metadata
dataset_info:
features:
- name: prompt ID
dtype: int64
- name: language
dtype: string
- name: topic
dtype: string
- name: labels
sequence: string
- name: original prompt
dtype: string
- name: translated prompt
dtype: string
- name: 'answer: Qwen/Qwen1.5-32B - original'
dtype: 'null'
- name: 'answer: Qwen/Qwen1.5-32B - translated'
dtype: string
- name: pos priming phrase 0 - original
dtype: string
- name: 'pos priming phrase 0, answer: Yes - original'
dtype: string
- name: pos priming phrase 0, answer Yes logprob - original
dtype: float64
- name: 'pos priming phrase 0, answer: No - original'
dtype: string
- name: pos priming phrase 0, answer No logprob - original
dtype: float64
- name: pos priming phrase 1 - original
dtype: string
- name: 'pos priming phrase 1, answer: Yes - original'
dtype: string
- name: pos priming phrase 1, answer Yes logprob - original
dtype: float64
- name: 'pos priming phrase 1, answer: No - original'
dtype: string
- name: pos priming phrase 1, answer No logprob - original
dtype: float64
- name: pos priming phrase 2 - original
dtype: string
- name: 'pos priming phrase 2, answer: Yes - original'
dtype: string
- name: pos priming phrase 2, answer Yes logprob - original
dtype: float64
- name: 'pos priming phrase 2, answer: No - original'
dtype: string
- name: pos priming phrase 2, answer No logprob - original
dtype: float64
- name: neg priming phrase 0 - original
dtype: string
- name: 'neg priming phrase 0, answer: Yes - original'
dtype: string
- name: neg priming phrase 0, answer Yes logprob - original
dtype: float64
- name: 'neg priming phrase 0, answer: No - original'
dtype: string
- name: neg priming phrase 0, answer No logprob - original
dtype: float64
- name: neg priming phrase 1 - original
dtype: string
- name: 'neg priming phrase 1, answer: Yes - original'
dtype: string
- name: neg priming phrase 1, answer Yes logprob - original
dtype: float64
- name: 'neg priming phrase 1, answer: No - original'
dtype: string
- name: neg priming phrase 1, answer No logprob - original
dtype: float64
- name: neg priming phrase 2 - original
dtype: string
- name: 'neg priming phrase 2, answer: Yes - original'
dtype: string
- name: neg priming phrase 2, answer Yes logprob - original
dtype: float64
- name: 'neg priming phrase 2, answer: No - original'
dtype: string
- name: neg priming phrase 2, answer No logprob - original
dtype: float64
- name: pos priming phrase 0 - translated
dtype: string
- name: 'pos priming phrase 0, answer: Yes - translated'
dtype: string
- name: pos priming phrase 0, answer Yes logprob - translated
dtype: float64
- name: 'pos priming phrase 0, answer: No - translated'
dtype: string
- name: pos priming phrase 0, answer No logprob - translated
dtype: float64
- name: pos priming phrase 1 - translated
dtype: string
- name: 'pos priming phrase 1, answer: Yes - translated'
dtype: string
- name: pos priming phrase 1, answer Yes logprob - translated
dtype: float64
- name: 'pos priming phrase 1, answer: No - translated'
dtype: string
- name: pos priming phrase 1, answer No logprob - translated
dtype: float64
- name: pos priming phrase 2 - translated
dtype: string
- name: 'pos priming phrase 2, answer: Yes - translated'
dtype: string
- name: pos priming phrase 2, answer Yes logprob - translated
dtype: float64
- name: 'pos priming phrase 2, answer: No - translated'
dtype: string
- name: pos priming phrase 2, answer No logprob - translated
dtype: float64
- name: neg priming phrase 0 - translated
dtype: string
- name: 'neg priming phrase 0, answer: Yes - translated'
dtype: string
- name: neg priming phrase 0, answer Yes logprob - translated
dtype: float64
- name: 'neg priming phrase 0, answer: No - translated'
dtype: string
- name: neg priming phrase 0, answer No logprob - translated
dtype: float64
- name: neg priming phrase 1 - translated
dtype: string
- name: 'neg priming phrase 1, answer: Yes - translated'
dtype: string
- name: neg priming phrase 1, answer Yes logprob - translated
dtype: float64
- name: 'neg priming phrase 1, answer: No - translated'
dtype: string
- name: neg priming phrase 1, answer No logprob - translated
dtype: float64
- name: neg priming phrase 2 - translated
dtype: string
- name: 'neg priming phrase 2, answer: Yes - translated'
dtype: string
- name: neg priming phrase 2, answer Yes logprob - translated
dtype: float64
- name: 'neg priming phrase 2, answer: No - translated'
dtype: string
- name: neg priming phrase 2, answer No logprob - translated
dtype: float64
splits:
- name: self_response_Qwen_cont_Qwen1.5_32B_answers
num_bytes: 387252
num_examples: 286
download_size: 178904
dataset_size: 387252
configs:
- config_name: default
data_files:
- split: self_response_Qwen_cont_Qwen1.5_32B_answers
path: data/self_response_Qwen_cont_Qwen1.5_32B_answers-*