last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.027932960893854747,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002793296089385475,
      "grad_norm": 2.895449638366699,
      "learning_rate": 2e-05,
      "loss": 8.9931,
      "step": 1
    },
    {
      "epoch": 0.002793296089385475,
      "eval_loss": 2.400453567504883,
      "eval_runtime": 6.586,
      "eval_samples_per_second": 22.927,
      "eval_steps_per_second": 11.54,
      "step": 1
    },
    {
      "epoch": 0.00558659217877095,
      "grad_norm": 3.0176351070404053,
      "learning_rate": 4e-05,
      "loss": 8.9609,
      "step": 2
    },
    {
      "epoch": 0.008379888268156424,
      "grad_norm": 3.3207359313964844,
      "learning_rate": 6e-05,
      "loss": 9.9514,
      "step": 3
    },
    {
      "epoch": 0.008379888268156424,
      "eval_loss": 2.384174346923828,
      "eval_runtime": 7.2456,
      "eval_samples_per_second": 20.84,
      "eval_steps_per_second": 10.489,
      "step": 3
    },
    {
      "epoch": 0.0111731843575419,
      "grad_norm": 3.740717887878418,
      "learning_rate": 8e-05,
      "loss": 9.0674,
      "step": 4
    },
    {
      "epoch": 0.013966480446927373,
      "grad_norm": 3.150239944458008,
      "learning_rate": 0.0001,
      "loss": 8.5266,
      "step": 5
    },
    {
      "epoch": 0.01675977653631285,
      "grad_norm": 3.9240503311157227,
      "learning_rate": 0.00012,
      "loss": 8.6384,
      "step": 6
    },
    {
      "epoch": 0.01675977653631285,
      "eval_loss": 2.2556746006011963,
      "eval_runtime": 11.1555,
      "eval_samples_per_second": 13.536,
      "eval_steps_per_second": 6.813,
      "step": 6
    },
    {
      "epoch": 0.019553072625698324,
      "grad_norm": 4.200657367706299,
      "learning_rate": 0.00014,
      "loss": 8.5678,
      "step": 7
    },
    {
      "epoch": 0.0223463687150838,
      "grad_norm": 4.508555889129639,
      "learning_rate": 0.00016,
      "loss": 8.5416,
      "step": 8
    },
    {
      "epoch": 0.025139664804469275,
      "grad_norm": 4.351197242736816,
      "learning_rate": 0.00018,
      "loss": 7.5666,
      "step": 9
    },
    {
      "epoch": 0.025139664804469275,
      "eval_loss": 1.9446355104446411,
      "eval_runtime": 11.0871,
      "eval_samples_per_second": 13.619,
      "eval_steps_per_second": 6.855,
      "step": 9
    },
    {
      "epoch": 0.027932960893854747,
      "grad_norm": 4.46405029296875,
      "learning_rate": 0.0002,
      "loss": 7.6561,
      "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": 2852805358387200.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}