{
  "best_metric": 0.7183207652218417,
  "best_model_checkpoint": "training_dir/checkpoint-8000",
  "epoch": 0.9536929108826958,
  "eval_steps": 1000,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1059658789869662,
      "grad_norm": 15.513019561767578,
      "learning_rate": 9.946734846063707e-06,
      "loss": 0.8105,
      "step": 1000
    },
    {
      "epoch": 0.1059658789869662,
      "eval_accuracy": 0.6267482517482518,
      "eval_f1": 0.5486997676789124,
      "eval_loss": 0.8523119688034058,
      "eval_precision": 0.6663256334906192,
      "eval_recall": 0.6267482517482518,
      "eval_runtime": 26.6488,
      "eval_samples_per_second": 85.857,
      "eval_steps_per_second": 10.732,
      "step": 1000
    },
    {
      "epoch": 0.2119317579739324,
      "grad_norm": 14.981549263000488,
      "learning_rate": 9.840204538191116e-06,
      "loss": 0.5828,
      "step": 2000
    },
    {
      "epoch": 0.2119317579739324,
      "eval_accuracy": 0.6682692307692307,
      "eval_f1": 0.6366336108184594,
      "eval_loss": 0.9371851086616516,
      "eval_precision": 0.6852715704153842,
      "eval_recall": 0.6682692307692307,
      "eval_runtime": 25.9805,
      "eval_samples_per_second": 88.066,
      "eval_steps_per_second": 11.008,
      "step": 2000
    },
    {
      "epoch": 0.31789763696089857,
      "grad_norm": 8.239383697509766,
      "learning_rate": 9.733674230318526e-06,
      "loss": 0.5349,
      "step": 3000
    },
    {
      "epoch": 0.31789763696089857,
      "eval_accuracy": 0.6914335664335665,
      "eval_f1": 0.669233772837662,
      "eval_loss": 0.8858013153076172,
      "eval_precision": 0.6967032053774095,
      "eval_recall": 0.6914335664335665,
      "eval_runtime": 25.9486,
      "eval_samples_per_second": 88.174,
      "eval_steps_per_second": 11.022,
      "step": 3000
    },
    {
      "epoch": 0.4238635159478648,
      "grad_norm": 9.338201522827148,
      "learning_rate": 9.627143922445936e-06,
      "loss": 0.5233,
      "step": 4000
    },
    {
      "epoch": 0.4238635159478648,
      "eval_accuracy": 0.7023601398601399,
      "eval_f1": 0.6877643727244785,
      "eval_loss": 0.9143859148025513,
      "eval_precision": 0.7132978004907108,
      "eval_recall": 0.7023601398601399,
      "eval_runtime": 26.1039,
      "eval_samples_per_second": 87.65,
      "eval_steps_per_second": 10.956,
      "step": 4000
    },
    {
      "epoch": 0.5298293949348309,
      "grad_norm": 7.548835754394531,
      "learning_rate": 9.520613614573348e-06,
      "loss": 0.5089,
      "step": 5000
    },
    {
      "epoch": 0.5298293949348309,
      "eval_accuracy": 0.7084790209790209,
      "eval_f1": 0.6979166210878224,
      "eval_loss": 0.8279535174369812,
      "eval_precision": 0.7081949635182484,
      "eval_recall": 0.7084790209790209,
      "eval_runtime": 25.9523,
      "eval_samples_per_second": 88.162,
      "eval_steps_per_second": 11.02,
      "step": 5000
    },
    {
      "epoch": 0.6357952739217971,
      "grad_norm": 7.738587379455566,
      "learning_rate": 9.414083306700758e-06,
      "loss": 0.4964,
      "step": 6000
    },
    {
      "epoch": 0.6357952739217971,
      "eval_accuracy": 0.7106643356643356,
      "eval_f1": 0.6933251320034988,
      "eval_loss": 0.7791977524757385,
      "eval_precision": 0.7149995516226468,
      "eval_recall": 0.7106643356643356,
      "eval_runtime": 26.0573,
      "eval_samples_per_second": 87.806,
      "eval_steps_per_second": 10.976,
      "step": 6000
    },
    {
      "epoch": 0.7417611529087634,
      "grad_norm": 2.134586811065674,
      "learning_rate": 9.307552998828167e-06,
      "loss": 0.4972,
      "step": 7000
    },
    {
      "epoch": 0.7417611529087634,
      "eval_accuracy": 0.715472027972028,
      "eval_f1": 0.7023119335505582,
      "eval_loss": 0.8132364749908447,
      "eval_precision": 0.7209094583908723,
      "eval_recall": 0.715472027972028,
      "eval_runtime": 26.036,
      "eval_samples_per_second": 87.878,
      "eval_steps_per_second": 10.985,
      "step": 7000
    },
    {
      "epoch": 0.8477270318957296,
      "grad_norm": 7.15505838394165,
      "learning_rate": 9.201022690955577e-06,
      "loss": 0.4598,
      "step": 8000
    },
    {
      "epoch": 0.8477270318957296,
      "eval_accuracy": 0.7263986013986014,
      "eval_f1": 0.7183207652218417,
      "eval_loss": 0.782526969909668,
      "eval_precision": 0.7270872567100993,
      "eval_recall": 0.7263986013986014,
      "eval_runtime": 26.0044,
      "eval_samples_per_second": 87.985,
      "eval_steps_per_second": 10.998,
      "step": 8000
    },
    {
      "epoch": 0.9536929108826958,
      "grad_norm": 4.955699920654297,
      "learning_rate": 9.094492383082987e-06,
      "loss": 0.4648,
      "step": 9000
    },
    {
      "epoch": 0.9536929108826958,
      "eval_accuracy": 0.7167832167832168,
      "eval_f1": 0.7085336559753441,
      "eval_loss": 0.8246528506278992,
      "eval_precision": 0.7230635739304796,
      "eval_recall": 0.7167832167832168,
      "eval_runtime": 26.0039,
      "eval_samples_per_second": 87.987,
      "eval_steps_per_second": 10.998,
      "step": 9000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 94370,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 8735478289285824.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}