last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0001379552955864652,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.3795529558646521e-05,
      "grad_norm": 0.6632211804389954,
      "learning_rate": 2e-05,
      "loss": 9.1361,
      "step": 1
    },
    {
      "epoch": 1.3795529558646521e-05,
      "eval_loss": 2.80515718460083,
      "eval_runtime": 1226.5126,
      "eval_samples_per_second": 24.884,
      "eval_steps_per_second": 12.443,
      "step": 1
    },
    {
      "epoch": 2.7591059117293042e-05,
      "grad_norm": 1.992271900177002,
      "learning_rate": 4e-05,
      "loss": 13.3555,
      "step": 2
    },
    {
      "epoch": 4.1386588675939564e-05,
      "grad_norm": 1.5762066841125488,
      "learning_rate": 6e-05,
      "loss": 13.7686,
      "step": 3
    },
    {
      "epoch": 4.1386588675939564e-05,
      "eval_loss": 2.804302930831909,
      "eval_runtime": 1220.861,
      "eval_samples_per_second": 25.0,
      "eval_steps_per_second": 12.5,
      "step": 3
    },
    {
      "epoch": 5.5182118234586084e-05,
      "grad_norm": 1.2571256160736084,
      "learning_rate": 8e-05,
      "loss": 11.4214,
      "step": 4
    },
    {
      "epoch": 6.89776477932326e-05,
      "grad_norm": 1.5989934206008911,
      "learning_rate": 0.0001,
      "loss": 12.4182,
      "step": 5
    },
    {
      "epoch": 8.277317735187913e-05,
      "grad_norm": 3.580453395843506,
      "learning_rate": 0.00012,
      "loss": 13.3373,
      "step": 6
    },
    {
      "epoch": 8.277317735187913e-05,
      "eval_loss": 2.794506788253784,
      "eval_runtime": 1225.5368,
      "eval_samples_per_second": 24.904,
      "eval_steps_per_second": 12.453,
      "step": 6
    },
    {
      "epoch": 9.656870691052564e-05,
      "grad_norm": 3.1573030948638916,
      "learning_rate": 0.00014,
      "loss": 13.9403,
      "step": 7
    },
    {
      "epoch": 0.00011036423646917217,
      "grad_norm": 3.094115734100342,
      "learning_rate": 0.00016,
      "loss": 13.1829,
      "step": 8
    },
    {
      "epoch": 0.0001241597660278187,
      "grad_norm": 2.3581008911132812,
      "learning_rate": 0.00018,
      "loss": 11.7588,
      "step": 9
    },
    {
      "epoch": 0.0001241597660278187,
      "eval_loss": 2.747213840484619,
      "eval_runtime": 1225.3883,
      "eval_samples_per_second": 24.907,
      "eval_steps_per_second": 12.454,
      "step": 9
    },
    {
      "epoch": 0.0001379552955864652,
      "grad_norm": 1.9102445840835571,
      "learning_rate": 0.0002,
      "loss": 12.4421,
      "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": 1995217821499392.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}