last-checkpoint/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.0013619803193843849,
  "eval_steps": 3,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0001361980319384385,
      "grad_norm": 15.731740951538086,
      "learning_rate": 2e-05,
      "loss": 11.4019,
      "step": 1
    },
    {
      "epoch": 0.0001361980319384385,
      "eval_loss": 11.31342601776123,
      "eval_runtime": 606.6334,
      "eval_samples_per_second": 2.548,
      "eval_steps_per_second": 2.548,
      "step": 1
    },
    {
      "epoch": 0.000272396063876877,
      "grad_norm": 13.67992877960205,
      "learning_rate": 4e-05,
      "loss": 11.3802,
      "step": 2
    },
    {
      "epoch": 0.00040859409581531545,
      "grad_norm": 14.387179374694824,
      "learning_rate": 6e-05,
      "loss": 12.3824,
      "step": 3
    },
    {
      "epoch": 0.00040859409581531545,
      "eval_loss": 10.988008499145508,
      "eval_runtime": 610.0431,
      "eval_samples_per_second": 2.534,
      "eval_steps_per_second": 2.534,
      "step": 3
    },
    {
      "epoch": 0.000544792127753754,
      "grad_norm": 16.716053009033203,
      "learning_rate": 8e-05,
      "loss": 11.0426,
      "step": 4
    },
    {
      "epoch": 0.0006809901596921924,
      "grad_norm": 20.072181701660156,
      "learning_rate": 0.0001,
      "loss": 10.3388,
      "step": 5
    },
    {
      "epoch": 0.0008171881916306309,
      "grad_norm": 19.392244338989258,
      "learning_rate": 0.00012,
      "loss": 9.2314,
      "step": 6
    },
    {
      "epoch": 0.0008171881916306309,
      "eval_loss": 7.863962173461914,
      "eval_runtime": 613.1792,
      "eval_samples_per_second": 2.521,
      "eval_steps_per_second": 2.521,
      "step": 6
    },
    {
      "epoch": 0.0009533862235690695,
      "grad_norm": 23.612119674682617,
      "learning_rate": 0.00014,
      "loss": 8.1276,
      "step": 7
    },
    {
      "epoch": 0.001089584255507508,
      "grad_norm": 29.537059783935547,
      "learning_rate": 0.00016,
      "loss": 6.5983,
      "step": 8
    },
    {
      "epoch": 0.0012257822874459463,
      "grad_norm": 21.244483947753906,
      "learning_rate": 0.00018,
      "loss": 4.3311,
      "step": 9
    },
    {
      "epoch": 0.0012257822874459463,
      "eval_loss": 2.8899948596954346,
      "eval_runtime": 608.9823,
      "eval_samples_per_second": 2.539,
      "eval_steps_per_second": 2.539,
      "step": 9
    },
    {
      "epoch": 0.0013619803193843849,
      "grad_norm": 20.49325942993164,
      "learning_rate": 0.0002,
      "loss": 3.2196,
      "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": 6534582019031040.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}