trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.03333333333333333,
      "grad_norm": 1.2589778900146484,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.1573,
      "mean_token_accuracy": 0.7906571626663208,
      "step": 1
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 0.6721312403678894,
      "learning_rate": 0.0001666666666666667,
      "loss": 1.0823,
      "mean_token_accuracy": 0.7963626533746719,
      "step": 5
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.4220508933067322,
      "learning_rate": 0.00019730448705798239,
      "loss": 0.6866,
      "mean_token_accuracy": 0.8463268637657165,
      "step": 10
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.3032156527042389,
      "learning_rate": 0.00018660254037844388,
      "loss": 0.5073,
      "mean_token_accuracy": 0.8762549757957458,
      "step": 15
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.28760814666748047,
      "learning_rate": 0.0001686241637868734,
      "loss": 0.4286,
      "mean_token_accuracy": 0.8931135952472686,
      "step": 20
    },
    {
      "epoch": 0.8333333333333334,
      "grad_norm": 0.44189342856407166,
      "learning_rate": 0.00014487991802004623,
      "loss": 0.369,
      "mean_token_accuracy": 0.9084921300411224,
      "step": 25
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.25721806287765503,
      "learning_rate": 0.00011736481776669306,
      "loss": 0.3024,
      "mean_token_accuracy": 0.9252148032188415,
      "step": 30
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.3562483489513397,
      "eval_mean_token_accuracy": 0.9118186314900716,
      "eval_runtime": 62.9293,
      "eval_samples_per_second": 1.907,
      "eval_steps_per_second": 0.238,
      "step": 30
    },
    {
      "epoch": 1.1666666666666667,
      "grad_norm": 0.18542297184467316,
      "learning_rate": 8.839070858747697e-05,
      "loss": 0.3044,
      "mean_token_accuracy": 0.9222678005695343,
      "step": 35
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.15972654521465302,
      "learning_rate": 6.039202339608432e-05,
      "loss": 0.2852,
      "mean_token_accuracy": 0.9260704517364502,
      "step": 40
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.21341268718242645,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.2792,
      "mean_token_accuracy": 0.9288078784942627,
      "step": 45
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.1644662618637085,
      "learning_rate": 1.6451218858706374e-05,
      "loss": 0.3164,
      "mean_token_accuracy": 0.9177724421024323,
      "step": 50
    },
    {
      "epoch": 1.8333333333333335,
      "grad_norm": 0.17668558657169342,
      "learning_rate": 4.20104876845111e-06,
      "loss": 0.3048,
      "mean_token_accuracy": 0.9214222669601441,
      "step": 55
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.22243957221508026,
      "learning_rate": 0.0,
      "loss": 0.2748,
      "mean_token_accuracy": 0.9267385125160217,
      "step": 60
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.344578355550766,
      "eval_mean_token_accuracy": 0.9140905062357585,
      "eval_runtime": 62.968,
      "eval_samples_per_second": 1.906,
      "eval_steps_per_second": 0.238,
      "step": 60
    },
    {
      "epoch": 2.0,
      "step": 60,
      "total_flos": 4.166042403838362e+16,
      "train_loss": 0.4296759645144145,
      "train_runtime": 1003.3823,
      "train_samples_per_second": 0.472,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 5,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 4.166042403838362e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}