Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,26 @@
|
|
1 |
---
|
2 |
license: apache-2.0
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- en
|
5 |
+
pipeline_tag: text-generation
|
6 |
---
|
7 |
+
![palmer](https://huggingface.co/appvoid/palmer-001/resolve/main/new-logo.jpg)
|
8 |
+
# palmer
|
9 |
+
### a better base model
|
10 |
+
This model is palmer-002-2401 scaled to 32k by merging and fine-tuning with TinyLlama-1.1B-32k by Doctor-Shotgun
|
11 |
+
|
12 |
+
### evaluation 🧪
|
13 |
+
note that this is a zero-shot setting as opposite to open llm leaderboard's few-shot evals
|
14 |
+
```
|
15 |
+
model ARC-C OBQA HellaSwag PIQA Winogrande Average
|
16 |
+
tinyllama | 0.3029 | 0.3600 | 0.5935 | 0.7329 | 0.5959 | 0.5170 |
|
17 |
+
palmer-002-2401 | 0.3294 | 0.3700 | 0.5950 | 0.7399 | 0.5896 | 0.5247 |
|
18 |
+
palmer-002-32k | 0.3268 | 0.3780 | 0.5785 | 0.7492 | 0.6251 | 0.5315 | (this)
|
19 |
+
babbage-002 | 0.3285 | 0.3620 | 0.6380 | 0.7606 | 0.6085 | 0.5395 |
|
20 |
+
```
|
21 |
+
|
22 |
+
### prompt 📝
|
23 |
+
```
|
24 |
+
no prompt 🚀
|
25 |
+
```
|
26 |
+
<a href="https://ko-fi.com/appvoid" target="_blank"><img src="https://cdn.buymeacoffee.com/buttons/v2/default-yellow.png" alt="Buy Me A Coffee" style="height: 48px !important;width: 180px !important; filter: invert(70%);" ></a>
|