Update README.md
Browse files
README.md
CHANGED
@@ -208,16 +208,19 @@ model-index:
|
|
208 |
<img src=https://huggingface.co/altomek/YiSM-34B-0rn/resolve/main/YiSM.png>
|
209 |
<a href="https://www.youtube.com/watch?v=a9dNpk9G5h0" title="P.T. Adamczyk - Never Looking Back | Cyberpunk 2077: Phantom Liberty (Original Score)" target="_blank">intro music...</a>
|
210 |
|
|
|
211 |
## YiSM-34B-0rn
|
212 |
|
213 |
This is Yi Self Merged. I wanted model that will follow most instuctions yet preserve its base model nature.
|
214 |
|
|
|
215 |
### Ingridients
|
216 |
|
217 |
- [Yi-1.5-34B-Chat](https://huggingface.co/01-ai/Yi-1.5-34B-Chat)
|
218 |
|
219 |
- [Yi-1.5-34B](https://huggingface.co/01-ai/Yi-1.5-34B-Chat/)
|
220 |
|
|
|
221 |
### Settings
|
222 |
|
223 |
I use max_seq_len 8K with alpha_value 2.65.
|
@@ -279,6 +282,7 @@ SillyTavern presets:
|
|
279 |
}
|
280 |
```
|
281 |
|
|
|
282 |
### Terms and Conditions of Use
|
283 |
|
284 |
The following table outlines the primary characteristics and intended uses of my YiSM-34B-0rn models:
|
@@ -302,6 +306,7 @@ Please remember that all YiSM-34B-0rn models operate under the apache-2.0 licens
|
|
302 |
- [3.2bpw](https://huggingface.co/altomek/YiSM-34B-0rn-3.2bpw-EXL2) -> Fits in 16GB VRAM but not recomended. Performance is significantly degraded in lower quants.
|
303 |
- [measurements](https://huggingface.co/altomek/measurements/resolve/main/YiSM-34B-0rn_measurement.json) --> ExLlamav2 measurments
|
304 |
|
|
|
305 |
### [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
|
306 |
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_altomek__YiSM-34B-0rn)
|
307 |
|
@@ -318,7 +323,8 @@ Detailed results can be found [here](https://huggingface.co/datasets/open-llm-le
|
|
318 |
5th in 34B size range excluding "Private or deleted" or 8th with all models included as of 2024-06-10 ;P
|
319 |
<img src=https://huggingface.co/altomek/YiSM-34B-0rn/resolve/main/5thIn34B.png>
|
320 |
|
321 |
-
|
|
|
322 |
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_altomek__YiSM-34B-0rn)
|
323 |
|
324 |
| Metric |Value|
|
|
|
208 |
<img src=https://huggingface.co/altomek/YiSM-34B-0rn/resolve/main/YiSM.png>
|
209 |
<a href="https://www.youtube.com/watch?v=a9dNpk9G5h0" title="P.T. Adamczyk - Never Looking Back | Cyberpunk 2077: Phantom Liberty (Original Score)" target="_blank">intro music...</a>
|
210 |
|
211 |
+
|
212 |
## YiSM-34B-0rn
|
213 |
|
214 |
This is Yi Self Merged. I wanted model that will follow most instuctions yet preserve its base model nature.
|
215 |
|
216 |
+
|
217 |
### Ingridients
|
218 |
|
219 |
- [Yi-1.5-34B-Chat](https://huggingface.co/01-ai/Yi-1.5-34B-Chat)
|
220 |
|
221 |
- [Yi-1.5-34B](https://huggingface.co/01-ai/Yi-1.5-34B-Chat/)
|
222 |
|
223 |
+
|
224 |
### Settings
|
225 |
|
226 |
I use max_seq_len 8K with alpha_value 2.65.
|
|
|
282 |
}
|
283 |
```
|
284 |
|
285 |
+
|
286 |
### Terms and Conditions of Use
|
287 |
|
288 |
The following table outlines the primary characteristics and intended uses of my YiSM-34B-0rn models:
|
|
|
306 |
- [3.2bpw](https://huggingface.co/altomek/YiSM-34B-0rn-3.2bpw-EXL2) -> Fits in 16GB VRAM but not recomended. Performance is significantly degraded in lower quants.
|
307 |
- [measurements](https://huggingface.co/altomek/measurements/resolve/main/YiSM-34B-0rn_measurement.json) --> ExLlamav2 measurments
|
308 |
|
309 |
+
|
310 |
### [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
|
311 |
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_altomek__YiSM-34B-0rn)
|
312 |
|
|
|
323 |
5th in 34B size range excluding "Private or deleted" or 8th with all models included as of 2024-06-10 ;P
|
324 |
<img src=https://huggingface.co/altomek/YiSM-34B-0rn/resolve/main/5thIn34B.png>
|
325 |
|
326 |
+
|
327 |
+
### [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard)
|
328 |
Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_altomek__YiSM-34B-0rn)
|
329 |
|
330 |
| Metric |Value|
|