{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5,
      "grad_norm": 14.416919708251953,
      "learning_rate": 9.75e-05,
      "loss": 0.3808,
      "step": 10
    },
    {
      "epoch": 1.0,
      "grad_norm": 14.942909240722656,
      "learning_rate": 9.5e-05,
      "loss": 0.1636,
      "step": 20
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.1972920000553131,
      "eval_mse": 0.1972920149564743,
      "eval_runtime": 1.8069,
      "eval_samples_per_second": 21.584,
      "eval_steps_per_second": 2.767,
      "step": 20
    },
    {
      "epoch": 1.5,
      "grad_norm": 2.4087517261505127,
      "learning_rate": 9.250000000000001e-05,
      "loss": 0.2302,
      "step": 30
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.778547286987305,
      "learning_rate": 9e-05,
      "loss": 0.1331,
      "step": 40
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.046439673751592636,
      "eval_mse": 0.046439677476882935,
      "eval_runtime": 1.622,
      "eval_samples_per_second": 24.044,
      "eval_steps_per_second": 3.083,
      "step": 40
    },
    {
      "epoch": 2.5,
      "grad_norm": 1.1269134283065796,
      "learning_rate": 8.75e-05,
      "loss": 0.05,
      "step": 50
    },
    {
      "epoch": 3.0,
      "grad_norm": 4.305675983428955,
      "learning_rate": 8.5e-05,
      "loss": 0.0289,
      "step": 60
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.03575298935174942,
      "eval_mse": 0.03575298190116882,
      "eval_runtime": 1.6208,
      "eval_samples_per_second": 24.063,
      "eval_steps_per_second": 3.085,
      "step": 60
    },
    {
      "epoch": 3.5,
      "grad_norm": 1.7014061212539673,
      "learning_rate": 8.25e-05,
      "loss": 0.0246,
      "step": 70
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.39174097776412964,
      "learning_rate": 8e-05,
      "loss": 0.0221,
      "step": 80
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.03326353803277016,
      "eval_mse": 0.03326353803277016,
      "eval_runtime": 1.6114,
      "eval_samples_per_second": 24.203,
      "eval_steps_per_second": 3.103,
      "step": 80
    },
    {
      "epoch": 4.5,
      "grad_norm": 2.6145267486572266,
      "learning_rate": 7.75e-05,
      "loss": 0.021,
      "step": 90
    },
    {
      "epoch": 5.0,
      "grad_norm": 13.847771644592285,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0223,
      "step": 100
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.03398064896464348,
      "eval_mse": 0.033980656415224075,
      "eval_runtime": 1.6515,
      "eval_samples_per_second": 23.615,
      "eval_steps_per_second": 3.028,
      "step": 100
    },
    {
      "epoch": 5.5,
      "grad_norm": 0.43660324811935425,
      "learning_rate": 7.25e-05,
      "loss": 0.0131,
      "step": 110
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.1774399280548096,
      "learning_rate": 7e-05,
      "loss": 0.0117,
      "step": 120
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.04457540437579155,
      "eval_mse": 0.04457540065050125,
      "eval_runtime": 1.6081,
      "eval_samples_per_second": 24.252,
      "eval_steps_per_second": 3.109,
      "step": 120
    },
    {
      "epoch": 6.5,
      "grad_norm": 2.61739182472229,
      "learning_rate": 6.750000000000001e-05,
      "loss": 0.0168,
      "step": 130
    },
    {
      "epoch": 7.0,
      "grad_norm": 2.7107529640197754,
      "learning_rate": 6.500000000000001e-05,
      "loss": 0.0107,
      "step": 140
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.03702976927161217,
      "eval_mse": 0.037029776722192764,
      "eval_runtime": 1.6027,
      "eval_samples_per_second": 24.334,
      "eval_steps_per_second": 3.12,
      "step": 140
    },
    {
      "epoch": 7.5,
      "grad_norm": 4.807140350341797,
      "learning_rate": 6.25e-05,
      "loss": 0.0112,
      "step": 150
    },
    {
      "epoch": 8.0,
      "grad_norm": 1.6699814796447754,
      "learning_rate": 6e-05,
      "loss": 0.0096,
      "step": 160
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.03073795698583126,
      "eval_mse": 0.030737943947315216,
      "eval_runtime": 1.678,
      "eval_samples_per_second": 23.242,
      "eval_steps_per_second": 2.98,
      "step": 160
    },
    {
      "epoch": 8.5,
      "grad_norm": 5.444133281707764,
      "learning_rate": 5.7499999999999995e-05,
      "loss": 0.0099,
      "step": 170
    },
    {
      "epoch": 9.0,
      "grad_norm": 1.5312561988830566,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.0142,
      "step": 180
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.03504549711942673,
      "eval_mse": 0.03504551202058792,
      "eval_runtime": 1.6103,
      "eval_samples_per_second": 24.218,
      "eval_steps_per_second": 3.105,
      "step": 180
    },
    {
      "epoch": 9.5,
      "grad_norm": 1.527550220489502,
      "learning_rate": 5.25e-05,
      "loss": 0.0051,
      "step": 190
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.0232219696044922,
      "learning_rate": 5e-05,
      "loss": 0.0069,
      "step": 200
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.032399099320173264,
      "eval_mse": 0.03239908814430237,
      "eval_runtime": 1.61,
      "eval_samples_per_second": 24.224,
      "eval_steps_per_second": 3.106,
      "step": 200
    },
    {
      "epoch": 10.5,
      "grad_norm": 1.1013288497924805,
      "learning_rate": 4.75e-05,
      "loss": 0.0034,
      "step": 210
    },
    {
      "epoch": 11.0,
      "grad_norm": 0.35051777958869934,
      "learning_rate": 4.5e-05,
      "loss": 0.0028,
      "step": 220
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.02933628484606743,
      "eval_mse": 0.029336294159293175,
      "eval_runtime": 1.7012,
      "eval_samples_per_second": 22.925,
      "eval_steps_per_second": 2.939,
      "step": 220
    },
    {
      "epoch": 11.5,
      "grad_norm": 1.1170843839645386,
      "learning_rate": 4.25e-05,
      "loss": 0.0019,
      "step": 230
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.3299288749694824,
      "learning_rate": 4e-05,
      "loss": 0.0044,
      "step": 240
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.028278259560465813,
      "eval_mse": 0.028278270736336708,
      "eval_runtime": 1.5914,
      "eval_samples_per_second": 24.506,
      "eval_steps_per_second": 3.142,
      "step": 240
    },
    {
      "epoch": 12.5,
      "grad_norm": 1.6604584455490112,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.002,
      "step": 250
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.2441127300262451,
      "learning_rate": 3.5e-05,
      "loss": 0.0011,
      "step": 260
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.029920559376478195,
      "eval_mse": 0.029920564964413643,
      "eval_runtime": 1.6282,
      "eval_samples_per_second": 23.953,
      "eval_steps_per_second": 3.071,
      "step": 260
    },
    {
      "epoch": 13.5,
      "grad_norm": 0.7714802026748657,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.0008,
      "step": 270
    },
    {
      "epoch": 14.0,
      "grad_norm": 0.5498138070106506,
      "learning_rate": 3e-05,
      "loss": 0.0005,
      "step": 280
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.027942122891545296,
      "eval_mse": 0.027942117303609848,
      "eval_runtime": 1.5994,
      "eval_samples_per_second": 24.384,
      "eval_steps_per_second": 3.126,
      "step": 280
    },
    {
      "epoch": 14.5,
      "grad_norm": 0.5462870001792908,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.0006,
      "step": 290
    },
    {
      "epoch": 15.0,
      "grad_norm": 0.32672354578971863,
      "learning_rate": 2.5e-05,
      "loss": 0.0005,
      "step": 300
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.029117202386260033,
      "eval_mse": 0.02911720797419548,
      "eval_runtime": 1.595,
      "eval_samples_per_second": 24.451,
      "eval_steps_per_second": 3.135,
      "step": 300
    },
    {
      "epoch": 15.5,
      "grad_norm": 0.7088171243667603,
      "learning_rate": 2.25e-05,
      "loss": 0.0012,
      "step": 310
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.3224898874759674,
      "learning_rate": 2e-05,
      "loss": 0.0011,
      "step": 320
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.028802577406167984,
      "eval_mse": 0.028802569955587387,
      "eval_runtime": 1.6242,
      "eval_samples_per_second": 24.012,
      "eval_steps_per_second": 3.078,
      "step": 320
    },
    {
      "epoch": 16.5,
      "grad_norm": 0.2536928951740265,
      "learning_rate": 1.75e-05,
      "loss": 0.0002,
      "step": 330
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.2693057060241699,
      "learning_rate": 1.5e-05,
      "loss": 0.0003,
      "step": 340
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.028974896296858788,
      "eval_mse": 0.02897489443421364,
      "eval_runtime": 1.5937,
      "eval_samples_per_second": 24.472,
      "eval_steps_per_second": 3.137,
      "step": 340
    },
    {
      "epoch": 17.5,
      "grad_norm": 0.22231905162334442,
      "learning_rate": 1.25e-05,
      "loss": 0.0001,
      "step": 350
    },
    {
      "epoch": 18.0,
      "grad_norm": 0.44173210859298706,
      "learning_rate": 1e-05,
      "loss": 0.0001,
      "step": 360
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.029911378398537636,
      "eval_mse": 0.029911383986473083,
      "eval_runtime": 1.5913,
      "eval_samples_per_second": 24.509,
      "eval_steps_per_second": 3.142,
      "step": 360
    },
    {
      "epoch": 18.5,
      "grad_norm": 0.2958744764328003,
      "learning_rate": 7.5e-06,
      "loss": 0.0001,
      "step": 370
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.41316938400268555,
      "learning_rate": 5e-06,
      "loss": 0.0001,
      "step": 380
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.029724078252911568,
      "eval_mse": 0.029724083840847015,
      "eval_runtime": 1.5907,
      "eval_samples_per_second": 24.518,
      "eval_steps_per_second": 3.143,
      "step": 380
    },
    {
      "epoch": 19.5,
      "grad_norm": 0.0391409695148468,
      "learning_rate": 2.5e-06,
      "loss": 0.0,
      "step": 390
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.021498844027519226,
      "learning_rate": 0.0,
      "loss": 0.0,
      "step": 400
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 10,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}