trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9872958257713247,
  "eval_steps": 500,
  "global_step": 102,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.29038112522686027,
      "grad_norm": 1.5343472890056573,
      "learning_rate": 5e-06,
      "loss": 0.9197,
      "step": 10
    },
    {
      "epoch": 0.5807622504537205,
      "grad_norm": 2.240679614803961,
      "learning_rate": 5e-06,
      "loss": 0.8131,
      "step": 20
    },
    {
      "epoch": 0.8711433756805808,
      "grad_norm": 3.4712434078338736,
      "learning_rate": 5e-06,
      "loss": 0.7844,
      "step": 30
    },
    {
      "epoch": 0.9872958257713249,
      "eval_loss": 0.7656542062759399,
      "eval_runtime": 114.0203,
      "eval_samples_per_second": 8.13,
      "eval_steps_per_second": 0.254,
      "step": 34
    },
    {
      "epoch": 1.161524500907441,
      "grad_norm": 0.7840407973392162,
      "learning_rate": 5e-06,
      "loss": 0.7443,
      "step": 40
    },
    {
      "epoch": 1.4519056261343013,
      "grad_norm": 1.4138928121402814,
      "learning_rate": 5e-06,
      "loss": 0.7169,
      "step": 50
    },
    {
      "epoch": 1.7422867513611615,
      "grad_norm": 0.6913878841678102,
      "learning_rate": 5e-06,
      "loss": 0.7088,
      "step": 60
    },
    {
      "epoch": 1.9745916515426498,
      "eval_loss": 0.7332684993743896,
      "eval_runtime": 112.5535,
      "eval_samples_per_second": 8.236,
      "eval_steps_per_second": 0.258,
      "step": 68
    },
    {
      "epoch": 2.058076225045372,
      "grad_norm": 1.845390252605406,
      "learning_rate": 5e-06,
      "loss": 0.6585,
      "step": 70
    },
    {
      "epoch": 2.3484573502722323,
      "grad_norm": 1.2356723897039843,
      "learning_rate": 5e-06,
      "loss": 0.648,
      "step": 80
    },
    {
      "epoch": 2.6388384754990923,
      "grad_norm": 1.9984492481741913,
      "learning_rate": 5e-06,
      "loss": 0.6451,
      "step": 90
    },
    {
      "epoch": 2.9292196007259528,
      "grad_norm": 0.9737870271644316,
      "learning_rate": 5e-06,
      "loss": 0.638,
      "step": 100
    },
    {
      "epoch": 2.9872958257713247,
      "eval_loss": 0.730301558971405,
      "eval_runtime": 110.6618,
      "eval_samples_per_second": 8.377,
      "eval_steps_per_second": 0.262,
      "step": 102
    },
    {
      "epoch": 2.9872958257713247,
      "step": 102,
      "total_flos": 172110076968960.0,
      "train_loss": 0.0,
      "train_runtime": 2.9305,
      "train_samples_per_second": 18030.854,
      "train_steps_per_second": 34.807
    }
  ],
  "logging_steps": 10,
  "max_steps": 102,
  "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": 172110076968960.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}