Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,109 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
|
| 2 |
+
---
|
| 3 |
+
|
| 4 |
+
base_model: []
|
| 5 |
+
library_name: transformers
|
| 6 |
+
tags:
|
| 7 |
+
- mergekit
|
| 8 |
+
- merge
|
| 9 |
+
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
|
| 13 |
+
|
| 14 |
+
# QuantFactory/NemoMix-Unleashed-12B-GGUF
|
| 15 |
+
This is quantized version of [MarinaraSpaghetti/NemoMix-Unleashed-12B](https://huggingface.co/MarinaraSpaghetti/NemoMix-Unleashed-12B) created using llama.cpp
|
| 16 |
+
|
| 17 |
+
# Original Model Card
|
| 18 |
+
|
| 19 |
+
|
| 20 |
+
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
# Information
|
| 25 |
+
## Details
|
| 26 |
+
|
| 27 |
+
Okay, I tried really hard to improve my ChatML merges, but that has gone terribly wrong. Everyone is adding special tokens with different IDs so can't even make a proper union tokenizer for them, damn. Not to mention, I made some... interesting discoveres in regards to some models' context lenghts. You can watch the breakdown of how it went down here: https://www.captiongenerator.com/v/2303039/marinaraspaghetti's-merging-experience.
|
| 28 |
+
|
| 29 |
+
This one feels a bit different to my previous attempts and seems less prone to repetition, especially on higher contexts, which is great for me! I'll probably improve on it even further, but for now, it feels rather nice. Great for RP and storytelling. All credits and thanks go to the amazing MistralAI, Intervitens, Sao10K and Nbeerbower for their amazing models! Plus, special shoutouts to Parasitic Rogue for ideas and Prodeus Unity for cool exl2 quants of my previous merges. Have a good one, everyone.
|
| 30 |
+
|
| 31 |
+
## Instruct
|
| 32 |
+
|
| 33 |
+
|
| 34 |
+
|
| 35 |
+
*Sigh,* Mistral Instruct, I'm afraid.
|
| 36 |
+
|
| 37 |
+
```
|
| 38 |
+
<s>[INST] {system} [/INST]{response}</s>[INST] {user's message} [/INST]{response}</s>
|
| 39 |
+
```
|
| 40 |
+
|
| 41 |
+
## Parameters
|
| 42 |
+
|
| 43 |
+
I recommend running Temperature 1.0-1.25 with 0.1 Top A or 0.01-0.1 Min P, and with 0.8/1.75/2/0 DRY. Also works with lower Temperatures below 1.0. Nothing more needed.
|
| 44 |
+
|
| 45 |
+
### Settings
|
| 46 |
+
|
| 47 |
+
You can use my exact settings from here (use the ones from the Mistral Base/Customized folder, I also recommend checking the Mistral Improved folder): https://huggingface.co/MarinaraSpaghetti/SillyTavern-Settings/tree/main.
|
| 48 |
+
|
| 49 |
+
## GGUF
|
| 50 |
+
|
| 51 |
+
https://huggingface.co/bartowski/NemoMix-Unleashed-12B-GGUF
|
| 52 |
+
|
| 53 |
+
# NemoMix-Unleashed-12B
|
| 54 |
+
|
| 55 |
+
This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
|
| 56 |
+
|
| 57 |
+
## Merge Details
|
| 58 |
+
### Merge Method
|
| 59 |
+
|
| 60 |
+
This model was merged using the della_linear merge method using E:\mergekit\mistralaiMistral-Nemo-Base-2407 as a base.
|
| 61 |
+
|
| 62 |
+
### Models Merged
|
| 63 |
+
|
| 64 |
+
The following models were included in the merge:
|
| 65 |
+
* E:\mergekit\intervitens_mini-magnum-12b-v1.1
|
| 66 |
+
* E:\mergekit\nbeerbower_mistral-nemo-bophades-12B
|
| 67 |
+
* E:\mergekit\Sao10K_MN-12B-Lyra-v1
|
| 68 |
+
* E:\mergekit\nbeerbower_mistral-nemo-gutenberg-12B
|
| 69 |
+
* E:\mergekit\mistralaiMistral-Nemo-Instruct-2407
|
| 70 |
+
|
| 71 |
+
### Configuration
|
| 72 |
+
|
| 73 |
+
The following YAML configuration was used to produce this model:
|
| 74 |
+
|
| 75 |
+
```yaml
|
| 76 |
+
models:
|
| 77 |
+
- model: E:\mergekit\mistralaiMistral-Nemo-Instruct-2407
|
| 78 |
+
parameters:
|
| 79 |
+
weight: 0.1
|
| 80 |
+
density: 0.4
|
| 81 |
+
- model: E:\mergekit\nbeerbower_mistral-nemo-bophades-12B
|
| 82 |
+
parameters:
|
| 83 |
+
weight: 0.12
|
| 84 |
+
density: 0.5
|
| 85 |
+
- model: E:\mergekit\nbeerbower_mistral-nemo-gutenberg-12B
|
| 86 |
+
parameters:
|
| 87 |
+
weight: 0.2
|
| 88 |
+
density: 0.6
|
| 89 |
+
- model: E:\mergekit\Sao10K_MN-12B-Lyra-v1
|
| 90 |
+
parameters:
|
| 91 |
+
weight: 0.25
|
| 92 |
+
density: 0.7
|
| 93 |
+
- model: E:\mergekit\intervitens_mini-magnum-12b-v1.1
|
| 94 |
+
parameters:
|
| 95 |
+
weight: 0.33
|
| 96 |
+
density: 0.8
|
| 97 |
+
merge_method: della_linear
|
| 98 |
+
base_model: E:\mergekit\mistralaiMistral-Nemo-Base-2407
|
| 99 |
+
parameters:
|
| 100 |
+
epsilon: 0.05
|
| 101 |
+
lambda: 1
|
| 102 |
+
dtype: bfloat16
|
| 103 |
+
tokenizer_source: base
|
| 104 |
+
```
|
| 105 |
+
|
| 106 |
+
# Ko-fi
|
| 107 |
+
## Enjoying what I do? Consider donating here, thank you!
|
| 108 |
+
|
| 109 |
+
https://ko-fi.com/spicy_marinara
|