Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,50 @@
|
|
1 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2 |
license: cc-by-4.0
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
pipeline_tag: text-generation
|
5 |
+
tags:
|
6 |
+
- text-generation-inference
|
7 |
+
- instruct
|
8 |
+
- conversational
|
9 |
+
- roleplay
|
10 |
license: cc-by-4.0
|
11 |
---
|
12 |
+
|
13 |
+
<h1 style="text-align: center">Erosumika-7B-v3</h1>
|
14 |
+
|
15 |
+
<div style="display: flex; justify-content: center;">
|
16 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/6512681f4151fb1fa719e033/ZX5NLfB2CctdwuctS9W8A.gif" alt="Header GIF">
|
17 |
+
</div>
|
18 |
+
|
19 |
+
|
20 |
+
4.0bpw exl2 quant. great for 16k+ context on 6GB GPUS!
|
21 |
+
|
22 |
+
Original Model : (https://huggingface.co/localfultonextractor/Erosumika-7B-v3)
|
23 |
+
|
24 |
+
## Model Details
|
25 |
+
A DARE TIES merge between Nitral's [Kunocchini-7b](https://huggingface.co/Nitral-AI/Kunocchini-7b), Endevor's [InfinityRP-v1-7B](https://huggingface.co/Endevor/InfinityRP-v1-7B) and my [FlatErosAlpha](https://huggingface.co/localfultonextractor/FlatErosAlpha), a flattened(in order to keep the vocab size 32000) version of tavtav's [eros-7B-ALPHA](https://huggingface.co/tavtav/eros-7B-ALPHA). Alpaca and ChatML work best.
|
26 |
+
|
27 |
+
[GGUF quants](https://huggingface.co/localfultonextractor/Erosumika-7B-v3-GGUF)
|
28 |
+
|
29 |
+
|
30 |
+
## Limitations and biases
|
31 |
+
The intended use-case for this model is fictional writing for entertainment purposes. Any other sort of usage is out of scope.
|
32 |
+
It may produce socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive. Outputs might often be factually wrong or misleading.
|
33 |
+
|
34 |
+
|
35 |
+
```yaml
|
36 |
+
base_model: localfultonextractor/FlatErosAlpha
|
37 |
+
models:
|
38 |
+
- model: localfultonextractor/FlatErosAlpha
|
39 |
+
- model: Epiculous/InfinityRP-v1-7B
|
40 |
+
parameters:
|
41 |
+
density: 0.4
|
42 |
+
weight: 0.25
|
43 |
+
- model: Nitral-AI/Kunocchini-7b
|
44 |
+
parameters:
|
45 |
+
density: 0.3
|
46 |
+
weight: 0.35
|
47 |
+
merge_method: dare_ties
|
48 |
+
dtype: bfloat16
|
49 |
+
```
|
50 |
+
Note: Copied the tokenizer from InfinityRP-v1-7B.
|