{
  "best_metric": 1.0689845085144043,
  "best_model_checkpoint": "./llama3/26-07-24-Weni-WeniGPT-Agents-Llama3-5.1.24-SFT_Experiment with DPO and Llama3 70b-2_max_steps-32_batch_8_2024-07-26/checkpoint-8",
  "epoch": 7.111111111111111,
  "eval_steps": 2,
  "global_step": 8,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.7777777777777777,
      "eval_bleu": 0.02170523150063329,
      "eval_loss": 1.5546669960021973,
      "eval_rouge1": 0.5658290942329046,
      "eval_rouge2": 0.371216207944968,
      "eval_rougeL": 0.5544380522179795,
      "eval_rougeLsum": 0.557866697701094,
      "eval_runtime": 8.903,
      "eval_samples_per_second": 0.899,
      "eval_steps_per_second": 0.112,
      "step": 2
    },
    {
      "epoch": 3.5555555555555554,
      "eval_bleu": 0.016226404911743514,
      "eval_loss": 1.2373424768447876,
      "eval_rouge1": 0.1880159013464857,
      "eval_rouge2": 0.110501293225719,
      "eval_rougeL": 0.16774584781013288,
      "eval_rougeLsum": 0.18374679825109685,
      "eval_runtime": 7.2666,
      "eval_samples_per_second": 1.101,
      "eval_steps_per_second": 0.138,
      "step": 4
    },
    {
      "epoch": 5.333333333333333,
      "eval_bleu": 0.023706516954741297,
      "eval_loss": 1.123119592666626,
      "eval_rouge1": 0.6910290907363565,
      "eval_rouge2": 0.4711964975736579,
      "eval_rougeL": 0.673625245352441,
      "eval_rougeLsum": 0.6773221910352027,
      "eval_runtime": 5.1525,
      "eval_samples_per_second": 1.553,
      "eval_steps_per_second": 0.194,
      "step": 6
    },
    {
      "epoch": 7.111111111111111,
      "eval_bleu": 0.02402168983627918,
      "eval_loss": 1.0689845085144043,
      "eval_rouge1": 0.6850615792026293,
      "eval_rouge2": 0.4793755526234388,
      "eval_rougeL": 0.6702061146380454,
      "eval_rougeLsum": 0.6743158691399876,
      "eval_runtime": 3.5693,
      "eval_samples_per_second": 2.241,
      "eval_steps_per_second": 0.28,
      "step": 8
    }
  ],
  "logging_steps": 10,
  "max_steps": 32,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 32,
  "save_steps": 2,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5712648339456.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}