trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4198895027624308,
  "eval_steps": 22,
  "global_step": 220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.011049723756906077,
      "grad_norm": 0.24585983157157898,
      "learning_rate": 0.0005,
      "loss": 0.7653,
      "step": 1
    },
    {
      "epoch": 0.2430939226519337,
      "grad_norm": 0.32683706283569336,
      "learning_rate": 0.0005,
      "loss": 0.6442,
      "step": 22
    },
    {
      "epoch": 0.2430939226519337,
      "eval_loss": 0.6002205014228821,
      "eval_runtime": 264.7454,
      "eval_samples_per_second": 0.317,
      "eval_steps_per_second": 0.317,
      "step": 22
    },
    {
      "epoch": 0.4861878453038674,
      "grad_norm": 0.2470276653766632,
      "learning_rate": 0.0005,
      "loss": 0.5889,
      "step": 44
    },
    {
      "epoch": 0.4861878453038674,
      "eval_loss": 0.5837110877037048,
      "eval_runtime": 264.3289,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 44
    },
    {
      "epoch": 0.7292817679558011,
      "grad_norm": 0.2570318579673767,
      "learning_rate": 0.0005,
      "loss": 0.5855,
      "step": 66
    },
    {
      "epoch": 0.7292817679558011,
      "eval_loss": 0.5752720832824707,
      "eval_runtime": 264.2764,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 66
    },
    {
      "epoch": 0.9723756906077348,
      "grad_norm": 0.25765955448150635,
      "learning_rate": 0.0005,
      "loss": 0.5533,
      "step": 88
    },
    {
      "epoch": 0.9723756906077348,
      "eval_loss": 0.5659109950065613,
      "eval_runtime": 264.6141,
      "eval_samples_per_second": 0.317,
      "eval_steps_per_second": 0.317,
      "step": 88
    },
    {
      "epoch": 1.2099447513812154,
      "grad_norm": 0.36834636330604553,
      "learning_rate": 0.0005,
      "loss": 0.4701,
      "step": 110
    },
    {
      "epoch": 1.2099447513812154,
      "eval_loss": 0.5801011919975281,
      "eval_runtime": 264.7721,
      "eval_samples_per_second": 0.317,
      "eval_steps_per_second": 0.317,
      "step": 110
    },
    {
      "epoch": 1.4530386740331491,
      "grad_norm": 0.32389357686042786,
      "learning_rate": 0.0005,
      "loss": 0.4702,
      "step": 132
    },
    {
      "epoch": 1.4530386740331491,
      "eval_loss": 0.5771863460540771,
      "eval_runtime": 264.2468,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 132
    },
    {
      "epoch": 1.6961325966850829,
      "grad_norm": 0.3019094169139862,
      "learning_rate": 0.0005,
      "loss": 0.4717,
      "step": 154
    },
    {
      "epoch": 1.6961325966850829,
      "eval_loss": 0.5790910720825195,
      "eval_runtime": 264.4069,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 154
    },
    {
      "epoch": 1.9392265193370166,
      "grad_norm": 0.32115763425827026,
      "learning_rate": 0.0005,
      "loss": 0.4667,
      "step": 176
    },
    {
      "epoch": 1.9392265193370166,
      "eval_loss": 0.585096001625061,
      "eval_runtime": 264.4847,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 176
    },
    {
      "epoch": 2.1767955801104972,
      "grad_norm": 0.42335832118988037,
      "learning_rate": 0.0005,
      "loss": 0.3635,
      "step": 198
    },
    {
      "epoch": 2.1767955801104972,
      "eval_loss": 0.6495711207389832,
      "eval_runtime": 264.3041,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 198
    },
    {
      "epoch": 2.4198895027624308,
      "grad_norm": 0.36611926555633545,
      "learning_rate": 0.0005,
      "loss": 0.3412,
      "step": 220
    },
    {
      "epoch": 2.4198895027624308,
      "eval_loss": 0.6240765452384949,
      "eval_runtime": 264.2788,
      "eval_samples_per_second": 0.318,
      "eval_steps_per_second": 0.318,
      "step": 220
    }
  ],
  "logging_steps": 22,
  "max_steps": 270,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 22,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.767511683803668e+17,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}