{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 10,
  "global_step": 471,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.06369426751592357,
      "grad_norm": 5.655877590179443,
      "learning_rate": 4.893842887473461e-05,
      "loss": 0.9888,
      "step": 10
    },
    {
      "epoch": 0.06369426751592357,
      "eval_loss": 0.6204623579978943,
      "eval_runtime": 260.4945,
      "eval_samples_per_second": 1.332,
      "eval_steps_per_second": 0.334,
      "step": 10
    },
    {
      "epoch": 0.12738853503184713,
      "grad_norm": 5.2134904861450195,
      "learning_rate": 4.787685774946922e-05,
      "loss": 0.6134,
      "step": 20
    },
    {
      "epoch": 0.12738853503184713,
      "eval_loss": 0.5212348699569702,
      "eval_runtime": 261.046,
      "eval_samples_per_second": 1.329,
      "eval_steps_per_second": 0.333,
      "step": 20
    },
    {
      "epoch": 0.1910828025477707,
      "grad_norm": 6.658833026885986,
      "learning_rate": 4.681528662420383e-05,
      "loss": 0.5704,
      "step": 30
    },
    {
      "epoch": 0.1910828025477707,
      "eval_loss": 0.5708295106887817,
      "eval_runtime": 266.4556,
      "eval_samples_per_second": 1.302,
      "eval_steps_per_second": 0.327,
      "step": 30
    },
    {
      "epoch": 0.25477707006369427,
      "grad_norm": 4.93190336227417,
      "learning_rate": 4.575371549893843e-05,
      "loss": 0.5118,
      "step": 40
    },
    {
      "epoch": 0.25477707006369427,
      "eval_loss": 0.45317262411117554,
      "eval_runtime": 263.8344,
      "eval_samples_per_second": 1.315,
      "eval_steps_per_second": 0.33,
      "step": 40
    },
    {
      "epoch": 0.3184713375796178,
      "grad_norm": 6.612313747406006,
      "learning_rate": 4.469214437367304e-05,
      "loss": 0.5768,
      "step": 50
    },
    {
      "epoch": 0.3184713375796178,
      "eval_loss": 0.5027761459350586,
      "eval_runtime": 317.3646,
      "eval_samples_per_second": 1.093,
      "eval_steps_per_second": 0.274,
      "step": 50
    },
    {
      "epoch": 0.3821656050955414,
      "grad_norm": 6.43605375289917,
      "learning_rate": 4.3630573248407646e-05,
      "loss": 0.4945,
      "step": 60
    },
    {
      "epoch": 0.3821656050955414,
      "eval_loss": 0.45588889718055725,
      "eval_runtime": 269.9949,
      "eval_samples_per_second": 1.285,
      "eval_steps_per_second": 0.322,
      "step": 60
    },
    {
      "epoch": 0.445859872611465,
      "grad_norm": 3.744384527206421,
      "learning_rate": 4.256900212314226e-05,
      "loss": 0.4173,
      "step": 70
    },
    {
      "epoch": 0.445859872611465,
      "eval_loss": 0.44179391860961914,
      "eval_runtime": 263.2415,
      "eval_samples_per_second": 1.318,
      "eval_steps_per_second": 0.33,
      "step": 70
    },
    {
      "epoch": 0.5095541401273885,
      "grad_norm": 4.705707550048828,
      "learning_rate": 4.150743099787686e-05,
      "loss": 0.4277,
      "step": 80
    },
    {
      "epoch": 0.5095541401273885,
      "eval_loss": 0.37640380859375,
      "eval_runtime": 264.4226,
      "eval_samples_per_second": 1.312,
      "eval_steps_per_second": 0.329,
      "step": 80
    },
    {
      "epoch": 0.5732484076433121,
      "grad_norm": 2.7133395671844482,
      "learning_rate": 4.044585987261147e-05,
      "loss": 0.4374,
      "step": 90
    },
    {
      "epoch": 0.5732484076433121,
      "eval_loss": 0.35762831568717957,
      "eval_runtime": 264.9493,
      "eval_samples_per_second": 1.31,
      "eval_steps_per_second": 0.328,
      "step": 90
    },
    {
      "epoch": 0.6369426751592356,
      "grad_norm": 4.4099650382995605,
      "learning_rate": 3.9384288747346076e-05,
      "loss": 0.3717,
      "step": 100
    },
    {
      "epoch": 0.6369426751592356,
      "eval_loss": 0.35183703899383545,
      "eval_runtime": 278.8216,
      "eval_samples_per_second": 1.245,
      "eval_steps_per_second": 0.312,
      "step": 100
    },
    {
      "epoch": 0.7006369426751592,
      "grad_norm": 15.860236167907715,
      "learning_rate": 3.8322717622080686e-05,
      "loss": 0.4415,
      "step": 110
    },
    {
      "epoch": 0.7006369426751592,
      "eval_loss": 0.6824913620948792,
      "eval_runtime": 281.0438,
      "eval_samples_per_second": 1.235,
      "eval_steps_per_second": 0.31,
      "step": 110
    },
    {
      "epoch": 0.7643312101910829,
      "grad_norm": 5.011375427246094,
      "learning_rate": 3.7261146496815283e-05,
      "loss": 0.3639,
      "step": 120
    },
    {
      "epoch": 0.7643312101910829,
      "eval_loss": 0.3454379737377167,
      "eval_runtime": 281.28,
      "eval_samples_per_second": 1.234,
      "eval_steps_per_second": 0.309,
      "step": 120
    },
    {
      "epoch": 0.8280254777070064,
      "grad_norm": 3.8196499347686768,
      "learning_rate": 3.6199575371549894e-05,
      "loss": 0.3604,
      "step": 130
    },
    {
      "epoch": 0.8280254777070064,
      "eval_loss": 0.33031439781188965,
      "eval_runtime": 266.1981,
      "eval_samples_per_second": 1.304,
      "eval_steps_per_second": 0.327,
      "step": 130
    },
    {
      "epoch": 0.89171974522293,
      "grad_norm": 8.475601196289062,
      "learning_rate": 3.51380042462845e-05,
      "loss": 0.3782,
      "step": 140
    },
    {
      "epoch": 0.89171974522293,
      "eval_loss": 0.33289575576782227,
      "eval_runtime": 265.8273,
      "eval_samples_per_second": 1.305,
      "eval_steps_per_second": 0.327,
      "step": 140
    },
    {
      "epoch": 0.9554140127388535,
      "grad_norm": 4.658049583435059,
      "learning_rate": 3.407643312101911e-05,
      "loss": 0.36,
      "step": 150
    },
    {
      "epoch": 0.9554140127388535,
      "eval_loss": 0.33607012033462524,
      "eval_runtime": 277.0995,
      "eval_samples_per_second": 1.252,
      "eval_steps_per_second": 0.314,
      "step": 150
    },
    {
      "epoch": 1.019108280254777,
      "grad_norm": 2.7650649547576904,
      "learning_rate": 3.301486199575371e-05,
      "loss": 0.3244,
      "step": 160
    },
    {
      "epoch": 1.019108280254777,
      "eval_loss": 0.3019406497478485,
      "eval_runtime": 281.1357,
      "eval_samples_per_second": 1.234,
      "eval_steps_per_second": 0.309,
      "step": 160
    },
    {
      "epoch": 1.0828025477707006,
      "grad_norm": 7.399433612823486,
      "learning_rate": 3.1953290870488323e-05,
      "loss": 0.3337,
      "step": 170
    },
    {
      "epoch": 1.0828025477707006,
      "eval_loss": 0.33017274737358093,
      "eval_runtime": 279.943,
      "eval_samples_per_second": 1.24,
      "eval_steps_per_second": 0.311,
      "step": 170
    },
    {
      "epoch": 1.1464968152866242,
      "grad_norm": 6.1573567390441895,
      "learning_rate": 3.089171974522293e-05,
      "loss": 0.2967,
      "step": 180
    },
    {
      "epoch": 1.1464968152866242,
      "eval_loss": 0.286271870136261,
      "eval_runtime": 279.1484,
      "eval_samples_per_second": 1.243,
      "eval_steps_per_second": 0.312,
      "step": 180
    },
    {
      "epoch": 1.2101910828025477,
      "grad_norm": 5.677030086517334,
      "learning_rate": 2.9830148619957538e-05,
      "loss": 0.3236,
      "step": 190
    },
    {
      "epoch": 1.2101910828025477,
      "eval_loss": 0.3013075590133667,
      "eval_runtime": 280.1463,
      "eval_samples_per_second": 1.239,
      "eval_steps_per_second": 0.311,
      "step": 190
    },
    {
      "epoch": 1.2738853503184713,
      "grad_norm": 3.6576144695281982,
      "learning_rate": 2.8768577494692145e-05,
      "loss": 0.2848,
      "step": 200
    },
    {
      "epoch": 1.2738853503184713,
      "eval_loss": 0.28381872177124023,
      "eval_runtime": 278.4374,
      "eval_samples_per_second": 1.246,
      "eval_steps_per_second": 0.312,
      "step": 200
    },
    {
      "epoch": 1.3375796178343948,
      "grad_norm": 3.5609078407287598,
      "learning_rate": 2.7707006369426753e-05,
      "loss": 0.2138,
      "step": 210
    },
    {
      "epoch": 1.3375796178343948,
      "eval_loss": 0.27076515555381775,
      "eval_runtime": 269.6079,
      "eval_samples_per_second": 1.287,
      "eval_steps_per_second": 0.323,
      "step": 210
    },
    {
      "epoch": 1.4012738853503186,
      "grad_norm": 5.47592306137085,
      "learning_rate": 2.664543524416136e-05,
      "loss": 0.2646,
      "step": 220
    },
    {
      "epoch": 1.4012738853503186,
      "eval_loss": 0.2966860830783844,
      "eval_runtime": 267.6504,
      "eval_samples_per_second": 1.296,
      "eval_steps_per_second": 0.325,
      "step": 220
    },
    {
      "epoch": 1.4649681528662422,
      "grad_norm": 5.770505428314209,
      "learning_rate": 2.5583864118895967e-05,
      "loss": 0.2784,
      "step": 230
    },
    {
      "epoch": 1.4649681528662422,
      "eval_loss": 0.25654563307762146,
      "eval_runtime": 270.188,
      "eval_samples_per_second": 1.284,
      "eval_steps_per_second": 0.322,
      "step": 230
    },
    {
      "epoch": 1.5286624203821657,
      "grad_norm": 9.245145797729492,
      "learning_rate": 2.4522292993630575e-05,
      "loss": 0.3004,
      "step": 240
    },
    {
      "epoch": 1.5286624203821657,
      "eval_loss": 0.2588840425014496,
      "eval_runtime": 270.9119,
      "eval_samples_per_second": 1.281,
      "eval_steps_per_second": 0.321,
      "step": 240
    },
    {
      "epoch": 1.5923566878980893,
      "grad_norm": 6.825544834136963,
      "learning_rate": 2.3460721868365182e-05,
      "loss": 0.2585,
      "step": 250
    },
    {
      "epoch": 1.5923566878980893,
      "eval_loss": 0.2777319848537445,
      "eval_runtime": 271.4967,
      "eval_samples_per_second": 1.278,
      "eval_steps_per_second": 0.32,
      "step": 250
    },
    {
      "epoch": 1.6560509554140128,
      "grad_norm": 5.504091262817383,
      "learning_rate": 2.239915074309979e-05,
      "loss": 0.239,
      "step": 260
    },
    {
      "epoch": 1.6560509554140128,
      "eval_loss": 0.2845991551876068,
      "eval_runtime": 275.4719,
      "eval_samples_per_second": 1.26,
      "eval_steps_per_second": 0.316,
      "step": 260
    },
    {
      "epoch": 1.7197452229299364,
      "grad_norm": 1.7976171970367432,
      "learning_rate": 2.1337579617834397e-05,
      "loss": 0.324,
      "step": 270
    },
    {
      "epoch": 1.7197452229299364,
      "eval_loss": 0.2418244183063507,
      "eval_runtime": 273.3976,
      "eval_samples_per_second": 1.269,
      "eval_steps_per_second": 0.318,
      "step": 270
    },
    {
      "epoch": 1.78343949044586,
      "grad_norm": 7.518586158752441,
      "learning_rate": 2.0276008492569004e-05,
      "loss": 0.2661,
      "step": 280
    },
    {
      "epoch": 1.78343949044586,
      "eval_loss": 0.2677413821220398,
      "eval_runtime": 273.2159,
      "eval_samples_per_second": 1.27,
      "eval_steps_per_second": 0.318,
      "step": 280
    },
    {
      "epoch": 1.8471337579617835,
      "grad_norm": 3.743008613586426,
      "learning_rate": 1.921443736730361e-05,
      "loss": 0.2763,
      "step": 290
    },
    {
      "epoch": 1.8471337579617835,
      "eval_loss": 0.2228710651397705,
      "eval_runtime": 272.9055,
      "eval_samples_per_second": 1.272,
      "eval_steps_per_second": 0.319,
      "step": 290
    },
    {
      "epoch": 1.910828025477707,
      "grad_norm": 7.198996543884277,
      "learning_rate": 1.8152866242038215e-05,
      "loss": 0.1957,
      "step": 300
    },
    {
      "epoch": 1.910828025477707,
      "eval_loss": 0.25310221314430237,
      "eval_runtime": 272.8452,
      "eval_samples_per_second": 1.272,
      "eval_steps_per_second": 0.319,
      "step": 300
    },
    {
      "epoch": 1.9745222929936306,
      "grad_norm": 7.609084606170654,
      "learning_rate": 1.7091295116772823e-05,
      "loss": 0.251,
      "step": 310
    },
    {
      "epoch": 1.9745222929936306,
      "eval_loss": 0.250274658203125,
      "eval_runtime": 274.2439,
      "eval_samples_per_second": 1.265,
      "eval_steps_per_second": 0.317,
      "step": 310
    },
    {
      "epoch": 2.038216560509554,
      "grad_norm": 5.631246566772461,
      "learning_rate": 1.602972399150743e-05,
      "loss": 0.2124,
      "step": 320
    },
    {
      "epoch": 2.038216560509554,
      "eval_loss": 0.23672077059745789,
      "eval_runtime": 273.0475,
      "eval_samples_per_second": 1.271,
      "eval_steps_per_second": 0.319,
      "step": 320
    },
    {
      "epoch": 2.1019108280254777,
      "grad_norm": 8.399256706237793,
      "learning_rate": 1.4968152866242039e-05,
      "loss": 0.1832,
      "step": 330
    },
    {
      "epoch": 2.1019108280254777,
      "eval_loss": 0.2170046865940094,
      "eval_runtime": 274.911,
      "eval_samples_per_second": 1.262,
      "eval_steps_per_second": 0.316,
      "step": 330
    },
    {
      "epoch": 2.1656050955414012,
      "grad_norm": 6.787008285522461,
      "learning_rate": 1.3906581740976646e-05,
      "loss": 0.1968,
      "step": 340
    },
    {
      "epoch": 2.1656050955414012,
      "eval_loss": 0.25750404596328735,
      "eval_runtime": 274.5796,
      "eval_samples_per_second": 1.264,
      "eval_steps_per_second": 0.317,
      "step": 340
    },
    {
      "epoch": 2.229299363057325,
      "grad_norm": 4.561034202575684,
      "learning_rate": 1.2845010615711253e-05,
      "loss": 0.2169,
      "step": 350
    },
    {
      "epoch": 2.229299363057325,
      "eval_loss": 0.21437929570674896,
      "eval_runtime": 274.1716,
      "eval_samples_per_second": 1.266,
      "eval_steps_per_second": 0.317,
      "step": 350
    },
    {
      "epoch": 2.2929936305732483,
      "grad_norm": 2.4580085277557373,
      "learning_rate": 1.178343949044586e-05,
      "loss": 0.1645,
      "step": 360
    },
    {
      "epoch": 2.2929936305732483,
      "eval_loss": 0.21549557149410248,
      "eval_runtime": 274.4038,
      "eval_samples_per_second": 1.265,
      "eval_steps_per_second": 0.317,
      "step": 360
    },
    {
      "epoch": 2.356687898089172,
      "grad_norm": 5.2749786376953125,
      "learning_rate": 1.0721868365180468e-05,
      "loss": 0.1292,
      "step": 370
    },
    {
      "epoch": 2.356687898089172,
      "eval_loss": 0.24459926784038544,
      "eval_runtime": 275.4779,
      "eval_samples_per_second": 1.26,
      "eval_steps_per_second": 0.316,
      "step": 370
    },
    {
      "epoch": 2.4203821656050954,
      "grad_norm": 11.909132957458496,
      "learning_rate": 9.660297239915075e-06,
      "loss": 0.2064,
      "step": 380
    },
    {
      "epoch": 2.4203821656050954,
      "eval_loss": 0.2202015370130539,
      "eval_runtime": 276.3076,
      "eval_samples_per_second": 1.256,
      "eval_steps_per_second": 0.315,
      "step": 380
    },
    {
      "epoch": 2.484076433121019,
      "grad_norm": 8.857806205749512,
      "learning_rate": 8.598726114649681e-06,
      "loss": 0.2967,
      "step": 390
    },
    {
      "epoch": 2.484076433121019,
      "eval_loss": 0.2075081467628479,
      "eval_runtime": 275.5237,
      "eval_samples_per_second": 1.259,
      "eval_steps_per_second": 0.316,
      "step": 390
    },
    {
      "epoch": 2.5477707006369426,
      "grad_norm": 8.795026779174805,
      "learning_rate": 7.537154989384289e-06,
      "loss": 0.2349,
      "step": 400
    },
    {
      "epoch": 2.5477707006369426,
      "eval_loss": 0.2141169160604477,
      "eval_runtime": 274.8429,
      "eval_samples_per_second": 1.263,
      "eval_steps_per_second": 0.317,
      "step": 400
    },
    {
      "epoch": 2.611464968152866,
      "grad_norm": 4.121735572814941,
      "learning_rate": 6.4755838641188965e-06,
      "loss": 0.161,
      "step": 410
    },
    {
      "epoch": 2.611464968152866,
      "eval_loss": 0.2091439962387085,
      "eval_runtime": 275.865,
      "eval_samples_per_second": 1.258,
      "eval_steps_per_second": 0.315,
      "step": 410
    },
    {
      "epoch": 2.6751592356687897,
      "grad_norm": 7.995586395263672,
      "learning_rate": 5.414012738853504e-06,
      "loss": 0.1786,
      "step": 420
    },
    {
      "epoch": 2.6751592356687897,
      "eval_loss": 0.21283170580863953,
      "eval_runtime": 274.9723,
      "eval_samples_per_second": 1.262,
      "eval_steps_per_second": 0.316,
      "step": 420
    },
    {
      "epoch": 2.738853503184713,
      "grad_norm": 6.569371700286865,
      "learning_rate": 4.35244161358811e-06,
      "loss": 0.2665,
      "step": 430
    },
    {
      "epoch": 2.738853503184713,
      "eval_loss": 0.19985607266426086,
      "eval_runtime": 275.6861,
      "eval_samples_per_second": 1.259,
      "eval_steps_per_second": 0.316,
      "step": 430
    },
    {
      "epoch": 2.802547770700637,
      "grad_norm": 2.53202223777771,
      "learning_rate": 3.2908704883227177e-06,
      "loss": 0.1377,
      "step": 440
    },
    {
      "epoch": 2.802547770700637,
      "eval_loss": 0.2025289386510849,
      "eval_runtime": 275.6519,
      "eval_samples_per_second": 1.259,
      "eval_steps_per_second": 0.316,
      "step": 440
    },
    {
      "epoch": 2.8662420382165603,
      "grad_norm": 2.821432590484619,
      "learning_rate": 2.229299363057325e-06,
      "loss": 0.2082,
      "step": 450
    },
    {
      "epoch": 2.8662420382165603,
      "eval_loss": 0.20033259689807892,
      "eval_runtime": 276.9112,
      "eval_samples_per_second": 1.253,
      "eval_steps_per_second": 0.314,
      "step": 450
    },
    {
      "epoch": 2.9299363057324843,
      "grad_norm": 3.187981605529785,
      "learning_rate": 1.167728237791932e-06,
      "loss": 0.1408,
      "step": 460
    },
    {
      "epoch": 2.9299363057324843,
      "eval_loss": 0.19696064293384552,
      "eval_runtime": 276.0427,
      "eval_samples_per_second": 1.257,
      "eval_steps_per_second": 0.315,
      "step": 460
    },
    {
      "epoch": 2.9936305732484074,
      "grad_norm": 3.8218815326690674,
      "learning_rate": 1.0615711252653928e-07,
      "loss": 0.1506,
      "step": 470
    },
    {
      "epoch": 2.9936305732484074,
      "eval_loss": 0.19708757102489471,
      "eval_runtime": 274.6555,
      "eval_samples_per_second": 1.263,
      "eval_steps_per_second": 0.317,
      "step": 470
    }
  ],
  "logging_steps": 10,
  "max_steps": 471,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 495041961535488.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}