last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.005044772354647497,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0005044772354647496,
      "grad_norm": NaN,
      "learning_rate": 2e-05,
      "loss": 2.6568,
      "step": 1
    },
    {
      "epoch": 0.0005044772354647496,
      "eval_loss": NaN,
      "eval_runtime": 20.1538,
      "eval_samples_per_second": 41.431,
      "eval_steps_per_second": 20.741,
      "step": 1
    },
    {
      "epoch": 0.0010089544709294992,
      "grad_norm": NaN,
      "learning_rate": 4e-05,
      "loss": 1.0934,
      "step": 2
    },
    {
      "epoch": 0.001513431706394249,
      "grad_norm": NaN,
      "learning_rate": 6e-05,
      "loss": 0.9605,
      "step": 3
    },
    {
      "epoch": 0.001513431706394249,
      "eval_loss": NaN,
      "eval_runtime": 19.2299,
      "eval_samples_per_second": 43.422,
      "eval_steps_per_second": 21.737,
      "step": 3
    },
    {
      "epoch": 0.0020179089418589985,
      "grad_norm": 0.08474644273519516,
      "learning_rate": 8e-05,
      "loss": 1.2304,
      "step": 4
    },
    {
      "epoch": 0.0025223861773237483,
      "grad_norm": NaN,
      "learning_rate": 0.0001,
      "loss": 2.1614,
      "step": 5
    },
    {
      "epoch": 0.003026863412788498,
      "grad_norm": NaN,
      "learning_rate": 0.00012,
      "loss": 2.0522,
      "step": 6
    },
    {
      "epoch": 0.003026863412788498,
      "eval_loss": NaN,
      "eval_runtime": 19.262,
      "eval_samples_per_second": 43.35,
      "eval_steps_per_second": 21.701,
      "step": 6
    },
    {
      "epoch": 0.0035313406482532475,
      "grad_norm": 0.08714216202497482,
      "learning_rate": 0.00014,
      "loss": 1.4391,
      "step": 7
    },
    {
      "epoch": 0.004035817883717997,
      "grad_norm": NaN,
      "learning_rate": 0.00016,
      "loss": 1.3039,
      "step": 8
    },
    {
      "epoch": 0.004540295119182747,
      "grad_norm": NaN,
      "learning_rate": 0.00018,
      "loss": 2.2039,
      "step": 9
    },
    {
      "epoch": 0.004540295119182747,
      "eval_loss": NaN,
      "eval_runtime": 19.2672,
      "eval_samples_per_second": 43.338,
      "eval_steps_per_second": 21.695,
      "step": 9
    },
    {
      "epoch": 0.005044772354647497,
      "grad_norm": 0.09138359874486923,
      "learning_rate": 0.0002,
      "loss": 1.6505,
      "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": 796142988165120.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}