checkpoint-562/trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.997333333333333,
  "eval_steps": 500,
  "global_step": 562,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5333333333333333,
      "grad_norm": 0.271308958530426,
      "learning_rate": 5e-05,
      "loss": 1.8262,
      "step": 100
    },
    {
      "epoch": 0.9973333333333333,
      "eval_accuracy": 0.6047174887892377,
      "eval_loss": 1.7027028799057007,
      "eval_runtime": 6.6338,
      "eval_samples_per_second": 75.372,
      "eval_steps_per_second": 9.497,
      "step": 187
    },
    {
      "epoch": 0.9973333333333333,
      "eval_exact_match": 14.6,
      "eval_f1": 23.665714285714298,
      "eval_qa_bleu": 10.24028390049623,
      "eval_qa_exact_match": 0.114,
      "eval_recite_bleu": 12.525534079250063,
      "eval_recite_exact_match": 0.0,
      "step": 187
    },
    {
      "epoch": 1.0666666666666667,
      "grad_norm": 0.301651269197464,
      "learning_rate": 5e-05,
      "loss": 1.7424,
      "step": 200
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.35949045419692993,
      "learning_rate": 5e-05,
      "loss": 1.6978,
      "step": 300
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6068699551569506,
      "eval_loss": 1.6845285892486572,
      "eval_runtime": 5.7524,
      "eval_samples_per_second": 86.92,
      "eval_steps_per_second": 10.952,
      "step": 375
    },
    {
      "epoch": 2.0,
      "eval_exact_match": 14.6,
      "eval_f1": 23.74190476190477,
      "eval_qa_bleu": 9.70432698897927,
      "eval_qa_exact_match": 0.114,
      "eval_recite_bleu": 13.014822566317982,
      "eval_recite_exact_match": 0.006,
      "step": 375
    },
    {
      "epoch": 2.1333333333333333,
      "grad_norm": 0.41301965713500977,
      "learning_rate": 5e-05,
      "loss": 1.694,
      "step": 400
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.5057485699653625,
      "learning_rate": 5e-05,
      "loss": 1.6324,
      "step": 500
    },
    {
      "epoch": 2.997333333333333,
      "eval_accuracy": 0.6079551569506726,
      "eval_loss": 1.6851012706756592,
      "eval_runtime": 5.5314,
      "eval_samples_per_second": 90.394,
      "eval_steps_per_second": 11.39,
      "step": 562
    },
    {
      "epoch": 2.997333333333333,
      "eval_exact_match": 14.2,
      "eval_f1": 23.29190476190477,
      "eval_qa_bleu": 10.724307586340888,
      "eval_qa_exact_match": 0.104,
      "eval_recite_bleu": 12.83531066630853,
      "eval_recite_exact_match": 0.004,
      "step": 562
    }
  ],
  "logging_steps": 100,
  "max_steps": 9350,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 50,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.665963591807795e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}