{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 30,
  "global_step": 3128,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.00128,
      "grad_norm": 5.575794696807861,
      "learning_rate": 0.0,
      "loss": 1.0255,
      "step": 1
    },
    {
      "epoch": 0.00256,
      "grad_norm": 5.579328536987305,
      "learning_rate": 3.3898305084745766e-07,
      "loss": 0.9787,
      "step": 2
    },
    {
      "epoch": 0.00384,
      "grad_norm": 6.020977973937988,
      "learning_rate": 6.779661016949153e-07,
      "loss": 1.014,
      "step": 3
    },
    {
      "epoch": 0.00512,
      "grad_norm": 5.791597366333008,
      "learning_rate": 1.016949152542373e-06,
      "loss": 0.9377,
      "step": 4
    },
    {
      "epoch": 0.0064,
      "grad_norm": 5.563808441162109,
      "learning_rate": 1.3559322033898307e-06,
      "loss": 0.9765,
      "step": 5
    },
    {
      "epoch": 0.00768,
      "grad_norm": 5.490054607391357,
      "learning_rate": 1.6949152542372882e-06,
      "loss": 0.9175,
      "step": 6
    },
    {
      "epoch": 0.00896,
      "grad_norm": 4.674930572509766,
      "learning_rate": 2.033898305084746e-06,
      "loss": 0.9783,
      "step": 7
    },
    {
      "epoch": 0.01024,
      "grad_norm": 4.439324378967285,
      "learning_rate": 2.372881355932204e-06,
      "loss": 0.9672,
      "step": 8
    },
    {
      "epoch": 0.01152,
      "grad_norm": 3.3595175743103027,
      "learning_rate": 2.7118644067796613e-06,
      "loss": 1.0079,
      "step": 9
    },
    {
      "epoch": 0.0128,
      "grad_norm": 3.0522096157073975,
      "learning_rate": 3.0508474576271192e-06,
      "loss": 0.9425,
      "step": 10
    },
    {
      "epoch": 0.01408,
      "grad_norm": 2.8281033039093018,
      "learning_rate": 3.3898305084745763e-06,
      "loss": 0.9666,
      "step": 11
    },
    {
      "epoch": 0.01536,
      "grad_norm": 3.4895122051239014,
      "learning_rate": 3.7288135593220342e-06,
      "loss": 0.9399,
      "step": 12
    },
    {
      "epoch": 0.01664,
      "grad_norm": 3.4156854152679443,
      "learning_rate": 4.067796610169492e-06,
      "loss": 0.9253,
      "step": 13
    },
    {
      "epoch": 0.01792,
      "grad_norm": 3.1675727367401123,
      "learning_rate": 4.40677966101695e-06,
      "loss": 0.9185,
      "step": 14
    },
    {
      "epoch": 0.0192,
      "grad_norm": 2.15570068359375,
      "learning_rate": 4.745762711864408e-06,
      "loss": 0.8582,
      "step": 15
    },
    {
      "epoch": 0.02048,
      "grad_norm": 2.1219990253448486,
      "learning_rate": 5.084745762711865e-06,
      "loss": 0.9036,
      "step": 16
    },
    {
      "epoch": 0.02176,
      "grad_norm": 1.9810932874679565,
      "learning_rate": 5.423728813559323e-06,
      "loss": 0.897,
      "step": 17
    },
    {
      "epoch": 0.02304,
      "grad_norm": 1.596920132637024,
      "learning_rate": 5.7627118644067805e-06,
      "loss": 0.8757,
      "step": 18
    },
    {
      "epoch": 0.02432,
      "grad_norm": 1.31644868850708,
      "learning_rate": 6.1016949152542385e-06,
      "loss": 0.9057,
      "step": 19
    },
    {
      "epoch": 0.0256,
      "grad_norm": 1.2668453454971313,
      "learning_rate": 6.440677966101695e-06,
      "loss": 0.8404,
      "step": 20
    },
    {
      "epoch": 0.02688,
      "grad_norm": 1.502737045288086,
      "learning_rate": 6.779661016949153e-06,
      "loss": 0.8391,
      "step": 21
    },
    {
      "epoch": 0.02816,
      "grad_norm": 1.6580373048782349,
      "learning_rate": 7.1186440677966106e-06,
      "loss": 0.8693,
      "step": 22
    },
    {
      "epoch": 0.02944,
      "grad_norm": 1.4991168975830078,
      "learning_rate": 7.4576271186440685e-06,
      "loss": 0.8866,
      "step": 23
    },
    {
      "epoch": 0.03072,
      "grad_norm": 1.2355320453643799,
      "learning_rate": 7.796610169491526e-06,
      "loss": 0.8427,
      "step": 24
    },
    {
      "epoch": 0.032,
      "grad_norm": 1.173351764678955,
      "learning_rate": 8.135593220338983e-06,
      "loss": 0.8484,
      "step": 25
    },
    {
      "epoch": 0.03328,
      "grad_norm": 1.1595443487167358,
      "learning_rate": 8.47457627118644e-06,
      "loss": 0.8631,
      "step": 26
    },
    {
      "epoch": 0.03456,
      "grad_norm": 1.0706431865692139,
      "learning_rate": 8.8135593220339e-06,
      "loss": 0.7995,
      "step": 27
    },
    {
      "epoch": 0.03584,
      "grad_norm": 1.0414636135101318,
      "learning_rate": 9.152542372881356e-06,
      "loss": 0.8492,
      "step": 28
    },
    {
      "epoch": 0.03712,
      "grad_norm": 0.9365304112434387,
      "learning_rate": 9.491525423728815e-06,
      "loss": 0.8673,
      "step": 29
    },
    {
      "epoch": 0.0384,
      "grad_norm": 0.8955578804016113,
      "learning_rate": 9.830508474576272e-06,
      "loss": 0.817,
      "step": 30
    },
    {
      "epoch": 0.0384,
      "eval_loss": 0.8556124567985535,
      "eval_runtime": 19.5989,
      "eval_samples_per_second": 25.614,
      "eval_steps_per_second": 3.214,
      "step": 30
    },
    {
      "epoch": 0.03968,
      "grad_norm": 0.8397582769393921,
      "learning_rate": 1.016949152542373e-05,
      "loss": 0.8675,
      "step": 31
    },
    {
      "epoch": 0.04096,
      "grad_norm": 0.7109919190406799,
      "learning_rate": 1.0508474576271188e-05,
      "loss": 0.821,
      "step": 32
    },
    {
      "epoch": 0.04224,
      "grad_norm": 0.7386817336082458,
      "learning_rate": 1.0847457627118645e-05,
      "loss": 0.7783,
      "step": 33
    },
    {
      "epoch": 0.04352,
      "grad_norm": 0.8160094022750854,
      "learning_rate": 1.1186440677966102e-05,
      "loss": 0.8154,
      "step": 34
    },
    {
      "epoch": 0.0448,
      "grad_norm": 0.6477371454238892,
      "learning_rate": 1.1525423728813561e-05,
      "loss": 0.8157,
      "step": 35
    },
    {
      "epoch": 0.04608,
      "grad_norm": 0.5607908368110657,
      "learning_rate": 1.1864406779661018e-05,
      "loss": 0.7755,
      "step": 36
    },
    {
      "epoch": 0.04736,
      "grad_norm": 0.6505053043365479,
      "learning_rate": 1.2203389830508477e-05,
      "loss": 0.7971,
      "step": 37
    },
    {
      "epoch": 0.04864,
      "grad_norm": 0.5999047160148621,
      "learning_rate": 1.2542372881355932e-05,
      "loss": 0.7651,
      "step": 38
    },
    {
      "epoch": 0.04992,
      "grad_norm": 0.579596221446991,
      "learning_rate": 1.288135593220339e-05,
      "loss": 0.7833,
      "step": 39
    },
    {
      "epoch": 0.0512,
      "grad_norm": 0.5459831953048706,
      "learning_rate": 1.3220338983050848e-05,
      "loss": 0.7855,
      "step": 40
    },
    {
      "epoch": 0.05248,
      "grad_norm": 0.5510517358779907,
      "learning_rate": 1.3559322033898305e-05,
      "loss": 0.7795,
      "step": 41
    },
    {
      "epoch": 0.05376,
      "grad_norm": 0.5561262965202332,
      "learning_rate": 1.3898305084745764e-05,
      "loss": 0.7874,
      "step": 42
    },
    {
      "epoch": 0.05504,
      "grad_norm": 0.5355618000030518,
      "learning_rate": 1.4237288135593221e-05,
      "loss": 0.7995,
      "step": 43
    },
    {
      "epoch": 0.05632,
      "grad_norm": 0.6200373768806458,
      "learning_rate": 1.4576271186440678e-05,
      "loss": 0.8353,
      "step": 44
    },
    {
      "epoch": 0.0576,
      "grad_norm": 0.552998423576355,
      "learning_rate": 1.4915254237288137e-05,
      "loss": 0.8248,
      "step": 45
    },
    {
      "epoch": 0.05888,
      "grad_norm": 0.5586068630218506,
      "learning_rate": 1.5254237288135594e-05,
      "loss": 0.7416,
      "step": 46
    },
    {
      "epoch": 0.06016,
      "grad_norm": 0.5793619751930237,
      "learning_rate": 1.5593220338983053e-05,
      "loss": 0.7313,
      "step": 47
    },
    {
      "epoch": 0.06144,
      "grad_norm": 0.5604162216186523,
      "learning_rate": 1.593220338983051e-05,
      "loss": 0.7881,
      "step": 48
    },
    {
      "epoch": 0.06272,
      "grad_norm": 0.5856514573097229,
      "learning_rate": 1.6271186440677967e-05,
      "loss": 0.7468,
      "step": 49
    },
    {
      "epoch": 0.064,
      "grad_norm": 0.5885648727416992,
      "learning_rate": 1.6610169491525424e-05,
      "loss": 0.7944,
      "step": 50
    },
    {
      "epoch": 0.06528,
      "grad_norm": 0.6263200640678406,
      "learning_rate": 1.694915254237288e-05,
      "loss": 0.7463,
      "step": 51
    },
    {
      "epoch": 0.06656,
      "grad_norm": 0.5430126190185547,
      "learning_rate": 1.728813559322034e-05,
      "loss": 0.7754,
      "step": 52
    },
    {
      "epoch": 0.06784,
      "grad_norm": 0.6850333213806152,
      "learning_rate": 1.76271186440678e-05,
      "loss": 0.7677,
      "step": 53
    },
    {
      "epoch": 0.06912,
      "grad_norm": 0.683407187461853,
      "learning_rate": 1.7966101694915256e-05,
      "loss": 0.7857,
      "step": 54
    },
    {
      "epoch": 0.0704,
      "grad_norm": 0.5834031105041504,
      "learning_rate": 1.8305084745762713e-05,
      "loss": 0.7972,
      "step": 55
    },
    {
      "epoch": 0.07168,
      "grad_norm": 0.698237419128418,
      "learning_rate": 1.864406779661017e-05,
      "loss": 0.7525,
      "step": 56
    },
    {
      "epoch": 0.07296,
      "grad_norm": 0.5469260811805725,
      "learning_rate": 1.898305084745763e-05,
      "loss": 0.7435,
      "step": 57
    },
    {
      "epoch": 0.07424,
      "grad_norm": 0.7364252805709839,
      "learning_rate": 1.9322033898305087e-05,
      "loss": 0.7505,
      "step": 58
    },
    {
      "epoch": 0.07552,
      "grad_norm": 0.7219014763832092,
      "learning_rate": 1.9661016949152545e-05,
      "loss": 0.7619,
      "step": 59
    },
    {
      "epoch": 0.0768,
      "grad_norm": 0.6183701753616333,
      "learning_rate": 2e-05,
      "loss": 0.7317,
      "step": 60
    },
    {
      "epoch": 0.0768,
      "eval_loss": 0.7924284338951111,
      "eval_runtime": 19.5923,
      "eval_samples_per_second": 25.622,
      "eval_steps_per_second": 3.216,
      "step": 60
    },
    {
      "epoch": 0.07808,
      "grad_norm": 0.6832749247550964,
      "learning_rate": 2.033898305084746e-05,
      "loss": 0.7981,
      "step": 61
    },
    {
      "epoch": 0.07936,
      "grad_norm": 0.6453201174736023,
      "learning_rate": 2.0677966101694916e-05,
      "loss": 0.7711,
      "step": 62
    },
    {
      "epoch": 0.08064,
      "grad_norm": 0.6775647401809692,
      "learning_rate": 2.1016949152542376e-05,
      "loss": 0.7449,
      "step": 63
    },
    {
      "epoch": 0.08192,
      "grad_norm": 0.5813889503479004,
      "learning_rate": 2.1355932203389833e-05,
      "loss": 0.7393,
      "step": 64
    },
    {
      "epoch": 0.0832,
      "grad_norm": 0.6688156723976135,
      "learning_rate": 2.169491525423729e-05,
      "loss": 0.7511,
      "step": 65
    },
    {
      "epoch": 0.08448,
      "grad_norm": 0.6027134656906128,
      "learning_rate": 2.2033898305084748e-05,
      "loss": 0.7699,
      "step": 66
    },
    {
      "epoch": 0.08576,
      "grad_norm": 0.6834599375724792,
      "learning_rate": 2.2372881355932205e-05,
      "loss": 0.7522,
      "step": 67
    },
    {
      "epoch": 0.08704,
      "grad_norm": 0.6269927024841309,
      "learning_rate": 2.2711864406779665e-05,
      "loss": 0.7559,
      "step": 68
    },
    {
      "epoch": 0.08832,
      "grad_norm": 0.6447550058364868,
      "learning_rate": 2.3050847457627122e-05,
      "loss": 0.7821,
      "step": 69
    },
    {
      "epoch": 0.0896,
      "grad_norm": 0.6334463357925415,
      "learning_rate": 2.338983050847458e-05,
      "loss": 0.7555,
      "step": 70
    },
    {
      "epoch": 0.09088,
      "grad_norm": 0.5855139493942261,
      "learning_rate": 2.3728813559322036e-05,
      "loss": 0.7455,
      "step": 71
    },
    {
      "epoch": 0.09216,
      "grad_norm": 0.6558398008346558,
      "learning_rate": 2.406779661016949e-05,
      "loss": 0.7608,
      "step": 72
    },
    {
      "epoch": 0.09344,
      "grad_norm": 0.6356144547462463,
      "learning_rate": 2.4406779661016954e-05,
      "loss": 0.7179,
      "step": 73
    },
    {
      "epoch": 0.09472,
      "grad_norm": 0.5918307304382324,
      "learning_rate": 2.474576271186441e-05,
      "loss": 0.6638,
      "step": 74
    },
    {
      "epoch": 0.096,
      "grad_norm": 0.5894814729690552,
      "learning_rate": 2.5084745762711865e-05,
      "loss": 0.7639,
      "step": 75
    },
    {
      "epoch": 0.09728,
      "grad_norm": 0.5442491173744202,
      "learning_rate": 2.5423728813559322e-05,
      "loss": 0.7273,
      "step": 76
    },
    {
      "epoch": 0.09856,
      "grad_norm": 0.5948322415351868,
      "learning_rate": 2.576271186440678e-05,
      "loss": 0.7907,
      "step": 77
    },
    {
      "epoch": 0.09984,
      "grad_norm": 0.5570871233940125,
      "learning_rate": 2.610169491525424e-05,
      "loss": 0.698,
      "step": 78
    },
    {
      "epoch": 0.10112,
      "grad_norm": 0.5739171504974365,
      "learning_rate": 2.6440677966101696e-05,
      "loss": 0.7399,
      "step": 79
    },
    {
      "epoch": 0.1024,
      "grad_norm": 0.6343491077423096,
      "learning_rate": 2.6779661016949153e-05,
      "loss": 0.748,
      "step": 80
    },
    {
      "epoch": 0.10368,
      "grad_norm": 0.5615505576133728,
      "learning_rate": 2.711864406779661e-05,
      "loss": 0.7119,
      "step": 81
    },
    {
      "epoch": 0.10496,
      "grad_norm": 0.6687747240066528,
      "learning_rate": 2.7457627118644068e-05,
      "loss": 0.7364,
      "step": 82
    },
    {
      "epoch": 0.10624,
      "grad_norm": 0.6406599283218384,
      "learning_rate": 2.7796610169491528e-05,
      "loss": 0.7379,
      "step": 83
    },
    {
      "epoch": 0.10752,
      "grad_norm": 0.6296663880348206,
      "learning_rate": 2.8135593220338985e-05,
      "loss": 0.7276,
      "step": 84
    },
    {
      "epoch": 0.1088,
      "grad_norm": 0.6627721190452576,
      "learning_rate": 2.8474576271186442e-05,
      "loss": 0.7595,
      "step": 85
    },
    {
      "epoch": 0.11008,
      "grad_norm": 0.5808905363082886,
      "learning_rate": 2.88135593220339e-05,
      "loss": 0.7365,
      "step": 86
    },
    {
      "epoch": 0.11136,
      "grad_norm": 0.7772956490516663,
      "learning_rate": 2.9152542372881356e-05,
      "loss": 0.7491,
      "step": 87
    },
    {
      "epoch": 0.11264,
      "grad_norm": 0.6019331812858582,
      "learning_rate": 2.9491525423728817e-05,
      "loss": 0.7542,
      "step": 88
    },
    {
      "epoch": 0.11392,
      "grad_norm": 0.7027397751808167,
      "learning_rate": 2.9830508474576274e-05,
      "loss": 0.7413,
      "step": 89
    },
    {
      "epoch": 0.1152,
      "grad_norm": 0.6041712164878845,
      "learning_rate": 3.016949152542373e-05,
      "loss": 0.7232,
      "step": 90
    },
    {
      "epoch": 0.1152,
      "eval_loss": 0.7702571153640747,
      "eval_runtime": 19.5941,
      "eval_samples_per_second": 25.62,
      "eval_steps_per_second": 3.215,
      "step": 90
    },
    {
      "epoch": 0.11648,
      "grad_norm": 0.6485719680786133,
      "learning_rate": 3.0508474576271188e-05,
      "loss": 0.7666,
      "step": 91
    },
    {
      "epoch": 0.11776,
      "grad_norm": 0.6684650182723999,
      "learning_rate": 3.084745762711865e-05,
      "loss": 0.731,
      "step": 92
    },
    {
      "epoch": 0.11904,
      "grad_norm": 0.5569056272506714,
      "learning_rate": 3.1186440677966106e-05,
      "loss": 0.7148,
      "step": 93
    },
    {
      "epoch": 0.12032,
      "grad_norm": 0.6821630001068115,
      "learning_rate": 3.152542372881356e-05,
      "loss": 0.7213,
      "step": 94
    },
    {
      "epoch": 0.1216,
      "grad_norm": 0.7205902338027954,
      "learning_rate": 3.186440677966102e-05,
      "loss": 0.7382,
      "step": 95
    },
    {
      "epoch": 0.12288,
      "grad_norm": 0.7073635458946228,
      "learning_rate": 3.2203389830508473e-05,
      "loss": 0.7535,
      "step": 96
    },
    {
      "epoch": 0.12416,
      "grad_norm": 0.6599916219711304,
      "learning_rate": 3.2542372881355934e-05,
      "loss": 0.7329,
      "step": 97
    },
    {
      "epoch": 0.12544,
      "grad_norm": 0.7599468231201172,
      "learning_rate": 3.2881355932203394e-05,
      "loss": 0.6969,
      "step": 98
    },
    {
      "epoch": 0.12672,
      "grad_norm": 0.6323254108428955,
      "learning_rate": 3.322033898305085e-05,
      "loss": 0.7418,
      "step": 99
    },
    {
      "epoch": 0.128,
      "grad_norm": 0.694995641708374,
      "learning_rate": 3.355932203389831e-05,
      "loss": 0.7393,
      "step": 100
    },
    {
      "epoch": 0.12928,
      "grad_norm": 0.584010899066925,
      "learning_rate": 3.389830508474576e-05,
      "loss": 0.7245,
      "step": 101
    },
    {
      "epoch": 0.13056,
      "grad_norm": 0.7612520456314087,
      "learning_rate": 3.423728813559322e-05,
      "loss": 0.7576,
      "step": 102
    },
    {
      "epoch": 0.13184,
      "grad_norm": 0.6741138100624084,
      "learning_rate": 3.457627118644068e-05,
      "loss": 0.7317,
      "step": 103
    },
    {
      "epoch": 0.13312,
      "grad_norm": 0.744069516658783,
      "learning_rate": 3.491525423728814e-05,
      "loss": 0.7601,
      "step": 104
    },
    {
      "epoch": 0.1344,
      "grad_norm": 0.8154940605163574,
      "learning_rate": 3.52542372881356e-05,
      "loss": 0.7363,
      "step": 105
    },
    {
      "epoch": 0.13568,
      "grad_norm": 0.697728157043457,
      "learning_rate": 3.559322033898305e-05,
      "loss": 0.7392,
      "step": 106
    },
    {
      "epoch": 0.13696,
      "grad_norm": 0.8070045709609985,
      "learning_rate": 3.593220338983051e-05,
      "loss": 0.7376,
      "step": 107
    },
    {
      "epoch": 0.13824,
      "grad_norm": 0.6787447929382324,
      "learning_rate": 3.627118644067797e-05,
      "loss": 0.7204,
      "step": 108
    },
    {
      "epoch": 0.13952,
      "grad_norm": 0.8469722867012024,
      "learning_rate": 3.6610169491525426e-05,
      "loss": 0.7347,
      "step": 109
    },
    {
      "epoch": 0.1408,
      "grad_norm": 0.7555689811706543,
      "learning_rate": 3.6949152542372886e-05,
      "loss": 0.7203,
      "step": 110
    },
    {
      "epoch": 0.14208,
      "grad_norm": 0.6541507244110107,
      "learning_rate": 3.728813559322034e-05,
      "loss": 0.7116,
      "step": 111
    },
    {
      "epoch": 0.14336,
      "grad_norm": 0.7556775212287903,
      "learning_rate": 3.76271186440678e-05,
      "loss": 0.7118,
      "step": 112
    },
    {
      "epoch": 0.14464,
      "grad_norm": 0.6805198192596436,
      "learning_rate": 3.796610169491526e-05,
      "loss": 0.7281,
      "step": 113
    },
    {
      "epoch": 0.14592,
      "grad_norm": 0.6481510996818542,
      "learning_rate": 3.8305084745762714e-05,
      "loss": 0.7049,
      "step": 114
    },
    {
      "epoch": 0.1472,
      "grad_norm": 0.8636525869369507,
      "learning_rate": 3.8644067796610175e-05,
      "loss": 0.7306,
      "step": 115
    },
    {
      "epoch": 0.14848,
      "grad_norm": 0.6602344512939453,
      "learning_rate": 3.898305084745763e-05,
      "loss": 0.7546,
      "step": 116
    },
    {
      "epoch": 0.14976,
      "grad_norm": 0.7977970242500305,
      "learning_rate": 3.932203389830509e-05,
      "loss": 0.6979,
      "step": 117
    },
    {
      "epoch": 0.15104,
      "grad_norm": 0.7211064100265503,
      "learning_rate": 3.966101694915255e-05,
      "loss": 0.7607,
      "step": 118
    },
    {
      "epoch": 0.15232,
      "grad_norm": 0.9843719005584717,
      "learning_rate": 4e-05,
      "loss": 0.7308,
      "step": 119
    },
    {
      "epoch": 0.1536,
      "grad_norm": 0.8833848834037781,
      "learning_rate": 3.999999382260215e-05,
      "loss": 0.7609,
      "step": 120
    },
    {
      "epoch": 0.1536,
      "eval_loss": 0.7598698735237122,
      "eval_runtime": 19.5857,
      "eval_samples_per_second": 25.631,
      "eval_steps_per_second": 3.217,
      "step": 120
    },
    {
      "epoch": 0.15488,
      "grad_norm": 0.8075857758522034,
      "learning_rate": 3.9999975290412844e-05,
      "loss": 0.7324,
      "step": 121
    },
    {
      "epoch": 0.15616,
      "grad_norm": 0.7772446870803833,
      "learning_rate": 3.9999944403444785e-05,
      "loss": 0.6793,
      "step": 122
    },
    {
      "epoch": 0.15744,
      "grad_norm": 0.6814696788787842,
      "learning_rate": 3.999990116171919e-05,
      "loss": 0.7117,
      "step": 123
    },
    {
      "epoch": 0.15872,
      "grad_norm": 0.881449282169342,
      "learning_rate": 3.999984556526574e-05,
      "loss": 0.664,
      "step": 124
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.6054081320762634,
      "learning_rate": 3.9999777614122576e-05,
      "loss": 0.7515,
      "step": 125
    },
    {
      "epoch": 0.16128,
      "grad_norm": 0.897861123085022,
      "learning_rate": 3.9999697308336365e-05,
      "loss": 0.7238,
      "step": 126
    },
    {
      "epoch": 0.16256,
      "grad_norm": 0.6182326674461365,
      "learning_rate": 3.99996046479622e-05,
      "loss": 0.7571,
      "step": 127
    },
    {
      "epoch": 0.16384,
      "grad_norm": 0.7834803462028503,
      "learning_rate": 3.99994996330637e-05,
      "loss": 0.7465,
      "step": 128
    },
    {
      "epoch": 0.16512,
      "grad_norm": 0.658669650554657,
      "learning_rate": 3.999938226371294e-05,
      "loss": 0.7263,
      "step": 129
    },
    {
      "epoch": 0.1664,
      "grad_norm": 0.7675626873970032,
      "learning_rate": 3.999925253999048e-05,
      "loss": 0.7021,
      "step": 130
    },
    {
      "epoch": 0.16768,
      "grad_norm": 0.7073703408241272,
      "learning_rate": 3.999911046198536e-05,
      "loss": 0.7103,
      "step": 131
    },
    {
      "epoch": 0.16896,
      "grad_norm": 0.7621209025382996,
      "learning_rate": 3.9998956029795096e-05,
      "loss": 0.7122,
      "step": 132
    },
    {
      "epoch": 0.17024,
      "grad_norm": 0.732304036617279,
      "learning_rate": 3.999878924352568e-05,
      "loss": 0.7566,
      "step": 133
    },
    {
      "epoch": 0.17152,
      "grad_norm": 0.683933436870575,
      "learning_rate": 3.99986101032916e-05,
      "loss": 0.7343,
      "step": 134
    },
    {
      "epoch": 0.1728,
      "grad_norm": 0.6686181426048279,
      "learning_rate": 3.9998418609215826e-05,
      "loss": 0.7234,
      "step": 135
    },
    {
      "epoch": 0.17408,
      "grad_norm": 0.6578155159950256,
      "learning_rate": 3.999821476142977e-05,
      "loss": 0.7474,
      "step": 136
    },
    {
      "epoch": 0.17536,
      "grad_norm": 0.5755041241645813,
      "learning_rate": 3.9997998560073364e-05,
      "loss": 0.7251,
      "step": 137
    },
    {
      "epoch": 0.17664,
      "grad_norm": 0.665770947933197,
      "learning_rate": 3.9997770005294996e-05,
      "loss": 0.7235,
      "step": 138
    },
    {
      "epoch": 0.17792,
      "grad_norm": 0.5992425680160522,
      "learning_rate": 3.999752909725154e-05,
      "loss": 0.6807,
      "step": 139
    },
    {
      "epoch": 0.1792,
      "grad_norm": 0.6816670298576355,
      "learning_rate": 3.9997275836108366e-05,
      "loss": 0.6825,
      "step": 140
    },
    {
      "epoch": 0.18048,
      "grad_norm": 0.5996707677841187,
      "learning_rate": 3.9997010222039297e-05,
      "loss": 0.6961,
      "step": 141
    },
    {
      "epoch": 0.18176,
      "grad_norm": 0.5953992009162903,
      "learning_rate": 3.999673225522664e-05,
      "loss": 0.7227,
      "step": 142
    },
    {
      "epoch": 0.18304,
      "grad_norm": 0.6428366899490356,
      "learning_rate": 3.999644193586118e-05,
      "loss": 0.7649,
      "step": 143
    },
    {
      "epoch": 0.18432,
      "grad_norm": 0.7236312627792358,
      "learning_rate": 3.9996139264142205e-05,
      "loss": 0.7387,
      "step": 144
    },
    {
      "epoch": 0.1856,
      "grad_norm": 0.6506896615028381,
      "learning_rate": 3.999582424027744e-05,
      "loss": 0.7318,
      "step": 145
    },
    {
      "epoch": 0.18688,
      "grad_norm": 0.7359318733215332,
      "learning_rate": 3.999549686448313e-05,
      "loss": 0.705,
      "step": 146
    },
    {
      "epoch": 0.18816,
      "grad_norm": 0.7491677403450012,
      "learning_rate": 3.9995157136983966e-05,
      "loss": 0.7272,
      "step": 147
    },
    {
      "epoch": 0.18944,
      "grad_norm": 0.7297288179397583,
      "learning_rate": 3.999480505801313e-05,
      "loss": 0.7346,
      "step": 148
    },
    {
      "epoch": 0.19072,
      "grad_norm": 0.8053178787231445,
      "learning_rate": 3.999444062781229e-05,
      "loss": 0.7583,
      "step": 149
    },
    {
      "epoch": 0.192,
      "grad_norm": 0.6667599678039551,
      "learning_rate": 3.9994063846631565e-05,
      "loss": 0.6777,
      "step": 150
    },
    {
      "epoch": 0.192,
      "eval_loss": 0.7519004344940186,
      "eval_runtime": 19.5884,
      "eval_samples_per_second": 25.627,
      "eval_steps_per_second": 3.216,
      "step": 150
    },
    {
      "epoch": 0.19328,
      "grad_norm": 0.6819199919700623,
      "learning_rate": 3.99936747147296e-05,
      "loss": 0.7239,
      "step": 151
    },
    {
      "epoch": 0.19456,
      "grad_norm": 0.6432419419288635,
      "learning_rate": 3.9993273232373455e-05,
      "loss": 0.7198,
      "step": 152
    },
    {
      "epoch": 0.19584,
      "grad_norm": 0.6553572416305542,
      "learning_rate": 3.999285939983871e-05,
      "loss": 0.7195,
      "step": 153
    },
    {
      "epoch": 0.19712,
      "grad_norm": 0.6333758234977722,
      "learning_rate": 3.9992433217409414e-05,
      "loss": 0.7126,
      "step": 154
    },
    {
      "epoch": 0.1984,
      "grad_norm": 0.649039089679718,
      "learning_rate": 3.999199468537809e-05,
      "loss": 0.7413,
      "step": 155
    },
    {
      "epoch": 0.19968,
      "grad_norm": 0.6506604552268982,
      "learning_rate": 3.999154380404572e-05,
      "loss": 0.7115,
      "step": 156
    },
    {
      "epoch": 0.20096,
      "grad_norm": 0.6044195890426636,
      "learning_rate": 3.9991080573721805e-05,
      "loss": 0.7035,
      "step": 157
    },
    {
      "epoch": 0.20224,
      "grad_norm": 0.6326198577880859,
      "learning_rate": 3.9990604994724276e-05,
      "loss": 0.7138,
      "step": 158
    },
    {
      "epoch": 0.20352,
      "grad_norm": 0.7158970236778259,
      "learning_rate": 3.999011706737957e-05,
      "loss": 0.7224,
      "step": 159
    },
    {
      "epoch": 0.2048,
      "grad_norm": 0.8039223551750183,
      "learning_rate": 3.998961679202258e-05,
      "loss": 0.7015,
      "step": 160
    },
    {
      "epoch": 0.20608,
      "grad_norm": 0.6923312544822693,
      "learning_rate": 3.998910416899669e-05,
      "loss": 0.7168,
      "step": 161
    },
    {
      "epoch": 0.20736,
      "grad_norm": 0.7452057003974915,
      "learning_rate": 3.998857919865376e-05,
      "loss": 0.752,
      "step": 162
    },
    {
      "epoch": 0.20864,
      "grad_norm": 0.7468007802963257,
      "learning_rate": 3.9988041881354095e-05,
      "loss": 0.7323,
      "step": 163
    },
    {
      "epoch": 0.20992,
      "grad_norm": 0.7677522301673889,
      "learning_rate": 3.998749221746652e-05,
      "loss": 0.7293,
      "step": 164
    },
    {
      "epoch": 0.2112,
      "grad_norm": 0.7293217182159424,
      "learning_rate": 3.9986930207368306e-05,
      "loss": 0.6973,
      "step": 165
    },
    {
      "epoch": 0.21248,
      "grad_norm": 0.5631901621818542,
      "learning_rate": 3.998635585144521e-05,
      "loss": 0.7176,
      "step": 166
    },
    {
      "epoch": 0.21376,
      "grad_norm": 0.7135562896728516,
      "learning_rate": 3.9985769150091434e-05,
      "loss": 0.7005,
      "step": 167
    },
    {
      "epoch": 0.21504,
      "grad_norm": 0.5999516248703003,
      "learning_rate": 3.99851701037097e-05,
      "loss": 0.7132,
      "step": 168
    },
    {
      "epoch": 0.21632,
      "grad_norm": 0.6648542881011963,
      "learning_rate": 3.998455871271117e-05,
      "loss": 0.6966,
      "step": 169
    },
    {
      "epoch": 0.2176,
      "grad_norm": 0.7289178371429443,
      "learning_rate": 3.99839349775155e-05,
      "loss": 0.7506,
      "step": 170
    },
    {
      "epoch": 0.21888,
      "grad_norm": 0.5922276973724365,
      "learning_rate": 3.9983298898550794e-05,
      "loss": 0.6882,
      "step": 171
    },
    {
      "epoch": 0.22016,
      "grad_norm": 0.7117915749549866,
      "learning_rate": 3.998265047625364e-05,
      "loss": 0.7065,
      "step": 172
    },
    {
      "epoch": 0.22144,
      "grad_norm": 0.6005136966705322,
      "learning_rate": 3.998198971106912e-05,
      "loss": 0.7249,
      "step": 173
    },
    {
      "epoch": 0.22272,
      "grad_norm": 0.7030836939811707,
      "learning_rate": 3.998131660345075e-05,
      "loss": 0.6945,
      "step": 174
    },
    {
      "epoch": 0.224,
      "grad_norm": 0.6256377100944519,
      "learning_rate": 3.998063115386054e-05,
      "loss": 0.6946,
      "step": 175
    },
    {
      "epoch": 0.22528,
      "grad_norm": 0.6633886098861694,
      "learning_rate": 3.997993336276897e-05,
      "loss": 0.7134,
      "step": 176
    },
    {
      "epoch": 0.22656,
      "grad_norm": 0.6532630920410156,
      "learning_rate": 3.9979223230654994e-05,
      "loss": 0.6836,
      "step": 177
    },
    {
      "epoch": 0.22784,
      "grad_norm": 0.662128210067749,
      "learning_rate": 3.997850075800602e-05,
      "loss": 0.7235,
      "step": 178
    },
    {
      "epoch": 0.22912,
      "grad_norm": 0.6497978568077087,
      "learning_rate": 3.9977765945317934e-05,
      "loss": 0.7059,
      "step": 179
    },
    {
      "epoch": 0.2304,
      "grad_norm": 0.6233092546463013,
      "learning_rate": 3.997701879309511e-05,
      "loss": 0.6832,
      "step": 180
    },
    {
      "epoch": 0.2304,
      "eval_loss": 0.7454701662063599,
      "eval_runtime": 19.5832,
      "eval_samples_per_second": 25.634,
      "eval_steps_per_second": 3.217,
      "step": 180
    },
    {
      "epoch": 0.23168,
      "grad_norm": 0.685046911239624,
      "learning_rate": 3.9976259301850375e-05,
      "loss": 0.7217,
      "step": 181
    },
    {
      "epoch": 0.23296,
      "grad_norm": 0.5651426315307617,
      "learning_rate": 3.997548747210501e-05,
      "loss": 0.7003,
      "step": 182
    },
    {
      "epoch": 0.23424,
      "grad_norm": 0.7419203519821167,
      "learning_rate": 3.997470330438879e-05,
      "loss": 0.7204,
      "step": 183
    },
    {
      "epoch": 0.23552,
      "grad_norm": 0.5573960542678833,
      "learning_rate": 3.9973906799239964e-05,
      "loss": 0.714,
      "step": 184
    },
    {
      "epoch": 0.2368,
      "grad_norm": 0.701404333114624,
      "learning_rate": 3.997309795720521e-05,
      "loss": 0.7199,
      "step": 185
    },
    {
      "epoch": 0.23808,
      "grad_norm": 0.6078205108642578,
      "learning_rate": 3.9972276778839725e-05,
      "loss": 0.6842,
      "step": 186
    },
    {
      "epoch": 0.23936,
      "grad_norm": 0.6672995686531067,
      "learning_rate": 3.997144326470713e-05,
      "loss": 0.6844,
      "step": 187
    },
    {
      "epoch": 0.24064,
      "grad_norm": 0.6733210682868958,
      "learning_rate": 3.9970597415379535e-05,
      "loss": 0.6862,
      "step": 188
    },
    {
      "epoch": 0.24192,
      "grad_norm": 0.6919593811035156,
      "learning_rate": 3.9969739231437514e-05,
      "loss": 0.7051,
      "step": 189
    },
    {
      "epoch": 0.2432,
      "grad_norm": 0.756440281867981,
      "learning_rate": 3.99688687134701e-05,
      "loss": 0.7278,
      "step": 190
    },
    {
      "epoch": 0.24448,
      "grad_norm": 0.6633826494216919,
      "learning_rate": 3.9967985862074795e-05,
      "loss": 0.7298,
      "step": 191
    },
    {
      "epoch": 0.24576,
      "grad_norm": 0.631435751914978,
      "learning_rate": 3.996709067785758e-05,
      "loss": 0.7064,
      "step": 192
    },
    {
      "epoch": 0.24704,
      "grad_norm": 0.6971995234489441,
      "learning_rate": 3.996618316143288e-05,
      "loss": 0.6691,
      "step": 193
    },
    {
      "epoch": 0.24832,
      "grad_norm": 0.6438106298446655,
      "learning_rate": 3.996526331342361e-05,
      "loss": 0.7211,
      "step": 194
    },
    {
      "epoch": 0.2496,
      "grad_norm": 0.840981662273407,
      "learning_rate": 3.9964331134461104e-05,
      "loss": 0.6943,
      "step": 195
    },
    {
      "epoch": 0.25088,
      "grad_norm": 0.6285365223884583,
      "learning_rate": 3.996338662518521e-05,
      "loss": 0.703,
      "step": 196
    },
    {
      "epoch": 0.25216,
      "grad_norm": 0.6623023748397827,
      "learning_rate": 3.996242978624421e-05,
      "loss": 0.645,
      "step": 197
    },
    {
      "epoch": 0.25344,
      "grad_norm": 0.6104528307914734,
      "learning_rate": 3.996146061829487e-05,
      "loss": 0.6837,
      "step": 198
    },
    {
      "epoch": 0.25472,
      "grad_norm": 0.6691868305206299,
      "learning_rate": 3.9960479122002384e-05,
      "loss": 0.6958,
      "step": 199
    },
    {
      "epoch": 0.256,
      "grad_norm": 0.677840530872345,
      "learning_rate": 3.9959485298040436e-05,
      "loss": 0.7057,
      "step": 200
    },
    {
      "epoch": 0.25728,
      "grad_norm": 0.6167845726013184,
      "learning_rate": 3.995847914709118e-05,
      "loss": 0.6864,
      "step": 201
    },
    {
      "epoch": 0.25856,
      "grad_norm": 0.5558090806007385,
      "learning_rate": 3.9957460669845204e-05,
      "loss": 0.6922,
      "step": 202
    },
    {
      "epoch": 0.25984,
      "grad_norm": 0.6730953454971313,
      "learning_rate": 3.9956429867001554e-05,
      "loss": 0.731,
      "step": 203
    },
    {
      "epoch": 0.26112,
      "grad_norm": 0.6932212114334106,
      "learning_rate": 3.9955386739267766e-05,
      "loss": 0.6889,
      "step": 204
    },
    {
      "epoch": 0.2624,
      "grad_norm": 0.6830756664276123,
      "learning_rate": 3.9954331287359827e-05,
      "loss": 0.7464,
      "step": 205
    },
    {
      "epoch": 0.26368,
      "grad_norm": 0.627577543258667,
      "learning_rate": 3.995326351200215e-05,
      "loss": 0.6922,
      "step": 206
    },
    {
      "epoch": 0.26496,
      "grad_norm": 0.6913477778434753,
      "learning_rate": 3.995218341392766e-05,
      "loss": 0.73,
      "step": 207
    },
    {
      "epoch": 0.26624,
      "grad_norm": 0.6398324966430664,
      "learning_rate": 3.995109099387769e-05,
      "loss": 0.735,
      "step": 208
    },
    {
      "epoch": 0.26752,
      "grad_norm": 0.6085021495819092,
      "learning_rate": 3.9949986252602074e-05,
      "loss": 0.701,
      "step": 209
    },
    {
      "epoch": 0.2688,
      "grad_norm": 0.8021439909934998,
      "learning_rate": 3.994886919085906e-05,
      "loss": 0.7138,
      "step": 210
    },
    {
      "epoch": 0.2688,
      "eval_loss": 0.7406783699989319,
      "eval_runtime": 19.5682,
      "eval_samples_per_second": 25.654,
      "eval_steps_per_second": 3.22,
      "step": 210
    },
    {
      "epoch": 0.27008,
      "grad_norm": 0.6654173731803894,
      "learning_rate": 3.9947739809415384e-05,
      "loss": 0.6765,
      "step": 211
    },
    {
      "epoch": 0.27136,
      "grad_norm": 0.7146822214126587,
      "learning_rate": 3.994659810904623e-05,
      "loss": 0.677,
      "step": 212
    },
    {
      "epoch": 0.27264,
      "grad_norm": 0.565392255783081,
      "learning_rate": 3.994544409053523e-05,
      "loss": 0.7423,
      "step": 213
    },
    {
      "epoch": 0.27392,
      "grad_norm": 0.6871966123580933,
      "learning_rate": 3.994427775467448e-05,
      "loss": 0.7129,
      "step": 214
    },
    {
      "epoch": 0.2752,
      "grad_norm": 0.6078714728355408,
      "learning_rate": 3.994309910226453e-05,
      "loss": 0.6752,
      "step": 215
    },
    {
      "epoch": 0.27648,
      "grad_norm": 0.6045188307762146,
      "learning_rate": 3.994190813411437e-05,
      "loss": 0.6956,
      "step": 216
    },
    {
      "epoch": 0.27776,
      "grad_norm": 0.6105816960334778,
      "learning_rate": 3.9940704851041466e-05,
      "loss": 0.7053,
      "step": 217
    },
    {
      "epoch": 0.27904,
      "grad_norm": 0.6666073203086853,
      "learning_rate": 3.9939489253871715e-05,
      "loss": 0.7389,
      "step": 218
    },
    {
      "epoch": 0.28032,
      "grad_norm": 0.6191258430480957,
      "learning_rate": 3.993826134343949e-05,
      "loss": 0.7422,
      "step": 219
    },
    {
      "epoch": 0.2816,
      "grad_norm": 0.6105319857597351,
      "learning_rate": 3.993702112058757e-05,
      "loss": 0.6996,
      "step": 220
    },
    {
      "epoch": 0.28288,
      "grad_norm": 0.6659864783287048,
      "learning_rate": 3.9935768586167254e-05,
      "loss": 0.725,
      "step": 221
    },
    {
      "epoch": 0.28416,
      "grad_norm": 0.605975866317749,
      "learning_rate": 3.993450374103823e-05,
      "loss": 0.72,
      "step": 222
    },
    {
      "epoch": 0.28544,
      "grad_norm": 0.6666836738586426,
      "learning_rate": 3.9933226586068666e-05,
      "loss": 0.6969,
      "step": 223
    },
    {
      "epoch": 0.28672,
      "grad_norm": 0.6767602562904358,
      "learning_rate": 3.9931937122135176e-05,
      "loss": 0.7264,
      "step": 224
    },
    {
      "epoch": 0.288,
      "grad_norm": 0.6438207030296326,
      "learning_rate": 3.9930635350122806e-05,
      "loss": 0.7237,
      "step": 225
    },
    {
      "epoch": 0.28928,
      "grad_norm": 0.6157245635986328,
      "learning_rate": 3.992932127092508e-05,
      "loss": 0.7292,
      "step": 226
    },
    {
      "epoch": 0.29056,
      "grad_norm": 0.6455304026603699,
      "learning_rate": 3.9927994885443945e-05,
      "loss": 0.7175,
      "step": 227
    },
    {
      "epoch": 0.29184,
      "grad_norm": 0.7229188084602356,
      "learning_rate": 3.99266561945898e-05,
      "loss": 0.7244,
      "step": 228
    },
    {
      "epoch": 0.29312,
      "grad_norm": 0.5951524376869202,
      "learning_rate": 3.99253051992815e-05,
      "loss": 0.7234,
      "step": 229
    },
    {
      "epoch": 0.2944,
      "grad_norm": 0.6291482448577881,
      "learning_rate": 3.992394190044632e-05,
      "loss": 0.7122,
      "step": 230
    },
    {
      "epoch": 0.29568,
      "grad_norm": 0.5866241455078125,
      "learning_rate": 3.9922566299020015e-05,
      "loss": 0.6872,
      "step": 231
    },
    {
      "epoch": 0.29696,
      "grad_norm": 0.5411866903305054,
      "learning_rate": 3.992117839594676e-05,
      "loss": 0.6871,
      "step": 232
    },
    {
      "epoch": 0.29824,
      "grad_norm": 0.6554620862007141,
      "learning_rate": 3.991977819217919e-05,
      "loss": 0.6731,
      "step": 233
    },
    {
      "epoch": 0.29952,
      "grad_norm": 0.6732813715934753,
      "learning_rate": 3.9918365688678354e-05,
      "loss": 0.7271,
      "step": 234
    },
    {
      "epoch": 0.3008,
      "grad_norm": 0.7116442322731018,
      "learning_rate": 3.991694088641377e-05,
      "loss": 0.7123,
      "step": 235
    },
    {
      "epoch": 0.30208,
      "grad_norm": 0.5864224433898926,
      "learning_rate": 3.991550378636341e-05,
      "loss": 0.6614,
      "step": 236
    },
    {
      "epoch": 0.30336,
      "grad_norm": 0.630328357219696,
      "learning_rate": 3.991405438951365e-05,
      "loss": 0.7007,
      "step": 237
    },
    {
      "epoch": 0.30464,
      "grad_norm": 0.6472702026367188,
      "learning_rate": 3.991259269685932e-05,
      "loss": 0.6796,
      "step": 238
    },
    {
      "epoch": 0.30592,
      "grad_norm": 0.7071817517280579,
      "learning_rate": 3.9911118709403694e-05,
      "loss": 0.7207,
      "step": 239
    },
    {
      "epoch": 0.3072,
      "grad_norm": 0.7071561813354492,
      "learning_rate": 3.99096324281585e-05,
      "loss": 0.6707,
      "step": 240
    },
    {
      "epoch": 0.3072,
      "eval_loss": 0.7365993857383728,
      "eval_runtime": 19.6647,
      "eval_samples_per_second": 25.528,
      "eval_steps_per_second": 3.204,
      "step": 240
    },
    {
      "epoch": 0.30848,
      "grad_norm": 0.6155024170875549,
      "learning_rate": 3.9908133854143865e-05,
      "loss": 0.6712,
      "step": 241
    },
    {
      "epoch": 0.30976,
      "grad_norm": 0.6253347396850586,
      "learning_rate": 3.9906622988388396e-05,
      "loss": 0.7008,
      "step": 242
    },
    {
      "epoch": 0.31104,
      "grad_norm": 0.7121750712394714,
      "learning_rate": 3.99050998319291e-05,
      "loss": 0.7262,
      "step": 243
    },
    {
      "epoch": 0.31232,
      "grad_norm": 0.6147091388702393,
      "learning_rate": 3.9903564385811456e-05,
      "loss": 0.6519,
      "step": 244
    },
    {
      "epoch": 0.3136,
      "grad_norm": 0.6230913996696472,
      "learning_rate": 3.990201665108934e-05,
      "loss": 0.6903,
      "step": 245
    },
    {
      "epoch": 0.31488,
      "grad_norm": 0.6431207656860352,
      "learning_rate": 3.99004566288251e-05,
      "loss": 0.7014,
      "step": 246
    },
    {
      "epoch": 0.31616,
      "grad_norm": 0.6160919666290283,
      "learning_rate": 3.989888432008948e-05,
      "loss": 0.6572,
      "step": 247
    },
    {
      "epoch": 0.31744,
      "grad_norm": 0.5764616131782532,
      "learning_rate": 3.989729972596171e-05,
      "loss": 0.7235,
      "step": 248
    },
    {
      "epoch": 0.31872,
      "grad_norm": 0.6176016330718994,
      "learning_rate": 3.989570284752939e-05,
      "loss": 0.6847,
      "step": 249
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.6176882982254028,
      "learning_rate": 3.9894093685888584e-05,
      "loss": 0.7034,
      "step": 250
    },
    {
      "epoch": 0.32128,
      "grad_norm": 0.5789080262184143,
      "learning_rate": 3.98924722421438e-05,
      "loss": 0.6661,
      "step": 251
    },
    {
      "epoch": 0.32256,
      "grad_norm": 0.6819998621940613,
      "learning_rate": 3.989083851740795e-05,
      "loss": 0.7062,
      "step": 252
    },
    {
      "epoch": 0.32384,
      "grad_norm": 0.6409872770309448,
      "learning_rate": 3.9889192512802395e-05,
      "loss": 0.6699,
      "step": 253
    },
    {
      "epoch": 0.32512,
      "grad_norm": 0.6627818942070007,
      "learning_rate": 3.9887534229456896e-05,
      "loss": 0.6948,
      "step": 254
    },
    {
      "epoch": 0.3264,
      "grad_norm": 0.655550479888916,
      "learning_rate": 3.988586366850968e-05,
      "loss": 0.6881,
      "step": 255
    },
    {
      "epoch": 0.32768,
      "grad_norm": 0.6181684732437134,
      "learning_rate": 3.988418083110738e-05,
      "loss": 0.6787,
      "step": 256
    },
    {
      "epoch": 0.32896,
      "grad_norm": 0.6393156051635742,
      "learning_rate": 3.9882485718405054e-05,
      "loss": 0.7067,
      "step": 257
    },
    {
      "epoch": 0.33024,
      "grad_norm": 0.6367058157920837,
      "learning_rate": 3.988077833156619e-05,
      "loss": 0.7334,
      "step": 258
    },
    {
      "epoch": 0.33152,
      "grad_norm": 0.6381621360778809,
      "learning_rate": 3.9879058671762695e-05,
      "loss": 0.6963,
      "step": 259
    },
    {
      "epoch": 0.3328,
      "grad_norm": 0.5958847999572754,
      "learning_rate": 3.987732674017491e-05,
      "loss": 0.6991,
      "step": 260
    },
    {
      "epoch": 0.33408,
      "grad_norm": 0.7162778973579407,
      "learning_rate": 3.98755825379916e-05,
      "loss": 0.7111,
      "step": 261
    },
    {
      "epoch": 0.33536,
      "grad_norm": 0.5961620807647705,
      "learning_rate": 3.987382606640993e-05,
      "loss": 0.6884,
      "step": 262
    },
    {
      "epoch": 0.33664,
      "grad_norm": 0.5822495222091675,
      "learning_rate": 3.9872057326635525e-05,
      "loss": 0.6905,
      "step": 263
    },
    {
      "epoch": 0.33792,
      "grad_norm": 0.6093654632568359,
      "learning_rate": 3.987027631988239e-05,
      "loss": 0.7056,
      "step": 264
    },
    {
      "epoch": 0.3392,
      "grad_norm": 0.586662769317627,
      "learning_rate": 3.9868483047372974e-05,
      "loss": 0.6826,
      "step": 265
    },
    {
      "epoch": 0.34048,
      "grad_norm": 0.7516249418258667,
      "learning_rate": 3.9866677510338136e-05,
      "loss": 0.7076,
      "step": 266
    },
    {
      "epoch": 0.34176,
      "grad_norm": 0.674939751625061,
      "learning_rate": 3.986485971001716e-05,
      "loss": 0.7475,
      "step": 267
    },
    {
      "epoch": 0.34304,
      "grad_norm": 0.6539055109024048,
      "learning_rate": 3.986302964765774e-05,
      "loss": 0.6582,
      "step": 268
    },
    {
      "epoch": 0.34432,
      "grad_norm": 0.6921095252037048,
      "learning_rate": 3.9861187324516e-05,
      "loss": 0.6839,
      "step": 269
    },
    {
      "epoch": 0.3456,
      "grad_norm": 0.6666824221611023,
      "learning_rate": 3.985933274185646e-05,
      "loss": 0.7113,
      "step": 270
    },
    {
      "epoch": 0.3456,
      "eval_loss": 0.7319334149360657,
      "eval_runtime": 19.5976,
      "eval_samples_per_second": 25.615,
      "eval_steps_per_second": 3.215,
      "step": 270
    },
    {
      "epoch": 0.34688,
      "grad_norm": 0.6145063638687134,
      "learning_rate": 3.985746590095206e-05,
      "loss": 0.7013,
      "step": 271
    },
    {
      "epoch": 0.34816,
      "grad_norm": 0.6586883664131165,
      "learning_rate": 3.985558680308416e-05,
      "loss": 0.7018,
      "step": 272
    },
    {
      "epoch": 0.34944,
      "grad_norm": 0.5831804871559143,
      "learning_rate": 3.985369544954254e-05,
      "loss": 0.6788,
      "step": 273
    },
    {
      "epoch": 0.35072,
      "grad_norm": 0.5836542248725891,
      "learning_rate": 3.985179184162537e-05,
      "loss": 0.6745,
      "step": 274
    },
    {
      "epoch": 0.352,
      "grad_norm": 0.5316975116729736,
      "learning_rate": 3.984987598063925e-05,
      "loss": 0.7079,
      "step": 275
    },
    {
      "epoch": 0.35328,
      "grad_norm": 0.6127880811691284,
      "learning_rate": 3.984794786789919e-05,
      "loss": 0.6666,
      "step": 276
    },
    {
      "epoch": 0.35456,
      "grad_norm": 0.5104806423187256,
      "learning_rate": 3.9846007504728593e-05,
      "loss": 0.6921,
      "step": 277
    },
    {
      "epoch": 0.35584,
      "grad_norm": 0.5813133120536804,
      "learning_rate": 3.984405489245928e-05,
      "loss": 0.7116,
      "step": 278
    },
    {
      "epoch": 0.35712,
      "grad_norm": 0.5278999209403992,
      "learning_rate": 3.984209003243149e-05,
      "loss": 0.6825,
      "step": 279
    },
    {
      "epoch": 0.3584,
      "grad_norm": 0.5866681933403015,
      "learning_rate": 3.984011292599385e-05,
      "loss": 0.7064,
      "step": 280
    },
    {
      "epoch": 0.35968,
      "grad_norm": 0.6471099257469177,
      "learning_rate": 3.983812357450341e-05,
      "loss": 0.6994,
      "step": 281
    },
    {
      "epoch": 0.36096,
      "grad_norm": 0.6071333885192871,
      "learning_rate": 3.983612197932561e-05,
      "loss": 0.6975,
      "step": 282
    },
    {
      "epoch": 0.36224,
      "grad_norm": 0.7360506057739258,
      "learning_rate": 3.98341081418343e-05,
      "loss": 0.6861,
      "step": 283
    },
    {
      "epoch": 0.36352,
      "grad_norm": 0.6293416619300842,
      "learning_rate": 3.983208206341173e-05,
      "loss": 0.7041,
      "step": 284
    },
    {
      "epoch": 0.3648,
      "grad_norm": 0.7838633060455322,
      "learning_rate": 3.983004374544856e-05,
      "loss": 0.7048,
      "step": 285
    },
    {
      "epoch": 0.36608,
      "grad_norm": 0.589284360408783,
      "learning_rate": 3.982799318934385e-05,
      "loss": 0.6855,
      "step": 286
    },
    {
      "epoch": 0.36736,
      "grad_norm": 0.7723178863525391,
      "learning_rate": 3.9825930396505036e-05,
      "loss": 0.696,
      "step": 287
    },
    {
      "epoch": 0.36864,
      "grad_norm": 0.6082919239997864,
      "learning_rate": 3.982385536834799e-05,
      "loss": 0.6873,
      "step": 288
    },
    {
      "epoch": 0.36992,
      "grad_norm": 0.6642741560935974,
      "learning_rate": 3.9821768106296975e-05,
      "loss": 0.697,
      "step": 289
    },
    {
      "epoch": 0.3712,
      "grad_norm": 0.6599016785621643,
      "learning_rate": 3.981966861178462e-05,
      "loss": 0.6671,
      "step": 290
    },
    {
      "epoch": 0.37248,
      "grad_norm": 0.6891383528709412,
      "learning_rate": 3.981755688625197e-05,
      "loss": 0.7084,
      "step": 291
    },
    {
      "epoch": 0.37376,
      "grad_norm": 0.6851657629013062,
      "learning_rate": 3.981543293114849e-05,
      "loss": 0.7116,
      "step": 292
    },
    {
      "epoch": 0.37504,
      "grad_norm": 0.6203416585922241,
      "learning_rate": 3.981329674793198e-05,
      "loss": 0.6705,
      "step": 293
    },
    {
      "epoch": 0.37632,
      "grad_norm": 0.6133400797843933,
      "learning_rate": 3.98111483380687e-05,
      "loss": 0.6665,
      "step": 294
    },
    {
      "epoch": 0.3776,
      "grad_norm": 0.6637143492698669,
      "learning_rate": 3.9808987703033256e-05,
      "loss": 0.694,
      "step": 295
    },
    {
      "epoch": 0.37888,
      "grad_norm": 0.610291063785553,
      "learning_rate": 3.980681484430866e-05,
      "loss": 0.6958,
      "step": 296
    },
    {
      "epoch": 0.38016,
      "grad_norm": 0.5253980159759521,
      "learning_rate": 3.980462976338631e-05,
      "loss": 0.6814,
      "step": 297
    },
    {
      "epoch": 0.38144,
      "grad_norm": 0.6023359894752502,
      "learning_rate": 3.9802432461766006e-05,
      "loss": 0.701,
      "step": 298
    },
    {
      "epoch": 0.38272,
      "grad_norm": 0.6062554717063904,
      "learning_rate": 3.980022294095592e-05,
      "loss": 0.688,
      "step": 299
    },
    {
      "epoch": 0.384,
      "grad_norm": 0.6049817204475403,
      "learning_rate": 3.979800120247262e-05,
      "loss": 0.694,
      "step": 300
    },
    {
      "epoch": 0.384,
      "eval_loss": 0.7287601232528687,
      "eval_runtime": 19.6225,
      "eval_samples_per_second": 25.583,
      "eval_steps_per_second": 3.211,
      "step": 300
    },
    {
      "epoch": 0.38528,
      "grad_norm": 0.7107629179954529,
      "learning_rate": 3.979576724784105e-05,
      "loss": 0.6962,
      "step": 301
    },
    {
      "epoch": 0.38656,
      "grad_norm": 0.6316274404525757,
      "learning_rate": 3.9793521078594556e-05,
      "loss": 0.6981,
      "step": 302
    },
    {
      "epoch": 0.38784,
      "grad_norm": 0.7024062275886536,
      "learning_rate": 3.9791262696274854e-05,
      "loss": 0.6898,
      "step": 303
    },
    {
      "epoch": 0.38912,
      "grad_norm": 0.5016635060310364,
      "learning_rate": 3.9788992102432046e-05,
      "loss": 0.6602,
      "step": 304
    },
    {
      "epoch": 0.3904,
      "grad_norm": 0.6374686360359192,
      "learning_rate": 3.978670929862462e-05,
      "loss": 0.6877,
      "step": 305
    },
    {
      "epoch": 0.39168,
      "grad_norm": 0.5335717797279358,
      "learning_rate": 3.978441428641944e-05,
      "loss": 0.7204,
      "step": 306
    },
    {
      "epoch": 0.39296,
      "grad_norm": 0.6253091096878052,
      "learning_rate": 3.978210706739175e-05,
      "loss": 0.6462,
      "step": 307
    },
    {
      "epoch": 0.39424,
      "grad_norm": 0.5884978771209717,
      "learning_rate": 3.9779787643125174e-05,
      "loss": 0.6812,
      "step": 308
    },
    {
      "epoch": 0.39552,
      "grad_norm": 0.6102194786071777,
      "learning_rate": 3.977745601521171e-05,
      "loss": 0.7192,
      "step": 309
    },
    {
      "epoch": 0.3968,
      "grad_norm": 0.7592268586158752,
      "learning_rate": 3.9775112185251734e-05,
      "loss": 0.6889,
      "step": 310
    },
    {
      "epoch": 0.39808,
      "grad_norm": 0.6071754693984985,
      "learning_rate": 3.977275615485401e-05,
      "loss": 0.696,
      "step": 311
    },
    {
      "epoch": 0.39936,
      "grad_norm": 0.5783090591430664,
      "learning_rate": 3.9770387925635645e-05,
      "loss": 0.6728,
      "step": 312
    },
    {
      "epoch": 0.40064,
      "grad_norm": 0.5993475317955017,
      "learning_rate": 3.976800749922215e-05,
      "loss": 0.7099,
      "step": 313
    },
    {
      "epoch": 0.40192,
      "grad_norm": 0.6230868101119995,
      "learning_rate": 3.97656148772474e-05,
      "loss": 0.6572,
      "step": 314
    },
    {
      "epoch": 0.4032,
      "grad_norm": 0.5437166094779968,
      "learning_rate": 3.976321006135363e-05,
      "loss": 0.6567,
      "step": 315
    },
    {
      "epoch": 0.40448,
      "grad_norm": 0.6500023603439331,
      "learning_rate": 3.9760793053191455e-05,
      "loss": 0.719,
      "step": 316
    },
    {
      "epoch": 0.40576,
      "grad_norm": 0.6185916066169739,
      "learning_rate": 3.975836385441986e-05,
      "loss": 0.677,
      "step": 317
    },
    {
      "epoch": 0.40704,
      "grad_norm": 0.6805869936943054,
      "learning_rate": 3.975592246670618e-05,
      "loss": 0.7064,
      "step": 318
    },
    {
      "epoch": 0.40832,
      "grad_norm": 0.5612990856170654,
      "learning_rate": 3.975346889172613e-05,
      "loss": 0.6647,
      "step": 319
    },
    {
      "epoch": 0.4096,
      "grad_norm": 0.6046521067619324,
      "learning_rate": 3.9751003131163805e-05,
      "loss": 0.6905,
      "step": 320
    },
    {
      "epoch": 0.41088,
      "grad_norm": 0.5969988703727722,
      "learning_rate": 3.974852518671163e-05,
      "loss": 0.7239,
      "step": 321
    },
    {
      "epoch": 0.41216,
      "grad_norm": 0.6520891189575195,
      "learning_rate": 3.974603506007042e-05,
      "loss": 0.6966,
      "step": 322
    },
    {
      "epoch": 0.41344,
      "grad_norm": 0.5954484939575195,
      "learning_rate": 3.9743532752949336e-05,
      "loss": 0.6918,
      "step": 323
    },
    {
      "epoch": 0.41472,
      "grad_norm": 0.571968674659729,
      "learning_rate": 3.9741018267065904e-05,
      "loss": 0.6879,
      "step": 324
    },
    {
      "epoch": 0.416,
      "grad_norm": 0.6033818125724792,
      "learning_rate": 3.973849160414603e-05,
      "loss": 0.6935,
      "step": 325
    },
    {
      "epoch": 0.41728,
      "grad_norm": 0.5529316663742065,
      "learning_rate": 3.9735952765923934e-05,
      "loss": 0.6807,
      "step": 326
    },
    {
      "epoch": 0.41856,
      "grad_norm": 0.5921478271484375,
      "learning_rate": 3.973340175414222e-05,
      "loss": 0.6997,
      "step": 327
    },
    {
      "epoch": 0.41984,
      "grad_norm": 0.5832169651985168,
      "learning_rate": 3.973083857055186e-05,
      "loss": 0.6912,
      "step": 328
    },
    {
      "epoch": 0.42112,
      "grad_norm": 0.596961259841919,
      "learning_rate": 3.9728263216912164e-05,
      "loss": 0.6633,
      "step": 329
    },
    {
      "epoch": 0.4224,
      "grad_norm": 0.6029351353645325,
      "learning_rate": 3.972567569499078e-05,
      "loss": 0.6958,
      "step": 330
    },
    {
      "epoch": 0.4224,
      "eval_loss": 0.7257757782936096,
      "eval_runtime": 19.5944,
      "eval_samples_per_second": 25.62,
      "eval_steps_per_second": 3.215,
      "step": 330
    },
    {
      "epoch": 0.42368,
      "grad_norm": 0.5937738418579102,
      "learning_rate": 3.972307600656374e-05,
      "loss": 0.6983,
      "step": 331
    },
    {
      "epoch": 0.42496,
      "grad_norm": 0.6081376075744629,
      "learning_rate": 3.9720464153415416e-05,
      "loss": 0.7025,
      "step": 332
    },
    {
      "epoch": 0.42624,
      "grad_norm": 0.5982778668403625,
      "learning_rate": 3.97178401373385e-05,
      "loss": 0.6825,
      "step": 333
    },
    {
      "epoch": 0.42752,
      "grad_norm": 0.5828783512115479,
      "learning_rate": 3.971520396013409e-05,
      "loss": 0.6773,
      "step": 334
    },
    {
      "epoch": 0.4288,
      "grad_norm": 0.5288313627243042,
      "learning_rate": 3.9712555623611575e-05,
      "loss": 0.6932,
      "step": 335
    },
    {
      "epoch": 0.43008,
      "grad_norm": 0.5914018750190735,
      "learning_rate": 3.9709895129588726e-05,
      "loss": 0.7074,
      "step": 336
    },
    {
      "epoch": 0.43136,
      "grad_norm": 0.6078402400016785,
      "learning_rate": 3.970722247989164e-05,
      "loss": 0.6911,
      "step": 337
    },
    {
      "epoch": 0.43264,
      "grad_norm": 0.6958378553390503,
      "learning_rate": 3.9704537676354765e-05,
      "loss": 0.6992,
      "step": 338
    },
    {
      "epoch": 0.43392,
      "grad_norm": 0.6617015600204468,
      "learning_rate": 3.9701840720820894e-05,
      "loss": 0.7045,
      "step": 339
    },
    {
      "epoch": 0.4352,
      "grad_norm": 0.5927851796150208,
      "learning_rate": 3.9699131615141156e-05,
      "loss": 0.6996,
      "step": 340
    },
    {
      "epoch": 0.43648,
      "grad_norm": 0.6079074144363403,
      "learning_rate": 3.9696410361175014e-05,
      "loss": 0.6745,
      "step": 341
    },
    {
      "epoch": 0.43776,
      "grad_norm": 0.6404287815093994,
      "learning_rate": 3.9693676960790275e-05,
      "loss": 0.701,
      "step": 342
    },
    {
      "epoch": 0.43904,
      "grad_norm": 0.61528480052948,
      "learning_rate": 3.96909314158631e-05,
      "loss": 0.7006,
      "step": 343
    },
    {
      "epoch": 0.44032,
      "grad_norm": 0.5665475726127625,
      "learning_rate": 3.968817372827794e-05,
      "loss": 0.6756,
      "step": 344
    },
    {
      "epoch": 0.4416,
      "grad_norm": 0.6530044078826904,
      "learning_rate": 3.968540389992765e-05,
      "loss": 0.6937,
      "step": 345
    },
    {
      "epoch": 0.44288,
      "grad_norm": 0.6078184843063354,
      "learning_rate": 3.968262193271334e-05,
      "loss": 0.7183,
      "step": 346
    },
    {
      "epoch": 0.44416,
      "grad_norm": 0.6617047190666199,
      "learning_rate": 3.967982782854451e-05,
      "loss": 0.7179,
      "step": 347
    },
    {
      "epoch": 0.44544,
      "grad_norm": 0.5911338329315186,
      "learning_rate": 3.967702158933896e-05,
      "loss": 0.6544,
      "step": 348
    },
    {
      "epoch": 0.44672,
      "grad_norm": 0.6430078744888306,
      "learning_rate": 3.967420321702285e-05,
      "loss": 0.7036,
      "step": 349
    },
    {
      "epoch": 0.448,
      "grad_norm": 0.5376421213150024,
      "learning_rate": 3.967137271353061e-05,
      "loss": 0.717,
      "step": 350
    },
    {
      "epoch": 0.44928,
      "grad_norm": 0.623209536075592,
      "learning_rate": 3.9668530080805076e-05,
      "loss": 0.6985,
      "step": 351
    },
    {
      "epoch": 0.45056,
      "grad_norm": 0.5676966905593872,
      "learning_rate": 3.966567532079734e-05,
      "loss": 0.7104,
      "step": 352
    },
    {
      "epoch": 0.45184,
      "grad_norm": 0.658064603805542,
      "learning_rate": 3.966280843546685e-05,
      "loss": 0.713,
      "step": 353
    },
    {
      "epoch": 0.45312,
      "grad_norm": 0.6175843477249146,
      "learning_rate": 3.965992942678139e-05,
      "loss": 0.7099,
      "step": 354
    },
    {
      "epoch": 0.4544,
      "grad_norm": 0.5926572680473328,
      "learning_rate": 3.965703829671701e-05,
      "loss": 0.7067,
      "step": 355
    },
    {
      "epoch": 0.45568,
      "grad_norm": 0.6756188869476318,
      "learning_rate": 3.965413504725815e-05,
      "loss": 0.6908,
      "step": 356
    },
    {
      "epoch": 0.45696,
      "grad_norm": 0.5718277096748352,
      "learning_rate": 3.965121968039753e-05,
      "loss": 0.667,
      "step": 357
    },
    {
      "epoch": 0.45824,
      "grad_norm": 0.6279222965240479,
      "learning_rate": 3.9648292198136174e-05,
      "loss": 0.6883,
      "step": 358
    },
    {
      "epoch": 0.45952,
      "grad_norm": 0.6190250515937805,
      "learning_rate": 3.9645352602483456e-05,
      "loss": 0.6909,
      "step": 359
    },
    {
      "epoch": 0.4608,
      "grad_norm": 0.6703121662139893,
      "learning_rate": 3.964240089545704e-05,
      "loss": 0.7042,
      "step": 360
    },
    {
      "epoch": 0.4608,
      "eval_loss": 0.7227815985679626,
      "eval_runtime": 19.5866,
      "eval_samples_per_second": 25.63,
      "eval_steps_per_second": 3.216,
      "step": 360
    },
    {
      "epoch": 0.46208,
      "grad_norm": 0.5657631158828735,
      "learning_rate": 3.9639437079082926e-05,
      "loss": 0.6622,
      "step": 361
    },
    {
      "epoch": 0.46336,
      "grad_norm": 0.6389573216438293,
      "learning_rate": 3.963646115539539e-05,
      "loss": 0.7135,
      "step": 362
    },
    {
      "epoch": 0.46464,
      "grad_norm": 0.7077627182006836,
      "learning_rate": 3.9633473126437054e-05,
      "loss": 0.7222,
      "step": 363
    },
    {
      "epoch": 0.46592,
      "grad_norm": 0.5947174429893494,
      "learning_rate": 3.963047299425884e-05,
      "loss": 0.671,
      "step": 364
    },
    {
      "epoch": 0.4672,
      "grad_norm": 0.6566045880317688,
      "learning_rate": 3.962746076091994e-05,
      "loss": 0.6885,
      "step": 365
    },
    {
      "epoch": 0.46848,
      "grad_norm": 0.5595127940177917,
      "learning_rate": 3.9624436428487925e-05,
      "loss": 0.6918,
      "step": 366
    },
    {
      "epoch": 0.46976,
      "grad_norm": 0.6398352384567261,
      "learning_rate": 3.96213999990386e-05,
      "loss": 0.6491,
      "step": 367
    },
    {
      "epoch": 0.47104,
      "grad_norm": 0.5487276315689087,
      "learning_rate": 3.961835147465611e-05,
      "loss": 0.6972,
      "step": 368
    },
    {
      "epoch": 0.47232,
      "grad_norm": 0.6809778809547424,
      "learning_rate": 3.961529085743289e-05,
      "loss": 0.7026,
      "step": 369
    },
    {
      "epoch": 0.4736,
      "grad_norm": 0.6166555881500244,
      "learning_rate": 3.961221814946969e-05,
      "loss": 0.6509,
      "step": 370
    },
    {
      "epoch": 0.47488,
      "grad_norm": 0.583348274230957,
      "learning_rate": 3.960913335287554e-05,
      "loss": 0.692,
      "step": 371
    },
    {
      "epoch": 0.47616,
      "grad_norm": 0.645103394985199,
      "learning_rate": 3.960603646976777e-05,
      "loss": 0.6845,
      "step": 372
    },
    {
      "epoch": 0.47744,
      "grad_norm": 0.6492411494255066,
      "learning_rate": 3.9602927502272025e-05,
      "loss": 0.6863,
      "step": 373
    },
    {
      "epoch": 0.47872,
      "grad_norm": 0.6270129680633545,
      "learning_rate": 3.959980645252222e-05,
      "loss": 0.6937,
      "step": 374
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.6296427249908447,
      "learning_rate": 3.959667332266057e-05,
      "loss": 0.6845,
      "step": 375
    },
    {
      "epoch": 0.48128,
      "grad_norm": 0.6152610182762146,
      "learning_rate": 3.95935281148376e-05,
      "loss": 0.6634,
      "step": 376
    },
    {
      "epoch": 0.48256,
      "grad_norm": 0.6302235126495361,
      "learning_rate": 3.959037083121209e-05,
      "loss": 0.6914,
      "step": 377
    },
    {
      "epoch": 0.48384,
      "grad_norm": 0.6253993511199951,
      "learning_rate": 3.958720147395115e-05,
      "loss": 0.6804,
      "step": 378
    },
    {
      "epoch": 0.48512,
      "grad_norm": 0.6146043539047241,
      "learning_rate": 3.958402004523014e-05,
      "loss": 0.7009,
      "step": 379
    },
    {
      "epoch": 0.4864,
      "grad_norm": 0.5274761319160461,
      "learning_rate": 3.958082654723273e-05,
      "loss": 0.6973,
      "step": 380
    },
    {
      "epoch": 0.48768,
      "grad_norm": 0.5693843364715576,
      "learning_rate": 3.957762098215085e-05,
      "loss": 0.6738,
      "step": 381
    },
    {
      "epoch": 0.48896,
      "grad_norm": 0.5581604242324829,
      "learning_rate": 3.9574403352184746e-05,
      "loss": 0.6991,
      "step": 382
    },
    {
      "epoch": 0.49024,
      "grad_norm": 0.5842587351799011,
      "learning_rate": 3.957117365954292e-05,
      "loss": 0.6926,
      "step": 383
    },
    {
      "epoch": 0.49152,
      "grad_norm": 0.5771941542625427,
      "learning_rate": 3.956793190644216e-05,
      "loss": 0.6454,
      "step": 384
    },
    {
      "epoch": 0.4928,
      "grad_norm": 0.6340916752815247,
      "learning_rate": 3.956467809510753e-05,
      "loss": 0.6997,
      "step": 385
    },
    {
      "epoch": 0.49408,
      "grad_norm": 0.5437546372413635,
      "learning_rate": 3.956141222777239e-05,
      "loss": 0.6846,
      "step": 386
    },
    {
      "epoch": 0.49536,
      "grad_norm": 0.5618076920509338,
      "learning_rate": 3.955813430667833e-05,
      "loss": 0.6737,
      "step": 387
    },
    {
      "epoch": 0.49664,
      "grad_norm": 0.5618740916252136,
      "learning_rate": 3.955484433407526e-05,
      "loss": 0.6609,
      "step": 388
    },
    {
      "epoch": 0.49792,
      "grad_norm": 0.4951269328594208,
      "learning_rate": 3.955154231222134e-05,
      "loss": 0.6405,
      "step": 389
    },
    {
      "epoch": 0.4992,
      "grad_norm": 0.5335937738418579,
      "learning_rate": 3.9548228243383e-05,
      "loss": 0.6991,
      "step": 390
    },
    {
      "epoch": 0.4992,
      "eval_loss": 0.7202399969100952,
      "eval_runtime": 19.5682,
      "eval_samples_per_second": 25.654,
      "eval_steps_per_second": 3.22,
      "step": 390
    },
    {
      "epoch": 0.50048,
      "grad_norm": 0.5206847786903381,
      "learning_rate": 3.9544902129834945e-05,
      "loss": 0.6852,
      "step": 391
    },
    {
      "epoch": 0.50176,
      "grad_norm": 0.5636817216873169,
      "learning_rate": 3.9541563973860146e-05,
      "loss": 0.7226,
      "step": 392
    },
    {
      "epoch": 0.50304,
      "grad_norm": 0.5307664275169373,
      "learning_rate": 3.953821377774984e-05,
      "loss": 0.6679,
      "step": 393
    },
    {
      "epoch": 0.50432,
      "grad_norm": 0.5746545195579529,
      "learning_rate": 3.953485154380352e-05,
      "loss": 0.6591,
      "step": 394
    },
    {
      "epoch": 0.5056,
      "grad_norm": 0.5888547897338867,
      "learning_rate": 3.953147727432896e-05,
      "loss": 0.694,
      "step": 395
    },
    {
      "epoch": 0.50688,
      "grad_norm": 0.5708815455436707,
      "learning_rate": 3.952809097164216e-05,
      "loss": 0.6809,
      "step": 396
    },
    {
      "epoch": 0.50816,
      "grad_norm": 0.6028029322624207,
      "learning_rate": 3.952469263806742e-05,
      "loss": 0.7125,
      "step": 397
    },
    {
      "epoch": 0.50944,
      "grad_norm": 0.5291039347648621,
      "learning_rate": 3.952128227593728e-05,
      "loss": 0.6677,
      "step": 398
    },
    {
      "epoch": 0.51072,
      "grad_norm": 0.6337804198265076,
      "learning_rate": 3.951785988759253e-05,
      "loss": 0.6845,
      "step": 399
    },
    {
      "epoch": 0.512,
      "grad_norm": 0.5936557650566101,
      "learning_rate": 3.9514425475382216e-05,
      "loss": 0.6977,
      "step": 400
    },
    {
      "epoch": 0.51328,
      "grad_norm": 0.6255313754081726,
      "learning_rate": 3.951097904166366e-05,
      "loss": 0.6505,
      "step": 401
    },
    {
      "epoch": 0.51456,
      "grad_norm": 0.5913993120193481,
      "learning_rate": 3.9507520588802386e-05,
      "loss": 0.6434,
      "step": 402
    },
    {
      "epoch": 0.51584,
      "grad_norm": 0.6816836595535278,
      "learning_rate": 3.950405011917223e-05,
      "loss": 0.6957,
      "step": 403
    },
    {
      "epoch": 0.51712,
      "grad_norm": 0.5961104035377502,
      "learning_rate": 3.9500567635155234e-05,
      "loss": 0.6839,
      "step": 404
    },
    {
      "epoch": 0.5184,
      "grad_norm": 0.6731733679771423,
      "learning_rate": 3.949707313914169e-05,
      "loss": 0.706,
      "step": 405
    },
    {
      "epoch": 0.51968,
      "grad_norm": 0.5992302894592285,
      "learning_rate": 3.9493566633530146e-05,
      "loss": 0.691,
      "step": 406
    },
    {
      "epoch": 0.52096,
      "grad_norm": 0.6435731649398804,
      "learning_rate": 3.949004812072738e-05,
      "loss": 0.7285,
      "step": 407
    },
    {
      "epoch": 0.52224,
      "grad_norm": 0.5671555399894714,
      "learning_rate": 3.948651760314844e-05,
      "loss": 0.6769,
      "step": 408
    },
    {
      "epoch": 0.52352,
      "grad_norm": 0.6329907178878784,
      "learning_rate": 3.9482975083216575e-05,
      "loss": 0.665,
      "step": 409
    },
    {
      "epoch": 0.5248,
      "grad_norm": 0.5716051459312439,
      "learning_rate": 3.94794205633633e-05,
      "loss": 0.6845,
      "step": 410
    },
    {
      "epoch": 0.52608,
      "grad_norm": 0.5738784074783325,
      "learning_rate": 3.947585404602836e-05,
      "loss": 0.6838,
      "step": 411
    },
    {
      "epoch": 0.52736,
      "grad_norm": 0.5675833821296692,
      "learning_rate": 3.947227553365973e-05,
      "loss": 0.699,
      "step": 412
    },
    {
      "epoch": 0.52864,
      "grad_norm": 0.5291326642036438,
      "learning_rate": 3.9468685028713607e-05,
      "loss": 0.7165,
      "step": 413
    },
    {
      "epoch": 0.52992,
      "grad_norm": 0.6508097052574158,
      "learning_rate": 3.9465082533654453e-05,
      "loss": 0.674,
      "step": 414
    },
    {
      "epoch": 0.5312,
      "grad_norm": 0.6031058430671692,
      "learning_rate": 3.9461468050954935e-05,
      "loss": 0.6832,
      "step": 415
    },
    {
      "epoch": 0.53248,
      "grad_norm": 0.6708024740219116,
      "learning_rate": 3.945784158309594e-05,
      "loss": 0.6623,
      "step": 416
    },
    {
      "epoch": 0.53376,
      "grad_norm": 0.5872268676757812,
      "learning_rate": 3.945420313256661e-05,
      "loss": 0.6761,
      "step": 417
    },
    {
      "epoch": 0.53504,
      "grad_norm": 0.6734101176261902,
      "learning_rate": 3.9450552701864294e-05,
      "loss": 0.6567,
      "step": 418
    },
    {
      "epoch": 0.53632,
      "grad_norm": 0.5773875713348389,
      "learning_rate": 3.9446890293494554e-05,
      "loss": 0.7148,
      "step": 419
    },
    {
      "epoch": 0.5376,
      "grad_norm": 0.6625387668609619,
      "learning_rate": 3.944321590997119e-05,
      "loss": 0.6561,
      "step": 420
    },
    {
      "epoch": 0.5376,
      "eval_loss": 0.7188262343406677,
      "eval_runtime": 19.5724,
      "eval_samples_per_second": 25.648,
      "eval_steps_per_second": 3.219,
      "step": 420
    },
    {
      "epoch": 0.53888,
      "grad_norm": 0.5858486890792847,
      "learning_rate": 3.943952955381622e-05,
      "loss": 0.7004,
      "step": 421
    },
    {
      "epoch": 0.54016,
      "grad_norm": 0.6575307846069336,
      "learning_rate": 3.943583122755987e-05,
      "loss": 0.7175,
      "step": 422
    },
    {
      "epoch": 0.54144,
      "grad_norm": 0.624356746673584,
      "learning_rate": 3.94321209337406e-05,
      "loss": 0.7068,
      "step": 423
    },
    {
      "epoch": 0.54272,
      "grad_norm": 0.6155917048454285,
      "learning_rate": 3.942839867490506e-05,
      "loss": 0.6794,
      "step": 424
    },
    {
      "epoch": 0.544,
      "grad_norm": 0.6458774209022522,
      "learning_rate": 3.942466445360813e-05,
      "loss": 0.696,
      "step": 425
    },
    {
      "epoch": 0.54528,
      "grad_norm": 0.5345732569694519,
      "learning_rate": 3.9420918272412894e-05,
      "loss": 0.6726,
      "step": 426
    },
    {
      "epoch": 0.54656,
      "grad_norm": 0.5263603329658508,
      "learning_rate": 3.941716013389065e-05,
      "loss": 0.6786,
      "step": 427
    },
    {
      "epoch": 0.54784,
      "grad_norm": 0.5550932288169861,
      "learning_rate": 3.941339004062089e-05,
      "loss": 0.6994,
      "step": 428
    },
    {
      "epoch": 0.54912,
      "grad_norm": 0.5684278011322021,
      "learning_rate": 3.940960799519134e-05,
      "loss": 0.6977,
      "step": 429
    },
    {
      "epoch": 0.5504,
      "grad_norm": 0.5063207149505615,
      "learning_rate": 3.9405814000197894e-05,
      "loss": 0.66,
      "step": 430
    },
    {
      "epoch": 0.55168,
      "grad_norm": 0.6308528184890747,
      "learning_rate": 3.940200805824468e-05,
      "loss": 0.6713,
      "step": 431
    },
    {
      "epoch": 0.55296,
      "grad_norm": 0.5000202655792236,
      "learning_rate": 3.9398190171943986e-05,
      "loss": 0.644,
      "step": 432
    },
    {
      "epoch": 0.55424,
      "grad_norm": 0.6493459343910217,
      "learning_rate": 3.939436034391634e-05,
      "loss": 0.6793,
      "step": 433
    },
    {
      "epoch": 0.55552,
      "grad_norm": 0.5449095964431763,
      "learning_rate": 3.939051857679046e-05,
      "loss": 0.6762,
      "step": 434
    },
    {
      "epoch": 0.5568,
      "grad_norm": 0.5830295085906982,
      "learning_rate": 3.938666487320323e-05,
      "loss": 0.6738,
      "step": 435
    },
    {
      "epoch": 0.55808,
      "grad_norm": 0.5074776411056519,
      "learning_rate": 3.938279923579976e-05,
      "loss": 0.6671,
      "step": 436
    },
    {
      "epoch": 0.55936,
      "grad_norm": 0.5655557513237,
      "learning_rate": 3.937892166723332e-05,
      "loss": 0.6813,
      "step": 437
    },
    {
      "epoch": 0.56064,
      "grad_norm": 0.588187038898468,
      "learning_rate": 3.9375032170165405e-05,
      "loss": 0.6795,
      "step": 438
    },
    {
      "epoch": 0.56192,
      "grad_norm": 0.47244301438331604,
      "learning_rate": 3.937113074726567e-05,
      "loss": 0.6662,
      "step": 439
    },
    {
      "epoch": 0.5632,
      "grad_norm": 0.5566319227218628,
      "learning_rate": 3.936721740121196e-05,
      "loss": 0.653,
      "step": 440
    },
    {
      "epoch": 0.56448,
      "grad_norm": 0.549518346786499,
      "learning_rate": 3.9363292134690315e-05,
      "loss": 0.6907,
      "step": 441
    },
    {
      "epoch": 0.56576,
      "grad_norm": 0.5767434239387512,
      "learning_rate": 3.935935495039494e-05,
      "loss": 0.6638,
      "step": 442
    },
    {
      "epoch": 0.56704,
      "grad_norm": 0.5820361971855164,
      "learning_rate": 3.935540585102824e-05,
      "loss": 0.7077,
      "step": 443
    },
    {
      "epoch": 0.56832,
      "grad_norm": 0.4665572941303253,
      "learning_rate": 3.935144483930078e-05,
      "loss": 0.7076,
      "step": 444
    },
    {
      "epoch": 0.5696,
      "grad_norm": 0.5414321422576904,
      "learning_rate": 3.9347471917931316e-05,
      "loss": 0.6995,
      "step": 445
    },
    {
      "epoch": 0.57088,
      "grad_norm": 0.48185208439826965,
      "learning_rate": 3.9343487089646774e-05,
      "loss": 0.7012,
      "step": 446
    },
    {
      "epoch": 0.57216,
      "grad_norm": 0.5999348759651184,
      "learning_rate": 3.933949035718224e-05,
      "loss": 0.6776,
      "step": 447
    },
    {
      "epoch": 0.57344,
      "grad_norm": 0.4932726323604584,
      "learning_rate": 3.933548172328099e-05,
      "loss": 0.6717,
      "step": 448
    },
    {
      "epoch": 0.57472,
      "grad_norm": 0.6640354990959167,
      "learning_rate": 3.933146119069446e-05,
      "loss": 0.6594,
      "step": 449
    },
    {
      "epoch": 0.576,
      "grad_norm": 0.531021237373352,
      "learning_rate": 3.9327428762182244e-05,
      "loss": 0.6646,
      "step": 450
    },
    {
      "epoch": 0.576,
      "eval_loss": 0.7171094417572021,
      "eval_runtime": 19.577,
      "eval_samples_per_second": 25.642,
      "eval_steps_per_second": 3.218,
      "step": 450
    },
    {
      "epoch": 0.57728,
      "grad_norm": 0.6199612617492676,
      "learning_rate": 3.932338444051213e-05,
      "loss": 0.6655,
      "step": 451
    },
    {
      "epoch": 0.57856,
      "grad_norm": 0.5710664391517639,
      "learning_rate": 3.9319328228460024e-05,
      "loss": 0.6703,
      "step": 452
    },
    {
      "epoch": 0.57984,
      "grad_norm": 0.6318474411964417,
      "learning_rate": 3.931526012881004e-05,
      "loss": 0.6987,
      "step": 453
    },
    {
      "epoch": 0.58112,
      "grad_norm": 0.6198699474334717,
      "learning_rate": 3.931118014435442e-05,
      "loss": 0.6707,
      "step": 454
    },
    {
      "epoch": 0.5824,
      "grad_norm": 0.5880622267723083,
      "learning_rate": 3.930708827789357e-05,
      "loss": 0.6716,
      "step": 455
    },
    {
      "epoch": 0.58368,
      "grad_norm": 0.5723118782043457,
      "learning_rate": 3.930298453223607e-05,
      "loss": 0.6651,
      "step": 456
    },
    {
      "epoch": 0.58496,
      "grad_norm": 0.6389984488487244,
      "learning_rate": 3.929886891019862e-05,
      "loss": 0.681,
      "step": 457
    },
    {
      "epoch": 0.58624,
      "grad_norm": 0.6376423835754395,
      "learning_rate": 3.929474141460611e-05,
      "loss": 0.7048,
      "step": 458
    },
    {
      "epoch": 0.58752,
      "grad_norm": 0.6106829643249512,
      "learning_rate": 3.929060204829155e-05,
      "loss": 0.65,
      "step": 459
    },
    {
      "epoch": 0.5888,
      "grad_norm": 0.673686683177948,
      "learning_rate": 3.9286450814096106e-05,
      "loss": 0.6656,
      "step": 460
    },
    {
      "epoch": 0.59008,
      "grad_norm": 0.6399636268615723,
      "learning_rate": 3.928228771486909e-05,
      "loss": 0.6881,
      "step": 461
    },
    {
      "epoch": 0.59136,
      "grad_norm": 0.6681706309318542,
      "learning_rate": 3.927811275346797e-05,
      "loss": 0.6686,
      "step": 462
    },
    {
      "epoch": 0.59264,
      "grad_norm": 0.5817286968231201,
      "learning_rate": 3.927392593275834e-05,
      "loss": 0.704,
      "step": 463
    },
    {
      "epoch": 0.59392,
      "grad_norm": 0.6522300243377686,
      "learning_rate": 3.926972725561394e-05,
      "loss": 0.6547,
      "step": 464
    },
    {
      "epoch": 0.5952,
      "grad_norm": 0.5608401298522949,
      "learning_rate": 3.926551672491665e-05,
      "loss": 0.6824,
      "step": 465
    },
    {
      "epoch": 0.59648,
      "grad_norm": 0.6735158562660217,
      "learning_rate": 3.926129434355647e-05,
      "loss": 0.669,
      "step": 466
    },
    {
      "epoch": 0.59776,
      "grad_norm": 0.5486454963684082,
      "learning_rate": 3.925706011443157e-05,
      "loss": 0.679,
      "step": 467
    },
    {
      "epoch": 0.59904,
      "grad_norm": 0.5701844096183777,
      "learning_rate": 3.925281404044821e-05,
      "loss": 0.6619,
      "step": 468
    },
    {
      "epoch": 0.60032,
      "grad_norm": 0.5257014632225037,
      "learning_rate": 3.924855612452082e-05,
      "loss": 0.6159,
      "step": 469
    },
    {
      "epoch": 0.6016,
      "grad_norm": 0.6779845952987671,
      "learning_rate": 3.924428636957192e-05,
      "loss": 0.6676,
      "step": 470
    },
    {
      "epoch": 0.60288,
      "grad_norm": 0.6293267607688904,
      "learning_rate": 3.924000477853218e-05,
      "loss": 0.6735,
      "step": 471
    },
    {
      "epoch": 0.60416,
      "grad_norm": 0.6325689554214478,
      "learning_rate": 3.923571135434039e-05,
      "loss": 0.6729,
      "step": 472
    },
    {
      "epoch": 0.60544,
      "grad_norm": 0.6593401432037354,
      "learning_rate": 3.923140609994346e-05,
      "loss": 0.6857,
      "step": 473
    },
    {
      "epoch": 0.60672,
      "grad_norm": 0.648293673992157,
      "learning_rate": 3.922708901829641e-05,
      "loss": 0.6892,
      "step": 474
    },
    {
      "epoch": 0.608,
      "grad_norm": 0.5773738026618958,
      "learning_rate": 3.92227601123624e-05,
      "loss": 0.672,
      "step": 475
    },
    {
      "epoch": 0.60928,
      "grad_norm": 0.7121182084083557,
      "learning_rate": 3.9218419385112693e-05,
      "loss": 0.685,
      "step": 476
    },
    {
      "epoch": 0.61056,
      "grad_norm": 0.6572108864784241,
      "learning_rate": 3.921406683952666e-05,
      "loss": 0.6871,
      "step": 477
    },
    {
      "epoch": 0.61184,
      "grad_norm": 0.709013044834137,
      "learning_rate": 3.920970247859179e-05,
      "loss": 0.6882,
      "step": 478
    },
    {
      "epoch": 0.61312,
      "grad_norm": 0.6492271423339844,
      "learning_rate": 3.9205326305303695e-05,
      "loss": 0.684,
      "step": 479
    },
    {
      "epoch": 0.6144,
      "grad_norm": 0.6692129969596863,
      "learning_rate": 3.920093832266607e-05,
      "loss": 0.6823,
      "step": 480
    },
    {
      "epoch": 0.6144,
      "eval_loss": 0.7130095362663269,
      "eval_runtime": 19.6288,
      "eval_samples_per_second": 25.575,
      "eval_steps_per_second": 3.21,
      "step": 480
    },
    {
      "epoch": 0.61568,
      "grad_norm": 0.5876495242118835,
      "learning_rate": 3.9196538533690734e-05,
      "loss": 0.6754,
      "step": 481
    },
    {
      "epoch": 0.61696,
      "grad_norm": 0.5922191739082336,
      "learning_rate": 3.91921269413976e-05,
      "loss": 0.6819,
      "step": 482
    },
    {
      "epoch": 0.61824,
      "grad_norm": 0.6059567332267761,
      "learning_rate": 3.918770354881468e-05,
      "loss": 0.6755,
      "step": 483
    },
    {
      "epoch": 0.61952,
      "grad_norm": 0.7806392908096313,
      "learning_rate": 3.9183268358978107e-05,
      "loss": 0.6873,
      "step": 484
    },
    {
      "epoch": 0.6208,
      "grad_norm": 0.5761833786964417,
      "learning_rate": 3.9178821374932076e-05,
      "loss": 0.6729,
      "step": 485
    },
    {
      "epoch": 0.62208,
      "grad_norm": 0.7536304593086243,
      "learning_rate": 3.917436259972891e-05,
      "loss": 0.6667,
      "step": 486
    },
    {
      "epoch": 0.62336,
      "grad_norm": 0.5851329565048218,
      "learning_rate": 3.9169892036429014e-05,
      "loss": 0.6424,
      "step": 487
    },
    {
      "epoch": 0.62464,
      "grad_norm": 0.7353109121322632,
      "learning_rate": 3.916540968810087e-05,
      "loss": 0.6799,
      "step": 488
    },
    {
      "epoch": 0.62592,
      "grad_norm": 0.5898123979568481,
      "learning_rate": 3.916091555782108e-05,
      "loss": 0.6631,
      "step": 489
    },
    {
      "epoch": 0.6272,
      "grad_norm": 0.6608595848083496,
      "learning_rate": 3.915640964867429e-05,
      "loss": 0.632,
      "step": 490
    },
    {
      "epoch": 0.62848,
      "grad_norm": 0.7140365242958069,
      "learning_rate": 3.915189196375327e-05,
      "loss": 0.6872,
      "step": 491
    },
    {
      "epoch": 0.62976,
      "grad_norm": 0.6551097631454468,
      "learning_rate": 3.914736250615886e-05,
      "loss": 0.6673,
      "step": 492
    },
    {
      "epoch": 0.63104,
      "grad_norm": 0.6790564656257629,
      "learning_rate": 3.914282127899998e-05,
      "loss": 0.6798,
      "step": 493
    },
    {
      "epoch": 0.63232,
      "grad_norm": 0.7467402815818787,
      "learning_rate": 3.9138268285393606e-05,
      "loss": 0.6946,
      "step": 494
    },
    {
      "epoch": 0.6336,
      "grad_norm": 0.6852535009384155,
      "learning_rate": 3.913370352846483e-05,
      "loss": 0.6663,
      "step": 495
    },
    {
      "epoch": 0.63488,
      "grad_norm": 0.6170719861984253,
      "learning_rate": 3.91291270113468e-05,
      "loss": 0.6459,
      "step": 496
    },
    {
      "epoch": 0.63616,
      "grad_norm": 0.6421520709991455,
      "learning_rate": 3.9124538737180714e-05,
      "loss": 0.643,
      "step": 497
    },
    {
      "epoch": 0.63744,
      "grad_norm": 0.5826987028121948,
      "learning_rate": 3.911993870911588e-05,
      "loss": 0.7147,
      "step": 498
    },
    {
      "epoch": 0.63872,
      "grad_norm": 0.6224043965339661,
      "learning_rate": 3.911532693030965e-05,
      "loss": 0.701,
      "step": 499
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.5442000031471252,
      "learning_rate": 3.9110703403927434e-05,
      "loss": 0.6621,
      "step": 500
    },
    {
      "epoch": 0.64128,
      "grad_norm": 0.5441639423370361,
      "learning_rate": 3.910606813314273e-05,
      "loss": 0.6441,
      "step": 501
    },
    {
      "epoch": 0.64256,
      "grad_norm": 0.4996682107448578,
      "learning_rate": 3.9101421121137084e-05,
      "loss": 0.6891,
      "step": 502
    },
    {
      "epoch": 0.64384,
      "grad_norm": 0.5748485326766968,
      "learning_rate": 3.9096762371100094e-05,
      "loss": 0.6971,
      "step": 503
    },
    {
      "epoch": 0.64512,
      "grad_norm": 0.5196304321289062,
      "learning_rate": 3.909209188622942e-05,
      "loss": 0.6368,
      "step": 504
    },
    {
      "epoch": 0.6464,
      "grad_norm": 0.5985638499259949,
      "learning_rate": 3.9087409669730777e-05,
      "loss": 0.6658,
      "step": 505
    },
    {
      "epoch": 0.64768,
      "grad_norm": 0.5176037549972534,
      "learning_rate": 3.908271572481795e-05,
      "loss": 0.6931,
      "step": 506
    },
    {
      "epoch": 0.64896,
      "grad_norm": 0.5685039758682251,
      "learning_rate": 3.907801005471273e-05,
      "loss": 0.6843,
      "step": 507
    },
    {
      "epoch": 0.65024,
      "grad_norm": 0.5779643058776855,
      "learning_rate": 3.907329266264501e-05,
      "loss": 0.6864,
      "step": 508
    },
    {
      "epoch": 0.65152,
      "grad_norm": 0.5429275631904602,
      "learning_rate": 3.906856355185269e-05,
      "loss": 0.642,
      "step": 509
    },
    {
      "epoch": 0.6528,
      "grad_norm": 0.5564545392990112,
      "learning_rate": 3.906382272558172e-05,
      "loss": 0.6707,
      "step": 510
    },
    {
      "epoch": 0.6528,
      "eval_loss": 0.7112646698951721,
      "eval_runtime": 19.5809,
      "eval_samples_per_second": 25.637,
      "eval_steps_per_second": 3.217,
      "step": 510
    },
    {
      "epoch": 0.65408,
      "grad_norm": 0.5477917194366455,
      "learning_rate": 3.90590701870861e-05,
      "loss": 0.6717,
      "step": 511
    },
    {
      "epoch": 0.65536,
      "grad_norm": 0.5378879904747009,
      "learning_rate": 3.9054305939627875e-05,
      "loss": 0.6737,
      "step": 512
    },
    {
      "epoch": 0.65664,
      "grad_norm": 0.5589343309402466,
      "learning_rate": 3.904952998647711e-05,
      "loss": 0.697,
      "step": 513
    },
    {
      "epoch": 0.65792,
      "grad_norm": 0.6126993894577026,
      "learning_rate": 3.9044742330911904e-05,
      "loss": 0.6749,
      "step": 514
    },
    {
      "epoch": 0.6592,
      "grad_norm": 0.6147490739822388,
      "learning_rate": 3.9039942976218415e-05,
      "loss": 0.6998,
      "step": 515
    },
    {
      "epoch": 0.66048,
      "grad_norm": 0.5947819948196411,
      "learning_rate": 3.903513192569079e-05,
      "loss": 0.6811,
      "step": 516
    },
    {
      "epoch": 0.66176,
      "grad_norm": 0.6416747570037842,
      "learning_rate": 3.903030918263124e-05,
      "loss": 0.681,
      "step": 517
    },
    {
      "epoch": 0.66304,
      "grad_norm": 0.571851909160614,
      "learning_rate": 3.9025474750349994e-05,
      "loss": 0.7122,
      "step": 518
    },
    {
      "epoch": 0.66432,
      "grad_norm": 0.637311577796936,
      "learning_rate": 3.902062863216528e-05,
      "loss": 0.6744,
      "step": 519
    },
    {
      "epoch": 0.6656,
      "grad_norm": 0.4946390688419342,
      "learning_rate": 3.9015770831403385e-05,
      "loss": 0.6577,
      "step": 520
    },
    {
      "epoch": 0.66688,
      "grad_norm": 0.6272170543670654,
      "learning_rate": 3.9010901351398576e-05,
      "loss": 0.6527,
      "step": 521
    },
    {
      "epoch": 0.66816,
      "grad_norm": 0.5592139363288879,
      "learning_rate": 3.900602019549316e-05,
      "loss": 0.6642,
      "step": 522
    },
    {
      "epoch": 0.66944,
      "grad_norm": 0.7104748487472534,
      "learning_rate": 3.900112736703746e-05,
      "loss": 0.6785,
      "step": 523
    },
    {
      "epoch": 0.67072,
      "grad_norm": 0.576612651348114,
      "learning_rate": 3.8996222869389784e-05,
      "loss": 0.6801,
      "step": 524
    },
    {
      "epoch": 0.672,
      "grad_norm": 0.6686196327209473,
      "learning_rate": 3.89913067059165e-05,
      "loss": 0.6712,
      "step": 525
    },
    {
      "epoch": 0.67328,
      "grad_norm": 0.5267242789268494,
      "learning_rate": 3.8986378879991924e-05,
      "loss": 0.6761,
      "step": 526
    },
    {
      "epoch": 0.67456,
      "grad_norm": 0.7319815754890442,
      "learning_rate": 3.8981439394998425e-05,
      "loss": 0.6776,
      "step": 527
    },
    {
      "epoch": 0.67584,
      "grad_norm": 0.5937008857727051,
      "learning_rate": 3.897648825432634e-05,
      "loss": 0.6946,
      "step": 528
    },
    {
      "epoch": 0.67712,
      "grad_norm": 0.6719886064529419,
      "learning_rate": 3.897152546137403e-05,
      "loss": 0.6533,
      "step": 529
    },
    {
      "epoch": 0.6784,
      "grad_norm": 0.6564842462539673,
      "learning_rate": 3.896655101954783e-05,
      "loss": 0.6328,
      "step": 530
    },
    {
      "epoch": 0.67968,
      "grad_norm": 0.5917929410934448,
      "learning_rate": 3.896156493226211e-05,
      "loss": 0.6096,
      "step": 531
    },
    {
      "epoch": 0.68096,
      "grad_norm": 0.6879952549934387,
      "learning_rate": 3.8956567202939196e-05,
      "loss": 0.6981,
      "step": 532
    },
    {
      "epoch": 0.68224,
      "grad_norm": 0.5507075190544128,
      "learning_rate": 3.895155783500941e-05,
      "loss": 0.6777,
      "step": 533
    },
    {
      "epoch": 0.68352,
      "grad_norm": 0.7529070377349854,
      "learning_rate": 3.8946536831911074e-05,
      "loss": 0.6817,
      "step": 534
    },
    {
      "epoch": 0.6848,
      "grad_norm": 0.4982717037200928,
      "learning_rate": 3.89415041970905e-05,
      "loss": 0.6644,
      "step": 535
    },
    {
      "epoch": 0.68608,
      "grad_norm": 0.7063624858856201,
      "learning_rate": 3.893645993400196e-05,
      "loss": 0.6633,
      "step": 536
    },
    {
      "epoch": 0.68736,
      "grad_norm": 0.6446038484573364,
      "learning_rate": 3.893140404610773e-05,
      "loss": 0.6911,
      "step": 537
    },
    {
      "epoch": 0.68864,
      "grad_norm": 0.6866803169250488,
      "learning_rate": 3.892633653687807e-05,
      "loss": 0.6888,
      "step": 538
    },
    {
      "epoch": 0.68992,
      "grad_norm": 0.5922403335571289,
      "learning_rate": 3.892125740979119e-05,
      "loss": 0.6749,
      "step": 539
    },
    {
      "epoch": 0.6912,
      "grad_norm": 0.5578379034996033,
      "learning_rate": 3.891616666833329e-05,
      "loss": 0.72,
      "step": 540
    },
    {
      "epoch": 0.6912,
      "eval_loss": 0.7098914980888367,
      "eval_runtime": 19.6033,
      "eval_samples_per_second": 25.608,
      "eval_steps_per_second": 3.214,
      "step": 540
    },
    {
      "epoch": 0.69248,
      "grad_norm": 0.6193397045135498,
      "learning_rate": 3.891106431599854e-05,
      "loss": 0.6594,
      "step": 541
    },
    {
      "epoch": 0.69376,
      "grad_norm": 0.4924187660217285,
      "learning_rate": 3.8905950356289095e-05,
      "loss": 0.6737,
      "step": 542
    },
    {
      "epoch": 0.69504,
      "grad_norm": 0.6688212752342224,
      "learning_rate": 3.890082479271504e-05,
      "loss": 0.6656,
      "step": 543
    },
    {
      "epoch": 0.69632,
      "grad_norm": 0.5867065787315369,
      "learning_rate": 3.889568762879446e-05,
      "loss": 0.7095,
      "step": 544
    },
    {
      "epoch": 0.6976,
      "grad_norm": 0.6487833261489868,
      "learning_rate": 3.889053886805339e-05,
      "loss": 0.634,
      "step": 545
    },
    {
      "epoch": 0.69888,
      "grad_norm": 0.5362858176231384,
      "learning_rate": 3.888537851402582e-05,
      "loss": 0.6422,
      "step": 546
    },
    {
      "epoch": 0.70016,
      "grad_norm": 0.7044833898544312,
      "learning_rate": 3.8880206570253694e-05,
      "loss": 0.6711,
      "step": 547
    },
    {
      "epoch": 0.70144,
      "grad_norm": 0.6324202418327332,
      "learning_rate": 3.8875023040286926e-05,
      "loss": 0.6747,
      "step": 548
    },
    {
      "epoch": 0.70272,
      "grad_norm": 0.5976712703704834,
      "learning_rate": 3.886982792768338e-05,
      "loss": 0.7266,
      "step": 549
    },
    {
      "epoch": 0.704,
      "grad_norm": 0.6165710091590881,
      "learning_rate": 3.886462123600885e-05,
      "loss": 0.6654,
      "step": 550
    },
    {
      "epoch": 0.70528,
      "grad_norm": 0.5874255895614624,
      "learning_rate": 3.8859402968837106e-05,
      "loss": 0.6658,
      "step": 551
    },
    {
      "epoch": 0.70656,
      "grad_norm": 0.5603928565979004,
      "learning_rate": 3.8854173129749846e-05,
      "loss": 0.6241,
      "step": 552
    },
    {
      "epoch": 0.70784,
      "grad_norm": 0.5711289048194885,
      "learning_rate": 3.8848931722336705e-05,
      "loss": 0.6594,
      "step": 553
    },
    {
      "epoch": 0.70912,
      "grad_norm": 0.6343725919723511,
      "learning_rate": 3.884367875019528e-05,
      "loss": 0.6886,
      "step": 554
    },
    {
      "epoch": 0.7104,
      "grad_norm": 0.5955490469932556,
      "learning_rate": 3.8838414216931076e-05,
      "loss": 0.6676,
      "step": 555
    },
    {
      "epoch": 0.71168,
      "grad_norm": 0.659564197063446,
      "learning_rate": 3.883313812615757e-05,
      "loss": 0.6777,
      "step": 556
    },
    {
      "epoch": 0.71296,
      "grad_norm": 0.6436776518821716,
      "learning_rate": 3.882785048149615e-05,
      "loss": 0.6936,
      "step": 557
    },
    {
      "epoch": 0.71424,
      "grad_norm": 0.6461824774742126,
      "learning_rate": 3.882255128657612e-05,
      "loss": 0.6759,
      "step": 558
    },
    {
      "epoch": 0.71552,
      "grad_norm": 0.5778611898422241,
      "learning_rate": 3.881724054503474e-05,
      "loss": 0.6672,
      "step": 559
    },
    {
      "epoch": 0.7168,
      "grad_norm": 0.5813550353050232,
      "learning_rate": 3.881191826051719e-05,
      "loss": 0.6437,
      "step": 560
    },
    {
      "epoch": 0.71808,
      "grad_norm": 0.6578578948974609,
      "learning_rate": 3.880658443667655e-05,
      "loss": 0.6722,
      "step": 561
    },
    {
      "epoch": 0.71936,
      "grad_norm": 0.5632246136665344,
      "learning_rate": 3.8801239077173847e-05,
      "loss": 0.6318,
      "step": 562
    },
    {
      "epoch": 0.72064,
      "grad_norm": 0.6688623428344727,
      "learning_rate": 3.879588218567802e-05,
      "loss": 0.6969,
      "step": 563
    },
    {
      "epoch": 0.72192,
      "grad_norm": 0.7001153826713562,
      "learning_rate": 3.879051376586591e-05,
      "loss": 0.6694,
      "step": 564
    },
    {
      "epoch": 0.7232,
      "grad_norm": 0.5424560904502869,
      "learning_rate": 3.878513382142228e-05,
      "loss": 0.6651,
      "step": 565
    },
    {
      "epoch": 0.72448,
      "grad_norm": 0.6218736171722412,
      "learning_rate": 3.8779742356039815e-05,
      "loss": 0.6977,
      "step": 566
    },
    {
      "epoch": 0.72576,
      "grad_norm": 0.6319977045059204,
      "learning_rate": 3.877433937341908e-05,
      "loss": 0.7277,
      "step": 567
    },
    {
      "epoch": 0.72704,
      "grad_norm": 0.5249434113502502,
      "learning_rate": 3.876892487726857e-05,
      "loss": 0.6756,
      "step": 568
    },
    {
      "epoch": 0.72832,
      "grad_norm": 0.619577944278717,
      "learning_rate": 3.876349887130467e-05,
      "loss": 0.6771,
      "step": 569
    },
    {
      "epoch": 0.7296,
      "grad_norm": 0.4973960220813751,
      "learning_rate": 3.8758061359251675e-05,
      "loss": 0.6707,
      "step": 570
    },
    {
      "epoch": 0.7296,
      "eval_loss": 0.7068474292755127,
      "eval_runtime": 19.575,
      "eval_samples_per_second": 25.645,
      "eval_steps_per_second": 3.218,
      "step": 570
    },
    {
      "epoch": 0.73088,
      "grad_norm": 0.6298916339874268,
      "learning_rate": 3.875261234484176e-05,
      "loss": 0.727,
      "step": 571
    },
    {
      "epoch": 0.73216,
      "grad_norm": 0.5114372968673706,
      "learning_rate": 3.874715183181502e-05,
      "loss": 0.6867,
      "step": 572
    },
    {
      "epoch": 0.73344,
      "grad_norm": 0.5665026307106018,
      "learning_rate": 3.8741679823919415e-05,
      "loss": 0.6639,
      "step": 573
    },
    {
      "epoch": 0.73472,
      "grad_norm": 0.4909069240093231,
      "learning_rate": 3.873619632491081e-05,
      "loss": 0.6355,
      "step": 574
    },
    {
      "epoch": 0.736,
      "grad_norm": 0.564416229724884,
      "learning_rate": 3.8730701338552965e-05,
      "loss": 0.6829,
      "step": 575
    },
    {
      "epoch": 0.73728,
      "grad_norm": 0.5256968140602112,
      "learning_rate": 3.872519486861752e-05,
      "loss": 0.6954,
      "step": 576
    },
    {
      "epoch": 0.73856,
      "grad_norm": 0.5014488101005554,
      "learning_rate": 3.871967691888397e-05,
      "loss": 0.675,
      "step": 577
    },
    {
      "epoch": 0.73984,
      "grad_norm": 0.6175253987312317,
      "learning_rate": 3.871414749313972e-05,
      "loss": 0.6729,
      "step": 578
    },
    {
      "epoch": 0.74112,
      "grad_norm": 0.5787009000778198,
      "learning_rate": 3.870860659518006e-05,
      "loss": 0.6742,
      "step": 579
    },
    {
      "epoch": 0.7424,
      "grad_norm": 0.5943652987480164,
      "learning_rate": 3.870305422880812e-05,
      "loss": 0.6736,
      "step": 580
    },
    {
      "epoch": 0.74368,
      "grad_norm": 0.5780078768730164,
      "learning_rate": 3.869749039783492e-05,
      "loss": 0.6556,
      "step": 581
    },
    {
      "epoch": 0.74496,
      "grad_norm": 0.5442749261856079,
      "learning_rate": 3.869191510607936e-05,
      "loss": 0.6842,
      "step": 582
    },
    {
      "epoch": 0.74624,
      "grad_norm": 0.6305468082427979,
      "learning_rate": 3.868632835736819e-05,
      "loss": 0.7086,
      "step": 583
    },
    {
      "epoch": 0.74752,
      "grad_norm": 0.4775775372982025,
      "learning_rate": 3.8680730155536024e-05,
      "loss": 0.7332,
      "step": 584
    },
    {
      "epoch": 0.7488,
      "grad_norm": 0.5703808069229126,
      "learning_rate": 3.8675120504425345e-05,
      "loss": 0.6516,
      "step": 585
    },
    {
      "epoch": 0.75008,
      "grad_norm": 0.5440189838409424,
      "learning_rate": 3.86694994078865e-05,
      "loss": 0.672,
      "step": 586
    },
    {
      "epoch": 0.75136,
      "grad_norm": 0.590533971786499,
      "learning_rate": 3.866386686977766e-05,
      "loss": 0.6925,
      "step": 587
    },
    {
      "epoch": 0.75264,
      "grad_norm": 0.5902044773101807,
      "learning_rate": 3.86582228939649e-05,
      "loss": 0.7004,
      "step": 588
    },
    {
      "epoch": 0.75392,
      "grad_norm": 0.5268256664276123,
      "learning_rate": 3.86525674843221e-05,
      "loss": 0.6637,
      "step": 589
    },
    {
      "epoch": 0.7552,
      "grad_norm": 0.5215715169906616,
      "learning_rate": 3.864690064473102e-05,
      "loss": 0.6444,
      "step": 590
    },
    {
      "epoch": 0.75648,
      "grad_norm": 0.5789802074432373,
      "learning_rate": 3.864122237908123e-05,
      "loss": 0.6712,
      "step": 591
    },
    {
      "epoch": 0.75776,
      "grad_norm": 0.5239521861076355,
      "learning_rate": 3.8635532691270196e-05,
      "loss": 0.671,
      "step": 592
    },
    {
      "epoch": 0.75904,
      "grad_norm": 0.5557687282562256,
      "learning_rate": 3.862983158520316e-05,
      "loss": 0.7021,
      "step": 593
    },
    {
      "epoch": 0.76032,
      "grad_norm": 0.5416465401649475,
      "learning_rate": 3.862411906479326e-05,
      "loss": 0.6775,
      "step": 594
    },
    {
      "epoch": 0.7616,
      "grad_norm": 0.5547333359718323,
      "learning_rate": 3.861839513396142e-05,
      "loss": 0.6834,
      "step": 595
    },
    {
      "epoch": 0.76288,
      "grad_norm": 0.5584958791732788,
      "learning_rate": 3.861265979663643e-05,
      "loss": 0.7008,
      "step": 596
    },
    {
      "epoch": 0.76416,
      "grad_norm": 0.5268974900245667,
      "learning_rate": 3.860691305675489e-05,
      "loss": 0.6551,
      "step": 597
    },
    {
      "epoch": 0.76544,
      "grad_norm": 0.5101964473724365,
      "learning_rate": 3.8601154918261235e-05,
      "loss": 0.6968,
      "step": 598
    },
    {
      "epoch": 0.76672,
      "grad_norm": 0.48142826557159424,
      "learning_rate": 3.859538538510772e-05,
      "loss": 0.6683,
      "step": 599
    },
    {
      "epoch": 0.768,
      "grad_norm": 0.5648514628410339,
      "learning_rate": 3.858960446125443e-05,
      "loss": 0.6746,
      "step": 600
    },
    {
      "epoch": 0.768,
      "eval_loss": 0.7071773409843445,
      "eval_runtime": 19.563,
      "eval_samples_per_second": 25.661,
      "eval_steps_per_second": 3.22,
      "step": 600
    },
    {
      "epoch": 0.76928,
      "grad_norm": 0.5475654602050781,
      "learning_rate": 3.858381215066926e-05,
      "loss": 0.668,
      "step": 601
    },
    {
      "epoch": 0.77056,
      "grad_norm": 0.6000843644142151,
      "learning_rate": 3.857800845732792e-05,
      "loss": 0.6454,
      "step": 602
    },
    {
      "epoch": 0.77184,
      "grad_norm": 0.5896192789077759,
      "learning_rate": 3.857219338521393e-05,
      "loss": 0.6738,
      "step": 603
    },
    {
      "epoch": 0.77312,
      "grad_norm": 0.603815495967865,
      "learning_rate": 3.856636693831863e-05,
      "loss": 0.6816,
      "step": 604
    },
    {
      "epoch": 0.7744,
      "grad_norm": 0.6198026537895203,
      "learning_rate": 3.856052912064116e-05,
      "loss": 0.6978,
      "step": 605
    },
    {
      "epoch": 0.77568,
      "grad_norm": 0.6077916026115417,
      "learning_rate": 3.855467993618847e-05,
      "loss": 0.6782,
      "step": 606
    },
    {
      "epoch": 0.77696,
      "grad_norm": 0.6056360006332397,
      "learning_rate": 3.854881938897531e-05,
      "loss": 0.688,
      "step": 607
    },
    {
      "epoch": 0.77824,
      "grad_norm": 0.6548430919647217,
      "learning_rate": 3.854294748302422e-05,
      "loss": 0.6455,
      "step": 608
    },
    {
      "epoch": 0.77952,
      "grad_norm": 0.5651297569274902,
      "learning_rate": 3.8537064222365545e-05,
      "loss": 0.6663,
      "step": 609
    },
    {
      "epoch": 0.7808,
      "grad_norm": 0.5523439645767212,
      "learning_rate": 3.853116961103744e-05,
      "loss": 0.6905,
      "step": 610
    },
    {
      "epoch": 0.78208,
      "grad_norm": 0.5329164266586304,
      "learning_rate": 3.852526365308581e-05,
      "loss": 0.6482,
      "step": 611
    },
    {
      "epoch": 0.78336,
      "grad_norm": 0.5494553446769714,
      "learning_rate": 3.851934635256438e-05,
      "loss": 0.6756,
      "step": 612
    },
    {
      "epoch": 0.78464,
      "grad_norm": 0.6761051416397095,
      "learning_rate": 3.8513417713534676e-05,
      "loss": 0.6223,
      "step": 613
    },
    {
      "epoch": 0.78592,
      "grad_norm": 0.5757414698600769,
      "learning_rate": 3.850747774006594e-05,
      "loss": 0.6784,
      "step": 614
    },
    {
      "epoch": 0.7872,
      "grad_norm": 0.601269006729126,
      "learning_rate": 3.850152643623527e-05,
      "loss": 0.6973,
      "step": 615
    },
    {
      "epoch": 0.78848,
      "grad_norm": 0.5504854917526245,
      "learning_rate": 3.849556380612749e-05,
      "loss": 0.6432,
      "step": 616
    },
    {
      "epoch": 0.78976,
      "grad_norm": 0.5931381583213806,
      "learning_rate": 3.848958985383522e-05,
      "loss": 0.6561,
      "step": 617
    },
    {
      "epoch": 0.79104,
      "grad_norm": 0.5603706240653992,
      "learning_rate": 3.848360458345885e-05,
      "loss": 0.6847,
      "step": 618
    },
    {
      "epoch": 0.79232,
      "grad_norm": 0.5120813250541687,
      "learning_rate": 3.847760799910654e-05,
      "loss": 0.678,
      "step": 619
    },
    {
      "epoch": 0.7936,
      "grad_norm": 0.527540385723114,
      "learning_rate": 3.847160010489419e-05,
      "loss": 0.6656,
      "step": 620
    },
    {
      "epoch": 0.79488,
      "grad_norm": 0.56707763671875,
      "learning_rate": 3.84655809049455e-05,
      "loss": 0.6491,
      "step": 621
    },
    {
      "epoch": 0.79616,
      "grad_norm": 0.6506879329681396,
      "learning_rate": 3.8459550403391916e-05,
      "loss": 0.7003,
      "step": 622
    },
    {
      "epoch": 0.79744,
      "grad_norm": 0.5204246640205383,
      "learning_rate": 3.8453508604372627e-05,
      "loss": 0.6757,
      "step": 623
    },
    {
      "epoch": 0.79872,
      "grad_norm": 0.6690553426742554,
      "learning_rate": 3.84474555120346e-05,
      "loss": 0.6508,
      "step": 624
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.5945473909378052,
      "learning_rate": 3.844139113053253e-05,
      "loss": 0.6537,
      "step": 625
    },
    {
      "epoch": 0.80128,
      "grad_norm": 0.553322434425354,
      "learning_rate": 3.843531546402889e-05,
      "loss": 0.6519,
      "step": 626
    },
    {
      "epoch": 0.80256,
      "grad_norm": 0.6136388778686523,
      "learning_rate": 3.8429228516693856e-05,
      "loss": 0.6974,
      "step": 627
    },
    {
      "epoch": 0.80384,
      "grad_norm": 0.6032586693763733,
      "learning_rate": 3.842313029270539e-05,
      "loss": 0.647,
      "step": 628
    },
    {
      "epoch": 0.80512,
      "grad_norm": 0.6201940774917603,
      "learning_rate": 3.841702079624918e-05,
      "loss": 0.6592,
      "step": 629
    },
    {
      "epoch": 0.8064,
      "grad_norm": 0.5738290548324585,
      "learning_rate": 3.841090003151863e-05,
      "loss": 0.6922,
      "step": 630
    },
    {
      "epoch": 0.8064,
      "eval_loss": 0.7047855854034424,
      "eval_runtime": 19.5703,
      "eval_samples_per_second": 25.651,
      "eval_steps_per_second": 3.219,
      "step": 630
    },
    {
      "epoch": 0.80768,
      "grad_norm": 0.5257089734077454,
      "learning_rate": 3.8404768002714904e-05,
      "loss": 0.6427,
      "step": 631
    },
    {
      "epoch": 0.80896,
      "grad_norm": 0.5059003829956055,
      "learning_rate": 3.839862471404689e-05,
      "loss": 0.6689,
      "step": 632
    },
    {
      "epoch": 0.81024,
      "grad_norm": 0.5169146656990051,
      "learning_rate": 3.8392470169731206e-05,
      "loss": 0.6424,
      "step": 633
    },
    {
      "epoch": 0.81152,
      "grad_norm": 0.5781958699226379,
      "learning_rate": 3.8386304373992185e-05,
      "loss": 0.6489,
      "step": 634
    },
    {
      "epoch": 0.8128,
      "grad_norm": 0.513576090335846,
      "learning_rate": 3.8380127331061895e-05,
      "loss": 0.6409,
      "step": 635
    },
    {
      "epoch": 0.81408,
      "grad_norm": 0.5294614434242249,
      "learning_rate": 3.837393904518012e-05,
      "loss": 0.6377,
      "step": 636
    },
    {
      "epoch": 0.81536,
      "grad_norm": 0.5955902934074402,
      "learning_rate": 3.836773952059436e-05,
      "loss": 0.668,
      "step": 637
    },
    {
      "epoch": 0.81664,
      "grad_norm": 0.6294943690299988,
      "learning_rate": 3.8361528761559834e-05,
      "loss": 0.6963,
      "step": 638
    },
    {
      "epoch": 0.81792,
      "grad_norm": 0.5575120449066162,
      "learning_rate": 3.835530677233946e-05,
      "loss": 0.6493,
      "step": 639
    },
    {
      "epoch": 0.8192,
      "grad_norm": 0.5605102777481079,
      "learning_rate": 3.8349073557203875e-05,
      "loss": 0.6646,
      "step": 640
    },
    {
      "epoch": 0.82048,
      "grad_norm": 0.5291647911071777,
      "learning_rate": 3.834282912043141e-05,
      "loss": 0.7001,
      "step": 641
    },
    {
      "epoch": 0.82176,
      "grad_norm": 0.5542118549346924,
      "learning_rate": 3.833657346630812e-05,
      "loss": 0.6155,
      "step": 642
    },
    {
      "epoch": 0.82304,
      "grad_norm": 0.5809057950973511,
      "learning_rate": 3.833030659912774e-05,
      "loss": 0.6546,
      "step": 643
    },
    {
      "epoch": 0.82432,
      "grad_norm": 0.6260568499565125,
      "learning_rate": 3.8324028523191703e-05,
      "loss": 0.6766,
      "step": 644
    },
    {
      "epoch": 0.8256,
      "grad_norm": 0.55931156873703,
      "learning_rate": 3.8317739242809144e-05,
      "loss": 0.661,
      "step": 645
    },
    {
      "epoch": 0.82688,
      "grad_norm": 0.6160814166069031,
      "learning_rate": 3.831143876229688e-05,
      "loss": 0.6672,
      "step": 646
    },
    {
      "epoch": 0.82816,
      "grad_norm": 0.5560668110847473,
      "learning_rate": 3.830512708597942e-05,
      "loss": 0.6382,
      "step": 647
    },
    {
      "epoch": 0.82944,
      "grad_norm": 0.6235952377319336,
      "learning_rate": 3.8298804218188956e-05,
      "loss": 0.6576,
      "step": 648
    },
    {
      "epoch": 0.83072,
      "grad_norm": 0.5899331569671631,
      "learning_rate": 3.8292470163265375e-05,
      "loss": 0.6287,
      "step": 649
    },
    {
      "epoch": 0.832,
      "grad_norm": 0.5356541872024536,
      "learning_rate": 3.8286124925556206e-05,
      "loss": 0.6717,
      "step": 650
    },
    {
      "epoch": 0.83328,
      "grad_norm": 0.5748461484909058,
      "learning_rate": 3.8279768509416705e-05,
      "loss": 0.6449,
      "step": 651
    },
    {
      "epoch": 0.83456,
      "grad_norm": 0.5012470483779907,
      "learning_rate": 3.827340091920975e-05,
      "loss": 0.643,
      "step": 652
    },
    {
      "epoch": 0.83584,
      "grad_norm": 0.5592685341835022,
      "learning_rate": 3.826702215930593e-05,
      "loss": 0.6746,
      "step": 653
    },
    {
      "epoch": 0.83712,
      "grad_norm": 0.489939421415329,
      "learning_rate": 3.826063223408346e-05,
      "loss": 0.6581,
      "step": 654
    },
    {
      "epoch": 0.8384,
      "grad_norm": 0.6027989983558655,
      "learning_rate": 3.825423114792826e-05,
      "loss": 0.6709,
      "step": 655
    },
    {
      "epoch": 0.83968,
      "grad_norm": 0.5708503127098083,
      "learning_rate": 3.8247818905233895e-05,
      "loss": 0.696,
      "step": 656
    },
    {
      "epoch": 0.84096,
      "grad_norm": 0.5238654017448425,
      "learning_rate": 3.824139551040157e-05,
      "loss": 0.606,
      "step": 657
    },
    {
      "epoch": 0.84224,
      "grad_norm": 0.5396076440811157,
      "learning_rate": 3.823496096784017e-05,
      "loss": 0.6719,
      "step": 658
    },
    {
      "epoch": 0.84352,
      "grad_norm": 0.5923845767974854,
      "learning_rate": 3.822851528196621e-05,
      "loss": 0.6584,
      "step": 659
    },
    {
      "epoch": 0.8448,
      "grad_norm": 0.6149125099182129,
      "learning_rate": 3.8222058457203875e-05,
      "loss": 0.6749,
      "step": 660
    },
    {
      "epoch": 0.8448,
      "eval_loss": 0.7043057680130005,
      "eval_runtime": 19.5741,
      "eval_samples_per_second": 25.646,
      "eval_steps_per_second": 3.219,
      "step": 660
    },
    {
      "epoch": 0.84608,
      "grad_norm": 0.6996484398841858,
      "learning_rate": 3.821559049798497e-05,
      "loss": 0.6701,
      "step": 661
    },
    {
      "epoch": 0.84736,
      "grad_norm": 0.6439191699028015,
      "learning_rate": 3.820911140874897e-05,
      "loss": 0.696,
      "step": 662
    },
    {
      "epoch": 0.84864,
      "grad_norm": 0.5901380181312561,
      "learning_rate": 3.8202621193942976e-05,
      "loss": 0.6944,
      "step": 663
    },
    {
      "epoch": 0.84992,
      "grad_norm": 0.6192618608474731,
      "learning_rate": 3.819611985802172e-05,
      "loss": 0.6484,
      "step": 664
    },
    {
      "epoch": 0.8512,
      "grad_norm": 0.6373607516288757,
      "learning_rate": 3.818960740544757e-05,
      "loss": 0.7137,
      "step": 665
    },
    {
      "epoch": 0.85248,
      "grad_norm": 0.5928336977958679,
      "learning_rate": 3.818308384069054e-05,
      "loss": 0.678,
      "step": 666
    },
    {
      "epoch": 0.85376,
      "grad_norm": 0.5486979484558105,
      "learning_rate": 3.817654916822825e-05,
      "loss": 0.7139,
      "step": 667
    },
    {
      "epoch": 0.85504,
      "grad_norm": 0.5797085165977478,
      "learning_rate": 3.8170003392545944e-05,
      "loss": 0.6089,
      "step": 668
    },
    {
      "epoch": 0.85632,
      "grad_norm": 0.5493759512901306,
      "learning_rate": 3.816344651813651e-05,
      "loss": 0.6599,
      "step": 669
    },
    {
      "epoch": 0.8576,
      "grad_norm": 0.588843584060669,
      "learning_rate": 3.815687854950044e-05,
      "loss": 0.6466,
      "step": 670
    },
    {
      "epoch": 0.85888,
      "grad_norm": 0.5860204696655273,
      "learning_rate": 3.8150299491145824e-05,
      "loss": 0.6753,
      "step": 671
    },
    {
      "epoch": 0.86016,
      "grad_norm": 0.5386679172515869,
      "learning_rate": 3.8143709347588396e-05,
      "loss": 0.6688,
      "step": 672
    },
    {
      "epoch": 0.86144,
      "grad_norm": 0.531771719455719,
      "learning_rate": 3.813710812335147e-05,
      "loss": 0.6509,
      "step": 673
    },
    {
      "epoch": 0.86272,
      "grad_norm": 0.4643087685108185,
      "learning_rate": 3.8130495822965987e-05,
      "loss": 0.6498,
      "step": 674
    },
    {
      "epoch": 0.864,
      "grad_norm": 0.5194730162620544,
      "learning_rate": 3.812387245097048e-05,
      "loss": 0.6658,
      "step": 675
    },
    {
      "epoch": 0.86528,
      "grad_norm": 0.5443143844604492,
      "learning_rate": 3.811723801191108e-05,
      "loss": 0.6398,
      "step": 676
    },
    {
      "epoch": 0.86656,
      "grad_norm": 0.49592623114585876,
      "learning_rate": 3.811059251034152e-05,
      "loss": 0.6802,
      "step": 677
    },
    {
      "epoch": 0.86784,
      "grad_norm": 0.5415541529655457,
      "learning_rate": 3.8103935950823115e-05,
      "loss": 0.6577,
      "step": 678
    },
    {
      "epoch": 0.86912,
      "grad_norm": 0.5195705890655518,
      "learning_rate": 3.809726833792479e-05,
      "loss": 0.6502,
      "step": 679
    },
    {
      "epoch": 0.8704,
      "grad_norm": 0.5425269603729248,
      "learning_rate": 3.8090589676223035e-05,
      "loss": 0.6574,
      "step": 680
    },
    {
      "epoch": 0.87168,
      "grad_norm": 0.568422257900238,
      "learning_rate": 3.8083899970301944e-05,
      "loss": 0.6687,
      "step": 681
    },
    {
      "epoch": 0.87296,
      "grad_norm": 0.5313860774040222,
      "learning_rate": 3.807719922475317e-05,
      "loss": 0.6836,
      "step": 682
    },
    {
      "epoch": 0.87424,
      "grad_norm": 0.48866012692451477,
      "learning_rate": 3.8070487444175965e-05,
      "loss": 0.7003,
      "step": 683
    },
    {
      "epoch": 0.87552,
      "grad_norm": 0.5410491228103638,
      "learning_rate": 3.8063764633177126e-05,
      "loss": 0.653,
      "step": 684
    },
    {
      "epoch": 0.8768,
      "grad_norm": 0.4900820553302765,
      "learning_rate": 3.805703079637106e-05,
      "loss": 0.6761,
      "step": 685
    },
    {
      "epoch": 0.87808,
      "grad_norm": 0.576065182685852,
      "learning_rate": 3.8050285938379714e-05,
      "loss": 0.6575,
      "step": 686
    },
    {
      "epoch": 0.87936,
      "grad_norm": 0.5464766025543213,
      "learning_rate": 3.80435300638326e-05,
      "loss": 0.6523,
      "step": 687
    },
    {
      "epoch": 0.88064,
      "grad_norm": 0.4830459654331207,
      "learning_rate": 3.803676317736681e-05,
      "loss": 0.6446,
      "step": 688
    },
    {
      "epoch": 0.88192,
      "grad_norm": 0.5860381722450256,
      "learning_rate": 3.802998528362697e-05,
      "loss": 0.6855,
      "step": 689
    },
    {
      "epoch": 0.8832,
      "grad_norm": 0.5201774835586548,
      "learning_rate": 3.802319638726528e-05,
      "loss": 0.6373,
      "step": 690
    },
    {
      "epoch": 0.8832,
      "eval_loss": 0.703089714050293,
      "eval_runtime": 19.5988,
      "eval_samples_per_second": 25.614,
      "eval_steps_per_second": 3.214,
      "step": 690
    },
    {
      "epoch": 0.88448,
      "grad_norm": 0.5178946256637573,
      "learning_rate": 3.8016396492941496e-05,
      "loss": 0.6662,
      "step": 691
    },
    {
      "epoch": 0.88576,
      "grad_norm": 0.49665337800979614,
      "learning_rate": 3.80095856053229e-05,
      "loss": 0.6355,
      "step": 692
    },
    {
      "epoch": 0.88704,
      "grad_norm": 0.5232968330383301,
      "learning_rate": 3.8002763729084335e-05,
      "loss": 0.6794,
      "step": 693
    },
    {
      "epoch": 0.88832,
      "grad_norm": 0.5046104192733765,
      "learning_rate": 3.799593086890819e-05,
      "loss": 0.6728,
      "step": 694
    },
    {
      "epoch": 0.8896,
      "grad_norm": 0.5464974641799927,
      "learning_rate": 3.798908702948437e-05,
      "loss": 0.6525,
      "step": 695
    },
    {
      "epoch": 0.89088,
      "grad_norm": 0.5119357705116272,
      "learning_rate": 3.798223221551036e-05,
      "loss": 0.6845,
      "step": 696
    },
    {
      "epoch": 0.89216,
      "grad_norm": 0.5061988830566406,
      "learning_rate": 3.7975366431691124e-05,
      "loss": 0.6637,
      "step": 697
    },
    {
      "epoch": 0.89344,
      "grad_norm": 0.5115570425987244,
      "learning_rate": 3.7968489682739195e-05,
      "loss": 0.6611,
      "step": 698
    },
    {
      "epoch": 0.89472,
      "grad_norm": 0.45419949293136597,
      "learning_rate": 3.796160197337462e-05,
      "loss": 0.6395,
      "step": 699
    },
    {
      "epoch": 0.896,
      "grad_norm": 0.49113744497299194,
      "learning_rate": 3.795470330832496e-05,
      "loss": 0.6822,
      "step": 700
    },
    {
      "epoch": 0.89728,
      "grad_norm": 0.5135698914527893,
      "learning_rate": 3.794779369232531e-05,
      "loss": 0.6856,
      "step": 701
    },
    {
      "epoch": 0.89856,
      "grad_norm": 0.5226943492889404,
      "learning_rate": 3.7940873130118275e-05,
      "loss": 0.6544,
      "step": 702
    },
    {
      "epoch": 0.89984,
      "grad_norm": 0.5407574772834778,
      "learning_rate": 3.7933941626453965e-05,
      "loss": 0.65,
      "step": 703
    },
    {
      "epoch": 0.90112,
      "grad_norm": 0.538567066192627,
      "learning_rate": 3.792699918609002e-05,
      "loss": 0.6602,
      "step": 704
    },
    {
      "epoch": 0.9024,
      "grad_norm": 0.47789710760116577,
      "learning_rate": 3.792004581379157e-05,
      "loss": 0.6311,
      "step": 705
    },
    {
      "epoch": 0.90368,
      "grad_norm": 0.674965500831604,
      "learning_rate": 3.7913081514331256e-05,
      "loss": 0.6535,
      "step": 706
    },
    {
      "epoch": 0.90496,
      "grad_norm": 0.5215507745742798,
      "learning_rate": 3.7906106292489214e-05,
      "loss": 0.6734,
      "step": 707
    },
    {
      "epoch": 0.90624,
      "grad_norm": 0.6158815026283264,
      "learning_rate": 3.789912015305308e-05,
      "loss": 0.6454,
      "step": 708
    },
    {
      "epoch": 0.90752,
      "grad_norm": 0.5726962089538574,
      "learning_rate": 3.789212310081798e-05,
      "loss": 0.6675,
      "step": 709
    },
    {
      "epoch": 0.9088,
      "grad_norm": 0.6040086150169373,
      "learning_rate": 3.7885115140586546e-05,
      "loss": 0.6578,
      "step": 710
    },
    {
      "epoch": 0.91008,
      "grad_norm": 0.5612017512321472,
      "learning_rate": 3.787809627716887e-05,
      "loss": 0.6598,
      "step": 711
    },
    {
      "epoch": 0.91136,
      "grad_norm": 0.5215588212013245,
      "learning_rate": 3.787106651538254e-05,
      "loss": 0.6707,
      "step": 712
    },
    {
      "epoch": 0.91264,
      "grad_norm": 0.6304693818092346,
      "learning_rate": 3.786402586005264e-05,
      "loss": 0.6541,
      "step": 713
    },
    {
      "epoch": 0.91392,
      "grad_norm": 0.6233842968940735,
      "learning_rate": 3.785697431601172e-05,
      "loss": 0.6435,
      "step": 714
    },
    {
      "epoch": 0.9152,
      "grad_norm": 0.5794985890388489,
      "learning_rate": 3.784991188809979e-05,
      "loss": 0.6699,
      "step": 715
    },
    {
      "epoch": 0.91648,
      "grad_norm": 0.5338553786277771,
      "learning_rate": 3.784283858116434e-05,
      "loss": 0.6983,
      "step": 716
    },
    {
      "epoch": 0.91776,
      "grad_norm": 0.603832483291626,
      "learning_rate": 3.783575440006033e-05,
      "loss": 0.6347,
      "step": 717
    },
    {
      "epoch": 0.91904,
      "grad_norm": 0.5867237448692322,
      "learning_rate": 3.782865934965019e-05,
      "loss": 0.6749,
      "step": 718
    },
    {
      "epoch": 0.92032,
      "grad_norm": 0.6062390208244324,
      "learning_rate": 3.7821553434803806e-05,
      "loss": 0.664,
      "step": 719
    },
    {
      "epoch": 0.9216,
      "grad_norm": 0.6165828704833984,
      "learning_rate": 3.7814436660398514e-05,
      "loss": 0.6743,
      "step": 720
    },
    {
      "epoch": 0.9216,
      "eval_loss": 0.701291024684906,
      "eval_runtime": 19.6418,
      "eval_samples_per_second": 25.558,
      "eval_steps_per_second": 3.207,
      "step": 720
    },
    {
      "epoch": 0.92288,
      "grad_norm": 0.5896676778793335,
      "learning_rate": 3.78073090313191e-05,
      "loss": 0.6626,
      "step": 721
    },
    {
      "epoch": 0.92416,
      "grad_norm": 0.6335607171058655,
      "learning_rate": 3.780017055245782e-05,
      "loss": 0.696,
      "step": 722
    },
    {
      "epoch": 0.92544,
      "grad_norm": 0.5937784314155579,
      "learning_rate": 3.7793021228714366e-05,
      "loss": 0.6555,
      "step": 723
    },
    {
      "epoch": 0.92672,
      "grad_norm": 0.6331490874290466,
      "learning_rate": 3.7785861064995854e-05,
      "loss": 0.6792,
      "step": 724
    },
    {
      "epoch": 0.928,
      "grad_norm": 0.6025753617286682,
      "learning_rate": 3.7778690066216886e-05,
      "loss": 0.6706,
      "step": 725
    },
    {
      "epoch": 0.92928,
      "grad_norm": 0.5935916900634766,
      "learning_rate": 3.7771508237299456e-05,
      "loss": 0.6859,
      "step": 726
    },
    {
      "epoch": 0.93056,
      "grad_norm": 0.6255166530609131,
      "learning_rate": 3.776431558317302e-05,
      "loss": 0.6517,
      "step": 727
    },
    {
      "epoch": 0.93184,
      "grad_norm": 0.5103235244750977,
      "learning_rate": 3.775711210877445e-05,
      "loss": 0.677,
      "step": 728
    },
    {
      "epoch": 0.93312,
      "grad_norm": 0.5949177145957947,
      "learning_rate": 3.7749897819048044e-05,
      "loss": 0.6572,
      "step": 729
    },
    {
      "epoch": 0.9344,
      "grad_norm": 0.5092359781265259,
      "learning_rate": 3.774267271894554e-05,
      "loss": 0.6758,
      "step": 730
    },
    {
      "epoch": 0.93568,
      "grad_norm": 0.6258443593978882,
      "learning_rate": 3.773543681342607e-05,
      "loss": 0.6492,
      "step": 731
    },
    {
      "epoch": 0.93696,
      "grad_norm": 0.5788440108299255,
      "learning_rate": 3.7728190107456214e-05,
      "loss": 0.6611,
      "step": 732
    },
    {
      "epoch": 0.93824,
      "grad_norm": 0.6272332072257996,
      "learning_rate": 3.772093260600993e-05,
      "loss": 0.6311,
      "step": 733
    },
    {
      "epoch": 0.93952,
      "grad_norm": 0.5995723009109497,
      "learning_rate": 3.771366431406863e-05,
      "loss": 0.6965,
      "step": 734
    },
    {
      "epoch": 0.9408,
      "grad_norm": 0.5540239810943604,
      "learning_rate": 3.7706385236621074e-05,
      "loss": 0.6498,
      "step": 735
    },
    {
      "epoch": 0.94208,
      "grad_norm": 0.6140071153640747,
      "learning_rate": 3.7699095378663476e-05,
      "loss": 0.6528,
      "step": 736
    },
    {
      "epoch": 0.94336,
      "grad_norm": 0.5888604521751404,
      "learning_rate": 3.7691794745199435e-05,
      "loss": 0.647,
      "step": 737
    },
    {
      "epoch": 0.94464,
      "grad_norm": 0.6129213571548462,
      "learning_rate": 3.768448334123993e-05,
      "loss": 0.6889,
      "step": 738
    },
    {
      "epoch": 0.94592,
      "grad_norm": 0.611226499080658,
      "learning_rate": 3.7677161171803346e-05,
      "loss": 0.6904,
      "step": 739
    },
    {
      "epoch": 0.9472,
      "grad_norm": 0.5999221205711365,
      "learning_rate": 3.766982824191547e-05,
      "loss": 0.709,
      "step": 740
    },
    {
      "epoch": 0.94848,
      "grad_norm": 0.6189354658126831,
      "learning_rate": 3.766248455660944e-05,
      "loss": 0.6894,
      "step": 741
    },
    {
      "epoch": 0.94976,
      "grad_norm": 0.6292353868484497,
      "learning_rate": 3.7655130120925805e-05,
      "loss": 0.6698,
      "step": 742
    },
    {
      "epoch": 0.95104,
      "grad_norm": 0.6458495259284973,
      "learning_rate": 3.764776493991249e-05,
      "loss": 0.6616,
      "step": 743
    },
    {
      "epoch": 0.95232,
      "grad_norm": 0.5995420217514038,
      "learning_rate": 3.764038901862478e-05,
      "loss": 0.6622,
      "step": 744
    },
    {
      "epoch": 0.9536,
      "grad_norm": 0.7107582688331604,
      "learning_rate": 3.763300236212535e-05,
      "loss": 0.6801,
      "step": 745
    },
    {
      "epoch": 0.95488,
      "grad_norm": 0.6244316101074219,
      "learning_rate": 3.762560497548424e-05,
      "loss": 0.6667,
      "step": 746
    },
    {
      "epoch": 0.95616,
      "grad_norm": 0.5940176844596863,
      "learning_rate": 3.761819686377882e-05,
      "loss": 0.6635,
      "step": 747
    },
    {
      "epoch": 0.95744,
      "grad_norm": 0.6233631372451782,
      "learning_rate": 3.761077803209389e-05,
      "loss": 0.6654,
      "step": 748
    },
    {
      "epoch": 0.95872,
      "grad_norm": 0.5411273837089539,
      "learning_rate": 3.7603348485521546e-05,
      "loss": 0.6763,
      "step": 749
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.5749450325965881,
      "learning_rate": 3.759590822916128e-05,
      "loss": 0.6626,
      "step": 750
    },
    {
      "epoch": 0.96,
      "eval_loss": 0.6987097859382629,
      "eval_runtime": 19.5806,
      "eval_samples_per_second": 25.638,
      "eval_steps_per_second": 3.217,
      "step": 750
    },
    {
      "epoch": 0.96128,
      "grad_norm": 0.5283013582229614,
      "learning_rate": 3.75884572681199e-05,
      "loss": 0.6583,
      "step": 751
    },
    {
      "epoch": 0.96256,
      "grad_norm": 0.5731936693191528,
      "learning_rate": 3.7580995607511584e-05,
      "loss": 0.6848,
      "step": 752
    },
    {
      "epoch": 0.96384,
      "grad_norm": 0.4962114989757538,
      "learning_rate": 3.757352325245784e-05,
      "loss": 0.6839,
      "step": 753
    },
    {
      "epoch": 0.96512,
      "grad_norm": 0.5401961803436279,
      "learning_rate": 3.7566040208087544e-05,
      "loss": 0.6697,
      "step": 754
    },
    {
      "epoch": 0.9664,
      "grad_norm": 0.47949647903442383,
      "learning_rate": 3.755854647953687e-05,
      "loss": 0.6765,
      "step": 755
    },
    {
      "epoch": 0.96768,
      "grad_norm": 0.5169327259063721,
      "learning_rate": 3.755104207194936e-05,
      "loss": 0.6796,
      "step": 756
    },
    {
      "epoch": 0.96896,
      "grad_norm": 0.483396977186203,
      "learning_rate": 3.7543526990475864e-05,
      "loss": 0.6704,
      "step": 757
    },
    {
      "epoch": 0.97024,
      "grad_norm": 0.4732683300971985,
      "learning_rate": 3.753600124027456e-05,
      "loss": 0.6563,
      "step": 758
    },
    {
      "epoch": 0.97152,
      "grad_norm": 0.45995524525642395,
      "learning_rate": 3.7528464826510965e-05,
      "loss": 0.6622,
      "step": 759
    },
    {
      "epoch": 0.9728,
      "grad_norm": 0.4743465483188629,
      "learning_rate": 3.7520917754357895e-05,
      "loss": 0.6649,
      "step": 760
    },
    {
      "epoch": 0.97408,
      "grad_norm": 0.47582364082336426,
      "learning_rate": 3.751336002899549e-05,
      "loss": 0.646,
      "step": 761
    },
    {
      "epoch": 0.97536,
      "grad_norm": 0.5059706568717957,
      "learning_rate": 3.750579165561121e-05,
      "loss": 0.6705,
      "step": 762
    },
    {
      "epoch": 0.97664,
      "grad_norm": 0.5064669251441956,
      "learning_rate": 3.7498212639399814e-05,
      "loss": 0.6889,
      "step": 763
    },
    {
      "epoch": 0.97792,
      "grad_norm": 0.542224645614624,
      "learning_rate": 3.749062298556337e-05,
      "loss": 0.654,
      "step": 764
    },
    {
      "epoch": 0.9792,
      "grad_norm": 0.5148495435714722,
      "learning_rate": 3.748302269931124e-05,
      "loss": 0.653,
      "step": 765
    },
    {
      "epoch": 0.98048,
      "grad_norm": 0.5403793454170227,
      "learning_rate": 3.74754117858601e-05,
      "loss": 0.6809,
      "step": 766
    },
    {
      "epoch": 0.98176,
      "grad_norm": 0.47193753719329834,
      "learning_rate": 3.7467790250433903e-05,
      "loss": 0.6623,
      "step": 767
    },
    {
      "epoch": 0.98304,
      "grad_norm": 0.5113906264305115,
      "learning_rate": 3.7460158098263895e-05,
      "loss": 0.6771,
      "step": 768
    },
    {
      "epoch": 0.98432,
      "grad_norm": 0.5458548069000244,
      "learning_rate": 3.745251533458863e-05,
      "loss": 0.6613,
      "step": 769
    },
    {
      "epoch": 0.9856,
      "grad_norm": 0.513218343257904,
      "learning_rate": 3.7444861964653905e-05,
      "loss": 0.6556,
      "step": 770
    },
    {
      "epoch": 0.98688,
      "grad_norm": 0.4523468613624573,
      "learning_rate": 3.743719799371284e-05,
      "loss": 0.6672,
      "step": 771
    },
    {
      "epoch": 0.98816,
      "grad_norm": 0.4986381530761719,
      "learning_rate": 3.7429523427025813e-05,
      "loss": 0.6633,
      "step": 772
    },
    {
      "epoch": 0.98944,
      "grad_norm": 0.567969799041748,
      "learning_rate": 3.742183826986046e-05,
      "loss": 0.6544,
      "step": 773
    },
    {
      "epoch": 0.99072,
      "grad_norm": 0.48580843210220337,
      "learning_rate": 3.741414252749171e-05,
      "loss": 0.6823,
      "step": 774
    },
    {
      "epoch": 0.992,
      "grad_norm": 0.6315404176712036,
      "learning_rate": 3.7406436205201755e-05,
      "loss": 0.6685,
      "step": 775
    },
    {
      "epoch": 0.99328,
      "grad_norm": 0.5436789989471436,
      "learning_rate": 3.739871930828002e-05,
      "loss": 0.6394,
      "step": 776
    },
    {
      "epoch": 0.99456,
      "grad_norm": 0.6198738813400269,
      "learning_rate": 3.739099184202324e-05,
      "loss": 0.6526,
      "step": 777
    },
    {
      "epoch": 0.99584,
      "grad_norm": 0.5751096606254578,
      "learning_rate": 3.7383253811735346e-05,
      "loss": 0.6997,
      "step": 778
    },
    {
      "epoch": 0.99712,
      "grad_norm": 0.6743948459625244,
      "learning_rate": 3.737550522272756e-05,
      "loss": 0.664,
      "step": 779
    },
    {
      "epoch": 0.9984,
      "grad_norm": 0.6169484257698059,
      "learning_rate": 3.736774608031834e-05,
      "loss": 0.6935,
      "step": 780
    },
    {
      "epoch": 0.9984,
      "eval_loss": 0.6998871564865112,
      "eval_runtime": 19.5699,
      "eval_samples_per_second": 25.652,
      "eval_steps_per_second": 3.219,
      "step": 780
    },
    {
      "epoch": 0.99968,
      "grad_norm": 0.7273963689804077,
      "learning_rate": 3.7359976389833384e-05,
      "loss": 0.6746,
      "step": 781
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.9516610503196716,
      "learning_rate": 3.735219615660564e-05,
      "loss": 0.6863,
      "step": 782
    },
    {
      "epoch": 1.00128,
      "grad_norm": 0.6978998780250549,
      "learning_rate": 3.734440538597528e-05,
      "loss": 0.6197,
      "step": 783
    },
    {
      "epoch": 1.00256,
      "grad_norm": 0.7508107423782349,
      "learning_rate": 3.733660408328971e-05,
      "loss": 0.5911,
      "step": 784
    },
    {
      "epoch": 1.00384,
      "grad_norm": 0.6375359296798706,
      "learning_rate": 3.732879225390357e-05,
      "loss": 0.6053,
      "step": 785
    },
    {
      "epoch": 1.00512,
      "grad_norm": 0.6797159314155579,
      "learning_rate": 3.7320969903178735e-05,
      "loss": 0.6546,
      "step": 786
    },
    {
      "epoch": 1.0064,
      "grad_norm": 0.6292569637298584,
      "learning_rate": 3.731313703648427e-05,
      "loss": 0.6426,
      "step": 787
    },
    {
      "epoch": 1.00768,
      "grad_norm": 0.5453099012374878,
      "learning_rate": 3.730529365919651e-05,
      "loss": 0.625,
      "step": 788
    },
    {
      "epoch": 1.00896,
      "grad_norm": 0.5575852990150452,
      "learning_rate": 3.729743977669895e-05,
      "loss": 0.6587,
      "step": 789
    },
    {
      "epoch": 1.01024,
      "grad_norm": 0.5153116583824158,
      "learning_rate": 3.728957539438231e-05,
      "loss": 0.6439,
      "step": 790
    },
    {
      "epoch": 1.01152,
      "grad_norm": 0.5157707929611206,
      "learning_rate": 3.7281700517644554e-05,
      "loss": 0.6368,
      "step": 791
    },
    {
      "epoch": 1.0128,
      "grad_norm": 0.546253502368927,
      "learning_rate": 3.72738151518908e-05,
      "loss": 0.6295,
      "step": 792
    },
    {
      "epoch": 1.01408,
      "grad_norm": 0.6052794456481934,
      "learning_rate": 3.726591930253339e-05,
      "loss": 0.6571,
      "step": 793
    },
    {
      "epoch": 1.01536,
      "grad_norm": 0.5201693773269653,
      "learning_rate": 3.7258012974991864e-05,
      "loss": 0.6432,
      "step": 794
    },
    {
      "epoch": 1.01664,
      "grad_norm": 0.5689214468002319,
      "learning_rate": 3.7250096174692936e-05,
      "loss": 0.633,
      "step": 795
    },
    {
      "epoch": 1.01792,
      "grad_norm": 0.5149937868118286,
      "learning_rate": 3.7242168907070534e-05,
      "loss": 0.6737,
      "step": 796
    },
    {
      "epoch": 1.0192,
      "grad_norm": 0.5797461867332458,
      "learning_rate": 3.723423117756574e-05,
      "loss": 0.6252,
      "step": 797
    },
    {
      "epoch": 1.02048,
      "grad_norm": 0.5616918206214905,
      "learning_rate": 3.722628299162684e-05,
      "loss": 0.6158,
      "step": 798
    },
    {
      "epoch": 1.02176,
      "grad_norm": 0.5574386119842529,
      "learning_rate": 3.7218324354709305e-05,
      "loss": 0.6285,
      "step": 799
    },
    {
      "epoch": 1.02304,
      "grad_norm": 0.5712234973907471,
      "learning_rate": 3.721035527227575e-05,
      "loss": 0.5965,
      "step": 800
    },
    {
      "epoch": 1.02432,
      "grad_norm": 0.5263935327529907,
      "learning_rate": 3.720237574979597e-05,
      "loss": 0.6395,
      "step": 801
    },
    {
      "epoch": 1.0256,
      "grad_norm": 0.535115122795105,
      "learning_rate": 3.7194385792746934e-05,
      "loss": 0.6174,
      "step": 802
    },
    {
      "epoch": 1.02688,
      "grad_norm": 0.5689954161643982,
      "learning_rate": 3.718638540661278e-05,
      "loss": 0.6144,
      "step": 803
    },
    {
      "epoch": 1.02816,
      "grad_norm": 0.5647488236427307,
      "learning_rate": 3.7178374596884784e-05,
      "loss": 0.6105,
      "step": 804
    },
    {
      "epoch": 1.02944,
      "grad_norm": 0.5726591944694519,
      "learning_rate": 3.717035336906138e-05,
      "loss": 0.6229,
      "step": 805
    },
    {
      "epoch": 1.03072,
      "grad_norm": 0.52170729637146,
      "learning_rate": 3.716232172864817e-05,
      "loss": 0.5891,
      "step": 806
    },
    {
      "epoch": 1.032,
      "grad_norm": 0.6672565937042236,
      "learning_rate": 3.715427968115789e-05,
      "loss": 0.6208,
      "step": 807
    },
    {
      "epoch": 1.03328,
      "grad_norm": 0.5876906514167786,
      "learning_rate": 3.714622723211041e-05,
      "loss": 0.6124,
      "step": 808
    },
    {
      "epoch": 1.03456,
      "grad_norm": 0.6233872771263123,
      "learning_rate": 3.713816438703276e-05,
      "loss": 0.6126,
      "step": 809
    },
    {
      "epoch": 1.03584,
      "grad_norm": 0.5635946989059448,
      "learning_rate": 3.7130091151459094e-05,
      "loss": 0.6172,
      "step": 810
    },
    {
      "epoch": 1.03584,
      "eval_loss": 0.700531542301178,
      "eval_runtime": 19.5808,
      "eval_samples_per_second": 25.637,
      "eval_steps_per_second": 3.217,
      "step": 810
    },
    {
      "epoch": 1.03712,
      "grad_norm": 0.5930481553077698,
      "learning_rate": 3.71220075309307e-05,
      "loss": 0.6004,
      "step": 811
    },
    {
      "epoch": 1.0384,
      "grad_norm": 0.5178161263465881,
      "learning_rate": 3.711391353099599e-05,
      "loss": 0.6173,
      "step": 812
    },
    {
      "epoch": 1.03968,
      "grad_norm": 0.5803300738334656,
      "learning_rate": 3.710580915721051e-05,
      "loss": 0.6643,
      "step": 813
    },
    {
      "epoch": 1.04096,
      "grad_norm": 0.5607686638832092,
      "learning_rate": 3.709769441513691e-05,
      "loss": 0.6292,
      "step": 814
    },
    {
      "epoch": 1.04224,
      "grad_norm": 0.5866657495498657,
      "learning_rate": 3.708956931034498e-05,
      "loss": 0.6465,
      "step": 815
    },
    {
      "epoch": 1.04352,
      "grad_norm": 0.5063027143478394,
      "learning_rate": 3.7081433848411596e-05,
      "loss": 0.6063,
      "step": 816
    },
    {
      "epoch": 1.0448,
      "grad_norm": 0.539871335029602,
      "learning_rate": 3.707328803492077e-05,
      "loss": 0.6072,
      "step": 817
    },
    {
      "epoch": 1.04608,
      "grad_norm": 0.5645202994346619,
      "learning_rate": 3.70651318754636e-05,
      "loss": 0.6349,
      "step": 818
    },
    {
      "epoch": 1.04736,
      "grad_norm": 0.5289549827575684,
      "learning_rate": 3.705696537563828e-05,
      "loss": 0.5972,
      "step": 819
    },
    {
      "epoch": 1.04864,
      "grad_norm": 0.5740439891815186,
      "learning_rate": 3.7048788541050126e-05,
      "loss": 0.6096,
      "step": 820
    },
    {
      "epoch": 1.04992,
      "grad_norm": 0.5493366122245789,
      "learning_rate": 3.7040601377311526e-05,
      "loss": 0.6252,
      "step": 821
    },
    {
      "epoch": 1.0512,
      "grad_norm": 0.54609215259552,
      "learning_rate": 3.703240389004197e-05,
      "loss": 0.6351,
      "step": 822
    },
    {
      "epoch": 1.05248,
      "grad_norm": 0.6107680797576904,
      "learning_rate": 3.702419608486803e-05,
      "loss": 0.6373,
      "step": 823
    },
    {
      "epoch": 1.05376,
      "grad_norm": 0.5993803143501282,
      "learning_rate": 3.7015977967423356e-05,
      "loss": 0.6426,
      "step": 824
    },
    {
      "epoch": 1.05504,
      "grad_norm": 0.5063743591308594,
      "learning_rate": 3.700774954334868e-05,
      "loss": 0.6193,
      "step": 825
    },
    {
      "epoch": 1.05632,
      "grad_norm": 0.5079895853996277,
      "learning_rate": 3.69995108182918e-05,
      "loss": 0.5982,
      "step": 826
    },
    {
      "epoch": 1.0576,
      "grad_norm": 0.5126475095748901,
      "learning_rate": 3.6991261797907604e-05,
      "loss": 0.6003,
      "step": 827
    },
    {
      "epoch": 1.05888,
      "grad_norm": 0.5092754364013672,
      "learning_rate": 3.698300248785803e-05,
      "loss": 0.6639,
      "step": 828
    },
    {
      "epoch": 1.06016,
      "grad_norm": 0.5253211855888367,
      "learning_rate": 3.6974732893812074e-05,
      "loss": 0.6379,
      "step": 829
    },
    {
      "epoch": 1.06144,
      "grad_norm": 0.5032366514205933,
      "learning_rate": 3.696645302144582e-05,
      "loss": 0.6232,
      "step": 830
    },
    {
      "epoch": 1.06272,
      "grad_norm": 0.5466593503952026,
      "learning_rate": 3.695816287644236e-05,
      "loss": 0.6399,
      "step": 831
    },
    {
      "epoch": 1.064,
      "grad_norm": 0.48845332860946655,
      "learning_rate": 3.694986246449189e-05,
      "loss": 0.6158,
      "step": 832
    },
    {
      "epoch": 1.06528,
      "grad_norm": 0.47722238302230835,
      "learning_rate": 3.6941551791291604e-05,
      "loss": 0.6138,
      "step": 833
    },
    {
      "epoch": 1.06656,
      "grad_norm": 0.5120546221733093,
      "learning_rate": 3.693323086254578e-05,
      "loss": 0.6494,
      "step": 834
    },
    {
      "epoch": 1.06784,
      "grad_norm": 0.4693053364753723,
      "learning_rate": 3.69248996839657e-05,
      "loss": 0.6451,
      "step": 835
    },
    {
      "epoch": 1.06912,
      "grad_norm": 0.5126748085021973,
      "learning_rate": 3.691655826126971e-05,
      "loss": 0.6385,
      "step": 836
    },
    {
      "epoch": 1.0704,
      "grad_norm": 0.5147609114646912,
      "learning_rate": 3.690820660018317e-05,
      "loss": 0.6156,
      "step": 837
    },
    {
      "epoch": 1.07168,
      "grad_norm": 0.526591956615448,
      "learning_rate": 3.689984470643847e-05,
      "loss": 0.5892,
      "step": 838
    },
    {
      "epoch": 1.07296,
      "grad_norm": 0.5142748951911926,
      "learning_rate": 3.6891472585775034e-05,
      "loss": 0.6232,
      "step": 839
    },
    {
      "epoch": 1.07424,
      "grad_norm": 0.5200185775756836,
      "learning_rate": 3.688309024393929e-05,
      "loss": 0.6019,
      "step": 840
    },
    {
      "epoch": 1.07424,
      "eval_loss": 0.6997473239898682,
      "eval_runtime": 19.5888,
      "eval_samples_per_second": 25.627,
      "eval_steps_per_second": 3.216,
      "step": 840
    },
    {
      "epoch": 1.07552,
      "grad_norm": 0.581644594669342,
      "learning_rate": 3.687469768668469e-05,
      "loss": 0.617,
      "step": 841
    },
    {
      "epoch": 1.0768,
      "grad_norm": 0.5425085425376892,
      "learning_rate": 3.6866294919771705e-05,
      "loss": 0.6332,
      "step": 842
    },
    {
      "epoch": 1.07808,
      "grad_norm": 0.5750662088394165,
      "learning_rate": 3.685788194896779e-05,
      "loss": 0.5956,
      "step": 843
    },
    {
      "epoch": 1.07936,
      "grad_norm": 0.5308177471160889,
      "learning_rate": 3.684945878004744e-05,
      "loss": 0.6251,
      "step": 844
    },
    {
      "epoch": 1.08064,
      "grad_norm": 0.5591775178909302,
      "learning_rate": 3.6841025418792114e-05,
      "loss": 0.6319,
      "step": 845
    },
    {
      "epoch": 1.08192,
      "grad_norm": 0.5252130627632141,
      "learning_rate": 3.683258187099028e-05,
      "loss": 0.6287,
      "step": 846
    },
    {
      "epoch": 1.0832,
      "grad_norm": 0.49837878346443176,
      "learning_rate": 3.682412814243741e-05,
      "loss": 0.5966,
      "step": 847
    },
    {
      "epoch": 1.08448,
      "grad_norm": 0.4612179100513458,
      "learning_rate": 3.681566423893594e-05,
      "loss": 0.6112,
      "step": 848
    },
    {
      "epoch": 1.08576,
      "grad_norm": 0.5672823786735535,
      "learning_rate": 3.6807190166295324e-05,
      "loss": 0.5963,
      "step": 849
    },
    {
      "epoch": 1.08704,
      "grad_norm": 0.4686615765094757,
      "learning_rate": 3.6798705930331954e-05,
      "loss": 0.6188,
      "step": 850
    },
    {
      "epoch": 1.08832,
      "grad_norm": 0.4858367443084717,
      "learning_rate": 3.679021153686923e-05,
      "loss": 0.6321,
      "step": 851
    },
    {
      "epoch": 1.0896,
      "grad_norm": 0.5434247851371765,
      "learning_rate": 3.678170699173751e-05,
      "loss": 0.5925,
      "step": 852
    },
    {
      "epoch": 1.09088,
      "grad_norm": 0.5434045195579529,
      "learning_rate": 3.677319230077412e-05,
      "loss": 0.6173,
      "step": 853
    },
    {
      "epoch": 1.09216,
      "grad_norm": 0.5496405363082886,
      "learning_rate": 3.676466746982336e-05,
      "loss": 0.616,
      "step": 854
    },
    {
      "epoch": 1.09344,
      "grad_norm": 0.5053821802139282,
      "learning_rate": 3.675613250473648e-05,
      "loss": 0.6082,
      "step": 855
    },
    {
      "epoch": 1.09472,
      "grad_norm": 0.4990701675415039,
      "learning_rate": 3.674758741137168e-05,
      "loss": 0.5924,
      "step": 856
    },
    {
      "epoch": 1.096,
      "grad_norm": 0.469681054353714,
      "learning_rate": 3.6739032195594134e-05,
      "loss": 0.6148,
      "step": 857
    },
    {
      "epoch": 1.09728,
      "grad_norm": 0.4692353308200836,
      "learning_rate": 3.673046686327594e-05,
      "loss": 0.6264,
      "step": 858
    },
    {
      "epoch": 1.09856,
      "grad_norm": 0.497469037771225,
      "learning_rate": 3.672189142029614e-05,
      "loss": 0.6255,
      "step": 859
    },
    {
      "epoch": 1.09984,
      "grad_norm": 0.5303905606269836,
      "learning_rate": 3.671330587254076e-05,
      "loss": 0.6499,
      "step": 860
    },
    {
      "epoch": 1.10112,
      "grad_norm": 0.5445454120635986,
      "learning_rate": 3.670471022590269e-05,
      "loss": 0.606,
      "step": 861
    },
    {
      "epoch": 1.1024,
      "grad_norm": 0.45052316784858704,
      "learning_rate": 3.669610448628181e-05,
      "loss": 0.6055,
      "step": 862
    },
    {
      "epoch": 1.10368,
      "grad_norm": 0.6136205792427063,
      "learning_rate": 3.66874886595849e-05,
      "loss": 0.6422,
      "step": 863
    },
    {
      "epoch": 1.10496,
      "grad_norm": 0.5431349277496338,
      "learning_rate": 3.667886275172567e-05,
      "loss": 0.6313,
      "step": 864
    },
    {
      "epoch": 1.1062400000000001,
      "grad_norm": 0.5774427652359009,
      "learning_rate": 3.6670226768624745e-05,
      "loss": 0.609,
      "step": 865
    },
    {
      "epoch": 1.10752,
      "grad_norm": 0.5447267293930054,
      "learning_rate": 3.6661580716209684e-05,
      "loss": 0.6384,
      "step": 866
    },
    {
      "epoch": 1.1088,
      "grad_norm": 0.5456610321998596,
      "learning_rate": 3.665292460041492e-05,
      "loss": 0.6345,
      "step": 867
    },
    {
      "epoch": 1.11008,
      "grad_norm": 0.5874724984169006,
      "learning_rate": 3.6644258427181845e-05,
      "loss": 0.6335,
      "step": 868
    },
    {
      "epoch": 1.11136,
      "grad_norm": 0.5571343898773193,
      "learning_rate": 3.66355822024587e-05,
      "loss": 0.6491,
      "step": 869
    },
    {
      "epoch": 1.11264,
      "grad_norm": 0.5616312026977539,
      "learning_rate": 3.662689593220066e-05,
      "loss": 0.6338,
      "step": 870
    },
    {
      "epoch": 1.11264,
      "eval_loss": 0.6996856331825256,
      "eval_runtime": 19.5761,
      "eval_samples_per_second": 25.643,
      "eval_steps_per_second": 3.218,
      "step": 870
    },
    {
      "epoch": 1.11392,
      "grad_norm": 0.5741679072380066,
      "learning_rate": 3.66181996223698e-05,
      "loss": 0.6308,
      "step": 871
    },
    {
      "epoch": 1.1152,
      "grad_norm": 0.5313056707382202,
      "learning_rate": 3.660949327893505e-05,
      "loss": 0.6278,
      "step": 872
    },
    {
      "epoch": 1.11648,
      "grad_norm": 0.6152394413948059,
      "learning_rate": 3.660077690787226e-05,
      "loss": 0.6501,
      "step": 873
    },
    {
      "epoch": 1.11776,
      "grad_norm": 0.5823473930358887,
      "learning_rate": 3.659205051516414e-05,
      "loss": 0.6229,
      "step": 874
    },
    {
      "epoch": 1.11904,
      "grad_norm": 0.5507427453994751,
      "learning_rate": 3.658331410680031e-05,
      "loss": 0.5995,
      "step": 875
    },
    {
      "epoch": 1.12032,
      "grad_norm": 0.5569491982460022,
      "learning_rate": 3.657456768877723e-05,
      "loss": 0.6264,
      "step": 876
    },
    {
      "epoch": 1.1216,
      "grad_norm": 0.544539749622345,
      "learning_rate": 3.6565811267098245e-05,
      "loss": 0.6001,
      "step": 877
    },
    {
      "epoch": 1.12288,
      "grad_norm": 0.5211987495422363,
      "learning_rate": 3.6557044847773575e-05,
      "loss": 0.5917,
      "step": 878
    },
    {
      "epoch": 1.12416,
      "grad_norm": 0.5341512560844421,
      "learning_rate": 3.654826843682028e-05,
      "loss": 0.6182,
      "step": 879
    },
    {
      "epoch": 1.12544,
      "grad_norm": 0.47990870475769043,
      "learning_rate": 3.6539482040262304e-05,
      "loss": 0.6057,
      "step": 880
    },
    {
      "epoch": 1.12672,
      "grad_norm": 0.5681471824645996,
      "learning_rate": 3.653068566413043e-05,
      "loss": 0.6017,
      "step": 881
    },
    {
      "epoch": 1.1280000000000001,
      "grad_norm": 0.4682222306728363,
      "learning_rate": 3.6521879314462285e-05,
      "loss": 0.6192,
      "step": 882
    },
    {
      "epoch": 1.12928,
      "grad_norm": 0.5238298773765564,
      "learning_rate": 3.6513062997302356e-05,
      "loss": 0.6145,
      "step": 883
    },
    {
      "epoch": 1.13056,
      "grad_norm": 0.45715412497520447,
      "learning_rate": 3.650423671870197e-05,
      "loss": 0.5703,
      "step": 884
    },
    {
      "epoch": 1.13184,
      "grad_norm": 0.5793402791023254,
      "learning_rate": 3.649540048471927e-05,
      "loss": 0.6081,
      "step": 885
    },
    {
      "epoch": 1.13312,
      "grad_norm": 0.506584107875824,
      "learning_rate": 3.648655430141927e-05,
      "loss": 0.6014,
      "step": 886
    },
    {
      "epoch": 1.1344,
      "grad_norm": 0.602027416229248,
      "learning_rate": 3.647769817487377e-05,
      "loss": 0.6088,
      "step": 887
    },
    {
      "epoch": 1.13568,
      "grad_norm": 0.5856837630271912,
      "learning_rate": 3.646883211116143e-05,
      "loss": 0.6525,
      "step": 888
    },
    {
      "epoch": 1.13696,
      "grad_norm": 0.6262993216514587,
      "learning_rate": 3.645995611636772e-05,
      "loss": 0.6051,
      "step": 889
    },
    {
      "epoch": 1.13824,
      "grad_norm": 0.5652395486831665,
      "learning_rate": 3.645107019658491e-05,
      "loss": 0.6224,
      "step": 890
    },
    {
      "epoch": 1.13952,
      "grad_norm": 0.559037983417511,
      "learning_rate": 3.64421743579121e-05,
      "loss": 0.6392,
      "step": 891
    },
    {
      "epoch": 1.1408,
      "grad_norm": 0.63148432970047,
      "learning_rate": 3.643326860645521e-05,
      "loss": 0.6376,
      "step": 892
    },
    {
      "epoch": 1.14208,
      "grad_norm": 0.5662648677825928,
      "learning_rate": 3.642435294832692e-05,
      "loss": 0.62,
      "step": 893
    },
    {
      "epoch": 1.14336,
      "grad_norm": 0.6114415526390076,
      "learning_rate": 3.641542738964676e-05,
      "loss": 0.6087,
      "step": 894
    },
    {
      "epoch": 1.1446399999999999,
      "grad_norm": 0.5445461273193359,
      "learning_rate": 3.640649193654102e-05,
      "loss": 0.6097,
      "step": 895
    },
    {
      "epoch": 1.14592,
      "grad_norm": 0.667611300945282,
      "learning_rate": 3.639754659514281e-05,
      "loss": 0.6237,
      "step": 896
    },
    {
      "epoch": 1.1472,
      "grad_norm": 0.5402193665504456,
      "learning_rate": 3.638859137159199e-05,
      "loss": 0.6222,
      "step": 897
    },
    {
      "epoch": 1.14848,
      "grad_norm": 0.5603547096252441,
      "learning_rate": 3.6379626272035233e-05,
      "loss": 0.6168,
      "step": 898
    },
    {
      "epoch": 1.1497600000000001,
      "grad_norm": 0.5758335590362549,
      "learning_rate": 3.637065130262599e-05,
      "loss": 0.6093,
      "step": 899
    },
    {
      "epoch": 1.15104,
      "grad_norm": 0.5867034792900085,
      "learning_rate": 3.636166646952448e-05,
      "loss": 0.6363,
      "step": 900
    },
    {
      "epoch": 1.15104,
      "eval_loss": 0.698606550693512,
      "eval_runtime": 19.5797,
      "eval_samples_per_second": 25.639,
      "eval_steps_per_second": 3.218,
      "step": 900
    },
    {
      "epoch": 1.15232,
      "grad_norm": 0.5570172071456909,
      "learning_rate": 3.635267177889767e-05,
      "loss": 0.6584,
      "step": 901
    },
    {
      "epoch": 1.1536,
      "grad_norm": 0.4862557053565979,
      "learning_rate": 3.6343667236919336e-05,
      "loss": 0.5745,
      "step": 902
    },
    {
      "epoch": 1.15488,
      "grad_norm": 0.6052874326705933,
      "learning_rate": 3.633465284976998e-05,
      "loss": 0.6047,
      "step": 903
    },
    {
      "epoch": 1.15616,
      "grad_norm": 0.5398381948471069,
      "learning_rate": 3.632562862363688e-05,
      "loss": 0.6281,
      "step": 904
    },
    {
      "epoch": 1.15744,
      "grad_norm": 0.593108594417572,
      "learning_rate": 3.6316594564714055e-05,
      "loss": 0.6211,
      "step": 905
    },
    {
      "epoch": 1.15872,
      "grad_norm": 0.495141863822937,
      "learning_rate": 3.630755067920229e-05,
      "loss": 0.6155,
      "step": 906
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.56572026014328,
      "learning_rate": 3.629849697330911e-05,
      "loss": 0.6289,
      "step": 907
    },
    {
      "epoch": 1.16128,
      "grad_norm": 0.5484992265701294,
      "learning_rate": 3.6289433453248755e-05,
      "loss": 0.6124,
      "step": 908
    },
    {
      "epoch": 1.16256,
      "grad_norm": 0.5302249193191528,
      "learning_rate": 3.6280360125242234e-05,
      "loss": 0.5973,
      "step": 909
    },
    {
      "epoch": 1.16384,
      "grad_norm": 0.6414827704429626,
      "learning_rate": 3.6271276995517275e-05,
      "loss": 0.6596,
      "step": 910
    },
    {
      "epoch": 1.16512,
      "grad_norm": 0.5986211895942688,
      "learning_rate": 3.626218407030834e-05,
      "loss": 0.6436,
      "step": 911
    },
    {
      "epoch": 1.1663999999999999,
      "grad_norm": 0.5722213983535767,
      "learning_rate": 3.62530813558566e-05,
      "loss": 0.6588,
      "step": 912
    },
    {
      "epoch": 1.16768,
      "grad_norm": 0.5355356931686401,
      "learning_rate": 3.6243968858409955e-05,
      "loss": 0.6551,
      "step": 913
    },
    {
      "epoch": 1.16896,
      "grad_norm": 0.5994006395339966,
      "learning_rate": 3.623484658422302e-05,
      "loss": 0.6507,
      "step": 914
    },
    {
      "epoch": 1.17024,
      "grad_norm": 0.5505383014678955,
      "learning_rate": 3.622571453955712e-05,
      "loss": 0.6513,
      "step": 915
    },
    {
      "epoch": 1.1715200000000001,
      "grad_norm": 0.5609574913978577,
      "learning_rate": 3.6216572730680285e-05,
      "loss": 0.6831,
      "step": 916
    },
    {
      "epoch": 1.1728,
      "grad_norm": 0.5898633003234863,
      "learning_rate": 3.620742116386724e-05,
      "loss": 0.6423,
      "step": 917
    },
    {
      "epoch": 1.17408,
      "grad_norm": 0.5470394492149353,
      "learning_rate": 3.6198259845399424e-05,
      "loss": 0.6458,
      "step": 918
    },
    {
      "epoch": 1.17536,
      "grad_norm": 0.528093695640564,
      "learning_rate": 3.618908878156496e-05,
      "loss": 0.6353,
      "step": 919
    },
    {
      "epoch": 1.17664,
      "grad_norm": 0.5467092394828796,
      "learning_rate": 3.617990797865866e-05,
      "loss": 0.6065,
      "step": 920
    },
    {
      "epoch": 1.17792,
      "grad_norm": 0.5420581102371216,
      "learning_rate": 3.617071744298201e-05,
      "loss": 0.6048,
      "step": 921
    },
    {
      "epoch": 1.1792,
      "grad_norm": 0.5671415328979492,
      "learning_rate": 3.6161517180843204e-05,
      "loss": 0.6199,
      "step": 922
    },
    {
      "epoch": 1.18048,
      "grad_norm": 0.45265355706214905,
      "learning_rate": 3.615230719855709e-05,
      "loss": 0.5986,
      "step": 923
    },
    {
      "epoch": 1.18176,
      "grad_norm": 0.5522210597991943,
      "learning_rate": 3.614308750244518e-05,
      "loss": 0.6722,
      "step": 924
    },
    {
      "epoch": 1.18304,
      "grad_norm": 0.5389830470085144,
      "learning_rate": 3.613385809883569e-05,
      "loss": 0.6317,
      "step": 925
    },
    {
      "epoch": 1.18432,
      "grad_norm": 0.48764264583587646,
      "learning_rate": 3.6124618994063446e-05,
      "loss": 0.5848,
      "step": 926
    },
    {
      "epoch": 1.1856,
      "grad_norm": 0.507773756980896,
      "learning_rate": 3.611537019446999e-05,
      "loss": 0.5849,
      "step": 927
    },
    {
      "epoch": 1.18688,
      "grad_norm": 0.5008213520050049,
      "learning_rate": 3.610611170640347e-05,
      "loss": 0.6242,
      "step": 928
    },
    {
      "epoch": 1.1881599999999999,
      "grad_norm": 0.43542903661727905,
      "learning_rate": 3.6096843536218714e-05,
      "loss": 0.6135,
      "step": 929
    },
    {
      "epoch": 1.18944,
      "grad_norm": 0.5166139006614685,
      "learning_rate": 3.608756569027719e-05,
      "loss": 0.6674,
      "step": 930
    },
    {
      "epoch": 1.18944,
      "eval_loss": 0.6970492005348206,
      "eval_runtime": 19.5847,
      "eval_samples_per_second": 25.632,
      "eval_steps_per_second": 3.217,
      "step": 930
    },
    {
      "epoch": 1.19072,
      "grad_norm": 0.5094209909439087,
      "learning_rate": 3.607827817494699e-05,
      "loss": 0.6026,
      "step": 931
    },
    {
      "epoch": 1.192,
      "grad_norm": 0.48830729722976685,
      "learning_rate": 3.6068980996602874e-05,
      "loss": 0.6241,
      "step": 932
    },
    {
      "epoch": 1.1932800000000001,
      "grad_norm": 0.5363035202026367,
      "learning_rate": 3.605967416162619e-05,
      "loss": 0.6038,
      "step": 933
    },
    {
      "epoch": 1.19456,
      "grad_norm": 0.4707103371620178,
      "learning_rate": 3.605035767640497e-05,
      "loss": 0.612,
      "step": 934
    },
    {
      "epoch": 1.19584,
      "grad_norm": 0.5514024496078491,
      "learning_rate": 3.604103154733382e-05,
      "loss": 0.6244,
      "step": 935
    },
    {
      "epoch": 1.19712,
      "grad_norm": 0.5555980205535889,
      "learning_rate": 3.6031695780813994e-05,
      "loss": 0.6179,
      "step": 936
    },
    {
      "epoch": 1.1984,
      "grad_norm": 0.5490615963935852,
      "learning_rate": 3.602235038325335e-05,
      "loss": 0.6473,
      "step": 937
    },
    {
      "epoch": 1.19968,
      "grad_norm": 0.5086703300476074,
      "learning_rate": 3.601299536106636e-05,
      "loss": 0.6592,
      "step": 938
    },
    {
      "epoch": 1.20096,
      "grad_norm": 0.5780916810035706,
      "learning_rate": 3.60036307206741e-05,
      "loss": 0.6486,
      "step": 939
    },
    {
      "epoch": 1.20224,
      "grad_norm": 0.45233604311943054,
      "learning_rate": 3.5994256468504245e-05,
      "loss": 0.5867,
      "step": 940
    },
    {
      "epoch": 1.20352,
      "grad_norm": 0.5436866879463196,
      "learning_rate": 3.598487261099108e-05,
      "loss": 0.6326,
      "step": 941
    },
    {
      "epoch": 1.2048,
      "grad_norm": 0.5055412650108337,
      "learning_rate": 3.597547915457547e-05,
      "loss": 0.6237,
      "step": 942
    },
    {
      "epoch": 1.20608,
      "grad_norm": 0.5268957614898682,
      "learning_rate": 3.596607610570487e-05,
      "loss": 0.6123,
      "step": 943
    },
    {
      "epoch": 1.20736,
      "grad_norm": 0.5263251662254333,
      "learning_rate": 3.5956663470833326e-05,
      "loss": 0.6405,
      "step": 944
    },
    {
      "epoch": 1.20864,
      "grad_norm": 0.5110918283462524,
      "learning_rate": 3.594724125642145e-05,
      "loss": 0.5997,
      "step": 945
    },
    {
      "epoch": 1.2099199999999999,
      "grad_norm": 0.50609290599823,
      "learning_rate": 3.593780946893645e-05,
      "loss": 0.6166,
      "step": 946
    },
    {
      "epoch": 1.2112,
      "grad_norm": 0.5139018297195435,
      "learning_rate": 3.592836811485209e-05,
      "loss": 0.6088,
      "step": 947
    },
    {
      "epoch": 1.21248,
      "grad_norm": 0.540569543838501,
      "learning_rate": 3.59189172006487e-05,
      "loss": 0.6405,
      "step": 948
    },
    {
      "epoch": 1.21376,
      "grad_norm": 0.6000634431838989,
      "learning_rate": 3.5909456732813175e-05,
      "loss": 0.6101,
      "step": 949
    },
    {
      "epoch": 1.2150400000000001,
      "grad_norm": 0.5738515257835388,
      "learning_rate": 3.589998671783897e-05,
      "loss": 0.6204,
      "step": 950
    },
    {
      "epoch": 1.21632,
      "grad_norm": 0.5462875366210938,
      "learning_rate": 3.589050716222609e-05,
      "loss": 0.6022,
      "step": 951
    },
    {
      "epoch": 1.2176,
      "grad_norm": 0.6359582543373108,
      "learning_rate": 3.588101807248109e-05,
      "loss": 0.6238,
      "step": 952
    },
    {
      "epoch": 1.21888,
      "grad_norm": 0.4347775876522064,
      "learning_rate": 3.587151945511707e-05,
      "loss": 0.6062,
      "step": 953
    },
    {
      "epoch": 1.22016,
      "grad_norm": 0.5784165263175964,
      "learning_rate": 3.586201131665366e-05,
      "loss": 0.5993,
      "step": 954
    },
    {
      "epoch": 1.22144,
      "grad_norm": 0.5360873937606812,
      "learning_rate": 3.585249366361703e-05,
      "loss": 0.6223,
      "step": 955
    },
    {
      "epoch": 1.22272,
      "grad_norm": 0.5133347511291504,
      "learning_rate": 3.58429665025399e-05,
      "loss": 0.6142,
      "step": 956
    },
    {
      "epoch": 1.224,
      "grad_norm": 0.5823657512664795,
      "learning_rate": 3.583342983996149e-05,
      "loss": 0.6314,
      "step": 957
    },
    {
      "epoch": 1.22528,
      "grad_norm": 0.4947901964187622,
      "learning_rate": 3.5823883682427545e-05,
      "loss": 0.6403,
      "step": 958
    },
    {
      "epoch": 1.22656,
      "grad_norm": 0.5722011923789978,
      "learning_rate": 3.5814328036490346e-05,
      "loss": 0.6395,
      "step": 959
    },
    {
      "epoch": 1.22784,
      "grad_norm": 0.5118551254272461,
      "learning_rate": 3.5804762908708665e-05,
      "loss": 0.5985,
      "step": 960
    },
    {
      "epoch": 1.22784,
      "eval_loss": 0.6965128779411316,
      "eval_runtime": 19.6575,
      "eval_samples_per_second": 25.537,
      "eval_steps_per_second": 3.205,
      "step": 960
    },
    {
      "epoch": 1.22912,
      "grad_norm": 0.5312982201576233,
      "learning_rate": 3.5795188305647786e-05,
      "loss": 0.6154,
      "step": 961
    },
    {
      "epoch": 1.2304,
      "grad_norm": 0.547223687171936,
      "learning_rate": 3.5785604233879515e-05,
      "loss": 0.5913,
      "step": 962
    },
    {
      "epoch": 1.2316799999999999,
      "grad_norm": 0.49858781695365906,
      "learning_rate": 3.577601069998214e-05,
      "loss": 0.6093,
      "step": 963
    },
    {
      "epoch": 1.23296,
      "grad_norm": 0.593039333820343,
      "learning_rate": 3.5766407710540435e-05,
      "loss": 0.6488,
      "step": 964
    },
    {
      "epoch": 1.23424,
      "grad_norm": 0.45781368017196655,
      "learning_rate": 3.575679527214569e-05,
      "loss": 0.6349,
      "step": 965
    },
    {
      "epoch": 1.23552,
      "grad_norm": 0.6406901478767395,
      "learning_rate": 3.5747173391395665e-05,
      "loss": 0.6227,
      "step": 966
    },
    {
      "epoch": 1.2368000000000001,
      "grad_norm": 0.5171946883201599,
      "learning_rate": 3.5737542074894595e-05,
      "loss": 0.6233,
      "step": 967
    },
    {
      "epoch": 1.23808,
      "grad_norm": 0.5127807259559631,
      "learning_rate": 3.57279013292532e-05,
      "loss": 0.5928,
      "step": 968
    },
    {
      "epoch": 1.23936,
      "grad_norm": 0.5854061841964722,
      "learning_rate": 3.571825116108868e-05,
      "loss": 0.628,
      "step": 969
    },
    {
      "epoch": 1.24064,
      "grad_norm": 0.4943411648273468,
      "learning_rate": 3.570859157702469e-05,
      "loss": 0.6013,
      "step": 970
    },
    {
      "epoch": 1.24192,
      "grad_norm": 0.5105839371681213,
      "learning_rate": 3.569892258369135e-05,
      "loss": 0.6559,
      "step": 971
    },
    {
      "epoch": 1.2432,
      "grad_norm": 0.522589921951294,
      "learning_rate": 3.568924418772524e-05,
      "loss": 0.6098,
      "step": 972
    },
    {
      "epoch": 1.24448,
      "grad_norm": 0.5029391646385193,
      "learning_rate": 3.567955639576939e-05,
      "loss": 0.5926,
      "step": 973
    },
    {
      "epoch": 1.24576,
      "grad_norm": 0.4692816436290741,
      "learning_rate": 3.56698592144733e-05,
      "loss": 0.6099,
      "step": 974
    },
    {
      "epoch": 1.24704,
      "grad_norm": 0.5588908195495605,
      "learning_rate": 3.5660152650492874e-05,
      "loss": 0.6235,
      "step": 975
    },
    {
      "epoch": 1.24832,
      "grad_norm": 0.4936743378639221,
      "learning_rate": 3.565043671049049e-05,
      "loss": 0.6042,
      "step": 976
    },
    {
      "epoch": 1.2496,
      "grad_norm": 0.5436892509460449,
      "learning_rate": 3.564071140113495e-05,
      "loss": 0.6449,
      "step": 977
    },
    {
      "epoch": 1.25088,
      "grad_norm": 0.5177863240242004,
      "learning_rate": 3.563097672910149e-05,
      "loss": 0.6341,
      "step": 978
    },
    {
      "epoch": 1.25216,
      "grad_norm": 0.5292809009552002,
      "learning_rate": 3.562123270107177e-05,
      "loss": 0.6413,
      "step": 979
    },
    {
      "epoch": 1.2534399999999999,
      "grad_norm": 0.6132381558418274,
      "learning_rate": 3.561147932373387e-05,
      "loss": 0.6349,
      "step": 980
    },
    {
      "epoch": 1.25472,
      "grad_norm": 0.5383648872375488,
      "learning_rate": 3.5601716603782295e-05,
      "loss": 0.6436,
      "step": 981
    },
    {
      "epoch": 1.256,
      "grad_norm": 0.5148691534996033,
      "learning_rate": 3.559194454791795e-05,
      "loss": 0.6268,
      "step": 982
    },
    {
      "epoch": 1.25728,
      "grad_norm": 0.587352991104126,
      "learning_rate": 3.558216316284815e-05,
      "loss": 0.6208,
      "step": 983
    },
    {
      "epoch": 1.2585600000000001,
      "grad_norm": 0.4563191831111908,
      "learning_rate": 3.557237245528662e-05,
      "loss": 0.6048,
      "step": 984
    },
    {
      "epoch": 1.25984,
      "grad_norm": 0.5396101474761963,
      "learning_rate": 3.556257243195349e-05,
      "loss": 0.6106,
      "step": 985
    },
    {
      "epoch": 1.26112,
      "grad_norm": 0.5514459013938904,
      "learning_rate": 3.555276309957528e-05,
      "loss": 0.6365,
      "step": 986
    },
    {
      "epoch": 1.2624,
      "grad_norm": 0.5700104832649231,
      "learning_rate": 3.554294446488488e-05,
      "loss": 0.6643,
      "step": 987
    },
    {
      "epoch": 1.26368,
      "grad_norm": 0.5242200493812561,
      "learning_rate": 3.55331165346216e-05,
      "loss": 0.6082,
      "step": 988
    },
    {
      "epoch": 1.2649599999999999,
      "grad_norm": 0.5709508061408997,
      "learning_rate": 3.5523279315531084e-05,
      "loss": 0.6472,
      "step": 989
    },
    {
      "epoch": 1.26624,
      "grad_norm": 0.5788748860359192,
      "learning_rate": 3.55134328143654e-05,
      "loss": 0.6538,
      "step": 990
    },
    {
      "epoch": 1.26624,
      "eval_loss": 0.6963632106781006,
      "eval_runtime": 19.5744,
      "eval_samples_per_second": 25.646,
      "eval_steps_per_second": 3.218,
      "step": 990
    },
    {
      "epoch": 1.26752,
      "grad_norm": 0.5791829228401184,
      "learning_rate": 3.550357703788295e-05,
      "loss": 0.6379,
      "step": 991
    },
    {
      "epoch": 1.2688,
      "grad_norm": 0.5710839629173279,
      "learning_rate": 3.5493711992848536e-05,
      "loss": 0.6348,
      "step": 992
    },
    {
      "epoch": 1.27008,
      "grad_norm": 0.5680004954338074,
      "learning_rate": 3.5483837686033284e-05,
      "loss": 0.6371,
      "step": 993
    },
    {
      "epoch": 1.27136,
      "grad_norm": 0.6242326498031616,
      "learning_rate": 3.547395412421471e-05,
      "loss": 0.651,
      "step": 994
    },
    {
      "epoch": 1.27264,
      "grad_norm": 0.5174843072891235,
      "learning_rate": 3.546406131417666e-05,
      "loss": 0.6073,
      "step": 995
    },
    {
      "epoch": 1.27392,
      "grad_norm": 0.6075764298439026,
      "learning_rate": 3.5454159262709354e-05,
      "loss": 0.5872,
      "step": 996
    },
    {
      "epoch": 1.2752,
      "grad_norm": 0.47644922137260437,
      "learning_rate": 3.544424797660931e-05,
      "loss": 0.6211,
      "step": 997
    },
    {
      "epoch": 1.27648,
      "grad_norm": 0.5990807414054871,
      "learning_rate": 3.543432746267943e-05,
      "loss": 0.6105,
      "step": 998
    },
    {
      "epoch": 1.27776,
      "grad_norm": 0.555572509765625,
      "learning_rate": 3.542439772772893e-05,
      "loss": 0.6154,
      "step": 999
    },
    {
      "epoch": 1.27904,
      "grad_norm": 0.4940454065799713,
      "learning_rate": 3.541445877857335e-05,
      "loss": 0.6439,
      "step": 1000
    },
    {
      "epoch": 1.2803200000000001,
      "grad_norm": 0.5553591847419739,
      "learning_rate": 3.540451062203457e-05,
      "loss": 0.6261,
      "step": 1001
    },
    {
      "epoch": 1.2816,
      "grad_norm": 0.43316078186035156,
      "learning_rate": 3.539455326494078e-05,
      "loss": 0.6242,
      "step": 1002
    },
    {
      "epoch": 1.28288,
      "grad_norm": 0.491619348526001,
      "learning_rate": 3.5384586714126486e-05,
      "loss": 0.6303,
      "step": 1003
    },
    {
      "epoch": 1.28416,
      "grad_norm": 0.48678329586982727,
      "learning_rate": 3.5374610976432496e-05,
      "loss": 0.614,
      "step": 1004
    },
    {
      "epoch": 1.28544,
      "grad_norm": 0.47967180609703064,
      "learning_rate": 3.5364626058705944e-05,
      "loss": 0.5969,
      "step": 1005
    },
    {
      "epoch": 1.2867199999999999,
      "grad_norm": 0.5006229877471924,
      "learning_rate": 3.535463196780025e-05,
      "loss": 0.6339,
      "step": 1006
    },
    {
      "epoch": 1.288,
      "grad_norm": 0.5121135711669922,
      "learning_rate": 3.534462871057514e-05,
      "loss": 0.6084,
      "step": 1007
    },
    {
      "epoch": 1.28928,
      "grad_norm": 0.5153780579566956,
      "learning_rate": 3.5334616293896596e-05,
      "loss": 0.6134,
      "step": 1008
    },
    {
      "epoch": 1.29056,
      "grad_norm": 0.5448907017707825,
      "learning_rate": 3.532459472463695e-05,
      "loss": 0.6679,
      "step": 1009
    },
    {
      "epoch": 1.29184,
      "grad_norm": 0.5324149131774902,
      "learning_rate": 3.531456400967477e-05,
      "loss": 0.6166,
      "step": 1010
    },
    {
      "epoch": 1.29312,
      "grad_norm": 0.46375569701194763,
      "learning_rate": 3.530452415589491e-05,
      "loss": 0.593,
      "step": 1011
    },
    {
      "epoch": 1.2944,
      "grad_norm": 0.4514268934726715,
      "learning_rate": 3.52944751701885e-05,
      "loss": 0.6042,
      "step": 1012
    },
    {
      "epoch": 1.29568,
      "grad_norm": 0.4675837457180023,
      "learning_rate": 3.528441705945294e-05,
      "loss": 0.6186,
      "step": 1013
    },
    {
      "epoch": 1.29696,
      "grad_norm": 0.48428410291671753,
      "learning_rate": 3.52743498305919e-05,
      "loss": 0.6145,
      "step": 1014
    },
    {
      "epoch": 1.29824,
      "grad_norm": 0.4517797827720642,
      "learning_rate": 3.526427349051528e-05,
      "loss": 0.6032,
      "step": 1015
    },
    {
      "epoch": 1.29952,
      "grad_norm": 0.4889644980430603,
      "learning_rate": 3.525418804613928e-05,
      "loss": 0.6051,
      "step": 1016
    },
    {
      "epoch": 1.3008,
      "grad_norm": 0.5002602338790894,
      "learning_rate": 3.52440935043863e-05,
      "loss": 0.6506,
      "step": 1017
    },
    {
      "epoch": 1.3020800000000001,
      "grad_norm": 0.5081722736358643,
      "learning_rate": 3.523398987218502e-05,
      "loss": 0.6092,
      "step": 1018
    },
    {
      "epoch": 1.30336,
      "grad_norm": 0.5410910248756409,
      "learning_rate": 3.5223877156470333e-05,
      "loss": 0.5803,
      "step": 1019
    },
    {
      "epoch": 1.30464,
      "grad_norm": 0.5094830393791199,
      "learning_rate": 3.521375536418339e-05,
      "loss": 0.6372,
      "step": 1020
    },
    {
      "epoch": 1.30464,
      "eval_loss": 0.6951881051063538,
      "eval_runtime": 19.589,
      "eval_samples_per_second": 25.627,
      "eval_steps_per_second": 3.216,
      "step": 1020
    },
    {
      "epoch": 1.30592,
      "grad_norm": 0.491112619638443,
      "learning_rate": 3.520362450227155e-05,
      "loss": 0.6375,
      "step": 1021
    },
    {
      "epoch": 1.3072,
      "grad_norm": 0.5696011185646057,
      "learning_rate": 3.519348457768844e-05,
      "loss": 0.6124,
      "step": 1022
    },
    {
      "epoch": 1.3084799999999999,
      "grad_norm": 0.5652121901512146,
      "learning_rate": 3.518333559739384e-05,
      "loss": 0.6174,
      "step": 1023
    },
    {
      "epoch": 1.30976,
      "grad_norm": 0.5401461124420166,
      "learning_rate": 3.5173177568353796e-05,
      "loss": 0.6268,
      "step": 1024
    },
    {
      "epoch": 1.31104,
      "grad_norm": 0.5565940737724304,
      "learning_rate": 3.5163010497540556e-05,
      "loss": 0.6318,
      "step": 1025
    },
    {
      "epoch": 1.31232,
      "grad_norm": 0.5271297693252563,
      "learning_rate": 3.515283439193257e-05,
      "loss": 0.6172,
      "step": 1026
    },
    {
      "epoch": 1.3136,
      "grad_norm": 0.5320202708244324,
      "learning_rate": 3.514264925851448e-05,
      "loss": 0.6002,
      "step": 1027
    },
    {
      "epoch": 1.31488,
      "grad_norm": 0.596748411655426,
      "learning_rate": 3.5132455104277135e-05,
      "loss": 0.6283,
      "step": 1028
    },
    {
      "epoch": 1.31616,
      "grad_norm": 0.5227011442184448,
      "learning_rate": 3.5122251936217575e-05,
      "loss": 0.5963,
      "step": 1029
    },
    {
      "epoch": 1.31744,
      "grad_norm": 0.5354651808738708,
      "learning_rate": 3.511203976133903e-05,
      "loss": 0.6141,
      "step": 1030
    },
    {
      "epoch": 1.31872,
      "grad_norm": 0.5774796009063721,
      "learning_rate": 3.51018185866509e-05,
      "loss": 0.6192,
      "step": 1031
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5244203209877014,
      "learning_rate": 3.509158841916877e-05,
      "loss": 0.6121,
      "step": 1032
    },
    {
      "epoch": 1.32128,
      "grad_norm": 0.5258281826972961,
      "learning_rate": 3.50813492659144e-05,
      "loss": 0.6145,
      "step": 1033
    },
    {
      "epoch": 1.32256,
      "grad_norm": 0.45995625853538513,
      "learning_rate": 3.507110113391571e-05,
      "loss": 0.5802,
      "step": 1034
    },
    {
      "epoch": 1.3238400000000001,
      "grad_norm": 0.5954358577728271,
      "learning_rate": 3.50608440302068e-05,
      "loss": 0.6144,
      "step": 1035
    },
    {
      "epoch": 1.32512,
      "grad_norm": 0.502953290939331,
      "learning_rate": 3.5050577961827904e-05,
      "loss": 0.6278,
      "step": 1036
    },
    {
      "epoch": 1.3264,
      "grad_norm": 0.5640849471092224,
      "learning_rate": 3.504030293582543e-05,
      "loss": 0.6198,
      "step": 1037
    },
    {
      "epoch": 1.32768,
      "grad_norm": 0.47054290771484375,
      "learning_rate": 3.5030018959251915e-05,
      "loss": 0.6064,
      "step": 1038
    },
    {
      "epoch": 1.32896,
      "grad_norm": 0.4852832555770874,
      "learning_rate": 3.5019726039166055e-05,
      "loss": 0.5979,
      "step": 1039
    },
    {
      "epoch": 1.3302399999999999,
      "grad_norm": 0.5038312673568726,
      "learning_rate": 3.500942418263267e-05,
      "loss": 0.6138,
      "step": 1040
    },
    {
      "epoch": 1.33152,
      "grad_norm": 0.5502773523330688,
      "learning_rate": 3.499911339672274e-05,
      "loss": 0.6213,
      "step": 1041
    },
    {
      "epoch": 1.3328,
      "grad_norm": 0.5487447381019592,
      "learning_rate": 3.498879368851335e-05,
      "loss": 0.6464,
      "step": 1042
    },
    {
      "epoch": 1.33408,
      "grad_norm": 0.5230963230133057,
      "learning_rate": 3.497846506508771e-05,
      "loss": 0.6214,
      "step": 1043
    },
    {
      "epoch": 1.33536,
      "grad_norm": 0.44461992383003235,
      "learning_rate": 3.496812753353516e-05,
      "loss": 0.5936,
      "step": 1044
    },
    {
      "epoch": 1.33664,
      "grad_norm": 0.5880510807037354,
      "learning_rate": 3.495778110095114e-05,
      "loss": 0.6243,
      "step": 1045
    },
    {
      "epoch": 1.33792,
      "grad_norm": 0.5631331205368042,
      "learning_rate": 3.494742577443723e-05,
      "loss": 0.6283,
      "step": 1046
    },
    {
      "epoch": 1.3392,
      "grad_norm": 0.578289806842804,
      "learning_rate": 3.493706156110107e-05,
      "loss": 0.6031,
      "step": 1047
    },
    {
      "epoch": 1.34048,
      "grad_norm": 0.5098040103912354,
      "learning_rate": 3.492668846805644e-05,
      "loss": 0.6074,
      "step": 1048
    },
    {
      "epoch": 1.34176,
      "grad_norm": 0.5466323494911194,
      "learning_rate": 3.491630650242319e-05,
      "loss": 0.6427,
      "step": 1049
    },
    {
      "epoch": 1.34304,
      "grad_norm": 0.5978725552558899,
      "learning_rate": 3.4905915671327274e-05,
      "loss": 0.6275,
      "step": 1050
    },
    {
      "epoch": 1.34304,
      "eval_loss": 0.6936712265014648,
      "eval_runtime": 19.5664,
      "eval_samples_per_second": 25.656,
      "eval_steps_per_second": 3.22,
      "step": 1050
    },
    {
      "epoch": 1.34432,
      "grad_norm": 0.5109580159187317,
      "learning_rate": 3.489551598190072e-05,
      "loss": 0.5999,
      "step": 1051
    },
    {
      "epoch": 1.3456000000000001,
      "grad_norm": 0.5138126611709595,
      "learning_rate": 3.488510744128164e-05,
      "loss": 0.6155,
      "step": 1052
    },
    {
      "epoch": 1.34688,
      "grad_norm": 0.5189249515533447,
      "learning_rate": 3.4874690056614225e-05,
      "loss": 0.6347,
      "step": 1053
    },
    {
      "epoch": 1.34816,
      "grad_norm": 0.5142697691917419,
      "learning_rate": 3.4864263835048735e-05,
      "loss": 0.6096,
      "step": 1054
    },
    {
      "epoch": 1.34944,
      "grad_norm": 0.476410448551178,
      "learning_rate": 3.48538287837415e-05,
      "loss": 0.5797,
      "step": 1055
    },
    {
      "epoch": 1.35072,
      "grad_norm": 0.5054307579994202,
      "learning_rate": 3.484338490985489e-05,
      "loss": 0.6241,
      "step": 1056
    },
    {
      "epoch": 1.3519999999999999,
      "grad_norm": 0.48359036445617676,
      "learning_rate": 3.4832932220557355e-05,
      "loss": 0.6301,
      "step": 1057
    },
    {
      "epoch": 1.35328,
      "grad_norm": 0.4742510914802551,
      "learning_rate": 3.4822470723023385e-05,
      "loss": 0.5892,
      "step": 1058
    },
    {
      "epoch": 1.35456,
      "grad_norm": 0.5535560846328735,
      "learning_rate": 3.481200042443352e-05,
      "loss": 0.6313,
      "step": 1059
    },
    {
      "epoch": 1.35584,
      "grad_norm": 0.47079795598983765,
      "learning_rate": 3.480152133197433e-05,
      "loss": 0.5919,
      "step": 1060
    },
    {
      "epoch": 1.35712,
      "grad_norm": 0.5069361329078674,
      "learning_rate": 3.479103345283843e-05,
      "loss": 0.6088,
      "step": 1061
    },
    {
      "epoch": 1.3584,
      "grad_norm": 0.4823542833328247,
      "learning_rate": 3.478053679422447e-05,
      "loss": 0.5995,
      "step": 1062
    },
    {
      "epoch": 1.35968,
      "grad_norm": 0.4795922338962555,
      "learning_rate": 3.4770031363337126e-05,
      "loss": 0.6263,
      "step": 1063
    },
    {
      "epoch": 1.36096,
      "grad_norm": 0.48971349000930786,
      "learning_rate": 3.475951716738708e-05,
      "loss": 0.651,
      "step": 1064
    },
    {
      "epoch": 1.36224,
      "grad_norm": 0.5213609933853149,
      "learning_rate": 3.474899421359103e-05,
      "loss": 0.6064,
      "step": 1065
    },
    {
      "epoch": 1.36352,
      "grad_norm": 0.5053293704986572,
      "learning_rate": 3.473846250917172e-05,
      "loss": 0.6366,
      "step": 1066
    },
    {
      "epoch": 1.3648,
      "grad_norm": 0.4546721875667572,
      "learning_rate": 3.472792206135786e-05,
      "loss": 0.5996,
      "step": 1067
    },
    {
      "epoch": 1.36608,
      "grad_norm": 0.5525872707366943,
      "learning_rate": 3.4717372877384175e-05,
      "loss": 0.6398,
      "step": 1068
    },
    {
      "epoch": 1.3673600000000001,
      "grad_norm": 0.4695347547531128,
      "learning_rate": 3.4706814964491395e-05,
      "loss": 0.5904,
      "step": 1069
    },
    {
      "epoch": 1.36864,
      "grad_norm": 0.5401780009269714,
      "learning_rate": 3.469624832992623e-05,
      "loss": 0.646,
      "step": 1070
    },
    {
      "epoch": 1.36992,
      "grad_norm": 0.5475379824638367,
      "learning_rate": 3.4685672980941384e-05,
      "loss": 0.6029,
      "step": 1071
    },
    {
      "epoch": 1.3712,
      "grad_norm": 0.5604281425476074,
      "learning_rate": 3.4675088924795535e-05,
      "loss": 0.5991,
      "step": 1072
    },
    {
      "epoch": 1.37248,
      "grad_norm": 0.5799840688705444,
      "learning_rate": 3.466449616875335e-05,
      "loss": 0.6004,
      "step": 1073
    },
    {
      "epoch": 1.3737599999999999,
      "grad_norm": 0.4855911135673523,
      "learning_rate": 3.4653894720085443e-05,
      "loss": 0.604,
      "step": 1074
    },
    {
      "epoch": 1.37504,
      "grad_norm": 0.5208277106285095,
      "learning_rate": 3.464328458606843e-05,
      "loss": 0.6133,
      "step": 1075
    },
    {
      "epoch": 1.37632,
      "grad_norm": 0.5047687888145447,
      "learning_rate": 3.463266577398485e-05,
      "loss": 0.6162,
      "step": 1076
    },
    {
      "epoch": 1.3776,
      "grad_norm": 0.5023367404937744,
      "learning_rate": 3.462203829112323e-05,
      "loss": 0.6345,
      "step": 1077
    },
    {
      "epoch": 1.37888,
      "grad_norm": 0.4942294657230377,
      "learning_rate": 3.4611402144778027e-05,
      "loss": 0.605,
      "step": 1078
    },
    {
      "epoch": 1.38016,
      "grad_norm": 0.48462584614753723,
      "learning_rate": 3.460075734224966e-05,
      "loss": 0.6179,
      "step": 1079
    },
    {
      "epoch": 1.38144,
      "grad_norm": 0.5583162903785706,
      "learning_rate": 3.459010389084448e-05,
      "loss": 0.6331,
      "step": 1080
    },
    {
      "epoch": 1.38144,
      "eval_loss": 0.6926246285438538,
      "eval_runtime": 19.5831,
      "eval_samples_per_second": 25.634,
      "eval_steps_per_second": 3.217,
      "step": 1080
    },
    {
      "epoch": 1.38272,
      "grad_norm": 0.4886813461780548,
      "learning_rate": 3.457944179787477e-05,
      "loss": 0.5988,
      "step": 1081
    },
    {
      "epoch": 1.384,
      "grad_norm": 0.5811007618904114,
      "learning_rate": 3.4568771070658764e-05,
      "loss": 0.6153,
      "step": 1082
    },
    {
      "epoch": 1.38528,
      "grad_norm": 0.5323143601417542,
      "learning_rate": 3.45580917165206e-05,
      "loss": 0.6386,
      "step": 1083
    },
    {
      "epoch": 1.38656,
      "grad_norm": 0.5751598477363586,
      "learning_rate": 3.4547403742790346e-05,
      "loss": 0.6196,
      "step": 1084
    },
    {
      "epoch": 1.38784,
      "grad_norm": 0.558295726776123,
      "learning_rate": 3.4536707156803996e-05,
      "loss": 0.6185,
      "step": 1085
    },
    {
      "epoch": 1.3891200000000001,
      "grad_norm": 0.5496747493743896,
      "learning_rate": 3.452600196590344e-05,
      "loss": 0.6187,
      "step": 1086
    },
    {
      "epoch": 1.3904,
      "grad_norm": 0.5160159468650818,
      "learning_rate": 3.4515288177436474e-05,
      "loss": 0.6131,
      "step": 1087
    },
    {
      "epoch": 1.39168,
      "grad_norm": 0.582779586315155,
      "learning_rate": 3.450456579875682e-05,
      "loss": 0.6852,
      "step": 1088
    },
    {
      "epoch": 1.39296,
      "grad_norm": 0.49982917308807373,
      "learning_rate": 3.4493834837224067e-05,
      "loss": 0.6254,
      "step": 1089
    },
    {
      "epoch": 1.39424,
      "grad_norm": 0.6003973484039307,
      "learning_rate": 3.44830953002037e-05,
      "loss": 0.6118,
      "step": 1090
    },
    {
      "epoch": 1.3955199999999999,
      "grad_norm": 0.4994569420814514,
      "learning_rate": 3.44723471950671e-05,
      "loss": 0.6379,
      "step": 1091
    },
    {
      "epoch": 1.3968,
      "grad_norm": 0.5951939821243286,
      "learning_rate": 3.446159052919153e-05,
      "loss": 0.6024,
      "step": 1092
    },
    {
      "epoch": 1.39808,
      "grad_norm": 0.5928317308425903,
      "learning_rate": 3.445082530996012e-05,
      "loss": 0.6062,
      "step": 1093
    },
    {
      "epoch": 1.39936,
      "grad_norm": 0.5187568068504333,
      "learning_rate": 3.4440051544761876e-05,
      "loss": 0.6394,
      "step": 1094
    },
    {
      "epoch": 1.40064,
      "grad_norm": 0.5280294418334961,
      "learning_rate": 3.442926924099167e-05,
      "loss": 0.5905,
      "step": 1095
    },
    {
      "epoch": 1.40192,
      "grad_norm": 0.5829623937606812,
      "learning_rate": 3.441847840605023e-05,
      "loss": 0.6426,
      "step": 1096
    },
    {
      "epoch": 1.4032,
      "grad_norm": 0.5589150190353394,
      "learning_rate": 3.440767904734414e-05,
      "loss": 0.6271,
      "step": 1097
    },
    {
      "epoch": 1.40448,
      "grad_norm": 0.5500954389572144,
      "learning_rate": 3.439687117228585e-05,
      "loss": 0.6192,
      "step": 1098
    },
    {
      "epoch": 1.40576,
      "grad_norm": 0.6051462292671204,
      "learning_rate": 3.4386054788293625e-05,
      "loss": 0.6216,
      "step": 1099
    },
    {
      "epoch": 1.40704,
      "grad_norm": 0.49434930086135864,
      "learning_rate": 3.437522990279161e-05,
      "loss": 0.6248,
      "step": 1100
    },
    {
      "epoch": 1.40832,
      "grad_norm": 0.5195952653884888,
      "learning_rate": 3.436439652320973e-05,
      "loss": 0.6022,
      "step": 1101
    },
    {
      "epoch": 1.4096,
      "grad_norm": 0.4615444242954254,
      "learning_rate": 3.435355465698381e-05,
      "loss": 0.5863,
      "step": 1102
    },
    {
      "epoch": 1.4108800000000001,
      "grad_norm": 0.6040896773338318,
      "learning_rate": 3.434270431155544e-05,
      "loss": 0.6094,
      "step": 1103
    },
    {
      "epoch": 1.41216,
      "grad_norm": 0.547451376914978,
      "learning_rate": 3.433184549437206e-05,
      "loss": 0.5989,
      "step": 1104
    },
    {
      "epoch": 1.41344,
      "grad_norm": 0.49374184012413025,
      "learning_rate": 3.432097821288692e-05,
      "loss": 0.6389,
      "step": 1105
    },
    {
      "epoch": 1.41472,
      "grad_norm": 0.5352110862731934,
      "learning_rate": 3.431010247455908e-05,
      "loss": 0.6164,
      "step": 1106
    },
    {
      "epoch": 1.416,
      "grad_norm": 0.515762984752655,
      "learning_rate": 3.4299218286853394e-05,
      "loss": 0.6436,
      "step": 1107
    },
    {
      "epoch": 1.4172799999999999,
      "grad_norm": 0.5139670372009277,
      "learning_rate": 3.428832565724053e-05,
      "loss": 0.5848,
      "step": 1108
    },
    {
      "epoch": 1.41856,
      "grad_norm": 0.5706183314323425,
      "learning_rate": 3.427742459319694e-05,
      "loss": 0.6112,
      "step": 1109
    },
    {
      "epoch": 1.41984,
      "grad_norm": 0.48209401965141296,
      "learning_rate": 3.426651510220488e-05,
      "loss": 0.6046,
      "step": 1110
    },
    {
      "epoch": 1.41984,
      "eval_loss": 0.6923678517341614,
      "eval_runtime": 19.5715,
      "eval_samples_per_second": 25.65,
      "eval_steps_per_second": 3.219,
      "step": 1110
    },
    {
      "epoch": 1.42112,
      "grad_norm": 0.4488082528114319,
      "learning_rate": 3.4255597191752365e-05,
      "loss": 0.639,
      "step": 1111
    },
    {
      "epoch": 1.4224,
      "grad_norm": 0.4447503685951233,
      "learning_rate": 3.424467086933322e-05,
      "loss": 0.6207,
      "step": 1112
    },
    {
      "epoch": 1.42368,
      "grad_norm": 0.47931990027427673,
      "learning_rate": 3.423373614244702e-05,
      "loss": 0.6213,
      "step": 1113
    },
    {
      "epoch": 1.42496,
      "grad_norm": 0.4454458951950073,
      "learning_rate": 3.42227930185991e-05,
      "loss": 0.6374,
      "step": 1114
    },
    {
      "epoch": 1.42624,
      "grad_norm": 0.43798914551734924,
      "learning_rate": 3.42118415053006e-05,
      "loss": 0.616,
      "step": 1115
    },
    {
      "epoch": 1.42752,
      "grad_norm": 0.4445764422416687,
      "learning_rate": 3.420088161006837e-05,
      "loss": 0.6488,
      "step": 1116
    },
    {
      "epoch": 1.4288,
      "grad_norm": 0.4533645808696747,
      "learning_rate": 3.4189913340425054e-05,
      "loss": 0.586,
      "step": 1117
    },
    {
      "epoch": 1.43008,
      "grad_norm": 0.548818826675415,
      "learning_rate": 3.417893670389902e-05,
      "loss": 0.6296,
      "step": 1118
    },
    {
      "epoch": 1.43136,
      "grad_norm": 0.5249385237693787,
      "learning_rate": 3.4167951708024386e-05,
      "loss": 0.6454,
      "step": 1119
    },
    {
      "epoch": 1.4326400000000001,
      "grad_norm": 0.44476747512817383,
      "learning_rate": 3.4156958360341006e-05,
      "loss": 0.6071,
      "step": 1120
    },
    {
      "epoch": 1.43392,
      "grad_norm": 0.5426071286201477,
      "learning_rate": 3.414595666839446e-05,
      "loss": 0.652,
      "step": 1121
    },
    {
      "epoch": 1.4352,
      "grad_norm": 0.5567651987075806,
      "learning_rate": 3.413494663973606e-05,
      "loss": 0.6385,
      "step": 1122
    },
    {
      "epoch": 1.43648,
      "grad_norm": 0.5511852502822876,
      "learning_rate": 3.4123928281922865e-05,
      "loss": 0.6363,
      "step": 1123
    },
    {
      "epoch": 1.43776,
      "grad_norm": 0.4523511528968811,
      "learning_rate": 3.41129016025176e-05,
      "loss": 0.6204,
      "step": 1124
    },
    {
      "epoch": 1.4390399999999999,
      "grad_norm": 0.6237048506736755,
      "learning_rate": 3.4101866609088755e-05,
      "loss": 0.5974,
      "step": 1125
    },
    {
      "epoch": 1.44032,
      "grad_norm": 0.5577255487442017,
      "learning_rate": 3.409082330921048e-05,
      "loss": 0.6526,
      "step": 1126
    },
    {
      "epoch": 1.4416,
      "grad_norm": 0.46286851167678833,
      "learning_rate": 3.407977171046267e-05,
      "loss": 0.6376,
      "step": 1127
    },
    {
      "epoch": 1.44288,
      "grad_norm": 0.6102807521820068,
      "learning_rate": 3.4068711820430866e-05,
      "loss": 0.5857,
      "step": 1128
    },
    {
      "epoch": 1.44416,
      "grad_norm": 0.5619245767593384,
      "learning_rate": 3.405764364670635e-05,
      "loss": 0.6344,
      "step": 1129
    },
    {
      "epoch": 1.44544,
      "grad_norm": 0.5678418874740601,
      "learning_rate": 3.4046567196886076e-05,
      "loss": 0.6421,
      "step": 1130
    },
    {
      "epoch": 1.44672,
      "grad_norm": 0.5226830244064331,
      "learning_rate": 3.403548247857264e-05,
      "loss": 0.5939,
      "step": 1131
    },
    {
      "epoch": 1.448,
      "grad_norm": 0.5145910978317261,
      "learning_rate": 3.402438949937436e-05,
      "loss": 0.6389,
      "step": 1132
    },
    {
      "epoch": 1.44928,
      "grad_norm": 0.549931526184082,
      "learning_rate": 3.4013288266905214e-05,
      "loss": 0.6487,
      "step": 1133
    },
    {
      "epoch": 1.45056,
      "grad_norm": 0.5213527679443359,
      "learning_rate": 3.4002178788784834e-05,
      "loss": 0.6428,
      "step": 1134
    },
    {
      "epoch": 1.45184,
      "grad_norm": 0.4854603707790375,
      "learning_rate": 3.399106107263851e-05,
      "loss": 0.62,
      "step": 1135
    },
    {
      "epoch": 1.45312,
      "grad_norm": 0.49584925174713135,
      "learning_rate": 3.39799351260972e-05,
      "loss": 0.6026,
      "step": 1136
    },
    {
      "epoch": 1.4544000000000001,
      "grad_norm": 0.5149769186973572,
      "learning_rate": 3.396880095679749e-05,
      "loss": 0.6147,
      "step": 1137
    },
    {
      "epoch": 1.45568,
      "grad_norm": 0.5082680583000183,
      "learning_rate": 3.3957658572381654e-05,
      "loss": 0.6139,
      "step": 1138
    },
    {
      "epoch": 1.45696,
      "grad_norm": 0.5457879304885864,
      "learning_rate": 3.394650798049755e-05,
      "loss": 0.5926,
      "step": 1139
    },
    {
      "epoch": 1.45824,
      "grad_norm": 0.49039509892463684,
      "learning_rate": 3.393534918879869e-05,
      "loss": 0.6254,
      "step": 1140
    },
    {
      "epoch": 1.45824,
      "eval_loss": 0.6923708319664001,
      "eval_runtime": 19.5942,
      "eval_samples_per_second": 25.62,
      "eval_steps_per_second": 3.215,
      "step": 1140
    },
    {
      "epoch": 1.45952,
      "grad_norm": 0.5348886251449585,
      "learning_rate": 3.392418220494423e-05,
      "loss": 0.6014,
      "step": 1141
    },
    {
      "epoch": 1.4607999999999999,
      "grad_norm": 0.5121163725852966,
      "learning_rate": 3.3913007036598936e-05,
      "loss": 0.6106,
      "step": 1142
    },
    {
      "epoch": 1.46208,
      "grad_norm": 0.4890391528606415,
      "learning_rate": 3.390182369143319e-05,
      "loss": 0.6234,
      "step": 1143
    },
    {
      "epoch": 1.46336,
      "grad_norm": 0.47267991304397583,
      "learning_rate": 3.389063217712299e-05,
      "loss": 0.6272,
      "step": 1144
    },
    {
      "epoch": 1.46464,
      "grad_norm": 0.4694819152355194,
      "learning_rate": 3.3879432501349945e-05,
      "loss": 0.6105,
      "step": 1145
    },
    {
      "epoch": 1.4659200000000001,
      "grad_norm": 0.47761106491088867,
      "learning_rate": 3.386822467180124e-05,
      "loss": 0.5839,
      "step": 1146
    },
    {
      "epoch": 1.4672,
      "grad_norm": 0.5500559210777283,
      "learning_rate": 3.3857008696169705e-05,
      "loss": 0.6362,
      "step": 1147
    },
    {
      "epoch": 1.46848,
      "grad_norm": 0.49265584349632263,
      "learning_rate": 3.384578458215371e-05,
      "loss": 0.5896,
      "step": 1148
    },
    {
      "epoch": 1.46976,
      "grad_norm": 0.49351248145103455,
      "learning_rate": 3.383455233745727e-05,
      "loss": 0.6143,
      "step": 1149
    },
    {
      "epoch": 1.47104,
      "grad_norm": 0.5843918323516846,
      "learning_rate": 3.38233119697899e-05,
      "loss": 0.608,
      "step": 1150
    },
    {
      "epoch": 1.47232,
      "grad_norm": 0.47140052914619446,
      "learning_rate": 3.3812063486866766e-05,
      "loss": 0.6067,
      "step": 1151
    },
    {
      "epoch": 1.4736,
      "grad_norm": 0.5267847180366516,
      "learning_rate": 3.3800806896408564e-05,
      "loss": 0.6228,
      "step": 1152
    },
    {
      "epoch": 1.47488,
      "grad_norm": 0.526216447353363,
      "learning_rate": 3.378954220614156e-05,
      "loss": 0.6133,
      "step": 1153
    },
    {
      "epoch": 1.4761600000000001,
      "grad_norm": 0.5503297448158264,
      "learning_rate": 3.3778269423797604e-05,
      "loss": 0.6078,
      "step": 1154
    },
    {
      "epoch": 1.47744,
      "grad_norm": 0.6197400689125061,
      "learning_rate": 3.376698855711406e-05,
      "loss": 0.6092,
      "step": 1155
    },
    {
      "epoch": 1.47872,
      "grad_norm": 0.5008931159973145,
      "learning_rate": 3.375569961383387e-05,
      "loss": 0.6177,
      "step": 1156
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.5343114733695984,
      "learning_rate": 3.374440260170551e-05,
      "loss": 0.5988,
      "step": 1157
    },
    {
      "epoch": 1.48128,
      "grad_norm": 0.4977648854255676,
      "learning_rate": 3.3733097528482994e-05,
      "loss": 0.5942,
      "step": 1158
    },
    {
      "epoch": 1.4825599999999999,
      "grad_norm": 0.5336430072784424,
      "learning_rate": 3.372178440192588e-05,
      "loss": 0.6071,
      "step": 1159
    },
    {
      "epoch": 1.48384,
      "grad_norm": 0.5108422636985779,
      "learning_rate": 3.3710463229799226e-05,
      "loss": 0.6221,
      "step": 1160
    },
    {
      "epoch": 1.48512,
      "grad_norm": 0.49816861748695374,
      "learning_rate": 3.369913401987364e-05,
      "loss": 0.6061,
      "step": 1161
    },
    {
      "epoch": 1.4864,
      "grad_norm": 0.4848283529281616,
      "learning_rate": 3.368779677992524e-05,
      "loss": 0.5809,
      "step": 1162
    },
    {
      "epoch": 1.4876800000000001,
      "grad_norm": 0.4750574827194214,
      "learning_rate": 3.367645151773565e-05,
      "loss": 0.6049,
      "step": 1163
    },
    {
      "epoch": 1.48896,
      "grad_norm": 0.4983043670654297,
      "learning_rate": 3.366509824109201e-05,
      "loss": 0.6087,
      "step": 1164
    },
    {
      "epoch": 1.49024,
      "grad_norm": 0.4696747064590454,
      "learning_rate": 3.365373695778694e-05,
      "loss": 0.6444,
      "step": 1165
    },
    {
      "epoch": 1.49152,
      "grad_norm": 0.4505400061607361,
      "learning_rate": 3.3642367675618576e-05,
      "loss": 0.627,
      "step": 1166
    },
    {
      "epoch": 1.4928,
      "grad_norm": 0.5065871477127075,
      "learning_rate": 3.363099040239055e-05,
      "loss": 0.6176,
      "step": 1167
    },
    {
      "epoch": 1.49408,
      "grad_norm": 0.46014344692230225,
      "learning_rate": 3.3619605145911935e-05,
      "loss": 0.589,
      "step": 1168
    },
    {
      "epoch": 1.49536,
      "grad_norm": 0.4853391945362091,
      "learning_rate": 3.360821191399736e-05,
      "loss": 0.6404,
      "step": 1169
    },
    {
      "epoch": 1.49664,
      "grad_norm": 0.4962824583053589,
      "learning_rate": 3.359681071446685e-05,
      "loss": 0.6808,
      "step": 1170
    },
    {
      "epoch": 1.49664,
      "eval_loss": 0.6906577348709106,
      "eval_runtime": 19.5875,
      "eval_samples_per_second": 25.629,
      "eval_steps_per_second": 3.216,
      "step": 1170
    },
    {
      "epoch": 1.49792,
      "grad_norm": 0.5387783050537109,
      "learning_rate": 3.358540155514594e-05,
      "loss": 0.6146,
      "step": 1171
    },
    {
      "epoch": 1.4992,
      "grad_norm": 0.4829972982406616,
      "learning_rate": 3.357398444386562e-05,
      "loss": 0.6077,
      "step": 1172
    },
    {
      "epoch": 1.50048,
      "grad_norm": 0.558963418006897,
      "learning_rate": 3.356255938846234e-05,
      "loss": 0.6302,
      "step": 1173
    },
    {
      "epoch": 1.50176,
      "grad_norm": 0.5465085506439209,
      "learning_rate": 3.3551126396778004e-05,
      "loss": 0.6664,
      "step": 1174
    },
    {
      "epoch": 1.50304,
      "grad_norm": 0.5838433504104614,
      "learning_rate": 3.3539685476659964e-05,
      "loss": 0.6233,
      "step": 1175
    },
    {
      "epoch": 1.5043199999999999,
      "grad_norm": 0.5447558760643005,
      "learning_rate": 3.3528236635961004e-05,
      "loss": 0.6586,
      "step": 1176
    },
    {
      "epoch": 1.5056,
      "grad_norm": 0.487019419670105,
      "learning_rate": 3.3516779882539355e-05,
      "loss": 0.615,
      "step": 1177
    },
    {
      "epoch": 1.50688,
      "grad_norm": 0.48308447003364563,
      "learning_rate": 3.3505315224258675e-05,
      "loss": 0.6169,
      "step": 1178
    },
    {
      "epoch": 1.50816,
      "grad_norm": 0.47541332244873047,
      "learning_rate": 3.349384266898804e-05,
      "loss": 0.6105,
      "step": 1179
    },
    {
      "epoch": 1.5094400000000001,
      "grad_norm": 0.505120575428009,
      "learning_rate": 3.348236222460196e-05,
      "loss": 0.6304,
      "step": 1180
    },
    {
      "epoch": 1.51072,
      "grad_norm": 0.5058416128158569,
      "learning_rate": 3.347087389898036e-05,
      "loss": 0.6039,
      "step": 1181
    },
    {
      "epoch": 1.512,
      "grad_norm": 0.5037918090820312,
      "learning_rate": 3.3459377700008564e-05,
      "loss": 0.608,
      "step": 1182
    },
    {
      "epoch": 1.51328,
      "grad_norm": 0.4896526038646698,
      "learning_rate": 3.34478736355773e-05,
      "loss": 0.5955,
      "step": 1183
    },
    {
      "epoch": 1.51456,
      "grad_norm": 0.5761212110519409,
      "learning_rate": 3.343636171358271e-05,
      "loss": 0.6595,
      "step": 1184
    },
    {
      "epoch": 1.5158399999999999,
      "grad_norm": 0.5094326734542847,
      "learning_rate": 3.3424841941926304e-05,
      "loss": 0.6279,
      "step": 1185
    },
    {
      "epoch": 1.51712,
      "grad_norm": 0.5421015620231628,
      "learning_rate": 3.3413314328515e-05,
      "loss": 0.6391,
      "step": 1186
    },
    {
      "epoch": 1.5184,
      "grad_norm": 0.5173119902610779,
      "learning_rate": 3.3401778881261104e-05,
      "loss": 0.5696,
      "step": 1187
    },
    {
      "epoch": 1.5196800000000001,
      "grad_norm": 0.5335983037948608,
      "learning_rate": 3.3390235608082276e-05,
      "loss": 0.6225,
      "step": 1188
    },
    {
      "epoch": 1.52096,
      "grad_norm": 0.5069127082824707,
      "learning_rate": 3.337868451690157e-05,
      "loss": 0.6088,
      "step": 1189
    },
    {
      "epoch": 1.52224,
      "grad_norm": 0.5019217133522034,
      "learning_rate": 3.336712561564738e-05,
      "loss": 0.6049,
      "step": 1190
    },
    {
      "epoch": 1.52352,
      "grad_norm": 0.4802045226097107,
      "learning_rate": 3.335555891225349e-05,
      "loss": 0.6145,
      "step": 1191
    },
    {
      "epoch": 1.5248,
      "grad_norm": 0.5522844791412354,
      "learning_rate": 3.334398441465901e-05,
      "loss": 0.5897,
      "step": 1192
    },
    {
      "epoch": 1.5260799999999999,
      "grad_norm": 0.4938916265964508,
      "learning_rate": 3.3332402130808435e-05,
      "loss": 0.6462,
      "step": 1193
    },
    {
      "epoch": 1.52736,
      "grad_norm": 0.46159783005714417,
      "learning_rate": 3.332081206865158e-05,
      "loss": 0.6365,
      "step": 1194
    },
    {
      "epoch": 1.52864,
      "grad_norm": 0.5437260866165161,
      "learning_rate": 3.3309214236143585e-05,
      "loss": 0.6372,
      "step": 1195
    },
    {
      "epoch": 1.52992,
      "grad_norm": 0.43438720703125,
      "learning_rate": 3.329760864124496e-05,
      "loss": 0.6019,
      "step": 1196
    },
    {
      "epoch": 1.5312000000000001,
      "grad_norm": 0.5130593180656433,
      "learning_rate": 3.3285995291921515e-05,
      "loss": 0.593,
      "step": 1197
    },
    {
      "epoch": 1.53248,
      "grad_norm": 0.48909106850624084,
      "learning_rate": 3.327437419614441e-05,
      "loss": 0.594,
      "step": 1198
    },
    {
      "epoch": 1.53376,
      "grad_norm": 0.4905202090740204,
      "learning_rate": 3.326274536189008e-05,
      "loss": 0.6212,
      "step": 1199
    },
    {
      "epoch": 1.53504,
      "grad_norm": 0.5448431968688965,
      "learning_rate": 3.3251108797140314e-05,
      "loss": 0.6216,
      "step": 1200
    },
    {
      "epoch": 1.53504,
      "eval_loss": 0.6900917887687683,
      "eval_runtime": 19.634,
      "eval_samples_per_second": 25.568,
      "eval_steps_per_second": 3.209,
      "step": 1200
    },
    {
      "epoch": 1.53632,
      "grad_norm": 0.5616306066513062,
      "learning_rate": 3.323946450988217e-05,
      "loss": 0.6247,
      "step": 1201
    },
    {
      "epoch": 1.5375999999999999,
      "grad_norm": 0.5488273501396179,
      "learning_rate": 3.322781250810805e-05,
      "loss": 0.6022,
      "step": 1202
    },
    {
      "epoch": 1.53888,
      "grad_norm": 0.5273008942604065,
      "learning_rate": 3.32161527998156e-05,
      "loss": 0.6567,
      "step": 1203
    },
    {
      "epoch": 1.54016,
      "grad_norm": 0.5553897023200989,
      "learning_rate": 3.32044853930078e-05,
      "loss": 0.6348,
      "step": 1204
    },
    {
      "epoch": 1.5414400000000001,
      "grad_norm": 0.47348931431770325,
      "learning_rate": 3.3192810295692885e-05,
      "loss": 0.6045,
      "step": 1205
    },
    {
      "epoch": 1.54272,
      "grad_norm": 0.5469029545783997,
      "learning_rate": 3.3181127515884394e-05,
      "loss": 0.6116,
      "step": 1206
    },
    {
      "epoch": 1.544,
      "grad_norm": 0.5251120924949646,
      "learning_rate": 3.316943706160111e-05,
      "loss": 0.6412,
      "step": 1207
    },
    {
      "epoch": 1.54528,
      "grad_norm": 0.4996357560157776,
      "learning_rate": 3.3157738940867115e-05,
      "loss": 0.614,
      "step": 1208
    },
    {
      "epoch": 1.54656,
      "grad_norm": 0.4736141860485077,
      "learning_rate": 3.314603316171171e-05,
      "loss": 0.605,
      "step": 1209
    },
    {
      "epoch": 1.5478399999999999,
      "grad_norm": 0.5199579000473022,
      "learning_rate": 3.3134319732169504e-05,
      "loss": 0.6493,
      "step": 1210
    },
    {
      "epoch": 1.54912,
      "grad_norm": 0.4983501434326172,
      "learning_rate": 3.312259866028032e-05,
      "loss": 0.6164,
      "step": 1211
    },
    {
      "epoch": 1.5504,
      "grad_norm": 0.46590107679367065,
      "learning_rate": 3.3110869954089235e-05,
      "loss": 0.6162,
      "step": 1212
    },
    {
      "epoch": 1.55168,
      "grad_norm": 0.5510172843933105,
      "learning_rate": 3.3099133621646574e-05,
      "loss": 0.6263,
      "step": 1213
    },
    {
      "epoch": 1.5529600000000001,
      "grad_norm": 0.48461389541625977,
      "learning_rate": 3.30873896710079e-05,
      "loss": 0.6072,
      "step": 1214
    },
    {
      "epoch": 1.55424,
      "grad_norm": 0.6475604176521301,
      "learning_rate": 3.307563811023399e-05,
      "loss": 0.6462,
      "step": 1215
    },
    {
      "epoch": 1.55552,
      "grad_norm": 0.5574936866760254,
      "learning_rate": 3.306387894739085e-05,
      "loss": 0.6071,
      "step": 1216
    },
    {
      "epoch": 1.5568,
      "grad_norm": 0.5765414834022522,
      "learning_rate": 3.3052112190549705e-05,
      "loss": 0.6354,
      "step": 1217
    },
    {
      "epoch": 1.55808,
      "grad_norm": 0.5568042397499084,
      "learning_rate": 3.3040337847786994e-05,
      "loss": 0.6045,
      "step": 1218
    },
    {
      "epoch": 1.5593599999999999,
      "grad_norm": 0.5488018989562988,
      "learning_rate": 3.302855592718436e-05,
      "loss": 0.5961,
      "step": 1219
    },
    {
      "epoch": 1.56064,
      "grad_norm": 0.5399571657180786,
      "learning_rate": 3.301676643682866e-05,
      "loss": 0.6424,
      "step": 1220
    },
    {
      "epoch": 1.56192,
      "grad_norm": 0.5144947171211243,
      "learning_rate": 3.3004969384811915e-05,
      "loss": 0.6263,
      "step": 1221
    },
    {
      "epoch": 1.5632000000000001,
      "grad_norm": 0.49894604086875916,
      "learning_rate": 3.299316477923137e-05,
      "loss": 0.6038,
      "step": 1222
    },
    {
      "epoch": 1.56448,
      "grad_norm": 0.5087131261825562,
      "learning_rate": 3.298135262818944e-05,
      "loss": 0.5994,
      "step": 1223
    },
    {
      "epoch": 1.56576,
      "grad_norm": 0.5045850276947021,
      "learning_rate": 3.296953293979372e-05,
      "loss": 0.6134,
      "step": 1224
    },
    {
      "epoch": 1.56704,
      "grad_norm": 0.47476693987846375,
      "learning_rate": 3.295770572215697e-05,
      "loss": 0.6549,
      "step": 1225
    },
    {
      "epoch": 1.56832,
      "grad_norm": 0.5600761771202087,
      "learning_rate": 3.2945870983397144e-05,
      "loss": 0.6455,
      "step": 1226
    },
    {
      "epoch": 1.5695999999999999,
      "grad_norm": 0.5433741211891174,
      "learning_rate": 3.293402873163732e-05,
      "loss": 0.6318,
      "step": 1227
    },
    {
      "epoch": 1.57088,
      "grad_norm": 0.540194571018219,
      "learning_rate": 3.292217897500577e-05,
      "loss": 0.6426,
      "step": 1228
    },
    {
      "epoch": 1.57216,
      "grad_norm": 0.4946892559528351,
      "learning_rate": 3.29103217216359e-05,
      "loss": 0.6307,
      "step": 1229
    },
    {
      "epoch": 1.57344,
      "grad_norm": 0.5083387494087219,
      "learning_rate": 3.289845697966625e-05,
      "loss": 0.6093,
      "step": 1230
    },
    {
      "epoch": 1.57344,
      "eval_loss": 0.6880496144294739,
      "eval_runtime": 19.5897,
      "eval_samples_per_second": 25.626,
      "eval_steps_per_second": 3.216,
      "step": 1230
    },
    {
      "epoch": 1.5747200000000001,
      "grad_norm": 0.4483705759048462,
      "learning_rate": 3.288658475724052e-05,
      "loss": 0.6035,
      "step": 1231
    },
    {
      "epoch": 1.576,
      "grad_norm": 0.5203142762184143,
      "learning_rate": 3.2874705062507534e-05,
      "loss": 0.6565,
      "step": 1232
    },
    {
      "epoch": 1.57728,
      "grad_norm": 0.5592670440673828,
      "learning_rate": 3.286281790362125e-05,
      "loss": 0.6188,
      "step": 1233
    },
    {
      "epoch": 1.57856,
      "grad_norm": 0.5178285241127014,
      "learning_rate": 3.285092328874076e-05,
      "loss": 0.6499,
      "step": 1234
    },
    {
      "epoch": 1.57984,
      "grad_norm": 0.540086567401886,
      "learning_rate": 3.283902122603023e-05,
      "loss": 0.6447,
      "step": 1235
    },
    {
      "epoch": 1.5811199999999999,
      "grad_norm": 0.5154386162757874,
      "learning_rate": 3.2827111723659e-05,
      "loss": 0.6008,
      "step": 1236
    },
    {
      "epoch": 1.5824,
      "grad_norm": 0.4988633394241333,
      "learning_rate": 3.2815194789801466e-05,
      "loss": 0.6309,
      "step": 1237
    },
    {
      "epoch": 1.58368,
      "grad_norm": 0.5124436616897583,
      "learning_rate": 3.2803270432637144e-05,
      "loss": 0.6237,
      "step": 1238
    },
    {
      "epoch": 1.5849600000000001,
      "grad_norm": 0.4727044403553009,
      "learning_rate": 3.279133866035065e-05,
      "loss": 0.616,
      "step": 1239
    },
    {
      "epoch": 1.58624,
      "grad_norm": 0.4578966796398163,
      "learning_rate": 3.277939948113169e-05,
      "loss": 0.5946,
      "step": 1240
    },
    {
      "epoch": 1.58752,
      "grad_norm": 0.5531896352767944,
      "learning_rate": 3.276745290317504e-05,
      "loss": 0.6217,
      "step": 1241
    },
    {
      "epoch": 1.5888,
      "grad_norm": 0.474204957485199,
      "learning_rate": 3.275549893468057e-05,
      "loss": 0.5823,
      "step": 1242
    },
    {
      "epoch": 1.59008,
      "grad_norm": 0.45534083247184753,
      "learning_rate": 3.27435375838532e-05,
      "loss": 0.596,
      "step": 1243
    },
    {
      "epoch": 1.5913599999999999,
      "grad_norm": 0.5464202761650085,
      "learning_rate": 3.273156885890295e-05,
      "loss": 0.6526,
      "step": 1244
    },
    {
      "epoch": 1.5926399999999998,
      "grad_norm": 0.5128631591796875,
      "learning_rate": 3.271959276804488e-05,
      "loss": 0.6221,
      "step": 1245
    },
    {
      "epoch": 1.59392,
      "grad_norm": 0.49729597568511963,
      "learning_rate": 3.27076093194991e-05,
      "loss": 0.6068,
      "step": 1246
    },
    {
      "epoch": 1.5952,
      "grad_norm": 0.5025321245193481,
      "learning_rate": 3.269561852149079e-05,
      "loss": 0.601,
      "step": 1247
    },
    {
      "epoch": 1.5964800000000001,
      "grad_norm": 0.4665898382663727,
      "learning_rate": 3.268362038225017e-05,
      "loss": 0.6117,
      "step": 1248
    },
    {
      "epoch": 1.59776,
      "grad_norm": 0.46728044748306274,
      "learning_rate": 3.2671614910012475e-05,
      "loss": 0.6204,
      "step": 1249
    },
    {
      "epoch": 1.59904,
      "grad_norm": 0.5080220103263855,
      "learning_rate": 3.2659602113018e-05,
      "loss": 0.6153,
      "step": 1250
    },
    {
      "epoch": 1.60032,
      "grad_norm": 0.5096635818481445,
      "learning_rate": 3.264758199951206e-05,
      "loss": 0.5875,
      "step": 1251
    },
    {
      "epoch": 1.6016,
      "grad_norm": 0.46891364455223083,
      "learning_rate": 3.2635554577744996e-05,
      "loss": 0.5924,
      "step": 1252
    },
    {
      "epoch": 1.6028799999999999,
      "grad_norm": 0.5028663873672485,
      "learning_rate": 3.2623519855972156e-05,
      "loss": 0.6162,
      "step": 1253
    },
    {
      "epoch": 1.60416,
      "grad_norm": 0.4675734043121338,
      "learning_rate": 3.26114778424539e-05,
      "loss": 0.5979,
      "step": 1254
    },
    {
      "epoch": 1.60544,
      "grad_norm": 0.49414971470832825,
      "learning_rate": 3.259942854545559e-05,
      "loss": 0.6501,
      "step": 1255
    },
    {
      "epoch": 1.6067200000000001,
      "grad_norm": 0.5172702670097351,
      "learning_rate": 3.258737197324761e-05,
      "loss": 0.6231,
      "step": 1256
    },
    {
      "epoch": 1.608,
      "grad_norm": 0.5125314593315125,
      "learning_rate": 3.2575308134105314e-05,
      "loss": 0.6196,
      "step": 1257
    },
    {
      "epoch": 1.60928,
      "grad_norm": 0.5494481325149536,
      "learning_rate": 3.256323703630904e-05,
      "loss": 0.6484,
      "step": 1258
    },
    {
      "epoch": 1.61056,
      "grad_norm": 0.5024045705795288,
      "learning_rate": 3.255115868814413e-05,
      "loss": 0.5804,
      "step": 1259
    },
    {
      "epoch": 1.61184,
      "grad_norm": 0.5750858783721924,
      "learning_rate": 3.253907309790088e-05,
      "loss": 0.6417,
      "step": 1260
    },
    {
      "epoch": 1.61184,
      "eval_loss": 0.6883153319358826,
      "eval_runtime": 19.5956,
      "eval_samples_per_second": 25.618,
      "eval_steps_per_second": 3.215,
      "step": 1260
    },
    {
      "epoch": 1.6131199999999999,
      "grad_norm": 0.5532737970352173,
      "learning_rate": 3.252698027387458e-05,
      "loss": 0.6213,
      "step": 1261
    },
    {
      "epoch": 1.6143999999999998,
      "grad_norm": 0.5151821970939636,
      "learning_rate": 3.2514880224365456e-05,
      "loss": 0.6058,
      "step": 1262
    },
    {
      "epoch": 1.61568,
      "grad_norm": 0.5075077414512634,
      "learning_rate": 3.250277295767873e-05,
      "loss": 0.6165,
      "step": 1263
    },
    {
      "epoch": 1.61696,
      "grad_norm": 0.5011141300201416,
      "learning_rate": 3.249065848212455e-05,
      "loss": 0.5764,
      "step": 1264
    },
    {
      "epoch": 1.6182400000000001,
      "grad_norm": 0.5065586566925049,
      "learning_rate": 3.247853680601802e-05,
      "loss": 0.642,
      "step": 1265
    },
    {
      "epoch": 1.61952,
      "grad_norm": 0.5200608372688293,
      "learning_rate": 3.246640793767918e-05,
      "loss": 0.6268,
      "step": 1266
    },
    {
      "epoch": 1.6208,
      "grad_norm": 0.5661983489990234,
      "learning_rate": 3.245427188543302e-05,
      "loss": 0.635,
      "step": 1267
    },
    {
      "epoch": 1.62208,
      "grad_norm": 0.48517996072769165,
      "learning_rate": 3.2442128657609454e-05,
      "loss": 0.6153,
      "step": 1268
    },
    {
      "epoch": 1.62336,
      "grad_norm": 0.5886812806129456,
      "learning_rate": 3.2429978262543316e-05,
      "loss": 0.6263,
      "step": 1269
    },
    {
      "epoch": 1.6246399999999999,
      "grad_norm": 0.5839177966117859,
      "learning_rate": 3.241782070857436e-05,
      "loss": 0.6249,
      "step": 1270
    },
    {
      "epoch": 1.62592,
      "grad_norm": 0.49129006266593933,
      "learning_rate": 3.240565600404727e-05,
      "loss": 0.6069,
      "step": 1271
    },
    {
      "epoch": 1.6272,
      "grad_norm": 0.5378501415252686,
      "learning_rate": 3.239348415731163e-05,
      "loss": 0.6341,
      "step": 1272
    },
    {
      "epoch": 1.6284800000000001,
      "grad_norm": 0.5638146996498108,
      "learning_rate": 3.23813051767219e-05,
      "loss": 0.6163,
      "step": 1273
    },
    {
      "epoch": 1.62976,
      "grad_norm": 0.5095300674438477,
      "learning_rate": 3.236911907063747e-05,
      "loss": 0.6237,
      "step": 1274
    },
    {
      "epoch": 1.63104,
      "grad_norm": 0.5042887330055237,
      "learning_rate": 3.235692584742262e-05,
      "loss": 0.6303,
      "step": 1275
    },
    {
      "epoch": 1.63232,
      "grad_norm": 0.5169423818588257,
      "learning_rate": 3.2344725515446495e-05,
      "loss": 0.6173,
      "step": 1276
    },
    {
      "epoch": 1.6336,
      "grad_norm": 0.5292079448699951,
      "learning_rate": 3.233251808308312e-05,
      "loss": 0.6338,
      "step": 1277
    },
    {
      "epoch": 1.6348799999999999,
      "grad_norm": 0.4532772898674011,
      "learning_rate": 3.232030355871143e-05,
      "loss": 0.5816,
      "step": 1278
    },
    {
      "epoch": 1.6361599999999998,
      "grad_norm": 0.4767000079154968,
      "learning_rate": 3.2308081950715165e-05,
      "loss": 0.6336,
      "step": 1279
    },
    {
      "epoch": 1.63744,
      "grad_norm": 0.4941141605377197,
      "learning_rate": 3.229585326748298e-05,
      "loss": 0.6286,
      "step": 1280
    },
    {
      "epoch": 1.63872,
      "grad_norm": 0.4403752088546753,
      "learning_rate": 3.228361751740837e-05,
      "loss": 0.5824,
      "step": 1281
    },
    {
      "epoch": 1.6400000000000001,
      "grad_norm": 0.4952940344810486,
      "learning_rate": 3.2271374708889686e-05,
      "loss": 0.6332,
      "step": 1282
    },
    {
      "epoch": 1.64128,
      "grad_norm": 0.4949045479297638,
      "learning_rate": 3.22591248503301e-05,
      "loss": 0.6385,
      "step": 1283
    },
    {
      "epoch": 1.64256,
      "grad_norm": 0.4811001718044281,
      "learning_rate": 3.224686795013765e-05,
      "loss": 0.6048,
      "step": 1284
    },
    {
      "epoch": 1.64384,
      "grad_norm": 0.5035277009010315,
      "learning_rate": 3.223460401672518e-05,
      "loss": 0.625,
      "step": 1285
    },
    {
      "epoch": 1.64512,
      "grad_norm": 0.48320814967155457,
      "learning_rate": 3.222233305851041e-05,
      "loss": 0.603,
      "step": 1286
    },
    {
      "epoch": 1.6463999999999999,
      "grad_norm": 0.5264683961868286,
      "learning_rate": 3.221005508391583e-05,
      "loss": 0.6515,
      "step": 1287
    },
    {
      "epoch": 1.64768,
      "grad_norm": 0.45450764894485474,
      "learning_rate": 3.219777010136876e-05,
      "loss": 0.6079,
      "step": 1288
    },
    {
      "epoch": 1.64896,
      "grad_norm": 0.4834478199481964,
      "learning_rate": 3.218547811930136e-05,
      "loss": 0.6079,
      "step": 1289
    },
    {
      "epoch": 1.6502400000000002,
      "grad_norm": 0.5050206184387207,
      "learning_rate": 3.2173179146150536e-05,
      "loss": 0.5805,
      "step": 1290
    },
    {
      "epoch": 1.6502400000000002,
      "eval_loss": 0.6879907250404358,
      "eval_runtime": 19.5893,
      "eval_samples_per_second": 25.626,
      "eval_steps_per_second": 3.216,
      "step": 1290
    },
    {
      "epoch": 1.65152,
      "grad_norm": 0.4392125904560089,
      "learning_rate": 3.2160873190358056e-05,
      "loss": 0.583,
      "step": 1291
    },
    {
      "epoch": 1.6528,
      "grad_norm": 0.514826774597168,
      "learning_rate": 3.214856026037044e-05,
      "loss": 0.6349,
      "step": 1292
    },
    {
      "epoch": 1.65408,
      "grad_norm": 0.48852506279945374,
      "learning_rate": 3.213624036463901e-05,
      "loss": 0.6209,
      "step": 1293
    },
    {
      "epoch": 1.65536,
      "grad_norm": 0.4920100271701813,
      "learning_rate": 3.212391351161987e-05,
      "loss": 0.646,
      "step": 1294
    },
    {
      "epoch": 1.65664,
      "grad_norm": 0.46628084778785706,
      "learning_rate": 3.211157970977388e-05,
      "loss": 0.5843,
      "step": 1295
    },
    {
      "epoch": 1.6579199999999998,
      "grad_norm": 0.48271092772483826,
      "learning_rate": 3.2099238967566694e-05,
      "loss": 0.5936,
      "step": 1296
    },
    {
      "epoch": 1.6592,
      "grad_norm": 0.4579281508922577,
      "learning_rate": 3.208689129346872e-05,
      "loss": 0.6226,
      "step": 1297
    },
    {
      "epoch": 1.66048,
      "grad_norm": 0.4769735038280487,
      "learning_rate": 3.207453669595513e-05,
      "loss": 0.6189,
      "step": 1298
    },
    {
      "epoch": 1.6617600000000001,
      "grad_norm": 0.5103779435157776,
      "learning_rate": 3.206217518350584e-05,
      "loss": 0.627,
      "step": 1299
    },
    {
      "epoch": 1.66304,
      "grad_norm": 0.4973532259464264,
      "learning_rate": 3.20498067646055e-05,
      "loss": 0.6337,
      "step": 1300
    },
    {
      "epoch": 1.66432,
      "grad_norm": 0.4906511604785919,
      "learning_rate": 3.2037431447743524e-05,
      "loss": 0.6337,
      "step": 1301
    },
    {
      "epoch": 1.6656,
      "grad_norm": 0.47455281019210815,
      "learning_rate": 3.202504924141406e-05,
      "loss": 0.5967,
      "step": 1302
    },
    {
      "epoch": 1.66688,
      "grad_norm": 0.4619572162628174,
      "learning_rate": 3.2012660154115974e-05,
      "loss": 0.6355,
      "step": 1303
    },
    {
      "epoch": 1.6681599999999999,
      "grad_norm": 0.4846312999725342,
      "learning_rate": 3.200026419435284e-05,
      "loss": 0.6115,
      "step": 1304
    },
    {
      "epoch": 1.66944,
      "grad_norm": 0.4814051687717438,
      "learning_rate": 3.198786137063299e-05,
      "loss": 0.6128,
      "step": 1305
    },
    {
      "epoch": 1.67072,
      "grad_norm": 0.5108335614204407,
      "learning_rate": 3.197545169146943e-05,
      "loss": 0.6333,
      "step": 1306
    },
    {
      "epoch": 1.6720000000000002,
      "grad_norm": 0.49154427647590637,
      "learning_rate": 3.1963035165379884e-05,
      "loss": 0.5901,
      "step": 1307
    },
    {
      "epoch": 1.67328,
      "grad_norm": 0.5177683234214783,
      "learning_rate": 3.195061180088679e-05,
      "loss": 0.6174,
      "step": 1308
    },
    {
      "epoch": 1.67456,
      "grad_norm": 0.4472145736217499,
      "learning_rate": 3.1938181606517244e-05,
      "loss": 0.6261,
      "step": 1309
    },
    {
      "epoch": 1.67584,
      "grad_norm": 0.48267704248428345,
      "learning_rate": 3.1925744590803074e-05,
      "loss": 0.5934,
      "step": 1310
    },
    {
      "epoch": 1.67712,
      "grad_norm": 0.4444316029548645,
      "learning_rate": 3.191330076228076e-05,
      "loss": 0.5868,
      "step": 1311
    },
    {
      "epoch": 1.6784,
      "grad_norm": 0.49386894702911377,
      "learning_rate": 3.190085012949146e-05,
      "loss": 0.6276,
      "step": 1312
    },
    {
      "epoch": 1.6796799999999998,
      "grad_norm": 0.5394115447998047,
      "learning_rate": 3.188839270098101e-05,
      "loss": 0.5742,
      "step": 1313
    },
    {
      "epoch": 1.68096,
      "grad_norm": 0.5885350704193115,
      "learning_rate": 3.1875928485299925e-05,
      "loss": 0.6405,
      "step": 1314
    },
    {
      "epoch": 1.68224,
      "grad_norm": 0.4990473985671997,
      "learning_rate": 3.186345749100335e-05,
      "loss": 0.6047,
      "step": 1315
    },
    {
      "epoch": 1.6835200000000001,
      "grad_norm": 0.49828919768333435,
      "learning_rate": 3.185097972665109e-05,
      "loss": 0.575,
      "step": 1316
    },
    {
      "epoch": 1.6848,
      "grad_norm": 0.5319384932518005,
      "learning_rate": 3.1838495200807615e-05,
      "loss": 0.6276,
      "step": 1317
    },
    {
      "epoch": 1.68608,
      "grad_norm": 0.49893200397491455,
      "learning_rate": 3.1826003922042016e-05,
      "loss": 0.6371,
      "step": 1318
    },
    {
      "epoch": 1.68736,
      "grad_norm": 0.47414883971214294,
      "learning_rate": 3.1813505898928017e-05,
      "loss": 0.6431,
      "step": 1319
    },
    {
      "epoch": 1.68864,
      "grad_norm": 0.5123562812805176,
      "learning_rate": 3.1801001140044e-05,
      "loss": 0.6454,
      "step": 1320
    },
    {
      "epoch": 1.68864,
      "eval_loss": 0.6868248581886292,
      "eval_runtime": 19.5941,
      "eval_samples_per_second": 25.62,
      "eval_steps_per_second": 3.215,
      "step": 1320
    },
    {
      "epoch": 1.6899199999999999,
      "grad_norm": 0.4761483669281006,
      "learning_rate": 3.178848965397294e-05,
      "loss": 0.6072,
      "step": 1321
    },
    {
      "epoch": 1.6912,
      "grad_norm": 0.46802818775177,
      "learning_rate": 3.177597144930244e-05,
      "loss": 0.6195,
      "step": 1322
    },
    {
      "epoch": 1.69248,
      "grad_norm": 0.45173075795173645,
      "learning_rate": 3.176344653462472e-05,
      "loss": 0.582,
      "step": 1323
    },
    {
      "epoch": 1.6937600000000002,
      "grad_norm": 0.48264431953430176,
      "learning_rate": 3.175091491853659e-05,
      "loss": 0.6133,
      "step": 1324
    },
    {
      "epoch": 1.69504,
      "grad_norm": 0.4820813834667206,
      "learning_rate": 3.173837660963948e-05,
      "loss": 0.6084,
      "step": 1325
    },
    {
      "epoch": 1.69632,
      "grad_norm": 0.4754944443702698,
      "learning_rate": 3.172583161653939e-05,
      "loss": 0.6095,
      "step": 1326
    },
    {
      "epoch": 1.6976,
      "grad_norm": 0.4983585476875305,
      "learning_rate": 3.171327994784694e-05,
      "loss": 0.6281,
      "step": 1327
    },
    {
      "epoch": 1.69888,
      "grad_norm": 0.460040420293808,
      "learning_rate": 3.17007216121773e-05,
      "loss": 0.5892,
      "step": 1328
    },
    {
      "epoch": 1.70016,
      "grad_norm": 0.4526059627532959,
      "learning_rate": 3.168815661815024e-05,
      "loss": 0.577,
      "step": 1329
    },
    {
      "epoch": 1.7014399999999998,
      "grad_norm": 0.48954951763153076,
      "learning_rate": 3.167558497439007e-05,
      "loss": 0.6082,
      "step": 1330
    },
    {
      "epoch": 1.70272,
      "grad_norm": 0.49873214960098267,
      "learning_rate": 3.166300668952571e-05,
      "loss": 0.6366,
      "step": 1331
    },
    {
      "epoch": 1.704,
      "grad_norm": 0.4797149896621704,
      "learning_rate": 3.165042177219059e-05,
      "loss": 0.609,
      "step": 1332
    },
    {
      "epoch": 1.7052800000000001,
      "grad_norm": 0.4961213171482086,
      "learning_rate": 3.1637830231022727e-05,
      "loss": 0.5831,
      "step": 1333
    },
    {
      "epoch": 1.70656,
      "grad_norm": 0.43097934126853943,
      "learning_rate": 3.162523207466468e-05,
      "loss": 0.6042,
      "step": 1334
    },
    {
      "epoch": 1.70784,
      "grad_norm": 0.4459404945373535,
      "learning_rate": 3.161262731176351e-05,
      "loss": 0.6332,
      "step": 1335
    },
    {
      "epoch": 1.70912,
      "grad_norm": 0.5193769335746765,
      "learning_rate": 3.160001595097087e-05,
      "loss": 0.6488,
      "step": 1336
    },
    {
      "epoch": 1.7104,
      "grad_norm": 0.43740200996398926,
      "learning_rate": 3.1587398000942907e-05,
      "loss": 0.6118,
      "step": 1337
    },
    {
      "epoch": 1.7116799999999999,
      "grad_norm": 0.5154653191566467,
      "learning_rate": 3.15747734703403e-05,
      "loss": 0.6237,
      "step": 1338
    },
    {
      "epoch": 1.71296,
      "grad_norm": 0.5538183450698853,
      "learning_rate": 3.156214236782825e-05,
      "loss": 0.6354,
      "step": 1339
    },
    {
      "epoch": 1.71424,
      "grad_norm": 0.4525747001171112,
      "learning_rate": 3.154950470207644e-05,
      "loss": 0.6057,
      "step": 1340
    },
    {
      "epoch": 1.7155200000000002,
      "grad_norm": 0.456769734621048,
      "learning_rate": 3.1536860481759106e-05,
      "loss": 0.625,
      "step": 1341
    },
    {
      "epoch": 1.7168,
      "grad_norm": 0.48772838711738586,
      "learning_rate": 3.1524209715554945e-05,
      "loss": 0.6231,
      "step": 1342
    },
    {
      "epoch": 1.71808,
      "grad_norm": 0.46725308895111084,
      "learning_rate": 3.151155241214715e-05,
      "loss": 0.6155,
      "step": 1343
    },
    {
      "epoch": 1.71936,
      "grad_norm": 0.5146908760070801,
      "learning_rate": 3.149888858022343e-05,
      "loss": 0.6234,
      "step": 1344
    },
    {
      "epoch": 1.72064,
      "grad_norm": 0.5232139229774475,
      "learning_rate": 3.1486218228475936e-05,
      "loss": 0.6011,
      "step": 1345
    },
    {
      "epoch": 1.72192,
      "grad_norm": 0.44400662183761597,
      "learning_rate": 3.1473541365601316e-05,
      "loss": 0.6031,
      "step": 1346
    },
    {
      "epoch": 1.7231999999999998,
      "grad_norm": 0.4403976500034332,
      "learning_rate": 3.14608580003007e-05,
      "loss": 0.5998,
      "step": 1347
    },
    {
      "epoch": 1.72448,
      "grad_norm": 0.4710434675216675,
      "learning_rate": 3.144816814127964e-05,
      "loss": 0.6126,
      "step": 1348
    },
    {
      "epoch": 1.72576,
      "grad_norm": 0.4653922915458679,
      "learning_rate": 3.143547179724819e-05,
      "loss": 0.6176,
      "step": 1349
    },
    {
      "epoch": 1.7270400000000001,
      "grad_norm": 0.46456798911094666,
      "learning_rate": 3.142276897692083e-05,
      "loss": 0.6014,
      "step": 1350
    },
    {
      "epoch": 1.7270400000000001,
      "eval_loss": 0.6859357953071594,
      "eval_runtime": 19.5876,
      "eval_samples_per_second": 25.628,
      "eval_steps_per_second": 3.216,
      "step": 1350
    },
    {
      "epoch": 1.72832,
      "grad_norm": 0.5146026015281677,
      "learning_rate": 3.1410059689016485e-05,
      "loss": 0.602,
      "step": 1351
    },
    {
      "epoch": 1.7296,
      "grad_norm": 0.4607435464859009,
      "learning_rate": 3.1397343942258535e-05,
      "loss": 0.6238,
      "step": 1352
    },
    {
      "epoch": 1.73088,
      "grad_norm": 0.5064708590507507,
      "learning_rate": 3.138462174537477e-05,
      "loss": 0.6001,
      "step": 1353
    },
    {
      "epoch": 1.73216,
      "grad_norm": 0.4971449375152588,
      "learning_rate": 3.1371893107097434e-05,
      "loss": 0.6276,
      "step": 1354
    },
    {
      "epoch": 1.7334399999999999,
      "grad_norm": 0.4626976549625397,
      "learning_rate": 3.1359158036163165e-05,
      "loss": 0.6115,
      "step": 1355
    },
    {
      "epoch": 1.73472,
      "grad_norm": 0.4901287853717804,
      "learning_rate": 3.134641654131304e-05,
      "loss": 0.6109,
      "step": 1356
    },
    {
      "epoch": 1.736,
      "grad_norm": 0.48633086681365967,
      "learning_rate": 3.133366863129253e-05,
      "loss": 0.6197,
      "step": 1357
    },
    {
      "epoch": 1.7372800000000002,
      "grad_norm": 0.47582384943962097,
      "learning_rate": 3.132091431485152e-05,
      "loss": 0.6363,
      "step": 1358
    },
    {
      "epoch": 1.73856,
      "grad_norm": 0.46250778436660767,
      "learning_rate": 3.130815360074428e-05,
      "loss": 0.6316,
      "step": 1359
    },
    {
      "epoch": 1.73984,
      "grad_norm": 0.464727520942688,
      "learning_rate": 3.1295386497729476e-05,
      "loss": 0.5974,
      "step": 1360
    },
    {
      "epoch": 1.74112,
      "grad_norm": 0.4753580093383789,
      "learning_rate": 3.1282613014570166e-05,
      "loss": 0.6252,
      "step": 1361
    },
    {
      "epoch": 1.7424,
      "grad_norm": 0.4359992742538452,
      "learning_rate": 3.126983316003378e-05,
      "loss": 0.6002,
      "step": 1362
    },
    {
      "epoch": 1.74368,
      "grad_norm": 0.46032530069351196,
      "learning_rate": 3.125704694289212e-05,
      "loss": 0.6022,
      "step": 1363
    },
    {
      "epoch": 1.7449599999999998,
      "grad_norm": 0.4996948540210724,
      "learning_rate": 3.124425437192137e-05,
      "loss": 0.6371,
      "step": 1364
    },
    {
      "epoch": 1.74624,
      "grad_norm": 0.47237974405288696,
      "learning_rate": 3.123145545590204e-05,
      "loss": 0.6254,
      "step": 1365
    },
    {
      "epoch": 1.74752,
      "grad_norm": 0.45381543040275574,
      "learning_rate": 3.121865020361904e-05,
      "loss": 0.5992,
      "step": 1366
    },
    {
      "epoch": 1.7488000000000001,
      "grad_norm": 0.45342546701431274,
      "learning_rate": 3.12058386238616e-05,
      "loss": 0.5973,
      "step": 1367
    },
    {
      "epoch": 1.75008,
      "grad_norm": 0.509502649307251,
      "learning_rate": 3.11930207254233e-05,
      "loss": 0.6245,
      "step": 1368
    },
    {
      "epoch": 1.75136,
      "grad_norm": 0.5161288380622864,
      "learning_rate": 3.118019651710207e-05,
      "loss": 0.6313,
      "step": 1369
    },
    {
      "epoch": 1.75264,
      "grad_norm": 0.4396875202655792,
      "learning_rate": 3.1167366007700135e-05,
      "loss": 0.6554,
      "step": 1370
    },
    {
      "epoch": 1.75392,
      "grad_norm": 0.5149477124214172,
      "learning_rate": 3.1154529206024075e-05,
      "loss": 0.6472,
      "step": 1371
    },
    {
      "epoch": 1.7551999999999999,
      "grad_norm": 0.4774300158023834,
      "learning_rate": 3.1141686120884804e-05,
      "loss": 0.612,
      "step": 1372
    },
    {
      "epoch": 1.75648,
      "grad_norm": 0.44212377071380615,
      "learning_rate": 3.112883676109751e-05,
      "loss": 0.5879,
      "step": 1373
    },
    {
      "epoch": 1.75776,
      "grad_norm": 0.4760090708732605,
      "learning_rate": 3.11159811354817e-05,
      "loss": 0.6374,
      "step": 1374
    },
    {
      "epoch": 1.7590400000000002,
      "grad_norm": 0.44294464588165283,
      "learning_rate": 3.110311925286119e-05,
      "loss": 0.6228,
      "step": 1375
    },
    {
      "epoch": 1.76032,
      "grad_norm": 0.48608365654945374,
      "learning_rate": 3.10902511220641e-05,
      "loss": 0.6427,
      "step": 1376
    },
    {
      "epoch": 1.7616,
      "grad_norm": 0.4990738332271576,
      "learning_rate": 3.1077376751922805e-05,
      "loss": 0.5794,
      "step": 1377
    },
    {
      "epoch": 1.76288,
      "grad_norm": 0.46069759130477905,
      "learning_rate": 3.1064496151274e-05,
      "loss": 0.6006,
      "step": 1378
    },
    {
      "epoch": 1.76416,
      "grad_norm": 0.46551960706710815,
      "learning_rate": 3.105160932895864e-05,
      "loss": 0.5893,
      "step": 1379
    },
    {
      "epoch": 1.76544,
      "grad_norm": 0.5440554022789001,
      "learning_rate": 3.103871629382193e-05,
      "loss": 0.6425,
      "step": 1380
    },
    {
      "epoch": 1.76544,
      "eval_loss": 0.6858192086219788,
      "eval_runtime": 19.5945,
      "eval_samples_per_second": 25.619,
      "eval_steps_per_second": 3.215,
      "step": 1380
    },
    {
      "epoch": 1.7667199999999998,
      "grad_norm": 0.5122801065444946,
      "learning_rate": 3.1025817054713385e-05,
      "loss": 0.6086,
      "step": 1381
    },
    {
      "epoch": 1.768,
      "grad_norm": 0.519063413143158,
      "learning_rate": 3.101291162048675e-05,
      "loss": 0.66,
      "step": 1382
    },
    {
      "epoch": 1.76928,
      "grad_norm": 0.5404231548309326,
      "learning_rate": 3.1e-05,
      "loss": 0.6209,
      "step": 1383
    },
    {
      "epoch": 1.7705600000000001,
      "grad_norm": 0.5387503504753113,
      "learning_rate": 3.098708220211541e-05,
      "loss": 0.6075,
      "step": 1384
    },
    {
      "epoch": 1.77184,
      "grad_norm": 0.4766044020652771,
      "learning_rate": 3.097415823569946e-05,
      "loss": 0.6263,
      "step": 1385
    },
    {
      "epoch": 1.77312,
      "grad_norm": 0.5069013833999634,
      "learning_rate": 3.0961228109622856e-05,
      "loss": 0.6121,
      "step": 1386
    },
    {
      "epoch": 1.7744,
      "grad_norm": 0.4816810190677643,
      "learning_rate": 3.094829183276056e-05,
      "loss": 0.6086,
      "step": 1387
    },
    {
      "epoch": 1.77568,
      "grad_norm": 0.5198071599006653,
      "learning_rate": 3.093534941399174e-05,
      "loss": 0.6095,
      "step": 1388
    },
    {
      "epoch": 1.7769599999999999,
      "grad_norm": 0.5285684466362,
      "learning_rate": 3.092240086219978e-05,
      "loss": 0.6263,
      "step": 1389
    },
    {
      "epoch": 1.77824,
      "grad_norm": 0.48517653346061707,
      "learning_rate": 3.090944618627227e-05,
      "loss": 0.621,
      "step": 1390
    },
    {
      "epoch": 1.77952,
      "grad_norm": 0.5991851687431335,
      "learning_rate": 3.0896485395101016e-05,
      "loss": 0.639,
      "step": 1391
    },
    {
      "epoch": 1.7808000000000002,
      "grad_norm": 0.5169137120246887,
      "learning_rate": 3.088351849758201e-05,
      "loss": 0.6225,
      "step": 1392
    },
    {
      "epoch": 1.78208,
      "grad_norm": 0.49744483828544617,
      "learning_rate": 3.087054550261544e-05,
      "loss": 0.6081,
      "step": 1393
    },
    {
      "epoch": 1.78336,
      "grad_norm": 0.6128389835357666,
      "learning_rate": 3.0857566419105676e-05,
      "loss": 0.6276,
      "step": 1394
    },
    {
      "epoch": 1.78464,
      "grad_norm": 0.5034732818603516,
      "learning_rate": 3.0844581255961275e-05,
      "loss": 0.5961,
      "step": 1395
    },
    {
      "epoch": 1.78592,
      "grad_norm": 0.4583236575126648,
      "learning_rate": 3.083159002209497e-05,
      "loss": 0.5829,
      "step": 1396
    },
    {
      "epoch": 1.7872,
      "grad_norm": 0.5208564400672913,
      "learning_rate": 3.0818592726423616e-05,
      "loss": 0.5884,
      "step": 1397
    },
    {
      "epoch": 1.7884799999999998,
      "grad_norm": 0.45216360688209534,
      "learning_rate": 3.080558937786831e-05,
      "loss": 0.6208,
      "step": 1398
    },
    {
      "epoch": 1.78976,
      "grad_norm": 0.49808359146118164,
      "learning_rate": 3.079257998535423e-05,
      "loss": 0.6208,
      "step": 1399
    },
    {
      "epoch": 1.79104,
      "grad_norm": 0.4901629090309143,
      "learning_rate": 3.077956455781073e-05,
      "loss": 0.614,
      "step": 1400
    },
    {
      "epoch": 1.7923200000000001,
      "grad_norm": 0.48698392510414124,
      "learning_rate": 3.0766543104171326e-05,
      "loss": 0.5861,
      "step": 1401
    },
    {
      "epoch": 1.7936,
      "grad_norm": 0.480970174074173,
      "learning_rate": 3.0753515633373634e-05,
      "loss": 0.6155,
      "step": 1402
    },
    {
      "epoch": 1.79488,
      "grad_norm": 0.6318512558937073,
      "learning_rate": 3.0740482154359434e-05,
      "loss": 0.6516,
      "step": 1403
    },
    {
      "epoch": 1.79616,
      "grad_norm": 0.46645238995552063,
      "learning_rate": 3.0727442676074595e-05,
      "loss": 0.6439,
      "step": 1404
    },
    {
      "epoch": 1.79744,
      "grad_norm": 0.5154925584793091,
      "learning_rate": 3.071439720746913e-05,
      "loss": 0.6119,
      "step": 1405
    },
    {
      "epoch": 1.7987199999999999,
      "grad_norm": 0.48363927006721497,
      "learning_rate": 3.070134575749717e-05,
      "loss": 0.613,
      "step": 1406
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.4920346736907959,
      "learning_rate": 3.0688288335116906e-05,
      "loss": 0.6045,
      "step": 1407
    },
    {
      "epoch": 1.80128,
      "grad_norm": 0.46684446930885315,
      "learning_rate": 3.067522494929069e-05,
      "loss": 0.6114,
      "step": 1408
    },
    {
      "epoch": 1.8025600000000002,
      "grad_norm": 0.4765709340572357,
      "learning_rate": 3.066215560898491e-05,
      "loss": 0.5943,
      "step": 1409
    },
    {
      "epoch": 1.80384,
      "grad_norm": 0.4900147616863251,
      "learning_rate": 3.064908032317009e-05,
      "loss": 0.6279,
      "step": 1410
    },
    {
      "epoch": 1.80384,
      "eval_loss": 0.6844485402107239,
      "eval_runtime": 19.6013,
      "eval_samples_per_second": 25.611,
      "eval_steps_per_second": 3.214,
      "step": 1410
    },
    {
      "epoch": 1.80512,
      "grad_norm": 0.4650956988334656,
      "learning_rate": 3.06359991008208e-05,
      "loss": 0.5891,
      "step": 1411
    },
    {
      "epoch": 1.8064,
      "grad_norm": 0.4766197204589844,
      "learning_rate": 3.06229119509157e-05,
      "loss": 0.6105,
      "step": 1412
    },
    {
      "epoch": 1.80768,
      "grad_norm": 0.47516539692878723,
      "learning_rate": 3.060981888243752e-05,
      "loss": 0.6218,
      "step": 1413
    },
    {
      "epoch": 1.80896,
      "grad_norm": 0.49431338906288147,
      "learning_rate": 3.0596719904373036e-05,
      "loss": 0.6333,
      "step": 1414
    },
    {
      "epoch": 1.8102399999999998,
      "grad_norm": 0.47089746594429016,
      "learning_rate": 3.058361502571312e-05,
      "loss": 0.6386,
      "step": 1415
    },
    {
      "epoch": 1.81152,
      "grad_norm": 0.5137251019477844,
      "learning_rate": 3.057050425545263e-05,
      "loss": 0.6264,
      "step": 1416
    },
    {
      "epoch": 1.8128,
      "grad_norm": 0.45938169956207275,
      "learning_rate": 3.0557387602590526e-05,
      "loss": 0.6202,
      "step": 1417
    },
    {
      "epoch": 1.8140800000000001,
      "grad_norm": 0.44876766204833984,
      "learning_rate": 3.054426507612977e-05,
      "loss": 0.621,
      "step": 1418
    },
    {
      "epoch": 1.81536,
      "grad_norm": 0.4695495069026947,
      "learning_rate": 3.0531136685077397e-05,
      "loss": 0.6041,
      "step": 1419
    },
    {
      "epoch": 1.81664,
      "grad_norm": 0.5342169404029846,
      "learning_rate": 3.0518002438444406e-05,
      "loss": 0.6073,
      "step": 1420
    },
    {
      "epoch": 1.81792,
      "grad_norm": 0.506734311580658,
      "learning_rate": 3.0504862345245875e-05,
      "loss": 0.6153,
      "step": 1421
    },
    {
      "epoch": 1.8192,
      "grad_norm": 0.4909009635448456,
      "learning_rate": 3.0491716414500847e-05,
      "loss": 0.605,
      "step": 1422
    },
    {
      "epoch": 1.8204799999999999,
      "grad_norm": 0.5573544502258301,
      "learning_rate": 3.0478564655232414e-05,
      "loss": 0.6745,
      "step": 1423
    },
    {
      "epoch": 1.82176,
      "grad_norm": 0.452993243932724,
      "learning_rate": 3.0465407076467638e-05,
      "loss": 0.6159,
      "step": 1424
    },
    {
      "epoch": 1.82304,
      "grad_norm": 0.504835307598114,
      "learning_rate": 3.045224368723758e-05,
      "loss": 0.6126,
      "step": 1425
    },
    {
      "epoch": 1.8243200000000002,
      "grad_norm": 0.5175802111625671,
      "learning_rate": 3.043907449657731e-05,
      "loss": 0.6054,
      "step": 1426
    },
    {
      "epoch": 1.8256000000000001,
      "grad_norm": 0.4913342595100403,
      "learning_rate": 3.042589951352585e-05,
      "loss": 0.6109,
      "step": 1427
    },
    {
      "epoch": 1.82688,
      "grad_norm": 0.5518784523010254,
      "learning_rate": 3.0412718747126218e-05,
      "loss": 0.6118,
      "step": 1428
    },
    {
      "epoch": 1.82816,
      "grad_norm": 0.479159414768219,
      "learning_rate": 3.0399532206425402e-05,
      "loss": 0.5905,
      "step": 1429
    },
    {
      "epoch": 1.82944,
      "grad_norm": 0.49373456835746765,
      "learning_rate": 3.0386339900474337e-05,
      "loss": 0.6082,
      "step": 1430
    },
    {
      "epoch": 1.83072,
      "grad_norm": 0.5093984603881836,
      "learning_rate": 3.037314183832793e-05,
      "loss": 0.6189,
      "step": 1431
    },
    {
      "epoch": 1.8319999999999999,
      "grad_norm": 0.4808180630207062,
      "learning_rate": 3.035993802904504e-05,
      "loss": 0.6454,
      "step": 1432
    },
    {
      "epoch": 1.83328,
      "grad_norm": 0.5686001777648926,
      "learning_rate": 3.0346728481688452e-05,
      "loss": 0.6321,
      "step": 1433
    },
    {
      "epoch": 1.83456,
      "grad_norm": 0.4634808897972107,
      "learning_rate": 3.0333513205324916e-05,
      "loss": 0.6023,
      "step": 1434
    },
    {
      "epoch": 1.8358400000000001,
      "grad_norm": 0.5245974659919739,
      "learning_rate": 3.0320292209025092e-05,
      "loss": 0.6061,
      "step": 1435
    },
    {
      "epoch": 1.83712,
      "grad_norm": 0.4950631260871887,
      "learning_rate": 3.0307065501863576e-05,
      "loss": 0.6513,
      "step": 1436
    },
    {
      "epoch": 1.8384,
      "grad_norm": 0.5015468597412109,
      "learning_rate": 3.0293833092918886e-05,
      "loss": 0.6308,
      "step": 1437
    },
    {
      "epoch": 1.83968,
      "grad_norm": 0.4837664067745209,
      "learning_rate": 3.0280594991273452e-05,
      "loss": 0.5915,
      "step": 1438
    },
    {
      "epoch": 1.84096,
      "grad_norm": 0.4777585566043854,
      "learning_rate": 3.0267351206013602e-05,
      "loss": 0.5945,
      "step": 1439
    },
    {
      "epoch": 1.8422399999999999,
      "grad_norm": 0.4659727215766907,
      "learning_rate": 3.025410174622958e-05,
      "loss": 0.6155,
      "step": 1440
    },
    {
      "epoch": 1.8422399999999999,
      "eval_loss": 0.6829977035522461,
      "eval_runtime": 19.6488,
      "eval_samples_per_second": 25.549,
      "eval_steps_per_second": 3.206,
      "step": 1440
    },
    {
      "epoch": 1.84352,
      "grad_norm": 0.4669051766395569,
      "learning_rate": 3.0240846621015523e-05,
      "loss": 0.6208,
      "step": 1441
    },
    {
      "epoch": 1.8448,
      "grad_norm": 0.4984658360481262,
      "learning_rate": 3.022758583946943e-05,
      "loss": 0.5836,
      "step": 1442
    },
    {
      "epoch": 1.8460800000000002,
      "grad_norm": 0.4953427016735077,
      "learning_rate": 3.0214319410693223e-05,
      "loss": 0.6319,
      "step": 1443
    },
    {
      "epoch": 1.8473600000000001,
      "grad_norm": 0.46328282356262207,
      "learning_rate": 3.0201047343792672e-05,
      "loss": 0.5975,
      "step": 1444
    },
    {
      "epoch": 1.84864,
      "grad_norm": 0.4950697720050812,
      "learning_rate": 3.018776964787743e-05,
      "loss": 0.625,
      "step": 1445
    },
    {
      "epoch": 1.84992,
      "grad_norm": 0.49767038226127625,
      "learning_rate": 3.017448633206101e-05,
      "loss": 0.6006,
      "step": 1446
    },
    {
      "epoch": 1.8512,
      "grad_norm": 0.4775907099246979,
      "learning_rate": 3.0161197405460774e-05,
      "loss": 0.6445,
      "step": 1447
    },
    {
      "epoch": 1.85248,
      "grad_norm": 0.5154576301574707,
      "learning_rate": 3.0147902877197954e-05,
      "loss": 0.6092,
      "step": 1448
    },
    {
      "epoch": 1.8537599999999999,
      "grad_norm": 0.5414725542068481,
      "learning_rate": 3.0134602756397603e-05,
      "loss": 0.6108,
      "step": 1449
    },
    {
      "epoch": 1.85504,
      "grad_norm": 0.4950706958770752,
      "learning_rate": 3.0121297052188638e-05,
      "loss": 0.6283,
      "step": 1450
    },
    {
      "epoch": 1.85632,
      "grad_norm": 0.4902404248714447,
      "learning_rate": 3.0107985773703783e-05,
      "loss": 0.6115,
      "step": 1451
    },
    {
      "epoch": 1.8576000000000001,
      "grad_norm": 0.4769650399684906,
      "learning_rate": 3.0094668930079606e-05,
      "loss": 0.6055,
      "step": 1452
    },
    {
      "epoch": 1.85888,
      "grad_norm": 0.5179007053375244,
      "learning_rate": 3.0081346530456496e-05,
      "loss": 0.6099,
      "step": 1453
    },
    {
      "epoch": 1.86016,
      "grad_norm": 0.5024033188819885,
      "learning_rate": 3.006801858397864e-05,
      "loss": 0.586,
      "step": 1454
    },
    {
      "epoch": 1.86144,
      "grad_norm": 0.4630223214626312,
      "learning_rate": 3.005468509979404e-05,
      "loss": 0.6157,
      "step": 1455
    },
    {
      "epoch": 1.86272,
      "grad_norm": 0.45166951417922974,
      "learning_rate": 3.0041346087054508e-05,
      "loss": 0.6229,
      "step": 1456
    },
    {
      "epoch": 1.8639999999999999,
      "grad_norm": 0.461215615272522,
      "learning_rate": 3.002800155491564e-05,
      "loss": 0.6301,
      "step": 1457
    },
    {
      "epoch": 1.86528,
      "grad_norm": 0.46575331687927246,
      "learning_rate": 3.0014651512536826e-05,
      "loss": 0.598,
      "step": 1458
    },
    {
      "epoch": 1.86656,
      "grad_norm": 0.4878748655319214,
      "learning_rate": 3.000129596908122e-05,
      "loss": 0.5859,
      "step": 1459
    },
    {
      "epoch": 1.86784,
      "grad_norm": 0.48017966747283936,
      "learning_rate": 2.9987934933715786e-05,
      "loss": 0.61,
      "step": 1460
    },
    {
      "epoch": 1.8691200000000001,
      "grad_norm": 0.4690764546394348,
      "learning_rate": 2.997456841561123e-05,
      "loss": 0.6089,
      "step": 1461
    },
    {
      "epoch": 1.8704,
      "grad_norm": 0.4682995676994324,
      "learning_rate": 2.9961196423942026e-05,
      "loss": 0.6058,
      "step": 1462
    },
    {
      "epoch": 1.87168,
      "grad_norm": 0.46251216530799866,
      "learning_rate": 2.9947818967886416e-05,
      "loss": 0.6556,
      "step": 1463
    },
    {
      "epoch": 1.87296,
      "grad_norm": 0.4609954059123993,
      "learning_rate": 2.9934436056626386e-05,
      "loss": 0.6094,
      "step": 1464
    },
    {
      "epoch": 1.87424,
      "grad_norm": 0.4351571202278137,
      "learning_rate": 2.9921047699347653e-05,
      "loss": 0.6288,
      "step": 1465
    },
    {
      "epoch": 1.8755199999999999,
      "grad_norm": 0.48737555742263794,
      "learning_rate": 2.9907653905239703e-05,
      "loss": 0.601,
      "step": 1466
    },
    {
      "epoch": 1.8768,
      "grad_norm": 0.4811854958534241,
      "learning_rate": 2.9894254683495724e-05,
      "loss": 0.6436,
      "step": 1467
    },
    {
      "epoch": 1.87808,
      "grad_norm": 0.4843752086162567,
      "learning_rate": 2.988085004331264e-05,
      "loss": 0.6219,
      "step": 1468
    },
    {
      "epoch": 1.8793600000000001,
      "grad_norm": 0.4829710125923157,
      "learning_rate": 2.9867439993891104e-05,
      "loss": 0.5944,
      "step": 1469
    },
    {
      "epoch": 1.88064,
      "grad_norm": 0.48306459188461304,
      "learning_rate": 2.9854024544435463e-05,
      "loss": 0.5932,
      "step": 1470
    },
    {
      "epoch": 1.88064,
      "eval_loss": 0.682723343372345,
      "eval_runtime": 19.5929,
      "eval_samples_per_second": 25.621,
      "eval_steps_per_second": 3.215,
      "step": 1470
    },
    {
      "epoch": 1.88192,
      "grad_norm": 0.5078936219215393,
      "learning_rate": 2.9840603704153785e-05,
      "loss": 0.6452,
      "step": 1471
    },
    {
      "epoch": 1.8832,
      "grad_norm": 0.460633784532547,
      "learning_rate": 2.9827177482257834e-05,
      "loss": 0.6068,
      "step": 1472
    },
    {
      "epoch": 1.88448,
      "grad_norm": 0.45168301463127136,
      "learning_rate": 2.9813745887963065e-05,
      "loss": 0.6407,
      "step": 1473
    },
    {
      "epoch": 1.8857599999999999,
      "grad_norm": 0.46906033158302307,
      "learning_rate": 2.980030893048862e-05,
      "loss": 0.629,
      "step": 1474
    },
    {
      "epoch": 1.88704,
      "grad_norm": 0.415548175573349,
      "learning_rate": 2.9786866619057328e-05,
      "loss": 0.5876,
      "step": 1475
    },
    {
      "epoch": 1.88832,
      "grad_norm": 0.4550069570541382,
      "learning_rate": 2.9773418962895695e-05,
      "loss": 0.6157,
      "step": 1476
    },
    {
      "epoch": 1.8896,
      "grad_norm": 0.4906242787837982,
      "learning_rate": 2.975996597123388e-05,
      "loss": 0.6298,
      "step": 1477
    },
    {
      "epoch": 1.8908800000000001,
      "grad_norm": 0.4912736713886261,
      "learning_rate": 2.974650765330572e-05,
      "loss": 0.6157,
      "step": 1478
    },
    {
      "epoch": 1.89216,
      "grad_norm": 0.4961681067943573,
      "learning_rate": 2.9733044018348707e-05,
      "loss": 0.613,
      "step": 1479
    },
    {
      "epoch": 1.89344,
      "grad_norm": 0.5538224577903748,
      "learning_rate": 2.971957507560397e-05,
      "loss": 0.6276,
      "step": 1480
    },
    {
      "epoch": 1.89472,
      "grad_norm": 0.5274718999862671,
      "learning_rate": 2.9706100834316282e-05,
      "loss": 0.6181,
      "step": 1481
    },
    {
      "epoch": 1.896,
      "grad_norm": 0.5358492732048035,
      "learning_rate": 2.9692621303734073e-05,
      "loss": 0.6346,
      "step": 1482
    },
    {
      "epoch": 1.8972799999999999,
      "grad_norm": 0.42724448442459106,
      "learning_rate": 2.9679136493109392e-05,
      "loss": 0.5998,
      "step": 1483
    },
    {
      "epoch": 1.89856,
      "grad_norm": 0.4989132285118103,
      "learning_rate": 2.96656464116979e-05,
      "loss": 0.5958,
      "step": 1484
    },
    {
      "epoch": 1.89984,
      "grad_norm": 0.46765923500061035,
      "learning_rate": 2.9652151068758882e-05,
      "loss": 0.5829,
      "step": 1485
    },
    {
      "epoch": 1.9011200000000001,
      "grad_norm": 0.4639012813568115,
      "learning_rate": 2.963865047355525e-05,
      "loss": 0.6324,
      "step": 1486
    },
    {
      "epoch": 1.9024,
      "grad_norm": 0.4307348430156708,
      "learning_rate": 2.9625144635353514e-05,
      "loss": 0.5869,
      "step": 1487
    },
    {
      "epoch": 1.90368,
      "grad_norm": 0.488514244556427,
      "learning_rate": 2.9611633563423755e-05,
      "loss": 0.5991,
      "step": 1488
    },
    {
      "epoch": 1.90496,
      "grad_norm": 0.4567602574825287,
      "learning_rate": 2.959811726703969e-05,
      "loss": 0.6364,
      "step": 1489
    },
    {
      "epoch": 1.90624,
      "grad_norm": 0.44587045907974243,
      "learning_rate": 2.9584595755478597e-05,
      "loss": 0.6328,
      "step": 1490
    },
    {
      "epoch": 1.9075199999999999,
      "grad_norm": 0.46266430616378784,
      "learning_rate": 2.9571069038021324e-05,
      "loss": 0.6359,
      "step": 1491
    },
    {
      "epoch": 1.9088,
      "grad_norm": 0.4900732934474945,
      "learning_rate": 2.955753712395233e-05,
      "loss": 0.5796,
      "step": 1492
    },
    {
      "epoch": 1.91008,
      "grad_norm": 0.4824363589286804,
      "learning_rate": 2.9544000022559598e-05,
      "loss": 0.5966,
      "step": 1493
    },
    {
      "epoch": 1.91136,
      "grad_norm": 0.4921255111694336,
      "learning_rate": 2.9530457743134694e-05,
      "loss": 0.6204,
      "step": 1494
    },
    {
      "epoch": 1.9126400000000001,
      "grad_norm": 0.46053797006607056,
      "learning_rate": 2.9516910294972745e-05,
      "loss": 0.6208,
      "step": 1495
    },
    {
      "epoch": 1.91392,
      "grad_norm": 0.45902523398399353,
      "learning_rate": 2.9503357687372402e-05,
      "loss": 0.5942,
      "step": 1496
    },
    {
      "epoch": 1.9152,
      "grad_norm": 0.44390127062797546,
      "learning_rate": 2.9489799929635875e-05,
      "loss": 0.5929,
      "step": 1497
    },
    {
      "epoch": 1.91648,
      "grad_norm": 0.5113105177879333,
      "learning_rate": 2.947623703106891e-05,
      "loss": 0.6325,
      "step": 1498
    },
    {
      "epoch": 1.91776,
      "grad_norm": 0.47255396842956543,
      "learning_rate": 2.9462669000980763e-05,
      "loss": 0.6341,
      "step": 1499
    },
    {
      "epoch": 1.9190399999999999,
      "grad_norm": 0.4689778983592987,
      "learning_rate": 2.9449095848684233e-05,
      "loss": 0.6311,
      "step": 1500
    },
    {
      "epoch": 1.9190399999999999,
      "eval_loss": 0.6820359230041504,
      "eval_runtime": 19.5787,
      "eval_samples_per_second": 25.64,
      "eval_steps_per_second": 3.218,
      "step": 1500
    },
    {
      "epoch": 1.92032,
      "grad_norm": 0.5162354111671448,
      "learning_rate": 2.9435517583495633e-05,
      "loss": 0.5997,
      "step": 1501
    },
    {
      "epoch": 1.9216,
      "grad_norm": 0.5373241901397705,
      "learning_rate": 2.9421934214734768e-05,
      "loss": 0.6314,
      "step": 1502
    },
    {
      "epoch": 1.9228800000000001,
      "grad_norm": 0.5147911906242371,
      "learning_rate": 2.9408345751724967e-05,
      "loss": 0.6424,
      "step": 1503
    },
    {
      "epoch": 1.92416,
      "grad_norm": 0.4841938614845276,
      "learning_rate": 2.9394752203793038e-05,
      "loss": 0.5884,
      "step": 1504
    },
    {
      "epoch": 1.92544,
      "grad_norm": 0.487347811460495,
      "learning_rate": 2.938115358026929e-05,
      "loss": 0.6152,
      "step": 1505
    },
    {
      "epoch": 1.92672,
      "grad_norm": 0.4829878807067871,
      "learning_rate": 2.9367549890487513e-05,
      "loss": 0.6064,
      "step": 1506
    },
    {
      "epoch": 1.928,
      "grad_norm": 0.4968488812446594,
      "learning_rate": 2.9353941143784977e-05,
      "loss": 0.6111,
      "step": 1507
    },
    {
      "epoch": 1.9292799999999999,
      "grad_norm": 0.47948139905929565,
      "learning_rate": 2.934032734950241e-05,
      "loss": 0.597,
      "step": 1508
    },
    {
      "epoch": 1.93056,
      "grad_norm": 0.48983055353164673,
      "learning_rate": 2.9326708516984032e-05,
      "loss": 0.6159,
      "step": 1509
    },
    {
      "epoch": 1.93184,
      "grad_norm": 0.4299463629722595,
      "learning_rate": 2.9313084655577486e-05,
      "loss": 0.5964,
      "step": 1510
    },
    {
      "epoch": 1.93312,
      "grad_norm": 0.4318764805793762,
      "learning_rate": 2.9299455774633894e-05,
      "loss": 0.6226,
      "step": 1511
    },
    {
      "epoch": 1.9344000000000001,
      "grad_norm": 0.44073593616485596,
      "learning_rate": 2.928582188350781e-05,
      "loss": 0.5932,
      "step": 1512
    },
    {
      "epoch": 1.93568,
      "grad_norm": 0.41787245869636536,
      "learning_rate": 2.927218299155723e-05,
      "loss": 0.6165,
      "step": 1513
    },
    {
      "epoch": 1.93696,
      "grad_norm": 0.5000843405723572,
      "learning_rate": 2.9258539108143588e-05,
      "loss": 0.6145,
      "step": 1514
    },
    {
      "epoch": 1.93824,
      "grad_norm": 0.45677292346954346,
      "learning_rate": 2.924489024263174e-05,
      "loss": 0.6511,
      "step": 1515
    },
    {
      "epoch": 1.93952,
      "grad_norm": 0.5070911049842834,
      "learning_rate": 2.923123640438996e-05,
      "loss": 0.6405,
      "step": 1516
    },
    {
      "epoch": 1.9407999999999999,
      "grad_norm": 0.4740370213985443,
      "learning_rate": 2.921757760278992e-05,
      "loss": 0.6107,
      "step": 1517
    },
    {
      "epoch": 1.94208,
      "grad_norm": 0.4753107726573944,
      "learning_rate": 2.9203913847206737e-05,
      "loss": 0.5975,
      "step": 1518
    },
    {
      "epoch": 1.94336,
      "grad_norm": 0.4732353091239929,
      "learning_rate": 2.9190245147018892e-05,
      "loss": 0.6006,
      "step": 1519
    },
    {
      "epoch": 1.9446400000000001,
      "grad_norm": 0.479711651802063,
      "learning_rate": 2.9176571511608278e-05,
      "loss": 0.5974,
      "step": 1520
    },
    {
      "epoch": 1.94592,
      "grad_norm": 0.5606484413146973,
      "learning_rate": 2.9162892950360175e-05,
      "loss": 0.6359,
      "step": 1521
    },
    {
      "epoch": 1.9472,
      "grad_norm": 0.4636978209018707,
      "learning_rate": 2.914920947266324e-05,
      "loss": 0.6259,
      "step": 1522
    },
    {
      "epoch": 1.94848,
      "grad_norm": 0.4966209828853607,
      "learning_rate": 2.9135521087909494e-05,
      "loss": 0.6247,
      "step": 1523
    },
    {
      "epoch": 1.94976,
      "grad_norm": 0.5154779553413391,
      "learning_rate": 2.9121827805494347e-05,
      "loss": 0.6491,
      "step": 1524
    },
    {
      "epoch": 1.9510399999999999,
      "grad_norm": 0.4724636971950531,
      "learning_rate": 2.910812963481656e-05,
      "loss": 0.6139,
      "step": 1525
    },
    {
      "epoch": 1.95232,
      "grad_norm": 0.49538785219192505,
      "learning_rate": 2.9094426585278247e-05,
      "loss": 0.6155,
      "step": 1526
    },
    {
      "epoch": 1.9536,
      "grad_norm": 0.49748390913009644,
      "learning_rate": 2.908071866628487e-05,
      "loss": 0.6154,
      "step": 1527
    },
    {
      "epoch": 1.95488,
      "grad_norm": 0.4940442740917206,
      "learning_rate": 2.906700588724525e-05,
      "loss": 0.623,
      "step": 1528
    },
    {
      "epoch": 1.9561600000000001,
      "grad_norm": 0.5037598013877869,
      "learning_rate": 2.9053288257571514e-05,
      "loss": 0.6512,
      "step": 1529
    },
    {
      "epoch": 1.95744,
      "grad_norm": 0.4246531128883362,
      "learning_rate": 2.9039565786679142e-05,
      "loss": 0.5874,
      "step": 1530
    },
    {
      "epoch": 1.95744,
      "eval_loss": 0.6807330250740051,
      "eval_runtime": 19.5816,
      "eval_samples_per_second": 25.636,
      "eval_steps_per_second": 3.217,
      "step": 1530
    },
    {
      "epoch": 1.95872,
      "grad_norm": 0.45062705874443054,
      "learning_rate": 2.9025838483986928e-05,
      "loss": 0.5932,
      "step": 1531
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.4469867944717407,
      "learning_rate": 2.9012106358916994e-05,
      "loss": 0.6393,
      "step": 1532
    },
    {
      "epoch": 1.96128,
      "grad_norm": 0.43591511249542236,
      "learning_rate": 2.8998369420894746e-05,
      "loss": 0.6273,
      "step": 1533
    },
    {
      "epoch": 1.9625599999999999,
      "grad_norm": 0.43304944038391113,
      "learning_rate": 2.8984627679348922e-05,
      "loss": 0.6046,
      "step": 1534
    },
    {
      "epoch": 1.96384,
      "grad_norm": 0.5330939292907715,
      "learning_rate": 2.8970881143711542e-05,
      "loss": 0.6273,
      "step": 1535
    },
    {
      "epoch": 1.96512,
      "grad_norm": 0.43347227573394775,
      "learning_rate": 2.8957129823417916e-05,
      "loss": 0.5996,
      "step": 1536
    },
    {
      "epoch": 1.9664000000000001,
      "grad_norm": 0.5185285210609436,
      "learning_rate": 2.894337372790664e-05,
      "loss": 0.5892,
      "step": 1537
    },
    {
      "epoch": 1.96768,
      "grad_norm": 0.4612886309623718,
      "learning_rate": 2.8929612866619603e-05,
      "loss": 0.6085,
      "step": 1538
    },
    {
      "epoch": 1.96896,
      "grad_norm": 0.5051066875457764,
      "learning_rate": 2.8915847249001934e-05,
      "loss": 0.6055,
      "step": 1539
    },
    {
      "epoch": 1.97024,
      "grad_norm": 0.5502194762229919,
      "learning_rate": 2.890207688450205e-05,
      "loss": 0.6408,
      "step": 1540
    },
    {
      "epoch": 1.97152,
      "grad_norm": 0.43735823035240173,
      "learning_rate": 2.8888301782571618e-05,
      "loss": 0.6157,
      "step": 1541
    },
    {
      "epoch": 1.9727999999999999,
      "grad_norm": 0.5195667147636414,
      "learning_rate": 2.8874521952665566e-05,
      "loss": 0.6215,
      "step": 1542
    },
    {
      "epoch": 1.9740799999999998,
      "grad_norm": 0.4594319760799408,
      "learning_rate": 2.886073740424205e-05,
      "loss": 0.6078,
      "step": 1543
    },
    {
      "epoch": 1.97536,
      "grad_norm": 0.4490661323070526,
      "learning_rate": 2.8846948146762476e-05,
      "loss": 0.6028,
      "step": 1544
    },
    {
      "epoch": 1.97664,
      "grad_norm": 0.4709550440311432,
      "learning_rate": 2.8833154189691495e-05,
      "loss": 0.6287,
      "step": 1545
    },
    {
      "epoch": 1.9779200000000001,
      "grad_norm": 0.46984657645225525,
      "learning_rate": 2.8819355542496946e-05,
      "loss": 0.5836,
      "step": 1546
    },
    {
      "epoch": 1.9792,
      "grad_norm": 0.46593722701072693,
      "learning_rate": 2.880555221464993e-05,
      "loss": 0.6104,
      "step": 1547
    },
    {
      "epoch": 1.98048,
      "grad_norm": 0.4238821864128113,
      "learning_rate": 2.8791744215624737e-05,
      "loss": 0.6033,
      "step": 1548
    },
    {
      "epoch": 1.98176,
      "grad_norm": 0.46578243374824524,
      "learning_rate": 2.8777931554898855e-05,
      "loss": 0.6026,
      "step": 1549
    },
    {
      "epoch": 1.98304,
      "grad_norm": 0.5095911026000977,
      "learning_rate": 2.8764114241952996e-05,
      "loss": 0.6147,
      "step": 1550
    },
    {
      "epoch": 1.9843199999999999,
      "grad_norm": 0.47689592838287354,
      "learning_rate": 2.8750292286271047e-05,
      "loss": 0.6172,
      "step": 1551
    },
    {
      "epoch": 1.9856,
      "grad_norm": 0.4785391688346863,
      "learning_rate": 2.87364656973401e-05,
      "loss": 0.5976,
      "step": 1552
    },
    {
      "epoch": 1.98688,
      "grad_norm": 0.5094159245491028,
      "learning_rate": 2.8722634484650397e-05,
      "loss": 0.5918,
      "step": 1553
    },
    {
      "epoch": 1.9881600000000001,
      "grad_norm": 0.46542686223983765,
      "learning_rate": 2.8708798657695384e-05,
      "loss": 0.5875,
      "step": 1554
    },
    {
      "epoch": 1.98944,
      "grad_norm": 0.45332208275794983,
      "learning_rate": 2.8694958225971664e-05,
      "loss": 0.5913,
      "step": 1555
    },
    {
      "epoch": 1.99072,
      "grad_norm": 0.5411492586135864,
      "learning_rate": 2.8681113198978984e-05,
      "loss": 0.6312,
      "step": 1556
    },
    {
      "epoch": 1.992,
      "grad_norm": 0.48426690697669983,
      "learning_rate": 2.866726358622028e-05,
      "loss": 0.6176,
      "step": 1557
    },
    {
      "epoch": 1.99328,
      "grad_norm": 0.47467929124832153,
      "learning_rate": 2.8653409397201596e-05,
      "loss": 0.6304,
      "step": 1558
    },
    {
      "epoch": 1.9945599999999999,
      "grad_norm": 0.4745519161224365,
      "learning_rate": 2.8639550641432145e-05,
      "loss": 0.6002,
      "step": 1559
    },
    {
      "epoch": 1.9958399999999998,
      "grad_norm": 0.50393146276474,
      "learning_rate": 2.8625687328424266e-05,
      "loss": 0.629,
      "step": 1560
    },
    {
      "epoch": 1.9958399999999998,
      "eval_loss": 0.680648684501648,
      "eval_runtime": 19.6154,
      "eval_samples_per_second": 25.592,
      "eval_steps_per_second": 3.212,
      "step": 1560
    },
    {
      "epoch": 1.99712,
      "grad_norm": 0.5005529522895813,
      "learning_rate": 2.8611819467693424e-05,
      "loss": 0.619,
      "step": 1561
    },
    {
      "epoch": 1.9984,
      "grad_norm": 0.4954354465007782,
      "learning_rate": 2.8597947068758206e-05,
      "loss": 0.6038,
      "step": 1562
    },
    {
      "epoch": 1.9996800000000001,
      "grad_norm": 0.4475495517253876,
      "learning_rate": 2.858407014114032e-05,
      "loss": 0.6361,
      "step": 1563
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6442052125930786,
      "learning_rate": 2.8570188694364573e-05,
      "loss": 0.5444,
      "step": 1564
    },
    {
      "epoch": 2.00128,
      "grad_norm": 0.6374321579933167,
      "learning_rate": 2.855630273795888e-05,
      "loss": 0.5656,
      "step": 1565
    },
    {
      "epoch": 2.00256,
      "grad_norm": 0.5605729222297668,
      "learning_rate": 2.8542412281454245e-05,
      "loss": 0.5853,
      "step": 1566
    },
    {
      "epoch": 2.00384,
      "grad_norm": 0.5113856792449951,
      "learning_rate": 2.8528517334384775e-05,
      "loss": 0.5361,
      "step": 1567
    },
    {
      "epoch": 2.00512,
      "grad_norm": 0.5127249956130981,
      "learning_rate": 2.8514617906287645e-05,
      "loss": 0.5799,
      "step": 1568
    },
    {
      "epoch": 2.0064,
      "grad_norm": 0.5437144637107849,
      "learning_rate": 2.85007140067031e-05,
      "loss": 0.5832,
      "step": 1569
    },
    {
      "epoch": 2.00768,
      "grad_norm": 0.5008301138877869,
      "learning_rate": 2.8486805645174486e-05,
      "loss": 0.5785,
      "step": 1570
    },
    {
      "epoch": 2.00896,
      "grad_norm": 0.547435998916626,
      "learning_rate": 2.847289283124817e-05,
      "loss": 0.5645,
      "step": 1571
    },
    {
      "epoch": 2.01024,
      "grad_norm": 0.48056700825691223,
      "learning_rate": 2.8458975574473596e-05,
      "loss": 0.5268,
      "step": 1572
    },
    {
      "epoch": 2.01152,
      "grad_norm": 0.493370920419693,
      "learning_rate": 2.8445053884403276e-05,
      "loss": 0.5518,
      "step": 1573
    },
    {
      "epoch": 2.0128,
      "grad_norm": 0.4865136742591858,
      "learning_rate": 2.8431127770592723e-05,
      "loss": 0.562,
      "step": 1574
    },
    {
      "epoch": 2.01408,
      "grad_norm": 0.5080748200416565,
      "learning_rate": 2.841719724260052e-05,
      "loss": 0.5578,
      "step": 1575
    },
    {
      "epoch": 2.01536,
      "grad_norm": 0.5026775002479553,
      "learning_rate": 2.840326230998827e-05,
      "loss": 0.5542,
      "step": 1576
    },
    {
      "epoch": 2.01664,
      "grad_norm": 0.5246357321739197,
      "learning_rate": 2.8389322982320592e-05,
      "loss": 0.5764,
      "step": 1577
    },
    {
      "epoch": 2.01792,
      "grad_norm": 0.494767963886261,
      "learning_rate": 2.8375379269165128e-05,
      "loss": 0.5678,
      "step": 1578
    },
    {
      "epoch": 2.0192,
      "grad_norm": 0.46353206038475037,
      "learning_rate": 2.8361431180092534e-05,
      "loss": 0.5489,
      "step": 1579
    },
    {
      "epoch": 2.02048,
      "grad_norm": 0.4915197193622589,
      "learning_rate": 2.8347478724676462e-05,
      "loss": 0.5538,
      "step": 1580
    },
    {
      "epoch": 2.02176,
      "grad_norm": 0.45937174558639526,
      "learning_rate": 2.8333521912493557e-05,
      "loss": 0.582,
      "step": 1581
    },
    {
      "epoch": 2.02304,
      "grad_norm": 0.48977726697921753,
      "learning_rate": 2.831956075312348e-05,
      "loss": 0.5802,
      "step": 1582
    },
    {
      "epoch": 2.02432,
      "grad_norm": 0.48102542757987976,
      "learning_rate": 2.830559525614884e-05,
      "loss": 0.5818,
      "step": 1583
    },
    {
      "epoch": 2.0256,
      "grad_norm": 0.4348898231983185,
      "learning_rate": 2.8291625431155244e-05,
      "loss": 0.5722,
      "step": 1584
    },
    {
      "epoch": 2.02688,
      "grad_norm": 0.5017533302307129,
      "learning_rate": 2.8277651287731273e-05,
      "loss": 0.5741,
      "step": 1585
    },
    {
      "epoch": 2.02816,
      "grad_norm": 0.45297515392303467,
      "learning_rate": 2.8263672835468456e-05,
      "loss": 0.5905,
      "step": 1586
    },
    {
      "epoch": 2.02944,
      "grad_norm": 0.5268329977989197,
      "learning_rate": 2.82496900839613e-05,
      "loss": 0.571,
      "step": 1587
    },
    {
      "epoch": 2.03072,
      "grad_norm": 0.4581129252910614,
      "learning_rate": 2.8235703042807246e-05,
      "loss": 0.5604,
      "step": 1588
    },
    {
      "epoch": 2.032,
      "grad_norm": 0.5102109313011169,
      "learning_rate": 2.822171172160667e-05,
      "loss": 0.5468,
      "step": 1589
    },
    {
      "epoch": 2.03328,
      "grad_norm": 0.5498456954956055,
      "learning_rate": 2.8207716129962926e-05,
      "loss": 0.5839,
      "step": 1590
    },
    {
      "epoch": 2.03328,
      "eval_loss": 0.6889892816543579,
      "eval_runtime": 19.5877,
      "eval_samples_per_second": 25.628,
      "eval_steps_per_second": 3.216,
      "step": 1590
    },
    {
      "epoch": 2.03456,
      "grad_norm": 0.46910062432289124,
      "learning_rate": 2.819371627748226e-05,
      "loss": 0.5501,
      "step": 1591
    },
    {
      "epoch": 2.03584,
      "grad_norm": 0.6276616454124451,
      "learning_rate": 2.817971217377386e-05,
      "loss": 0.6124,
      "step": 1592
    },
    {
      "epoch": 2.03712,
      "grad_norm": 0.45304033160209656,
      "learning_rate": 2.8165703828449826e-05,
      "loss": 0.5843,
      "step": 1593
    },
    {
      "epoch": 2.0384,
      "grad_norm": 0.4703642427921295,
      "learning_rate": 2.815169125112517e-05,
      "loss": 0.5706,
      "step": 1594
    },
    {
      "epoch": 2.03968,
      "grad_norm": 0.5255608558654785,
      "learning_rate": 2.813767445141781e-05,
      "loss": 0.5721,
      "step": 1595
    },
    {
      "epoch": 2.04096,
      "grad_norm": 0.4225960373878479,
      "learning_rate": 2.812365343894857e-05,
      "loss": 0.5384,
      "step": 1596
    },
    {
      "epoch": 2.04224,
      "grad_norm": 0.5019941926002502,
      "learning_rate": 2.810962822334115e-05,
      "loss": 0.5614,
      "step": 1597
    },
    {
      "epoch": 2.04352,
      "grad_norm": 0.47851160168647766,
      "learning_rate": 2.8095598814222147e-05,
      "loss": 0.5701,
      "step": 1598
    },
    {
      "epoch": 2.0448,
      "grad_norm": 0.4844616949558258,
      "learning_rate": 2.808156522122103e-05,
      "loss": 0.5664,
      "step": 1599
    },
    {
      "epoch": 2.04608,
      "grad_norm": 0.49759793281555176,
      "learning_rate": 2.806752745397015e-05,
      "loss": 0.632,
      "step": 1600
    },
    {
      "epoch": 2.04736,
      "grad_norm": 0.4634726941585541,
      "learning_rate": 2.8053485522104706e-05,
      "loss": 0.5388,
      "step": 1601
    },
    {
      "epoch": 2.04864,
      "grad_norm": 0.49071577191352844,
      "learning_rate": 2.803943943526276e-05,
      "loss": 0.5601,
      "step": 1602
    },
    {
      "epoch": 2.04992,
      "grad_norm": 0.46740227937698364,
      "learning_rate": 2.802538920308524e-05,
      "loss": 0.5711,
      "step": 1603
    },
    {
      "epoch": 2.0512,
      "grad_norm": 0.46447139978408813,
      "learning_rate": 2.801133483521591e-05,
      "loss": 0.5612,
      "step": 1604
    },
    {
      "epoch": 2.05248,
      "grad_norm": 0.4782780408859253,
      "learning_rate": 2.799727634130137e-05,
      "loss": 0.5895,
      "step": 1605
    },
    {
      "epoch": 2.05376,
      "grad_norm": 0.5044755339622498,
      "learning_rate": 2.7983213730991055e-05,
      "loss": 0.5855,
      "step": 1606
    },
    {
      "epoch": 2.05504,
      "grad_norm": 0.5155549645423889,
      "learning_rate": 2.7969147013937227e-05,
      "loss": 0.5639,
      "step": 1607
    },
    {
      "epoch": 2.05632,
      "grad_norm": 0.4956282079219818,
      "learning_rate": 2.7955076199794954e-05,
      "loss": 0.5876,
      "step": 1608
    },
    {
      "epoch": 2.0576,
      "grad_norm": 0.5118156671524048,
      "learning_rate": 2.794100129822214e-05,
      "loss": 0.5723,
      "step": 1609
    },
    {
      "epoch": 2.05888,
      "grad_norm": 0.4918240010738373,
      "learning_rate": 2.7926922318879474e-05,
      "loss": 0.5536,
      "step": 1610
    },
    {
      "epoch": 2.06016,
      "grad_norm": 0.5174325108528137,
      "learning_rate": 2.791283927143045e-05,
      "loss": 0.5565,
      "step": 1611
    },
    {
      "epoch": 2.06144,
      "grad_norm": 0.49884384870529175,
      "learning_rate": 2.7898752165541365e-05,
      "loss": 0.5848,
      "step": 1612
    },
    {
      "epoch": 2.06272,
      "grad_norm": 0.5088360905647278,
      "learning_rate": 2.7884661010881284e-05,
      "loss": 0.5771,
      "step": 1613
    },
    {
      "epoch": 2.064,
      "grad_norm": 0.49027571082115173,
      "learning_rate": 2.7870565817122054e-05,
      "loss": 0.5543,
      "step": 1614
    },
    {
      "epoch": 2.06528,
      "grad_norm": 0.4648645520210266,
      "learning_rate": 2.785646659393831e-05,
      "loss": 0.5623,
      "step": 1615
    },
    {
      "epoch": 2.06656,
      "grad_norm": 0.5602195858955383,
      "learning_rate": 2.7842363351007443e-05,
      "loss": 0.5819,
      "step": 1616
    },
    {
      "epoch": 2.06784,
      "grad_norm": 0.5447742342948914,
      "learning_rate": 2.7828256098009595e-05,
      "loss": 0.5984,
      "step": 1617
    },
    {
      "epoch": 2.06912,
      "grad_norm": 0.5072308778762817,
      "learning_rate": 2.7814144844627664e-05,
      "loss": 0.5798,
      "step": 1618
    },
    {
      "epoch": 2.0704,
      "grad_norm": 0.5437843799591064,
      "learning_rate": 2.7800029600547312e-05,
      "loss": 0.5875,
      "step": 1619
    },
    {
      "epoch": 2.07168,
      "grad_norm": 0.4389694333076477,
      "learning_rate": 2.7785910375456914e-05,
      "loss": 0.5536,
      "step": 1620
    },
    {
      "epoch": 2.07168,
      "eval_loss": 0.6892576813697815,
      "eval_runtime": 19.5759,
      "eval_samples_per_second": 25.644,
      "eval_steps_per_second": 3.218,
      "step": 1620
    },
    {
      "epoch": 2.07296,
      "grad_norm": 0.5112296342849731,
      "learning_rate": 2.7771787179047588e-05,
      "loss": 0.5509,
      "step": 1621
    },
    {
      "epoch": 2.07424,
      "grad_norm": 0.42263558506965637,
      "learning_rate": 2.7757660021013184e-05,
      "loss": 0.5501,
      "step": 1622
    },
    {
      "epoch": 2.07552,
      "grad_norm": 0.5040010809898376,
      "learning_rate": 2.7743528911050267e-05,
      "loss": 0.5666,
      "step": 1623
    },
    {
      "epoch": 2.0768,
      "grad_norm": 0.4713045358657837,
      "learning_rate": 2.7729393858858107e-05,
      "loss": 0.5608,
      "step": 1624
    },
    {
      "epoch": 2.07808,
      "grad_norm": 0.49158063530921936,
      "learning_rate": 2.7715254874138696e-05,
      "loss": 0.5848,
      "step": 1625
    },
    {
      "epoch": 2.07936,
      "grad_norm": 0.5084195137023926,
      "learning_rate": 2.7701111966596704e-05,
      "loss": 0.6099,
      "step": 1626
    },
    {
      "epoch": 2.08064,
      "grad_norm": 0.44276511669158936,
      "learning_rate": 2.768696514593952e-05,
      "loss": 0.5551,
      "step": 1627
    },
    {
      "epoch": 2.08192,
      "grad_norm": 0.467803418636322,
      "learning_rate": 2.7672814421877182e-05,
      "loss": 0.5543,
      "step": 1628
    },
    {
      "epoch": 2.0832,
      "grad_norm": 0.4726732075214386,
      "learning_rate": 2.7658659804122452e-05,
      "loss": 0.5858,
      "step": 1629
    },
    {
      "epoch": 2.08448,
      "grad_norm": 0.4821057915687561,
      "learning_rate": 2.7644501302390727e-05,
      "loss": 0.5486,
      "step": 1630
    },
    {
      "epoch": 2.08576,
      "grad_norm": 0.5125967860221863,
      "learning_rate": 2.7630338926400093e-05,
      "loss": 0.5647,
      "step": 1631
    },
    {
      "epoch": 2.08704,
      "grad_norm": 0.47355136275291443,
      "learning_rate": 2.761617268587128e-05,
      "loss": 0.609,
      "step": 1632
    },
    {
      "epoch": 2.08832,
      "grad_norm": 0.48122119903564453,
      "learning_rate": 2.7602002590527684e-05,
      "loss": 0.5904,
      "step": 1633
    },
    {
      "epoch": 2.0896,
      "grad_norm": 0.4593060314655304,
      "learning_rate": 2.758782865009533e-05,
      "loss": 0.5508,
      "step": 1634
    },
    {
      "epoch": 2.09088,
      "grad_norm": 0.46918314695358276,
      "learning_rate": 2.7573650874302905e-05,
      "loss": 0.5894,
      "step": 1635
    },
    {
      "epoch": 2.09216,
      "grad_norm": 0.4836597740650177,
      "learning_rate": 2.7559469272881708e-05,
      "loss": 0.591,
      "step": 1636
    },
    {
      "epoch": 2.09344,
      "grad_norm": 0.4591413140296936,
      "learning_rate": 2.754528385556567e-05,
      "loss": 0.5516,
      "step": 1637
    },
    {
      "epoch": 2.09472,
      "grad_norm": 0.4661344885826111,
      "learning_rate": 2.7531094632091352e-05,
      "loss": 0.5399,
      "step": 1638
    },
    {
      "epoch": 2.096,
      "grad_norm": 0.49129122495651245,
      "learning_rate": 2.751690161219791e-05,
      "loss": 0.5757,
      "step": 1639
    },
    {
      "epoch": 2.09728,
      "grad_norm": 0.49361559748649597,
      "learning_rate": 2.750270480562712e-05,
      "loss": 0.5776,
      "step": 1640
    },
    {
      "epoch": 2.09856,
      "grad_norm": 0.44980013370513916,
      "learning_rate": 2.7488504222123342e-05,
      "loss": 0.5749,
      "step": 1641
    },
    {
      "epoch": 2.09984,
      "grad_norm": 0.5004571676254272,
      "learning_rate": 2.7474299871433543e-05,
      "loss": 0.5859,
      "step": 1642
    },
    {
      "epoch": 2.10112,
      "grad_norm": 0.4927177429199219,
      "learning_rate": 2.7460091763307275e-05,
      "loss": 0.548,
      "step": 1643
    },
    {
      "epoch": 2.1024,
      "grad_norm": 0.43971818685531616,
      "learning_rate": 2.7445879907496656e-05,
      "loss": 0.5727,
      "step": 1644
    },
    {
      "epoch": 2.1036799999999998,
      "grad_norm": 0.4913136661052704,
      "learning_rate": 2.7431664313756394e-05,
      "loss": 0.5769,
      "step": 1645
    },
    {
      "epoch": 2.10496,
      "grad_norm": 0.48149263858795166,
      "learning_rate": 2.7417444991843746e-05,
      "loss": 0.5902,
      "step": 1646
    },
    {
      "epoch": 2.10624,
      "grad_norm": 0.4909886121749878,
      "learning_rate": 2.7403221951518533e-05,
      "loss": 0.5745,
      "step": 1647
    },
    {
      "epoch": 2.10752,
      "grad_norm": 0.4584827125072479,
      "learning_rate": 2.738899520254314e-05,
      "loss": 0.562,
      "step": 1648
    },
    {
      "epoch": 2.1088,
      "grad_norm": 0.4532663822174072,
      "learning_rate": 2.7374764754682482e-05,
      "loss": 0.6013,
      "step": 1649
    },
    {
      "epoch": 2.11008,
      "grad_norm": 0.4460378587245941,
      "learning_rate": 2.736053061770401e-05,
      "loss": 0.5805,
      "step": 1650
    },
    {
      "epoch": 2.11008,
      "eval_loss": 0.6897958517074585,
      "eval_runtime": 19.5769,
      "eval_samples_per_second": 25.642,
      "eval_steps_per_second": 3.218,
      "step": 1650
    },
    {
      "epoch": 2.11136,
      "grad_norm": 0.4545774757862091,
      "learning_rate": 2.7346292801377737e-05,
      "loss": 0.5858,
      "step": 1651
    },
    {
      "epoch": 2.11264,
      "grad_norm": 0.42635059356689453,
      "learning_rate": 2.7332051315476166e-05,
      "loss": 0.5485,
      "step": 1652
    },
    {
      "epoch": 2.11392,
      "grad_norm": 0.4749841094017029,
      "learning_rate": 2.7317806169774325e-05,
      "loss": 0.5758,
      "step": 1653
    },
    {
      "epoch": 2.1152,
      "grad_norm": 0.46972087025642395,
      "learning_rate": 2.730355737404978e-05,
      "loss": 0.5893,
      "step": 1654
    },
    {
      "epoch": 2.11648,
      "grad_norm": 0.44887983798980713,
      "learning_rate": 2.7289304938082583e-05,
      "loss": 0.5828,
      "step": 1655
    },
    {
      "epoch": 2.11776,
      "grad_norm": 0.5040025115013123,
      "learning_rate": 2.727504887165527e-05,
      "loss": 0.5406,
      "step": 1656
    },
    {
      "epoch": 2.11904,
      "grad_norm": 0.4109235107898712,
      "learning_rate": 2.7260789184552902e-05,
      "loss": 0.5595,
      "step": 1657
    },
    {
      "epoch": 2.12032,
      "grad_norm": 0.47293850779533386,
      "learning_rate": 2.7246525886562995e-05,
      "loss": 0.5782,
      "step": 1658
    },
    {
      "epoch": 2.1216,
      "grad_norm": 0.500037431716919,
      "learning_rate": 2.723225898747556e-05,
      "loss": 0.566,
      "step": 1659
    },
    {
      "epoch": 2.12288,
      "grad_norm": 0.4618556797504425,
      "learning_rate": 2.721798849708308e-05,
      "loss": 0.5551,
      "step": 1660
    },
    {
      "epoch": 2.12416,
      "grad_norm": 0.5106038451194763,
      "learning_rate": 2.7203714425180497e-05,
      "loss": 0.6139,
      "step": 1661
    },
    {
      "epoch": 2.12544,
      "grad_norm": 0.5301266312599182,
      "learning_rate": 2.7189436781565218e-05,
      "loss": 0.5958,
      "step": 1662
    },
    {
      "epoch": 2.12672,
      "grad_norm": 0.46651196479797363,
      "learning_rate": 2.717515557603709e-05,
      "loss": 0.5486,
      "step": 1663
    },
    {
      "epoch": 2.128,
      "grad_norm": 0.5061004757881165,
      "learning_rate": 2.716087081839841e-05,
      "loss": 0.5677,
      "step": 1664
    },
    {
      "epoch": 2.12928,
      "grad_norm": 0.5145121216773987,
      "learning_rate": 2.7146582518453918e-05,
      "loss": 0.5568,
      "step": 1665
    },
    {
      "epoch": 2.13056,
      "grad_norm": 0.48494312167167664,
      "learning_rate": 2.7132290686010795e-05,
      "loss": 0.5669,
      "step": 1666
    },
    {
      "epoch": 2.13184,
      "grad_norm": 0.4836319386959076,
      "learning_rate": 2.7117995330878604e-05,
      "loss": 0.5544,
      "step": 1667
    },
    {
      "epoch": 2.13312,
      "grad_norm": 0.5250100493431091,
      "learning_rate": 2.7103696462869386e-05,
      "loss": 0.5754,
      "step": 1668
    },
    {
      "epoch": 2.1344,
      "grad_norm": 0.4753483235836029,
      "learning_rate": 2.7089394091797547e-05,
      "loss": 0.581,
      "step": 1669
    },
    {
      "epoch": 2.13568,
      "grad_norm": 0.52092045545578,
      "learning_rate": 2.7075088227479912e-05,
      "loss": 0.5861,
      "step": 1670
    },
    {
      "epoch": 2.13696,
      "grad_norm": 0.5094035267829895,
      "learning_rate": 2.7060778879735714e-05,
      "loss": 0.561,
      "step": 1671
    },
    {
      "epoch": 2.13824,
      "grad_norm": 0.47683224081993103,
      "learning_rate": 2.7046466058386567e-05,
      "loss": 0.5755,
      "step": 1672
    },
    {
      "epoch": 2.13952,
      "grad_norm": 0.5220264792442322,
      "learning_rate": 2.7032149773256454e-05,
      "loss": 0.5871,
      "step": 1673
    },
    {
      "epoch": 2.1408,
      "grad_norm": 0.497913658618927,
      "learning_rate": 2.7017830034171772e-05,
      "loss": 0.6103,
      "step": 1674
    },
    {
      "epoch": 2.14208,
      "grad_norm": 0.4961109757423401,
      "learning_rate": 2.7003506850961262e-05,
      "loss": 0.5547,
      "step": 1675
    },
    {
      "epoch": 2.14336,
      "grad_norm": 0.5179574489593506,
      "learning_rate": 2.6989180233456034e-05,
      "loss": 0.5998,
      "step": 1676
    },
    {
      "epoch": 2.14464,
      "grad_norm": 0.44227835536003113,
      "learning_rate": 2.697485019148955e-05,
      "loss": 0.5528,
      "step": 1677
    },
    {
      "epoch": 2.14592,
      "grad_norm": 0.4586826264858246,
      "learning_rate": 2.6960516734897638e-05,
      "loss": 0.5469,
      "step": 1678
    },
    {
      "epoch": 2.1471999999999998,
      "grad_norm": 0.4850342571735382,
      "learning_rate": 2.6946179873518453e-05,
      "loss": 0.5651,
      "step": 1679
    },
    {
      "epoch": 2.14848,
      "grad_norm": 0.4509904980659485,
      "learning_rate": 2.6931839617192507e-05,
      "loss": 0.5557,
      "step": 1680
    },
    {
      "epoch": 2.14848,
      "eval_loss": 0.6887514591217041,
      "eval_runtime": 19.5737,
      "eval_samples_per_second": 25.647,
      "eval_steps_per_second": 3.219,
      "step": 1680
    },
    {
      "epoch": 2.14976,
      "grad_norm": 0.4966086447238922,
      "learning_rate": 2.6917495975762623e-05,
      "loss": 0.5907,
      "step": 1681
    },
    {
      "epoch": 2.15104,
      "grad_norm": 0.4586782455444336,
      "learning_rate": 2.6903148959073947e-05,
      "loss": 0.5943,
      "step": 1682
    },
    {
      "epoch": 2.15232,
      "grad_norm": 0.4662559926509857,
      "learning_rate": 2.6888798576973955e-05,
      "loss": 0.5872,
      "step": 1683
    },
    {
      "epoch": 2.1536,
      "grad_norm": 0.47413530945777893,
      "learning_rate": 2.6874444839312427e-05,
      "loss": 0.5575,
      "step": 1684
    },
    {
      "epoch": 2.15488,
      "grad_norm": 0.4715486168861389,
      "learning_rate": 2.6860087755941454e-05,
      "loss": 0.5904,
      "step": 1685
    },
    {
      "epoch": 2.15616,
      "grad_norm": 0.4761320650577545,
      "learning_rate": 2.6845727336715393e-05,
      "loss": 0.5826,
      "step": 1686
    },
    {
      "epoch": 2.15744,
      "grad_norm": 0.5202406048774719,
      "learning_rate": 2.6831363591490945e-05,
      "loss": 0.5759,
      "step": 1687
    },
    {
      "epoch": 2.15872,
      "grad_norm": 0.4305674135684967,
      "learning_rate": 2.681699653012704e-05,
      "loss": 0.5737,
      "step": 1688
    },
    {
      "epoch": 2.16,
      "grad_norm": 0.504291832447052,
      "learning_rate": 2.68026261624849e-05,
      "loss": 0.5843,
      "step": 1689
    },
    {
      "epoch": 2.16128,
      "grad_norm": 0.44171032309532166,
      "learning_rate": 2.6788252498428048e-05,
      "loss": 0.5668,
      "step": 1690
    },
    {
      "epoch": 2.16256,
      "grad_norm": 0.4176311492919922,
      "learning_rate": 2.677387554782222e-05,
      "loss": 0.5755,
      "step": 1691
    },
    {
      "epoch": 2.16384,
      "grad_norm": 0.4424145519733429,
      "learning_rate": 2.6759495320535448e-05,
      "loss": 0.586,
      "step": 1692
    },
    {
      "epoch": 2.16512,
      "grad_norm": 0.45824339985847473,
      "learning_rate": 2.6745111826437987e-05,
      "loss": 0.5432,
      "step": 1693
    },
    {
      "epoch": 2.1664,
      "grad_norm": 0.45858070254325867,
      "learning_rate": 2.6730725075402358e-05,
      "loss": 0.598,
      "step": 1694
    },
    {
      "epoch": 2.16768,
      "grad_norm": 0.43888765573501587,
      "learning_rate": 2.671633507730329e-05,
      "loss": 0.5938,
      "step": 1695
    },
    {
      "epoch": 2.16896,
      "grad_norm": 0.5001144409179688,
      "learning_rate": 2.6701941842017755e-05,
      "loss": 0.5766,
      "step": 1696
    },
    {
      "epoch": 2.17024,
      "grad_norm": 0.48221108317375183,
      "learning_rate": 2.6687545379424954e-05,
      "loss": 0.5817,
      "step": 1697
    },
    {
      "epoch": 2.17152,
      "grad_norm": 0.4804971218109131,
      "learning_rate": 2.667314569940629e-05,
      "loss": 0.57,
      "step": 1698
    },
    {
      "epoch": 2.1728,
      "grad_norm": 0.49444660544395447,
      "learning_rate": 2.6658742811845377e-05,
      "loss": 0.5859,
      "step": 1699
    },
    {
      "epoch": 2.17408,
      "grad_norm": 0.4761796295642853,
      "learning_rate": 2.664433672662804e-05,
      "loss": 0.5489,
      "step": 1700
    },
    {
      "epoch": 2.17536,
      "grad_norm": 0.44284963607788086,
      "learning_rate": 2.6629927453642293e-05,
      "loss": 0.5825,
      "step": 1701
    },
    {
      "epoch": 2.17664,
      "grad_norm": 0.47245386242866516,
      "learning_rate": 2.661551500277833e-05,
      "loss": 0.5571,
      "step": 1702
    },
    {
      "epoch": 2.17792,
      "grad_norm": 0.47504687309265137,
      "learning_rate": 2.6601099383928545e-05,
      "loss": 0.5843,
      "step": 1703
    },
    {
      "epoch": 2.1792,
      "grad_norm": 0.4557231366634369,
      "learning_rate": 2.6586680606987485e-05,
      "loss": 0.5732,
      "step": 1704
    },
    {
      "epoch": 2.18048,
      "grad_norm": 0.4854606091976166,
      "learning_rate": 2.6572258681851878e-05,
      "loss": 0.5876,
      "step": 1705
    },
    {
      "epoch": 2.18176,
      "grad_norm": 0.43298855423927307,
      "learning_rate": 2.6557833618420614e-05,
      "loss": 0.5391,
      "step": 1706
    },
    {
      "epoch": 2.18304,
      "grad_norm": 0.40423235297203064,
      "learning_rate": 2.6543405426594725e-05,
      "loss": 0.5497,
      "step": 1707
    },
    {
      "epoch": 2.18432,
      "grad_norm": 0.4659039378166199,
      "learning_rate": 2.65289741162774e-05,
      "loss": 0.5599,
      "step": 1708
    },
    {
      "epoch": 2.1856,
      "grad_norm": 0.43222808837890625,
      "learning_rate": 2.6514539697373964e-05,
      "loss": 0.5408,
      "step": 1709
    },
    {
      "epoch": 2.18688,
      "grad_norm": 0.46446728706359863,
      "learning_rate": 2.6500102179791887e-05,
      "loss": 0.5691,
      "step": 1710
    },
    {
      "epoch": 2.18688,
      "eval_loss": 0.6889845132827759,
      "eval_runtime": 19.6262,
      "eval_samples_per_second": 25.578,
      "eval_steps_per_second": 3.21,
      "step": 1710
    },
    {
      "epoch": 2.18816,
      "grad_norm": 0.4668131470680237,
      "learning_rate": 2.6485661573440746e-05,
      "loss": 0.5634,
      "step": 1711
    },
    {
      "epoch": 2.18944,
      "grad_norm": 0.5059604644775391,
      "learning_rate": 2.6471217888232255e-05,
      "loss": 0.5896,
      "step": 1712
    },
    {
      "epoch": 2.19072,
      "grad_norm": 0.461904913187027,
      "learning_rate": 2.645677113408023e-05,
      "loss": 0.5835,
      "step": 1713
    },
    {
      "epoch": 2.192,
      "grad_norm": 0.4850727915763855,
      "learning_rate": 2.644232132090061e-05,
      "loss": 0.5374,
      "step": 1714
    },
    {
      "epoch": 2.19328,
      "grad_norm": 0.44838401675224304,
      "learning_rate": 2.6427868458611404e-05,
      "loss": 0.59,
      "step": 1715
    },
    {
      "epoch": 2.19456,
      "grad_norm": 0.44102972745895386,
      "learning_rate": 2.6413412557132753e-05,
      "loss": 0.5629,
      "step": 1716
    },
    {
      "epoch": 2.19584,
      "grad_norm": 0.4629192352294922,
      "learning_rate": 2.6398953626386852e-05,
      "loss": 0.5471,
      "step": 1717
    },
    {
      "epoch": 2.19712,
      "grad_norm": 0.4437639117240906,
      "learning_rate": 2.638449167629798e-05,
      "loss": 0.5718,
      "step": 1718
    },
    {
      "epoch": 2.1984,
      "grad_norm": 0.4644782543182373,
      "learning_rate": 2.6370026716792505e-05,
      "loss": 0.5887,
      "step": 1719
    },
    {
      "epoch": 2.19968,
      "grad_norm": 0.5009134411811829,
      "learning_rate": 2.6355558757798843e-05,
      "loss": 0.5787,
      "step": 1720
    },
    {
      "epoch": 2.20096,
      "grad_norm": 0.42760053277015686,
      "learning_rate": 2.634108780924749e-05,
      "loss": 0.5639,
      "step": 1721
    },
    {
      "epoch": 2.20224,
      "grad_norm": 0.47184401750564575,
      "learning_rate": 2.6326613881070958e-05,
      "loss": 0.5991,
      "step": 1722
    },
    {
      "epoch": 2.20352,
      "grad_norm": 0.46054160594940186,
      "learning_rate": 2.6312136983203848e-05,
      "loss": 0.5831,
      "step": 1723
    },
    {
      "epoch": 2.2048,
      "grad_norm": 0.4219911992549896,
      "learning_rate": 2.6297657125582765e-05,
      "loss": 0.5662,
      "step": 1724
    },
    {
      "epoch": 2.20608,
      "grad_norm": 0.4911462962627411,
      "learning_rate": 2.6283174318146352e-05,
      "loss": 0.5918,
      "step": 1725
    },
    {
      "epoch": 2.20736,
      "grad_norm": 0.46021342277526855,
      "learning_rate": 2.6268688570835302e-05,
      "loss": 0.585,
      "step": 1726
    },
    {
      "epoch": 2.20864,
      "grad_norm": 0.44681957364082336,
      "learning_rate": 2.6254199893592293e-05,
      "loss": 0.5665,
      "step": 1727
    },
    {
      "epoch": 2.20992,
      "grad_norm": 0.439086377620697,
      "learning_rate": 2.623970829636203e-05,
      "loss": 0.5622,
      "step": 1728
    },
    {
      "epoch": 2.2112,
      "grad_norm": 0.4608093202114105,
      "learning_rate": 2.6225213789091218e-05,
      "loss": 0.563,
      "step": 1729
    },
    {
      "epoch": 2.2124800000000002,
      "grad_norm": 0.4593770205974579,
      "learning_rate": 2.621071638172857e-05,
      "loss": 0.5564,
      "step": 1730
    },
    {
      "epoch": 2.21376,
      "grad_norm": 0.4569006562232971,
      "learning_rate": 2.619621608422477e-05,
      "loss": 0.5615,
      "step": 1731
    },
    {
      "epoch": 2.21504,
      "grad_norm": 0.5438253283500671,
      "learning_rate": 2.6181712906532502e-05,
      "loss": 0.5846,
      "step": 1732
    },
    {
      "epoch": 2.21632,
      "grad_norm": 0.4294547140598297,
      "learning_rate": 2.616720685860642e-05,
      "loss": 0.5817,
      "step": 1733
    },
    {
      "epoch": 2.2176,
      "grad_norm": 0.611320436000824,
      "learning_rate": 2.615269795040315e-05,
      "loss": 0.5402,
      "step": 1734
    },
    {
      "epoch": 2.21888,
      "grad_norm": 0.5203279256820679,
      "learning_rate": 2.613818619188129e-05,
      "loss": 0.5582,
      "step": 1735
    },
    {
      "epoch": 2.22016,
      "grad_norm": 0.5088427662849426,
      "learning_rate": 2.612367159300137e-05,
      "loss": 0.5943,
      "step": 1736
    },
    {
      "epoch": 2.22144,
      "grad_norm": 0.5930956602096558,
      "learning_rate": 2.610915416372588e-05,
      "loss": 0.5772,
      "step": 1737
    },
    {
      "epoch": 2.22272,
      "grad_norm": 0.4942617118358612,
      "learning_rate": 2.6094633914019277e-05,
      "loss": 0.5858,
      "step": 1738
    },
    {
      "epoch": 2.224,
      "grad_norm": 0.45109954476356506,
      "learning_rate": 2.6080110853847922e-05,
      "loss": 0.5628,
      "step": 1739
    },
    {
      "epoch": 2.22528,
      "grad_norm": 0.48240992426872253,
      "learning_rate": 2.6065584993180123e-05,
      "loss": 0.5516,
      "step": 1740
    },
    {
      "epoch": 2.22528,
      "eval_loss": 0.6892181038856506,
      "eval_runtime": 19.5995,
      "eval_samples_per_second": 25.613,
      "eval_steps_per_second": 3.214,
      "step": 1740
    },
    {
      "epoch": 2.22656,
      "grad_norm": 0.44723188877105713,
      "learning_rate": 2.60510563419861e-05,
      "loss": 0.565,
      "step": 1741
    },
    {
      "epoch": 2.22784,
      "grad_norm": 0.49325910210609436,
      "learning_rate": 2.6036524910237993e-05,
      "loss": 0.5609,
      "step": 1742
    },
    {
      "epoch": 2.22912,
      "grad_norm": 0.4392958879470825,
      "learning_rate": 2.6021990707909858e-05,
      "loss": 0.5701,
      "step": 1743
    },
    {
      "epoch": 2.2304,
      "grad_norm": 0.44813546538352966,
      "learning_rate": 2.600745374497764e-05,
      "loss": 0.5797,
      "step": 1744
    },
    {
      "epoch": 2.23168,
      "grad_norm": 0.4627484381198883,
      "learning_rate": 2.5992914031419174e-05,
      "loss": 0.5748,
      "step": 1745
    },
    {
      "epoch": 2.23296,
      "grad_norm": 0.4794831871986389,
      "learning_rate": 2.5978371577214213e-05,
      "loss": 0.5601,
      "step": 1746
    },
    {
      "epoch": 2.23424,
      "grad_norm": 0.44142457842826843,
      "learning_rate": 2.596382639234435e-05,
      "loss": 0.5528,
      "step": 1747
    },
    {
      "epoch": 2.23552,
      "grad_norm": 0.44726330041885376,
      "learning_rate": 2.5949278486793086e-05,
      "loss": 0.5882,
      "step": 1748
    },
    {
      "epoch": 2.2368,
      "grad_norm": 0.4589160084724426,
      "learning_rate": 2.593472787054578e-05,
      "loss": 0.5338,
      "step": 1749
    },
    {
      "epoch": 2.23808,
      "grad_norm": 0.492593377828598,
      "learning_rate": 2.5920174553589633e-05,
      "loss": 0.6025,
      "step": 1750
    },
    {
      "epoch": 2.23936,
      "grad_norm": 0.48570048809051514,
      "learning_rate": 2.5905618545913728e-05,
      "loss": 0.5687,
      "step": 1751
    },
    {
      "epoch": 2.24064,
      "grad_norm": 0.4951373040676117,
      "learning_rate": 2.5891059857508974e-05,
      "loss": 0.5836,
      "step": 1752
    },
    {
      "epoch": 2.24192,
      "grad_norm": 0.44068455696105957,
      "learning_rate": 2.5876498498368135e-05,
      "loss": 0.5672,
      "step": 1753
    },
    {
      "epoch": 2.2432,
      "grad_norm": 0.49540549516677856,
      "learning_rate": 2.586193447848579e-05,
      "loss": 0.6101,
      "step": 1754
    },
    {
      "epoch": 2.24448,
      "grad_norm": 0.4484197795391083,
      "learning_rate": 2.584736780785837e-05,
      "loss": 0.5762,
      "step": 1755
    },
    {
      "epoch": 2.24576,
      "grad_norm": 0.47352614998817444,
      "learning_rate": 2.5832798496484102e-05,
      "loss": 0.5579,
      "step": 1756
    },
    {
      "epoch": 2.24704,
      "grad_norm": 0.4246540069580078,
      "learning_rate": 2.5818226554363027e-05,
      "loss": 0.5684,
      "step": 1757
    },
    {
      "epoch": 2.24832,
      "grad_norm": 0.490526020526886,
      "learning_rate": 2.580365199149701e-05,
      "loss": 0.5715,
      "step": 1758
    },
    {
      "epoch": 2.2496,
      "grad_norm": 0.503845751285553,
      "learning_rate": 2.5789074817889704e-05,
      "loss": 0.582,
      "step": 1759
    },
    {
      "epoch": 2.25088,
      "grad_norm": 0.4746897220611572,
      "learning_rate": 2.5774495043546548e-05,
      "loss": 0.5913,
      "step": 1760
    },
    {
      "epoch": 2.25216,
      "grad_norm": 0.4539012610912323,
      "learning_rate": 2.575991267847477e-05,
      "loss": 0.5639,
      "step": 1761
    },
    {
      "epoch": 2.25344,
      "grad_norm": 0.41014665365219116,
      "learning_rate": 2.5745327732683384e-05,
      "loss": 0.5675,
      "step": 1762
    },
    {
      "epoch": 2.25472,
      "grad_norm": 0.4564160704612732,
      "learning_rate": 2.5730740216183168e-05,
      "loss": 0.5711,
      "step": 1763
    },
    {
      "epoch": 2.2560000000000002,
      "grad_norm": 0.4823758900165558,
      "learning_rate": 2.5716150138986654e-05,
      "loss": 0.5721,
      "step": 1764
    },
    {
      "epoch": 2.25728,
      "grad_norm": 0.4670538306236267,
      "learning_rate": 2.570155751110816e-05,
      "loss": 0.5843,
      "step": 1765
    },
    {
      "epoch": 2.25856,
      "grad_norm": 0.4602479338645935,
      "learning_rate": 2.568696234256373e-05,
      "loss": 0.5758,
      "step": 1766
    },
    {
      "epoch": 2.25984,
      "grad_norm": 0.43574029207229614,
      "learning_rate": 2.567236464337117e-05,
      "loss": 0.5579,
      "step": 1767
    },
    {
      "epoch": 2.26112,
      "grad_norm": 0.45511406660079956,
      "learning_rate": 2.5657764423549994e-05,
      "loss": 0.5705,
      "step": 1768
    },
    {
      "epoch": 2.2624,
      "grad_norm": 0.44675129652023315,
      "learning_rate": 2.5643161693121476e-05,
      "loss": 0.5571,
      "step": 1769
    },
    {
      "epoch": 2.26368,
      "grad_norm": 0.42243945598602295,
      "learning_rate": 2.56285564621086e-05,
      "loss": 0.5813,
      "step": 1770
    },
    {
      "epoch": 2.26368,
      "eval_loss": 0.6885350346565247,
      "eval_runtime": 19.5961,
      "eval_samples_per_second": 25.617,
      "eval_steps_per_second": 3.215,
      "step": 1770
    },
    {
      "epoch": 2.26496,
      "grad_norm": 0.44185778498649597,
      "learning_rate": 2.561394874053607e-05,
      "loss": 0.553,
      "step": 1771
    },
    {
      "epoch": 2.26624,
      "grad_norm": 0.4599118232727051,
      "learning_rate": 2.5599338538430304e-05,
      "loss": 0.5634,
      "step": 1772
    },
    {
      "epoch": 2.26752,
      "grad_norm": 0.46443474292755127,
      "learning_rate": 2.55847258658194e-05,
      "loss": 0.5988,
      "step": 1773
    },
    {
      "epoch": 2.2688,
      "grad_norm": 0.45918551087379456,
      "learning_rate": 2.5570110732733185e-05,
      "loss": 0.5744,
      "step": 1774
    },
    {
      "epoch": 2.27008,
      "grad_norm": 0.4701629877090454,
      "learning_rate": 2.555549314920315e-05,
      "loss": 0.5996,
      "step": 1775
    },
    {
      "epoch": 2.27136,
      "grad_norm": 0.4913133680820465,
      "learning_rate": 2.554087312526248e-05,
      "loss": 0.5626,
      "step": 1776
    },
    {
      "epoch": 2.27264,
      "grad_norm": 0.46578091382980347,
      "learning_rate": 2.5526250670946026e-05,
      "loss": 0.5534,
      "step": 1777
    },
    {
      "epoch": 2.27392,
      "grad_norm": 0.43767064809799194,
      "learning_rate": 2.5511625796290314e-05,
      "loss": 0.5758,
      "step": 1778
    },
    {
      "epoch": 2.2752,
      "grad_norm": 0.47960004210472107,
      "learning_rate": 2.5496998511333527e-05,
      "loss": 0.5682,
      "step": 1779
    },
    {
      "epoch": 2.27648,
      "grad_norm": 0.4404206871986389,
      "learning_rate": 2.5482368826115514e-05,
      "loss": 0.581,
      "step": 1780
    },
    {
      "epoch": 2.27776,
      "grad_norm": 0.48881250619888306,
      "learning_rate": 2.546773675067775e-05,
      "loss": 0.5554,
      "step": 1781
    },
    {
      "epoch": 2.27904,
      "grad_norm": 0.4786207675933838,
      "learning_rate": 2.545310229506338e-05,
      "loss": 0.578,
      "step": 1782
    },
    {
      "epoch": 2.28032,
      "grad_norm": 0.48750537633895874,
      "learning_rate": 2.5438465469317142e-05,
      "loss": 0.5951,
      "step": 1783
    },
    {
      "epoch": 2.2816,
      "grad_norm": 0.43644851446151733,
      "learning_rate": 2.542382628348544e-05,
      "loss": 0.5832,
      "step": 1784
    },
    {
      "epoch": 2.28288,
      "grad_norm": 0.44002875685691833,
      "learning_rate": 2.5409184747616284e-05,
      "loss": 0.5727,
      "step": 1785
    },
    {
      "epoch": 2.28416,
      "grad_norm": 0.47104984521865845,
      "learning_rate": 2.5394540871759294e-05,
      "loss": 0.5759,
      "step": 1786
    },
    {
      "epoch": 2.28544,
      "grad_norm": 0.5222973227500916,
      "learning_rate": 2.5379894665965686e-05,
      "loss": 0.5569,
      "step": 1787
    },
    {
      "epoch": 2.28672,
      "grad_norm": 0.44273924827575684,
      "learning_rate": 2.5365246140288302e-05,
      "loss": 0.6018,
      "step": 1788
    },
    {
      "epoch": 2.288,
      "grad_norm": 0.5452260375022888,
      "learning_rate": 2.5350595304781557e-05,
      "loss": 0.5623,
      "step": 1789
    },
    {
      "epoch": 2.2892799999999998,
      "grad_norm": 0.4626634120941162,
      "learning_rate": 2.533594216950144e-05,
      "loss": 0.5613,
      "step": 1790
    },
    {
      "epoch": 2.29056,
      "grad_norm": 0.4667629301548004,
      "learning_rate": 2.5321286744505558e-05,
      "loss": 0.6043,
      "step": 1791
    },
    {
      "epoch": 2.29184,
      "grad_norm": 0.465975821018219,
      "learning_rate": 2.530662903985305e-05,
      "loss": 0.5459,
      "step": 1792
    },
    {
      "epoch": 2.29312,
      "grad_norm": 0.43520575761795044,
      "learning_rate": 2.5291969065604636e-05,
      "loss": 0.5606,
      "step": 1793
    },
    {
      "epoch": 2.2944,
      "grad_norm": 0.4742392301559448,
      "learning_rate": 2.52773068318226e-05,
      "loss": 0.5668,
      "step": 1794
    },
    {
      "epoch": 2.29568,
      "grad_norm": 0.480435848236084,
      "learning_rate": 2.526264234857077e-05,
      "loss": 0.5385,
      "step": 1795
    },
    {
      "epoch": 2.29696,
      "grad_norm": 0.42459115386009216,
      "learning_rate": 2.524797562591451e-05,
      "loss": 0.5863,
      "step": 1796
    },
    {
      "epoch": 2.29824,
      "grad_norm": 0.4652448892593384,
      "learning_rate": 2.5233306673920735e-05,
      "loss": 0.5706,
      "step": 1797
    },
    {
      "epoch": 2.2995200000000002,
      "grad_norm": 0.45632845163345337,
      "learning_rate": 2.521863550265788e-05,
      "loss": 0.552,
      "step": 1798
    },
    {
      "epoch": 2.3008,
      "grad_norm": 0.41214942932128906,
      "learning_rate": 2.5203962122195915e-05,
      "loss": 0.5324,
      "step": 1799
    },
    {
      "epoch": 2.30208,
      "grad_norm": 0.46558645367622375,
      "learning_rate": 2.5189286542606302e-05,
      "loss": 0.5624,
      "step": 1800
    },
    {
      "epoch": 2.30208,
      "eval_loss": 0.6883463263511658,
      "eval_runtime": 19.5976,
      "eval_samples_per_second": 25.615,
      "eval_steps_per_second": 3.215,
      "step": 1800
    },
    {
      "epoch": 2.30336,
      "grad_norm": 0.4548110365867615,
      "learning_rate": 2.517460877396205e-05,
      "loss": 0.5416,
      "step": 1801
    },
    {
      "epoch": 2.30464,
      "grad_norm": 0.41525647044181824,
      "learning_rate": 2.5159928826337638e-05,
      "loss": 0.5508,
      "step": 1802
    },
    {
      "epoch": 2.30592,
      "grad_norm": 0.4663832187652588,
      "learning_rate": 2.514524670980905e-05,
      "loss": 0.6007,
      "step": 1803
    },
    {
      "epoch": 2.3072,
      "grad_norm": 0.455738365650177,
      "learning_rate": 2.5130562434453775e-05,
      "loss": 0.5324,
      "step": 1804
    },
    {
      "epoch": 2.30848,
      "grad_norm": 0.4471772015094757,
      "learning_rate": 2.5115876010350762e-05,
      "loss": 0.5461,
      "step": 1805
    },
    {
      "epoch": 2.30976,
      "grad_norm": 0.44468411803245544,
      "learning_rate": 2.5101187447580433e-05,
      "loss": 0.558,
      "step": 1806
    },
    {
      "epoch": 2.31104,
      "grad_norm": 0.4384734034538269,
      "learning_rate": 2.50864967562247e-05,
      "loss": 0.5583,
      "step": 1807
    },
    {
      "epoch": 2.31232,
      "grad_norm": 0.44412410259246826,
      "learning_rate": 2.507180394636692e-05,
      "loss": 0.5463,
      "step": 1808
    },
    {
      "epoch": 2.3136,
      "grad_norm": 0.4444560110569,
      "learning_rate": 2.50571090280919e-05,
      "loss": 0.5807,
      "step": 1809
    },
    {
      "epoch": 2.31488,
      "grad_norm": 0.45088663697242737,
      "learning_rate": 2.5042412011485918e-05,
      "loss": 0.5603,
      "step": 1810
    },
    {
      "epoch": 2.31616,
      "grad_norm": 0.4372834861278534,
      "learning_rate": 2.5027712906636664e-05,
      "loss": 0.5694,
      "step": 1811
    },
    {
      "epoch": 2.31744,
      "grad_norm": 0.44497573375701904,
      "learning_rate": 2.501301172363327e-05,
      "loss": 0.6058,
      "step": 1812
    },
    {
      "epoch": 2.31872,
      "grad_norm": 0.46659722924232483,
      "learning_rate": 2.4998308472566315e-05,
      "loss": 0.5667,
      "step": 1813
    },
    {
      "epoch": 2.32,
      "grad_norm": 0.4408104121685028,
      "learning_rate": 2.4983603163527765e-05,
      "loss": 0.5428,
      "step": 1814
    },
    {
      "epoch": 2.32128,
      "grad_norm": 0.43959736824035645,
      "learning_rate": 2.4968895806611014e-05,
      "loss": 0.554,
      "step": 1815
    },
    {
      "epoch": 2.32256,
      "grad_norm": 0.5164245963096619,
      "learning_rate": 2.4954186411910876e-05,
      "loss": 0.6254,
      "step": 1816
    },
    {
      "epoch": 2.32384,
      "grad_norm": 0.451420396566391,
      "learning_rate": 2.493947498952353e-05,
      "loss": 0.5477,
      "step": 1817
    },
    {
      "epoch": 2.32512,
      "grad_norm": 0.521676778793335,
      "learning_rate": 2.4924761549546576e-05,
      "loss": 0.5703,
      "step": 1818
    },
    {
      "epoch": 2.3264,
      "grad_norm": 0.4797092378139496,
      "learning_rate": 2.4910046102078993e-05,
      "loss": 0.5807,
      "step": 1819
    },
    {
      "epoch": 2.32768,
      "grad_norm": 0.5207264423370361,
      "learning_rate": 2.4895328657221116e-05,
      "loss": 0.5972,
      "step": 1820
    },
    {
      "epoch": 2.32896,
      "grad_norm": 0.5188628435134888,
      "learning_rate": 2.488060922507469e-05,
      "loss": 0.5516,
      "step": 1821
    },
    {
      "epoch": 2.33024,
      "grad_norm": 0.4875158965587616,
      "learning_rate": 2.4865887815742794e-05,
      "loss": 0.5781,
      "step": 1822
    },
    {
      "epoch": 2.33152,
      "grad_norm": 0.5407564640045166,
      "learning_rate": 2.4851164439329872e-05,
      "loss": 0.5635,
      "step": 1823
    },
    {
      "epoch": 2.3327999999999998,
      "grad_norm": 0.4618186950683594,
      "learning_rate": 2.4836439105941716e-05,
      "loss": 0.5905,
      "step": 1824
    },
    {
      "epoch": 2.33408,
      "grad_norm": 0.4825805127620697,
      "learning_rate": 2.4821711825685467e-05,
      "loss": 0.5967,
      "step": 1825
    },
    {
      "epoch": 2.33536,
      "grad_norm": 0.5035352110862732,
      "learning_rate": 2.48069826086696e-05,
      "loss": 0.6003,
      "step": 1826
    },
    {
      "epoch": 2.33664,
      "grad_norm": 0.4618520736694336,
      "learning_rate": 2.479225146500392e-05,
      "loss": 0.5953,
      "step": 1827
    },
    {
      "epoch": 2.33792,
      "grad_norm": 0.47905200719833374,
      "learning_rate": 2.4777518404799553e-05,
      "loss": 0.5597,
      "step": 1828
    },
    {
      "epoch": 2.3392,
      "grad_norm": 0.4741435647010803,
      "learning_rate": 2.476278343816893e-05,
      "loss": 0.5423,
      "step": 1829
    },
    {
      "epoch": 2.34048,
      "grad_norm": 0.5330893993377686,
      "learning_rate": 2.4748046575225817e-05,
      "loss": 0.5608,
      "step": 1830
    },
    {
      "epoch": 2.34048,
      "eval_loss": 0.6872444152832031,
      "eval_runtime": 19.6005,
      "eval_samples_per_second": 25.612,
      "eval_steps_per_second": 3.214,
      "step": 1830
    },
    {
      "epoch": 2.34176,
      "grad_norm": 0.45255693793296814,
      "learning_rate": 2.4733307826085265e-05,
      "loss": 0.549,
      "step": 1831
    },
    {
      "epoch": 2.3430400000000002,
      "grad_norm": 0.5184780955314636,
      "learning_rate": 2.4718567200863605e-05,
      "loss": 0.5652,
      "step": 1832
    },
    {
      "epoch": 2.34432,
      "grad_norm": 0.47143077850341797,
      "learning_rate": 2.470382470967848e-05,
      "loss": 0.593,
      "step": 1833
    },
    {
      "epoch": 2.3456,
      "grad_norm": 0.4462724030017853,
      "learning_rate": 2.468908036264881e-05,
      "loss": 0.5655,
      "step": 1834
    },
    {
      "epoch": 2.34688,
      "grad_norm": 0.5066423416137695,
      "learning_rate": 2.467433416989477e-05,
      "loss": 0.5925,
      "step": 1835
    },
    {
      "epoch": 2.34816,
      "grad_norm": 0.48381438851356506,
      "learning_rate": 2.465958614153783e-05,
      "loss": 0.5916,
      "step": 1836
    },
    {
      "epoch": 2.34944,
      "grad_norm": 0.4324931502342224,
      "learning_rate": 2.4644836287700704e-05,
      "loss": 0.5523,
      "step": 1837
    },
    {
      "epoch": 2.35072,
      "grad_norm": 0.506111741065979,
      "learning_rate": 2.463008461850735e-05,
      "loss": 0.5622,
      "step": 1838
    },
    {
      "epoch": 2.352,
      "grad_norm": 0.45325300097465515,
      "learning_rate": 2.4615331144082984e-05,
      "loss": 0.5649,
      "step": 1839
    },
    {
      "epoch": 2.35328,
      "grad_norm": 0.45014071464538574,
      "learning_rate": 2.4600575874554072e-05,
      "loss": 0.5577,
      "step": 1840
    },
    {
      "epoch": 2.35456,
      "grad_norm": 0.5864941477775574,
      "learning_rate": 2.458581882004829e-05,
      "loss": 0.6189,
      "step": 1841
    },
    {
      "epoch": 2.35584,
      "grad_norm": 0.4674026072025299,
      "learning_rate": 2.4571059990694547e-05,
      "loss": 0.5852,
      "step": 1842
    },
    {
      "epoch": 2.35712,
      "grad_norm": 0.46268466114997864,
      "learning_rate": 2.4556299396622976e-05,
      "loss": 0.6293,
      "step": 1843
    },
    {
      "epoch": 2.3584,
      "grad_norm": 0.5257765650749207,
      "learning_rate": 2.4541537047964918e-05,
      "loss": 0.5812,
      "step": 1844
    },
    {
      "epoch": 2.35968,
      "grad_norm": 0.42285916209220886,
      "learning_rate": 2.452677295485291e-05,
      "loss": 0.5807,
      "step": 1845
    },
    {
      "epoch": 2.36096,
      "grad_norm": 0.42277437448501587,
      "learning_rate": 2.4512007127420705e-05,
      "loss": 0.5514,
      "step": 1846
    },
    {
      "epoch": 2.36224,
      "grad_norm": 0.41128724813461304,
      "learning_rate": 2.4497239575803233e-05,
      "loss": 0.5776,
      "step": 1847
    },
    {
      "epoch": 2.36352,
      "grad_norm": 0.4611054062843323,
      "learning_rate": 2.4482470310136597e-05,
      "loss": 0.5703,
      "step": 1848
    },
    {
      "epoch": 2.3648,
      "grad_norm": 0.4351324141025543,
      "learning_rate": 2.4467699340558108e-05,
      "loss": 0.5587,
      "step": 1849
    },
    {
      "epoch": 2.36608,
      "grad_norm": 0.437563419342041,
      "learning_rate": 2.4452926677206215e-05,
      "loss": 0.5601,
      "step": 1850
    },
    {
      "epoch": 2.36736,
      "grad_norm": 0.43191543221473694,
      "learning_rate": 2.4438152330220546e-05,
      "loss": 0.5536,
      "step": 1851
    },
    {
      "epoch": 2.36864,
      "grad_norm": 0.45357435941696167,
      "learning_rate": 2.4423376309741878e-05,
      "loss": 0.6048,
      "step": 1852
    },
    {
      "epoch": 2.36992,
      "grad_norm": 0.43874821066856384,
      "learning_rate": 2.4408598625912154e-05,
      "loss": 0.5751,
      "step": 1853
    },
    {
      "epoch": 2.3712,
      "grad_norm": 0.42184025049209595,
      "learning_rate": 2.4393819288874422e-05,
      "loss": 0.5645,
      "step": 1854
    },
    {
      "epoch": 2.37248,
      "grad_norm": 0.45755261182785034,
      "learning_rate": 2.4379038308772897e-05,
      "loss": 0.5678,
      "step": 1855
    },
    {
      "epoch": 2.37376,
      "grad_norm": 0.42253774404525757,
      "learning_rate": 2.4364255695752917e-05,
      "loss": 0.5734,
      "step": 1856
    },
    {
      "epoch": 2.37504,
      "grad_norm": 0.4399717450141907,
      "learning_rate": 2.4349471459960935e-05,
      "loss": 0.556,
      "step": 1857
    },
    {
      "epoch": 2.3763199999999998,
      "grad_norm": 0.45479288697242737,
      "learning_rate": 2.4334685611544505e-05,
      "loss": 0.581,
      "step": 1858
    },
    {
      "epoch": 2.3776,
      "grad_norm": 0.45053523778915405,
      "learning_rate": 2.4319898160652313e-05,
      "loss": 0.5502,
      "step": 1859
    },
    {
      "epoch": 2.37888,
      "grad_norm": 0.43532097339630127,
      "learning_rate": 2.430510911743414e-05,
      "loss": 0.5463,
      "step": 1860
    },
    {
      "epoch": 2.37888,
      "eval_loss": 0.6879814267158508,
      "eval_runtime": 19.6392,
      "eval_samples_per_second": 25.561,
      "eval_steps_per_second": 3.208,
      "step": 1860
    },
    {
      "epoch": 2.38016,
      "grad_norm": 0.4721943438053131,
      "learning_rate": 2.4290318492040835e-05,
      "loss": 0.5633,
      "step": 1861
    },
    {
      "epoch": 2.38144,
      "grad_norm": 0.47411856055259705,
      "learning_rate": 2.427552629462436e-05,
      "loss": 0.5639,
      "step": 1862
    },
    {
      "epoch": 2.38272,
      "grad_norm": 0.43705224990844727,
      "learning_rate": 2.426073253533775e-05,
      "loss": 0.5768,
      "step": 1863
    },
    {
      "epoch": 2.384,
      "grad_norm": 0.4831198751926422,
      "learning_rate": 2.4245937224335103e-05,
      "loss": 0.5831,
      "step": 1864
    },
    {
      "epoch": 2.38528,
      "grad_norm": 0.4628940224647522,
      "learning_rate": 2.4231140371771592e-05,
      "loss": 0.5804,
      "step": 1865
    },
    {
      "epoch": 2.3865600000000002,
      "grad_norm": 0.42717164754867554,
      "learning_rate": 2.421634198780345e-05,
      "loss": 0.5849,
      "step": 1866
    },
    {
      "epoch": 2.38784,
      "grad_norm": 0.4303050935268402,
      "learning_rate": 2.4201542082587942e-05,
      "loss": 0.5649,
      "step": 1867
    },
    {
      "epoch": 2.38912,
      "grad_norm": 0.43048736453056335,
      "learning_rate": 2.4186740666283397e-05,
      "loss": 0.553,
      "step": 1868
    },
    {
      "epoch": 2.3904,
      "grad_norm": 0.446988970041275,
      "learning_rate": 2.4171937749049175e-05,
      "loss": 0.5492,
      "step": 1869
    },
    {
      "epoch": 2.39168,
      "grad_norm": 0.4533318877220154,
      "learning_rate": 2.415713334104567e-05,
      "loss": 0.5985,
      "step": 1870
    },
    {
      "epoch": 2.39296,
      "grad_norm": 0.44291216135025024,
      "learning_rate": 2.4142327452434292e-05,
      "loss": 0.5639,
      "step": 1871
    },
    {
      "epoch": 2.39424,
      "grad_norm": 0.4459875524044037,
      "learning_rate": 2.4127520093377465e-05,
      "loss": 0.5882,
      "step": 1872
    },
    {
      "epoch": 2.39552,
      "grad_norm": 0.4398691654205322,
      "learning_rate": 2.411271127403864e-05,
      "loss": 0.5554,
      "step": 1873
    },
    {
      "epoch": 2.3968,
      "grad_norm": 0.3956506848335266,
      "learning_rate": 2.4097901004582233e-05,
      "loss": 0.5516,
      "step": 1874
    },
    {
      "epoch": 2.39808,
      "grad_norm": 0.4817471504211426,
      "learning_rate": 2.4083089295173716e-05,
      "loss": 0.5822,
      "step": 1875
    },
    {
      "epoch": 2.39936,
      "grad_norm": 0.4335651099681854,
      "learning_rate": 2.4068276155979486e-05,
      "loss": 0.5877,
      "step": 1876
    },
    {
      "epoch": 2.40064,
      "grad_norm": 0.4682128429412842,
      "learning_rate": 2.405346159716696e-05,
      "loss": 0.562,
      "step": 1877
    },
    {
      "epoch": 2.40192,
      "grad_norm": 0.45321351289749146,
      "learning_rate": 2.4038645628904515e-05,
      "loss": 0.5786,
      "step": 1878
    },
    {
      "epoch": 2.4032,
      "grad_norm": 0.45631060004234314,
      "learning_rate": 2.40238282613615e-05,
      "loss": 0.5799,
      "step": 1879
    },
    {
      "epoch": 2.40448,
      "grad_norm": 0.43943092226982117,
      "learning_rate": 2.4009009504708232e-05,
      "loss": 0.5812,
      "step": 1880
    },
    {
      "epoch": 2.40576,
      "grad_norm": 0.4432222843170166,
      "learning_rate": 2.3994189369115953e-05,
      "loss": 0.5344,
      "step": 1881
    },
    {
      "epoch": 2.40704,
      "grad_norm": 0.46579059958457947,
      "learning_rate": 2.3979367864756887e-05,
      "loss": 0.593,
      "step": 1882
    },
    {
      "epoch": 2.40832,
      "grad_norm": 0.48050373792648315,
      "learning_rate": 2.396454500180418e-05,
      "loss": 0.5901,
      "step": 1883
    },
    {
      "epoch": 2.4096,
      "grad_norm": 0.44765594601631165,
      "learning_rate": 2.3949720790431905e-05,
      "loss": 0.5742,
      "step": 1884
    },
    {
      "epoch": 2.41088,
      "grad_norm": 0.45086538791656494,
      "learning_rate": 2.3934895240815075e-05,
      "loss": 0.5604,
      "step": 1885
    },
    {
      "epoch": 2.41216,
      "grad_norm": 0.5598390698432922,
      "learning_rate": 2.3920068363129612e-05,
      "loss": 0.6001,
      "step": 1886
    },
    {
      "epoch": 2.41344,
      "grad_norm": 0.445596307516098,
      "learning_rate": 2.390524016755235e-05,
      "loss": 0.5402,
      "step": 1887
    },
    {
      "epoch": 2.41472,
      "grad_norm": 0.4994315803050995,
      "learning_rate": 2.3890410664261033e-05,
      "loss": 0.587,
      "step": 1888
    },
    {
      "epoch": 2.416,
      "grad_norm": 0.5435822606086731,
      "learning_rate": 2.3875579863434295e-05,
      "loss": 0.5403,
      "step": 1889
    },
    {
      "epoch": 2.41728,
      "grad_norm": 0.44386982917785645,
      "learning_rate": 2.386074777525166e-05,
      "loss": 0.5797,
      "step": 1890
    },
    {
      "epoch": 2.41728,
      "eval_loss": 0.6864021420478821,
      "eval_runtime": 19.5972,
      "eval_samples_per_second": 25.616,
      "eval_steps_per_second": 3.215,
      "step": 1890
    },
    {
      "epoch": 2.41856,
      "grad_norm": 0.49713876843452454,
      "learning_rate": 2.3845914409893554e-05,
      "loss": 0.566,
      "step": 1891
    },
    {
      "epoch": 2.4198399999999998,
      "grad_norm": 0.4760511815547943,
      "learning_rate": 2.383107977754126e-05,
      "loss": 0.5999,
      "step": 1892
    },
    {
      "epoch": 2.42112,
      "grad_norm": 0.47576361894607544,
      "learning_rate": 2.3816243888376925e-05,
      "loss": 0.5681,
      "step": 1893
    },
    {
      "epoch": 2.4224,
      "grad_norm": 0.46428993344306946,
      "learning_rate": 2.380140675258359e-05,
      "loss": 0.5627,
      "step": 1894
    },
    {
      "epoch": 2.42368,
      "grad_norm": 0.4505695700645447,
      "learning_rate": 2.378656838034511e-05,
      "loss": 0.5831,
      "step": 1895
    },
    {
      "epoch": 2.42496,
      "grad_norm": 0.48643916845321655,
      "learning_rate": 2.3771728781846228e-05,
      "loss": 0.5771,
      "step": 1896
    },
    {
      "epoch": 2.42624,
      "grad_norm": 0.4123098850250244,
      "learning_rate": 2.3756887967272514e-05,
      "loss": 0.5595,
      "step": 1897
    },
    {
      "epoch": 2.42752,
      "grad_norm": 0.4496588408946991,
      "learning_rate": 2.3742045946810346e-05,
      "loss": 0.5855,
      "step": 1898
    },
    {
      "epoch": 2.4288,
      "grad_norm": 0.4408184289932251,
      "learning_rate": 2.3727202730646976e-05,
      "loss": 0.5623,
      "step": 1899
    },
    {
      "epoch": 2.4300800000000002,
      "grad_norm": 0.4345359802246094,
      "learning_rate": 2.3712358328970448e-05,
      "loss": 0.551,
      "step": 1900
    },
    {
      "epoch": 2.43136,
      "grad_norm": 0.4054926633834839,
      "learning_rate": 2.3697512751969617e-05,
      "loss": 0.5521,
      "step": 1901
    },
    {
      "epoch": 2.43264,
      "grad_norm": 0.44658681750297546,
      "learning_rate": 2.3682666009834164e-05,
      "loss": 0.5596,
      "step": 1902
    },
    {
      "epoch": 2.43392,
      "grad_norm": 0.4469820559024811,
      "learning_rate": 2.3667818112754563e-05,
      "loss": 0.5715,
      "step": 1903
    },
    {
      "epoch": 2.4352,
      "grad_norm": 0.4233338236808777,
      "learning_rate": 2.3652969070922062e-05,
      "loss": 0.5879,
      "step": 1904
    },
    {
      "epoch": 2.43648,
      "grad_norm": 0.4192318022251129,
      "learning_rate": 2.363811889452872e-05,
      "loss": 0.5605,
      "step": 1905
    },
    {
      "epoch": 2.43776,
      "grad_norm": 0.4243566393852234,
      "learning_rate": 2.3623267593767363e-05,
      "loss": 0.5992,
      "step": 1906
    },
    {
      "epoch": 2.43904,
      "grad_norm": 0.4443212151527405,
      "learning_rate": 2.360841517883159e-05,
      "loss": 0.6051,
      "step": 1907
    },
    {
      "epoch": 2.44032,
      "grad_norm": 0.4523197412490845,
      "learning_rate": 2.3593561659915767e-05,
      "loss": 0.5765,
      "step": 1908
    },
    {
      "epoch": 2.4416,
      "grad_norm": 0.4022792875766754,
      "learning_rate": 2.3578707047215013e-05,
      "loss": 0.5689,
      "step": 1909
    },
    {
      "epoch": 2.44288,
      "grad_norm": 0.47122398018836975,
      "learning_rate": 2.35638513509252e-05,
      "loss": 0.5709,
      "step": 1910
    },
    {
      "epoch": 2.44416,
      "grad_norm": 0.4721255600452423,
      "learning_rate": 2.3548994581242948e-05,
      "loss": 0.6031,
      "step": 1911
    },
    {
      "epoch": 2.44544,
      "grad_norm": 0.4392406940460205,
      "learning_rate": 2.3534136748365604e-05,
      "loss": 0.5971,
      "step": 1912
    },
    {
      "epoch": 2.44672,
      "grad_norm": 0.4260604679584503,
      "learning_rate": 2.3519277862491255e-05,
      "loss": 0.5545,
      "step": 1913
    },
    {
      "epoch": 2.448,
      "grad_norm": 0.4519716203212738,
      "learning_rate": 2.3504417933818708e-05,
      "loss": 0.5799,
      "step": 1914
    },
    {
      "epoch": 2.44928,
      "grad_norm": 0.43348559737205505,
      "learning_rate": 2.348955697254748e-05,
      "loss": 0.5671,
      "step": 1915
    },
    {
      "epoch": 2.45056,
      "grad_norm": 0.48234179615974426,
      "learning_rate": 2.34746949888778e-05,
      "loss": 0.5705,
      "step": 1916
    },
    {
      "epoch": 2.45184,
      "grad_norm": 0.5109958648681641,
      "learning_rate": 2.3459831993010605e-05,
      "loss": 0.5895,
      "step": 1917
    },
    {
      "epoch": 2.45312,
      "grad_norm": 0.42233631014823914,
      "learning_rate": 2.3444967995147524e-05,
      "loss": 0.5698,
      "step": 1918
    },
    {
      "epoch": 2.4544,
      "grad_norm": 0.4705251157283783,
      "learning_rate": 2.3430103005490863e-05,
      "loss": 0.5653,
      "step": 1919
    },
    {
      "epoch": 2.45568,
      "grad_norm": 0.45902517437934875,
      "learning_rate": 2.3415237034243618e-05,
      "loss": 0.5817,
      "step": 1920
    },
    {
      "epoch": 2.45568,
      "eval_loss": 0.6859309077262878,
      "eval_runtime": 19.6029,
      "eval_samples_per_second": 25.609,
      "eval_steps_per_second": 3.214,
      "step": 1920
    },
    {
      "epoch": 2.45696,
      "grad_norm": 0.4393598437309265,
      "learning_rate": 2.3400370091609464e-05,
      "loss": 0.5843,
      "step": 1921
    },
    {
      "epoch": 2.45824,
      "grad_norm": 0.46856045722961426,
      "learning_rate": 2.338550218779273e-05,
      "loss": 0.5565,
      "step": 1922
    },
    {
      "epoch": 2.45952,
      "grad_norm": 0.4586985111236572,
      "learning_rate": 2.3370633332998413e-05,
      "loss": 0.5698,
      "step": 1923
    },
    {
      "epoch": 2.4608,
      "grad_norm": 0.4334520399570465,
      "learning_rate": 2.3355763537432173e-05,
      "loss": 0.5812,
      "step": 1924
    },
    {
      "epoch": 2.46208,
      "grad_norm": 0.4530614912509918,
      "learning_rate": 2.3340892811300283e-05,
      "loss": 0.5784,
      "step": 1925
    },
    {
      "epoch": 2.4633599999999998,
      "grad_norm": 0.5027890205383301,
      "learning_rate": 2.3326021164809688e-05,
      "loss": 0.5759,
      "step": 1926
    },
    {
      "epoch": 2.46464,
      "grad_norm": 0.4158819019794464,
      "learning_rate": 2.3311148608167944e-05,
      "loss": 0.5622,
      "step": 1927
    },
    {
      "epoch": 2.46592,
      "grad_norm": 0.4853421151638031,
      "learning_rate": 2.329627515158325e-05,
      "loss": 0.5926,
      "step": 1928
    },
    {
      "epoch": 2.4672,
      "grad_norm": 0.4470900595188141,
      "learning_rate": 2.3281400805264408e-05,
      "loss": 0.5613,
      "step": 1929
    },
    {
      "epoch": 2.46848,
      "grad_norm": 0.510113000869751,
      "learning_rate": 2.3266525579420833e-05,
      "loss": 0.5909,
      "step": 1930
    },
    {
      "epoch": 2.46976,
      "grad_norm": 0.5097657442092896,
      "learning_rate": 2.3251649484262546e-05,
      "loss": 0.5729,
      "step": 1931
    },
    {
      "epoch": 2.47104,
      "grad_norm": 0.4624604880809784,
      "learning_rate": 2.323677253000017e-05,
      "loss": 0.5883,
      "step": 1932
    },
    {
      "epoch": 2.47232,
      "grad_norm": 0.4557540714740753,
      "learning_rate": 2.322189472684491e-05,
      "loss": 0.5555,
      "step": 1933
    },
    {
      "epoch": 2.4736000000000002,
      "grad_norm": 0.44446662068367004,
      "learning_rate": 2.320701608500855e-05,
      "loss": 0.5867,
      "step": 1934
    },
    {
      "epoch": 2.47488,
      "grad_norm": 0.4319787621498108,
      "learning_rate": 2.319213661470346e-05,
      "loss": 0.5708,
      "step": 1935
    },
    {
      "epoch": 2.47616,
      "grad_norm": 0.45995891094207764,
      "learning_rate": 2.3177256326142578e-05,
      "loss": 0.5638,
      "step": 1936
    },
    {
      "epoch": 2.47744,
      "grad_norm": 0.4343652129173279,
      "learning_rate": 2.3162375229539395e-05,
      "loss": 0.5815,
      "step": 1937
    },
    {
      "epoch": 2.47872,
      "grad_norm": 0.4371507465839386,
      "learning_rate": 2.3147493335107964e-05,
      "loss": 0.5756,
      "step": 1938
    },
    {
      "epoch": 2.48,
      "grad_norm": 0.41909360885620117,
      "learning_rate": 2.3132610653062873e-05,
      "loss": 0.5988,
      "step": 1939
    },
    {
      "epoch": 2.48128,
      "grad_norm": 0.4791184663772583,
      "learning_rate": 2.3117727193619278e-05,
      "loss": 0.6049,
      "step": 1940
    },
    {
      "epoch": 2.48256,
      "grad_norm": 0.4705685079097748,
      "learning_rate": 2.3102842966992847e-05,
      "loss": 0.5884,
      "step": 1941
    },
    {
      "epoch": 2.48384,
      "grad_norm": 0.4478840231895447,
      "learning_rate": 2.3087957983399762e-05,
      "loss": 0.5908,
      "step": 1942
    },
    {
      "epoch": 2.48512,
      "grad_norm": 0.42873358726501465,
      "learning_rate": 2.307307225305676e-05,
      "loss": 0.5504,
      "step": 1943
    },
    {
      "epoch": 2.4864,
      "grad_norm": 0.45527398586273193,
      "learning_rate": 2.3058185786181068e-05,
      "loss": 0.568,
      "step": 1944
    },
    {
      "epoch": 2.48768,
      "grad_norm": 0.4955735504627228,
      "learning_rate": 2.304329859299042e-05,
      "loss": 0.5888,
      "step": 1945
    },
    {
      "epoch": 2.48896,
      "grad_norm": 0.47625014185905457,
      "learning_rate": 2.3028410683703045e-05,
      "loss": 0.5887,
      "step": 1946
    },
    {
      "epoch": 2.49024,
      "grad_norm": 0.4415814280509949,
      "learning_rate": 2.301352206853768e-05,
      "loss": 0.5665,
      "step": 1947
    },
    {
      "epoch": 2.49152,
      "grad_norm": 0.4647148847579956,
      "learning_rate": 2.2998632757713538e-05,
      "loss": 0.5282,
      "step": 1948
    },
    {
      "epoch": 2.4928,
      "grad_norm": 0.4453596770763397,
      "learning_rate": 2.2983742761450286e-05,
      "loss": 0.559,
      "step": 1949
    },
    {
      "epoch": 2.49408,
      "grad_norm": 0.4447324872016907,
      "learning_rate": 2.296885208996811e-05,
      "loss": 0.5681,
      "step": 1950
    },
    {
      "epoch": 2.49408,
      "eval_loss": 0.6846315860748291,
      "eval_runtime": 19.6122,
      "eval_samples_per_second": 25.596,
      "eval_steps_per_second": 3.212,
      "step": 1950
    },
    {
      "epoch": 2.49536,
      "grad_norm": 0.44272559881210327,
      "learning_rate": 2.2953960753487623e-05,
      "loss": 0.5659,
      "step": 1951
    },
    {
      "epoch": 2.49664,
      "grad_norm": 0.44563981890678406,
      "learning_rate": 2.2939068762229894e-05,
      "loss": 0.5671,
      "step": 1952
    },
    {
      "epoch": 2.49792,
      "grad_norm": 0.4674242436885834,
      "learning_rate": 2.2924176126416463e-05,
      "loss": 0.6073,
      "step": 1953
    },
    {
      "epoch": 2.4992,
      "grad_norm": 0.4552050232887268,
      "learning_rate": 2.2909282856269298e-05,
      "loss": 0.5679,
      "step": 1954
    },
    {
      "epoch": 2.50048,
      "grad_norm": 0.5002722144126892,
      "learning_rate": 2.2894388962010798e-05,
      "loss": 0.5771,
      "step": 1955
    },
    {
      "epoch": 2.50176,
      "grad_norm": 0.4574168026447296,
      "learning_rate": 2.287949445386381e-05,
      "loss": 0.528,
      "step": 1956
    },
    {
      "epoch": 2.50304,
      "grad_norm": 0.47306182980537415,
      "learning_rate": 2.2864599342051584e-05,
      "loss": 0.5493,
      "step": 1957
    },
    {
      "epoch": 2.50432,
      "grad_norm": 0.4465441107749939,
      "learning_rate": 2.2849703636797788e-05,
      "loss": 0.5527,
      "step": 1958
    },
    {
      "epoch": 2.5056000000000003,
      "grad_norm": 0.4296610355377197,
      "learning_rate": 2.2834807348326497e-05,
      "loss": 0.579,
      "step": 1959
    },
    {
      "epoch": 2.5068799999999998,
      "grad_norm": 0.4300537407398224,
      "learning_rate": 2.2819910486862194e-05,
      "loss": 0.5758,
      "step": 1960
    },
    {
      "epoch": 2.50816,
      "grad_norm": 0.4490695893764496,
      "learning_rate": 2.280501306262975e-05,
      "loss": 0.5505,
      "step": 1961
    },
    {
      "epoch": 2.50944,
      "grad_norm": 0.4072880446910858,
      "learning_rate": 2.279011508585442e-05,
      "loss": 0.5602,
      "step": 1962
    },
    {
      "epoch": 2.51072,
      "grad_norm": 0.42263078689575195,
      "learning_rate": 2.2775216566761843e-05,
      "loss": 0.5755,
      "step": 1963
    },
    {
      "epoch": 2.512,
      "grad_norm": 0.46272796392440796,
      "learning_rate": 2.2760317515578024e-05,
      "loss": 0.5512,
      "step": 1964
    },
    {
      "epoch": 2.51328,
      "grad_norm": 0.46716970205307007,
      "learning_rate": 2.2745417942529343e-05,
      "loss": 0.5855,
      "step": 1965
    },
    {
      "epoch": 2.51456,
      "grad_norm": 0.42793014645576477,
      "learning_rate": 2.2730517857842526e-05,
      "loss": 0.5526,
      "step": 1966
    },
    {
      "epoch": 2.51584,
      "grad_norm": 0.45595625042915344,
      "learning_rate": 2.271561727174467e-05,
      "loss": 0.5666,
      "step": 1967
    },
    {
      "epoch": 2.5171200000000002,
      "grad_norm": 0.4583331346511841,
      "learning_rate": 2.2700716194463177e-05,
      "loss": 0.5858,
      "step": 1968
    },
    {
      "epoch": 2.5183999999999997,
      "grad_norm": 0.48306262493133545,
      "learning_rate": 2.2685814636225837e-05,
      "loss": 0.5691,
      "step": 1969
    },
    {
      "epoch": 2.51968,
      "grad_norm": 0.4364309310913086,
      "learning_rate": 2.2670912607260732e-05,
      "loss": 0.5731,
      "step": 1970
    },
    {
      "epoch": 2.52096,
      "grad_norm": 0.47410571575164795,
      "learning_rate": 2.2656010117796278e-05,
      "loss": 0.5652,
      "step": 1971
    },
    {
      "epoch": 2.52224,
      "grad_norm": 0.4599086046218872,
      "learning_rate": 2.2641107178061225e-05,
      "loss": 0.5783,
      "step": 1972
    },
    {
      "epoch": 2.52352,
      "grad_norm": 0.4437457323074341,
      "learning_rate": 2.26262037982846e-05,
      "loss": 0.5843,
      "step": 1973
    },
    {
      "epoch": 2.5248,
      "grad_norm": 0.4504581391811371,
      "learning_rate": 2.2611299988695744e-05,
      "loss": 0.5786,
      "step": 1974
    },
    {
      "epoch": 2.52608,
      "grad_norm": 0.4429946839809418,
      "learning_rate": 2.2596395759524316e-05,
      "loss": 0.572,
      "step": 1975
    },
    {
      "epoch": 2.52736,
      "grad_norm": 0.4308660328388214,
      "learning_rate": 2.258149112100023e-05,
      "loss": 0.5473,
      "step": 1976
    },
    {
      "epoch": 2.52864,
      "grad_norm": 0.45771992206573486,
      "learning_rate": 2.25665860833537e-05,
      "loss": 0.5695,
      "step": 1977
    },
    {
      "epoch": 2.5299199999999997,
      "grad_norm": 0.47256872057914734,
      "learning_rate": 2.2551680656815217e-05,
      "loss": 0.5642,
      "step": 1978
    },
    {
      "epoch": 2.5312,
      "grad_norm": 0.4202207624912262,
      "learning_rate": 2.2536774851615507e-05,
      "loss": 0.5735,
      "step": 1979
    },
    {
      "epoch": 2.53248,
      "grad_norm": 0.43840059638023376,
      "learning_rate": 2.252186867798561e-05,
      "loss": 0.5369,
      "step": 1980
    },
    {
      "epoch": 2.53248,
      "eval_loss": 0.6849427819252014,
      "eval_runtime": 19.5935,
      "eval_samples_per_second": 25.621,
      "eval_steps_per_second": 3.215,
      "step": 1980
    },
    {
      "epoch": 2.53376,
      "grad_norm": 0.4783090054988861,
      "learning_rate": 2.2506962146156764e-05,
      "loss": 0.6176,
      "step": 1981
    },
    {
      "epoch": 2.53504,
      "grad_norm": 0.4283246397972107,
      "learning_rate": 2.249205526636049e-05,
      "loss": 0.5774,
      "step": 1982
    },
    {
      "epoch": 2.53632,
      "grad_norm": 0.45895466208457947,
      "learning_rate": 2.2477148048828536e-05,
      "loss": 0.5747,
      "step": 1983
    },
    {
      "epoch": 2.5376,
      "grad_norm": 0.4468008279800415,
      "learning_rate": 2.2462240503792882e-05,
      "loss": 0.5482,
      "step": 1984
    },
    {
      "epoch": 2.53888,
      "grad_norm": 0.4108228087425232,
      "learning_rate": 2.2447332641485717e-05,
      "loss": 0.547,
      "step": 1985
    },
    {
      "epoch": 2.54016,
      "grad_norm": 0.4353698194026947,
      "learning_rate": 2.2432424472139483e-05,
      "loss": 0.5734,
      "step": 1986
    },
    {
      "epoch": 2.54144,
      "grad_norm": 0.4353506863117218,
      "learning_rate": 2.2417516005986806e-05,
      "loss": 0.5695,
      "step": 1987
    },
    {
      "epoch": 2.54272,
      "grad_norm": 0.4309532642364502,
      "learning_rate": 2.2402607253260514e-05,
      "loss": 0.595,
      "step": 1988
    },
    {
      "epoch": 2.544,
      "grad_norm": 0.4385818839073181,
      "learning_rate": 2.2387698224193653e-05,
      "loss": 0.5556,
      "step": 1989
    },
    {
      "epoch": 2.54528,
      "grad_norm": 0.44252559542655945,
      "learning_rate": 2.2372788929019432e-05,
      "loss": 0.5887,
      "step": 1990
    },
    {
      "epoch": 2.54656,
      "grad_norm": 0.43037253618240356,
      "learning_rate": 2.2357879377971265e-05,
      "loss": 0.5861,
      "step": 1991
    },
    {
      "epoch": 2.54784,
      "grad_norm": 0.427584707736969,
      "learning_rate": 2.2342969581282734e-05,
      "loss": 0.5482,
      "step": 1992
    },
    {
      "epoch": 2.5491200000000003,
      "grad_norm": 0.4659539461135864,
      "learning_rate": 2.2328059549187582e-05,
      "loss": 0.5764,
      "step": 1993
    },
    {
      "epoch": 2.5504,
      "grad_norm": 0.42366841435432434,
      "learning_rate": 2.2313149291919726e-05,
      "loss": 0.5622,
      "step": 1994
    },
    {
      "epoch": 2.55168,
      "grad_norm": 0.44541725516319275,
      "learning_rate": 2.2298238819713234e-05,
      "loss": 0.587,
      "step": 1995
    },
    {
      "epoch": 2.55296,
      "grad_norm": 0.41166210174560547,
      "learning_rate": 2.228332814280231e-05,
      "loss": 0.5672,
      "step": 1996
    },
    {
      "epoch": 2.55424,
      "grad_norm": 0.4156389832496643,
      "learning_rate": 2.2268417271421315e-05,
      "loss": 0.5545,
      "step": 1997
    },
    {
      "epoch": 2.55552,
      "grad_norm": 0.4599660336971283,
      "learning_rate": 2.2253506215804733e-05,
      "loss": 0.5899,
      "step": 1998
    },
    {
      "epoch": 2.5568,
      "grad_norm": 0.3998776078224182,
      "learning_rate": 2.2238594986187177e-05,
      "loss": 0.5902,
      "step": 1999
    },
    {
      "epoch": 2.55808,
      "grad_norm": 0.4303065836429596,
      "learning_rate": 2.2223683592803383e-05,
      "loss": 0.5691,
      "step": 2000
    },
    {
      "epoch": 2.55936,
      "grad_norm": 0.4438993036746979,
      "learning_rate": 2.2208772045888193e-05,
      "loss": 0.5614,
      "step": 2001
    },
    {
      "epoch": 2.5606400000000002,
      "grad_norm": 0.4252774715423584,
      "learning_rate": 2.2193860355676563e-05,
      "loss": 0.5818,
      "step": 2002
    },
    {
      "epoch": 2.5619199999999998,
      "grad_norm": 0.4115821123123169,
      "learning_rate": 2.2178948532403538e-05,
      "loss": 0.5468,
      "step": 2003
    },
    {
      "epoch": 2.5632,
      "grad_norm": 0.45010432600975037,
      "learning_rate": 2.2164036586304254e-05,
      "loss": 0.5656,
      "step": 2004
    },
    {
      "epoch": 2.56448,
      "grad_norm": 0.42297306656837463,
      "learning_rate": 2.2149124527613953e-05,
      "loss": 0.5856,
      "step": 2005
    },
    {
      "epoch": 2.56576,
      "grad_norm": 0.4158904552459717,
      "learning_rate": 2.213421236656792e-05,
      "loss": 0.5745,
      "step": 2006
    },
    {
      "epoch": 2.56704,
      "grad_norm": 0.49635934829711914,
      "learning_rate": 2.2119300113401527e-05,
      "loss": 0.5775,
      "step": 2007
    },
    {
      "epoch": 2.56832,
      "grad_norm": 0.46382132172584534,
      "learning_rate": 2.2104387778350226e-05,
      "loss": 0.5569,
      "step": 2008
    },
    {
      "epoch": 2.5696,
      "grad_norm": 0.43032869696617126,
      "learning_rate": 2.2089475371649493e-05,
      "loss": 0.5576,
      "step": 2009
    },
    {
      "epoch": 2.57088,
      "grad_norm": 0.4703066051006317,
      "learning_rate": 2.2074562903534883e-05,
      "loss": 0.5696,
      "step": 2010
    },
    {
      "epoch": 2.57088,
      "eval_loss": 0.6837652921676636,
      "eval_runtime": 19.5958,
      "eval_samples_per_second": 25.618,
      "eval_steps_per_second": 3.215,
      "step": 2010
    },
    {
      "epoch": 2.5721600000000002,
      "grad_norm": 0.43766123056411743,
      "learning_rate": 2.2059650384241956e-05,
      "loss": 0.5794,
      "step": 2011
    },
    {
      "epoch": 2.5734399999999997,
      "grad_norm": 0.4614957273006439,
      "learning_rate": 2.2044737824006354e-05,
      "loss": 0.563,
      "step": 2012
    },
    {
      "epoch": 2.57472,
      "grad_norm": 0.4705134332180023,
      "learning_rate": 2.2029825233063716e-05,
      "loss": 0.6088,
      "step": 2013
    },
    {
      "epoch": 2.576,
      "grad_norm": 0.4462265074253082,
      "learning_rate": 2.2014912621649702e-05,
      "loss": 0.5475,
      "step": 2014
    },
    {
      "epoch": 2.57728,
      "grad_norm": 0.41511115431785583,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.5729,
      "step": 2015
    },
    {
      "epoch": 2.57856,
      "grad_norm": 0.42461708188056946,
      "learning_rate": 2.19850873783503e-05,
      "loss": 0.5849,
      "step": 2016
    },
    {
      "epoch": 2.57984,
      "grad_norm": 0.44178956747055054,
      "learning_rate": 2.19701747669363e-05,
      "loss": 0.5369,
      "step": 2017
    },
    {
      "epoch": 2.58112,
      "grad_norm": 0.4107958972454071,
      "learning_rate": 2.195526217599365e-05,
      "loss": 0.5677,
      "step": 2018
    },
    {
      "epoch": 2.5824,
      "grad_norm": 0.4347243905067444,
      "learning_rate": 2.1940349615758046e-05,
      "loss": 0.599,
      "step": 2019
    },
    {
      "epoch": 2.58368,
      "grad_norm": 0.4311941862106323,
      "learning_rate": 2.1925437096465133e-05,
      "loss": 0.5485,
      "step": 2020
    },
    {
      "epoch": 2.58496,
      "grad_norm": 0.47051113843917847,
      "learning_rate": 2.1910524628350513e-05,
      "loss": 0.5672,
      "step": 2021
    },
    {
      "epoch": 2.58624,
      "grad_norm": 0.47263777256011963,
      "learning_rate": 2.189561222164978e-05,
      "loss": 0.5759,
      "step": 2022
    },
    {
      "epoch": 2.58752,
      "grad_norm": 0.4554099142551422,
      "learning_rate": 2.188069988659848e-05,
      "loss": 0.5681,
      "step": 2023
    },
    {
      "epoch": 2.5888,
      "grad_norm": 0.5138896107673645,
      "learning_rate": 2.1865787633432086e-05,
      "loss": 0.5679,
      "step": 2024
    },
    {
      "epoch": 2.59008,
      "grad_norm": 0.46441352367401123,
      "learning_rate": 2.1850875472386056e-05,
      "loss": 0.5523,
      "step": 2025
    },
    {
      "epoch": 2.59136,
      "grad_norm": 0.43304264545440674,
      "learning_rate": 2.183596341369575e-05,
      "loss": 0.567,
      "step": 2026
    },
    {
      "epoch": 2.59264,
      "grad_norm": 0.4271076023578644,
      "learning_rate": 2.1821051467596467e-05,
      "loss": 0.5546,
      "step": 2027
    },
    {
      "epoch": 2.59392,
      "grad_norm": 0.45855385065078735,
      "learning_rate": 2.180613964432344e-05,
      "loss": 0.5644,
      "step": 2028
    },
    {
      "epoch": 2.5952,
      "grad_norm": 0.4627922475337982,
      "learning_rate": 2.1791227954111816e-05,
      "loss": 0.5651,
      "step": 2029
    },
    {
      "epoch": 2.59648,
      "grad_norm": 0.43553051352500916,
      "learning_rate": 2.1776316407196626e-05,
      "loss": 0.579,
      "step": 2030
    },
    {
      "epoch": 2.59776,
      "grad_norm": 0.4542708992958069,
      "learning_rate": 2.1761405013812836e-05,
      "loss": 0.5617,
      "step": 2031
    },
    {
      "epoch": 2.59904,
      "grad_norm": 0.4563835859298706,
      "learning_rate": 2.174649378419528e-05,
      "loss": 0.5897,
      "step": 2032
    },
    {
      "epoch": 2.60032,
      "grad_norm": 0.42525970935821533,
      "learning_rate": 2.1731582728578694e-05,
      "loss": 0.5325,
      "step": 2033
    },
    {
      "epoch": 2.6016,
      "grad_norm": 0.41682469844818115,
      "learning_rate": 2.17166718571977e-05,
      "loss": 0.5416,
      "step": 2034
    },
    {
      "epoch": 2.60288,
      "grad_norm": 0.4311223328113556,
      "learning_rate": 2.1701761180286775e-05,
      "loss": 0.5389,
      "step": 2035
    },
    {
      "epoch": 2.6041600000000003,
      "grad_norm": 0.45937106013298035,
      "learning_rate": 2.1686850708080276e-05,
      "loss": 0.5676,
      "step": 2036
    },
    {
      "epoch": 2.6054399999999998,
      "grad_norm": 0.4074769616127014,
      "learning_rate": 2.1671940450812427e-05,
      "loss": 0.567,
      "step": 2037
    },
    {
      "epoch": 2.60672,
      "grad_norm": 0.42753535509109497,
      "learning_rate": 2.1657030418717275e-05,
      "loss": 0.5798,
      "step": 2038
    },
    {
      "epoch": 2.608,
      "grad_norm": 0.42000672221183777,
      "learning_rate": 2.1642120622028737e-05,
      "loss": 0.5812,
      "step": 2039
    },
    {
      "epoch": 2.60928,
      "grad_norm": 0.4018554985523224,
      "learning_rate": 2.162721107098058e-05,
      "loss": 0.5609,
      "step": 2040
    },
    {
      "epoch": 2.60928,
      "eval_loss": 0.6842366456985474,
      "eval_runtime": 19.5997,
      "eval_samples_per_second": 25.613,
      "eval_steps_per_second": 3.214,
      "step": 2040
    },
    {
      "epoch": 2.61056,
      "grad_norm": 0.4556508958339691,
      "learning_rate": 2.161230177580636e-05,
      "loss": 0.5738,
      "step": 2041
    },
    {
      "epoch": 2.61184,
      "grad_norm": 0.4296879768371582,
      "learning_rate": 2.1597392746739492e-05,
      "loss": 0.5534,
      "step": 2042
    },
    {
      "epoch": 2.61312,
      "grad_norm": 0.40791013836860657,
      "learning_rate": 2.1582483994013207e-05,
      "loss": 0.5725,
      "step": 2043
    },
    {
      "epoch": 2.6144,
      "grad_norm": 0.4300387501716614,
      "learning_rate": 2.156757552786052e-05,
      "loss": 0.5897,
      "step": 2044
    },
    {
      "epoch": 2.6156800000000002,
      "grad_norm": 0.4825383722782135,
      "learning_rate": 2.1552667358514285e-05,
      "loss": 0.566,
      "step": 2045
    },
    {
      "epoch": 2.6169599999999997,
      "grad_norm": 0.42368200421333313,
      "learning_rate": 2.153775949620713e-05,
      "loss": 0.5801,
      "step": 2046
    },
    {
      "epoch": 2.61824,
      "grad_norm": 0.42963165044784546,
      "learning_rate": 2.152285195117147e-05,
      "loss": 0.5876,
      "step": 2047
    },
    {
      "epoch": 2.61952,
      "grad_norm": 0.3960072994232178,
      "learning_rate": 2.1507944733639513e-05,
      "loss": 0.5522,
      "step": 2048
    },
    {
      "epoch": 2.6208,
      "grad_norm": 0.4466218054294586,
      "learning_rate": 2.149303785384324e-05,
      "loss": 0.5482,
      "step": 2049
    },
    {
      "epoch": 2.62208,
      "grad_norm": 0.494499534368515,
      "learning_rate": 2.14781313220144e-05,
      "loss": 0.5726,
      "step": 2050
    },
    {
      "epoch": 2.62336,
      "grad_norm": 0.43378692865371704,
      "learning_rate": 2.1463225148384495e-05,
      "loss": 0.5813,
      "step": 2051
    },
    {
      "epoch": 2.62464,
      "grad_norm": 0.4379490315914154,
      "learning_rate": 2.1448319343184792e-05,
      "loss": 0.5599,
      "step": 2052
    },
    {
      "epoch": 2.62592,
      "grad_norm": 0.43849682807922363,
      "learning_rate": 2.1433413916646305e-05,
      "loss": 0.5538,
      "step": 2053
    },
    {
      "epoch": 2.6272,
      "grad_norm": 0.41164514422416687,
      "learning_rate": 2.1418508878999774e-05,
      "loss": 0.6029,
      "step": 2054
    },
    {
      "epoch": 2.62848,
      "grad_norm": 0.46228301525115967,
      "learning_rate": 2.140360424047569e-05,
      "loss": 0.5732,
      "step": 2055
    },
    {
      "epoch": 2.62976,
      "grad_norm": 0.46438297629356384,
      "learning_rate": 2.138870001130426e-05,
      "loss": 0.5616,
      "step": 2056
    },
    {
      "epoch": 2.63104,
      "grad_norm": 0.4063985347747803,
      "learning_rate": 2.1373796201715415e-05,
      "loss": 0.5461,
      "step": 2057
    },
    {
      "epoch": 2.63232,
      "grad_norm": 0.46617591381073,
      "learning_rate": 2.1358892821938784e-05,
      "loss": 0.5634,
      "step": 2058
    },
    {
      "epoch": 2.6336,
      "grad_norm": 0.47225233912467957,
      "learning_rate": 2.1343989882203728e-05,
      "loss": 0.566,
      "step": 2059
    },
    {
      "epoch": 2.63488,
      "grad_norm": 0.46407338976860046,
      "learning_rate": 2.1329087392739277e-05,
      "loss": 0.5609,
      "step": 2060
    },
    {
      "epoch": 2.63616,
      "grad_norm": 0.47127047181129456,
      "learning_rate": 2.1314185363774168e-05,
      "loss": 0.5613,
      "step": 2061
    },
    {
      "epoch": 2.63744,
      "grad_norm": 0.4126521050930023,
      "learning_rate": 2.1299283805536825e-05,
      "loss": 0.5525,
      "step": 2062
    },
    {
      "epoch": 2.63872,
      "grad_norm": 0.5020531415939331,
      "learning_rate": 2.1284382728255343e-05,
      "loss": 0.5691,
      "step": 2063
    },
    {
      "epoch": 2.64,
      "grad_norm": 0.4251871109008789,
      "learning_rate": 2.1269482142157476e-05,
      "loss": 0.5589,
      "step": 2064
    },
    {
      "epoch": 2.64128,
      "grad_norm": 0.39278876781463623,
      "learning_rate": 2.125458205747066e-05,
      "loss": 0.5502,
      "step": 2065
    },
    {
      "epoch": 2.64256,
      "grad_norm": 0.4942920207977295,
      "learning_rate": 2.123968248442198e-05,
      "loss": 0.5687,
      "step": 2066
    },
    {
      "epoch": 2.64384,
      "grad_norm": 0.4744316041469574,
      "learning_rate": 2.1224783433238163e-05,
      "loss": 0.5753,
      "step": 2067
    },
    {
      "epoch": 2.64512,
      "grad_norm": 0.39959436655044556,
      "learning_rate": 2.1209884914145588e-05,
      "loss": 0.5616,
      "step": 2068
    },
    {
      "epoch": 2.6464,
      "grad_norm": 0.48772791028022766,
      "learning_rate": 2.119498693737026e-05,
      "loss": 0.5931,
      "step": 2069
    },
    {
      "epoch": 2.6476800000000003,
      "grad_norm": 0.4168209433555603,
      "learning_rate": 2.1180089513137815e-05,
      "loss": 0.5322,
      "step": 2070
    },
    {
      "epoch": 2.6476800000000003,
      "eval_loss": 0.6838758587837219,
      "eval_runtime": 19.5967,
      "eval_samples_per_second": 25.617,
      "eval_steps_per_second": 3.215,
      "step": 2070
    },
    {
      "epoch": 2.6489599999999998,
      "grad_norm": 0.41104644536972046,
      "learning_rate": 2.116519265167351e-05,
      "loss": 0.5711,
      "step": 2071
    },
    {
      "epoch": 2.65024,
      "grad_norm": 0.45247265696525574,
      "learning_rate": 2.1150296363202224e-05,
      "loss": 0.6058,
      "step": 2072
    },
    {
      "epoch": 2.65152,
      "grad_norm": 0.40062829852104187,
      "learning_rate": 2.113540065794842e-05,
      "loss": 0.5782,
      "step": 2073
    },
    {
      "epoch": 2.6528,
      "grad_norm": 0.41871193051338196,
      "learning_rate": 2.112050554613619e-05,
      "loss": 0.5901,
      "step": 2074
    },
    {
      "epoch": 2.65408,
      "grad_norm": 0.4067140221595764,
      "learning_rate": 2.1105611037989208e-05,
      "loss": 0.5754,
      "step": 2075
    },
    {
      "epoch": 2.65536,
      "grad_norm": 0.43430429697036743,
      "learning_rate": 2.1090717143730708e-05,
      "loss": 0.5585,
      "step": 2076
    },
    {
      "epoch": 2.65664,
      "grad_norm": 0.4349079430103302,
      "learning_rate": 2.107582387358354e-05,
      "loss": 0.6004,
      "step": 2077
    },
    {
      "epoch": 2.65792,
      "grad_norm": 0.43396127223968506,
      "learning_rate": 2.106093123777011e-05,
      "loss": 0.5653,
      "step": 2078
    },
    {
      "epoch": 2.6592000000000002,
      "grad_norm": 0.43549713492393494,
      "learning_rate": 2.1046039246512383e-05,
      "loss": 0.5975,
      "step": 2079
    },
    {
      "epoch": 2.6604799999999997,
      "grad_norm": 0.40730002522468567,
      "learning_rate": 2.103114791003189e-05,
      "loss": 0.5547,
      "step": 2080
    },
    {
      "epoch": 2.66176,
      "grad_norm": 0.43441450595855713,
      "learning_rate": 2.1016257238549716e-05,
      "loss": 0.5569,
      "step": 2081
    },
    {
      "epoch": 2.66304,
      "grad_norm": 0.40992671251296997,
      "learning_rate": 2.100136724228647e-05,
      "loss": 0.5887,
      "step": 2082
    },
    {
      "epoch": 2.66432,
      "grad_norm": 0.44157418608665466,
      "learning_rate": 2.098647793146233e-05,
      "loss": 0.5756,
      "step": 2083
    },
    {
      "epoch": 2.6656,
      "grad_norm": 0.3957017660140991,
      "learning_rate": 2.097158931629696e-05,
      "loss": 0.5777,
      "step": 2084
    },
    {
      "epoch": 2.66688,
      "grad_norm": 0.45218905806541443,
      "learning_rate": 2.095670140700959e-05,
      "loss": 0.6073,
      "step": 2085
    },
    {
      "epoch": 2.66816,
      "grad_norm": 0.4223405122756958,
      "learning_rate": 2.0941814213818944e-05,
      "loss": 0.5706,
      "step": 2086
    },
    {
      "epoch": 2.66944,
      "grad_norm": 0.3951118290424347,
      "learning_rate": 2.0926927746943243e-05,
      "loss": 0.5513,
      "step": 2087
    },
    {
      "epoch": 2.67072,
      "grad_norm": 0.3917636275291443,
      "learning_rate": 2.0912042016600237e-05,
      "loss": 0.5402,
      "step": 2088
    },
    {
      "epoch": 2.672,
      "grad_norm": 0.4029698967933655,
      "learning_rate": 2.0897157033007166e-05,
      "loss": 0.5714,
      "step": 2089
    },
    {
      "epoch": 2.67328,
      "grad_norm": 0.4382835328578949,
      "learning_rate": 2.0882272806380728e-05,
      "loss": 0.5705,
      "step": 2090
    },
    {
      "epoch": 2.67456,
      "grad_norm": 0.414702445268631,
      "learning_rate": 2.086738934693713e-05,
      "loss": 0.5677,
      "step": 2091
    },
    {
      "epoch": 2.67584,
      "grad_norm": 0.4208332300186157,
      "learning_rate": 2.0852506664892048e-05,
      "loss": 0.5579,
      "step": 2092
    },
    {
      "epoch": 2.67712,
      "grad_norm": 0.4284907281398773,
      "learning_rate": 2.083762477046061e-05,
      "loss": 0.5797,
      "step": 2093
    },
    {
      "epoch": 2.6784,
      "grad_norm": 0.41864898800849915,
      "learning_rate": 2.0822743673857424e-05,
      "loss": 0.5402,
      "step": 2094
    },
    {
      "epoch": 2.67968,
      "grad_norm": 0.4506852328777313,
      "learning_rate": 2.0807863385296544e-05,
      "loss": 0.5524,
      "step": 2095
    },
    {
      "epoch": 2.68096,
      "grad_norm": 0.3982718586921692,
      "learning_rate": 2.0792983914991453e-05,
      "loss": 0.5204,
      "step": 2096
    },
    {
      "epoch": 2.68224,
      "grad_norm": 0.43679866194725037,
      "learning_rate": 2.0778105273155095e-05,
      "loss": 0.5996,
      "step": 2097
    },
    {
      "epoch": 2.68352,
      "grad_norm": 0.4286017417907715,
      "learning_rate": 2.0763227469999834e-05,
      "loss": 0.5547,
      "step": 2098
    },
    {
      "epoch": 2.6848,
      "grad_norm": 0.4179457426071167,
      "learning_rate": 2.0748350515737456e-05,
      "loss": 0.563,
      "step": 2099
    },
    {
      "epoch": 2.68608,
      "grad_norm": 0.4385077655315399,
      "learning_rate": 2.0733474420579173e-05,
      "loss": 0.5559,
      "step": 2100
    },
    {
      "epoch": 2.68608,
      "eval_loss": 0.6828464865684509,
      "eval_runtime": 19.6144,
      "eval_samples_per_second": 25.593,
      "eval_steps_per_second": 3.212,
      "step": 2100
    },
    {
      "epoch": 2.68736,
      "grad_norm": 0.42880794405937195,
      "learning_rate": 2.07185991947356e-05,
      "loss": 0.5747,
      "step": 2101
    },
    {
      "epoch": 2.68864,
      "grad_norm": 0.4291785657405853,
      "learning_rate": 2.0703724848416757e-05,
      "loss": 0.5918,
      "step": 2102
    },
    {
      "epoch": 2.68992,
      "grad_norm": 0.43131014704704285,
      "learning_rate": 2.0688851391832058e-05,
      "loss": 0.5642,
      "step": 2103
    },
    {
      "epoch": 2.6912000000000003,
      "grad_norm": 0.40932610630989075,
      "learning_rate": 2.0673978835190324e-05,
      "loss": 0.5516,
      "step": 2104
    },
    {
      "epoch": 2.6924799999999998,
      "grad_norm": 0.4489363431930542,
      "learning_rate": 2.0659107188699722e-05,
      "loss": 0.5851,
      "step": 2105
    },
    {
      "epoch": 2.69376,
      "grad_norm": 0.41983991861343384,
      "learning_rate": 2.0644236462567836e-05,
      "loss": 0.544,
      "step": 2106
    },
    {
      "epoch": 2.69504,
      "grad_norm": 0.4725874364376068,
      "learning_rate": 2.062936666700159e-05,
      "loss": 0.574,
      "step": 2107
    },
    {
      "epoch": 2.69632,
      "grad_norm": 0.48117056488990784,
      "learning_rate": 2.0614497812207275e-05,
      "loss": 0.5606,
      "step": 2108
    },
    {
      "epoch": 2.6976,
      "grad_norm": 0.4447472095489502,
      "learning_rate": 2.0599629908390545e-05,
      "loss": 0.5605,
      "step": 2109
    },
    {
      "epoch": 2.69888,
      "grad_norm": 0.4650132954120636,
      "learning_rate": 2.058476296575639e-05,
      "loss": 0.5598,
      "step": 2110
    },
    {
      "epoch": 2.70016,
      "grad_norm": 0.45209425687789917,
      "learning_rate": 2.0569896994509146e-05,
      "loss": 0.5623,
      "step": 2111
    },
    {
      "epoch": 2.70144,
      "grad_norm": 0.42609545588493347,
      "learning_rate": 2.055503200485249e-05,
      "loss": 0.583,
      "step": 2112
    },
    {
      "epoch": 2.7027200000000002,
      "grad_norm": 0.44554242491722107,
      "learning_rate": 2.05401680069894e-05,
      "loss": 0.5861,
      "step": 2113
    },
    {
      "epoch": 2.7039999999999997,
      "grad_norm": 0.4989047050476074,
      "learning_rate": 2.0525305011122202e-05,
      "loss": 0.5881,
      "step": 2114
    },
    {
      "epoch": 2.70528,
      "grad_norm": 0.42148852348327637,
      "learning_rate": 2.051044302745253e-05,
      "loss": 0.5728,
      "step": 2115
    },
    {
      "epoch": 2.70656,
      "grad_norm": 0.4493227005004883,
      "learning_rate": 2.04955820661813e-05,
      "loss": 0.5526,
      "step": 2116
    },
    {
      "epoch": 2.70784,
      "grad_norm": 0.4848760664463043,
      "learning_rate": 2.0480722137508754e-05,
      "loss": 0.5498,
      "step": 2117
    },
    {
      "epoch": 2.70912,
      "grad_norm": 0.4156774580478668,
      "learning_rate": 2.046586325163441e-05,
      "loss": 0.5561,
      "step": 2118
    },
    {
      "epoch": 2.7104,
      "grad_norm": 0.4166371524333954,
      "learning_rate": 2.0451005418757058e-05,
      "loss": 0.5814,
      "step": 2119
    },
    {
      "epoch": 2.71168,
      "grad_norm": 0.39696004986763,
      "learning_rate": 2.043614864907481e-05,
      "loss": 0.5549,
      "step": 2120
    },
    {
      "epoch": 2.71296,
      "grad_norm": 0.4294726848602295,
      "learning_rate": 2.0421292952784996e-05,
      "loss": 0.6017,
      "step": 2121
    },
    {
      "epoch": 2.71424,
      "grad_norm": 0.4472660720348358,
      "learning_rate": 2.040643834008424e-05,
      "loss": 0.5397,
      "step": 2122
    },
    {
      "epoch": 2.71552,
      "grad_norm": 0.44376182556152344,
      "learning_rate": 2.0391584821168413e-05,
      "loss": 0.5643,
      "step": 2123
    },
    {
      "epoch": 2.7168,
      "grad_norm": 0.413866251707077,
      "learning_rate": 2.037673240623264e-05,
      "loss": 0.5903,
      "step": 2124
    },
    {
      "epoch": 2.71808,
      "grad_norm": 0.4692688584327698,
      "learning_rate": 2.0361881105471286e-05,
      "loss": 0.5757,
      "step": 2125
    },
    {
      "epoch": 2.71936,
      "grad_norm": 0.5268343687057495,
      "learning_rate": 2.0347030929077943e-05,
      "loss": 0.5685,
      "step": 2126
    },
    {
      "epoch": 2.72064,
      "grad_norm": 0.471332311630249,
      "learning_rate": 2.0332181887245443e-05,
      "loss": 0.5503,
      "step": 2127
    },
    {
      "epoch": 2.72192,
      "grad_norm": 0.4262884557247162,
      "learning_rate": 2.0317333990165838e-05,
      "loss": 0.5659,
      "step": 2128
    },
    {
      "epoch": 2.7232,
      "grad_norm": 0.4254116117954254,
      "learning_rate": 2.0302487248030385e-05,
      "loss": 0.5457,
      "step": 2129
    },
    {
      "epoch": 2.72448,
      "grad_norm": 0.4963558316230774,
      "learning_rate": 2.0287641671029564e-05,
      "loss": 0.5716,
      "step": 2130
    },
    {
      "epoch": 2.72448,
      "eval_loss": 0.6825251579284668,
      "eval_runtime": 19.6074,
      "eval_samples_per_second": 25.603,
      "eval_steps_per_second": 3.213,
      "step": 2130
    },
    {
      "epoch": 2.72576,
      "grad_norm": 0.4824458658695221,
      "learning_rate": 2.0272797269353033e-05,
      "loss": 0.5571,
      "step": 2131
    },
    {
      "epoch": 2.72704,
      "grad_norm": 0.5310393571853638,
      "learning_rate": 2.0257954053189657e-05,
      "loss": 0.5735,
      "step": 2132
    },
    {
      "epoch": 2.72832,
      "grad_norm": 0.4398851692676544,
      "learning_rate": 2.02431120327275e-05,
      "loss": 0.5847,
      "step": 2133
    },
    {
      "epoch": 2.7296,
      "grad_norm": 0.5228756666183472,
      "learning_rate": 2.0228271218153774e-05,
      "loss": 0.5863,
      "step": 2134
    },
    {
      "epoch": 2.73088,
      "grad_norm": 0.4445500671863556,
      "learning_rate": 2.0213431619654894e-05,
      "loss": 0.5318,
      "step": 2135
    },
    {
      "epoch": 2.73216,
      "grad_norm": 0.44130969047546387,
      "learning_rate": 2.019859324741642e-05,
      "loss": 0.5987,
      "step": 2136
    },
    {
      "epoch": 2.73344,
      "grad_norm": 0.4789541959762573,
      "learning_rate": 2.0183756111623084e-05,
      "loss": 0.5546,
      "step": 2137
    },
    {
      "epoch": 2.7347200000000003,
      "grad_norm": 0.43957746028900146,
      "learning_rate": 2.0168920222458754e-05,
      "loss": 0.5501,
      "step": 2138
    },
    {
      "epoch": 2.7359999999999998,
      "grad_norm": 0.43163958191871643,
      "learning_rate": 2.0154085590106455e-05,
      "loss": 0.5479,
      "step": 2139
    },
    {
      "epoch": 2.73728,
      "grad_norm": 0.4709853231906891,
      "learning_rate": 2.013925222474835e-05,
      "loss": 0.5763,
      "step": 2140
    },
    {
      "epoch": 2.73856,
      "grad_norm": 0.41731154918670654,
      "learning_rate": 2.0124420136565717e-05,
      "loss": 0.594,
      "step": 2141
    },
    {
      "epoch": 2.73984,
      "grad_norm": 0.478981614112854,
      "learning_rate": 2.010958933573898e-05,
      "loss": 0.568,
      "step": 2142
    },
    {
      "epoch": 2.74112,
      "grad_norm": 0.39418473839759827,
      "learning_rate": 2.009475983244766e-05,
      "loss": 0.5764,
      "step": 2143
    },
    {
      "epoch": 2.7424,
      "grad_norm": 0.4343644976615906,
      "learning_rate": 2.00799316368704e-05,
      "loss": 0.5751,
      "step": 2144
    },
    {
      "epoch": 2.74368,
      "grad_norm": 0.41238731145858765,
      "learning_rate": 2.0065104759184934e-05,
      "loss": 0.5646,
      "step": 2145
    },
    {
      "epoch": 2.74496,
      "grad_norm": 0.47756287455558777,
      "learning_rate": 2.0050279209568097e-05,
      "loss": 0.6018,
      "step": 2146
    },
    {
      "epoch": 2.7462400000000002,
      "grad_norm": 0.4455742835998535,
      "learning_rate": 2.0035454998195832e-05,
      "loss": 0.5688,
      "step": 2147
    },
    {
      "epoch": 2.7475199999999997,
      "grad_norm": 0.4909248948097229,
      "learning_rate": 2.002063213524312e-05,
      "loss": 0.5552,
      "step": 2148
    },
    {
      "epoch": 2.7488,
      "grad_norm": 0.4947395920753479,
      "learning_rate": 2.000581063088405e-05,
      "loss": 0.5674,
      "step": 2149
    },
    {
      "epoch": 2.75008,
      "grad_norm": 0.4388936161994934,
      "learning_rate": 1.999099049529178e-05,
      "loss": 0.5524,
      "step": 2150
    },
    {
      "epoch": 2.75136,
      "grad_norm": 0.5129539370536804,
      "learning_rate": 1.9976171738638504e-05,
      "loss": 0.5744,
      "step": 2151
    },
    {
      "epoch": 2.75264,
      "grad_norm": 0.5023990869522095,
      "learning_rate": 1.996135437109549e-05,
      "loss": 0.5724,
      "step": 2152
    },
    {
      "epoch": 2.75392,
      "grad_norm": 0.48078176379203796,
      "learning_rate": 1.9946538402833047e-05,
      "loss": 0.5881,
      "step": 2153
    },
    {
      "epoch": 2.7552,
      "grad_norm": 0.5419201254844666,
      "learning_rate": 1.993172384402052e-05,
      "loss": 0.5798,
      "step": 2154
    },
    {
      "epoch": 2.75648,
      "grad_norm": 0.4865698516368866,
      "learning_rate": 1.991691070482629e-05,
      "loss": 0.5464,
      "step": 2155
    },
    {
      "epoch": 2.75776,
      "grad_norm": 0.4413692355155945,
      "learning_rate": 1.990209899541777e-05,
      "loss": 0.5466,
      "step": 2156
    },
    {
      "epoch": 2.75904,
      "grad_norm": 0.4955664575099945,
      "learning_rate": 1.9887288725961374e-05,
      "loss": 0.5858,
      "step": 2157
    },
    {
      "epoch": 2.76032,
      "grad_norm": 0.46951162815093994,
      "learning_rate": 1.9872479906622538e-05,
      "loss": 0.5482,
      "step": 2158
    },
    {
      "epoch": 2.7616,
      "grad_norm": 0.455692321062088,
      "learning_rate": 1.985767254756572e-05,
      "loss": 0.5937,
      "step": 2159
    },
    {
      "epoch": 2.76288,
      "grad_norm": 0.43875402212142944,
      "learning_rate": 1.9842866658954336e-05,
      "loss": 0.5905,
      "step": 2160
    },
    {
      "epoch": 2.76288,
      "eval_loss": 0.6822736859321594,
      "eval_runtime": 19.6276,
      "eval_samples_per_second": 25.576,
      "eval_steps_per_second": 3.21,
      "step": 2160
    },
    {
      "epoch": 2.76416,
      "grad_norm": 0.42841261625289917,
      "learning_rate": 1.982806225095083e-05,
      "loss": 0.5882,
      "step": 2161
    },
    {
      "epoch": 2.76544,
      "grad_norm": 0.45809510350227356,
      "learning_rate": 1.981325933371661e-05,
      "loss": 0.537,
      "step": 2162
    },
    {
      "epoch": 2.76672,
      "grad_norm": 0.4625266492366791,
      "learning_rate": 1.979845791741206e-05,
      "loss": 0.5703,
      "step": 2163
    },
    {
      "epoch": 2.768,
      "grad_norm": 0.43833640217781067,
      "learning_rate": 1.9783658012196563e-05,
      "loss": 0.5794,
      "step": 2164
    },
    {
      "epoch": 2.76928,
      "grad_norm": 0.46834054589271545,
      "learning_rate": 1.9768859628228414e-05,
      "loss": 0.5692,
      "step": 2165
    },
    {
      "epoch": 2.77056,
      "grad_norm": 0.42364701628685,
      "learning_rate": 1.97540627756649e-05,
      "loss": 0.5641,
      "step": 2166
    },
    {
      "epoch": 2.77184,
      "grad_norm": 0.4180537760257721,
      "learning_rate": 1.973926746466226e-05,
      "loss": 0.5532,
      "step": 2167
    },
    {
      "epoch": 2.77312,
      "grad_norm": 0.41021454334259033,
      "learning_rate": 1.9724473705375645e-05,
      "loss": 0.5595,
      "step": 2168
    },
    {
      "epoch": 2.7744,
      "grad_norm": 0.44835081696510315,
      "learning_rate": 1.9709681507959174e-05,
      "loss": 0.5715,
      "step": 2169
    },
    {
      "epoch": 2.77568,
      "grad_norm": 0.4295932352542877,
      "learning_rate": 1.9694890882565874e-05,
      "loss": 0.5776,
      "step": 2170
    },
    {
      "epoch": 2.77696,
      "grad_norm": 0.4161606729030609,
      "learning_rate": 1.968010183934769e-05,
      "loss": 0.5826,
      "step": 2171
    },
    {
      "epoch": 2.7782400000000003,
      "grad_norm": 0.42526739835739136,
      "learning_rate": 1.9665314388455498e-05,
      "loss": 0.5675,
      "step": 2172
    },
    {
      "epoch": 2.7795199999999998,
      "grad_norm": 0.43559908866882324,
      "learning_rate": 1.9650528540039077e-05,
      "loss": 0.5643,
      "step": 2173
    },
    {
      "epoch": 2.7808,
      "grad_norm": 0.42948082089424133,
      "learning_rate": 1.963574430424709e-05,
      "loss": 0.5515,
      "step": 2174
    },
    {
      "epoch": 2.78208,
      "grad_norm": 0.4642879068851471,
      "learning_rate": 1.9620961691227106e-05,
      "loss": 0.5703,
      "step": 2175
    },
    {
      "epoch": 2.78336,
      "grad_norm": 0.3809150755405426,
      "learning_rate": 1.960618071112559e-05,
      "loss": 0.553,
      "step": 2176
    },
    {
      "epoch": 2.78464,
      "grad_norm": 0.43620896339416504,
      "learning_rate": 1.9591401374087855e-05,
      "loss": 0.5755,
      "step": 2177
    },
    {
      "epoch": 2.78592,
      "grad_norm": 0.4368182122707367,
      "learning_rate": 1.9576623690258117e-05,
      "loss": 0.5684,
      "step": 2178
    },
    {
      "epoch": 2.7872,
      "grad_norm": 0.42477947473526,
      "learning_rate": 1.956184766977946e-05,
      "loss": 0.5541,
      "step": 2179
    },
    {
      "epoch": 2.78848,
      "grad_norm": 0.41406384110450745,
      "learning_rate": 1.9547073322793788e-05,
      "loss": 0.5656,
      "step": 2180
    },
    {
      "epoch": 2.7897600000000002,
      "grad_norm": 0.44187867641448975,
      "learning_rate": 1.9532300659441895e-05,
      "loss": 0.5345,
      "step": 2181
    },
    {
      "epoch": 2.7910399999999997,
      "grad_norm": 0.4068676829338074,
      "learning_rate": 1.9517529689863406e-05,
      "loss": 0.5535,
      "step": 2182
    },
    {
      "epoch": 2.79232,
      "grad_norm": 0.4106282889842987,
      "learning_rate": 1.9502760424196773e-05,
      "loss": 0.5715,
      "step": 2183
    },
    {
      "epoch": 2.7936,
      "grad_norm": 0.41989609599113464,
      "learning_rate": 1.9487992872579297e-05,
      "loss": 0.5858,
      "step": 2184
    },
    {
      "epoch": 2.79488,
      "grad_norm": 0.41973504424095154,
      "learning_rate": 1.9473227045147096e-05,
      "loss": 0.5844,
      "step": 2185
    },
    {
      "epoch": 2.79616,
      "grad_norm": 0.4133204519748688,
      "learning_rate": 1.9458462952035088e-05,
      "loss": 0.5656,
      "step": 2186
    },
    {
      "epoch": 2.79744,
      "grad_norm": 0.42403319478034973,
      "learning_rate": 1.9443700603377037e-05,
      "loss": 0.5592,
      "step": 2187
    },
    {
      "epoch": 2.79872,
      "grad_norm": 0.4255143404006958,
      "learning_rate": 1.9428940009305462e-05,
      "loss": 0.5739,
      "step": 2188
    },
    {
      "epoch": 2.8,
      "grad_norm": 0.4249035120010376,
      "learning_rate": 1.9414181179951715e-05,
      "loss": 0.5594,
      "step": 2189
    },
    {
      "epoch": 2.80128,
      "grad_norm": 0.4300551116466522,
      "learning_rate": 1.939942412544594e-05,
      "loss": 0.5895,
      "step": 2190
    },
    {
      "epoch": 2.80128,
      "eval_loss": 0.68100905418396,
      "eval_runtime": 19.622,
      "eval_samples_per_second": 25.584,
      "eval_steps_per_second": 3.211,
      "step": 2190
    },
    {
      "epoch": 2.80256,
      "grad_norm": 0.43244194984436035,
      "learning_rate": 1.938466885591702e-05,
      "loss": 0.5579,
      "step": 2191
    },
    {
      "epoch": 2.80384,
      "grad_norm": 0.4225500226020813,
      "learning_rate": 1.9369915381492655e-05,
      "loss": 0.5936,
      "step": 2192
    },
    {
      "epoch": 2.80512,
      "grad_norm": 0.4501644968986511,
      "learning_rate": 1.935516371229931e-05,
      "loss": 0.5708,
      "step": 2193
    },
    {
      "epoch": 2.8064,
      "grad_norm": 0.4456457793712616,
      "learning_rate": 1.9340413858462174e-05,
      "loss": 0.56,
      "step": 2194
    },
    {
      "epoch": 2.80768,
      "grad_norm": 0.470650851726532,
      "learning_rate": 1.9325665830105232e-05,
      "loss": 0.5653,
      "step": 2195
    },
    {
      "epoch": 2.80896,
      "grad_norm": 0.4132485091686249,
      "learning_rate": 1.9310919637351202e-05,
      "loss": 0.5718,
      "step": 2196
    },
    {
      "epoch": 2.81024,
      "grad_norm": 0.4300811290740967,
      "learning_rate": 1.9296175290321527e-05,
      "loss": 0.5555,
      "step": 2197
    },
    {
      "epoch": 2.81152,
      "grad_norm": 0.4188995361328125,
      "learning_rate": 1.9281432799136398e-05,
      "loss": 0.5621,
      "step": 2198
    },
    {
      "epoch": 2.8128,
      "grad_norm": 0.4360223710536957,
      "learning_rate": 1.9266692173914747e-05,
      "loss": 0.5783,
      "step": 2199
    },
    {
      "epoch": 2.81408,
      "grad_norm": 0.42980456352233887,
      "learning_rate": 1.9251953424774185e-05,
      "loss": 0.5538,
      "step": 2200
    },
    {
      "epoch": 2.81536,
      "grad_norm": 0.4526362717151642,
      "learning_rate": 1.923721656183107e-05,
      "loss": 0.5884,
      "step": 2201
    },
    {
      "epoch": 2.81664,
      "grad_norm": 0.4404382109642029,
      "learning_rate": 1.9222481595200456e-05,
      "loss": 0.607,
      "step": 2202
    },
    {
      "epoch": 2.81792,
      "grad_norm": 0.40499237179756165,
      "learning_rate": 1.9207748534996086e-05,
      "loss": 0.5651,
      "step": 2203
    },
    {
      "epoch": 2.8192,
      "grad_norm": 0.4323599934577942,
      "learning_rate": 1.91930173913304e-05,
      "loss": 0.5588,
      "step": 2204
    },
    {
      "epoch": 2.82048,
      "grad_norm": 0.44055265188217163,
      "learning_rate": 1.9178288174314535e-05,
      "loss": 0.5922,
      "step": 2205
    },
    {
      "epoch": 2.8217600000000003,
      "grad_norm": 0.4403923451900482,
      "learning_rate": 1.916356089405829e-05,
      "loss": 0.5808,
      "step": 2206
    },
    {
      "epoch": 2.8230399999999998,
      "grad_norm": 0.4582671821117401,
      "learning_rate": 1.914883556067013e-05,
      "loss": 0.584,
      "step": 2207
    },
    {
      "epoch": 2.82432,
      "grad_norm": 0.44535231590270996,
      "learning_rate": 1.9134112184257208e-05,
      "loss": 0.5549,
      "step": 2208
    },
    {
      "epoch": 2.8256,
      "grad_norm": 0.42243677377700806,
      "learning_rate": 1.9119390774925315e-05,
      "loss": 0.5496,
      "step": 2209
    },
    {
      "epoch": 2.82688,
      "grad_norm": 0.44139552116394043,
      "learning_rate": 1.9104671342778883e-05,
      "loss": 0.5653,
      "step": 2210
    },
    {
      "epoch": 2.82816,
      "grad_norm": 0.4750986695289612,
      "learning_rate": 1.908995389792102e-05,
      "loss": 0.5784,
      "step": 2211
    },
    {
      "epoch": 2.82944,
      "grad_norm": 0.38217708468437195,
      "learning_rate": 1.907523845045343e-05,
      "loss": 0.583,
      "step": 2212
    },
    {
      "epoch": 2.83072,
      "grad_norm": 0.47178032994270325,
      "learning_rate": 1.906052501047648e-05,
      "loss": 0.5883,
      "step": 2213
    },
    {
      "epoch": 2.832,
      "grad_norm": 0.41022613644599915,
      "learning_rate": 1.9045813588089133e-05,
      "loss": 0.5647,
      "step": 2214
    },
    {
      "epoch": 2.8332800000000002,
      "grad_norm": 0.4145445227622986,
      "learning_rate": 1.903110419338899e-05,
      "loss": 0.5895,
      "step": 2215
    },
    {
      "epoch": 2.8345599999999997,
      "grad_norm": 0.39259931445121765,
      "learning_rate": 1.9016396836472248e-05,
      "loss": 0.5506,
      "step": 2216
    },
    {
      "epoch": 2.83584,
      "grad_norm": 0.3897520899772644,
      "learning_rate": 1.900169152743369e-05,
      "loss": 0.5552,
      "step": 2217
    },
    {
      "epoch": 2.83712,
      "grad_norm": 0.4061276614665985,
      "learning_rate": 1.8986988276366732e-05,
      "loss": 0.5749,
      "step": 2218
    },
    {
      "epoch": 2.8384,
      "grad_norm": 0.41867735981941223,
      "learning_rate": 1.8972287093363345e-05,
      "loss": 0.5705,
      "step": 2219
    },
    {
      "epoch": 2.83968,
      "grad_norm": 0.43415945768356323,
      "learning_rate": 1.895758798851409e-05,
      "loss": 0.5645,
      "step": 2220
    },
    {
      "epoch": 2.83968,
      "eval_loss": 0.681098461151123,
      "eval_runtime": 19.5985,
      "eval_samples_per_second": 25.614,
      "eval_steps_per_second": 3.215,
      "step": 2220
    },
    {
      "epoch": 2.84096,
      "grad_norm": 0.42419615387916565,
      "learning_rate": 1.8942890971908105e-05,
      "loss": 0.5772,
      "step": 2221
    },
    {
      "epoch": 2.84224,
      "grad_norm": 0.44453462958335876,
      "learning_rate": 1.892819605363309e-05,
      "loss": 0.563,
      "step": 2222
    },
    {
      "epoch": 2.84352,
      "grad_norm": 0.4272880554199219,
      "learning_rate": 1.891350324377531e-05,
      "loss": 0.5457,
      "step": 2223
    },
    {
      "epoch": 2.8448,
      "grad_norm": 0.47611308097839355,
      "learning_rate": 1.889881255241957e-05,
      "loss": 0.566,
      "step": 2224
    },
    {
      "epoch": 2.84608,
      "grad_norm": 0.4296311140060425,
      "learning_rate": 1.888412398964925e-05,
      "loss": 0.5459,
      "step": 2225
    },
    {
      "epoch": 2.84736,
      "grad_norm": 0.44485270977020264,
      "learning_rate": 1.886943756554623e-05,
      "loss": 0.567,
      "step": 2226
    },
    {
      "epoch": 2.84864,
      "grad_norm": 0.46021559834480286,
      "learning_rate": 1.885475329019095e-05,
      "loss": 0.5828,
      "step": 2227
    },
    {
      "epoch": 2.84992,
      "grad_norm": 0.4361613392829895,
      "learning_rate": 1.8840071173662374e-05,
      "loss": 0.5744,
      "step": 2228
    },
    {
      "epoch": 2.8512,
      "grad_norm": 0.4244159162044525,
      "learning_rate": 1.882539122603796e-05,
      "loss": 0.5814,
      "step": 2229
    },
    {
      "epoch": 2.85248,
      "grad_norm": 0.4016723334789276,
      "learning_rate": 1.88107134573937e-05,
      "loss": 0.594,
      "step": 2230
    },
    {
      "epoch": 2.85376,
      "grad_norm": 0.4040704369544983,
      "learning_rate": 1.8796037877804098e-05,
      "loss": 0.5745,
      "step": 2231
    },
    {
      "epoch": 2.85504,
      "grad_norm": 0.4339129626750946,
      "learning_rate": 1.8781364497342128e-05,
      "loss": 0.5593,
      "step": 2232
    },
    {
      "epoch": 2.85632,
      "grad_norm": 0.45570647716522217,
      "learning_rate": 1.8766693326079274e-05,
      "loss": 0.5998,
      "step": 2233
    },
    {
      "epoch": 2.8576,
      "grad_norm": 0.43942537903785706,
      "learning_rate": 1.8752024374085503e-05,
      "loss": 0.5609,
      "step": 2234
    },
    {
      "epoch": 2.85888,
      "grad_norm": 0.43522512912750244,
      "learning_rate": 1.873735765142924e-05,
      "loss": 0.5637,
      "step": 2235
    },
    {
      "epoch": 2.86016,
      "grad_norm": 0.4488886892795563,
      "learning_rate": 1.87226931681774e-05,
      "loss": 0.5405,
      "step": 2236
    },
    {
      "epoch": 2.86144,
      "grad_norm": 0.4802055060863495,
      "learning_rate": 1.8708030934395373e-05,
      "loss": 0.5911,
      "step": 2237
    },
    {
      "epoch": 2.86272,
      "grad_norm": 0.46101969480514526,
      "learning_rate": 1.8693370960146958e-05,
      "loss": 0.5684,
      "step": 2238
    },
    {
      "epoch": 2.864,
      "grad_norm": 0.43480825424194336,
      "learning_rate": 1.8678713255494455e-05,
      "loss": 0.5452,
      "step": 2239
    },
    {
      "epoch": 2.8652800000000003,
      "grad_norm": 0.4606250524520874,
      "learning_rate": 1.8664057830498565e-05,
      "loss": 0.5546,
      "step": 2240
    },
    {
      "epoch": 2.8665599999999998,
      "grad_norm": 0.46391162276268005,
      "learning_rate": 1.8649404695218455e-05,
      "loss": 0.5796,
      "step": 2241
    },
    {
      "epoch": 2.86784,
      "grad_norm": 0.42897844314575195,
      "learning_rate": 1.8634753859711704e-05,
      "loss": 0.5764,
      "step": 2242
    },
    {
      "epoch": 2.86912,
      "grad_norm": 0.4244156777858734,
      "learning_rate": 1.862010533403432e-05,
      "loss": 0.5665,
      "step": 2243
    },
    {
      "epoch": 2.8704,
      "grad_norm": 0.46743395924568176,
      "learning_rate": 1.860545912824071e-05,
      "loss": 0.5768,
      "step": 2244
    },
    {
      "epoch": 2.87168,
      "grad_norm": 0.4175646901130676,
      "learning_rate": 1.8590815252383722e-05,
      "loss": 0.5613,
      "step": 2245
    },
    {
      "epoch": 2.87296,
      "grad_norm": 0.40576890110969543,
      "learning_rate": 1.857617371651456e-05,
      "loss": 0.5601,
      "step": 2246
    },
    {
      "epoch": 2.87424,
      "grad_norm": 0.4206882119178772,
      "learning_rate": 1.856153453068286e-05,
      "loss": 0.5557,
      "step": 2247
    },
    {
      "epoch": 2.87552,
      "grad_norm": 0.3663040101528168,
      "learning_rate": 1.8546897704936633e-05,
      "loss": 0.521,
      "step": 2248
    },
    {
      "epoch": 2.8768000000000002,
      "grad_norm": 0.4158974885940552,
      "learning_rate": 1.8532263249322255e-05,
      "loss": 0.5687,
      "step": 2249
    },
    {
      "epoch": 2.8780799999999997,
      "grad_norm": 0.3926350474357605,
      "learning_rate": 1.8517631173884495e-05,
      "loss": 0.5642,
      "step": 2250
    },
    {
      "epoch": 2.8780799999999997,
      "eval_loss": 0.68003910779953,
      "eval_runtime": 19.5859,
      "eval_samples_per_second": 25.631,
      "eval_steps_per_second": 3.217,
      "step": 2250
    },
    {
      "epoch": 2.87936,
      "grad_norm": 0.45268553495407104,
      "learning_rate": 1.850300148866648e-05,
      "loss": 0.5601,
      "step": 2251
    },
    {
      "epoch": 2.88064,
      "grad_norm": 0.4440188407897949,
      "learning_rate": 1.8488374203709692e-05,
      "loss": 0.5568,
      "step": 2252
    },
    {
      "epoch": 2.88192,
      "grad_norm": 0.3886945843696594,
      "learning_rate": 1.8473749329053977e-05,
      "loss": 0.5885,
      "step": 2253
    },
    {
      "epoch": 2.8832,
      "grad_norm": 0.41918349266052246,
      "learning_rate": 1.845912687473753e-05,
      "loss": 0.5356,
      "step": 2254
    },
    {
      "epoch": 2.88448,
      "grad_norm": 0.4112301170825958,
      "learning_rate": 1.8444506850796852e-05,
      "loss": 0.5809,
      "step": 2255
    },
    {
      "epoch": 2.88576,
      "grad_norm": 0.4210822880268097,
      "learning_rate": 1.8429889267266818e-05,
      "loss": 0.5741,
      "step": 2256
    },
    {
      "epoch": 2.88704,
      "grad_norm": 0.4469418525695801,
      "learning_rate": 1.8415274134180605e-05,
      "loss": 0.569,
      "step": 2257
    },
    {
      "epoch": 2.88832,
      "grad_norm": 0.4153585731983185,
      "learning_rate": 1.8400661461569705e-05,
      "loss": 0.5849,
      "step": 2258
    },
    {
      "epoch": 2.8895999999999997,
      "grad_norm": 0.4336307942867279,
      "learning_rate": 1.8386051259463928e-05,
      "loss": 0.5522,
      "step": 2259
    },
    {
      "epoch": 2.89088,
      "grad_norm": 0.3883611261844635,
      "learning_rate": 1.8371443537891406e-05,
      "loss": 0.5576,
      "step": 2260
    },
    {
      "epoch": 2.89216,
      "grad_norm": 0.425616979598999,
      "learning_rate": 1.835683830687853e-05,
      "loss": 0.5764,
      "step": 2261
    },
    {
      "epoch": 2.89344,
      "grad_norm": 0.39636480808258057,
      "learning_rate": 1.834223557645001e-05,
      "loss": 0.5164,
      "step": 2262
    },
    {
      "epoch": 2.89472,
      "grad_norm": 0.43521809577941895,
      "learning_rate": 1.832763535662884e-05,
      "loss": 0.5615,
      "step": 2263
    },
    {
      "epoch": 2.896,
      "grad_norm": 0.38103070855140686,
      "learning_rate": 1.831303765743627e-05,
      "loss": 0.5675,
      "step": 2264
    },
    {
      "epoch": 2.89728,
      "grad_norm": 0.3841995596885681,
      "learning_rate": 1.829844248889184e-05,
      "loss": 0.5652,
      "step": 2265
    },
    {
      "epoch": 2.89856,
      "grad_norm": 0.42928850650787354,
      "learning_rate": 1.828384986101335e-05,
      "loss": 0.5759,
      "step": 2266
    },
    {
      "epoch": 2.89984,
      "grad_norm": 0.43955132365226746,
      "learning_rate": 1.8269259783816838e-05,
      "loss": 0.5409,
      "step": 2267
    },
    {
      "epoch": 2.90112,
      "grad_norm": 0.4013601839542389,
      "learning_rate": 1.8254672267316625e-05,
      "loss": 0.567,
      "step": 2268
    },
    {
      "epoch": 2.9024,
      "grad_norm": 0.49128490686416626,
      "learning_rate": 1.8240087321525236e-05,
      "loss": 0.5693,
      "step": 2269
    },
    {
      "epoch": 2.90368,
      "grad_norm": 0.4039883017539978,
      "learning_rate": 1.822550495645346e-05,
      "loss": 0.5651,
      "step": 2270
    },
    {
      "epoch": 2.90496,
      "grad_norm": 0.4066225588321686,
      "learning_rate": 1.8210925182110305e-05,
      "loss": 0.519,
      "step": 2271
    },
    {
      "epoch": 2.90624,
      "grad_norm": 0.4312953054904938,
      "learning_rate": 1.8196348008502993e-05,
      "loss": 0.5723,
      "step": 2272
    },
    {
      "epoch": 2.90752,
      "grad_norm": 0.4298991560935974,
      "learning_rate": 1.8181773445636975e-05,
      "loss": 0.5644,
      "step": 2273
    },
    {
      "epoch": 2.9088000000000003,
      "grad_norm": 0.41935890913009644,
      "learning_rate": 1.816720150351591e-05,
      "loss": 0.5768,
      "step": 2274
    },
    {
      "epoch": 2.91008,
      "grad_norm": 0.44522449374198914,
      "learning_rate": 1.815263219214164e-05,
      "loss": 0.5439,
      "step": 2275
    },
    {
      "epoch": 2.91136,
      "grad_norm": 0.4181341528892517,
      "learning_rate": 1.813806552151421e-05,
      "loss": 0.5599,
      "step": 2276
    },
    {
      "epoch": 2.91264,
      "grad_norm": 0.4270663261413574,
      "learning_rate": 1.8123501501631874e-05,
      "loss": 0.5597,
      "step": 2277
    },
    {
      "epoch": 2.91392,
      "grad_norm": 0.3964351713657379,
      "learning_rate": 1.810894014249103e-05,
      "loss": 0.5514,
      "step": 2278
    },
    {
      "epoch": 2.9152,
      "grad_norm": 0.41637828946113586,
      "learning_rate": 1.8094381454086278e-05,
      "loss": 0.566,
      "step": 2279
    },
    {
      "epoch": 2.91648,
      "grad_norm": 0.41652923822402954,
      "learning_rate": 1.8079825446410376e-05,
      "loss": 0.5589,
      "step": 2280
    },
    {
      "epoch": 2.91648,
      "eval_loss": 0.6795461177825928,
      "eval_runtime": 19.5894,
      "eval_samples_per_second": 25.626,
      "eval_steps_per_second": 3.216,
      "step": 2280
    },
    {
      "epoch": 2.91776,
      "grad_norm": 0.4093642830848694,
      "learning_rate": 1.806527212945423e-05,
      "loss": 0.5491,
      "step": 2281
    },
    {
      "epoch": 2.91904,
      "grad_norm": 0.3952707350254059,
      "learning_rate": 1.8050721513206913e-05,
      "loss": 0.5415,
      "step": 2282
    },
    {
      "epoch": 2.9203200000000002,
      "grad_norm": 0.4046980142593384,
      "learning_rate": 1.803617360765565e-05,
      "loss": 0.5693,
      "step": 2283
    },
    {
      "epoch": 2.9215999999999998,
      "grad_norm": 0.47894731163978577,
      "learning_rate": 1.8021628422785796e-05,
      "loss": 0.5731,
      "step": 2284
    },
    {
      "epoch": 2.92288,
      "grad_norm": 0.45544588565826416,
      "learning_rate": 1.8007085968580825e-05,
      "loss": 0.5652,
      "step": 2285
    },
    {
      "epoch": 2.92416,
      "grad_norm": 0.3870166540145874,
      "learning_rate": 1.7992546255022366e-05,
      "loss": 0.5514,
      "step": 2286
    },
    {
      "epoch": 2.92544,
      "grad_norm": 0.47565239667892456,
      "learning_rate": 1.7978009292090145e-05,
      "loss": 0.5558,
      "step": 2287
    },
    {
      "epoch": 2.92672,
      "grad_norm": 0.41594594717025757,
      "learning_rate": 1.7963475089762002e-05,
      "loss": 0.5382,
      "step": 2288
    },
    {
      "epoch": 2.928,
      "grad_norm": 0.4429880678653717,
      "learning_rate": 1.79489436580139e-05,
      "loss": 0.5777,
      "step": 2289
    },
    {
      "epoch": 2.92928,
      "grad_norm": 0.43753665685653687,
      "learning_rate": 1.7934415006819882e-05,
      "loss": 0.5744,
      "step": 2290
    },
    {
      "epoch": 2.93056,
      "grad_norm": 0.3929087519645691,
      "learning_rate": 1.791988914615208e-05,
      "loss": 0.574,
      "step": 2291
    },
    {
      "epoch": 2.9318400000000002,
      "grad_norm": 0.45288240909576416,
      "learning_rate": 1.790536608598073e-05,
      "loss": 0.5926,
      "step": 2292
    },
    {
      "epoch": 2.9331199999999997,
      "grad_norm": 0.44063621759414673,
      "learning_rate": 1.7890845836274126e-05,
      "loss": 0.5793,
      "step": 2293
    },
    {
      "epoch": 2.9344,
      "grad_norm": 0.38857123255729675,
      "learning_rate": 1.7876328406998644e-05,
      "loss": 0.5492,
      "step": 2294
    },
    {
      "epoch": 2.93568,
      "grad_norm": 0.42111659049987793,
      "learning_rate": 1.786181380811872e-05,
      "loss": 0.5827,
      "step": 2295
    },
    {
      "epoch": 2.93696,
      "grad_norm": 0.47171756625175476,
      "learning_rate": 1.7847302049596855e-05,
      "loss": 0.5675,
      "step": 2296
    },
    {
      "epoch": 2.93824,
      "grad_norm": 0.4484410583972931,
      "learning_rate": 1.7832793141393584e-05,
      "loss": 0.5528,
      "step": 2297
    },
    {
      "epoch": 2.93952,
      "grad_norm": 0.4297260344028473,
      "learning_rate": 1.7818287093467504e-05,
      "loss": 0.577,
      "step": 2298
    },
    {
      "epoch": 2.9408,
      "grad_norm": 0.408372700214386,
      "learning_rate": 1.780378391577524e-05,
      "loss": 0.5666,
      "step": 2299
    },
    {
      "epoch": 2.94208,
      "grad_norm": 0.42754632234573364,
      "learning_rate": 1.778928361827144e-05,
      "loss": 0.5837,
      "step": 2300
    },
    {
      "epoch": 2.94336,
      "grad_norm": 0.49496400356292725,
      "learning_rate": 1.7774786210908784e-05,
      "loss": 0.5923,
      "step": 2301
    },
    {
      "epoch": 2.94464,
      "grad_norm": 0.40822485089302063,
      "learning_rate": 1.7760291703637973e-05,
      "loss": 0.5697,
      "step": 2302
    },
    {
      "epoch": 2.94592,
      "grad_norm": 0.42243319749832153,
      "learning_rate": 1.7745800106407717e-05,
      "loss": 0.5808,
      "step": 2303
    },
    {
      "epoch": 2.9472,
      "grad_norm": 0.4468790888786316,
      "learning_rate": 1.7731311429164707e-05,
      "loss": 0.5417,
      "step": 2304
    },
    {
      "epoch": 2.94848,
      "grad_norm": 0.4059481620788574,
      "learning_rate": 1.771682568185365e-05,
      "loss": 0.5458,
      "step": 2305
    },
    {
      "epoch": 2.94976,
      "grad_norm": 0.4323550760746002,
      "learning_rate": 1.770234287441725e-05,
      "loss": 0.6062,
      "step": 2306
    },
    {
      "epoch": 2.95104,
      "grad_norm": 0.43020176887512207,
      "learning_rate": 1.768786301679616e-05,
      "loss": 0.5742,
      "step": 2307
    },
    {
      "epoch": 2.9523200000000003,
      "grad_norm": 0.4239875078201294,
      "learning_rate": 1.7673386118929044e-05,
      "loss": 0.5801,
      "step": 2308
    },
    {
      "epoch": 2.9536,
      "grad_norm": 0.41328591108322144,
      "learning_rate": 1.7658912190752522e-05,
      "loss": 0.58,
      "step": 2309
    },
    {
      "epoch": 2.95488,
      "grad_norm": 0.4064328670501709,
      "learning_rate": 1.7644441242201162e-05,
      "loss": 0.5813,
      "step": 2310
    },
    {
      "epoch": 2.95488,
      "eval_loss": 0.6797740459442139,
      "eval_runtime": 19.6258,
      "eval_samples_per_second": 25.579,
      "eval_steps_per_second": 3.21,
      "step": 2310
    },
    {
      "epoch": 2.95616,
      "grad_norm": 0.42762699723243713,
      "learning_rate": 1.76299732832075e-05,
      "loss": 0.5465,
      "step": 2311
    },
    {
      "epoch": 2.95744,
      "grad_norm": 0.44760847091674805,
      "learning_rate": 1.761550832370203e-05,
      "loss": 0.575,
      "step": 2312
    },
    {
      "epoch": 2.95872,
      "grad_norm": 0.4181264638900757,
      "learning_rate": 1.760104637361316e-05,
      "loss": 0.5839,
      "step": 2313
    },
    {
      "epoch": 2.96,
      "grad_norm": 0.4497205913066864,
      "learning_rate": 1.758658744286725e-05,
      "loss": 0.5681,
      "step": 2314
    },
    {
      "epoch": 2.96128,
      "grad_norm": 0.4380146563053131,
      "learning_rate": 1.75721315413886e-05,
      "loss": 0.5659,
      "step": 2315
    },
    {
      "epoch": 2.96256,
      "grad_norm": 0.40518519282341003,
      "learning_rate": 1.75576786790994e-05,
      "loss": 0.5701,
      "step": 2316
    },
    {
      "epoch": 2.9638400000000003,
      "grad_norm": 0.4426787495613098,
      "learning_rate": 1.754322886591977e-05,
      "loss": 0.5676,
      "step": 2317
    },
    {
      "epoch": 2.9651199999999998,
      "grad_norm": 0.4340369701385498,
      "learning_rate": 1.7528782111767754e-05,
      "loss": 0.5747,
      "step": 2318
    },
    {
      "epoch": 2.9664,
      "grad_norm": 0.398754745721817,
      "learning_rate": 1.751433842655926e-05,
      "loss": 0.5522,
      "step": 2319
    },
    {
      "epoch": 2.96768,
      "grad_norm": 0.44134482741355896,
      "learning_rate": 1.7499897820208125e-05,
      "loss": 0.5798,
      "step": 2320
    },
    {
      "epoch": 2.96896,
      "grad_norm": 0.44437381625175476,
      "learning_rate": 1.7485460302626042e-05,
      "loss": 0.5817,
      "step": 2321
    },
    {
      "epoch": 2.97024,
      "grad_norm": 0.4558928906917572,
      "learning_rate": 1.7471025883722604e-05,
      "loss": 0.5531,
      "step": 2322
    },
    {
      "epoch": 2.97152,
      "grad_norm": 0.41100287437438965,
      "learning_rate": 1.7456594573405284e-05,
      "loss": 0.5795,
      "step": 2323
    },
    {
      "epoch": 2.9728,
      "grad_norm": 0.43487975001335144,
      "learning_rate": 1.7442166381579392e-05,
      "loss": 0.5876,
      "step": 2324
    },
    {
      "epoch": 2.97408,
      "grad_norm": 0.48762452602386475,
      "learning_rate": 1.7427741318148124e-05,
      "loss": 0.5719,
      "step": 2325
    },
    {
      "epoch": 2.9753600000000002,
      "grad_norm": 0.3819199204444885,
      "learning_rate": 1.7413319393012524e-05,
      "loss": 0.5553,
      "step": 2326
    },
    {
      "epoch": 2.9766399999999997,
      "grad_norm": 0.4146660566329956,
      "learning_rate": 1.7398900616071457e-05,
      "loss": 0.545,
      "step": 2327
    },
    {
      "epoch": 2.97792,
      "grad_norm": 0.4211333096027374,
      "learning_rate": 1.738448499722167e-05,
      "loss": 0.5408,
      "step": 2328
    },
    {
      "epoch": 2.9792,
      "grad_norm": 0.40616315603256226,
      "learning_rate": 1.7370072546357713e-05,
      "loss": 0.5591,
      "step": 2329
    },
    {
      "epoch": 2.98048,
      "grad_norm": 0.3882712423801422,
      "learning_rate": 1.7355663273371965e-05,
      "loss": 0.5434,
      "step": 2330
    },
    {
      "epoch": 2.98176,
      "grad_norm": 0.43389037251472473,
      "learning_rate": 1.7341257188154625e-05,
      "loss": 0.5691,
      "step": 2331
    },
    {
      "epoch": 2.98304,
      "grad_norm": 0.40227407217025757,
      "learning_rate": 1.7326854300593723e-05,
      "loss": 0.5505,
      "step": 2332
    },
    {
      "epoch": 2.98432,
      "grad_norm": 0.3852725327014923,
      "learning_rate": 1.7312454620575055e-05,
      "loss": 0.5815,
      "step": 2333
    },
    {
      "epoch": 2.9856,
      "grad_norm": 0.46819984912872314,
      "learning_rate": 1.729805815798225e-05,
      "loss": 0.5789,
      "step": 2334
    },
    {
      "epoch": 2.98688,
      "grad_norm": 0.4047696888446808,
      "learning_rate": 1.728366492269672e-05,
      "loss": 0.5849,
      "step": 2335
    },
    {
      "epoch": 2.98816,
      "grad_norm": 0.44423624873161316,
      "learning_rate": 1.7269274924597644e-05,
      "loss": 0.5495,
      "step": 2336
    },
    {
      "epoch": 2.98944,
      "grad_norm": 0.4462408721446991,
      "learning_rate": 1.725488817356201e-05,
      "loss": 0.558,
      "step": 2337
    },
    {
      "epoch": 2.99072,
      "grad_norm": 0.43430182337760925,
      "learning_rate": 1.7240504679464554e-05,
      "loss": 0.5523,
      "step": 2338
    },
    {
      "epoch": 2.992,
      "grad_norm": 0.4402843415737152,
      "learning_rate": 1.722612445217778e-05,
      "loss": 0.5848,
      "step": 2339
    },
    {
      "epoch": 2.99328,
      "grad_norm": 0.38917961716651917,
      "learning_rate": 1.7211747501571957e-05,
      "loss": 0.546,
      "step": 2340
    },
    {
      "epoch": 2.99328,
      "eval_loss": 0.679282009601593,
      "eval_runtime": 19.5991,
      "eval_samples_per_second": 25.613,
      "eval_steps_per_second": 3.214,
      "step": 2340
    },
    {
      "epoch": 2.99456,
      "grad_norm": 0.4574100077152252,
      "learning_rate": 1.7197373837515104e-05,
      "loss": 0.5402,
      "step": 2341
    },
    {
      "epoch": 2.99584,
      "grad_norm": 0.4329470694065094,
      "learning_rate": 1.718300346987297e-05,
      "loss": 0.5569,
      "step": 2342
    },
    {
      "epoch": 2.99712,
      "grad_norm": 0.42445385456085205,
      "learning_rate": 1.7168636408509064e-05,
      "loss": 0.548,
      "step": 2343
    },
    {
      "epoch": 2.9984,
      "grad_norm": 0.4329701066017151,
      "learning_rate": 1.715427266328461e-05,
      "loss": 0.5884,
      "step": 2344
    },
    {
      "epoch": 2.99968,
      "grad_norm": 0.4474468529224396,
      "learning_rate": 1.7139912244058555e-05,
      "loss": 0.5894,
      "step": 2345
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7302874326705933,
      "learning_rate": 1.7125555160687582e-05,
      "loss": 0.5362,
      "step": 2346
    },
    {
      "epoch": 3.00128,
      "grad_norm": 0.6048936247825623,
      "learning_rate": 1.7111201423026054e-05,
      "loss": 0.5308,
      "step": 2347
    },
    {
      "epoch": 3.00256,
      "grad_norm": 0.4857417345046997,
      "learning_rate": 1.7096851040926062e-05,
      "loss": 0.5495,
      "step": 2348
    },
    {
      "epoch": 3.00384,
      "grad_norm": 0.4502650201320648,
      "learning_rate": 1.708250402423739e-05,
      "loss": 0.5516,
      "step": 2349
    },
    {
      "epoch": 3.00512,
      "grad_norm": 0.47997909784317017,
      "learning_rate": 1.70681603828075e-05,
      "loss": 0.4878,
      "step": 2350
    },
    {
      "epoch": 3.0064,
      "grad_norm": 0.5411403179168701,
      "learning_rate": 1.7053820126481545e-05,
      "loss": 0.5512,
      "step": 2351
    },
    {
      "epoch": 3.00768,
      "grad_norm": 0.4429756700992584,
      "learning_rate": 1.703948326510237e-05,
      "loss": 0.5075,
      "step": 2352
    },
    {
      "epoch": 3.00896,
      "grad_norm": 0.5193628072738647,
      "learning_rate": 1.7025149808510456e-05,
      "loss": 0.5431,
      "step": 2353
    },
    {
      "epoch": 3.01024,
      "grad_norm": 0.44285666942596436,
      "learning_rate": 1.7010819766543972e-05,
      "loss": 0.5306,
      "step": 2354
    },
    {
      "epoch": 3.01152,
      "grad_norm": 0.4516311287879944,
      "learning_rate": 1.6996493149038747e-05,
      "loss": 0.4998,
      "step": 2355
    },
    {
      "epoch": 3.0128,
      "grad_norm": 0.4905199110507965,
      "learning_rate": 1.698216996582823e-05,
      "loss": 0.5225,
      "step": 2356
    },
    {
      "epoch": 3.01408,
      "grad_norm": 0.49010151624679565,
      "learning_rate": 1.6967850226743544e-05,
      "loss": 0.5323,
      "step": 2357
    },
    {
      "epoch": 3.01536,
      "grad_norm": 0.43194037675857544,
      "learning_rate": 1.6953533941613446e-05,
      "loss": 0.5421,
      "step": 2358
    },
    {
      "epoch": 3.01664,
      "grad_norm": 0.4935694932937622,
      "learning_rate": 1.693922112026429e-05,
      "loss": 0.5447,
      "step": 2359
    },
    {
      "epoch": 3.01792,
      "grad_norm": 0.46256211400032043,
      "learning_rate": 1.692491177252009e-05,
      "loss": 0.5229,
      "step": 2360
    },
    {
      "epoch": 3.0192,
      "grad_norm": 0.4362751543521881,
      "learning_rate": 1.691060590820246e-05,
      "loss": 0.5374,
      "step": 2361
    },
    {
      "epoch": 3.02048,
      "grad_norm": 0.43670937418937683,
      "learning_rate": 1.6896303537130617e-05,
      "loss": 0.5166,
      "step": 2362
    },
    {
      "epoch": 3.02176,
      "grad_norm": 0.4747062027454376,
      "learning_rate": 1.6882004669121394e-05,
      "loss": 0.5677,
      "step": 2363
    },
    {
      "epoch": 3.02304,
      "grad_norm": 0.45343372225761414,
      "learning_rate": 1.6867709313989217e-05,
      "loss": 0.5563,
      "step": 2364
    },
    {
      "epoch": 3.02432,
      "grad_norm": 0.4346565306186676,
      "learning_rate": 1.6853417481546084e-05,
      "loss": 0.5513,
      "step": 2365
    },
    {
      "epoch": 3.0256,
      "grad_norm": 0.39591914415359497,
      "learning_rate": 1.6839129181601596e-05,
      "loss": 0.5177,
      "step": 2366
    },
    {
      "epoch": 3.02688,
      "grad_norm": 0.496260404586792,
      "learning_rate": 1.682484442396292e-05,
      "loss": 0.5357,
      "step": 2367
    },
    {
      "epoch": 3.02816,
      "grad_norm": 0.448275625705719,
      "learning_rate": 1.681056321843479e-05,
      "loss": 0.5181,
      "step": 2368
    },
    {
      "epoch": 3.02944,
      "grad_norm": 0.4219822287559509,
      "learning_rate": 1.6796285574819505e-05,
      "loss": 0.5407,
      "step": 2369
    },
    {
      "epoch": 3.03072,
      "grad_norm": 0.4447540044784546,
      "learning_rate": 1.6782011502916925e-05,
      "loss": 0.5221,
      "step": 2370
    },
    {
      "epoch": 3.03072,
      "eval_loss": 0.6959467530250549,
      "eval_runtime": 19.5935,
      "eval_samples_per_second": 25.621,
      "eval_steps_per_second": 3.215,
      "step": 2370
    },
    {
      "epoch": 3.032,
      "grad_norm": 0.42596331238746643,
      "learning_rate": 1.6767741012524444e-05,
      "loss": 0.5411,
      "step": 2371
    },
    {
      "epoch": 3.03328,
      "grad_norm": 0.4144585430622101,
      "learning_rate": 1.6753474113437014e-05,
      "loss": 0.5197,
      "step": 2372
    },
    {
      "epoch": 3.03456,
      "grad_norm": 0.461469441652298,
      "learning_rate": 1.6739210815447107e-05,
      "loss": 0.5344,
      "step": 2373
    },
    {
      "epoch": 3.03584,
      "grad_norm": 0.4525810480117798,
      "learning_rate": 1.672495112834474e-05,
      "loss": 0.5316,
      "step": 2374
    },
    {
      "epoch": 3.03712,
      "grad_norm": 0.4148513078689575,
      "learning_rate": 1.671069506191743e-05,
      "loss": 0.5047,
      "step": 2375
    },
    {
      "epoch": 3.0384,
      "grad_norm": 0.4418170750141144,
      "learning_rate": 1.6696442625950223e-05,
      "loss": 0.5342,
      "step": 2376
    },
    {
      "epoch": 3.03968,
      "grad_norm": 0.4185195565223694,
      "learning_rate": 1.6682193830225678e-05,
      "loss": 0.5169,
      "step": 2377
    },
    {
      "epoch": 3.04096,
      "grad_norm": 0.44607657194137573,
      "learning_rate": 1.6667948684523847e-05,
      "loss": 0.5312,
      "step": 2378
    },
    {
      "epoch": 3.04224,
      "grad_norm": 0.43360334634780884,
      "learning_rate": 1.6653707198622272e-05,
      "loss": 0.5174,
      "step": 2379
    },
    {
      "epoch": 3.04352,
      "grad_norm": 0.4537777602672577,
      "learning_rate": 1.6639469382295992e-05,
      "loss": 0.5029,
      "step": 2380
    },
    {
      "epoch": 3.0448,
      "grad_norm": 0.42125779390335083,
      "learning_rate": 1.662523524531753e-05,
      "loss": 0.534,
      "step": 2381
    },
    {
      "epoch": 3.04608,
      "grad_norm": 0.42872855067253113,
      "learning_rate": 1.6611004797456866e-05,
      "loss": 0.5273,
      "step": 2382
    },
    {
      "epoch": 3.04736,
      "grad_norm": 0.47265881299972534,
      "learning_rate": 1.659677804848147e-05,
      "loss": 0.5403,
      "step": 2383
    },
    {
      "epoch": 3.04864,
      "grad_norm": 0.4314512312412262,
      "learning_rate": 1.658255500815626e-05,
      "loss": 0.5101,
      "step": 2384
    },
    {
      "epoch": 3.04992,
      "grad_norm": 0.3899570405483246,
      "learning_rate": 1.6568335686243608e-05,
      "loss": 0.5249,
      "step": 2385
    },
    {
      "epoch": 3.0512,
      "grad_norm": 0.40133556723594666,
      "learning_rate": 1.6554120092503346e-05,
      "loss": 0.5132,
      "step": 2386
    },
    {
      "epoch": 3.05248,
      "grad_norm": 0.45560187101364136,
      "learning_rate": 1.653990823669273e-05,
      "loss": 0.5113,
      "step": 2387
    },
    {
      "epoch": 3.05376,
      "grad_norm": 0.40980711579322815,
      "learning_rate": 1.652570012856646e-05,
      "loss": 0.4962,
      "step": 2388
    },
    {
      "epoch": 3.05504,
      "grad_norm": 0.4330582618713379,
      "learning_rate": 1.651149577787666e-05,
      "loss": 0.5181,
      "step": 2389
    },
    {
      "epoch": 3.05632,
      "grad_norm": 0.4352385103702545,
      "learning_rate": 1.649729519437289e-05,
      "loss": 0.5379,
      "step": 2390
    },
    {
      "epoch": 3.0576,
      "grad_norm": 0.4541664123535156,
      "learning_rate": 1.6483098387802098e-05,
      "loss": 0.5395,
      "step": 2391
    },
    {
      "epoch": 3.05888,
      "grad_norm": 0.4232214391231537,
      "learning_rate": 1.6468905367908653e-05,
      "loss": 0.5638,
      "step": 2392
    },
    {
      "epoch": 3.06016,
      "grad_norm": 0.44389379024505615,
      "learning_rate": 1.6454716144434334e-05,
      "loss": 0.5393,
      "step": 2393
    },
    {
      "epoch": 3.06144,
      "grad_norm": 0.4125765562057495,
      "learning_rate": 1.6440530727118295e-05,
      "loss": 0.5081,
      "step": 2394
    },
    {
      "epoch": 3.06272,
      "grad_norm": 0.41580402851104736,
      "learning_rate": 1.6426349125697097e-05,
      "loss": 0.539,
      "step": 2395
    },
    {
      "epoch": 3.064,
      "grad_norm": 0.40604615211486816,
      "learning_rate": 1.6412171349904675e-05,
      "loss": 0.5297,
      "step": 2396
    },
    {
      "epoch": 3.06528,
      "grad_norm": 0.406445175409317,
      "learning_rate": 1.6397997409472325e-05,
      "loss": 0.5072,
      "step": 2397
    },
    {
      "epoch": 3.06656,
      "grad_norm": 0.41381263732910156,
      "learning_rate": 1.6383827314128732e-05,
      "loss": 0.5262,
      "step": 2398
    },
    {
      "epoch": 3.06784,
      "grad_norm": 0.43235138058662415,
      "learning_rate": 1.6369661073599913e-05,
      "loss": 0.5635,
      "step": 2399
    },
    {
      "epoch": 3.06912,
      "grad_norm": 0.39751461148262024,
      "learning_rate": 1.6355498697609275e-05,
      "loss": 0.563,
      "step": 2400
    },
    {
      "epoch": 3.06912,
      "eval_loss": 0.696057140827179,
      "eval_runtime": 19.6424,
      "eval_samples_per_second": 25.557,
      "eval_steps_per_second": 3.207,
      "step": 2400
    },
    {
      "epoch": 3.0704,
      "grad_norm": 0.41560688614845276,
      "learning_rate": 1.6341340195877557e-05,
      "loss": 0.5503,
      "step": 2401
    },
    {
      "epoch": 3.07168,
      "grad_norm": 0.42766180634498596,
      "learning_rate": 1.6327185578122823e-05,
      "loss": 0.5266,
      "step": 2402
    },
    {
      "epoch": 3.07296,
      "grad_norm": 0.42232775688171387,
      "learning_rate": 1.631303485406049e-05,
      "loss": 0.5442,
      "step": 2403
    },
    {
      "epoch": 3.07424,
      "grad_norm": 0.44024574756622314,
      "learning_rate": 1.62988880334033e-05,
      "loss": 0.5343,
      "step": 2404
    },
    {
      "epoch": 3.07552,
      "grad_norm": 0.41560858488082886,
      "learning_rate": 1.6284745125861313e-05,
      "loss": 0.5143,
      "step": 2405
    },
    {
      "epoch": 3.0768,
      "grad_norm": 0.39409852027893066,
      "learning_rate": 1.6270606141141896e-05,
      "loss": 0.5283,
      "step": 2406
    },
    {
      "epoch": 3.07808,
      "grad_norm": 0.45898160338401794,
      "learning_rate": 1.6256471088949742e-05,
      "loss": 0.5105,
      "step": 2407
    },
    {
      "epoch": 3.07936,
      "grad_norm": 0.4474641978740692,
      "learning_rate": 1.624233997898682e-05,
      "loss": 0.5432,
      "step": 2408
    },
    {
      "epoch": 3.08064,
      "grad_norm": 0.4142138361930847,
      "learning_rate": 1.6228212820952414e-05,
      "loss": 0.53,
      "step": 2409
    },
    {
      "epoch": 3.08192,
      "grad_norm": 0.4177364110946655,
      "learning_rate": 1.62140896245431e-05,
      "loss": 0.5275,
      "step": 2410
    },
    {
      "epoch": 3.0832,
      "grad_norm": 0.42851269245147705,
      "learning_rate": 1.6199970399452697e-05,
      "loss": 0.5324,
      "step": 2411
    },
    {
      "epoch": 3.08448,
      "grad_norm": 0.3937895596027374,
      "learning_rate": 1.6185855155372338e-05,
      "loss": 0.514,
      "step": 2412
    },
    {
      "epoch": 3.08576,
      "grad_norm": 0.40375033020973206,
      "learning_rate": 1.6171743901990414e-05,
      "loss": 0.5227,
      "step": 2413
    },
    {
      "epoch": 3.08704,
      "grad_norm": 0.3945481777191162,
      "learning_rate": 1.6157636648992566e-05,
      "loss": 0.4931,
      "step": 2414
    },
    {
      "epoch": 3.08832,
      "grad_norm": 0.38710424304008484,
      "learning_rate": 1.6143533406061692e-05,
      "loss": 0.4944,
      "step": 2415
    },
    {
      "epoch": 3.0896,
      "grad_norm": 0.3928638696670532,
      "learning_rate": 1.6129434182877952e-05,
      "loss": 0.5569,
      "step": 2416
    },
    {
      "epoch": 3.09088,
      "grad_norm": 0.4271368086338043,
      "learning_rate": 1.611533898911872e-05,
      "loss": 0.5363,
      "step": 2417
    },
    {
      "epoch": 3.09216,
      "grad_norm": 0.3831751048564911,
      "learning_rate": 1.610124783445864e-05,
      "loss": 0.5089,
      "step": 2418
    },
    {
      "epoch": 3.09344,
      "grad_norm": 0.3982716202735901,
      "learning_rate": 1.6087160728569557e-05,
      "loss": 0.5233,
      "step": 2419
    },
    {
      "epoch": 3.09472,
      "grad_norm": 0.40576601028442383,
      "learning_rate": 1.6073077681120532e-05,
      "loss": 0.5368,
      "step": 2420
    },
    {
      "epoch": 3.096,
      "grad_norm": 0.4066865146160126,
      "learning_rate": 1.6058998701777864e-05,
      "loss": 0.5461,
      "step": 2421
    },
    {
      "epoch": 3.09728,
      "grad_norm": 0.38939401507377625,
      "learning_rate": 1.6044923800205055e-05,
      "loss": 0.5402,
      "step": 2422
    },
    {
      "epoch": 3.09856,
      "grad_norm": 0.40298181772232056,
      "learning_rate": 1.6030852986062782e-05,
      "loss": 0.5265,
      "step": 2423
    },
    {
      "epoch": 3.09984,
      "grad_norm": 0.4386041462421417,
      "learning_rate": 1.6016786269008954e-05,
      "loss": 0.5454,
      "step": 2424
    },
    {
      "epoch": 3.10112,
      "grad_norm": 0.41656941175460815,
      "learning_rate": 1.600272365869864e-05,
      "loss": 0.5432,
      "step": 2425
    },
    {
      "epoch": 3.1024,
      "grad_norm": 0.45001932978630066,
      "learning_rate": 1.5988665164784093e-05,
      "loss": 0.5692,
      "step": 2426
    },
    {
      "epoch": 3.1036799999999998,
      "grad_norm": 0.4216475188732147,
      "learning_rate": 1.5974610796914765e-05,
      "loss": 0.5299,
      "step": 2427
    },
    {
      "epoch": 3.10496,
      "grad_norm": 0.39157629013061523,
      "learning_rate": 1.5960560564737247e-05,
      "loss": 0.5263,
      "step": 2428
    },
    {
      "epoch": 3.10624,
      "grad_norm": 0.4138008952140808,
      "learning_rate": 1.5946514477895303e-05,
      "loss": 0.5144,
      "step": 2429
    },
    {
      "epoch": 3.10752,
      "grad_norm": 0.43523404002189636,
      "learning_rate": 1.5932472546029863e-05,
      "loss": 0.5591,
      "step": 2430
    },
    {
      "epoch": 3.10752,
      "eval_loss": 0.6968607306480408,
      "eval_runtime": 19.5811,
      "eval_samples_per_second": 25.637,
      "eval_steps_per_second": 3.217,
      "step": 2430
    },
    {
      "epoch": 3.1088,
      "grad_norm": 0.3952302634716034,
      "learning_rate": 1.5918434778778974e-05,
      "loss": 0.5207,
      "step": 2431
    },
    {
      "epoch": 3.11008,
      "grad_norm": 0.4308258295059204,
      "learning_rate": 1.5904401185777855e-05,
      "loss": 0.5117,
      "step": 2432
    },
    {
      "epoch": 3.11136,
      "grad_norm": 0.4379512369632721,
      "learning_rate": 1.5890371776658858e-05,
      "loss": 0.5073,
      "step": 2433
    },
    {
      "epoch": 3.11264,
      "grad_norm": 0.38857418298721313,
      "learning_rate": 1.5876346561051438e-05,
      "loss": 0.4998,
      "step": 2434
    },
    {
      "epoch": 3.11392,
      "grad_norm": 0.42681705951690674,
      "learning_rate": 1.5862325548582193e-05,
      "loss": 0.5013,
      "step": 2435
    },
    {
      "epoch": 3.1152,
      "grad_norm": 0.4107728898525238,
      "learning_rate": 1.584830874887484e-05,
      "loss": 0.5372,
      "step": 2436
    },
    {
      "epoch": 3.11648,
      "grad_norm": 0.4097538888454437,
      "learning_rate": 1.5834296171550187e-05,
      "loss": 0.5538,
      "step": 2437
    },
    {
      "epoch": 3.11776,
      "grad_norm": 0.42558732628822327,
      "learning_rate": 1.5820287826226145e-05,
      "loss": 0.5634,
      "step": 2438
    },
    {
      "epoch": 3.11904,
      "grad_norm": 0.4282431900501251,
      "learning_rate": 1.5806283722517747e-05,
      "loss": 0.5386,
      "step": 2439
    },
    {
      "epoch": 3.12032,
      "grad_norm": 0.40141505002975464,
      "learning_rate": 1.579228387003708e-05,
      "loss": 0.5284,
      "step": 2440
    },
    {
      "epoch": 3.1216,
      "grad_norm": 0.3922045826911926,
      "learning_rate": 1.577828827839333e-05,
      "loss": 0.5568,
      "step": 2441
    },
    {
      "epoch": 3.12288,
      "grad_norm": 0.4447406530380249,
      "learning_rate": 1.5764296957192763e-05,
      "loss": 0.4981,
      "step": 2442
    },
    {
      "epoch": 3.12416,
      "grad_norm": 0.41691941022872925,
      "learning_rate": 1.5750309916038707e-05,
      "loss": 0.5242,
      "step": 2443
    },
    {
      "epoch": 3.12544,
      "grad_norm": 0.3895879089832306,
      "learning_rate": 1.5736327164531543e-05,
      "loss": 0.5443,
      "step": 2444
    },
    {
      "epoch": 3.12672,
      "grad_norm": 0.43840885162353516,
      "learning_rate": 1.572234871226873e-05,
      "loss": 0.5315,
      "step": 2445
    },
    {
      "epoch": 3.128,
      "grad_norm": 0.39115333557128906,
      "learning_rate": 1.5708374568844758e-05,
      "loss": 0.5322,
      "step": 2446
    },
    {
      "epoch": 3.12928,
      "grad_norm": 0.3948315382003784,
      "learning_rate": 1.5694404743851162e-05,
      "loss": 0.5157,
      "step": 2447
    },
    {
      "epoch": 3.13056,
      "grad_norm": 0.42144012451171875,
      "learning_rate": 1.5680439246876528e-05,
      "loss": 0.5424,
      "step": 2448
    },
    {
      "epoch": 3.13184,
      "grad_norm": 0.42122355103492737,
      "learning_rate": 1.5666478087506445e-05,
      "loss": 0.5097,
      "step": 2449
    },
    {
      "epoch": 3.13312,
      "grad_norm": 0.4034767746925354,
      "learning_rate": 1.5652521275323547e-05,
      "loss": 0.5256,
      "step": 2450
    },
    {
      "epoch": 3.1344,
      "grad_norm": 0.41410744190216064,
      "learning_rate": 1.5638568819907475e-05,
      "loss": 0.5329,
      "step": 2451
    },
    {
      "epoch": 3.13568,
      "grad_norm": 0.44710052013397217,
      "learning_rate": 1.5624620730834878e-05,
      "loss": 0.5388,
      "step": 2452
    },
    {
      "epoch": 3.13696,
      "grad_norm": 0.4226559102535248,
      "learning_rate": 1.5610677017679417e-05,
      "loss": 0.5557,
      "step": 2453
    },
    {
      "epoch": 3.13824,
      "grad_norm": 0.403870165348053,
      "learning_rate": 1.5596737690011735e-05,
      "loss": 0.5066,
      "step": 2454
    },
    {
      "epoch": 3.13952,
      "grad_norm": 0.4367230534553528,
      "learning_rate": 1.5582802757399485e-05,
      "loss": 0.5295,
      "step": 2455
    },
    {
      "epoch": 3.1408,
      "grad_norm": 0.3972592353820801,
      "learning_rate": 1.5568872229407282e-05,
      "loss": 0.5199,
      "step": 2456
    },
    {
      "epoch": 3.14208,
      "grad_norm": 0.4336166977882385,
      "learning_rate": 1.5554946115596733e-05,
      "loss": 0.5417,
      "step": 2457
    },
    {
      "epoch": 3.14336,
      "grad_norm": 0.41951775550842285,
      "learning_rate": 1.5541024425526406e-05,
      "loss": 0.5189,
      "step": 2458
    },
    {
      "epoch": 3.14464,
      "grad_norm": 0.4169628918170929,
      "learning_rate": 1.5527107168751845e-05,
      "loss": 0.5567,
      "step": 2459
    },
    {
      "epoch": 3.14592,
      "grad_norm": 0.42372748255729675,
      "learning_rate": 1.5513194354825523e-05,
      "loss": 0.5423,
      "step": 2460
    },
    {
      "epoch": 3.14592,
      "eval_loss": 0.6958539485931396,
      "eval_runtime": 19.6155,
      "eval_samples_per_second": 25.592,
      "eval_steps_per_second": 3.212,
      "step": 2460
    },
    {
      "epoch": 3.1471999999999998,
      "grad_norm": 0.4191286265850067,
      "learning_rate": 1.54992859932969e-05,
      "loss": 0.5537,
      "step": 2461
    },
    {
      "epoch": 3.14848,
      "grad_norm": 0.39077678322792053,
      "learning_rate": 1.5485382093712368e-05,
      "loss": 0.5129,
      "step": 2462
    },
    {
      "epoch": 3.14976,
      "grad_norm": 0.41337355971336365,
      "learning_rate": 1.547148266561523e-05,
      "loss": 0.548,
      "step": 2463
    },
    {
      "epoch": 3.15104,
      "grad_norm": 0.44362136721611023,
      "learning_rate": 1.545758771854575e-05,
      "loss": 0.5394,
      "step": 2464
    },
    {
      "epoch": 3.15232,
      "grad_norm": 0.42797115445137024,
      "learning_rate": 1.5443697262041126e-05,
      "loss": 0.5475,
      "step": 2465
    },
    {
      "epoch": 3.1536,
      "grad_norm": 0.42144355177879333,
      "learning_rate": 1.5429811305635433e-05,
      "loss": 0.5308,
      "step": 2466
    },
    {
      "epoch": 3.15488,
      "grad_norm": 0.4248436689376831,
      "learning_rate": 1.541592985885968e-05,
      "loss": 0.5171,
      "step": 2467
    },
    {
      "epoch": 3.15616,
      "grad_norm": 0.3880794048309326,
      "learning_rate": 1.54020529312418e-05,
      "loss": 0.5297,
      "step": 2468
    },
    {
      "epoch": 3.15744,
      "grad_norm": 0.4033370912075043,
      "learning_rate": 1.538818053230658e-05,
      "loss": 0.5328,
      "step": 2469
    },
    {
      "epoch": 3.15872,
      "grad_norm": 0.4251076281070709,
      "learning_rate": 1.5374312671575736e-05,
      "loss": 0.5584,
      "step": 2470
    },
    {
      "epoch": 3.16,
      "grad_norm": 0.40172943472862244,
      "learning_rate": 1.5360449358567864e-05,
      "loss": 0.5393,
      "step": 2471
    },
    {
      "epoch": 3.16128,
      "grad_norm": 0.39594829082489014,
      "learning_rate": 1.5346590602798413e-05,
      "loss": 0.5643,
      "step": 2472
    },
    {
      "epoch": 3.16256,
      "grad_norm": 0.4096458852291107,
      "learning_rate": 1.5332736413779726e-05,
      "loss": 0.568,
      "step": 2473
    },
    {
      "epoch": 3.16384,
      "grad_norm": 0.4240380525588989,
      "learning_rate": 1.5318886801021018e-05,
      "loss": 0.5562,
      "step": 2474
    },
    {
      "epoch": 3.16512,
      "grad_norm": 0.402528315782547,
      "learning_rate": 1.5305041774028345e-05,
      "loss": 0.5227,
      "step": 2475
    },
    {
      "epoch": 3.1664,
      "grad_norm": 0.41248664259910583,
      "learning_rate": 1.529120134230462e-05,
      "loss": 0.5264,
      "step": 2476
    },
    {
      "epoch": 3.16768,
      "grad_norm": 0.4060472249984741,
      "learning_rate": 1.527736551534961e-05,
      "loss": 0.542,
      "step": 2477
    },
    {
      "epoch": 3.16896,
      "grad_norm": 0.41034382581710815,
      "learning_rate": 1.526353430265991e-05,
      "loss": 0.5133,
      "step": 2478
    },
    {
      "epoch": 3.17024,
      "grad_norm": 0.4624151289463043,
      "learning_rate": 1.524970771372896e-05,
      "loss": 0.5414,
      "step": 2479
    },
    {
      "epoch": 3.17152,
      "grad_norm": 0.4164605438709259,
      "learning_rate": 1.5235885758047011e-05,
      "loss": 0.5159,
      "step": 2480
    },
    {
      "epoch": 3.1728,
      "grad_norm": 0.42653587460517883,
      "learning_rate": 1.522206844510115e-05,
      "loss": 0.5351,
      "step": 2481
    },
    {
      "epoch": 3.17408,
      "grad_norm": 0.4258320927619934,
      "learning_rate": 1.5208255784375276e-05,
      "loss": 0.5324,
      "step": 2482
    },
    {
      "epoch": 3.17536,
      "grad_norm": 0.4585385322570801,
      "learning_rate": 1.5194447785350073e-05,
      "loss": 0.5559,
      "step": 2483
    },
    {
      "epoch": 3.17664,
      "grad_norm": 0.46017488837242126,
      "learning_rate": 1.5180644457503056e-05,
      "loss": 0.542,
      "step": 2484
    },
    {
      "epoch": 3.17792,
      "grad_norm": 0.40270373225212097,
      "learning_rate": 1.5166845810308518e-05,
      "loss": 0.5153,
      "step": 2485
    },
    {
      "epoch": 3.1792,
      "grad_norm": 0.40549492835998535,
      "learning_rate": 1.5153051853237526e-05,
      "loss": 0.5155,
      "step": 2486
    },
    {
      "epoch": 3.18048,
      "grad_norm": 0.4028909206390381,
      "learning_rate": 1.5139262595757954e-05,
      "loss": 0.5111,
      "step": 2487
    },
    {
      "epoch": 3.18176,
      "grad_norm": 0.4060007333755493,
      "learning_rate": 1.5125478047334445e-05,
      "loss": 0.5372,
      "step": 2488
    },
    {
      "epoch": 3.18304,
      "grad_norm": 0.41059330105781555,
      "learning_rate": 1.5111698217428385e-05,
      "loss": 0.5529,
      "step": 2489
    },
    {
      "epoch": 3.18432,
      "grad_norm": 0.4084486961364746,
      "learning_rate": 1.5097923115497955e-05,
      "loss": 0.5614,
      "step": 2490
    },
    {
      "epoch": 3.18432,
      "eval_loss": 0.6955298781394958,
      "eval_runtime": 19.5768,
      "eval_samples_per_second": 25.643,
      "eval_steps_per_second": 3.218,
      "step": 2490
    },
    {
      "epoch": 3.1856,
      "grad_norm": 0.4166280925273895,
      "learning_rate": 1.5084152750998079e-05,
      "loss": 0.5117,
      "step": 2491
    },
    {
      "epoch": 3.18688,
      "grad_norm": 0.39208823442459106,
      "learning_rate": 1.5070387133380408e-05,
      "loss": 0.518,
      "step": 2492
    },
    {
      "epoch": 3.18816,
      "grad_norm": 0.377246230840683,
      "learning_rate": 1.5056626272093358e-05,
      "loss": 0.5101,
      "step": 2493
    },
    {
      "epoch": 3.18944,
      "grad_norm": 0.4401876926422119,
      "learning_rate": 1.5042870176582094e-05,
      "loss": 0.5407,
      "step": 2494
    },
    {
      "epoch": 3.19072,
      "grad_norm": 0.4089777171611786,
      "learning_rate": 1.5029118856288465e-05,
      "loss": 0.5254,
      "step": 2495
    },
    {
      "epoch": 3.192,
      "grad_norm": 0.39873006939888,
      "learning_rate": 1.501537232065108e-05,
      "loss": 0.527,
      "step": 2496
    },
    {
      "epoch": 3.19328,
      "grad_norm": 0.4293418526649475,
      "learning_rate": 1.500163057910526e-05,
      "loss": 0.5539,
      "step": 2497
    },
    {
      "epoch": 3.19456,
      "grad_norm": 0.41150152683258057,
      "learning_rate": 1.4987893641083014e-05,
      "loss": 0.5766,
      "step": 2498
    },
    {
      "epoch": 3.19584,
      "grad_norm": 0.41213148832321167,
      "learning_rate": 1.4974161516013071e-05,
      "loss": 0.5572,
      "step": 2499
    },
    {
      "epoch": 3.19712,
      "grad_norm": 0.4287356436252594,
      "learning_rate": 1.496043421332087e-05,
      "loss": 0.5493,
      "step": 2500
    },
    {
      "epoch": 3.1984,
      "grad_norm": 0.4079543650150299,
      "learning_rate": 1.4946711742428495e-05,
      "loss": 0.5692,
      "step": 2501
    },
    {
      "epoch": 3.19968,
      "grad_norm": 0.40431660413742065,
      "learning_rate": 1.4932994112754755e-05,
      "loss": 0.5549,
      "step": 2502
    },
    {
      "epoch": 3.20096,
      "grad_norm": 0.3852398097515106,
      "learning_rate": 1.4919281333715136e-05,
      "loss": 0.5224,
      "step": 2503
    },
    {
      "epoch": 3.20224,
      "grad_norm": 0.394904226064682,
      "learning_rate": 1.4905573414721759e-05,
      "loss": 0.521,
      "step": 2504
    },
    {
      "epoch": 3.20352,
      "grad_norm": 0.42589011788368225,
      "learning_rate": 1.4891870365183447e-05,
      "loss": 0.519,
      "step": 2505
    },
    {
      "epoch": 3.2048,
      "grad_norm": 0.4415909945964813,
      "learning_rate": 1.4878172194505656e-05,
      "loss": 0.5311,
      "step": 2506
    },
    {
      "epoch": 3.20608,
      "grad_norm": 0.43666884303092957,
      "learning_rate": 1.4864478912090508e-05,
      "loss": 0.5192,
      "step": 2507
    },
    {
      "epoch": 3.20736,
      "grad_norm": 0.408772349357605,
      "learning_rate": 1.4850790527336773e-05,
      "loss": 0.5224,
      "step": 2508
    },
    {
      "epoch": 3.20864,
      "grad_norm": 0.4186095595359802,
      "learning_rate": 1.4837107049639832e-05,
      "loss": 0.5275,
      "step": 2509
    },
    {
      "epoch": 3.20992,
      "grad_norm": 0.39057090878486633,
      "learning_rate": 1.4823428488391724e-05,
      "loss": 0.5021,
      "step": 2510
    },
    {
      "epoch": 3.2112,
      "grad_norm": 0.4427454173564911,
      "learning_rate": 1.4809754852981115e-05,
      "loss": 0.5614,
      "step": 2511
    },
    {
      "epoch": 3.2124800000000002,
      "grad_norm": 0.4101174473762512,
      "learning_rate": 1.4796086152793274e-05,
      "loss": 0.528,
      "step": 2512
    },
    {
      "epoch": 3.21376,
      "grad_norm": 0.44704586267471313,
      "learning_rate": 1.4782422397210083e-05,
      "loss": 0.5042,
      "step": 2513
    },
    {
      "epoch": 3.21504,
      "grad_norm": 0.4616358280181885,
      "learning_rate": 1.4768763595610059e-05,
      "loss": 0.5296,
      "step": 2514
    },
    {
      "epoch": 3.21632,
      "grad_norm": 0.4259190261363983,
      "learning_rate": 1.4755109757368264e-05,
      "loss": 0.56,
      "step": 2515
    },
    {
      "epoch": 3.2176,
      "grad_norm": 0.41648736596107483,
      "learning_rate": 1.4741460891856415e-05,
      "loss": 0.5171,
      "step": 2516
    },
    {
      "epoch": 3.21888,
      "grad_norm": 0.4889039099216461,
      "learning_rate": 1.4727817008442777e-05,
      "loss": 0.5243,
      "step": 2517
    },
    {
      "epoch": 3.22016,
      "grad_norm": 0.478168785572052,
      "learning_rate": 1.4714178116492198e-05,
      "loss": 0.5164,
      "step": 2518
    },
    {
      "epoch": 3.22144,
      "grad_norm": 0.48705601692199707,
      "learning_rate": 1.4700544225366114e-05,
      "loss": 0.5607,
      "step": 2519
    },
    {
      "epoch": 3.22272,
      "grad_norm": 0.44704827666282654,
      "learning_rate": 1.4686915344422522e-05,
      "loss": 0.5158,
      "step": 2520
    },
    {
      "epoch": 3.22272,
      "eval_loss": 0.695722222328186,
      "eval_runtime": 19.5982,
      "eval_samples_per_second": 25.615,
      "eval_steps_per_second": 3.215,
      "step": 2520
    },
    {
      "epoch": 3.224,
      "grad_norm": 0.4274193048477173,
      "learning_rate": 1.4673291483015977e-05,
      "loss": 0.5353,
      "step": 2521
    },
    {
      "epoch": 3.22528,
      "grad_norm": 0.406403124332428,
      "learning_rate": 1.4659672650497591e-05,
      "loss": 0.5312,
      "step": 2522
    },
    {
      "epoch": 3.22656,
      "grad_norm": 0.45830878615379333,
      "learning_rate": 1.464605885621503e-05,
      "loss": 0.5437,
      "step": 2523
    },
    {
      "epoch": 3.22784,
      "grad_norm": 0.3769043982028961,
      "learning_rate": 1.463245010951249e-05,
      "loss": 0.491,
      "step": 2524
    },
    {
      "epoch": 3.22912,
      "grad_norm": 0.38698580861091614,
      "learning_rate": 1.4618846419730713e-05,
      "loss": 0.5271,
      "step": 2525
    },
    {
      "epoch": 3.2304,
      "grad_norm": 0.45829567313194275,
      "learning_rate": 1.4605247796206965e-05,
      "loss": 0.5608,
      "step": 2526
    },
    {
      "epoch": 3.23168,
      "grad_norm": 0.43296247720718384,
      "learning_rate": 1.459165424827504e-05,
      "loss": 0.5662,
      "step": 2527
    },
    {
      "epoch": 3.23296,
      "grad_norm": 0.39579641819000244,
      "learning_rate": 1.4578065785265234e-05,
      "loss": 0.5174,
      "step": 2528
    },
    {
      "epoch": 3.23424,
      "grad_norm": 0.40931564569473267,
      "learning_rate": 1.456448241650437e-05,
      "loss": 0.5292,
      "step": 2529
    },
    {
      "epoch": 3.23552,
      "grad_norm": 0.44076675176620483,
      "learning_rate": 1.455090415131577e-05,
      "loss": 0.5221,
      "step": 2530
    },
    {
      "epoch": 3.2368,
      "grad_norm": 0.4330349266529083,
      "learning_rate": 1.4537330999019248e-05,
      "loss": 0.5542,
      "step": 2531
    },
    {
      "epoch": 3.23808,
      "grad_norm": 0.3833218812942505,
      "learning_rate": 1.4523762968931103e-05,
      "loss": 0.523,
      "step": 2532
    },
    {
      "epoch": 3.23936,
      "grad_norm": 0.3918771743774414,
      "learning_rate": 1.4510200070364135e-05,
      "loss": 0.5273,
      "step": 2533
    },
    {
      "epoch": 3.24064,
      "grad_norm": 0.45978766679763794,
      "learning_rate": 1.4496642312627614e-05,
      "loss": 0.5463,
      "step": 2534
    },
    {
      "epoch": 3.24192,
      "grad_norm": 0.41567739844322205,
      "learning_rate": 1.4483089705027267e-05,
      "loss": 0.543,
      "step": 2535
    },
    {
      "epoch": 3.2432,
      "grad_norm": 0.3931529223918915,
      "learning_rate": 1.4469542256865312e-05,
      "loss": 0.5552,
      "step": 2536
    },
    {
      "epoch": 3.24448,
      "grad_norm": 0.42104965448379517,
      "learning_rate": 1.4455999977440414e-05,
      "loss": 0.5055,
      "step": 2537
    },
    {
      "epoch": 3.24576,
      "grad_norm": 0.41262951493263245,
      "learning_rate": 1.4442462876047682e-05,
      "loss": 0.5187,
      "step": 2538
    },
    {
      "epoch": 3.24704,
      "grad_norm": 0.4200323522090912,
      "learning_rate": 1.4428930961978678e-05,
      "loss": 0.5313,
      "step": 2539
    },
    {
      "epoch": 3.24832,
      "grad_norm": 0.3990638256072998,
      "learning_rate": 1.4415404244521419e-05,
      "loss": 0.5044,
      "step": 2540
    },
    {
      "epoch": 3.2496,
      "grad_norm": 0.4087105393409729,
      "learning_rate": 1.4401882732960317e-05,
      "loss": 0.5442,
      "step": 2541
    },
    {
      "epoch": 3.25088,
      "grad_norm": 0.4079061448574066,
      "learning_rate": 1.4388366436576253e-05,
      "loss": 0.5351,
      "step": 2542
    },
    {
      "epoch": 3.25216,
      "grad_norm": 0.44716742634773254,
      "learning_rate": 1.4374855364646503e-05,
      "loss": 0.5362,
      "step": 2543
    },
    {
      "epoch": 3.25344,
      "grad_norm": 0.40454334020614624,
      "learning_rate": 1.4361349526444754e-05,
      "loss": 0.4934,
      "step": 2544
    },
    {
      "epoch": 3.25472,
      "grad_norm": 0.4424201548099518,
      "learning_rate": 1.4347848931241122e-05,
      "loss": 0.5336,
      "step": 2545
    },
    {
      "epoch": 3.2560000000000002,
      "grad_norm": 0.4345995783805847,
      "learning_rate": 1.4334353588302113e-05,
      "loss": 0.554,
      "step": 2546
    },
    {
      "epoch": 3.25728,
      "grad_norm": 0.44795718789100647,
      "learning_rate": 1.4320863506890619e-05,
      "loss": 0.5454,
      "step": 2547
    },
    {
      "epoch": 3.25856,
      "grad_norm": 0.41713055968284607,
      "learning_rate": 1.4307378696265927e-05,
      "loss": 0.5356,
      "step": 2548
    },
    {
      "epoch": 3.25984,
      "grad_norm": 0.4336369037628174,
      "learning_rate": 1.4293899165683725e-05,
      "loss": 0.5238,
      "step": 2549
    },
    {
      "epoch": 3.26112,
      "grad_norm": 0.4080328941345215,
      "learning_rate": 1.4280424924396043e-05,
      "loss": 0.4951,
      "step": 2550
    },
    {
      "epoch": 3.26112,
      "eval_loss": 0.6953138113021851,
      "eval_runtime": 19.639,
      "eval_samples_per_second": 25.561,
      "eval_steps_per_second": 3.208,
      "step": 2550
    },
    {
      "epoch": 3.2624,
      "grad_norm": 0.40359893441200256,
      "learning_rate": 1.42669559816513e-05,
      "loss": 0.5103,
      "step": 2551
    },
    {
      "epoch": 3.26368,
      "grad_norm": 0.40651965141296387,
      "learning_rate": 1.4253492346694287e-05,
      "loss": 0.5431,
      "step": 2552
    },
    {
      "epoch": 3.26496,
      "grad_norm": 0.4102991819381714,
      "learning_rate": 1.4240034028766125e-05,
      "loss": 0.5843,
      "step": 2553
    },
    {
      "epoch": 3.26624,
      "grad_norm": 0.41700980067253113,
      "learning_rate": 1.4226581037104312e-05,
      "loss": 0.5628,
      "step": 2554
    },
    {
      "epoch": 3.26752,
      "grad_norm": 0.4093681871891022,
      "learning_rate": 1.4213133380942678e-05,
      "loss": 0.5269,
      "step": 2555
    },
    {
      "epoch": 3.2688,
      "grad_norm": 0.39923086762428284,
      "learning_rate": 1.4199691069511387e-05,
      "loss": 0.4892,
      "step": 2556
    },
    {
      "epoch": 3.27008,
      "grad_norm": 0.4558875858783722,
      "learning_rate": 1.4186254112036944e-05,
      "loss": 0.5307,
      "step": 2557
    },
    {
      "epoch": 3.27136,
      "grad_norm": 0.4435747265815735,
      "learning_rate": 1.4172822517742178e-05,
      "loss": 0.5436,
      "step": 2558
    },
    {
      "epoch": 3.27264,
      "grad_norm": 0.41236135363578796,
      "learning_rate": 1.4159396295846222e-05,
      "loss": 0.5348,
      "step": 2559
    },
    {
      "epoch": 3.27392,
      "grad_norm": 0.4405274987220764,
      "learning_rate": 1.4145975455564544e-05,
      "loss": 0.5427,
      "step": 2560
    },
    {
      "epoch": 3.2752,
      "grad_norm": 0.41194450855255127,
      "learning_rate": 1.4132560006108911e-05,
      "loss": 0.5314,
      "step": 2561
    },
    {
      "epoch": 3.27648,
      "grad_norm": 0.41615167260169983,
      "learning_rate": 1.4119149956687369e-05,
      "loss": 0.5084,
      "step": 2562
    },
    {
      "epoch": 3.27776,
      "grad_norm": 0.4054078757762909,
      "learning_rate": 1.4105745316504282e-05,
      "loss": 0.5345,
      "step": 2563
    },
    {
      "epoch": 3.27904,
      "grad_norm": 0.43417343497276306,
      "learning_rate": 1.4092346094760308e-05,
      "loss": 0.5202,
      "step": 2564
    },
    {
      "epoch": 3.28032,
      "grad_norm": 0.4273699223995209,
      "learning_rate": 1.4078952300652353e-05,
      "loss": 0.533,
      "step": 2565
    },
    {
      "epoch": 3.2816,
      "grad_norm": 0.44152095913887024,
      "learning_rate": 1.4065563943373626e-05,
      "loss": 0.5378,
      "step": 2566
    },
    {
      "epoch": 3.28288,
      "grad_norm": 0.4138184189796448,
      "learning_rate": 1.4052181032113596e-05,
      "loss": 0.5315,
      "step": 2567
    },
    {
      "epoch": 3.28416,
      "grad_norm": 0.4269736111164093,
      "learning_rate": 1.4038803576057985e-05,
      "loss": 0.557,
      "step": 2568
    },
    {
      "epoch": 3.28544,
      "grad_norm": 0.415829598903656,
      "learning_rate": 1.4025431584388779e-05,
      "loss": 0.5354,
      "step": 2569
    },
    {
      "epoch": 3.28672,
      "grad_norm": 0.4068555235862732,
      "learning_rate": 1.4012065066284218e-05,
      "loss": 0.5208,
      "step": 2570
    },
    {
      "epoch": 3.288,
      "grad_norm": 0.4118039608001709,
      "learning_rate": 1.3998704030918784e-05,
      "loss": 0.5275,
      "step": 2571
    },
    {
      "epoch": 3.2892799999999998,
      "grad_norm": 0.4198782444000244,
      "learning_rate": 1.3985348487463181e-05,
      "loss": 0.5498,
      "step": 2572
    },
    {
      "epoch": 3.29056,
      "grad_norm": 0.39979609847068787,
      "learning_rate": 1.397199844508436e-05,
      "loss": 0.54,
      "step": 2573
    },
    {
      "epoch": 3.29184,
      "grad_norm": 0.40346163511276245,
      "learning_rate": 1.3958653912945494e-05,
      "loss": 0.5318,
      "step": 2574
    },
    {
      "epoch": 3.29312,
      "grad_norm": 0.38004857301712036,
      "learning_rate": 1.3945314900205959e-05,
      "loss": 0.527,
      "step": 2575
    },
    {
      "epoch": 3.2944,
      "grad_norm": 0.415242463350296,
      "learning_rate": 1.3931981416021361e-05,
      "loss": 0.5517,
      "step": 2576
    },
    {
      "epoch": 3.29568,
      "grad_norm": 0.42500096559524536,
      "learning_rate": 1.3918653469543508e-05,
      "loss": 0.4956,
      "step": 2577
    },
    {
      "epoch": 3.29696,
      "grad_norm": 0.4031103253364563,
      "learning_rate": 1.3905331069920393e-05,
      "loss": 0.5012,
      "step": 2578
    },
    {
      "epoch": 3.29824,
      "grad_norm": 0.39542046189308167,
      "learning_rate": 1.3892014226296219e-05,
      "loss": 0.5517,
      "step": 2579
    },
    {
      "epoch": 3.2995200000000002,
      "grad_norm": 0.41314658522605896,
      "learning_rate": 1.3878702947811369e-05,
      "loss": 0.552,
      "step": 2580
    },
    {
      "epoch": 3.2995200000000002,
      "eval_loss": 0.6949470043182373,
      "eval_runtime": 19.6109,
      "eval_samples_per_second": 25.598,
      "eval_steps_per_second": 3.213,
      "step": 2580
    },
    {
      "epoch": 3.3008,
      "grad_norm": 0.4166899621486664,
      "learning_rate": 1.38653972436024e-05,
      "loss": 0.5432,
      "step": 2581
    },
    {
      "epoch": 3.30208,
      "grad_norm": 0.39990362524986267,
      "learning_rate": 1.3852097122802048e-05,
      "loss": 0.5117,
      "step": 2582
    },
    {
      "epoch": 3.30336,
      "grad_norm": 0.37661391496658325,
      "learning_rate": 1.3838802594539228e-05,
      "loss": 0.5324,
      "step": 2583
    },
    {
      "epoch": 3.30464,
      "grad_norm": 0.40337276458740234,
      "learning_rate": 1.3825513667938995e-05,
      "loss": 0.5615,
      "step": 2584
    },
    {
      "epoch": 3.30592,
      "grad_norm": 0.38411661982536316,
      "learning_rate": 1.3812230352122567e-05,
      "loss": 0.5292,
      "step": 2585
    },
    {
      "epoch": 3.3072,
      "grad_norm": 0.38597792387008667,
      "learning_rate": 1.3798952656207329e-05,
      "loss": 0.5362,
      "step": 2586
    },
    {
      "epoch": 3.30848,
      "grad_norm": 0.4000000059604645,
      "learning_rate": 1.378568058930678e-05,
      "loss": 0.5253,
      "step": 2587
    },
    {
      "epoch": 3.30976,
      "grad_norm": 0.4042285978794098,
      "learning_rate": 1.3772414160530569e-05,
      "loss": 0.5556,
      "step": 2588
    },
    {
      "epoch": 3.31104,
      "grad_norm": 0.4033012092113495,
      "learning_rate": 1.3759153378984487e-05,
      "loss": 0.5405,
      "step": 2589
    },
    {
      "epoch": 3.31232,
      "grad_norm": 0.41891181468963623,
      "learning_rate": 1.3745898253770418e-05,
      "loss": 0.516,
      "step": 2590
    },
    {
      "epoch": 3.3136,
      "grad_norm": 0.3944721817970276,
      "learning_rate": 1.3732648793986397e-05,
      "loss": 0.5463,
      "step": 2591
    },
    {
      "epoch": 3.31488,
      "grad_norm": 0.4229712188243866,
      "learning_rate": 1.3719405008726553e-05,
      "loss": 0.5387,
      "step": 2592
    },
    {
      "epoch": 3.31616,
      "grad_norm": 0.3930949866771698,
      "learning_rate": 1.3706166907081113e-05,
      "loss": 0.5363,
      "step": 2593
    },
    {
      "epoch": 3.31744,
      "grad_norm": 0.4258730411529541,
      "learning_rate": 1.3692934498136422e-05,
      "loss": 0.5659,
      "step": 2594
    },
    {
      "epoch": 3.31872,
      "grad_norm": 0.3744409680366516,
      "learning_rate": 1.3679707790974917e-05,
      "loss": 0.5152,
      "step": 2595
    },
    {
      "epoch": 3.32,
      "grad_norm": 0.35936257243156433,
      "learning_rate": 1.3666486794675092e-05,
      "loss": 0.5294,
      "step": 2596
    },
    {
      "epoch": 3.32128,
      "grad_norm": 0.4027958810329437,
      "learning_rate": 1.365327151831155e-05,
      "loss": 0.5246,
      "step": 2597
    },
    {
      "epoch": 3.32256,
      "grad_norm": 0.42305245995521545,
      "learning_rate": 1.364006197095497e-05,
      "loss": 0.5568,
      "step": 2598
    },
    {
      "epoch": 3.32384,
      "grad_norm": 0.3850555419921875,
      "learning_rate": 1.3626858161672074e-05,
      "loss": 0.5209,
      "step": 2599
    },
    {
      "epoch": 3.32512,
      "grad_norm": 0.3785809874534607,
      "learning_rate": 1.3613660099525665e-05,
      "loss": 0.5343,
      "step": 2600
    },
    {
      "epoch": 3.3264,
      "grad_norm": 0.40879759192466736,
      "learning_rate": 1.3600467793574605e-05,
      "loss": 0.542,
      "step": 2601
    },
    {
      "epoch": 3.32768,
      "grad_norm": 0.4017791748046875,
      "learning_rate": 1.3587281252873781e-05,
      "loss": 0.5655,
      "step": 2602
    },
    {
      "epoch": 3.32896,
      "grad_norm": 0.39526045322418213,
      "learning_rate": 1.3574100486474152e-05,
      "loss": 0.536,
      "step": 2603
    },
    {
      "epoch": 3.33024,
      "grad_norm": 0.418507844209671,
      "learning_rate": 1.3560925503422696e-05,
      "loss": 0.5313,
      "step": 2604
    },
    {
      "epoch": 3.33152,
      "grad_norm": 0.3925471901893616,
      "learning_rate": 1.3547756312762422e-05,
      "loss": 0.521,
      "step": 2605
    },
    {
      "epoch": 3.3327999999999998,
      "grad_norm": 0.3932124972343445,
      "learning_rate": 1.3534592923532364e-05,
      "loss": 0.5525,
      "step": 2606
    },
    {
      "epoch": 3.33408,
      "grad_norm": 0.40973982214927673,
      "learning_rate": 1.352143534476759e-05,
      "loss": 0.559,
      "step": 2607
    },
    {
      "epoch": 3.33536,
      "grad_norm": 0.4215923249721527,
      "learning_rate": 1.350828358549915e-05,
      "loss": 0.5137,
      "step": 2608
    },
    {
      "epoch": 3.33664,
      "grad_norm": 0.38590431213378906,
      "learning_rate": 1.3495137654754137e-05,
      "loss": 0.4889,
      "step": 2609
    },
    {
      "epoch": 3.33792,
      "grad_norm": 0.4656379222869873,
      "learning_rate": 1.34819975615556e-05,
      "loss": 0.553,
      "step": 2610
    },
    {
      "epoch": 3.33792,
      "eval_loss": 0.6948084235191345,
      "eval_runtime": 19.6058,
      "eval_samples_per_second": 25.605,
      "eval_steps_per_second": 3.213,
      "step": 2610
    },
    {
      "epoch": 3.3392,
      "grad_norm": 0.39811092615127563,
      "learning_rate": 1.3468863314922614e-05,
      "loss": 0.4941,
      "step": 2611
    },
    {
      "epoch": 3.34048,
      "grad_norm": 0.4159993529319763,
      "learning_rate": 1.3455734923870232e-05,
      "loss": 0.5094,
      "step": 2612
    },
    {
      "epoch": 3.34176,
      "grad_norm": 0.42880547046661377,
      "learning_rate": 1.3442612397409482e-05,
      "loss": 0.5345,
      "step": 2613
    },
    {
      "epoch": 3.3430400000000002,
      "grad_norm": 0.4012126326560974,
      "learning_rate": 1.3429495744547377e-05,
      "loss": 0.5338,
      "step": 2614
    },
    {
      "epoch": 3.34432,
      "grad_norm": 0.39612531661987305,
      "learning_rate": 1.3416384974286895e-05,
      "loss": 0.5686,
      "step": 2615
    },
    {
      "epoch": 3.3456,
      "grad_norm": 0.4013139009475708,
      "learning_rate": 1.3403280095626965e-05,
      "loss": 0.5157,
      "step": 2616
    },
    {
      "epoch": 3.34688,
      "grad_norm": 0.38638120889663696,
      "learning_rate": 1.3390181117562483e-05,
      "loss": 0.5345,
      "step": 2617
    },
    {
      "epoch": 3.34816,
      "grad_norm": 0.39033421874046326,
      "learning_rate": 1.3377088049084307e-05,
      "loss": 0.5189,
      "step": 2618
    },
    {
      "epoch": 3.34944,
      "grad_norm": 0.40154018998146057,
      "learning_rate": 1.3364000899179205e-05,
      "loss": 0.5193,
      "step": 2619
    },
    {
      "epoch": 3.35072,
      "grad_norm": 0.37580278515815735,
      "learning_rate": 1.3350919676829915e-05,
      "loss": 0.5309,
      "step": 2620
    },
    {
      "epoch": 3.352,
      "grad_norm": 0.4394700527191162,
      "learning_rate": 1.3337844391015096e-05,
      "loss": 0.542,
      "step": 2621
    },
    {
      "epoch": 3.35328,
      "grad_norm": 0.426400363445282,
      "learning_rate": 1.3324775050709321e-05,
      "loss": 0.56,
      "step": 2622
    },
    {
      "epoch": 3.35456,
      "grad_norm": 0.38398614525794983,
      "learning_rate": 1.3311711664883098e-05,
      "loss": 0.5591,
      "step": 2623
    },
    {
      "epoch": 3.35584,
      "grad_norm": 0.4110603630542755,
      "learning_rate": 1.3298654242502844e-05,
      "loss": 0.5162,
      "step": 2624
    },
    {
      "epoch": 3.35712,
      "grad_norm": 0.4002382457256317,
      "learning_rate": 1.3285602792530869e-05,
      "loss": 0.5492,
      "step": 2625
    },
    {
      "epoch": 3.3584,
      "grad_norm": 0.38603073358535767,
      "learning_rate": 1.3272557323925406e-05,
      "loss": 0.5239,
      "step": 2626
    },
    {
      "epoch": 3.35968,
      "grad_norm": 0.4215501844882965,
      "learning_rate": 1.3259517845640576e-05,
      "loss": 0.5463,
      "step": 2627
    },
    {
      "epoch": 3.36096,
      "grad_norm": 0.39184245467185974,
      "learning_rate": 1.3246484366626367e-05,
      "loss": 0.5139,
      "step": 2628
    },
    {
      "epoch": 3.36224,
      "grad_norm": 0.41844233870506287,
      "learning_rate": 1.3233456895828677e-05,
      "loss": 0.5025,
      "step": 2629
    },
    {
      "epoch": 3.36352,
      "grad_norm": 0.4184134602546692,
      "learning_rate": 1.3220435442189275e-05,
      "loss": 0.5256,
      "step": 2630
    },
    {
      "epoch": 3.3648,
      "grad_norm": 0.4049496054649353,
      "learning_rate": 1.320742001464578e-05,
      "loss": 0.5393,
      "step": 2631
    },
    {
      "epoch": 3.36608,
      "grad_norm": 0.4165656864643097,
      "learning_rate": 1.3194410622131698e-05,
      "loss": 0.5627,
      "step": 2632
    },
    {
      "epoch": 3.36736,
      "grad_norm": 0.40335169434547424,
      "learning_rate": 1.3181407273576388e-05,
      "loss": 0.5408,
      "step": 2633
    },
    {
      "epoch": 3.36864,
      "grad_norm": 0.3936487138271332,
      "learning_rate": 1.3168409977905044e-05,
      "loss": 0.5313,
      "step": 2634
    },
    {
      "epoch": 3.36992,
      "grad_norm": 0.442121684551239,
      "learning_rate": 1.315541874403873e-05,
      "loss": 0.5639,
      "step": 2635
    },
    {
      "epoch": 3.3712,
      "grad_norm": 0.4155605137348175,
      "learning_rate": 1.3142433580894328e-05,
      "loss": 0.5551,
      "step": 2636
    },
    {
      "epoch": 3.37248,
      "grad_norm": 0.38098034262657166,
      "learning_rate": 1.3129454497384565e-05,
      "loss": 0.5386,
      "step": 2637
    },
    {
      "epoch": 3.37376,
      "grad_norm": 0.42181500792503357,
      "learning_rate": 1.3116481502418e-05,
      "loss": 0.543,
      "step": 2638
    },
    {
      "epoch": 3.37504,
      "grad_norm": 0.44447389245033264,
      "learning_rate": 1.310351460489899e-05,
      "loss": 0.5307,
      "step": 2639
    },
    {
      "epoch": 3.3763199999999998,
      "grad_norm": 0.4100289046764374,
      "learning_rate": 1.3090553813727733e-05,
      "loss": 0.5347,
      "step": 2640
    },
    {
      "epoch": 3.3763199999999998,
      "eval_loss": 0.695135235786438,
      "eval_runtime": 19.6225,
      "eval_samples_per_second": 25.583,
      "eval_steps_per_second": 3.211,
      "step": 2640
    },
    {
      "epoch": 3.3776,
      "grad_norm": 0.3707532286643982,
      "learning_rate": 1.3077599137800232e-05,
      "loss": 0.5317,
      "step": 2641
    },
    {
      "epoch": 3.37888,
      "grad_norm": 0.4214347302913666,
      "learning_rate": 1.3064650586008268e-05,
      "loss": 0.5341,
      "step": 2642
    },
    {
      "epoch": 3.38016,
      "grad_norm": 0.44026046991348267,
      "learning_rate": 1.3051708167239441e-05,
      "loss": 0.5371,
      "step": 2643
    },
    {
      "epoch": 3.38144,
      "grad_norm": 0.3822016716003418,
      "learning_rate": 1.3038771890377151e-05,
      "loss": 0.5397,
      "step": 2644
    },
    {
      "epoch": 3.38272,
      "grad_norm": 0.37577933073043823,
      "learning_rate": 1.3025841764300548e-05,
      "loss": 0.5344,
      "step": 2645
    },
    {
      "epoch": 3.384,
      "grad_norm": 0.3922223746776581,
      "learning_rate": 1.3012917797884593e-05,
      "loss": 0.5156,
      "step": 2646
    },
    {
      "epoch": 3.38528,
      "grad_norm": 0.3870489299297333,
      "learning_rate": 1.3000000000000006e-05,
      "loss": 0.5425,
      "step": 2647
    },
    {
      "epoch": 3.3865600000000002,
      "grad_norm": 0.3950110077857971,
      "learning_rate": 1.2987088379513261e-05,
      "loss": 0.5246,
      "step": 2648
    },
    {
      "epoch": 3.38784,
      "grad_norm": 0.4124525487422943,
      "learning_rate": 1.2974182945286615e-05,
      "loss": 0.5352,
      "step": 2649
    },
    {
      "epoch": 3.38912,
      "grad_norm": 0.3822491765022278,
      "learning_rate": 1.2961283706178076e-05,
      "loss": 0.5219,
      "step": 2650
    },
    {
      "epoch": 3.3904,
      "grad_norm": 0.40268436074256897,
      "learning_rate": 1.2948390671041373e-05,
      "loss": 0.5462,
      "step": 2651
    },
    {
      "epoch": 3.39168,
      "grad_norm": 0.43101346492767334,
      "learning_rate": 1.2935503848726003e-05,
      "loss": 0.5495,
      "step": 2652
    },
    {
      "epoch": 3.39296,
      "grad_norm": 0.3869779407978058,
      "learning_rate": 1.2922623248077204e-05,
      "loss": 0.5159,
      "step": 2653
    },
    {
      "epoch": 3.39424,
      "grad_norm": 0.39610037207603455,
      "learning_rate": 1.2909748877935913e-05,
      "loss": 0.5108,
      "step": 2654
    },
    {
      "epoch": 3.39552,
      "grad_norm": 0.4522349238395691,
      "learning_rate": 1.2896880747138813e-05,
      "loss": 0.5184,
      "step": 2655
    },
    {
      "epoch": 3.3968,
      "grad_norm": 0.4347802400588989,
      "learning_rate": 1.2884018864518313e-05,
      "loss": 0.5196,
      "step": 2656
    },
    {
      "epoch": 3.39808,
      "grad_norm": 0.38997015357017517,
      "learning_rate": 1.28711632389025e-05,
      "loss": 0.4958,
      "step": 2657
    },
    {
      "epoch": 3.39936,
      "grad_norm": 0.40131741762161255,
      "learning_rate": 1.2858313879115198e-05,
      "loss": 0.526,
      "step": 2658
    },
    {
      "epoch": 3.40064,
      "grad_norm": 0.4053361117839813,
      "learning_rate": 1.2845470793975925e-05,
      "loss": 0.5018,
      "step": 2659
    },
    {
      "epoch": 3.40192,
      "grad_norm": 0.3757518231868744,
      "learning_rate": 1.2832633992299872e-05,
      "loss": 0.5349,
      "step": 2660
    },
    {
      "epoch": 3.4032,
      "grad_norm": 0.4149254262447357,
      "learning_rate": 1.2819803482897944e-05,
      "loss": 0.503,
      "step": 2661
    },
    {
      "epoch": 3.40448,
      "grad_norm": 0.452566921710968,
      "learning_rate": 1.2806979274576703e-05,
      "loss": 0.556,
      "step": 2662
    },
    {
      "epoch": 3.40576,
      "grad_norm": 0.43798044323921204,
      "learning_rate": 1.2794161376138403e-05,
      "loss": 0.5469,
      "step": 2663
    },
    {
      "epoch": 3.40704,
      "grad_norm": 0.406974196434021,
      "learning_rate": 1.2781349796380968e-05,
      "loss": 0.5313,
      "step": 2664
    },
    {
      "epoch": 3.40832,
      "grad_norm": 0.40056514739990234,
      "learning_rate": 1.2768544544097967e-05,
      "loss": 0.5166,
      "step": 2665
    },
    {
      "epoch": 3.4096,
      "grad_norm": 0.38950440287590027,
      "learning_rate": 1.2755745628078642e-05,
      "loss": 0.5121,
      "step": 2666
    },
    {
      "epoch": 3.41088,
      "grad_norm": 0.3675340414047241,
      "learning_rate": 1.2742953057107888e-05,
      "loss": 0.5468,
      "step": 2667
    },
    {
      "epoch": 3.41216,
      "grad_norm": 0.3688168227672577,
      "learning_rate": 1.2730166839966227e-05,
      "loss": 0.5076,
      "step": 2668
    },
    {
      "epoch": 3.41344,
      "grad_norm": 0.4008009433746338,
      "learning_rate": 1.271738698542984e-05,
      "loss": 0.5291,
      "step": 2669
    },
    {
      "epoch": 3.41472,
      "grad_norm": 0.36751988530158997,
      "learning_rate": 1.2704613502270533e-05,
      "loss": 0.5119,
      "step": 2670
    },
    {
      "epoch": 3.41472,
      "eval_loss": 0.6944306492805481,
      "eval_runtime": 19.5985,
      "eval_samples_per_second": 25.614,
      "eval_steps_per_second": 3.215,
      "step": 2670
    },
    {
      "epoch": 3.416,
      "grad_norm": 0.3984326422214508,
      "learning_rate": 1.2691846399255727e-05,
      "loss": 0.508,
      "step": 2671
    },
    {
      "epoch": 3.41728,
      "grad_norm": 0.41484883427619934,
      "learning_rate": 1.2679085685148484e-05,
      "loss": 0.5411,
      "step": 2672
    },
    {
      "epoch": 3.41856,
      "grad_norm": 0.41740524768829346,
      "learning_rate": 1.2666331368707475e-05,
      "loss": 0.5539,
      "step": 2673
    },
    {
      "epoch": 3.4198399999999998,
      "grad_norm": 0.3806936740875244,
      "learning_rate": 1.2653583458686964e-05,
      "loss": 0.5081,
      "step": 2674
    },
    {
      "epoch": 3.42112,
      "grad_norm": 0.38713473081588745,
      "learning_rate": 1.2640841963836836e-05,
      "loss": 0.512,
      "step": 2675
    },
    {
      "epoch": 3.4224,
      "grad_norm": 0.4212294816970825,
      "learning_rate": 1.2628106892902575e-05,
      "loss": 0.5534,
      "step": 2676
    },
    {
      "epoch": 3.42368,
      "grad_norm": 0.4053986668586731,
      "learning_rate": 1.2615378254625235e-05,
      "loss": 0.5491,
      "step": 2677
    },
    {
      "epoch": 3.42496,
      "grad_norm": 0.40320926904678345,
      "learning_rate": 1.2602656057741469e-05,
      "loss": 0.5315,
      "step": 2678
    },
    {
      "epoch": 3.42624,
      "grad_norm": 0.4245164692401886,
      "learning_rate": 1.2589940310983519e-05,
      "loss": 0.5521,
      "step": 2679
    },
    {
      "epoch": 3.42752,
      "grad_norm": 0.4037604033946991,
      "learning_rate": 1.2577231023079176e-05,
      "loss": 0.5345,
      "step": 2680
    },
    {
      "epoch": 3.4288,
      "grad_norm": 0.42574286460876465,
      "learning_rate": 1.2564528202751811e-05,
      "loss": 0.5117,
      "step": 2681
    },
    {
      "epoch": 3.4300800000000002,
      "grad_norm": 0.39327573776245117,
      "learning_rate": 1.2551831858720368e-05,
      "loss": 0.516,
      "step": 2682
    },
    {
      "epoch": 3.43136,
      "grad_norm": 0.3738795816898346,
      "learning_rate": 1.253914199969931e-05,
      "loss": 0.509,
      "step": 2683
    },
    {
      "epoch": 3.43264,
      "grad_norm": 0.4271436035633087,
      "learning_rate": 1.2526458634398683e-05,
      "loss": 0.5407,
      "step": 2684
    },
    {
      "epoch": 3.43392,
      "grad_norm": 0.41948506236076355,
      "learning_rate": 1.2513781771524073e-05,
      "loss": 0.5342,
      "step": 2685
    },
    {
      "epoch": 3.4352,
      "grad_norm": 0.4017522633075714,
      "learning_rate": 1.2501111419776579e-05,
      "loss": 0.5465,
      "step": 2686
    },
    {
      "epoch": 3.43648,
      "grad_norm": 0.39963582158088684,
      "learning_rate": 1.2488447587852859e-05,
      "loss": 0.5328,
      "step": 2687
    },
    {
      "epoch": 3.43776,
      "grad_norm": 0.38865578174591064,
      "learning_rate": 1.2475790284445064e-05,
      "loss": 0.5248,
      "step": 2688
    },
    {
      "epoch": 3.43904,
      "grad_norm": 0.39466941356658936,
      "learning_rate": 1.2463139518240896e-05,
      "loss": 0.5256,
      "step": 2689
    },
    {
      "epoch": 3.44032,
      "grad_norm": 0.4313627779483795,
      "learning_rate": 1.2450495297923566e-05,
      "loss": 0.509,
      "step": 2690
    },
    {
      "epoch": 3.4416,
      "grad_norm": 0.39986366033554077,
      "learning_rate": 1.2437857632171762e-05,
      "loss": 0.5036,
      "step": 2691
    },
    {
      "epoch": 3.44288,
      "grad_norm": 0.4173099994659424,
      "learning_rate": 1.2425226529659702e-05,
      "loss": 0.5491,
      "step": 2692
    },
    {
      "epoch": 3.44416,
      "grad_norm": 0.40632057189941406,
      "learning_rate": 1.2412601999057099e-05,
      "loss": 0.5171,
      "step": 2693
    },
    {
      "epoch": 3.44544,
      "grad_norm": 0.420969694852829,
      "learning_rate": 1.239998404902914e-05,
      "loss": 0.5357,
      "step": 2694
    },
    {
      "epoch": 3.44672,
      "grad_norm": 0.44641727209091187,
      "learning_rate": 1.2387372688236499e-05,
      "loss": 0.4988,
      "step": 2695
    },
    {
      "epoch": 3.448,
      "grad_norm": 0.4285929203033447,
      "learning_rate": 1.2374767925335343e-05,
      "loss": 0.5291,
      "step": 2696
    },
    {
      "epoch": 3.44928,
      "grad_norm": 0.3910332918167114,
      "learning_rate": 1.2362169768977284e-05,
      "loss": 0.556,
      "step": 2697
    },
    {
      "epoch": 3.45056,
      "grad_norm": 0.4395662248134613,
      "learning_rate": 1.2349578227809417e-05,
      "loss": 0.4944,
      "step": 2698
    },
    {
      "epoch": 3.45184,
      "grad_norm": 0.3824242353439331,
      "learning_rate": 1.2336993310474306e-05,
      "loss": 0.5211,
      "step": 2699
    },
    {
      "epoch": 3.45312,
      "grad_norm": 0.39866331219673157,
      "learning_rate": 1.2324415025609939e-05,
      "loss": 0.5255,
      "step": 2700
    },
    {
      "epoch": 3.45312,
      "eval_loss": 0.6942668557167053,
      "eval_runtime": 19.5978,
      "eval_samples_per_second": 25.615,
      "eval_steps_per_second": 3.215,
      "step": 2700
    },
    {
      "epoch": 3.4544,
      "grad_norm": 0.3997983932495117,
      "learning_rate": 1.2311843381849773e-05,
      "loss": 0.5055,
      "step": 2701
    },
    {
      "epoch": 3.45568,
      "grad_norm": 0.41547828912734985,
      "learning_rate": 1.2299278387822713e-05,
      "loss": 0.5637,
      "step": 2702
    },
    {
      "epoch": 3.45696,
      "grad_norm": 0.3631564974784851,
      "learning_rate": 1.2286720052153069e-05,
      "loss": 0.5096,
      "step": 2703
    },
    {
      "epoch": 3.45824,
      "grad_norm": 0.41522154211997986,
      "learning_rate": 1.2274168383460613e-05,
      "loss": 0.5296,
      "step": 2704
    },
    {
      "epoch": 3.45952,
      "grad_norm": 0.4168614149093628,
      "learning_rate": 1.2261623390360535e-05,
      "loss": 0.5181,
      "step": 2705
    },
    {
      "epoch": 3.4608,
      "grad_norm": 0.36784303188323975,
      "learning_rate": 1.224908508146342e-05,
      "loss": 0.5194,
      "step": 2706
    },
    {
      "epoch": 3.46208,
      "grad_norm": 0.3977992534637451,
      "learning_rate": 1.2236553465375289e-05,
      "loss": 0.5365,
      "step": 2707
    },
    {
      "epoch": 3.4633599999999998,
      "grad_norm": 0.4519791305065155,
      "learning_rate": 1.2224028550697571e-05,
      "loss": 0.5519,
      "step": 2708
    },
    {
      "epoch": 3.46464,
      "grad_norm": 0.4161181151866913,
      "learning_rate": 1.2211510346027066e-05,
      "loss": 0.4929,
      "step": 2709
    },
    {
      "epoch": 3.46592,
      "grad_norm": 0.39703962206840515,
      "learning_rate": 1.2198998859956004e-05,
      "loss": 0.5416,
      "step": 2710
    },
    {
      "epoch": 3.4672,
      "grad_norm": 0.42353489995002747,
      "learning_rate": 1.218649410107199e-05,
      "loss": 0.5428,
      "step": 2711
    },
    {
      "epoch": 3.46848,
      "grad_norm": 0.4087219536304474,
      "learning_rate": 1.2173996077957997e-05,
      "loss": 0.5188,
      "step": 2712
    },
    {
      "epoch": 3.46976,
      "grad_norm": 0.41100871562957764,
      "learning_rate": 1.2161504799192394e-05,
      "loss": 0.526,
      "step": 2713
    },
    {
      "epoch": 3.47104,
      "grad_norm": 0.40871357917785645,
      "learning_rate": 1.214902027334892e-05,
      "loss": 0.5155,
      "step": 2714
    },
    {
      "epoch": 3.47232,
      "grad_norm": 0.4100225269794464,
      "learning_rate": 1.2136542508996658e-05,
      "loss": 0.5083,
      "step": 2715
    },
    {
      "epoch": 3.4736000000000002,
      "grad_norm": 0.40770429372787476,
      "learning_rate": 1.2124071514700079e-05,
      "loss": 0.5349,
      "step": 2716
    },
    {
      "epoch": 3.47488,
      "grad_norm": 0.42250382900238037,
      "learning_rate": 1.2111607299018995e-05,
      "loss": 0.5476,
      "step": 2717
    },
    {
      "epoch": 3.47616,
      "grad_norm": 0.4113774001598358,
      "learning_rate": 1.2099149870508548e-05,
      "loss": 0.5197,
      "step": 2718
    },
    {
      "epoch": 3.47744,
      "grad_norm": 0.3929632306098938,
      "learning_rate": 1.2086699237719249e-05,
      "loss": 0.5327,
      "step": 2719
    },
    {
      "epoch": 3.47872,
      "grad_norm": 0.3873066008090973,
      "learning_rate": 1.2074255409196937e-05,
      "loss": 0.5395,
      "step": 2720
    },
    {
      "epoch": 3.48,
      "grad_norm": 0.3998710811138153,
      "learning_rate": 1.2061818393482763e-05,
      "loss": 0.5418,
      "step": 2721
    },
    {
      "epoch": 3.48128,
      "grad_norm": 0.41157862544059753,
      "learning_rate": 1.2049388199113218e-05,
      "loss": 0.558,
      "step": 2722
    },
    {
      "epoch": 3.48256,
      "grad_norm": 0.41144317388534546,
      "learning_rate": 1.2036964834620123e-05,
      "loss": 0.5376,
      "step": 2723
    },
    {
      "epoch": 3.48384,
      "grad_norm": 0.40244075655937195,
      "learning_rate": 1.2024548308530576e-05,
      "loss": 0.5155,
      "step": 2724
    },
    {
      "epoch": 3.48512,
      "grad_norm": 0.400608628988266,
      "learning_rate": 1.2012138629367013e-05,
      "loss": 0.5432,
      "step": 2725
    },
    {
      "epoch": 3.4864,
      "grad_norm": 0.38311827182769775,
      "learning_rate": 1.1999735805647165e-05,
      "loss": 0.5148,
      "step": 2726
    },
    {
      "epoch": 3.48768,
      "grad_norm": 0.3999381363391876,
      "learning_rate": 1.1987339845884036e-05,
      "loss": 0.5284,
      "step": 2727
    },
    {
      "epoch": 3.48896,
      "grad_norm": 0.38908156752586365,
      "learning_rate": 1.1974950758585941e-05,
      "loss": 0.5349,
      "step": 2728
    },
    {
      "epoch": 3.49024,
      "grad_norm": 0.40771716833114624,
      "learning_rate": 1.1962568552256476e-05,
      "loss": 0.5081,
      "step": 2729
    },
    {
      "epoch": 3.49152,
      "grad_norm": 0.43971920013427734,
      "learning_rate": 1.1950193235394507e-05,
      "loss": 0.5338,
      "step": 2730
    },
    {
      "epoch": 3.49152,
      "eval_loss": 0.6935325264930725,
      "eval_runtime": 19.5963,
      "eval_samples_per_second": 25.617,
      "eval_steps_per_second": 3.215,
      "step": 2730
    },
    {
      "epoch": 3.4928,
      "grad_norm": 0.3733380436897278,
      "learning_rate": 1.193782481649417e-05,
      "loss": 0.5386,
      "step": 2731
    },
    {
      "epoch": 3.49408,
      "grad_norm": 0.41533342003822327,
      "learning_rate": 1.192546330404487e-05,
      "loss": 0.5134,
      "step": 2732
    },
    {
      "epoch": 3.49536,
      "grad_norm": 0.42003339529037476,
      "learning_rate": 1.1913108706531284e-05,
      "loss": 0.5228,
      "step": 2733
    },
    {
      "epoch": 3.49664,
      "grad_norm": 0.408920019865036,
      "learning_rate": 1.1900761032433308e-05,
      "loss": 0.5508,
      "step": 2734
    },
    {
      "epoch": 3.49792,
      "grad_norm": 0.3803151845932007,
      "learning_rate": 1.1888420290226127e-05,
      "loss": 0.5339,
      "step": 2735
    },
    {
      "epoch": 3.4992,
      "grad_norm": 0.46105483174324036,
      "learning_rate": 1.187608648838014e-05,
      "loss": 0.5456,
      "step": 2736
    },
    {
      "epoch": 3.50048,
      "grad_norm": 0.4186250865459442,
      "learning_rate": 1.1863759635360994e-05,
      "loss": 0.526,
      "step": 2737
    },
    {
      "epoch": 3.50176,
      "grad_norm": 0.42904427647590637,
      "learning_rate": 1.185143973962956e-05,
      "loss": 0.5665,
      "step": 2738
    },
    {
      "epoch": 3.50304,
      "grad_norm": 0.42233777046203613,
      "learning_rate": 1.1839126809641953e-05,
      "loss": 0.5741,
      "step": 2739
    },
    {
      "epoch": 3.50432,
      "grad_norm": 0.39923760294914246,
      "learning_rate": 1.1826820853849465e-05,
      "loss": 0.5315,
      "step": 2740
    },
    {
      "epoch": 3.5056000000000003,
      "grad_norm": 0.4320422112941742,
      "learning_rate": 1.1814521880698649e-05,
      "loss": 0.5191,
      "step": 2741
    },
    {
      "epoch": 3.5068799999999998,
      "grad_norm": 0.40699610114097595,
      "learning_rate": 1.1802229898631246e-05,
      "loss": 0.4951,
      "step": 2742
    },
    {
      "epoch": 3.50816,
      "grad_norm": 0.4049845337867737,
      "learning_rate": 1.1789944916084177e-05,
      "loss": 0.5299,
      "step": 2743
    },
    {
      "epoch": 3.50944,
      "grad_norm": 0.4159267246723175,
      "learning_rate": 1.1777666941489594e-05,
      "loss": 0.535,
      "step": 2744
    },
    {
      "epoch": 3.51072,
      "grad_norm": 0.42587703466415405,
      "learning_rate": 1.176539598327482e-05,
      "loss": 0.5437,
      "step": 2745
    },
    {
      "epoch": 3.512,
      "grad_norm": 0.4023404121398926,
      "learning_rate": 1.1753132049862362e-05,
      "loss": 0.5374,
      "step": 2746
    },
    {
      "epoch": 3.51328,
      "grad_norm": 0.39031782746315,
      "learning_rate": 1.1740875149669907e-05,
      "loss": 0.532,
      "step": 2747
    },
    {
      "epoch": 3.51456,
      "grad_norm": 0.4122608006000519,
      "learning_rate": 1.1728625291110323e-05,
      "loss": 0.5412,
      "step": 2748
    },
    {
      "epoch": 3.51584,
      "grad_norm": 0.41631051898002625,
      "learning_rate": 1.1716382482591629e-05,
      "loss": 0.5244,
      "step": 2749
    },
    {
      "epoch": 3.5171200000000002,
      "grad_norm": 0.38406652212142944,
      "learning_rate": 1.170414673251702e-05,
      "loss": 0.5154,
      "step": 2750
    },
    {
      "epoch": 3.5183999999999997,
      "grad_norm": 0.38796326518058777,
      "learning_rate": 1.169191804928484e-05,
      "loss": 0.5154,
      "step": 2751
    },
    {
      "epoch": 3.51968,
      "grad_norm": 0.39699438214302063,
      "learning_rate": 1.1679696441288582e-05,
      "loss": 0.5342,
      "step": 2752
    },
    {
      "epoch": 3.52096,
      "grad_norm": 0.42582544684410095,
      "learning_rate": 1.1667481916916876e-05,
      "loss": 0.5764,
      "step": 2753
    },
    {
      "epoch": 3.52224,
      "grad_norm": 0.37467941641807556,
      "learning_rate": 1.165527448455351e-05,
      "loss": 0.5517,
      "step": 2754
    },
    {
      "epoch": 3.52352,
      "grad_norm": 0.3749716579914093,
      "learning_rate": 1.1643074152577382e-05,
      "loss": 0.5489,
      "step": 2755
    },
    {
      "epoch": 3.5248,
      "grad_norm": 0.43429484963417053,
      "learning_rate": 1.1630880929362526e-05,
      "loss": 0.4987,
      "step": 2756
    },
    {
      "epoch": 3.52608,
      "grad_norm": 0.39178505539894104,
      "learning_rate": 1.1618694823278104e-05,
      "loss": 0.5131,
      "step": 2757
    },
    {
      "epoch": 3.52736,
      "grad_norm": 0.3756830096244812,
      "learning_rate": 1.1606515842688379e-05,
      "loss": 0.5026,
      "step": 2758
    },
    {
      "epoch": 3.52864,
      "grad_norm": 0.3717840015888214,
      "learning_rate": 1.1594343995952728e-05,
      "loss": 0.5044,
      "step": 2759
    },
    {
      "epoch": 3.5299199999999997,
      "grad_norm": 0.4263432025909424,
      "learning_rate": 1.1582179291425644e-05,
      "loss": 0.5713,
      "step": 2760
    },
    {
      "epoch": 3.5299199999999997,
      "eval_loss": 0.6936967372894287,
      "eval_runtime": 19.6582,
      "eval_samples_per_second": 25.536,
      "eval_steps_per_second": 3.205,
      "step": 2760
    },
    {
      "epoch": 3.5312,
      "grad_norm": 0.4126444458961487,
      "learning_rate": 1.157002173745669e-05,
      "loss": 0.5315,
      "step": 2761
    },
    {
      "epoch": 3.53248,
      "grad_norm": 0.39197617769241333,
      "learning_rate": 1.1557871342390547e-05,
      "loss": 0.5604,
      "step": 2762
    },
    {
      "epoch": 3.53376,
      "grad_norm": 0.3963038921356201,
      "learning_rate": 1.1545728114566985e-05,
      "loss": 0.5167,
      "step": 2763
    },
    {
      "epoch": 3.53504,
      "grad_norm": 0.3763555586338043,
      "learning_rate": 1.1533592062320822e-05,
      "loss": 0.4986,
      "step": 2764
    },
    {
      "epoch": 3.53632,
      "grad_norm": 0.3591424226760864,
      "learning_rate": 1.1521463193981983e-05,
      "loss": 0.5127,
      "step": 2765
    },
    {
      "epoch": 3.5376,
      "grad_norm": 0.4119926989078522,
      "learning_rate": 1.1509341517875452e-05,
      "loss": 0.5435,
      "step": 2766
    },
    {
      "epoch": 3.53888,
      "grad_norm": 0.3975541591644287,
      "learning_rate": 1.1497227042321269e-05,
      "loss": 0.533,
      "step": 2767
    },
    {
      "epoch": 3.54016,
      "grad_norm": 0.3958500921726227,
      "learning_rate": 1.1485119775634548e-05,
      "loss": 0.5238,
      "step": 2768
    },
    {
      "epoch": 3.54144,
      "grad_norm": 0.3829525113105774,
      "learning_rate": 1.1473019726125427e-05,
      "loss": 0.5459,
      "step": 2769
    },
    {
      "epoch": 3.54272,
      "grad_norm": 0.3801780939102173,
      "learning_rate": 1.1460926902099123e-05,
      "loss": 0.5556,
      "step": 2770
    },
    {
      "epoch": 3.544,
      "grad_norm": 0.4168373942375183,
      "learning_rate": 1.1448841311855881e-05,
      "loss": 0.5076,
      "step": 2771
    },
    {
      "epoch": 3.54528,
      "grad_norm": 0.38964536786079407,
      "learning_rate": 1.1436762963690966e-05,
      "loss": 0.5245,
      "step": 2772
    },
    {
      "epoch": 3.54656,
      "grad_norm": 0.4036487340927124,
      "learning_rate": 1.142469186589469e-05,
      "loss": 0.5403,
      "step": 2773
    },
    {
      "epoch": 3.54784,
      "grad_norm": 0.3836483061313629,
      "learning_rate": 1.1412628026752395e-05,
      "loss": 0.5094,
      "step": 2774
    },
    {
      "epoch": 3.5491200000000003,
      "grad_norm": 0.3934383690357208,
      "learning_rate": 1.1400571454544412e-05,
      "loss": 0.5418,
      "step": 2775
    },
    {
      "epoch": 3.5504,
      "grad_norm": 0.3750290274620056,
      "learning_rate": 1.1388522157546107e-05,
      "loss": 0.5067,
      "step": 2776
    },
    {
      "epoch": 3.55168,
      "grad_norm": 0.4024461507797241,
      "learning_rate": 1.1376480144027853e-05,
      "loss": 0.5128,
      "step": 2777
    },
    {
      "epoch": 3.55296,
      "grad_norm": 0.3850504159927368,
      "learning_rate": 1.136444542225501e-05,
      "loss": 0.5306,
      "step": 2778
    },
    {
      "epoch": 3.55424,
      "grad_norm": 0.3832181692123413,
      "learning_rate": 1.1352418000487941e-05,
      "loss": 0.5302,
      "step": 2779
    },
    {
      "epoch": 3.55552,
      "grad_norm": 0.3990479111671448,
      "learning_rate": 1.1340397886982007e-05,
      "loss": 0.5314,
      "step": 2780
    },
    {
      "epoch": 3.5568,
      "grad_norm": 0.3864832818508148,
      "learning_rate": 1.1328385089987532e-05,
      "loss": 0.5387,
      "step": 2781
    },
    {
      "epoch": 3.55808,
      "grad_norm": 0.40432727336883545,
      "learning_rate": 1.1316379617749837e-05,
      "loss": 0.5374,
      "step": 2782
    },
    {
      "epoch": 3.55936,
      "grad_norm": 0.3554989695549011,
      "learning_rate": 1.1304381478509216e-05,
      "loss": 0.5057,
      "step": 2783
    },
    {
      "epoch": 3.5606400000000002,
      "grad_norm": 0.3895516097545624,
      "learning_rate": 1.1292390680500903e-05,
      "loss": 0.5479,
      "step": 2784
    },
    {
      "epoch": 3.5619199999999998,
      "grad_norm": 0.39768296480178833,
      "learning_rate": 1.1280407231955125e-05,
      "loss": 0.5539,
      "step": 2785
    },
    {
      "epoch": 3.5632,
      "grad_norm": 0.3957372307777405,
      "learning_rate": 1.1268431141097054e-05,
      "loss": 0.5305,
      "step": 2786
    },
    {
      "epoch": 3.56448,
      "grad_norm": 0.42029085755348206,
      "learning_rate": 1.1256462416146803e-05,
      "loss": 0.5444,
      "step": 2787
    },
    {
      "epoch": 3.56576,
      "grad_norm": 0.41207560896873474,
      "learning_rate": 1.1244501065319437e-05,
      "loss": 0.5405,
      "step": 2788
    },
    {
      "epoch": 3.56704,
      "grad_norm": 0.3855437636375427,
      "learning_rate": 1.1232547096824966e-05,
      "loss": 0.5237,
      "step": 2789
    },
    {
      "epoch": 3.56832,
      "grad_norm": 0.3809058666229248,
      "learning_rate": 1.1220600518868311e-05,
      "loss": 0.5275,
      "step": 2790
    },
    {
      "epoch": 3.56832,
      "eval_loss": 0.6925818920135498,
      "eval_runtime": 19.6318,
      "eval_samples_per_second": 25.571,
      "eval_steps_per_second": 3.209,
      "step": 2790
    },
    {
      "epoch": 3.5696,
      "grad_norm": 0.41823264956474304,
      "learning_rate": 1.1208661339649348e-05,
      "loss": 0.499,
      "step": 2791
    },
    {
      "epoch": 3.57088,
      "grad_norm": 0.4045957326889038,
      "learning_rate": 1.1196729567362863e-05,
      "loss": 0.5207,
      "step": 2792
    },
    {
      "epoch": 3.5721600000000002,
      "grad_norm": 0.3973578214645386,
      "learning_rate": 1.118480521019854e-05,
      "loss": 0.5278,
      "step": 2793
    },
    {
      "epoch": 3.5734399999999997,
      "grad_norm": 0.3847652077674866,
      "learning_rate": 1.1172888276341013e-05,
      "loss": 0.5148,
      "step": 2794
    },
    {
      "epoch": 3.57472,
      "grad_norm": 0.4217729866504669,
      "learning_rate": 1.1160978773969773e-05,
      "loss": 0.5163,
      "step": 2795
    },
    {
      "epoch": 3.576,
      "grad_norm": 0.39440542459487915,
      "learning_rate": 1.1149076711259248e-05,
      "loss": 0.5268,
      "step": 2796
    },
    {
      "epoch": 3.57728,
      "grad_norm": 0.3773360550403595,
      "learning_rate": 1.1137182096378751e-05,
      "loss": 0.5528,
      "step": 2797
    },
    {
      "epoch": 3.57856,
      "grad_norm": 0.3919049799442291,
      "learning_rate": 1.1125294937492468e-05,
      "loss": 0.5096,
      "step": 2798
    },
    {
      "epoch": 3.57984,
      "grad_norm": 0.4034450948238373,
      "learning_rate": 1.1113415242759484e-05,
      "loss": 0.5483,
      "step": 2799
    },
    {
      "epoch": 3.58112,
      "grad_norm": 0.41759708523750305,
      "learning_rate": 1.1101543020333757e-05,
      "loss": 0.5161,
      "step": 2800
    },
    {
      "epoch": 3.5824,
      "grad_norm": 0.3851338028907776,
      "learning_rate": 1.1089678278364108e-05,
      "loss": 0.5344,
      "step": 2801
    },
    {
      "epoch": 3.58368,
      "grad_norm": 0.4193148612976074,
      "learning_rate": 1.1077821024994231e-05,
      "loss": 0.5186,
      "step": 2802
    },
    {
      "epoch": 3.58496,
      "grad_norm": 0.4054766893386841,
      "learning_rate": 1.1065971268362685e-05,
      "loss": 0.4923,
      "step": 2803
    },
    {
      "epoch": 3.58624,
      "grad_norm": 0.3914055824279785,
      "learning_rate": 1.1054129016602863e-05,
      "loss": 0.524,
      "step": 2804
    },
    {
      "epoch": 3.58752,
      "grad_norm": 0.40777549147605896,
      "learning_rate": 1.1042294277843029e-05,
      "loss": 0.5376,
      "step": 2805
    },
    {
      "epoch": 3.5888,
      "grad_norm": 0.4160667955875397,
      "learning_rate": 1.1030467060206287e-05,
      "loss": 0.5286,
      "step": 2806
    },
    {
      "epoch": 3.59008,
      "grad_norm": 0.4037516415119171,
      "learning_rate": 1.1018647371810562e-05,
      "loss": 0.5361,
      "step": 2807
    },
    {
      "epoch": 3.59136,
      "grad_norm": 0.43503421545028687,
      "learning_rate": 1.100683522076863e-05,
      "loss": 0.5349,
      "step": 2808
    },
    {
      "epoch": 3.59264,
      "grad_norm": 0.41867363452911377,
      "learning_rate": 1.099503061518809e-05,
      "loss": 0.5298,
      "step": 2809
    },
    {
      "epoch": 3.59392,
      "grad_norm": 0.37841764092445374,
      "learning_rate": 1.0983233563171346e-05,
      "loss": 0.4951,
      "step": 2810
    },
    {
      "epoch": 3.5952,
      "grad_norm": 0.43736982345581055,
      "learning_rate": 1.0971444072815638e-05,
      "loss": 0.5203,
      "step": 2811
    },
    {
      "epoch": 3.59648,
      "grad_norm": 0.3952690362930298,
      "learning_rate": 1.095966215221301e-05,
      "loss": 0.5208,
      "step": 2812
    },
    {
      "epoch": 3.59776,
      "grad_norm": 0.4146837890148163,
      "learning_rate": 1.0947887809450299e-05,
      "loss": 0.5183,
      "step": 2813
    },
    {
      "epoch": 3.59904,
      "grad_norm": 0.39580023288726807,
      "learning_rate": 1.0936121052609155e-05,
      "loss": 0.5535,
      "step": 2814
    },
    {
      "epoch": 3.60032,
      "grad_norm": 0.4427098333835602,
      "learning_rate": 1.0924361889766017e-05,
      "loss": 0.5476,
      "step": 2815
    },
    {
      "epoch": 3.6016,
      "grad_norm": 0.3978303372859955,
      "learning_rate": 1.0912610328992103e-05,
      "loss": 0.5287,
      "step": 2816
    },
    {
      "epoch": 3.60288,
      "grad_norm": 0.4317154288291931,
      "learning_rate": 1.0900866378353424e-05,
      "loss": 0.542,
      "step": 2817
    },
    {
      "epoch": 3.6041600000000003,
      "grad_norm": 0.4113706648349762,
      "learning_rate": 1.0889130045910772e-05,
      "loss": 0.5302,
      "step": 2818
    },
    {
      "epoch": 3.6054399999999998,
      "grad_norm": 0.382451593875885,
      "learning_rate": 1.0877401339719688e-05,
      "loss": 0.5534,
      "step": 2819
    },
    {
      "epoch": 3.60672,
      "grad_norm": 0.35701024532318115,
      "learning_rate": 1.0865680267830507e-05,
      "loss": 0.4852,
      "step": 2820
    },
    {
      "epoch": 3.60672,
      "eval_loss": 0.69279545545578,
      "eval_runtime": 19.6048,
      "eval_samples_per_second": 25.606,
      "eval_steps_per_second": 3.214,
      "step": 2820
    },
    {
      "epoch": 3.608,
      "grad_norm": 0.38199862837791443,
      "learning_rate": 1.0853966838288296e-05,
      "loss": 0.5211,
      "step": 2821
    },
    {
      "epoch": 3.60928,
      "grad_norm": 0.4160092771053314,
      "learning_rate": 1.0842261059132894e-05,
      "loss": 0.5418,
      "step": 2822
    },
    {
      "epoch": 3.61056,
      "grad_norm": 0.41221144795417786,
      "learning_rate": 1.0830562938398893e-05,
      "loss": 0.5629,
      "step": 2823
    },
    {
      "epoch": 3.61184,
      "grad_norm": 0.38705974817276,
      "learning_rate": 1.081887248411561e-05,
      "loss": 0.5295,
      "step": 2824
    },
    {
      "epoch": 3.61312,
      "grad_norm": 0.3720438778400421,
      "learning_rate": 1.0807189704307115e-05,
      "loss": 0.5317,
      "step": 2825
    },
    {
      "epoch": 3.6144,
      "grad_norm": 0.43141815066337585,
      "learning_rate": 1.079551460699221e-05,
      "loss": 0.5444,
      "step": 2826
    },
    {
      "epoch": 3.6156800000000002,
      "grad_norm": 0.40104877948760986,
      "learning_rate": 1.0783847200184408e-05,
      "loss": 0.5744,
      "step": 2827
    },
    {
      "epoch": 3.6169599999999997,
      "grad_norm": 0.4276953339576721,
      "learning_rate": 1.077218749189196e-05,
      "loss": 0.5141,
      "step": 2828
    },
    {
      "epoch": 3.61824,
      "grad_norm": 0.41719603538513184,
      "learning_rate": 1.0760535490117836e-05,
      "loss": 0.5459,
      "step": 2829
    },
    {
      "epoch": 3.61952,
      "grad_norm": 0.393088161945343,
      "learning_rate": 1.0748891202859695e-05,
      "loss": 0.5465,
      "step": 2830
    },
    {
      "epoch": 3.6208,
      "grad_norm": 0.3777695298194885,
      "learning_rate": 1.0737254638109923e-05,
      "loss": 0.5235,
      "step": 2831
    },
    {
      "epoch": 3.62208,
      "grad_norm": 0.4007987082004547,
      "learning_rate": 1.07256258038556e-05,
      "loss": 0.5135,
      "step": 2832
    },
    {
      "epoch": 3.62336,
      "grad_norm": 0.42344895005226135,
      "learning_rate": 1.0714004708078485e-05,
      "loss": 0.5413,
      "step": 2833
    },
    {
      "epoch": 3.62464,
      "grad_norm": 0.37936583161354065,
      "learning_rate": 1.0702391358755041e-05,
      "loss": 0.5374,
      "step": 2834
    },
    {
      "epoch": 3.62592,
      "grad_norm": 0.3774566650390625,
      "learning_rate": 1.0690785763856421e-05,
      "loss": 0.524,
      "step": 2835
    },
    {
      "epoch": 3.6272,
      "grad_norm": 0.38538309931755066,
      "learning_rate": 1.0679187931348432e-05,
      "loss": 0.5351,
      "step": 2836
    },
    {
      "epoch": 3.62848,
      "grad_norm": 0.391896516084671,
      "learning_rate": 1.0667597869191567e-05,
      "loss": 0.5332,
      "step": 2837
    },
    {
      "epoch": 3.62976,
      "grad_norm": 0.3809085786342621,
      "learning_rate": 1.0656015585340995e-05,
      "loss": 0.5172,
      "step": 2838
    },
    {
      "epoch": 3.63104,
      "grad_norm": 0.39903533458709717,
      "learning_rate": 1.064444108774652e-05,
      "loss": 0.5113,
      "step": 2839
    },
    {
      "epoch": 3.63232,
      "grad_norm": 0.4094974398612976,
      "learning_rate": 1.0632874384352623e-05,
      "loss": 0.5428,
      "step": 2840
    },
    {
      "epoch": 3.6336,
      "grad_norm": 0.37527838349342346,
      "learning_rate": 1.0621315483098442e-05,
      "loss": 0.5543,
      "step": 2841
    },
    {
      "epoch": 3.63488,
      "grad_norm": 0.3820810914039612,
      "learning_rate": 1.0609764391917728e-05,
      "loss": 0.535,
      "step": 2842
    },
    {
      "epoch": 3.63616,
      "grad_norm": 0.3992846608161926,
      "learning_rate": 1.0598221118738898e-05,
      "loss": 0.5219,
      "step": 2843
    },
    {
      "epoch": 3.63744,
      "grad_norm": 0.4034469723701477,
      "learning_rate": 1.0586685671485003e-05,
      "loss": 0.5362,
      "step": 2844
    },
    {
      "epoch": 3.63872,
      "grad_norm": 0.382069855928421,
      "learning_rate": 1.0575158058073707e-05,
      "loss": 0.5402,
      "step": 2845
    },
    {
      "epoch": 3.64,
      "grad_norm": 0.36648836731910706,
      "learning_rate": 1.0563638286417305e-05,
      "loss": 0.509,
      "step": 2846
    },
    {
      "epoch": 3.64128,
      "grad_norm": 0.4011310935020447,
      "learning_rate": 1.055212636442271e-05,
      "loss": 0.5633,
      "step": 2847
    },
    {
      "epoch": 3.64256,
      "grad_norm": 0.41788607835769653,
      "learning_rate": 1.0540622299991444e-05,
      "loss": 0.5387,
      "step": 2848
    },
    {
      "epoch": 3.64384,
      "grad_norm": 0.3700264096260071,
      "learning_rate": 1.0529126101019651e-05,
      "loss": 0.5067,
      "step": 2849
    },
    {
      "epoch": 3.64512,
      "grad_norm": 0.39257314801216125,
      "learning_rate": 1.0517637775398045e-05,
      "loss": 0.5253,
      "step": 2850
    },
    {
      "epoch": 3.64512,
      "eval_loss": 0.6919096112251282,
      "eval_runtime": 19.6068,
      "eval_samples_per_second": 25.603,
      "eval_steps_per_second": 3.213,
      "step": 2850
    },
    {
      "epoch": 3.6464,
      "grad_norm": 0.4005648195743561,
      "learning_rate": 1.0506157331011965e-05,
      "loss": 0.52,
      "step": 2851
    },
    {
      "epoch": 3.6476800000000003,
      "grad_norm": 0.3984036445617676,
      "learning_rate": 1.0494684775741338e-05,
      "loss": 0.5305,
      "step": 2852
    },
    {
      "epoch": 3.6489599999999998,
      "grad_norm": 0.3855576813220978,
      "learning_rate": 1.0483220117460654e-05,
      "loss": 0.5485,
      "step": 2853
    },
    {
      "epoch": 3.65024,
      "grad_norm": 0.39826157689094543,
      "learning_rate": 1.0471763364039e-05,
      "loss": 0.5424,
      "step": 2854
    },
    {
      "epoch": 3.65152,
      "grad_norm": 0.3812386691570282,
      "learning_rate": 1.0460314523340045e-05,
      "loss": 0.5329,
      "step": 2855
    },
    {
      "epoch": 3.6528,
      "grad_norm": 0.3689391016960144,
      "learning_rate": 1.0448873603221998e-05,
      "loss": 0.5331,
      "step": 2856
    },
    {
      "epoch": 3.65408,
      "grad_norm": 0.3852054476737976,
      "learning_rate": 1.0437440611537663e-05,
      "loss": 0.5216,
      "step": 2857
    },
    {
      "epoch": 3.65536,
      "grad_norm": 0.3992381989955902,
      "learning_rate": 1.0426015556134392e-05,
      "loss": 0.5581,
      "step": 2858
    },
    {
      "epoch": 3.65664,
      "grad_norm": 0.41818341612815857,
      "learning_rate": 1.0414598444854074e-05,
      "loss": 0.5445,
      "step": 2859
    },
    {
      "epoch": 3.65792,
      "grad_norm": 0.39289405941963196,
      "learning_rate": 1.0403189285533162e-05,
      "loss": 0.553,
      "step": 2860
    },
    {
      "epoch": 3.6592000000000002,
      "grad_norm": 0.393294095993042,
      "learning_rate": 1.0391788086002655e-05,
      "loss": 0.5452,
      "step": 2861
    },
    {
      "epoch": 3.6604799999999997,
      "grad_norm": 0.3914821147918701,
      "learning_rate": 1.0380394854088064e-05,
      "loss": 0.5378,
      "step": 2862
    },
    {
      "epoch": 3.66176,
      "grad_norm": 0.40927696228027344,
      "learning_rate": 1.036900959760946e-05,
      "loss": 0.5092,
      "step": 2863
    },
    {
      "epoch": 3.66304,
      "grad_norm": 0.41068533062934875,
      "learning_rate": 1.0357632324381433e-05,
      "loss": 0.5696,
      "step": 2864
    },
    {
      "epoch": 3.66432,
      "grad_norm": 0.37779608368873596,
      "learning_rate": 1.0346263042213068e-05,
      "loss": 0.5065,
      "step": 2865
    },
    {
      "epoch": 3.6656,
      "grad_norm": 0.36545783281326294,
      "learning_rate": 1.0334901758907998e-05,
      "loss": 0.5019,
      "step": 2866
    },
    {
      "epoch": 3.66688,
      "grad_norm": 0.38986366987228394,
      "learning_rate": 1.0323548482264356e-05,
      "loss": 0.5186,
      "step": 2867
    },
    {
      "epoch": 3.66816,
      "grad_norm": 0.3853451907634735,
      "learning_rate": 1.0312203220074766e-05,
      "loss": 0.5292,
      "step": 2868
    },
    {
      "epoch": 3.66944,
      "grad_norm": 0.39557725191116333,
      "learning_rate": 1.0300865980126363e-05,
      "loss": 0.5218,
      "step": 2869
    },
    {
      "epoch": 3.67072,
      "grad_norm": 0.39430752396583557,
      "learning_rate": 1.0289536770200783e-05,
      "loss": 0.5205,
      "step": 2870
    },
    {
      "epoch": 3.672,
      "grad_norm": 0.3839975893497467,
      "learning_rate": 1.0278215598074133e-05,
      "loss": 0.5293,
      "step": 2871
    },
    {
      "epoch": 3.67328,
      "grad_norm": 0.3877027928829193,
      "learning_rate": 1.0266902471517008e-05,
      "loss": 0.5198,
      "step": 2872
    },
    {
      "epoch": 3.67456,
      "grad_norm": 0.3610535264015198,
      "learning_rate": 1.02555973982945e-05,
      "loss": 0.5054,
      "step": 2873
    },
    {
      "epoch": 3.67584,
      "grad_norm": 0.40294355154037476,
      "learning_rate": 1.0244300386166137e-05,
      "loss": 0.5072,
      "step": 2874
    },
    {
      "epoch": 3.67712,
      "grad_norm": 0.3884049355983734,
      "learning_rate": 1.0233011442885946e-05,
      "loss": 0.5479,
      "step": 2875
    },
    {
      "epoch": 3.6784,
      "grad_norm": 0.39016881585121155,
      "learning_rate": 1.0221730576202407e-05,
      "loss": 0.5246,
      "step": 2876
    },
    {
      "epoch": 3.67968,
      "grad_norm": 0.42978039383888245,
      "learning_rate": 1.0210457793858444e-05,
      "loss": 0.5499,
      "step": 2877
    },
    {
      "epoch": 3.68096,
      "grad_norm": 0.4034227430820465,
      "learning_rate": 1.019919310359144e-05,
      "loss": 0.541,
      "step": 2878
    },
    {
      "epoch": 3.68224,
      "grad_norm": 0.4031940698623657,
      "learning_rate": 1.0187936513133245e-05,
      "loss": 0.5565,
      "step": 2879
    },
    {
      "epoch": 3.68352,
      "grad_norm": 0.3943078815937042,
      "learning_rate": 1.0176688030210107e-05,
      "loss": 0.5405,
      "step": 2880
    },
    {
      "epoch": 3.68352,
      "eval_loss": 0.6921889781951904,
      "eval_runtime": 19.6151,
      "eval_samples_per_second": 25.592,
      "eval_steps_per_second": 3.212,
      "step": 2880
    },
    {
      "epoch": 3.6848,
      "grad_norm": 0.39938703179359436,
      "learning_rate": 1.016544766254274e-05,
      "loss": 0.5379,
      "step": 2881
    },
    {
      "epoch": 3.68608,
      "grad_norm": 0.39377644658088684,
      "learning_rate": 1.0154215417846292e-05,
      "loss": 0.5303,
      "step": 2882
    },
    {
      "epoch": 3.68736,
      "grad_norm": 0.3930898904800415,
      "learning_rate": 1.0142991303830301e-05,
      "loss": 0.5108,
      "step": 2883
    },
    {
      "epoch": 3.68864,
      "grad_norm": 0.3893067538738251,
      "learning_rate": 1.013177532819876e-05,
      "loss": 0.5339,
      "step": 2884
    },
    {
      "epoch": 3.68992,
      "grad_norm": 0.41059496998786926,
      "learning_rate": 1.0120567498650065e-05,
      "loss": 0.5392,
      "step": 2885
    },
    {
      "epoch": 3.6912000000000003,
      "grad_norm": 0.42212939262390137,
      "learning_rate": 1.010936782287702e-05,
      "loss": 0.5452,
      "step": 2886
    },
    {
      "epoch": 3.6924799999999998,
      "grad_norm": 0.3938245475292206,
      "learning_rate": 1.0098176308566817e-05,
      "loss": 0.522,
      "step": 2887
    },
    {
      "epoch": 3.69376,
      "grad_norm": 0.4170350432395935,
      "learning_rate": 1.0086992963401068e-05,
      "loss": 0.5389,
      "step": 2888
    },
    {
      "epoch": 3.69504,
      "grad_norm": 0.38585466146469116,
      "learning_rate": 1.0075817795055776e-05,
      "loss": 0.5186,
      "step": 2889
    },
    {
      "epoch": 3.69632,
      "grad_norm": 0.4402909278869629,
      "learning_rate": 1.0064650811201316e-05,
      "loss": 0.5472,
      "step": 2890
    },
    {
      "epoch": 3.6976,
      "grad_norm": 0.4163888394832611,
      "learning_rate": 1.0053492019502457e-05,
      "loss": 0.5201,
      "step": 2891
    },
    {
      "epoch": 3.69888,
      "grad_norm": 0.401815265417099,
      "learning_rate": 1.0042341427618357e-05,
      "loss": 0.5056,
      "step": 2892
    },
    {
      "epoch": 3.70016,
      "grad_norm": 0.39862966537475586,
      "learning_rate": 1.0031199043202506e-05,
      "loss": 0.5688,
      "step": 2893
    },
    {
      "epoch": 3.70144,
      "grad_norm": 0.37970712780952454,
      "learning_rate": 1.0020064873902803e-05,
      "loss": 0.5201,
      "step": 2894
    },
    {
      "epoch": 3.7027200000000002,
      "grad_norm": 0.3529803156852722,
      "learning_rate": 1.0008938927361495e-05,
      "loss": 0.5172,
      "step": 2895
    },
    {
      "epoch": 3.7039999999999997,
      "grad_norm": 0.3716217875480652,
      "learning_rate": 9.997821211215172e-06,
      "loss": 0.508,
      "step": 2896
    },
    {
      "epoch": 3.70528,
      "grad_norm": 0.3569638729095459,
      "learning_rate": 9.986711733094785e-06,
      "loss": 0.5477,
      "step": 2897
    },
    {
      "epoch": 3.70656,
      "grad_norm": 0.3688352406024933,
      "learning_rate": 9.975610500625642e-06,
      "loss": 0.5148,
      "step": 2898
    },
    {
      "epoch": 3.70784,
      "grad_norm": 0.39409464597702026,
      "learning_rate": 9.964517521427366e-06,
      "loss": 0.5328,
      "step": 2899
    },
    {
      "epoch": 3.70912,
      "grad_norm": 0.37420061230659485,
      "learning_rate": 9.953432803113931e-06,
      "loss": 0.5384,
      "step": 2900
    },
    {
      "epoch": 3.7104,
      "grad_norm": 0.43107661604881287,
      "learning_rate": 9.94235635329365e-06,
      "loss": 0.5522,
      "step": 2901
    },
    {
      "epoch": 3.71168,
      "grad_norm": 0.3913263976573944,
      "learning_rate": 9.931288179569135e-06,
      "loss": 0.5266,
      "step": 2902
    },
    {
      "epoch": 3.71296,
      "grad_norm": 0.38705945014953613,
      "learning_rate": 9.920228289537337e-06,
      "loss": 0.5172,
      "step": 2903
    },
    {
      "epoch": 3.71424,
      "grad_norm": 0.389431357383728,
      "learning_rate": 9.909176690789522e-06,
      "loss": 0.5367,
      "step": 2904
    },
    {
      "epoch": 3.71552,
      "grad_norm": 0.3860527276992798,
      "learning_rate": 9.89813339091125e-06,
      "loss": 0.5204,
      "step": 2905
    },
    {
      "epoch": 3.7168,
      "grad_norm": 0.4057573080062866,
      "learning_rate": 9.887098397482398e-06,
      "loss": 0.5515,
      "step": 2906
    },
    {
      "epoch": 3.71808,
      "grad_norm": 0.36535072326660156,
      "learning_rate": 9.876071718077142e-06,
      "loss": 0.5039,
      "step": 2907
    },
    {
      "epoch": 3.71936,
      "grad_norm": 0.38402247428894043,
      "learning_rate": 9.865053360263939e-06,
      "loss": 0.5348,
      "step": 2908
    },
    {
      "epoch": 3.72064,
      "grad_norm": 0.40012845396995544,
      "learning_rate": 9.854043331605544e-06,
      "loss": 0.534,
      "step": 2909
    },
    {
      "epoch": 3.72192,
      "grad_norm": 0.40505117177963257,
      "learning_rate": 9.843041639659003e-06,
      "loss": 0.5448,
      "step": 2910
    },
    {
      "epoch": 3.72192,
      "eval_loss": 0.6919758915901184,
      "eval_runtime": 19.5936,
      "eval_samples_per_second": 25.621,
      "eval_steps_per_second": 3.215,
      "step": 2910
    },
    {
      "epoch": 3.7232,
      "grad_norm": 0.3679109215736389,
      "learning_rate": 9.832048291975616e-06,
      "loss": 0.5192,
      "step": 2911
    },
    {
      "epoch": 3.72448,
      "grad_norm": 0.38582491874694824,
      "learning_rate": 9.821063296100979e-06,
      "loss": 0.5408,
      "step": 2912
    },
    {
      "epoch": 3.72576,
      "grad_norm": 0.38157039880752563,
      "learning_rate": 9.810086659574948e-06,
      "loss": 0.5151,
      "step": 2913
    },
    {
      "epoch": 3.72704,
      "grad_norm": 0.378989577293396,
      "learning_rate": 9.79911838993163e-06,
      "loss": 0.5232,
      "step": 2914
    },
    {
      "epoch": 3.72832,
      "grad_norm": 0.36829081177711487,
      "learning_rate": 9.788158494699405e-06,
      "loss": 0.5456,
      "step": 2915
    },
    {
      "epoch": 3.7296,
      "grad_norm": 0.38691553473472595,
      "learning_rate": 9.777206981400904e-06,
      "loss": 0.5455,
      "step": 2916
    },
    {
      "epoch": 3.73088,
      "grad_norm": 0.3702642619609833,
      "learning_rate": 9.76626385755299e-06,
      "loss": 0.5299,
      "step": 2917
    },
    {
      "epoch": 3.73216,
      "grad_norm": 0.38395193219184875,
      "learning_rate": 9.755329130666781e-06,
      "loss": 0.5349,
      "step": 2918
    },
    {
      "epoch": 3.73344,
      "grad_norm": 0.4038832485675812,
      "learning_rate": 9.744402808247635e-06,
      "loss": 0.5452,
      "step": 2919
    },
    {
      "epoch": 3.7347200000000003,
      "grad_norm": 0.36393991112709045,
      "learning_rate": 9.733484897795122e-06,
      "loss": 0.5349,
      "step": 2920
    },
    {
      "epoch": 3.7359999999999998,
      "grad_norm": 0.3992518484592438,
      "learning_rate": 9.722575406803056e-06,
      "loss": 0.5115,
      "step": 2921
    },
    {
      "epoch": 3.73728,
      "grad_norm": 0.3825940191745758,
      "learning_rate": 9.711674342759474e-06,
      "loss": 0.5162,
      "step": 2922
    },
    {
      "epoch": 3.73856,
      "grad_norm": 0.3796791434288025,
      "learning_rate": 9.700781713146608e-06,
      "loss": 0.5473,
      "step": 2923
    },
    {
      "epoch": 3.73984,
      "grad_norm": 0.36719459295272827,
      "learning_rate": 9.68989752544093e-06,
      "loss": 0.5164,
      "step": 2924
    },
    {
      "epoch": 3.74112,
      "grad_norm": 0.3616054356098175,
      "learning_rate": 9.679021787113083e-06,
      "loss": 0.5212,
      "step": 2925
    },
    {
      "epoch": 3.7424,
      "grad_norm": 0.40870460867881775,
      "learning_rate": 9.668154505627942e-06,
      "loss": 0.5223,
      "step": 2926
    },
    {
      "epoch": 3.74368,
      "grad_norm": 0.37021496891975403,
      "learning_rate": 9.657295688444567e-06,
      "loss": 0.5429,
      "step": 2927
    },
    {
      "epoch": 3.74496,
      "grad_norm": 0.3513473868370056,
      "learning_rate": 9.646445343016195e-06,
      "loss": 0.5188,
      "step": 2928
    },
    {
      "epoch": 3.7462400000000002,
      "grad_norm": 0.36925196647644043,
      "learning_rate": 9.635603476790268e-06,
      "loss": 0.5146,
      "step": 2929
    },
    {
      "epoch": 3.7475199999999997,
      "grad_norm": 0.40323886275291443,
      "learning_rate": 9.624770097208404e-06,
      "loss": 0.5066,
      "step": 2930
    },
    {
      "epoch": 3.7488,
      "grad_norm": 0.3805594742298126,
      "learning_rate": 9.613945211706377e-06,
      "loss": 0.5209,
      "step": 2931
    },
    {
      "epoch": 3.75008,
      "grad_norm": 0.38635683059692383,
      "learning_rate": 9.603128827714153e-06,
      "loss": 0.5502,
      "step": 2932
    },
    {
      "epoch": 3.75136,
      "grad_norm": 0.4128238558769226,
      "learning_rate": 9.592320952655863e-06,
      "loss": 0.5303,
      "step": 2933
    },
    {
      "epoch": 3.75264,
      "grad_norm": 0.3759922385215759,
      "learning_rate": 9.581521593949774e-06,
      "loss": 0.5448,
      "step": 2934
    },
    {
      "epoch": 3.75392,
      "grad_norm": 0.3930472135543823,
      "learning_rate": 9.570730759008331e-06,
      "loss": 0.5372,
      "step": 2935
    },
    {
      "epoch": 3.7552,
      "grad_norm": 0.3881407678127289,
      "learning_rate": 9.559948455238128e-06,
      "loss": 0.5615,
      "step": 2936
    },
    {
      "epoch": 3.75648,
      "grad_norm": 0.3940655291080475,
      "learning_rate": 9.549174690039884e-06,
      "loss": 0.5375,
      "step": 2937
    },
    {
      "epoch": 3.75776,
      "grad_norm": 0.3704267740249634,
      "learning_rate": 9.538409470808472e-06,
      "loss": 0.5197,
      "step": 2938
    },
    {
      "epoch": 3.75904,
      "grad_norm": 0.39302298426628113,
      "learning_rate": 9.527652804932907e-06,
      "loss": 0.5642,
      "step": 2939
    },
    {
      "epoch": 3.76032,
      "grad_norm": 0.4118668735027313,
      "learning_rate": 9.516904699796308e-06,
      "loss": 0.5114,
      "step": 2940
    },
    {
      "epoch": 3.76032,
      "eval_loss": 0.6915484070777893,
      "eval_runtime": 19.5927,
      "eval_samples_per_second": 25.622,
      "eval_steps_per_second": 3.215,
      "step": 2940
    },
    {
      "epoch": 3.7616,
      "grad_norm": 0.38111916184425354,
      "learning_rate": 9.50616516277594e-06,
      "loss": 0.5176,
      "step": 2941
    },
    {
      "epoch": 3.76288,
      "grad_norm": 0.37873461842536926,
      "learning_rate": 9.495434201243187e-06,
      "loss": 0.5488,
      "step": 2942
    },
    {
      "epoch": 3.76416,
      "grad_norm": 0.3987492620944977,
      "learning_rate": 9.484711822563527e-06,
      "loss": 0.553,
      "step": 2943
    },
    {
      "epoch": 3.76544,
      "grad_norm": 0.42262259125709534,
      "learning_rate": 9.473998034096565e-06,
      "loss": 0.5553,
      "step": 2944
    },
    {
      "epoch": 3.76672,
      "grad_norm": 0.39590519666671753,
      "learning_rate": 9.463292843196011e-06,
      "loss": 0.5451,
      "step": 2945
    },
    {
      "epoch": 3.768,
      "grad_norm": 0.3853332996368408,
      "learning_rate": 9.452596257209658e-06,
      "loss": 0.5115,
      "step": 2946
    },
    {
      "epoch": 3.76928,
      "grad_norm": 0.3893522620201111,
      "learning_rate": 9.441908283479404e-06,
      "loss": 0.5179,
      "step": 2947
    },
    {
      "epoch": 3.77056,
      "grad_norm": 0.4044470191001892,
      "learning_rate": 9.431228929341243e-06,
      "loss": 0.5324,
      "step": 2948
    },
    {
      "epoch": 3.77184,
      "grad_norm": 0.38433733582496643,
      "learning_rate": 9.420558202125232e-06,
      "loss": 0.518,
      "step": 2949
    },
    {
      "epoch": 3.77312,
      "grad_norm": 0.4074355661869049,
      "learning_rate": 9.40989610915553e-06,
      "loss": 0.5248,
      "step": 2950
    },
    {
      "epoch": 3.7744,
      "grad_norm": 0.3789268136024475,
      "learning_rate": 9.399242657750346e-06,
      "loss": 0.4921,
      "step": 2951
    },
    {
      "epoch": 3.77568,
      "grad_norm": 0.3961251378059387,
      "learning_rate": 9.388597855221976e-06,
      "loss": 0.5109,
      "step": 2952
    },
    {
      "epoch": 3.77696,
      "grad_norm": 0.41037222743034363,
      "learning_rate": 9.377961708876777e-06,
      "loss": 0.5346,
      "step": 2953
    },
    {
      "epoch": 3.7782400000000003,
      "grad_norm": 0.39118361473083496,
      "learning_rate": 9.367334226015154e-06,
      "loss": 0.5413,
      "step": 2954
    },
    {
      "epoch": 3.7795199999999998,
      "grad_norm": 0.40381744503974915,
      "learning_rate": 9.356715413931575e-06,
      "loss": 0.4914,
      "step": 2955
    },
    {
      "epoch": 3.7808,
      "grad_norm": 0.36784061789512634,
      "learning_rate": 9.34610527991456e-06,
      "loss": 0.5018,
      "step": 2956
    },
    {
      "epoch": 3.78208,
      "grad_norm": 0.35084378719329834,
      "learning_rate": 9.335503831246656e-06,
      "loss": 0.5222,
      "step": 2957
    },
    {
      "epoch": 3.78336,
      "grad_norm": 0.393728643655777,
      "learning_rate": 9.324911075204465e-06,
      "loss": 0.5238,
      "step": 2958
    },
    {
      "epoch": 3.78464,
      "grad_norm": 0.3760773539543152,
      "learning_rate": 9.314327019058621e-06,
      "loss": 0.4884,
      "step": 2959
    },
    {
      "epoch": 3.78592,
      "grad_norm": 0.3859328031539917,
      "learning_rate": 9.303751670073771e-06,
      "loss": 0.5531,
      "step": 2960
    },
    {
      "epoch": 3.7872,
      "grad_norm": 0.40965285897254944,
      "learning_rate": 9.293185035508607e-06,
      "loss": 0.5377,
      "step": 2961
    },
    {
      "epoch": 3.78848,
      "grad_norm": 0.3590520918369293,
      "learning_rate": 9.28262712261583e-06,
      "loss": 0.5162,
      "step": 2962
    },
    {
      "epoch": 3.7897600000000002,
      "grad_norm": 0.37786349654197693,
      "learning_rate": 9.272077938642147e-06,
      "loss": 0.5472,
      "step": 2963
    },
    {
      "epoch": 3.7910399999999997,
      "grad_norm": 0.38319623470306396,
      "learning_rate": 9.261537490828283e-06,
      "loss": 0.53,
      "step": 2964
    },
    {
      "epoch": 3.79232,
      "grad_norm": 0.3642309010028839,
      "learning_rate": 9.251005786408973e-06,
      "loss": 0.4983,
      "step": 2965
    },
    {
      "epoch": 3.7936,
      "grad_norm": 0.38622239232063293,
      "learning_rate": 9.240482832612931e-06,
      "loss": 0.5495,
      "step": 2966
    },
    {
      "epoch": 3.79488,
      "grad_norm": 0.36561107635498047,
      "learning_rate": 9.229968636662878e-06,
      "loss": 0.5448,
      "step": 2967
    },
    {
      "epoch": 3.79616,
      "grad_norm": 0.38697388768196106,
      "learning_rate": 9.219463205775532e-06,
      "loss": 0.4966,
      "step": 2968
    },
    {
      "epoch": 3.79744,
      "grad_norm": 0.3898349106311798,
      "learning_rate": 9.208966547161573e-06,
      "loss": 0.4984,
      "step": 2969
    },
    {
      "epoch": 3.79872,
      "grad_norm": 0.3959459364414215,
      "learning_rate": 9.198478668025675e-06,
      "loss": 0.5218,
      "step": 2970
    },
    {
      "epoch": 3.79872,
      "eval_loss": 0.692091166973114,
      "eval_runtime": 19.6136,
      "eval_samples_per_second": 25.594,
      "eval_steps_per_second": 3.212,
      "step": 2970
    },
    {
      "epoch": 3.8,
      "grad_norm": 0.38385024666786194,
      "learning_rate": 9.18799957556649e-06,
      "loss": 0.5456,
      "step": 2971
    },
    {
      "epoch": 3.80128,
      "grad_norm": 0.41451653838157654,
      "learning_rate": 9.177529276976619e-06,
      "loss": 0.5351,
      "step": 2972
    },
    {
      "epoch": 3.80256,
      "grad_norm": 0.37752336263656616,
      "learning_rate": 9.167067779442646e-06,
      "loss": 0.5241,
      "step": 2973
    },
    {
      "epoch": 3.80384,
      "grad_norm": 0.35137736797332764,
      "learning_rate": 9.156615090145116e-06,
      "loss": 0.524,
      "step": 2974
    },
    {
      "epoch": 3.80512,
      "grad_norm": 0.40610355138778687,
      "learning_rate": 9.146171216258506e-06,
      "loss": 0.5662,
      "step": 2975
    },
    {
      "epoch": 3.8064,
      "grad_norm": 0.3712407350540161,
      "learning_rate": 9.135736164951264e-06,
      "loss": 0.5243,
      "step": 2976
    },
    {
      "epoch": 3.80768,
      "grad_norm": 0.393951952457428,
      "learning_rate": 9.125309943385778e-06,
      "loss": 0.5632,
      "step": 2977
    },
    {
      "epoch": 3.80896,
      "grad_norm": 0.40164709091186523,
      "learning_rate": 9.114892558718363e-06,
      "loss": 0.5325,
      "step": 2978
    },
    {
      "epoch": 3.81024,
      "grad_norm": 0.3851127326488495,
      "learning_rate": 9.104484018099288e-06,
      "loss": 0.5453,
      "step": 2979
    },
    {
      "epoch": 3.81152,
      "grad_norm": 0.3681679964065552,
      "learning_rate": 9.094084328672732e-06,
      "loss": 0.5223,
      "step": 2980
    },
    {
      "epoch": 3.8128,
      "grad_norm": 0.38433828949928284,
      "learning_rate": 9.083693497576813e-06,
      "loss": 0.5287,
      "step": 2981
    },
    {
      "epoch": 3.81408,
      "grad_norm": 0.37378668785095215,
      "learning_rate": 9.073311531943569e-06,
      "loss": 0.5414,
      "step": 2982
    },
    {
      "epoch": 3.81536,
      "grad_norm": 0.38275086879730225,
      "learning_rate": 9.062938438898936e-06,
      "loss": 0.5196,
      "step": 2983
    },
    {
      "epoch": 3.81664,
      "grad_norm": 0.3679923415184021,
      "learning_rate": 9.052574225562778e-06,
      "loss": 0.543,
      "step": 2984
    },
    {
      "epoch": 3.81792,
      "grad_norm": 0.3831397294998169,
      "learning_rate": 9.042218899048865e-06,
      "loss": 0.5375,
      "step": 2985
    },
    {
      "epoch": 3.8192,
      "grad_norm": 0.3647843599319458,
      "learning_rate": 9.031872466464853e-06,
      "loss": 0.5144,
      "step": 2986
    },
    {
      "epoch": 3.82048,
      "grad_norm": 0.37163451313972473,
      "learning_rate": 9.0215349349123e-06,
      "loss": 0.4957,
      "step": 2987
    },
    {
      "epoch": 3.8217600000000003,
      "grad_norm": 0.36406153440475464,
      "learning_rate": 9.011206311486662e-06,
      "loss": 0.51,
      "step": 2988
    },
    {
      "epoch": 3.8230399999999998,
      "grad_norm": 0.38502752780914307,
      "learning_rate": 9.000886603277266e-06,
      "loss": 0.5621,
      "step": 2989
    },
    {
      "epoch": 3.82432,
      "grad_norm": 0.3868742287158966,
      "learning_rate": 8.990575817367331e-06,
      "loss": 0.5139,
      "step": 2990
    },
    {
      "epoch": 3.8256,
      "grad_norm": 0.3866227865219116,
      "learning_rate": 8.980273960833955e-06,
      "loss": 0.5362,
      "step": 2991
    },
    {
      "epoch": 3.82688,
      "grad_norm": 0.4030471444129944,
      "learning_rate": 8.969981040748094e-06,
      "loss": 0.5474,
      "step": 2992
    },
    {
      "epoch": 3.82816,
      "grad_norm": 0.3852604329586029,
      "learning_rate": 8.959697064174575e-06,
      "loss": 0.5269,
      "step": 2993
    },
    {
      "epoch": 3.82944,
      "grad_norm": 0.3811005651950836,
      "learning_rate": 8.9494220381721e-06,
      "loss": 0.5521,
      "step": 2994
    },
    {
      "epoch": 3.83072,
      "grad_norm": 0.38129404187202454,
      "learning_rate": 8.939155969793205e-06,
      "loss": 0.5362,
      "step": 2995
    },
    {
      "epoch": 3.832,
      "grad_norm": 0.3769567310810089,
      "learning_rate": 8.92889886608429e-06,
      "loss": 0.5292,
      "step": 2996
    },
    {
      "epoch": 3.8332800000000002,
      "grad_norm": 0.3777936100959778,
      "learning_rate": 8.918650734085608e-06,
      "loss": 0.5299,
      "step": 2997
    },
    {
      "epoch": 3.8345599999999997,
      "grad_norm": 0.4086422622203827,
      "learning_rate": 8.908411580831239e-06,
      "loss": 0.5643,
      "step": 2998
    },
    {
      "epoch": 3.83584,
      "grad_norm": 0.3899197578430176,
      "learning_rate": 8.898181413349107e-06,
      "loss": 0.5241,
      "step": 2999
    },
    {
      "epoch": 3.83712,
      "grad_norm": 0.36683395504951477,
      "learning_rate": 8.88796023866098e-06,
      "loss": 0.536,
      "step": 3000
    },
    {
      "epoch": 3.83712,
      "eval_loss": 0.6903300881385803,
      "eval_runtime": 19.631,
      "eval_samples_per_second": 25.572,
      "eval_steps_per_second": 3.209,
      "step": 3000
    },
    {
      "epoch": 3.8384,
      "grad_norm": 0.3655562698841095,
      "learning_rate": 8.877748063782429e-06,
      "loss": 0.5116,
      "step": 3001
    },
    {
      "epoch": 3.83968,
      "grad_norm": 0.4178914725780487,
      "learning_rate": 8.867544895722869e-06,
      "loss": 0.538,
      "step": 3002
    },
    {
      "epoch": 3.84096,
      "grad_norm": 0.39487767219543457,
      "learning_rate": 8.857350741485526e-06,
      "loss": 0.5527,
      "step": 3003
    },
    {
      "epoch": 3.84224,
      "grad_norm": 0.3724385201931,
      "learning_rate": 8.847165608067436e-06,
      "loss": 0.5123,
      "step": 3004
    },
    {
      "epoch": 3.84352,
      "grad_norm": 0.37117642164230347,
      "learning_rate": 8.83698950245945e-06,
      "loss": 0.5265,
      "step": 3005
    },
    {
      "epoch": 3.8448,
      "grad_norm": 0.370647132396698,
      "learning_rate": 8.826822431646208e-06,
      "loss": 0.4981,
      "step": 3006
    },
    {
      "epoch": 3.84608,
      "grad_norm": 0.38945552706718445,
      "learning_rate": 8.816664402606167e-06,
      "loss": 0.5249,
      "step": 3007
    },
    {
      "epoch": 3.84736,
      "grad_norm": 0.3807106614112854,
      "learning_rate": 8.806515422311575e-06,
      "loss": 0.5251,
      "step": 3008
    },
    {
      "epoch": 3.84864,
      "grad_norm": 0.38490813970565796,
      "learning_rate": 8.796375497728451e-06,
      "loss": 0.5208,
      "step": 3009
    },
    {
      "epoch": 3.84992,
      "grad_norm": 0.3844028413295746,
      "learning_rate": 8.78624463581662e-06,
      "loss": 0.4894,
      "step": 3010
    },
    {
      "epoch": 3.8512,
      "grad_norm": 0.36624544858932495,
      "learning_rate": 8.776122843529679e-06,
      "loss": 0.5238,
      "step": 3011
    },
    {
      "epoch": 3.85248,
      "grad_norm": 0.40320026874542236,
      "learning_rate": 8.766010127814994e-06,
      "loss": 0.5581,
      "step": 3012
    },
    {
      "epoch": 3.85376,
      "grad_norm": 0.3780871331691742,
      "learning_rate": 8.755906495613706e-06,
      "loss": 0.4852,
      "step": 3013
    },
    {
      "epoch": 3.85504,
      "grad_norm": 0.3710925877094269,
      "learning_rate": 8.74581195386073e-06,
      "loss": 0.5116,
      "step": 3014
    },
    {
      "epoch": 3.85632,
      "grad_norm": 0.3820582628250122,
      "learning_rate": 8.735726509484722e-06,
      "loss": 0.5121,
      "step": 3015
    },
    {
      "epoch": 3.8576,
      "grad_norm": 0.36904266476631165,
      "learning_rate": 8.725650169408106e-06,
      "loss": 0.5492,
      "step": 3016
    },
    {
      "epoch": 3.85888,
      "grad_norm": 0.3740033209323883,
      "learning_rate": 8.715582940547068e-06,
      "loss": 0.5422,
      "step": 3017
    },
    {
      "epoch": 3.86016,
      "grad_norm": 0.3828977048397064,
      "learning_rate": 8.705524829811508e-06,
      "loss": 0.5039,
      "step": 3018
    },
    {
      "epoch": 3.86144,
      "grad_norm": 0.39203783869743347,
      "learning_rate": 8.695475844105098e-06,
      "loss": 0.5492,
      "step": 3019
    },
    {
      "epoch": 3.86272,
      "grad_norm": 0.36637014150619507,
      "learning_rate": 8.68543599032524e-06,
      "loss": 0.5281,
      "step": 3020
    },
    {
      "epoch": 3.864,
      "grad_norm": 0.378307044506073,
      "learning_rate": 8.675405275363058e-06,
      "loss": 0.5344,
      "step": 3021
    },
    {
      "epoch": 3.8652800000000003,
      "grad_norm": 0.37208059430122375,
      "learning_rate": 8.665383706103408e-06,
      "loss": 0.5518,
      "step": 3022
    },
    {
      "epoch": 3.8665599999999998,
      "grad_norm": 0.38464826345443726,
      "learning_rate": 8.655371289424879e-06,
      "loss": 0.5368,
      "step": 3023
    },
    {
      "epoch": 3.86784,
      "grad_norm": 0.3738413155078888,
      "learning_rate": 8.645368032199757e-06,
      "loss": 0.535,
      "step": 3024
    },
    {
      "epoch": 3.86912,
      "grad_norm": 0.35537099838256836,
      "learning_rate": 8.63537394129406e-06,
      "loss": 0.5078,
      "step": 3025
    },
    {
      "epoch": 3.8704,
      "grad_norm": 0.35781610012054443,
      "learning_rate": 8.62538902356751e-06,
      "loss": 0.5181,
      "step": 3026
    },
    {
      "epoch": 3.87168,
      "grad_norm": 0.36739879846572876,
      "learning_rate": 8.615413285873523e-06,
      "loss": 0.5545,
      "step": 3027
    },
    {
      "epoch": 3.87296,
      "grad_norm": 0.3685036301612854,
      "learning_rate": 8.605446735059223e-06,
      "loss": 0.4984,
      "step": 3028
    },
    {
      "epoch": 3.87424,
      "grad_norm": 0.38391441106796265,
      "learning_rate": 8.595489377965438e-06,
      "loss": 0.517,
      "step": 3029
    },
    {
      "epoch": 3.87552,
      "grad_norm": 0.3499889373779297,
      "learning_rate": 8.585541221426657e-06,
      "loss": 0.5282,
      "step": 3030
    },
    {
      "epoch": 3.87552,
      "eval_loss": 0.6905120015144348,
      "eval_runtime": 19.6143,
      "eval_samples_per_second": 25.594,
      "eval_steps_per_second": 3.212,
      "step": 3030
    },
    {
      "epoch": 3.8768000000000002,
      "grad_norm": 0.36727920174598694,
      "learning_rate": 8.575602272271077e-06,
      "loss": 0.5478,
      "step": 3031
    },
    {
      "epoch": 3.8780799999999997,
      "grad_norm": 0.3841688632965088,
      "learning_rate": 8.565672537320581e-06,
      "loss": 0.5313,
      "step": 3032
    },
    {
      "epoch": 3.87936,
      "grad_norm": 0.3827265799045563,
      "learning_rate": 8.5557520233907e-06,
      "loss": 0.4906,
      "step": 3033
    },
    {
      "epoch": 3.88064,
      "grad_norm": 0.3875212073326111,
      "learning_rate": 8.545840737290655e-06,
      "loss": 0.5047,
      "step": 3034
    },
    {
      "epoch": 3.88192,
      "grad_norm": 0.3891419768333435,
      "learning_rate": 8.535938685823344e-06,
      "loss": 0.5526,
      "step": 3035
    },
    {
      "epoch": 3.8832,
      "grad_norm": 0.3931465446949005,
      "learning_rate": 8.526045875785293e-06,
      "loss": 0.5474,
      "step": 3036
    },
    {
      "epoch": 3.88448,
      "grad_norm": 0.37502387166023254,
      "learning_rate": 8.516162313966717e-06,
      "loss": 0.5131,
      "step": 3037
    },
    {
      "epoch": 3.88576,
      "grad_norm": 0.3653978705406189,
      "learning_rate": 8.506288007151473e-06,
      "loss": 0.5175,
      "step": 3038
    },
    {
      "epoch": 3.88704,
      "grad_norm": 0.3838379681110382,
      "learning_rate": 8.496422962117051e-06,
      "loss": 0.5444,
      "step": 3039
    },
    {
      "epoch": 3.88832,
      "grad_norm": 0.3978499472141266,
      "learning_rate": 8.486567185634606e-06,
      "loss": 0.5388,
      "step": 3040
    },
    {
      "epoch": 3.8895999999999997,
      "grad_norm": 0.4188674986362457,
      "learning_rate": 8.476720684468921e-06,
      "loss": 0.5465,
      "step": 3041
    },
    {
      "epoch": 3.89088,
      "grad_norm": 0.3850255310535431,
      "learning_rate": 8.466883465378411e-06,
      "loss": 0.4819,
      "step": 3042
    },
    {
      "epoch": 3.89216,
      "grad_norm": 0.37270432710647583,
      "learning_rate": 8.45705553511512e-06,
      "loss": 0.5157,
      "step": 3043
    },
    {
      "epoch": 3.89344,
      "grad_norm": 0.3700830340385437,
      "learning_rate": 8.44723690042472e-06,
      "loss": 0.5337,
      "step": 3044
    },
    {
      "epoch": 3.89472,
      "grad_norm": 0.3444601595401764,
      "learning_rate": 8.43742756804651e-06,
      "loss": 0.5049,
      "step": 3045
    },
    {
      "epoch": 3.896,
      "grad_norm": 0.3991895616054535,
      "learning_rate": 8.427627544713381e-06,
      "loss": 0.5271,
      "step": 3046
    },
    {
      "epoch": 3.89728,
      "grad_norm": 0.3815551698207855,
      "learning_rate": 8.417836837151857e-06,
      "loss": 0.5187,
      "step": 3047
    },
    {
      "epoch": 3.89856,
      "grad_norm": 0.35359781980514526,
      "learning_rate": 8.408055452082064e-06,
      "loss": 0.5002,
      "step": 3048
    },
    {
      "epoch": 3.89984,
      "grad_norm": 0.36077311635017395,
      "learning_rate": 8.398283396217712e-06,
      "loss": 0.5259,
      "step": 3049
    },
    {
      "epoch": 3.90112,
      "grad_norm": 0.39165806770324707,
      "learning_rate": 8.388520676266132e-06,
      "loss": 0.5621,
      "step": 3050
    },
    {
      "epoch": 3.9024,
      "grad_norm": 0.36504966020584106,
      "learning_rate": 8.378767298928237e-06,
      "loss": 0.5159,
      "step": 3051
    },
    {
      "epoch": 3.90368,
      "grad_norm": 0.3953617811203003,
      "learning_rate": 8.369023270898515e-06,
      "loss": 0.5614,
      "step": 3052
    },
    {
      "epoch": 3.90496,
      "grad_norm": 0.4084630608558655,
      "learning_rate": 8.359288598865052e-06,
      "loss": 0.5113,
      "step": 3053
    },
    {
      "epoch": 3.90624,
      "grad_norm": 0.36661767959594727,
      "learning_rate": 8.349563289509516e-06,
      "loss": 0.5068,
      "step": 3054
    },
    {
      "epoch": 3.90752,
      "grad_norm": 0.3621689975261688,
      "learning_rate": 8.339847349507132e-06,
      "loss": 0.5396,
      "step": 3055
    },
    {
      "epoch": 3.9088000000000003,
      "grad_norm": 0.3819715976715088,
      "learning_rate": 8.330140785526706e-06,
      "loss": 0.5256,
      "step": 3056
    },
    {
      "epoch": 3.91008,
      "grad_norm": 0.3973776400089264,
      "learning_rate": 8.32044360423061e-06,
      "loss": 0.5236,
      "step": 3057
    },
    {
      "epoch": 3.91136,
      "grad_norm": 0.37809520959854126,
      "learning_rate": 8.310755812274764e-06,
      "loss": 0.5267,
      "step": 3058
    },
    {
      "epoch": 3.91264,
      "grad_norm": 0.4008995592594147,
      "learning_rate": 8.301077416308651e-06,
      "loss": 0.5399,
      "step": 3059
    },
    {
      "epoch": 3.91392,
      "grad_norm": 0.38021284341812134,
      "learning_rate": 8.291408422975315e-06,
      "loss": 0.5252,
      "step": 3060
    },
    {
      "epoch": 3.91392,
      "eval_loss": 0.6899545788764954,
      "eval_runtime": 19.6046,
      "eval_samples_per_second": 25.606,
      "eval_steps_per_second": 3.214,
      "step": 3060
    },
    {
      "epoch": 3.9152,
      "grad_norm": 0.39704933762550354,
      "learning_rate": 8.281748838911323e-06,
      "loss": 0.5364,
      "step": 3061
    },
    {
      "epoch": 3.91648,
      "grad_norm": 0.38096025586128235,
      "learning_rate": 8.272098670746801e-06,
      "loss": 0.5503,
      "step": 3062
    },
    {
      "epoch": 3.91776,
      "grad_norm": 0.3867654800415039,
      "learning_rate": 8.262457925105414e-06,
      "loss": 0.4922,
      "step": 3063
    },
    {
      "epoch": 3.91904,
      "grad_norm": 0.3768038749694824,
      "learning_rate": 8.252826608604343e-06,
      "loss": 0.576,
      "step": 3064
    },
    {
      "epoch": 3.9203200000000002,
      "grad_norm": 0.4050436317920685,
      "learning_rate": 8.24320472785431e-06,
      "loss": 0.5363,
      "step": 3065
    },
    {
      "epoch": 3.9215999999999998,
      "grad_norm": 0.37319400906562805,
      "learning_rate": 8.23359228945957e-06,
      "loss": 0.5278,
      "step": 3066
    },
    {
      "epoch": 3.92288,
      "grad_norm": 0.4066287875175476,
      "learning_rate": 8.223989300017867e-06,
      "loss": 0.5352,
      "step": 3067
    },
    {
      "epoch": 3.92416,
      "grad_norm": 0.39050936698913574,
      "learning_rate": 8.214395766120484e-06,
      "loss": 0.5442,
      "step": 3068
    },
    {
      "epoch": 3.92544,
      "grad_norm": 0.3640260100364685,
      "learning_rate": 8.204811694352216e-06,
      "loss": 0.5199,
      "step": 3069
    },
    {
      "epoch": 3.92672,
      "grad_norm": 0.34316781163215637,
      "learning_rate": 8.19523709129134e-06,
      "loss": 0.5165,
      "step": 3070
    },
    {
      "epoch": 3.928,
      "grad_norm": 0.41491860151290894,
      "learning_rate": 8.185671963509655e-06,
      "loss": 0.5533,
      "step": 3071
    },
    {
      "epoch": 3.92928,
      "grad_norm": 0.3743742108345032,
      "learning_rate": 8.176116317572457e-06,
      "loss": 0.5596,
      "step": 3072
    },
    {
      "epoch": 3.93056,
      "grad_norm": 0.37523549795150757,
      "learning_rate": 8.166570160038514e-06,
      "loss": 0.5173,
      "step": 3073
    },
    {
      "epoch": 3.9318400000000002,
      "grad_norm": 0.3915380537509918,
      "learning_rate": 8.1570334974601e-06,
      "loss": 0.5357,
      "step": 3074
    },
    {
      "epoch": 3.9331199999999997,
      "grad_norm": 0.3859361708164215,
      "learning_rate": 8.147506336382973e-06,
      "loss": 0.5363,
      "step": 3075
    },
    {
      "epoch": 3.9344,
      "grad_norm": 0.3671061098575592,
      "learning_rate": 8.137988683346347e-06,
      "loss": 0.5199,
      "step": 3076
    },
    {
      "epoch": 3.93568,
      "grad_norm": 0.39495792984962463,
      "learning_rate": 8.128480544882934e-06,
      "loss": 0.5512,
      "step": 3077
    },
    {
      "epoch": 3.93696,
      "grad_norm": 0.38403016328811646,
      "learning_rate": 8.118981927518915e-06,
      "loss": 0.5358,
      "step": 3078
    },
    {
      "epoch": 3.93824,
      "grad_norm": 0.3983258903026581,
      "learning_rate": 8.109492837773912e-06,
      "loss": 0.5015,
      "step": 3079
    },
    {
      "epoch": 3.93952,
      "grad_norm": 0.3887477517127991,
      "learning_rate": 8.10001328216103e-06,
      "loss": 0.5595,
      "step": 3080
    },
    {
      "epoch": 3.9408,
      "grad_norm": 0.366597980260849,
      "learning_rate": 8.090543267186829e-06,
      "loss": 0.5125,
      "step": 3081
    },
    {
      "epoch": 3.94208,
      "grad_norm": 0.3792959153652191,
      "learning_rate": 8.081082799351301e-06,
      "loss": 0.5539,
      "step": 3082
    },
    {
      "epoch": 3.94336,
      "grad_norm": 0.4004511833190918,
      "learning_rate": 8.071631885147914e-06,
      "loss": 0.5613,
      "step": 3083
    },
    {
      "epoch": 3.94464,
      "grad_norm": 0.37375709414482117,
      "learning_rate": 8.062190531063549e-06,
      "loss": 0.5404,
      "step": 3084
    },
    {
      "epoch": 3.94592,
      "grad_norm": 0.3716925084590912,
      "learning_rate": 8.052758743578547e-06,
      "loss": 0.567,
      "step": 3085
    },
    {
      "epoch": 3.9472,
      "grad_norm": 0.3965913653373718,
      "learning_rate": 8.043336529166681e-06,
      "loss": 0.5375,
      "step": 3086
    },
    {
      "epoch": 3.94848,
      "grad_norm": 0.35229620337486267,
      "learning_rate": 8.033923894295132e-06,
      "loss": 0.5238,
      "step": 3087
    },
    {
      "epoch": 3.94976,
      "grad_norm": 0.37307727336883545,
      "learning_rate": 8.024520845424532e-06,
      "loss": 0.5498,
      "step": 3088
    },
    {
      "epoch": 3.95104,
      "grad_norm": 0.4185270071029663,
      "learning_rate": 8.015127389008923e-06,
      "loss": 0.5422,
      "step": 3089
    },
    {
      "epoch": 3.9523200000000003,
      "grad_norm": 0.34924566745758057,
      "learning_rate": 8.005743531495756e-06,
      "loss": 0.5016,
      "step": 3090
    },
    {
      "epoch": 3.9523200000000003,
      "eval_loss": 0.6905309557914734,
      "eval_runtime": 19.595,
      "eval_samples_per_second": 25.619,
      "eval_steps_per_second": 3.215,
      "step": 3090
    },
    {
      "epoch": 3.9536,
      "grad_norm": 0.34284985065460205,
      "learning_rate": 7.996369279325901e-06,
      "loss": 0.5113,
      "step": 3091
    },
    {
      "epoch": 3.95488,
      "grad_norm": 0.35125061869621277,
      "learning_rate": 7.987004638933646e-06,
      "loss": 0.5075,
      "step": 3092
    },
    {
      "epoch": 3.95616,
      "grad_norm": 0.3729866147041321,
      "learning_rate": 7.977649616746653e-06,
      "loss": 0.5127,
      "step": 3093
    },
    {
      "epoch": 3.95744,
      "grad_norm": 0.3505498468875885,
      "learning_rate": 7.968304219186006e-06,
      "loss": 0.5287,
      "step": 3094
    },
    {
      "epoch": 3.95872,
      "grad_norm": 0.3715629577636719,
      "learning_rate": 7.958968452666184e-06,
      "loss": 0.555,
      "step": 3095
    },
    {
      "epoch": 3.96,
      "grad_norm": 0.3551843464374542,
      "learning_rate": 7.949642323595035e-06,
      "loss": 0.5357,
      "step": 3096
    },
    {
      "epoch": 3.96128,
      "grad_norm": 0.37480103969573975,
      "learning_rate": 7.940325838373808e-06,
      "loss": 0.509,
      "step": 3097
    },
    {
      "epoch": 3.96256,
      "grad_norm": 0.3963179588317871,
      "learning_rate": 7.931019003397135e-06,
      "loss": 0.5065,
      "step": 3098
    },
    {
      "epoch": 3.9638400000000003,
      "grad_norm": 0.3952457904815674,
      "learning_rate": 7.921721825053013e-06,
      "loss": 0.5091,
      "step": 3099
    },
    {
      "epoch": 3.9651199999999998,
      "grad_norm": 0.4067932963371277,
      "learning_rate": 7.912434309722815e-06,
      "loss": 0.567,
      "step": 3100
    },
    {
      "epoch": 3.9664,
      "grad_norm": 0.38736796379089355,
      "learning_rate": 7.90315646378129e-06,
      "loss": 0.5403,
      "step": 3101
    },
    {
      "epoch": 3.96768,
      "grad_norm": 0.3655763268470764,
      "learning_rate": 7.893888293596534e-06,
      "loss": 0.5045,
      "step": 3102
    },
    {
      "epoch": 3.96896,
      "grad_norm": 0.3784528076648712,
      "learning_rate": 7.884629805530017e-06,
      "loss": 0.5357,
      "step": 3103
    },
    {
      "epoch": 3.97024,
      "grad_norm": 0.37929052114486694,
      "learning_rate": 7.87538100593656e-06,
      "loss": 0.534,
      "step": 3104
    },
    {
      "epoch": 3.97152,
      "grad_norm": 0.35935595631599426,
      "learning_rate": 7.866141901164324e-06,
      "loss": 0.5289,
      "step": 3105
    },
    {
      "epoch": 3.9728,
      "grad_norm": 0.37929806113243103,
      "learning_rate": 7.856912497554821e-06,
      "loss": 0.5392,
      "step": 3106
    },
    {
      "epoch": 3.97408,
      "grad_norm": 0.38722866773605347,
      "learning_rate": 7.847692801442921e-06,
      "loss": 0.5275,
      "step": 3107
    },
    {
      "epoch": 3.9753600000000002,
      "grad_norm": 0.3578529357910156,
      "learning_rate": 7.8384828191568e-06,
      "loss": 0.5403,
      "step": 3108
    },
    {
      "epoch": 3.9766399999999997,
      "grad_norm": 0.3992893099784851,
      "learning_rate": 7.829282557017994e-06,
      "loss": 0.5287,
      "step": 3109
    },
    {
      "epoch": 3.97792,
      "grad_norm": 0.36259108781814575,
      "learning_rate": 7.82009202134135e-06,
      "loss": 0.5325,
      "step": 3110
    },
    {
      "epoch": 3.9792,
      "grad_norm": 0.3608449101448059,
      "learning_rate": 7.810911218435044e-06,
      "loss": 0.5439,
      "step": 3111
    },
    {
      "epoch": 3.98048,
      "grad_norm": 0.35753127932548523,
      "learning_rate": 7.801740154600584e-06,
      "loss": 0.5205,
      "step": 3112
    },
    {
      "epoch": 3.98176,
      "grad_norm": 0.37365490198135376,
      "learning_rate": 7.792578836132767e-06,
      "loss": 0.5176,
      "step": 3113
    },
    {
      "epoch": 3.98304,
      "grad_norm": 0.38964807987213135,
      "learning_rate": 7.783427269319726e-06,
      "loss": 0.5681,
      "step": 3114
    },
    {
      "epoch": 3.98432,
      "grad_norm": 0.36225229501724243,
      "learning_rate": 7.77428546044289e-06,
      "loss": 0.5024,
      "step": 3115
    },
    {
      "epoch": 3.9856,
      "grad_norm": 0.3946057856082916,
      "learning_rate": 7.765153415776987e-06,
      "loss": 0.5311,
      "step": 3116
    },
    {
      "epoch": 3.98688,
      "grad_norm": 0.4129025936126709,
      "learning_rate": 7.756031141590051e-06,
      "loss": 0.5284,
      "step": 3117
    },
    {
      "epoch": 3.98816,
      "grad_norm": 0.40833184123039246,
      "learning_rate": 7.74691864414341e-06,
      "loss": 0.5454,
      "step": 3118
    },
    {
      "epoch": 3.98944,
      "grad_norm": 0.3792262077331543,
      "learning_rate": 7.737815929691665e-06,
      "loss": 0.496,
      "step": 3119
    },
    {
      "epoch": 3.99072,
      "grad_norm": 0.3732023239135742,
      "learning_rate": 7.728723004482726e-06,
      "loss": 0.5104,
      "step": 3120
    },
    {
      "epoch": 3.99072,
      "eval_loss": 0.6902841329574585,
      "eval_runtime": 19.6067,
      "eval_samples_per_second": 25.604,
      "eval_steps_per_second": 3.213,
      "step": 3120
    },
    {
      "epoch": 3.992,
      "grad_norm": 0.3809022307395935,
      "learning_rate": 7.71963987475777e-06,
      "loss": 0.5332,
      "step": 3121
    },
    {
      "epoch": 3.99328,
      "grad_norm": 0.3501242697238922,
      "learning_rate": 7.71056654675125e-06,
      "loss": 0.5258,
      "step": 3122
    },
    {
      "epoch": 3.99456,
      "grad_norm": 0.3587786555290222,
      "learning_rate": 7.701503026690898e-06,
      "loss": 0.5336,
      "step": 3123
    },
    {
      "epoch": 3.99584,
      "grad_norm": 0.41848596930503845,
      "learning_rate": 7.692449320797715e-06,
      "loss": 0.5513,
      "step": 3124
    },
    {
      "epoch": 3.99712,
      "grad_norm": 0.3665623664855957,
      "learning_rate": 7.68340543528595e-06,
      "loss": 0.5359,
      "step": 3125
    },
    {
      "epoch": 3.9984,
      "grad_norm": 0.3694246709346771,
      "learning_rate": 7.674371376363128e-06,
      "loss": 0.547,
      "step": 3126
    },
    {
      "epoch": 3.99968,
      "grad_norm": 0.3580736219882965,
      "learning_rate": 7.665347150230029e-06,
      "loss": 0.5132,
      "step": 3127
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6840554475784302,
      "learning_rate": 7.656332763080672e-06,
      "loss": 0.5805,
      "step": 3128
    }
  ],
  "logging_steps": 1,
  "max_steps": 3910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 1.4752511078986416e+19,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}