last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9302325581395349,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09302325581395349,
      "grad_norm": 9.457008361816406,
      "learning_rate": 2e-05,
      "loss": 3.2681,
      "step": 1
    },
    {
      "epoch": 0.09302325581395349,
      "eval_loss": 4.726832389831543,
      "eval_runtime": 0.8745,
      "eval_samples_per_second": 5.717,
      "eval_steps_per_second": 3.43,
      "step": 1
    },
    {
      "epoch": 0.18604651162790697,
      "grad_norm": 9.067631721496582,
      "learning_rate": 4e-05,
      "loss": 3.4566,
      "step": 2
    },
    {
      "epoch": 0.27906976744186046,
      "grad_norm": 14.187406539916992,
      "learning_rate": 6e-05,
      "loss": 4.1425,
      "step": 3
    },
    {
      "epoch": 0.27906976744186046,
      "eval_loss": 4.624197959899902,
      "eval_runtime": 0.0994,
      "eval_samples_per_second": 50.287,
      "eval_steps_per_second": 30.172,
      "step": 3
    },
    {
      "epoch": 0.37209302325581395,
      "grad_norm": 7.558735370635986,
      "learning_rate": 8e-05,
      "loss": 2.9593,
      "step": 4
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 10.164987564086914,
      "learning_rate": 0.0001,
      "loss": 4.0569,
      "step": 5
    },
    {
      "epoch": 0.5581395348837209,
      "grad_norm": 4.438661575317383,
      "learning_rate": 0.00012,
      "loss": 3.4217,
      "step": 6
    },
    {
      "epoch": 0.5581395348837209,
      "eval_loss": 3.70611834526062,
      "eval_runtime": 0.1013,
      "eval_samples_per_second": 49.356,
      "eval_steps_per_second": 29.614,
      "step": 6
    },
    {
      "epoch": 0.6511627906976745,
      "grad_norm": 3.077719211578369,
      "learning_rate": 0.00014,
      "loss": 2.9391,
      "step": 7
    },
    {
      "epoch": 0.7441860465116279,
      "grad_norm": 3.016050338745117,
      "learning_rate": 0.00016,
      "loss": 3.0178,
      "step": 8
    },
    {
      "epoch": 0.8372093023255814,
      "grad_norm": 2.7069358825683594,
      "learning_rate": 0.00018,
      "loss": 2.5267,
      "step": 9
    },
    {
      "epoch": 0.8372093023255814,
      "eval_loss": 3.011277914047241,
      "eval_runtime": 0.1011,
      "eval_samples_per_second": 49.447,
      "eval_steps_per_second": 29.668,
      "step": 9
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 2.937380313873291,
      "learning_rate": 0.0002,
      "loss": 3.1368,
      "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": 76723802603520.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}