
"epoch": 1.712,
"grad_norm": 1.7816462516784668,
"learning_rate": 6.201493049564883e-07,
"loss": 0.5902,
"step": 4280
},
{
"epoch": 1.72,
"grad_norm": 1.7035921812057495,
"learning_rate": 5.869018763200929e-07,
"loss": 0.5896,
"step": 4300
},
{
"epoch": 1.728,
"grad_norm": 1.6429486274719238,
"learning_rate": 5.545147899069836e-07,
"loss": 0.5918,
"step": 4320
},
{
"epoch": 1.736,
"grad_norm": 1.9174542427062988,
"learning_rate": 5.229943596438297e-07,
"loss": 0.5899,
"step": 4340
},
{
"epoch": 1.744,
"grad_norm": 1.6559209823608398,
"learning_rate": 4.92346730500966e-07,
"loss": 0.6075,
"step": 4360
},
{
"epoch": 1.752,
"grad_norm": 1.7558505535125732,
"learning_rate": 4.625778772944156e-07,
"loss": 0.5742,
"step": 4380
},
{
"epoch": 1.76,
"grad_norm": 1.6096889972686768,
"learning_rate": 4.33693603521097e-07,
"loss": 0.591,
"step": 4400
},
{
"epoch": 1.768,
"grad_norm": 1.9347445964813232,
"learning_rate": 4.056995402274122e-07,
"loss": 0.5884,
"step": 4420
},
{
"epoch": 1.776,
"grad_norm": 1.515364408493042,
"learning_rate": 3.7860114491147017e-07,
"loss": 0.5857,
"step": 4440
},
{
"epoch": 1.784,
"grad_norm": 1.7330504655838013,
"learning_rate": 3.524037004591274e-07,
"loss": 0.5812,
"step": 4460
},
{
"epoch": 1.792,
"grad_norm": 1.6101831197738647,
"learning_rate": 3.271123141140886e-07,
"loss": 0.5973,
"step": 4480
},
{
"epoch": 1.8,
"grad_norm": 1.7717736959457397,
"learning_rate": 3.027319164822329e-07,
"loss": 0.5982,
"step": 4500
},
{
"epoch": 1.808,
"grad_norm": 1.8308385610580444,
"learning_rate": 2.792672605703867e-07,
"loss": 0.5888,
"step": 4520
},
{
"epoch": 1.8159999999999998,
"grad_norm": 1.861782431602478,
"learning_rate": 2.5672292085971276e-07,
"loss": 0.5902,
"step": 4540
},
{
"epoch": 1.8239999999999998,
"grad_norm": 1.5745478868484497,
"learning_rate": 2.351032924139063e-07,
"loss": 0.597,
"step": 4560
},
{
"epoch": 1.8319999999999999,
"grad_norm": 1.7998218536376953,
"learning_rate": 2.1441259002236924e-07,
"loss": 0.5907,
"step": 4580
},
{
"epoch": 1.8399999999999999,
"grad_norm": 1.588487148284912,
"learning_rate": 1.9465484737853092e-07,
"loss": 0.5953,
"step": 4600
},
{
"epoch": 1.8479999999999999,
"grad_norm": 1.7415562868118286,
"learning_rate": 1.758339162934658e-07,
"loss": 0.5664,
"step": 4620
},
{
"epoch": 1.8559999999999999,
"grad_norm": 1.7705098390579224,
"learning_rate": 1.5795346594498162e-07,
"loss": 0.5895,
"step": 4640
},
{
"epoch": 1.8639999999999999,
"grad_norm": 1.7154279947280884,
"learning_rate": 1.4101698216230254e-07,
"loss": 0.5906,
"step": 4660
},
{
"epoch": 1.8719999999999999,
"grad_norm": 1.7364534139633179,
"learning_rate": 1.2502776674649776e-07,
"loss": 0.5729,
"step": 4680
},
{
"epoch": 1.88,
"grad_norm": 1.724892258644104,
"learning_rate": 1.0998893682679479e-07,
"loss": 0.5848,
"step": 4700
},
{
"epoch": 1.888,
"grad_norm": 1.6230823993682861,
"learning_rate": 9.590342425288446e-08,
"loss": 0.5693,
"step": 4720
},
{
"epoch": 1.896,
"grad_norm": 1.8447569608688354,
"learning_rate": 8.277397502335194e-08,
"loss": 0.5706,
"step": 4740
},
{
"epoch": 1.904,
"grad_norm": 1.7180930376052856,
"learning_rate": 7.060314875033836e-08,
"loss": 0.5884,
"step": 4760
},
{
"epoch": 1.912,
"grad_norm": 1.6014764308929443,
"learning_rate": 5.939331816054161e-08,
"loss": 0.5675,
"step": 4780
},
{
"epoch": 1.92,
"grad_norm": 1.7394343614578247,
"learning_rate": 4.914666863264528e-08,
"loss": 0.6013,
"step": 4800
},
{
"epoch": 1.928,
"grad_norm": 1.8303427696228027,
"learning_rate": 3.986519777127884e-08,
"loss": 0.5873,
"step": 4820
},
{
"epoch": 1.936,
"grad_norm": 1.592680811882019,
"learning_rate": 3.1550715017575895e-08,
"loss": 0.5846,
"step": 4840
},
{
"epoch": 1.944,
"grad_norm": 1.7676323652267456,
"learning_rate": 2.4204841296424086e-08,
"loss": 0.5773,
"step": 4860
},
{
"epoch": 1.952,
"grad_norm": 1.834705114364624,
"learning_rate": 1.7829008700460116e-08,
"loss": 0.5826,
"step": 4880
},
{
"epoch": 1.96,
"grad_norm": 1.8008114099502563,
"learning_rate": 1.2424460210881394e-08,
"loss": 0.5787,
"step": 4900
},
{
"epoch": 1.968,
"grad_norm": 1.7017130851745605,
"learning_rate": 7.992249455124889e-09,
"loss": 0.5897,
"step": 4920
},
{
"epoch": 1.976,
"grad_norm": 1.8904122114181519,
"learning_rate": 4.533240501459202e-09,
"loss": 0.5814,
"step": 4940
},
{
"epoch": 1.984,
"grad_norm": 1.7103095054626465,
"learning_rate": 2.0481076905332074e-09,
"loss": 0.5906,
"step": 4960
},
{
"epoch": 1.992,
"grad_norm": 1.902403712272644,
"learning_rate": 5.373355039128836e-10,
"loss": 0.5983,
"step": 4980
},
{
"epoch": 2.0,
"grad_norm": 1.680394172668457,
"learning_rate": 1.2184696296380083e-12,
"loss": 0.5883,
"step": 5000
},
{
"epoch": 2.0,
"step": 5000,
"total_flos": 2.46059940019431e+18,
"train_loss": 0.7171756315231324,
"train_runtime": 37151.4848,
"train_samples_per_second": 2.153,
"train_steps_per_second": 0.135
}
],
"logging_steps": 20,
"max_steps": 5000,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"save_steps": 5000,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.46059940019431e+18,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}