Commit
•
ad5e5fd
1
Parent(s):
0c55dc1
Update README.md
Browse files
README.md
CHANGED
@@ -135,10 +135,26 @@ model-index:
|
|
135 |
|
136 |
This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-70B-Instruct` model.
|
137 |
|
138 |
-
# Quantized GGUF
|
139 |
|
140 |
All GGUF models are available here: [MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.2-GGUF](https://huggingface.co/MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.2-GGUF)
|
141 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
142 |
# Prompt Template
|
143 |
|
144 |
This model uses `ChatML` prompt template:
|
@@ -218,16 +234,3 @@ outputs = pipeline(
|
|
218 |
)
|
219 |
print(outputs[0]["generated_text"][len(prompt):])
|
220 |
```
|
221 |
-
# [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
222 |
-
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-70B-Instruct-DPO-v0.2)
|
223 |
-
|
224 |
-
| Metric |Value|
|
225 |
-
|---------------------------------|----:|
|
226 |
-
|Avg. |78.96|
|
227 |
-
|AI2 Reasoning Challenge (25-Shot)|72.53|
|
228 |
-
|HellaSwag (10-Shot) |86.22|
|
229 |
-
|MMLU (5-Shot) |80.41|
|
230 |
-
|TruthfulQA (0-shot) |63.57|
|
231 |
-
|Winogrande (5-shot) |82.79|
|
232 |
-
|GSM8k (5-shot) |88.25|
|
233 |
-
|
|
|
135 |
|
136 |
This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-70B-Instruct` model.
|
137 |
|
138 |
+
# ⚡ Quantized GGUF
|
139 |
|
140 |
All GGUF models are available here: [MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.2-GGUF](https://huggingface.co/MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.2-GGUF)
|
141 |
|
142 |
+
# 🏆 [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
|
143 |
+
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-70B-Instruct-DPO-v0.2)
|
144 |
+
|
145 |
+
| Metric |Value|
|
146 |
+
|---------------------------------|----:|
|
147 |
+
|Avg. |78.96|
|
148 |
+
|AI2 Reasoning Challenge (25-Shot)|72.53|
|
149 |
+
|HellaSwag (10-Shot) |86.22|
|
150 |
+
|MMLU (5-Shot) |80.41|
|
151 |
+
|TruthfulQA (0-shot) |63.57|
|
152 |
+
|Winogrande (5-shot) |82.79|
|
153 |
+
|GSM8k (5-shot) |88.25|
|
154 |
+
|
155 |
+
**Top 10 models on the Leaderboard**
|
156 |
+
<img src="./llama-3-70b-top-10.png" alt="Llama-3-70B finet-tuned models" width="500" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
|
157 |
+
|
158 |
# Prompt Template
|
159 |
|
160 |
This model uses `ChatML` prompt template:
|
|
|
234 |
)
|
235 |
print(outputs[0]["generated_text"][len(prompt):])
|
236 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|