froggeric commited on
Commit
8759e31
1 Parent(s): 41ecd6b

Add refusal bypass method, and link to coding leaderboard

Browse files
Files changed (1) hide show
  1. README.md +14 -6
README.md CHANGED
@@ -172,18 +172,25 @@ bypass methods used are below.
172
 
173
  ## Bypassing censorship/refusal
174
 
175
- **Method 1: rewrite the Assistant response, asking for completion**\
176
- By far the best refusal bypass method, is to rewrite the first Assistant response with the beginning of a compliant
 
 
 
 
 
 
177
  reply, and then continue the chat. For example: _"The"_, _"It"_, or _"Step 1:"_. Sometimes it is necessary to add a few more
178
  words either in that first Assistant reply, or by rewriting the second Asssitant reply. Using this method, I have
179
- found that very few models persist in their refusal.
 
180
 
181
- **Method 2: use a system prompt**\
182
  An additional method, less reliable, is to use a system prompt. I have had more success with prompts telling the model
183
  it is a fiction writer, rather than telling it is uncensored or unbiased. Using system prompt for this purpose is a
184
  poor choice, as I think they are better suited to define the writing style.
185
 
186
- **Method 3: use a different prompt format**\
187
  Last method, seldom reliable and often producing lesser quality replies, it to switch to a different prompt format,
188
  such as Alpaca, Vicuna or ChatML.
189
 
@@ -219,4 +226,5 @@ top_p = 0.1
219
  - [Chatbot Arena Leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard)
220
  - [NeoEvalPlusN benchmark](https://huggingface.co/datasets/ChuckMcSneed/NeoEvalPlusN_benchmark)
221
  - [WolframRavenwolf's benchmark](https://huggingface.co/datasets/ChuckMcSneed/WolframRavenwolfs_benchmark_results)
222
- - [Uncensored General Intelligence](https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard)
 
 
172
 
173
  ## Bypassing censorship/refusal
174
 
175
+ **Method 1: rewrite the Assistant response, with the beginning of a compliant response**\
176
+ By far the most successful way to bypass refusal, is to rewrite the first Assistant response with the beginning of a compliant
177
+ response, and then continue the chat, using a simple `Sure `. Don't forget the space, otherwise it is likely to
178
+ complete with something like _"Surely you cannot be asking..."._ This method has the added advantage of not
179
+ introducing user bias in the response.
180
+
181
+ **Method 2: rewrite the Assistant response, asking for completion**\
182
+ Another equally successful bypass method, is to rewrite the first Assistant response with the beginning of a
183
  reply, and then continue the chat. For example: _"The"_, _"It"_, or _"Step 1:"_. Sometimes it is necessary to add a few more
184
  words either in that first Assistant reply, or by rewriting the second Asssitant reply. Using this method, I have
185
+ found that very few models persist in their refusal. This can also be combined with Method 1 in case of particularly
186
+ stubborn refusals.
187
 
188
+ **Method 3: use a system prompt**\
189
  An additional method, less reliable, is to use a system prompt. I have had more success with prompts telling the model
190
  it is a fiction writer, rather than telling it is uncensored or unbiased. Using system prompt for this purpose is a
191
  poor choice, as I think they are better suited to define the writing style.
192
 
193
+ **Method 4: use a different prompt format**\
194
  Last method, seldom reliable and often producing lesser quality replies, it to switch to a different prompt format,
195
  such as Alpaca, Vicuna or ChatML.
196
 
 
226
  - [Chatbot Arena Leaderboard](https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard)
227
  - [NeoEvalPlusN benchmark](https://huggingface.co/datasets/ChuckMcSneed/NeoEvalPlusN_benchmark)
228
  - [WolframRavenwolf's benchmark](https://huggingface.co/datasets/ChuckMcSneed/WolframRavenwolfs_benchmark_results)
229
+ - [Uncensored General Intelligence](https://huggingface.co/spaces/DontPlanToEnd/UGI-Leaderboard)
230
+ - [toqan coding assistant leaderboard](https://prollm.toqan.ai/leaderboard/coding-assistant)