trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9171270718232045,
  "eval_steps": 500,
  "global_step": 99,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2946593001841621,
      "grad_norm": 2.4692469653351874,
      "learning_rate": 5e-06,
      "loss": 0.8598,
      "step": 10
    },
    {
      "epoch": 0.5893186003683242,
      "grad_norm": 1.1829109342265023,
      "learning_rate": 5e-06,
      "loss": 0.7522,
      "step": 20
    },
    {
      "epoch": 0.8839779005524862,
      "grad_norm": 2.3810607640229704,
      "learning_rate": 5e-06,
      "loss": 0.7103,
      "step": 30
    },
    {
      "epoch": 0.9723756906077348,
      "eval_loss": 0.6921380162239075,
      "eval_runtime": 105.6301,
      "eval_samples_per_second": 8.653,
      "eval_steps_per_second": 0.275,
      "step": 33
    },
    {
      "epoch": 1.1786372007366483,
      "grad_norm": 2.070195253913277,
      "learning_rate": 5e-06,
      "loss": 0.6749,
      "step": 40
    },
    {
      "epoch": 1.4732965009208103,
      "grad_norm": 1.189689433003807,
      "learning_rate": 5e-06,
      "loss": 0.6447,
      "step": 50
    },
    {
      "epoch": 1.7679558011049723,
      "grad_norm": 1.2811868301588352,
      "learning_rate": 5e-06,
      "loss": 0.6364,
      "step": 60
    },
    {
      "epoch": 1.9742173112338857,
      "eval_loss": 0.6614336371421814,
      "eval_runtime": 105.345,
      "eval_samples_per_second": 8.676,
      "eval_steps_per_second": 0.275,
      "step": 67
    },
    {
      "epoch": 2.0626151012891345,
      "grad_norm": 0.796453744547378,
      "learning_rate": 5e-06,
      "loss": 0.6182,
      "step": 70
    },
    {
      "epoch": 2.3572744014732967,
      "grad_norm": 2.537451226370363,
      "learning_rate": 5e-06,
      "loss": 0.5796,
      "step": 80
    },
    {
      "epoch": 2.6519337016574585,
      "grad_norm": 0.9119097844237827,
      "learning_rate": 5e-06,
      "loss": 0.5746,
      "step": 90
    },
    {
      "epoch": 2.9171270718232045,
      "step": 99,
      "total_flos": 165723997470720.0,
      "train_loss": 0.0,
      "train_runtime": 2.9957,
      "train_samples_per_second": 17384.187,
      "train_steps_per_second": 33.048
    }
  ],
  "logging_steps": 10,
  "max_steps": 99,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 165723997470720.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}