Update README.md
Browse files
README.md
CHANGED
|
@@ -61,6 +61,7 @@ Weaponized blight.
|
|
| 61 |
---
|
| 62 |
|
| 63 |
### TL;DR
|
|
|
|
| 64 |
- **High IFeval** for an 8B model that is not too censored: **74.30**.
|
| 65 |
- **Strong Roleplay** internet RP format lovers will appriciate it, medium size paragraphs (as requested by some people).
|
| 66 |
- **Very coherent** in long context thanks to llama 3.1 models.
|
|
@@ -246,6 +247,21 @@ do_sample: True
|
|
| 246 |
|MuSR (0-shot) |10.89|
|
| 247 |
|MMLU-PRO (5-shot) |29.32|
|
| 248 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 249 |
---
|
| 250 |
|
| 251 |
## Other stuff
|
|
|
|
| 61 |
---
|
| 62 |
|
| 63 |
### TL;DR
|
| 64 |
+
- **Highest rated 8B model** according to a closed external benchmark. See details at the buttom of the page.
|
| 65 |
- **High IFeval** for an 8B model that is not too censored: **74.30**.
|
| 66 |
- **Strong Roleplay** internet RP format lovers will appriciate it, medium size paragraphs (as requested by some people).
|
| 67 |
- **Very coherent** in long context thanks to llama 3.1 models.
|
|
|
|
| 247 |
|MuSR (0-shot) |10.89|
|
| 248 |
|MMLU-PRO (5-shot) |29.32|
|
| 249 |
|
| 250 |
+
---
|
| 251 |
+
|
| 252 |
+
|
| 253 |
+
# Additional benchmarks
|
| 254 |
+
|
| 255 |
+
On the **17th of February, 2025**, I became aware that the model was ranked as the **1st place in the world** among **8B** models, in a closed external benchmark.
|
| 256 |
+
|
| 257 |
+
Bnechmarked on the following site:
|
| 258 |
+
```
|
| 259 |
+
https://moonride.hashnode.dev/biased-test-of-gpt-4-era-llms-300-models-deepseek-r1-included
|
| 260 |
+
```
|
| 261 |
+
|
| 262 |
+
<img src="https://huggingface.co/SicariusSicariiStuff/Wingless_Imp_8B/resolve/main/Images/Wingless_8B_Bench.png" alt="External Benchmark" style="width: 100%; min-width: 600px; display: block; margin: auto;">
|
| 263 |
+
|
| 264 |
+
|
| 265 |
---
|
| 266 |
|
| 267 |
## Other stuff
|