{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 50,
  "global_step": 708,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0423728813559322,
      "grad_norm": 1.7643470764160156,
      "learning_rate": 4.936440677966102e-05,
      "loss": 1.7478,
      "step": 10
    },
    {
      "epoch": 0.0847457627118644,
      "grad_norm": 1.352023959159851,
      "learning_rate": 4.8658192090395484e-05,
      "loss": 1.2556,
      "step": 20
    },
    {
      "epoch": 0.1271186440677966,
      "grad_norm": 1.205060601234436,
      "learning_rate": 4.795197740112994e-05,
      "loss": 1.1919,
      "step": 30
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 1.019239902496338,
      "learning_rate": 4.724576271186441e-05,
      "loss": 1.061,
      "step": 40
    },
    {
      "epoch": 0.211864406779661,
      "grad_norm": 1.0925650596618652,
      "learning_rate": 4.6539548022598875e-05,
      "loss": 1.1092,
      "step": 50
    },
    {
      "epoch": 0.211864406779661,
      "eval_loss": 1.0438729524612427,
      "eval_runtime": 22.5506,
      "eval_samples_per_second": 95.385,
      "eval_steps_per_second": 1.508,
      "step": 50
    },
    {
      "epoch": 0.2542372881355932,
      "grad_norm": 0.9862654209136963,
      "learning_rate": 4.5833333333333334e-05,
      "loss": 1.0876,
      "step": 60
    },
    {
      "epoch": 0.2966101694915254,
      "grad_norm": 0.9759798645973206,
      "learning_rate": 4.51271186440678e-05,
      "loss": 1.0237,
      "step": 70
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 1.0690054893493652,
      "learning_rate": 4.442090395480226e-05,
      "loss": 1.0092,
      "step": 80
    },
    {
      "epoch": 0.3813559322033898,
      "grad_norm": 1.1084680557250977,
      "learning_rate": 4.3714689265536725e-05,
      "loss": 1.0229,
      "step": 90
    },
    {
      "epoch": 0.423728813559322,
      "grad_norm": 1.081882357597351,
      "learning_rate": 4.300847457627119e-05,
      "loss": 1.0233,
      "step": 100
    },
    {
      "epoch": 0.423728813559322,
      "eval_loss": 1.003989815711975,
      "eval_runtime": 22.3839,
      "eval_samples_per_second": 96.096,
      "eval_steps_per_second": 1.519,
      "step": 100
    },
    {
      "epoch": 0.4661016949152542,
      "grad_norm": 1.1931794881820679,
      "learning_rate": 4.230225988700565e-05,
      "loss": 0.9807,
      "step": 110
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 1.1798979043960571,
      "learning_rate": 4.1596045197740115e-05,
      "loss": 1.037,
      "step": 120
    },
    {
      "epoch": 0.5508474576271186,
      "grad_norm": 1.236763596534729,
      "learning_rate": 4.088983050847458e-05,
      "loss": 1.0088,
      "step": 130
    },
    {
      "epoch": 0.5932203389830508,
      "grad_norm": 1.031708836555481,
      "learning_rate": 4.018361581920904e-05,
      "loss": 0.9605,
      "step": 140
    },
    {
      "epoch": 0.635593220338983,
      "grad_norm": 1.0881413221359253,
      "learning_rate": 3.9477401129943506e-05,
      "loss": 0.9669,
      "step": 150
    },
    {
      "epoch": 0.635593220338983,
      "eval_loss": 0.9852223992347717,
      "eval_runtime": 22.645,
      "eval_samples_per_second": 94.988,
      "eval_steps_per_second": 1.501,
      "step": 150
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 1.0248509645462036,
      "learning_rate": 3.877118644067797e-05,
      "loss": 0.9584,
      "step": 160
    },
    {
      "epoch": 0.7203389830508474,
      "grad_norm": 1.1003658771514893,
      "learning_rate": 3.806497175141243e-05,
      "loss": 1.0191,
      "step": 170
    },
    {
      "epoch": 0.7627118644067796,
      "grad_norm": 0.9632079005241394,
      "learning_rate": 3.735875706214689e-05,
      "loss": 1.0048,
      "step": 180
    },
    {
      "epoch": 0.8050847457627118,
      "grad_norm": 1.2166016101837158,
      "learning_rate": 3.665254237288136e-05,
      "loss": 1.0035,
      "step": 190
    },
    {
      "epoch": 0.847457627118644,
      "grad_norm": 1.132070779800415,
      "learning_rate": 3.594632768361582e-05,
      "loss": 0.9939,
      "step": 200
    },
    {
      "epoch": 0.847457627118644,
      "eval_loss": 0.9768873453140259,
      "eval_runtime": 23.2377,
      "eval_samples_per_second": 92.565,
      "eval_steps_per_second": 1.463,
      "step": 200
    },
    {
      "epoch": 0.8898305084745762,
      "grad_norm": 1.0186951160430908,
      "learning_rate": 3.524011299435028e-05,
      "loss": 0.9749,
      "step": 210
    },
    {
      "epoch": 0.9322033898305084,
      "grad_norm": 1.2069514989852905,
      "learning_rate": 3.4533898305084746e-05,
      "loss": 0.9756,
      "step": 220
    },
    {
      "epoch": 0.9745762711864406,
      "grad_norm": 1.1548621654510498,
      "learning_rate": 3.382768361581921e-05,
      "loss": 0.9723,
      "step": 230
    },
    {
      "epoch": 1.0169491525423728,
      "grad_norm": 1.0338218212127686,
      "learning_rate": 3.312146892655367e-05,
      "loss": 0.9304,
      "step": 240
    },
    {
      "epoch": 1.0593220338983051,
      "grad_norm": 1.1850943565368652,
      "learning_rate": 3.241525423728814e-05,
      "loss": 0.9512,
      "step": 250
    },
    {
      "epoch": 1.0593220338983051,
      "eval_loss": 0.9682268500328064,
      "eval_runtime": 22.9025,
      "eval_samples_per_second": 93.92,
      "eval_steps_per_second": 1.485,
      "step": 250
    },
    {
      "epoch": 1.1016949152542372,
      "grad_norm": 1.0937676429748535,
      "learning_rate": 3.17090395480226e-05,
      "loss": 0.9241,
      "step": 260
    },
    {
      "epoch": 1.1440677966101696,
      "grad_norm": 1.296743631362915,
      "learning_rate": 3.100282485875706e-05,
      "loss": 0.9508,
      "step": 270
    },
    {
      "epoch": 1.1864406779661016,
      "grad_norm": 1.145988941192627,
      "learning_rate": 3.0296610169491528e-05,
      "loss": 0.9629,
      "step": 280
    },
    {
      "epoch": 1.228813559322034,
      "grad_norm": 1.2783371210098267,
      "learning_rate": 2.959039548022599e-05,
      "loss": 0.9621,
      "step": 290
    },
    {
      "epoch": 1.271186440677966,
      "grad_norm": 1.3129595518112183,
      "learning_rate": 2.8884180790960453e-05,
      "loss": 0.9667,
      "step": 300
    },
    {
      "epoch": 1.271186440677966,
      "eval_loss": 0.9666466116905212,
      "eval_runtime": 23.5102,
      "eval_samples_per_second": 91.492,
      "eval_steps_per_second": 1.446,
      "step": 300
    },
    {
      "epoch": 1.3135593220338984,
      "grad_norm": 1.2464239597320557,
      "learning_rate": 2.817796610169492e-05,
      "loss": 0.9359,
      "step": 310
    },
    {
      "epoch": 1.3559322033898304,
      "grad_norm": 1.3320660591125488,
      "learning_rate": 2.747175141242938e-05,
      "loss": 0.9232,
      "step": 320
    },
    {
      "epoch": 1.3983050847457628,
      "grad_norm": 1.3213363885879517,
      "learning_rate": 2.6765536723163843e-05,
      "loss": 0.9064,
      "step": 330
    },
    {
      "epoch": 1.4406779661016949,
      "grad_norm": 1.4874022006988525,
      "learning_rate": 2.605932203389831e-05,
      "loss": 0.9367,
      "step": 340
    },
    {
      "epoch": 1.4830508474576272,
      "grad_norm": 1.308205246925354,
      "learning_rate": 2.535310734463277e-05,
      "loss": 0.9058,
      "step": 350
    },
    {
      "epoch": 1.4830508474576272,
      "eval_loss": 0.964046061038971,
      "eval_runtime": 22.9381,
      "eval_samples_per_second": 93.774,
      "eval_steps_per_second": 1.482,
      "step": 350
    },
    {
      "epoch": 1.5254237288135593,
      "grad_norm": 1.3957818746566772,
      "learning_rate": 2.464689265536723e-05,
      "loss": 0.9232,
      "step": 360
    },
    {
      "epoch": 1.5677966101694916,
      "grad_norm": 1.3040454387664795,
      "learning_rate": 2.3940677966101697e-05,
      "loss": 0.9376,
      "step": 370
    },
    {
      "epoch": 1.6101694915254239,
      "grad_norm": 1.4122378826141357,
      "learning_rate": 2.323446327683616e-05,
      "loss": 0.9115,
      "step": 380
    },
    {
      "epoch": 1.652542372881356,
      "grad_norm": 1.3442506790161133,
      "learning_rate": 2.252824858757062e-05,
      "loss": 0.9599,
      "step": 390
    },
    {
      "epoch": 1.694915254237288,
      "grad_norm": 1.5409705638885498,
      "learning_rate": 2.1822033898305087e-05,
      "loss": 0.9258,
      "step": 400
    },
    {
      "epoch": 1.694915254237288,
      "eval_loss": 0.9589577913284302,
      "eval_runtime": 22.6608,
      "eval_samples_per_second": 94.922,
      "eval_steps_per_second": 1.5,
      "step": 400
    },
    {
      "epoch": 1.7372881355932204,
      "grad_norm": 1.315023422241211,
      "learning_rate": 2.111581920903955e-05,
      "loss": 0.9088,
      "step": 410
    },
    {
      "epoch": 1.7796610169491527,
      "grad_norm": 1.4532520771026611,
      "learning_rate": 2.0409604519774012e-05,
      "loss": 0.9351,
      "step": 420
    },
    {
      "epoch": 1.8220338983050848,
      "grad_norm": 1.3708842992782593,
      "learning_rate": 1.9703389830508475e-05,
      "loss": 0.95,
      "step": 430
    },
    {
      "epoch": 1.8644067796610169,
      "grad_norm": 1.3897417783737183,
      "learning_rate": 1.899717514124294e-05,
      "loss": 0.8943,
      "step": 440
    },
    {
      "epoch": 1.9067796610169492,
      "grad_norm": 1.2795928716659546,
      "learning_rate": 1.8290960451977403e-05,
      "loss": 0.8856,
      "step": 450
    },
    {
      "epoch": 1.9067796610169492,
      "eval_loss": 0.9577222466468811,
      "eval_runtime": 23.1528,
      "eval_samples_per_second": 92.905,
      "eval_steps_per_second": 1.469,
      "step": 450
    },
    {
      "epoch": 1.9491525423728815,
      "grad_norm": 1.2920231819152832,
      "learning_rate": 1.7584745762711865e-05,
      "loss": 0.9415,
      "step": 460
    },
    {
      "epoch": 1.9915254237288136,
      "grad_norm": 1.3519489765167236,
      "learning_rate": 1.687853107344633e-05,
      "loss": 0.9708,
      "step": 470
    },
    {
      "epoch": 2.0338983050847457,
      "grad_norm": 1.3836076259613037,
      "learning_rate": 1.617231638418079e-05,
      "loss": 0.8801,
      "step": 480
    },
    {
      "epoch": 2.0762711864406778,
      "grad_norm": 1.478256344795227,
      "learning_rate": 1.5466101694915256e-05,
      "loss": 0.8908,
      "step": 490
    },
    {
      "epoch": 2.1186440677966103,
      "grad_norm": 1.571032166481018,
      "learning_rate": 1.475988700564972e-05,
      "loss": 0.9013,
      "step": 500
    },
    {
      "epoch": 2.1186440677966103,
      "eval_loss": 0.9608597755432129,
      "eval_runtime": 22.5872,
      "eval_samples_per_second": 95.231,
      "eval_steps_per_second": 1.505,
      "step": 500
    },
    {
      "epoch": 2.1610169491525424,
      "grad_norm": 1.466437578201294,
      "learning_rate": 1.4053672316384181e-05,
      "loss": 0.8926,
      "step": 510
    },
    {
      "epoch": 2.2033898305084745,
      "grad_norm": 1.6073771715164185,
      "learning_rate": 1.3347457627118645e-05,
      "loss": 0.8717,
      "step": 520
    },
    {
      "epoch": 2.2457627118644066,
      "grad_norm": 1.5890705585479736,
      "learning_rate": 1.2641242937853107e-05,
      "loss": 0.8633,
      "step": 530
    },
    {
      "epoch": 2.288135593220339,
      "grad_norm": 1.6749075651168823,
      "learning_rate": 1.1935028248587572e-05,
      "loss": 0.8941,
      "step": 540
    },
    {
      "epoch": 2.330508474576271,
      "grad_norm": 1.6727330684661865,
      "learning_rate": 1.1228813559322036e-05,
      "loss": 0.8929,
      "step": 550
    },
    {
      "epoch": 2.330508474576271,
      "eval_loss": 0.9595548510551453,
      "eval_runtime": 22.9206,
      "eval_samples_per_second": 93.846,
      "eval_steps_per_second": 1.483,
      "step": 550
    },
    {
      "epoch": 2.3728813559322033,
      "grad_norm": 1.5609760284423828,
      "learning_rate": 1.0522598870056498e-05,
      "loss": 0.8298,
      "step": 560
    },
    {
      "epoch": 2.415254237288136,
      "grad_norm": 1.7489114999771118,
      "learning_rate": 9.81638418079096e-06,
      "loss": 0.8618,
      "step": 570
    },
    {
      "epoch": 2.457627118644068,
      "grad_norm": 1.7662324905395508,
      "learning_rate": 9.110169491525423e-06,
      "loss": 0.8932,
      "step": 580
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.9018930196762085,
      "learning_rate": 8.403954802259887e-06,
      "loss": 0.8743,
      "step": 590
    },
    {
      "epoch": 2.542372881355932,
      "grad_norm": 1.6308236122131348,
      "learning_rate": 7.697740112994351e-06,
      "loss": 0.8752,
      "step": 600
    },
    {
      "epoch": 2.542372881355932,
      "eval_loss": 0.9617831707000732,
      "eval_runtime": 23.2374,
      "eval_samples_per_second": 92.566,
      "eval_steps_per_second": 1.463,
      "step": 600
    },
    {
      "epoch": 2.584745762711864,
      "grad_norm": 1.5900993347167969,
      "learning_rate": 6.991525423728814e-06,
      "loss": 0.8653,
      "step": 610
    },
    {
      "epoch": 2.6271186440677967,
      "grad_norm": 1.8414711952209473,
      "learning_rate": 6.285310734463278e-06,
      "loss": 0.8877,
      "step": 620
    },
    {
      "epoch": 2.669491525423729,
      "grad_norm": 1.7117115259170532,
      "learning_rate": 5.57909604519774e-06,
      "loss": 0.8691,
      "step": 630
    },
    {
      "epoch": 2.711864406779661,
      "grad_norm": 1.690845012664795,
      "learning_rate": 4.872881355932204e-06,
      "loss": 0.8728,
      "step": 640
    },
    {
      "epoch": 2.7542372881355934,
      "grad_norm": 1.727298378944397,
      "learning_rate": 4.166666666666667e-06,
      "loss": 0.9005,
      "step": 650
    },
    {
      "epoch": 2.7542372881355934,
      "eval_loss": 0.959761381149292,
      "eval_runtime": 22.8543,
      "eval_samples_per_second": 94.118,
      "eval_steps_per_second": 1.488,
      "step": 650
    },
    {
      "epoch": 2.7966101694915255,
      "grad_norm": 1.6035046577453613,
      "learning_rate": 3.46045197740113e-06,
      "loss": 0.9181,
      "step": 660
    },
    {
      "epoch": 2.8389830508474576,
      "grad_norm": 1.7235994338989258,
      "learning_rate": 2.7542372881355934e-06,
      "loss": 0.9003,
      "step": 670
    },
    {
      "epoch": 2.8813559322033897,
      "grad_norm": 1.7914451360702515,
      "learning_rate": 2.0480225988700563e-06,
      "loss": 0.951,
      "step": 680
    },
    {
      "epoch": 2.923728813559322,
      "grad_norm": 1.8638156652450562,
      "learning_rate": 1.3418079096045198e-06,
      "loss": 0.8739,
      "step": 690
    },
    {
      "epoch": 2.9661016949152543,
      "grad_norm": 1.6826245784759521,
      "learning_rate": 6.355932203389831e-07,
      "loss": 0.8375,
      "step": 700
    },
    {
      "epoch": 2.9661016949152543,
      "eval_loss": 0.9599277973175049,
      "eval_runtime": 22.8971,
      "eval_samples_per_second": 93.942,
      "eval_steps_per_second": 1.485,
      "step": 700
    }
  ],
  "logging_steps": 10,
  "max_steps": 708,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 50,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8604011452104704e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}