checkpoint-3/trainer_state.json

{
  "best_metric": 0.6624308228492737,
  "best_model_checkpoint": "./llama3/17-07-24-Weni-pipeline_test_Experiment with DPO and Llama3 8B, 1.0.2.-2_max_steps-37_batch_8_2024-07-17/checkpoint-3",
  "epoch": 0.08,
  "eval_steps": 1,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02666666666666667,
      "eval_logits/chosen": -0.40656229853630066,
      "eval_logits/rejected": -0.19044353067874908,
      "eval_logps/chosen": -149.38897705078125,
      "eval_logps/rejected": -200.58956909179688,
      "eval_loss": 0.6931471824645996,
      "eval_rewards/accuracies": 0.0,
      "eval_rewards/chosen": 0.0,
      "eval_rewards/margins": 0.0,
      "eval_rewards/rejected": 0.0,
      "eval_runtime": 340.4707,
      "eval_samples_per_second": 0.1,
      "eval_steps_per_second": 0.1,
      "step": 1
    },
    {
      "epoch": 0.05333333333333334,
      "eval_logits/chosen": -0.4069802761077881,
      "eval_logits/rejected": -0.1900046020746231,
      "eval_logps/chosen": -149.28421020507812,
      "eval_logps/rejected": -200.71835327148438,
      "eval_loss": 0.6726033687591553,
      "eval_rewards/accuracies": 0.6176470518112183,
      "eval_rewards/chosen": 0.02095092460513115,
      "eval_rewards/margins": 0.04670356214046478,
      "eval_rewards/rejected": -0.025752641260623932,
      "eval_runtime": 349.5604,
      "eval_samples_per_second": 0.097,
      "eval_steps_per_second": 0.097,
      "step": 2
    },
    {
      "epoch": 0.08,
      "eval_logits/chosen": -0.4091716408729553,
      "eval_logits/rejected": -0.1912209391593933,
      "eval_logps/chosen": -149.259765625,
      "eval_logps/rejected": -200.7961883544922,
      "eval_loss": 0.6624308228492737,
      "eval_rewards/accuracies": 0.7058823704719543,
      "eval_rewards/chosen": 0.025843007490038872,
      "eval_rewards/margins": 0.06716245412826538,
      "eval_rewards/rejected": -0.04131945222616196,
      "eval_runtime": 335.7836,
      "eval_samples_per_second": 0.101,
      "eval_steps_per_second": 0.101,
      "step": 3
    }
  ],
  "logging_steps": 10,
  "max_steps": 37,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 1,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}