last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.00023365714319093878,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.3365714319093876e-05,
      "grad_norm": 0.8085529804229736,
      "learning_rate": 2e-05,
      "loss": 2.9772,
      "step": 1
    },
    {
      "epoch": 2.3365714319093876e-05,
      "eval_loss": 0.8430490493774414,
      "eval_runtime": 1868.3465,
      "eval_samples_per_second": 9.645,
      "eval_steps_per_second": 4.823,
      "step": 1
    },
    {
      "epoch": 4.673142863818775e-05,
      "grad_norm": 0.8253940939903259,
      "learning_rate": 4e-05,
      "loss": 2.9822,
      "step": 2
    },
    {
      "epoch": 7.009714295728164e-05,
      "grad_norm": 0.7106623649597168,
      "learning_rate": 6e-05,
      "loss": 3.1731,
      "step": 3
    },
    {
      "epoch": 7.009714295728164e-05,
      "eval_loss": 0.8419922590255737,
      "eval_runtime": 935.4078,
      "eval_samples_per_second": 19.265,
      "eval_steps_per_second": 9.633,
      "step": 3
    },
    {
      "epoch": 9.34628572763755e-05,
      "grad_norm": 0.8510690927505493,
      "learning_rate": 8e-05,
      "loss": 2.6167,
      "step": 4
    },
    {
      "epoch": 0.00011682857159546939,
      "grad_norm": 0.9762678742408752,
      "learning_rate": 0.0001,
      "loss": 3.4279,
      "step": 5
    },
    {
      "epoch": 0.00014019428591456327,
      "grad_norm": 0.885900616645813,
      "learning_rate": 0.00012,
      "loss": 3.2001,
      "step": 6
    },
    {
      "epoch": 0.00014019428591456327,
      "eval_loss": 0.8309927582740784,
      "eval_runtime": 943.7787,
      "eval_samples_per_second": 19.095,
      "eval_steps_per_second": 9.548,
      "step": 6
    },
    {
      "epoch": 0.00016356000023365715,
      "grad_norm": 1.1779074668884277,
      "learning_rate": 0.00014,
      "loss": 3.1902,
      "step": 7
    },
    {
      "epoch": 0.000186925714552751,
      "grad_norm": 1.1536974906921387,
      "learning_rate": 0.00016,
      "loss": 3.2861,
      "step": 8
    },
    {
      "epoch": 0.0002102914288718449,
      "grad_norm": 1.2713385820388794,
      "learning_rate": 0.00018,
      "loss": 3.4101,
      "step": 9
    },
    {
      "epoch": 0.0002102914288718449,
      "eval_loss": 0.7832401990890503,
      "eval_runtime": 1120.8965,
      "eval_samples_per_second": 16.077,
      "eval_steps_per_second": 8.039,
      "step": 9
    },
    {
      "epoch": 0.00023365714319093878,
      "grad_norm": 1.4203972816467285,
      "learning_rate": 0.0002,
      "loss": 3.1491,
      "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": 2646691345858560.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}