last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.6348773841961854,
  "eval_steps": 500,
  "global_step": 600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1362397820163488,
      "grad_norm": 0.784312903881073,
      "learning_rate": 0.0004,
      "loss": 1.3927,
      "step": 50
    },
    {
      "epoch": 0.2724795640326976,
      "grad_norm": 0.5759815573692322,
      "learning_rate": 0.0004,
      "loss": 0.8469,
      "step": 100
    },
    {
      "epoch": 0.4087193460490463,
      "grad_norm": 0.7147420048713684,
      "learning_rate": 0.0004,
      "loss": 0.6428,
      "step": 150
    },
    {
      "epoch": 0.5449591280653951,
      "grad_norm": 1.0268043279647827,
      "learning_rate": 0.0004,
      "loss": 0.5406,
      "step": 200
    },
    {
      "epoch": 0.6811989100817438,
      "grad_norm": 0.5647836923599243,
      "learning_rate": 0.0004,
      "loss": 0.4279,
      "step": 250
    },
    {
      "epoch": 0.8174386920980926,
      "grad_norm": 0.48977088928222656,
      "learning_rate": 0.0004,
      "loss": 0.3726,
      "step": 300
    },
    {
      "epoch": 0.9536784741144414,
      "grad_norm": 0.3192068636417389,
      "learning_rate": 0.0004,
      "loss": 0.3033,
      "step": 350
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.2556329667568207,
      "eval_runtime": 4.2889,
      "eval_samples_per_second": 72.046,
      "eval_steps_per_second": 18.186,
      "step": 367
    },
    {
      "epoch": 1.0899182561307903,
      "grad_norm": 0.2055899053812027,
      "learning_rate": 0.0004,
      "loss": 0.2806,
      "step": 400
    },
    {
      "epoch": 1.226158038147139,
      "grad_norm": 0.28835996985435486,
      "learning_rate": 0.0004,
      "loss": 0.2058,
      "step": 450
    },
    {
      "epoch": 1.3623978201634879,
      "grad_norm": 0.30920594930648804,
      "learning_rate": 0.0004,
      "loss": 0.1817,
      "step": 500
    },
    {
      "epoch": 1.4986376021798364,
      "grad_norm": 0.18480294942855835,
      "learning_rate": 0.0004,
      "loss": 0.1557,
      "step": 550
    },
    {
      "epoch": 1.6348773841961854,
      "grad_norm": 0.6040850877761841,
      "learning_rate": 0.0004,
      "loss": 0.1337,
      "step": 600
    },
    {
      "epoch": 1.6348773841961854,
      "eval_loss": 0.14192721247673035,
      "eval_runtime": 4.3794,
      "eval_samples_per_second": 70.557,
      "eval_steps_per_second": 17.811,
      "step": 600
    }
  ],
  "logging_steps": 50,
  "max_steps": 600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.945276023865344e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}