trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.926829268292683,
  "eval_steps": 500,
  "global_step": 75,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.3902439024390244,
      "grad_norm": 6.465833920476205,
      "learning_rate": 5e-06,
      "loss": 1.1247,
      "step": 10
    },
    {
      "epoch": 0.7804878048780488,
      "grad_norm": 1.1862331772202763,
      "learning_rate": 5e-06,
      "loss": 1.018,
      "step": 20
    },
    {
      "epoch": 0.975609756097561,
      "eval_loss": 0.9846493005752563,
      "eval_runtime": 18.0366,
      "eval_samples_per_second": 38.311,
      "eval_steps_per_second": 0.61,
      "step": 25
    },
    {
      "epoch": 1.170731707317073,
      "grad_norm": 2.162191793586483,
      "learning_rate": 5e-06,
      "loss": 1.0085,
      "step": 30
    },
    {
      "epoch": 1.5609756097560976,
      "grad_norm": 1.9535821700087348,
      "learning_rate": 5e-06,
      "loss": 0.9331,
      "step": 40
    },
    {
      "epoch": 1.951219512195122,
      "grad_norm": 1.124539760660556,
      "learning_rate": 5e-06,
      "loss": 0.9171,
      "step": 50
    },
    {
      "epoch": 1.9902439024390244,
      "eval_loss": 0.9493349194526672,
      "eval_runtime": 18.0567,
      "eval_samples_per_second": 38.268,
      "eval_steps_per_second": 0.609,
      "step": 51
    },
    {
      "epoch": 2.341463414634146,
      "grad_norm": 0.9405210346947177,
      "learning_rate": 5e-06,
      "loss": 0.9107,
      "step": 60
    },
    {
      "epoch": 2.7317073170731705,
      "grad_norm": 1.8443181229715055,
      "learning_rate": 5e-06,
      "loss": 0.8594,
      "step": 70
    },
    {
      "epoch": 2.926829268292683,
      "eval_loss": 0.9366496801376343,
      "eval_runtime": 16.7048,
      "eval_samples_per_second": 41.365,
      "eval_steps_per_second": 0.658,
      "step": 75
    },
    {
      "epoch": 2.926829268292683,
      "step": 75,
      "total_flos": 125418413752320.0,
      "train_loss": 0.9610127830505371,
      "train_runtime": 2801.3394,
      "train_samples_per_second": 14.047,
      "train_steps_per_second": 0.027
    }
  ],
  "logging_steps": 10,
  "max_steps": 75,
  "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": 125418413752320.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}