trainer_state.json

{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 4.795750141143799,
      "learning_rate": 3.96e-06,
      "loss": 1.3996,
      "step": 100
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 4.383473873138428,
      "learning_rate": 7.960000000000002e-06,
      "loss": 1.1835,
      "step": 200
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 4.073297500610352,
      "learning_rate": 1.196e-05,
      "loss": 1.2475,
      "step": 300
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 4.537884712219238,
      "learning_rate": 1.5960000000000003e-05,
      "loss": 1.2569,
      "step": 400
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 3.899653196334839,
      "learning_rate": 1.9960000000000002e-05,
      "loss": 1.223,
      "step": 500
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 3.367648124694824,
      "learning_rate": 1.87016393442623e-05,
      "loss": 1.2074,
      "step": 600
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.287987232208252,
      "eval_runtime": 179.519,
      "eval_samples_per_second": 3.342,
      "eval_steps_per_second": 0.418,
      "step": 675
    },
    {
      "epoch": 1.037037037037037,
      "grad_norm": 3.876903533935547,
      "learning_rate": 1.7390163934426233e-05,
      "loss": 1.1854,
      "step": 700
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 3.6909120082855225,
      "learning_rate": 1.6078688524590164e-05,
      "loss": 0.9659,
      "step": 800
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 4.26814603805542,
      "learning_rate": 1.47672131147541e-05,
      "loss": 1.0073,
      "step": 900
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 3.909092664718628,
      "learning_rate": 1.3455737704918036e-05,
      "loss": 1.0126,
      "step": 1000
    },
    {
      "epoch": 1.6296296296296298,
      "grad_norm": 4.576954364776611,
      "learning_rate": 1.2144262295081968e-05,
      "loss": 1.0268,
      "step": 1100
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 4.566445827484131,
      "learning_rate": 1.0832786885245903e-05,
      "loss": 1.0,
      "step": 1200
    },
    {
      "epoch": 1.925925925925926,
      "grad_norm": 4.966795921325684,
      "learning_rate": 9.521311475409837e-06,
      "loss": 1.0073,
      "step": 1300
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.282468318939209,
      "eval_runtime": 179.5292,
      "eval_samples_per_second": 3.342,
      "eval_steps_per_second": 0.418,
      "step": 1350
    }
  ],
  "logging_steps": 100,
  "max_steps": 2025,
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.46136530681856e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}