last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.08316008316008316,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.008316008316008316,
      "grad_norm": 3.69275164604187,
      "learning_rate": 2e-05,
      "loss": 0.8451,
      "step": 1
    },
    {
      "epoch": 0.008316008316008316,
      "eval_loss": 0.845096230506897,
      "eval_runtime": 3.446,
      "eval_samples_per_second": 14.8,
      "eval_steps_per_second": 7.545,
      "step": 1
    },
    {
      "epoch": 0.016632016632016633,
      "grad_norm": 3.2952795028686523,
      "learning_rate": 4e-05,
      "loss": 0.9667,
      "step": 2
    },
    {
      "epoch": 0.02494802494802495,
      "grad_norm": 2.9131577014923096,
      "learning_rate": 6e-05,
      "loss": 0.8653,
      "step": 3
    },
    {
      "epoch": 0.02494802494802495,
      "eval_loss": 0.8450484871864319,
      "eval_runtime": 1.3979,
      "eval_samples_per_second": 36.482,
      "eval_steps_per_second": 18.599,
      "step": 3
    },
    {
      "epoch": 0.033264033264033266,
      "grad_norm": 2.838355779647827,
      "learning_rate": 8e-05,
      "loss": 0.7319,
      "step": 4
    },
    {
      "epoch": 0.04158004158004158,
      "grad_norm": 2.717000722885132,
      "learning_rate": 0.0001,
      "loss": 0.9133,
      "step": 5
    },
    {
      "epoch": 0.0498960498960499,
      "grad_norm": 3.1441810131073,
      "learning_rate": 0.00012,
      "loss": 0.8274,
      "step": 6
    },
    {
      "epoch": 0.0498960498960499,
      "eval_loss": 0.8306875824928284,
      "eval_runtime": 1.4115,
      "eval_samples_per_second": 36.132,
      "eval_steps_per_second": 18.42,
      "step": 6
    },
    {
      "epoch": 0.058212058212058215,
      "grad_norm": 3.0713138580322266,
      "learning_rate": 0.00014,
      "loss": 0.7636,
      "step": 7
    },
    {
      "epoch": 0.06652806652806653,
      "grad_norm": 3.598985195159912,
      "learning_rate": 0.00016,
      "loss": 0.9892,
      "step": 8
    },
    {
      "epoch": 0.07484407484407485,
      "grad_norm": 5.25281286239624,
      "learning_rate": 0.00018,
      "loss": 0.7988,
      "step": 9
    },
    {
      "epoch": 0.07484407484407485,
      "eval_loss": 0.7632012367248535,
      "eval_runtime": 1.4245,
      "eval_samples_per_second": 35.801,
      "eval_steps_per_second": 18.252,
      "step": 9
    },
    {
      "epoch": 0.08316008316008316,
      "grad_norm": 4.665212631225586,
      "learning_rate": 0.0002,
      "loss": 0.7269,
      "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": 784586338467840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}