logs/checkpoint-336/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.986666666666667,
  "eval_steps": 500,
  "global_step": 336,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 1.3438981771469116,
      "learning_rate": 0.0001990856842874641,
      "loss": 2.9111,
      "step": 25
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.23749998211860657,
      "learning_rate": 0.00019297764858882514,
      "loss": 1.3264,
      "step": 50
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.2705813944339752,
      "learning_rate": 0.00018146608991420534,
      "loss": 1.2225,
      "step": 75
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.21680688858032227,
      "learning_rate": 0.0001652200182109602,
      "loss": 1.206,
      "step": 100
    },
    {
      "epoch": 0.9955555555555555,
      "eval_loss": 1.3727788925170898,
      "eval_runtime": 13.0744,
      "eval_samples_per_second": 11.473,
      "eval_steps_per_second": 1.453,
      "step": 112
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.25154614448547363,
      "learning_rate": 0.0001451835961144145,
      "loss": 1.1144,
      "step": 125
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.42746973037719727,
      "learning_rate": 0.00012252126764738844,
      "loss": 1.116,
      "step": 150
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.4281701147556305,
      "learning_rate": 9.855008496617327e-05,
      "loss": 1.0742,
      "step": 175
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.521720290184021,
      "learning_rate": 7.466316607649738e-05,
      "loss": 1.0489,
      "step": 200
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3299340307712555,
      "learning_rate": 5.2248731878811365e-05,
      "loss": 0.9471,
      "step": 225
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3316137790679932,
      "eval_runtime": 12.8454,
      "eval_samples_per_second": 11.677,
      "eval_steps_per_second": 1.479,
      "step": 225
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.26803553104400635,
      "learning_rate": 3.2609427815531426e-05,
      "loss": 1.1612,
      "step": 250
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.26870790123939514,
      "learning_rate": 1.6886618852849724e-05,
      "loss": 1.1182,
      "step": 275
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.32614558935165405,
      "learning_rate": 5.994057497592031e-06,
      "loss": 1.0863,
      "step": 300
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.3909718096256256,
      "learning_rate": 5.647798228764156e-07,
      "loss": 1.0418,
      "step": 325
    }
  ],
  "logging_steps": 25,
  "max_steps": 336,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3468147388502016.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}