Upload folder using huggingface_hub
Browse files- .gitattributes +14 -0
- README.md +51 -0
- norobara-zloss-8x7b.Q2_K.gguf +3 -0
- norobara-zloss-8x7b.Q3_K_L.gguf +3 -0
- norobara-zloss-8x7b.Q3_K_M.gguf +3 -0
- norobara-zloss-8x7b.Q3_K_S.gguf +3 -0
- norobara-zloss-8x7b.Q4_0.gguf +3 -0
- norobara-zloss-8x7b.Q4_1.gguf +3 -0
- norobara-zloss-8x7b.Q4_K_M.gguf +3 -0
- norobara-zloss-8x7b.Q4_K_S.gguf +3 -0
- norobara-zloss-8x7b.Q5_0.gguf +3 -0
- norobara-zloss-8x7b.Q5_1.gguf +3 -0
- norobara-zloss-8x7b.Q5_K_M.gguf +3 -0
- norobara-zloss-8x7b.Q5_K_S.gguf +3 -0
- norobara-zloss-8x7b.Q6_K.gguf +3 -0
- norobara-zloss-8x7b.Q8_0.gguf +3 -0
.gitattributes
CHANGED
@@ -33,3 +33,17 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
33 |
*.zip filter=lfs diff=lfs merge=lfs -text
|
34 |
*.zst filter=lfs diff=lfs merge=lfs -text
|
35 |
*tfevents* filter=lfs diff=lfs merge=lfs -text
|
36 |
+
norobara-zloss-8x7b.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
|
37 |
+
norobara-zloss-8x7b.Q3_K_L.gguf filter=lfs diff=lfs merge=lfs -text
|
38 |
+
norobara-zloss-8x7b.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
39 |
+
norobara-zloss-8x7b.Q3_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
40 |
+
norobara-zloss-8x7b.Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
|
41 |
+
norobara-zloss-8x7b.Q4_1.gguf filter=lfs diff=lfs merge=lfs -text
|
42 |
+
norobara-zloss-8x7b.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
43 |
+
norobara-zloss-8x7b.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
44 |
+
norobara-zloss-8x7b.Q5_0.gguf filter=lfs diff=lfs merge=lfs -text
|
45 |
+
norobara-zloss-8x7b.Q5_1.gguf filter=lfs diff=lfs merge=lfs -text
|
46 |
+
norobara-zloss-8x7b.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
|
47 |
+
norobara-zloss-8x7b.Q5_K_S.gguf filter=lfs diff=lfs merge=lfs -text
|
48 |
+
norobara-zloss-8x7b.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
|
49 |
+
norobara-zloss-8x7b.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
README.md
ADDED
@@ -0,0 +1,51 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
title: "Norobara-ZLoss-8x7B Quantized in GGUF"
|
3 |
+
tags:
|
4 |
+
- GGUF
|
5 |
+
language: en
|
6 |
+
---
|
7 |
+
![Image description](https://i.postimg.cc/MGwhtFfF/tsune-fixed.png)
|
8 |
+
|
9 |
+
# Tsunemoto GGUF's of Norobara-ZLoss-8x7B
|
10 |
+
|
11 |
+
This is a GGUF quantization of Norobara-ZLoss-8x7B.
|
12 |
+
|
13 |
+
## Original Repo Link:
|
14 |
+
[Original Repository](https://huggingface.co/Doctor-Shotgun/Norobara-ZLoss-8x7B)
|
15 |
+
|
16 |
+
## Original Model Card:
|
17 |
+
---
|
18 |
+
|
19 |
+
# Norobara-ZLoss-8x7B
|
20 |
+
|
21 |
+
This is an experimental instruct-tuned [mistralai/Mixtral-8x7B-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-v0.1)-based model trained using [Charles Goddard](https://huggingface.co/chargoddard)'s ZLoss and Megablocks-based fork of transformers.
|
22 |
+
|
23 |
+
It primarily uses the Capybara and No Robots datasets (thus the name). The goal was to create an uncensored general instruction following model, as well as test various loss implementations while we figure out how the heck to train Mixtral properly.
|
24 |
+
|
25 |
+
[Exl2 Quants](https://huggingface.co/royallab/Norobara-ZLoss-8x7B-exl2)
|
26 |
+
|
27 |
+
## Usage:
|
28 |
+
The intended prompt format is a modified multi-turn Alpaca instruction format:
|
29 |
+
```
|
30 |
+
### Instruction:
|
31 |
+
{system prompt}
|
32 |
+
|
33 |
+
### Input:
|
34 |
+
{user message}
|
35 |
+
|
36 |
+
### Response:
|
37 |
+
{model response}
|
38 |
+
|
39 |
+
### Input:
|
40 |
+
{user message}
|
41 |
+
|
42 |
+
### Response:
|
43 |
+
{model response}
|
44 |
+
|
45 |
+
(etc.)
|
46 |
+
```
|
47 |
+
|
48 |
+
## Bias, Risks, and Limitations
|
49 |
+
The model will show biases present in the base model. No ethical alignment was applied to prevent the generation of toxic or harmful outputs (in fact the opposite, with examples from toxic-DPO included), so generate at your own risk.
|
50 |
+
## Training Details
|
51 |
+
This model was trained as a QLora adapter for 3 epochs using a single H100 GPU for around 13 hours.
|
norobara-zloss-8x7b.Q2_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:586b34708b37777bb89fc8a81624ee578af65f0b5ca0f4beab28c4cabd326f90
|
3 |
+
size 15459717536
|
norobara-zloss-8x7b.Q3_K_L.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a329509f16d2fab93f35015d7460acc4c512f293e41764b13e93f18476e537c1
|
3 |
+
size 20294488480
|
norobara-zloss-8x7b.Q3_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c1accc0a0d8caa6a2bdedb0e0ab302925f5f86c018f71dd891a728e6d1e88646
|
3 |
+
size 20211650976
|
norobara-zloss-8x7b.Q3_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:8bf4136c90ef2a9b483313b05f54b7a2dd147fe5e995c70dcbb9754c8a942517
|
3 |
+
size 20121473440
|
norobara-zloss-8x7b.Q4_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0cfec8e7925fe26a917a440dbfb58dfd30be1084a78b98be879ca4cd508cd2d6
|
3 |
+
size 26306744736
|
norobara-zloss-8x7b.Q4_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:893da5f2926412de4119e701975cae58c0adecdfa0805e5316c6c8249287351f
|
3 |
+
size 29217460640
|
norobara-zloss-8x7b.Q4_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:62de08fd91b9a334c88af29d7b9bc21da015a38fa68a5fb2d628828373c1c8a4
|
3 |
+
size 26324046240
|
norobara-zloss-8x7b.Q4_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:ae3900a36f68b168b3de30c2788f1d5846aa38e8fc14872c237ae7d1dd28d7f2
|
3 |
+
size 26308841888
|
norobara-zloss-8x7b.Q5_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e7f17cc8ccceb35ab652cfe5ad7b31df712867762530567039a8a4cfd325d89d
|
3 |
+
size 32128176544
|
norobara-zloss-8x7b.Q5_1.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e6fe0e7c09b6bf8e2be2faa3381ebcb887acbbd52d681202fdbe83f1fc69066e
|
3 |
+
size 35038892448
|
norobara-zloss-8x7b.Q5_K_M.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e00aec864c1f0e7a4a221517ee588c2177f25c7b6e773982b48f3d88051d814a
|
3 |
+
size 32137089440
|
norobara-zloss-8x7b.Q5_K_S.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:f68adb40f72699e2501bdb1f70bcab0a4b2b144f74dbe7c0694e87b0869ea590
|
3 |
+
size 32128176544
|
norobara-zloss-8x7b.Q6_K.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:36cc34eb6f837ae4da110353258f9bfa30e85ac1c2b4877a0b4e74fa16103392
|
3 |
+
size 38313447840
|
norobara-zloss-8x7b.Q8_0.gguf
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:0dd08db3005e77a2801fd6cb708bc8f342c589a107380c2d7feb5b87ff6268b3
|
3 |
+
size 49624215968
|