Transformers
llama
text-generation-inference
rombodawg commited on
Commit
55d7d16
1 Parent(s): 6b3331f

Update README.md

Browse files

Added more prompt templates to readme

Files changed (1) hide show
  1. README.md +41 -1
README.md CHANGED
@@ -156,6 +156,10 @@ This is one of the first models trained on the LosslessMegaCodeTrainingV2_1m_Evo
156
 
157
  - This model was made as a colaboration between me and andreaskoepf who is an affiliate of Open Assistant.
158
 
 
 
 
 
159
  Prompt template:
160
 
161
  - chatml format is used: "<|im_start|>system\n{system message}<|im_end|>\n<|im_start|>user\n{user prompt}<|im_end|>\n<|im_start|>assistant\n{Assistant answer}<|im_end|>\n"
@@ -169,6 +173,42 @@ multi-line:
169
  <|im_start|>assistant
170
  {Assistant answer}<|im_end|>
171
  ```
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
172
  Training data:
173
 
174
  - https://wandb.ai/open-assistant/epfl-mt-sft/runs/run34_megacode2_min100_13b
@@ -183,4 +223,4 @@ Link for the filtered dataset used to make this model are bellow:
183
 
184
  The original posting for this model was uploaded at the link bellow.
185
 
186
- - https://huggingface.co/andreaskoepf/llama2-13b-megacode2_min100
 
156
 
157
  - This model was made as a colaboration between me and andreaskoepf who is an affiliate of Open Assistant.
158
 
159
+ This Model score .29 on humaneval+ the same as LLaMA-2 70B Chat Link bellow (in this benchmark the model is called andreaskoepf/llama2-13b-megacode2_min100)
160
+
161
+ - https://tju01.github.io/FastEval-OpenAssistant/
162
+
163
  Prompt template:
164
 
165
  - chatml format is used: "<|im_start|>system\n{system message}<|im_end|>\n<|im_start|>user\n{user prompt}<|im_end|>\n<|im_start|>assistant\n{Assistant answer}<|im_end|>\n"
 
173
  <|im_start|>assistant
174
  {Assistant answer}<|im_end|>
175
  ```
176
+ Gpt4all template:
177
+
178
+ - System prompt
179
+ ```
180
+ <|im_start|>system
181
+ {system message}
182
+ ```
183
+ - Prompt template
184
+ ```
185
+ <|im_end|>
186
+ <|im_start|>user
187
+ %1<|im_end|>
188
+ <|im_start|>assistant
189
+ ```
190
+
191
+ Oobagooba Text-Generation-Webui Template
192
+ - user:
193
+ ```
194
+ <|im_start|>user
195
+ {User string}<|im_end|>
196
+ ```
197
+ - bot:
198
+ ```
199
+ <|im_start|>assistant
200
+ {Bot string}<|im_end|>
201
+ ```
202
+ - turn_template:
203
+ ```
204
+ <|user|>\n<|user-message|>\n\n<|bot|>\n<|bot-message|>\n\n
205
+ ```
206
+ - context:
207
+ ```
208
+ <|im_start|>system
209
+ Below is an instruction that describes a task. Write a response that appropriately completes the request.<|im_end|>
210
+ ```
211
+
212
  Training data:
213
 
214
  - https://wandb.ai/open-assistant/epfl-mt-sft/runs/run34_megacode2_min100_13b
 
223
 
224
  The original posting for this model was uploaded at the link bellow.
225
 
226
+ - https://huggingface.co/andreaskoepf/llama2-13b-megacode2_min100