SnakyMcSnekFace
commited on
Commit
•
fcb87f5
1
Parent(s):
98a2815
Upload 3 files
Browse files- Psyfighter2-13B-vore.Q4_K_M.gguf +1 -1
- Psyfighter2-13B-vore.Q8_0.gguf +1 -1
- README.md +57 -6
Psyfighter2-13B-vore.Q4_K_M.gguf
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 7865956448
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:906ee4c1fa60cd36949972bfc6d3b35302435b77afd7031b50ec61d38b0f1185
|
3 |
size 7865956448
|
Psyfighter2-13B-vore.Q8_0.gguf
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 13831319648
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:487f1febff821f9d5c03dac612edc3248af899aac1f4b122e8ed2385f87817de
|
3 |
size 13831319648
|
README.md
CHANGED
@@ -12,15 +12,15 @@ tags:
|
|
12 |
base_model: SnakyMcSnekFace/Psyfighter2-13B-vore
|
13 |
model_type: llama
|
14 |
prompt_template: >
|
|
|
|
|
15 |
Below is an instruction that describes a task. Write a response that
|
16 |
appropriately completes the request.
|
17 |
|
18 |
-
|
19 |
-
### Instruction:
|
20 |
|
21 |
{prompt}
|
22 |
|
23 |
-
|
24 |
### Response:
|
25 |
---
|
26 |
|
@@ -30,7 +30,7 @@ This is a quantized version of [SnakyMcSnekFace/Psyfighter2-13B-vore](https://hu
|
|
30 |
|
31 |
This model is a version of [KoboldAI/LLaMA2-13B-Psyfighter2](https://huggingface.co/KoboldAI/LLaMA2-13B-Psyfighter2) finetuned to better understand vore context. The primary purpose of this model is to be a storywriting assistant, a conversational model in a chat, and an interactive choose-your-own-adventure text game.
|
32 |
|
33 |
-
The Adventure Mode is still work in progress, and will be
|
34 |
|
35 |
## Model Details
|
36 |
|
@@ -55,13 +55,63 @@ The easiest way to try out the model is [Koboldcpp Colab Notebook](https://colab
|
|
55 |
- Open the notebook
|
56 |
- Paste the model URL into the `Model` field: `https://huggingface.co/SnakyMcSnekFace/Psyfighter2-13B-vore-GGUF/resolve/main/Psyfighter2-13B-vore.Q4_K_M.gguf`
|
57 |
- Start the notebook, wait for the URL to CloudFlare tunnel to appear at the bottom and click it
|
58 |
-
-
|
59 |
-
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
60 |
|
61 |
#### To run `Q8_0` model in Colab notebook
|
62 |
- Paste the model URL into the `Model` field: `https://huggingface.co/SnakyMcSnekFace/Psyfighter2-13B-vore-GGUF/resolve/main/Psyfighter2-13B-vore.Q8_0.gguf`
|
63 |
- Set `Layers` field to `30` (reduce this number if the model fails to start)
|
64 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
65 |
### Backyard AI
|
66 |
|
67 |
Another convenient way to use the model is [Backyard AI](https://backyard.ai/) application, which allows you to run the model locally on your computer. You'll need a graphics card with at least 8GB VRAM to use the model comfortably.
|
@@ -78,6 +128,7 @@ Download [Psyfighter2-13B-vore.Q4_K_M.gguf](https://huggingface.co/SnakyMcSnekFa
|
|
78 |
|
79 |
## Model updates
|
80 |
|
|
|
81 |
- 06/02/2024 - fixed errors in training and merging, significantly improving the overall prose quality
|
82 |
- 05/25/2024 - updated training process, making the model more coherent and improving the writing quality
|
83 |
- 04/13/2024 - uploaded the first version of the model
|
|
|
12 |
base_model: SnakyMcSnekFace/Psyfighter2-13B-vore
|
13 |
model_type: llama
|
14 |
prompt_template: >
|
15 |
+
### Instruction:
|
16 |
+
|
17 |
Below is an instruction that describes a task. Write a response that
|
18 |
appropriately completes the request.
|
19 |
|
20 |
+
### Input:
|
|
|
21 |
|
22 |
{prompt}
|
23 |
|
|
|
24 |
### Response:
|
25 |
---
|
26 |
|
|
|
30 |
|
31 |
This model is a version of [KoboldAI/LLaMA2-13B-Psyfighter2](https://huggingface.co/KoboldAI/LLaMA2-13B-Psyfighter2) finetuned to better understand vore context. The primary purpose of this model is to be a storywriting assistant, a conversational model in a chat, and an interactive choose-your-own-adventure text game.
|
32 |
|
33 |
+
The model has preliminary support for Adventure Mode. It is still work in progress, and the model will be further trained to be better at storytelling and to not require complicated configuration. **(See instructions below for configuring Adventure Mode for optimal performance.)**
|
34 |
|
35 |
## Model Details
|
36 |
|
|
|
55 |
- Open the notebook
|
56 |
- Paste the model URL into the `Model` field: `https://huggingface.co/SnakyMcSnekFace/Psyfighter2-13B-vore-GGUF/resolve/main/Psyfighter2-13B-vore.Q4_K_M.gguf`
|
57 |
- Start the notebook, wait for the URL to CloudFlare tunnel to appear at the bottom and click it
|
58 |
+
- Select "Settings" and configure them as follows:
|
59 |
+
- In "Basic" tab:
|
60 |
+
- Temperature = 0.8
|
61 |
+
- Amount to Gen. = 512
|
62 |
+
- Top p Sampling = 0.9
|
63 |
+
- Repetition Pentalty = 1.1
|
64 |
+
- In "Advanced" tab:
|
65 |
+
- Min-P = 0.1
|
66 |
+
- EOS Token Ban = Unban
|
67 |
+
- Placeholder Tags = Checked
|
68 |
+
- Select "Scenarios" -> "New Story" to use the model as a writing assistant
|
69 |
|
70 |
#### To run `Q8_0` model in Colab notebook
|
71 |
- Paste the model URL into the `Model` field: `https://huggingface.co/SnakyMcSnekFace/Psyfighter2-13B-vore-GGUF/resolve/main/Psyfighter2-13B-vore.Q8_0.gguf`
|
72 |
- Set `Layers` field to `30` (reduce this number if the model fails to start)
|
73 |
|
74 |
+
#### Adventure Mode
|
75 |
+
|
76 |
+
The model has **preliminary** support for the adventure mode. Select "Scenarios" -> "New Adventure" to start an adventure.
|
77 |
+
|
78 |
+
**IMPORTANT - The context must be configured exactly as described below or the model will not perform well.**
|
79 |
+
|
80 |
+
- Select "Context"
|
81 |
+
- Paste this text into "Memory" verbatim:
|
82 |
+
```
|
83 |
+
### Instruction:
|
84 |
+
|
85 |
+
Text transcript of a never-ending adventure story, written by the AI assistant. AI assistant uses vivid and evocative language to create a well-written novel. Characters are proactive and take initiative. Think about what goals the characters of the story have and write what they do to achieve those goals.
|
86 |
+
|
87 |
+
### Input:
|
88 |
+
```
|
89 |
+
- Ensure that "Newline After Memory" is checked
|
90 |
+
- Paste this text into "Author's note" verbatim **(IMPORTANT: the Author's Note must end with two empty lines)**:
|
91 |
+
```
|
92 |
+
Write a few paragraphs that advance the plot of the story.
|
93 |
+
|
94 |
+
### Response:
|
95 |
+
|
96 |
+
|
97 |
+
```
|
98 |
+
- Set "A/N Strength" to "Immediate"
|
99 |
+
- Set "Author's Note Template" to `<|>`
|
100 |
+
|
101 |
+
#### Adventure - creating premise and taking turns
|
102 |
+
|
103 |
+
Select "Story" in the bottom left corner to generate premise of the story, and "Action" to take actions with your character. In the adventure mode, the model expects all player actions to be written in second person. For example:
|
104 |
+
|
105 |
+
```
|
106 |
+
As you venture deeper into the damp cave, you come across a lone goblin. The vile creature mumbles something to itself as it stares at the glowing text on a cave wall. It doesn't notice your approach.
|
107 |
+
|
108 |
+
> You sneak behind the goblin and hit it with the sword.
|
109 |
+
|
110 |
+
```
|
111 |
+
|
112 |
+
Check "Allow Editing" to make edits to the story to overwrite and re-generate parts of the model's response. This is useful if the model makes a mistake or the story doesn't go in the direction that you like.
|
113 |
+
|
114 |
+
|
115 |
### Backyard AI
|
116 |
|
117 |
Another convenient way to use the model is [Backyard AI](https://backyard.ai/) application, which allows you to run the model locally on your computer. You'll need a graphics card with at least 8GB VRAM to use the model comfortably.
|
|
|
128 |
|
129 |
## Model updates
|
130 |
|
131 |
+
- 09/02/2024 - fine-tuned the model to follow Kobold AI Adventure Mode format
|
132 |
- 06/02/2024 - fixed errors in training and merging, significantly improving the overall prose quality
|
133 |
- 05/25/2024 - updated training process, making the model more coherent and improving the writing quality
|
134 |
- 04/13/2024 - uploaded the first version of the model
|