{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6694560669456067,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16736401673640167,
      "grad_norm": 1.0720182657241821,
      "learning_rate": 1.993822843360249e-05,
      "loss": 1.6067840576171875,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 4.98675,
      "step": 10,
      "tokens/total": 3932160,
      "tokens/train_per_sec_per_gpu": 16.35,
      "tokens/trainable": 1840181
    },
    {
      "epoch": 0.33472803347280333,
      "grad_norm": 0.36571913957595825,
      "learning_rate": 1.925203832797706e-05,
      "loss": 1.4119406700134278,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 4.10391,
      "step": 20,
      "tokens/total": 7864320,
      "tokens/train_per_sec_per_gpu": 17.64,
      "tokens/trainable": 3698601
    },
    {
      "epoch": 0.502092050209205,
      "grad_norm": 0.25172561407089233,
      "learning_rate": 1.785532003850676e-05,
      "loss": 1.3203330993652345,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.74467,
      "step": 30,
      "tokens/total": 11796480,
      "tokens/train_per_sec_per_gpu": 18.14,
      "tokens/trainable": 5537401
    },
    {
      "epoch": 0.6694560669456067,
      "grad_norm": 0.21719826757907867,
      "learning_rate": 1.5855337551329285e-05,
      "loss": 1.2934332847595216,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.64528,
      "step": 40,
      "tokens/total": 15728640,
      "tokens/train_per_sec_per_gpu": 18.49,
      "tokens/trainable": 7378776
    },
    {
      "epoch": 0.8368200836820083,
      "grad_norm": 0.18392807245254517,
      "learning_rate": 1.3405682693701237e-05,
      "loss": 1.268669319152832,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.55612,
      "step": 50,
      "tokens/total": 19660800,
      "tokens/train_per_sec_per_gpu": 17.87,
      "tokens/trainable": 9224452
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.17360800504684448,
      "learning_rate": 1.0694483535189647e-05,
      "loss": 1.2421661376953126,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.46311,
      "step": 60,
      "tokens/total": 23482368,
      "tokens/train_per_sec_per_gpu": 20.06,
      "tokens/trainable": 11016035
    },
    {
      "epoch": 1.1673640167364017,
      "grad_norm": 0.1570226103067398,
      "learning_rate": 7.929947969387285e-06,
      "loss": 1.2221103668212892,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.39434,
      "step": 70,
      "tokens/total": 27414528,
      "tokens/train_per_sec_per_gpu": 17.87,
      "tokens/trainable": 12851399
    },
    {
      "epoch": 1.3347280334728033,
      "grad_norm": 0.1533421128988266,
      "learning_rate": 5.324387529981323e-06,
      "loss": 1.2150437355041503,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.37044,
      "step": 80,
      "tokens/total": 31346688,
      "tokens/train_per_sec_per_gpu": 18.0,
      "tokens/trainable": 14712337
    },
    {
      "epoch": 1.502092050209205,
      "grad_norm": 0.1498780995607376,
      "learning_rate": 3.0778994641877944e-06,
      "loss": 1.2018341064453124,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.32621,
      "step": 90,
      "tokens/total": 35278848,
      "tokens/train_per_sec_per_gpu": 17.6,
      "tokens/trainable": 16552748
    },
    {
      "epoch": 1.6694560669456067,
      "grad_norm": 0.15034393966197968,
      "learning_rate": 1.3630069588543847e-06,
      "loss": 1.1949978828430177,
      "memory/device_reserved (GiB)": 33.31,
      "memory/max_active (GiB)": 32.17,
      "memory/max_allocated (GiB)": 32.17,
      "ppl": 3.30355,
      "step": 100,
      "tokens/total": 39211008,
      "tokens/train_per_sec_per_gpu": 17.79,
      "tokens/trainable": 18391700
    }
  ],
  "logging_steps": 10,
  "max_steps": 118,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 20,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.80112615568887e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}