last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.000827934510380229,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 8.27934510380229e-05,
      "grad_norm": 1.477283000946045,
      "learning_rate": 2e-05,
      "loss": 2.8657,
      "step": 1
    },
    {
      "epoch": 8.27934510380229e-05,
      "eval_loss": 0.6027017831802368,
      "eval_runtime": 763.0731,
      "eval_samples_per_second": 6.665,
      "eval_steps_per_second": 3.333,
      "step": 1
    },
    {
      "epoch": 0.0001655869020760458,
      "grad_norm": 0.8949412107467651,
      "learning_rate": 4e-05,
      "loss": 2.1009,
      "step": 2
    },
    {
      "epoch": 0.00024838035311406867,
      "grad_norm": 1.8197911977767944,
      "learning_rate": 6e-05,
      "loss": 2.5783,
      "step": 3
    },
    {
      "epoch": 0.00024838035311406867,
      "eval_loss": 0.5999210476875305,
      "eval_runtime": 653.8233,
      "eval_samples_per_second": 7.779,
      "eval_steps_per_second": 3.889,
      "step": 3
    },
    {
      "epoch": 0.0003311738041520916,
      "grad_norm": 1.435214638710022,
      "learning_rate": 8e-05,
      "loss": 2.3543,
      "step": 4
    },
    {
      "epoch": 0.0004139672551901145,
      "grad_norm": 1.3697088956832886,
      "learning_rate": 0.0001,
      "loss": 2.2362,
      "step": 5
    },
    {
      "epoch": 0.0004967607062281373,
      "grad_norm": 1.8139108419418335,
      "learning_rate": 0.00012,
      "loss": 3.0984,
      "step": 6
    },
    {
      "epoch": 0.0004967607062281373,
      "eval_loss": 0.541894257068634,
      "eval_runtime": 422.6787,
      "eval_samples_per_second": 12.033,
      "eval_steps_per_second": 6.016,
      "step": 6
    },
    {
      "epoch": 0.0005795541572661603,
      "grad_norm": 2.073357343673706,
      "learning_rate": 0.00014,
      "loss": 1.9979,
      "step": 7
    },
    {
      "epoch": 0.0006623476083041832,
      "grad_norm": 2.176180124282837,
      "learning_rate": 0.00016,
      "loss": 1.9623,
      "step": 8
    },
    {
      "epoch": 0.000745141059342206,
      "grad_norm": 2.046623706817627,
      "learning_rate": 0.00018,
      "loss": 1.6899,
      "step": 9
    },
    {
      "epoch": 0.000745141059342206,
      "eval_loss": 0.30938443541526794,
      "eval_runtime": 422.8667,
      "eval_samples_per_second": 12.027,
      "eval_steps_per_second": 6.014,
      "step": 9
    },
    {
      "epoch": 0.000827934510380229,
      "grad_norm": 1.7581552267074585,
      "learning_rate": 0.0002,
      "loss": 1.0762,
      "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": 4670180850401280.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}