Text Generation
Transformers
Safetensors
English
llama
fireplace
fireplace-2
valiant
valiant-labs
llama-3.1
llama-3.1-instruct
llama-3.1-instruct-8b
llama-3
llama-3-instruct
llama-3-instruct-8b
8b
function-calling
sql
database
data-visualization
matplotlib
json
conversational
chat
instruct
Eval Results
text-generation-inference
Inference Endpoints
model card
Browse files
README.md
CHANGED
@@ -204,6 +204,7 @@ These are supplemental to the existing special tokens used by Llama 3.1, such as
|
|
204 |
|
205 |
The 'request' tokens are used by the user to request a specific type of structured output. They should be appended to the end of the user's message and can be alternated with normal chat responses throughout the conversation.
|
206 |
|
|
|
207 |
## The Model
|
208 |
Fireplace 2 is built on top of Llama 3.1 8b Instruct.
|
209 |
|
@@ -219,6 +220,20 @@ This version of Fireplace 2 uses data from the following datasets:
|
|
219 |
Additional capabilities will be added to future releases.
|
220 |
|
221 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
222 |
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63444f2687964b331809eb55/VCJ8Fmefd8cdVhXSSxJiD.jpeg)
|
223 |
|
224 |
|
@@ -232,16 +247,4 @@ We care about open source.
|
|
232 |
For everyone to use.
|
233 |
|
234 |
We encourage others to finetune further from our models.
|
235 |
-
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
|
236 |
-
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ValiantLabs__Llama3.1-8B-Fireplace2)
|
237 |
-
|
238 |
-
| Metric |Value|
|
239 |
-
|-------------------|----:|
|
240 |
-
|Avg. |18.31|
|
241 |
-
|IFEval (0-Shot) |54.83|
|
242 |
-
|BBH (3-Shot) |24.07|
|
243 |
-
|MATH Lvl 5 (4-Shot)| 5.82|
|
244 |
-
|GPQA (0-shot) | 5.15|
|
245 |
-
|MuSR (0-shot) | 4.38|
|
246 |
-
|MMLU-PRO (5-shot) |15.63|
|
247 |
|
|
|
204 |
|
205 |
The 'request' tokens are used by the user to request a specific type of structured output. They should be appended to the end of the user's message and can be alternated with normal chat responses throughout the conversation.
|
206 |
|
207 |
+
|
208 |
## The Model
|
209 |
Fireplace 2 is built on top of Llama 3.1 8b Instruct.
|
210 |
|
|
|
220 |
Additional capabilities will be added to future releases.
|
221 |
|
222 |
|
223 |
+
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
|
224 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_ValiantLabs__Llama3.1-8B-Fireplace2)
|
225 |
+
|
226 |
+
| Metric |Value|
|
227 |
+
|-------------------|----:|
|
228 |
+
|Avg. |18.31|
|
229 |
+
|IFEval (0-Shot) |54.83|
|
230 |
+
|BBH (3-Shot) |24.07|
|
231 |
+
|MATH Lvl 5 (4-Shot)| 5.82|
|
232 |
+
|GPQA (0-shot) | 5.15|
|
233 |
+
|MuSR (0-shot) | 4.38|
|
234 |
+
|MMLU-PRO (5-shot) |15.63|
|
235 |
+
|
236 |
+
|
237 |
![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/63444f2687964b331809eb55/VCJ8Fmefd8cdVhXSSxJiD.jpeg)
|
238 |
|
239 |
|
|
|
247 |
For everyone to use.
|
248 |
|
249 |
We encourage others to finetune further from our models.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
250 |
|