last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.041884816753926704,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.004188481675392671,
      "grad_norm": 9.106295585632324,
      "learning_rate": 2e-05,
      "loss": 9.832,
      "step": 1
    },
    {
      "epoch": 0.004188481675392671,
      "eval_loss": 9.496695518493652,
      "eval_runtime": 1.0945,
      "eval_samples_per_second": 92.276,
      "eval_steps_per_second": 46.595,
      "step": 1
    },
    {
      "epoch": 0.008376963350785341,
      "grad_norm": 9.774275779724121,
      "learning_rate": 4e-05,
      "loss": 9.3052,
      "step": 2
    },
    {
      "epoch": 0.012565445026178011,
      "grad_norm": 10.53923225402832,
      "learning_rate": 6e-05,
      "loss": 9.4143,
      "step": 3
    },
    {
      "epoch": 0.012565445026178011,
      "eval_loss": 9.488880157470703,
      "eval_runtime": 1.103,
      "eval_samples_per_second": 91.57,
      "eval_steps_per_second": 46.238,
      "step": 3
    },
    {
      "epoch": 0.016753926701570682,
      "grad_norm": 8.774895668029785,
      "learning_rate": 8e-05,
      "loss": 9.3396,
      "step": 4
    },
    {
      "epoch": 0.020942408376963352,
      "grad_norm": 8.909485816955566,
      "learning_rate": 0.0001,
      "loss": 9.8732,
      "step": 5
    },
    {
      "epoch": 0.025130890052356022,
      "grad_norm": 7.478220462799072,
      "learning_rate": 0.00012,
      "loss": 9.4416,
      "step": 6
    },
    {
      "epoch": 0.025130890052356022,
      "eval_loss": 9.349822998046875,
      "eval_runtime": 1.14,
      "eval_samples_per_second": 88.599,
      "eval_steps_per_second": 44.738,
      "step": 6
    },
    {
      "epoch": 0.02931937172774869,
      "grad_norm": 9.638185501098633,
      "learning_rate": 0.00014,
      "loss": 9.8085,
      "step": 7
    },
    {
      "epoch": 0.033507853403141365,
      "grad_norm": 7.506699085235596,
      "learning_rate": 0.00016,
      "loss": 8.8918,
      "step": 8
    },
    {
      "epoch": 0.03769633507853403,
      "grad_norm": 8.663238525390625,
      "learning_rate": 0.00018,
      "loss": 9.3444,
      "step": 9
    },
    {
      "epoch": 0.03769633507853403,
      "eval_loss": 8.99866771697998,
      "eval_runtime": 1.1792,
      "eval_samples_per_second": 85.653,
      "eval_steps_per_second": 43.25,
      "step": 9
    },
    {
      "epoch": 0.041884816753926704,
      "grad_norm": 8.484256744384766,
      "learning_rate": 0.0002,
      "loss": 8.708,
      "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": 655585443840.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}