Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: gemma
|
| 3 |
+
base_model: unsloth/gemma-4-E4B-it-unsloth-bnb-4bit
|
| 4 |
+
tags:
|
| 5 |
+
- unsloth
|
| 6 |
+
- gguf
|
| 7 |
+
- q4_k_m
|
| 8 |
+
- lora
|
| 9 |
+
pipeline_tag: text-generation
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
# evilguy4
|
| 13 |
+
|
| 14 |
+
The newest evilguy. Now based off Gemma4 E4B.
|
| 15 |
+
|
| 16 |
+
## Changes
|
| 17 |
+
|
| 18 |
+
- Better at math
|
| 19 |
+
- Trained to handle more slang/substitutions/typos (hllo, ello, or complete gibberish)
|
| 20 |
+
|
| 21 |
+
Recommended sampling: temperature 1.0, top_p 0.95, top_k 64.
|
| 22 |
+
|
| 23 |
+
## Usage
|
| 24 |
+
|
| 25 |
+
```bash
|
| 26 |
+
llama-server -m evilguy4-q4-k-m.gguf --jinja
|
| 27 |
+
```
|
| 28 |
+
|
| 29 |
+
Run the llama.cpp server with `--jinja` so the embedded chat template (and tool calling) is used.
|