Commit
•
0322cc3
1
Parent(s):
e2b806a
uploaded readme
Browse files
README.md
ADDED
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Quantization made by Richard Erkhov.
|
2 |
+
|
3 |
+
[Github](https://github.com/RichardErkhov)
|
4 |
+
|
5 |
+
[Discord](https://discord.gg/pvy7H8DZMG)
|
6 |
+
|
7 |
+
[Request more models](https://github.com/RichardErkhov/quant_request)
|
8 |
+
|
9 |
+
|
10 |
+
Narumashi-RT-11B - GGUF
|
11 |
+
- Model creator: https://huggingface.co/Alsebay/
|
12 |
+
- Original model: https://huggingface.co/Alsebay/Narumashi-RT-11B/
|
13 |
+
|
14 |
+
|
15 |
+
| Name | Quant method | Size |
|
16 |
+
| ---- | ---- | ---- |
|
17 |
+
| [Narumashi-RT-11B.Q2_K.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q2_K.gguf) | Q2_K | 3.73GB |
|
18 |
+
| [Narumashi-RT-11B.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.IQ3_XS.gguf) | IQ3_XS | 4.14GB |
|
19 |
+
| [Narumashi-RT-11B.IQ3_S.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.IQ3_S.gguf) | IQ3_S | 4.37GB |
|
20 |
+
| [Narumashi-RT-11B.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q3_K_S.gguf) | Q3_K_S | 4.34GB |
|
21 |
+
| [Narumashi-RT-11B.IQ3_M.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.IQ3_M.gguf) | IQ3_M | 4.51GB |
|
22 |
+
| [Narumashi-RT-11B.Q3_K.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q3_K.gguf) | Q3_K | 4.84GB |
|
23 |
+
| [Narumashi-RT-11B.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q3_K_M.gguf) | Q3_K_M | 4.84GB |
|
24 |
+
| [Narumashi-RT-11B.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q3_K_L.gguf) | Q3_K_L | 5.26GB |
|
25 |
+
| [Narumashi-RT-11B.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.IQ4_XS.gguf) | IQ4_XS | 5.43GB |
|
26 |
+
| [Narumashi-RT-11B.Q4_0.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q4_0.gguf) | Q4_0 | 5.66GB |
|
27 |
+
| [Narumashi-RT-11B.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.IQ4_NL.gguf) | IQ4_NL | 5.72GB |
|
28 |
+
| [Narumashi-RT-11B.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q4_K_S.gguf) | Q4_K_S | 5.7GB |
|
29 |
+
| [Narumashi-RT-11B.Q4_K.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q4_K.gguf) | Q4_K | 6.02GB |
|
30 |
+
| [Narumashi-RT-11B.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q4_K_M.gguf) | Q4_K_M | 6.02GB |
|
31 |
+
| [Narumashi-RT-11B.Q4_1.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q4_1.gguf) | Q4_1 | 6.27GB |
|
32 |
+
| [Narumashi-RT-11B.Q5_0.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q5_0.gguf) | Q5_0 | 6.89GB |
|
33 |
+
| [Narumashi-RT-11B.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q5_K_S.gguf) | Q5_K_S | 6.89GB |
|
34 |
+
| [Narumashi-RT-11B.Q5_K.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q5_K.gguf) | Q5_K | 7.08GB |
|
35 |
+
| [Narumashi-RT-11B.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q5_K_M.gguf) | Q5_K_M | 7.08GB |
|
36 |
+
| [Narumashi-RT-11B.Q5_1.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q5_1.gguf) | Q5_1 | 7.51GB |
|
37 |
+
| [Narumashi-RT-11B.Q6_K.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q6_K.gguf) | Q6_K | 8.2GB |
|
38 |
+
| [Narumashi-RT-11B.Q8_0.gguf](https://huggingface.co/RichardErkhov/Alsebay_-_Narumashi-RT-11B-gguf/blob/main/Narumashi-RT-11B.Q8_0.gguf) | Q8_0 | 10.62GB |
|
39 |
+
|
40 |
+
|
41 |
+
|
42 |
+
|
43 |
+
Original model description:
|
44 |
+
---
|
45 |
+
language:
|
46 |
+
- en
|
47 |
+
license: cc-by-nc-4.0
|
48 |
+
tags:
|
49 |
+
- text-generation-inference
|
50 |
+
- transformers
|
51 |
+
- unsloth
|
52 |
+
- llama
|
53 |
+
- trl
|
54 |
+
- sft
|
55 |
+
- Roleplay
|
56 |
+
- roleplay
|
57 |
+
base_model: Sao10K/Fimbulvetr-11B-v2
|
58 |
+
---
|
59 |
+
> [!Important]
|
60 |
+
> Still in experiment
|
61 |
+
# About this model
|
62 |
+
|
63 |
+
This model now can handle (limited) TSF content. If you Character Card have complex plot, maybe you should try other model (maybe bigger parameter?).
|
64 |
+
|
65 |
+
**Update: I think it worse than original model: Sao10K/Fimbulvetr-11B-v2. This model was trained with rough translated dataset, so the responses is short, the IQ logic go down, also it will response wrong name, nonsense sentences sometimes...**
|
66 |
+
Anyways, if you find this is good, please let me know. Will have another update later.
|
67 |
+
|
68 |
+
Do you know TSF, TS, TG? A lot of model don't really know about that, so I do some experiment to finetune TSF dataset.
|
69 |
+
|
70 |
+
- **Finetuned with rough translate dataset, to increase the accuracy in TSF theme, which is not quite popular. (lewd dataset)**
|
71 |
+
- **Finetuned from model :** Sao10K/Fimbulvetr-11B-v2 . Thank Sao10K a lot :)
|
72 |
+
|
73 |
+
## Still testing, but seem it good enough for handle information. But the logic go down a bit because the rough translate dataset.
|
74 |
+
## GGUF version? [here is it](https://huggingface.co/Alsebay/Narumashi-RT-11B-GGUF).
|
75 |
+
## Dataset
|
76 |
+
Rough translated dataset, you could say that this is bad quality dataset.
|
77 |
+
```
|
78 |
+
Dataset(all are novels):
|
79 |
+
30% skinsuit
|
80 |
+
30% possession
|
81 |
+
35% transform(shapeshift)
|
82 |
+
5% other
|
83 |
+
```
|
84 |
+
|
85 |
+
# Thank Unsloth for good finetuning tool. This mistral model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
|
86 |
+
|
87 |
+
[<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
|
88 |
+
|