last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00659582264565775,
  "eval_steps": 3,
  "global_step": 9,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0007328691828508611,
      "grad_norm": Infinity,
      "learning_rate": 0.0,
      "loss": 15.4792,
      "step": 1
    },
    {
      "epoch": 0.0007328691828508611,
      "eval_loss": 12.304828643798828,
      "eval_runtime": 56.9499,
      "eval_samples_per_second": 5.057,
      "eval_steps_per_second": 5.057,
      "step": 1
    },
    {
      "epoch": 0.0014657383657017222,
      "grad_norm": 42.04989242553711,
      "learning_rate": 2e-05,
      "loss": 10.6283,
      "step": 2
    },
    {
      "epoch": 0.0021986075485525836,
      "grad_norm": 37.467315673828125,
      "learning_rate": 4e-05,
      "loss": 11.3833,
      "step": 3
    },
    {
      "epoch": 0.0021986075485525836,
      "eval_loss": 12.051217079162598,
      "eval_runtime": 56.814,
      "eval_samples_per_second": 5.069,
      "eval_steps_per_second": 5.069,
      "step": 3
    },
    {
      "epoch": 0.0029314767314034445,
      "grad_norm": 44.678558349609375,
      "learning_rate": 6e-05,
      "loss": 9.6695,
      "step": 4
    },
    {
      "epoch": 0.003664345914254306,
      "grad_norm": 48.171749114990234,
      "learning_rate": 8e-05,
      "loss": 14.795,
      "step": 5
    },
    {
      "epoch": 0.004397215097105167,
      "grad_norm": 32.893001556396484,
      "learning_rate": 0.0001,
      "loss": 7.3908,
      "step": 6
    },
    {
      "epoch": 0.004397215097105167,
      "eval_loss": 9.441511154174805,
      "eval_runtime": 56.7951,
      "eval_samples_per_second": 5.071,
      "eval_steps_per_second": 5.071,
      "step": 6
    },
    {
      "epoch": 0.005130084279956028,
      "grad_norm": 38.151817321777344,
      "learning_rate": 0.00012,
      "loss": 7.0317,
      "step": 7
    },
    {
      "epoch": 0.005862953462806889,
      "grad_norm": 32.140113830566406,
      "learning_rate": 0.00014,
      "loss": 5.01,
      "step": 8
    },
    {
      "epoch": 0.00659582264565775,
      "grad_norm": 26.151323318481445,
      "learning_rate": 0.00016,
      "loss": 6.6914,
      "step": 9
    },
    {
      "epoch": 0.00659582264565775,
      "eval_loss": 6.388987064361572,
      "eval_runtime": 56.8562,
      "eval_samples_per_second": 5.065,
      "eval_steps_per_second": 5.065,
      "step": 9
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1770795621679104.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}