MaziyarPanahi commited on
Commit
2fc0f91
1 Parent(s): 9452df4

Update README.md (#9)

Browse files

- Update README.md (25de977fe6a83c0edb7ad167dc3db60d516b30f5)

Files changed (1) hide show
  1. README.md +18 -28
README.md CHANGED
@@ -134,10 +134,27 @@ model-index:
134
 
135
  This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-70B-Instruct` model.
136
 
137
- # Quantized GGUF
138
 
139
  All GGUF models are available here: [MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.4-GGUF](https://huggingface.co/MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.4-GGUF)
140
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
141
  # Prompt Template
142
 
143
  This model uses `ChatML` prompt template:
@@ -259,30 +276,3 @@ Here are the pros and cons of the Docker system:
259
 
260
  Overall, Docker provides a powerful and flexible way to deploy and manage applications, but it requires careful planning, configuration, and management to ensure optimal performance and security.
261
  ```
262
- # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
263
- Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-70B-Instruct-DPO-v0.4)
264
-
265
- | Metric |Value|
266
- |---------------------------------|----:|
267
- |Avg. |78.89|
268
- |AI2 Reasoning Challenge (25-Shot)|72.61|
269
- |HellaSwag (10-Shot) |86.03|
270
- |MMLU (5-Shot) |80.50|
271
- |TruthfulQA (0-shot) |63.26|
272
- |Winogrande (5-shot) |83.58|
273
- |GSM8k (5-shot) |87.34|
274
-
275
-
276
- # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
277
- Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-70B-Instruct-DPO-v0.4)
278
-
279
- | Metric |Value|
280
- |---------------------------------|----:|
281
- |Avg. |78.89|
282
- |AI2 Reasoning Challenge (25-Shot)|72.61|
283
- |HellaSwag (10-Shot) |86.03|
284
- |MMLU (5-Shot) |80.50|
285
- |TruthfulQA (0-shot) |63.26|
286
- |Winogrande (5-shot) |83.58|
287
- |GSM8k (5-shot) |87.34|
288
-
 
134
 
135
  This model is a fine-tune (DPO) of `meta-llama/Meta-Llama-3-70B-Instruct` model.
136
 
137
+ # Quantized GGUF
138
 
139
  All GGUF models are available here: [MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.4-GGUF](https://huggingface.co/MaziyarPanahi/Llama-3-70B-Instruct-DPO-v0.4-GGUF)
140
 
141
+ # 🏆 [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)
142
+ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_MaziyarPanahi__Llama-3-70B-Instruct-DPO-v0.4)
143
+
144
+ | Metric |Value|
145
+ |---------------------------------|----:|
146
+ |Avg. |78.89|
147
+ |AI2 Reasoning Challenge (25-Shot)|72.61|
148
+ |HellaSwag (10-Shot) |86.03|
149
+ |MMLU (5-Shot) |80.50|
150
+ |TruthfulQA (0-shot) |63.26|
151
+ |Winogrande (5-shot) |83.58|
152
+ |GSM8k (5-shot) |87.34|
153
+
154
+ **Top 10 models on the Leaderboard**
155
+ <img src="./llama-3-70b-top-10.png" alt="Llama-3-70B finet-tuned models" style="margin-left:'auto' margin-right:'auto' display:'block'"/>
156
+
157
+
158
  # Prompt Template
159
 
160
  This model uses `ChatML` prompt template:
 
276
 
277
  Overall, Docker provides a powerful and flexible way to deploy and manage applications, but it requires careful planning, configuration, and management to ensure optimal performance and security.
278
  ```