{
  "best_metric": 5.4548899332682295,
  "best_model_checkpoint": "./results/checkpoint-5496",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 6412,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1091703056768559,
      "grad_norm": 50.38945388793945,
      "learning_rate": 9.863537117903931e-05,
      "loss": 46.0228,
      "step": 100
    },
    {
      "epoch": 0.2183406113537118,
      "grad_norm": 53.34562683105469,
      "learning_rate": 9.727074235807861e-05,
      "loss": 16.1163,
      "step": 200
    },
    {
      "epoch": 0.32751091703056767,
      "grad_norm": 47.549678802490234,
      "learning_rate": 9.59061135371179e-05,
      "loss": 10.3137,
      "step": 300
    },
    {
      "epoch": 0.4366812227074236,
      "grad_norm": 27.560810089111328,
      "learning_rate": 9.454148471615721e-05,
      "loss": 8.4484,
      "step": 400
    },
    {
      "epoch": 0.5458515283842795,
      "grad_norm": 43.655635833740234,
      "learning_rate": 9.317685589519652e-05,
      "loss": 7.5983,
      "step": 500
    },
    {
      "epoch": 0.6550218340611353,
      "grad_norm": 36.29502868652344,
      "learning_rate": 9.18122270742358e-05,
      "loss": 7.5018,
      "step": 600
    },
    {
      "epoch": 0.7641921397379913,
      "grad_norm": 46.47073745727539,
      "learning_rate": 9.044759825327511e-05,
      "loss": 7.1886,
      "step": 700
    },
    {
      "epoch": 0.8733624454148472,
      "grad_norm": 40.817569732666016,
      "learning_rate": 8.908296943231441e-05,
      "loss": 7.136,
      "step": 800
    },
    {
      "epoch": 0.982532751091703,
      "grad_norm": 40.99913024902344,
      "learning_rate": 8.771834061135371e-05,
      "loss": 6.9784,
      "step": 900
    },
    {
      "epoch": 1.0,
      "eval_avg_mae": 7.428911844889323,
      "eval_loss": 7.4289116859436035,
      "eval_mae_lex": 7.114953517913818,
      "eval_mae_sem": 5.620931625366211,
      "eval_mae_syn": 9.550849914550781,
      "eval_runtime": 27.041,
      "eval_samples_per_second": 270.959,
      "eval_steps_per_second": 8.469,
      "step": 916
    },
    {
      "epoch": 1.091703056768559,
      "grad_norm": 40.483001708984375,
      "learning_rate": 8.635371179039302e-05,
      "loss": 6.7681,
      "step": 1000
    },
    {
      "epoch": 1.2008733624454149,
      "grad_norm": 28.894540786743164,
      "learning_rate": 8.498908296943232e-05,
      "loss": 6.5716,
      "step": 1100
    },
    {
      "epoch": 1.3100436681222707,
      "grad_norm": 28.377840042114258,
      "learning_rate": 8.362445414847162e-05,
      "loss": 6.796,
      "step": 1200
    },
    {
      "epoch": 1.4192139737991267,
      "grad_norm": 39.30733108520508,
      "learning_rate": 8.225982532751092e-05,
      "loss": 6.6517,
      "step": 1300
    },
    {
      "epoch": 1.5283842794759825,
      "grad_norm": 29.63848304748535,
      "learning_rate": 8.089519650655023e-05,
      "loss": 6.7767,
      "step": 1400
    },
    {
      "epoch": 1.6375545851528384,
      "grad_norm": 52.78211975097656,
      "learning_rate": 7.953056768558951e-05,
      "loss": 6.6008,
      "step": 1500
    },
    {
      "epoch": 1.7467248908296944,
      "grad_norm": 52.6399040222168,
      "learning_rate": 7.816593886462883e-05,
      "loss": 6.5132,
      "step": 1600
    },
    {
      "epoch": 1.8558951965065502,
      "grad_norm": 25.399370193481445,
      "learning_rate": 7.680131004366813e-05,
      "loss": 6.5244,
      "step": 1700
    },
    {
      "epoch": 1.965065502183406,
      "grad_norm": 40.24644470214844,
      "learning_rate": 7.543668122270742e-05,
      "loss": 6.4174,
      "step": 1800
    },
    {
      "epoch": 2.0,
      "eval_avg_mae": 6.124305725097656,
      "eval_loss": 6.124305725097656,
      "eval_mae_lex": 5.557982921600342,
      "eval_mae_sem": 4.011363983154297,
      "eval_mae_syn": 8.803570747375488,
      "eval_runtime": 26.9945,
      "eval_samples_per_second": 271.426,
      "eval_steps_per_second": 8.483,
      "step": 1832
    },
    {
      "epoch": 2.074235807860262,
      "grad_norm": 72.3259048461914,
      "learning_rate": 7.407205240174672e-05,
      "loss": 6.362,
      "step": 1900
    },
    {
      "epoch": 2.183406113537118,
      "grad_norm": 44.780540466308594,
      "learning_rate": 7.270742358078603e-05,
      "loss": 6.2052,
      "step": 2000
    },
    {
      "epoch": 2.2925764192139737,
      "grad_norm": 42.22085952758789,
      "learning_rate": 7.134279475982533e-05,
      "loss": 6.0878,
      "step": 2100
    },
    {
      "epoch": 2.4017467248908297,
      "grad_norm": 31.405487060546875,
      "learning_rate": 6.997816593886463e-05,
      "loss": 6.1432,
      "step": 2200
    },
    {
      "epoch": 2.5109170305676853,
      "grad_norm": 40.08876037597656,
      "learning_rate": 6.861353711790393e-05,
      "loss": 6.0203,
      "step": 2300
    },
    {
      "epoch": 2.6200873362445414,
      "grad_norm": 25.91780662536621,
      "learning_rate": 6.724890829694324e-05,
      "loss": 6.05,
      "step": 2400
    },
    {
      "epoch": 2.7292576419213974,
      "grad_norm": 30.461589813232422,
      "learning_rate": 6.588427947598254e-05,
      "loss": 6.0259,
      "step": 2500
    },
    {
      "epoch": 2.8384279475982535,
      "grad_norm": 35.90027618408203,
      "learning_rate": 6.451965065502183e-05,
      "loss": 6.1552,
      "step": 2600
    },
    {
      "epoch": 2.947598253275109,
      "grad_norm": 26.988929748535156,
      "learning_rate": 6.315502183406113e-05,
      "loss": 5.8932,
      "step": 2700
    },
    {
      "epoch": 3.0,
      "eval_avg_mae": 5.737476348876953,
      "eval_loss": 5.737475872039795,
      "eval_mae_lex": 5.095954895019531,
      "eval_mae_sem": 3.588836669921875,
      "eval_mae_syn": 8.527636528015137,
      "eval_runtime": 27.0753,
      "eval_samples_per_second": 270.616,
      "eval_steps_per_second": 8.458,
      "step": 2748
    },
    {
      "epoch": 3.056768558951965,
      "grad_norm": 31.56240463256836,
      "learning_rate": 6.179039301310045e-05,
      "loss": 5.7562,
      "step": 2800
    },
    {
      "epoch": 3.165938864628821,
      "grad_norm": 30.02605628967285,
      "learning_rate": 6.042576419213974e-05,
      "loss": 5.7653,
      "step": 2900
    },
    {
      "epoch": 3.2751091703056767,
      "grad_norm": 37.670711517333984,
      "learning_rate": 5.9061135371179045e-05,
      "loss": 5.8259,
      "step": 3000
    },
    {
      "epoch": 3.3842794759825328,
      "grad_norm": 29.38449478149414,
      "learning_rate": 5.769650655021834e-05,
      "loss": 5.5527,
      "step": 3100
    },
    {
      "epoch": 3.493449781659389,
      "grad_norm": 41.96727752685547,
      "learning_rate": 5.633187772925764e-05,
      "loss": 5.7041,
      "step": 3200
    },
    {
      "epoch": 3.6026200873362444,
      "grad_norm": 40.971092224121094,
      "learning_rate": 5.4967248908296945e-05,
      "loss": 5.8564,
      "step": 3300
    },
    {
      "epoch": 3.7117903930131004,
      "grad_norm": 28.331939697265625,
      "learning_rate": 5.360262008733624e-05,
      "loss": 5.6168,
      "step": 3400
    },
    {
      "epoch": 3.8209606986899565,
      "grad_norm": 26.781084060668945,
      "learning_rate": 5.223799126637555e-05,
      "loss": 5.6631,
      "step": 3500
    },
    {
      "epoch": 3.930131004366812,
      "grad_norm": 32.894012451171875,
      "learning_rate": 5.087336244541485e-05,
      "loss": 5.6886,
      "step": 3600
    },
    {
      "epoch": 4.0,
      "eval_avg_mae": 6.235666910807292,
      "eval_loss": 6.2356672286987305,
      "eval_mae_lex": 5.677294731140137,
      "eval_mae_sem": 4.446841716766357,
      "eval_mae_syn": 8.582864761352539,
      "eval_runtime": 27.0543,
      "eval_samples_per_second": 270.826,
      "eval_steps_per_second": 8.464,
      "step": 3664
    },
    {
      "epoch": 4.039301310043668,
      "grad_norm": 35.94241714477539,
      "learning_rate": 4.950873362445415e-05,
      "loss": 5.6333,
      "step": 3700
    },
    {
      "epoch": 4.148471615720524,
      "grad_norm": 25.45247459411621,
      "learning_rate": 4.814410480349345e-05,
      "loss": 5.3055,
      "step": 3800
    },
    {
      "epoch": 4.25764192139738,
      "grad_norm": 30.626169204711914,
      "learning_rate": 4.6779475982532754e-05,
      "loss": 5.4211,
      "step": 3900
    },
    {
      "epoch": 4.366812227074236,
      "grad_norm": 25.812488555908203,
      "learning_rate": 4.5414847161572056e-05,
      "loss": 5.328,
      "step": 4000
    },
    {
      "epoch": 4.475982532751091,
      "grad_norm": 33.51716232299805,
      "learning_rate": 4.405021834061135e-05,
      "loss": 5.4752,
      "step": 4100
    },
    {
      "epoch": 4.585152838427947,
      "grad_norm": 32.362186431884766,
      "learning_rate": 4.268558951965066e-05,
      "loss": 5.3976,
      "step": 4200
    },
    {
      "epoch": 4.6943231441048034,
      "grad_norm": 32.369476318359375,
      "learning_rate": 4.1320960698689957e-05,
      "loss": 5.1836,
      "step": 4300
    },
    {
      "epoch": 4.8034934497816595,
      "grad_norm": 35.22241973876953,
      "learning_rate": 3.995633187772926e-05,
      "loss": 5.3904,
      "step": 4400
    },
    {
      "epoch": 4.9126637554585155,
      "grad_norm": 34.892250061035156,
      "learning_rate": 3.859170305676856e-05,
      "loss": 5.2038,
      "step": 4500
    },
    {
      "epoch": 5.0,
      "eval_avg_mae": 6.149158477783203,
      "eval_loss": 6.149158954620361,
      "eval_mae_lex": 5.426638126373291,
      "eval_mae_sem": 4.464701175689697,
      "eval_mae_syn": 8.556136131286621,
      "eval_runtime": 26.9461,
      "eval_samples_per_second": 271.913,
      "eval_steps_per_second": 8.498,
      "step": 4580
    },
    {
      "epoch": 5.021834061135372,
      "grad_norm": 30.598581314086914,
      "learning_rate": 3.7227074235807864e-05,
      "loss": 5.3014,
      "step": 4600
    },
    {
      "epoch": 5.131004366812227,
      "grad_norm": 25.072534561157227,
      "learning_rate": 3.586244541484716e-05,
      "loss": 4.9525,
      "step": 4700
    },
    {
      "epoch": 5.240174672489083,
      "grad_norm": 34.00102233886719,
      "learning_rate": 3.449781659388647e-05,
      "loss": 4.9499,
      "step": 4800
    },
    {
      "epoch": 5.349344978165939,
      "grad_norm": 30.76056480407715,
      "learning_rate": 3.3133187772925765e-05,
      "loss": 4.9841,
      "step": 4900
    },
    {
      "epoch": 5.458515283842795,
      "grad_norm": 28.28791618347168,
      "learning_rate": 3.176855895196507e-05,
      "loss": 5.1702,
      "step": 5000
    },
    {
      "epoch": 5.567685589519651,
      "grad_norm": 32.07628631591797,
      "learning_rate": 3.0403930131004366e-05,
      "loss": 5.0461,
      "step": 5100
    },
    {
      "epoch": 5.676855895196507,
      "grad_norm": 39.26410675048828,
      "learning_rate": 2.9039301310043672e-05,
      "loss": 4.8651,
      "step": 5200
    },
    {
      "epoch": 5.786026200873362,
      "grad_norm": 25.232053756713867,
      "learning_rate": 2.767467248908297e-05,
      "loss": 4.8736,
      "step": 5300
    },
    {
      "epoch": 5.895196506550218,
      "grad_norm": 29.347366333007812,
      "learning_rate": 2.631004366812227e-05,
      "loss": 4.9244,
      "step": 5400
    },
    {
      "epoch": 6.0,
      "eval_avg_mae": 5.4548899332682295,
      "eval_loss": 5.45488977432251,
      "eval_mae_lex": 4.807984352111816,
      "eval_mae_sem": 3.6341991424560547,
      "eval_mae_syn": 7.922485828399658,
      "eval_runtime": 27.0225,
      "eval_samples_per_second": 271.145,
      "eval_steps_per_second": 8.474,
      "step": 5496
    },
    {
      "epoch": 6.004366812227074,
      "grad_norm": 27.099462509155273,
      "learning_rate": 2.4945414847161576e-05,
      "loss": 5.0121,
      "step": 5500
    },
    {
      "epoch": 6.11353711790393,
      "grad_norm": 26.199542999267578,
      "learning_rate": 2.3580786026200875e-05,
      "loss": 4.6621,
      "step": 5600
    },
    {
      "epoch": 6.222707423580786,
      "grad_norm": 31.04909324645996,
      "learning_rate": 2.2216157205240178e-05,
      "loss": 4.6792,
      "step": 5700
    },
    {
      "epoch": 6.331877729257642,
      "grad_norm": 26.449748992919922,
      "learning_rate": 2.0851528384279477e-05,
      "loss": 4.6471,
      "step": 5800
    },
    {
      "epoch": 6.441048034934497,
      "grad_norm": 27.806798934936523,
      "learning_rate": 1.948689956331878e-05,
      "loss": 4.5161,
      "step": 5900
    },
    {
      "epoch": 6.550218340611353,
      "grad_norm": 39.113399505615234,
      "learning_rate": 1.812227074235808e-05,
      "loss": 4.5787,
      "step": 6000
    },
    {
      "epoch": 6.6593886462882095,
      "grad_norm": 30.493192672729492,
      "learning_rate": 1.675764192139738e-05,
      "loss": 4.6041,
      "step": 6100
    },
    {
      "epoch": 6.7685589519650655,
      "grad_norm": 35.86678695678711,
      "learning_rate": 1.5393013100436683e-05,
      "loss": 4.6345,
      "step": 6200
    },
    {
      "epoch": 6.877729257641922,
      "grad_norm": 44.66313934326172,
      "learning_rate": 1.4028384279475984e-05,
      "loss": 4.6063,
      "step": 6300
    },
    {
      "epoch": 6.986899563318778,
      "grad_norm": 34.84800720214844,
      "learning_rate": 1.2663755458515283e-05,
      "loss": 4.6152,
      "step": 6400
    },
    {
      "epoch": 7.0,
      "eval_avg_mae": 5.481756210327148,
      "eval_loss": 5.481756210327148,
      "eval_mae_lex": 4.816911220550537,
      "eval_mae_sem": 3.748530626296997,
      "eval_mae_syn": 7.87982702255249,
      "eval_runtime": 27.0121,
      "eval_samples_per_second": 271.248,
      "eval_steps_per_second": 8.478,
      "step": 6412
    }
  ],
  "logging_steps": 100,
  "max_steps": 7328,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.3494363633370368e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}