last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005608524957936063,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005608524957936063,
      "grad_norm": 0.013099128380417824,
      "learning_rate": 2e-05,
      "loss": 11.9315,
      "step": 1
    },
    {
      "epoch": 0.0005608524957936063,
      "eval_loss": 11.931306838989258,
      "eval_runtime": 3.0333,
      "eval_samples_per_second": 247.582,
      "eval_steps_per_second": 123.956,
      "step": 1
    },
    {
      "epoch": 0.0011217049915872126,
      "grad_norm": 0.008311104960739613,
      "learning_rate": 4e-05,
      "loss": 11.9304,
      "step": 2
    },
    {
      "epoch": 0.0016825574873808188,
      "grad_norm": 0.012805287726223469,
      "learning_rate": 6e-05,
      "loss": 11.9319,
      "step": 3
    },
    {
      "epoch": 0.0016825574873808188,
      "eval_loss": 11.931303977966309,
      "eval_runtime": 4.0939,
      "eval_samples_per_second": 183.444,
      "eval_steps_per_second": 91.844,
      "step": 3
    },
    {
      "epoch": 0.002243409983174425,
      "grad_norm": 0.016239454969763756,
      "learning_rate": 8e-05,
      "loss": 11.9328,
      "step": 4
    },
    {
      "epoch": 0.0028042624789680315,
      "grad_norm": 0.0150977224111557,
      "learning_rate": 0.0001,
      "loss": 11.9248,
      "step": 5
    },
    {
      "epoch": 0.0033651149747616375,
      "grad_norm": 0.008720125071704388,
      "learning_rate": 0.00012,
      "loss": 11.926,
      "step": 6
    },
    {
      "epoch": 0.0033651149747616375,
      "eval_loss": 11.931292533874512,
      "eval_runtime": 3.9788,
      "eval_samples_per_second": 188.751,
      "eval_steps_per_second": 94.501,
      "step": 6
    },
    {
      "epoch": 0.003925967470555244,
      "grad_norm": 0.010150888003408909,
      "learning_rate": 0.00014,
      "loss": 11.9324,
      "step": 7
    },
    {
      "epoch": 0.00448681996634885,
      "grad_norm": 0.0062065040692687035,
      "learning_rate": 0.00016,
      "loss": 11.935,
      "step": 8
    },
    {
      "epoch": 0.005047672462142457,
      "grad_norm": 0.012440411373972893,
      "learning_rate": 0.00018,
      "loss": 11.9287,
      "step": 9
    },
    {
      "epoch": 0.005047672462142457,
      "eval_loss": 11.931264877319336,
      "eval_runtime": 3.8023,
      "eval_samples_per_second": 197.511,
      "eval_steps_per_second": 98.887,
      "step": 9
    },
    {
      "epoch": 0.005608524957936063,
      "grad_norm": 0.00820008572191,
      "learning_rate": 0.0002,
      "loss": 11.9335,
      "step": 10
    }
  ],
  "logging_steps": 1,
  "max_steps": 10,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 3,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1211154432.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}