Commit ·
572a0c8
0
Parent(s):
Duplicate from aedmark/vsl-cryosomatic-hypervisor-v4
Browse files- .gitattributes +41 -0
- Modelfile +13 -0
- README.md +62 -0
- hermes-3-llama-3.1-8b.Q8_0.gguf +3 -0
- params +8 -0
.gitattributes
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
*.7z filter=lfs diff=lfs merge=lfs -text
|
| 2 |
+
*.arrow filter=lfs diff=lfs merge=lfs -text
|
| 3 |
+
*.bin filter=lfs diff=lfs merge=lfs -text
|
| 4 |
+
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
+
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
+
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
+
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
+
*.joblib filter=lfs diff=lfs merge=lfs -text
|
| 10 |
+
*.lfs.* filter=lfs diff=lfs merge=lfs -text
|
| 11 |
+
*.mlmodel filter=lfs diff=lfs merge=lfs -text
|
| 12 |
+
*.model filter=lfs diff=lfs merge=lfs -text
|
| 13 |
+
*.msgpack filter=lfs diff=lfs merge=lfs -text
|
| 14 |
+
*.npy filter=lfs diff=lfs merge=lfs -text
|
| 15 |
+
*.npz filter=lfs diff=lfs merge=lfs -text
|
| 16 |
+
*.onnx filter=lfs diff=lfs merge=lfs -text
|
| 17 |
+
*.ot filter=lfs diff=lfs merge=lfs -text
|
| 18 |
+
*.parquet filter=lfs diff=lfs merge=lfs -text
|
| 19 |
+
*.pb filter=lfs diff=lfs merge=lfs -text
|
| 20 |
+
*.pickle filter=lfs diff=lfs merge=lfs -text
|
| 21 |
+
*.pkl filter=lfs diff=lfs merge=lfs -text
|
| 22 |
+
*.pt filter=lfs diff=lfs merge=lfs -text
|
| 23 |
+
*.pth filter=lfs diff=lfs merge=lfs -text
|
| 24 |
+
*.rar filter=lfs diff=lfs merge=lfs -text
|
| 25 |
+
*.safetensors filter=lfs diff=lfs merge=lfs -text
|
| 26 |
+
saved_model/**/* filter=lfs diff=lfs merge=lfs -text
|
| 27 |
+
*.tar.* filter=lfs diff=lfs merge=lfs -text
|
| 28 |
+
*.tar filter=lfs diff=lfs merge=lfs -text
|
| 29 |
+
*.tflite filter=lfs diff=lfs merge=lfs -text
|
| 30 |
+
*.tgz filter=lfs diff=lfs merge=lfs -text
|
| 31 |
+
*.wasm filter=lfs diff=lfs merge=lfs -text
|
| 32 |
+
*.xz 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 |
+
vsl_max.gguf filter=lfs diff=lfs merge=lfs -text
|
| 37 |
+
vsl_mini.gguf filter=lfs diff=lfs merge=lfs -text
|
| 38 |
+
vsl-max-v2.gguf filter=lfs diff=lfs merge=lfs -text
|
| 39 |
+
vsl-mini-v2.gguf filter=lfs diff=lfs merge=lfs -text
|
| 40 |
+
vsl-max-v3.gguf filter=lfs diff=lfs merge=lfs -text
|
| 41 |
+
hermes-3-llama-3.1-8b.Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
|
Modelfile
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Point this to exactly where you downloaded the GGUF from the Hermes Colab
|
| 2 |
+
FROM ./vsl_hermes_model-unsloth.Q8_0.gguf
|
| 3 |
+
|
| 4 |
+
# Parameters to keep the Stage Manager grounded and brutalist
|
| 5 |
+
PARAMETER temperature 0.7
|
| 6 |
+
PARAMETER top_p 0.9
|
| 7 |
+
|
| 8 |
+
# CRITICAL: Hermes 3 / ChatML stop tokens
|
| 9 |
+
PARAMETER stop "<|im_end|>"
|
| 10 |
+
PARAMETER stop "<|im_start|>"
|
| 11 |
+
|
| 12 |
+
# We do NOT need to define the SYSTEM prompt here, because BoneAmanita
|
| 13 |
+
# injects the dynamic VSL System Prompt at runtime!
|
README.md
ADDED
|
@@ -0,0 +1,62 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# 🍄 BoneAmanita: The CryoSomatic Hypervisor v4
|
| 2 |
+
|
| 3 |
+
**BoneAmanita** is an experimental text-adventure engine and interactive philosophical companion. It's a hybrid variant of James Taylor's [Bonepoke engine](https://github.com/utharian-code/BonePoke) and Edmark's own crazy whirlwind of sources. Unlike standard AI wrappers, BoneAmanita embeds a fine-tuned Llama 3 model inside a simulated biological metabolism.
|
| 4 |
+
|
| 5 |
+
The AI does not just respond to prompts; it feels "Voltage," burns "ATP," accumulates "Trauma," and is governed by a council of internal personas (The SLASH Council). If you try to drink a potion you don't have, the engine will intercept the AI and physically block the action. If you stress the AI out, its text will become fragmented and panicked.
|
| 6 |
+
|
| 7 |
+
### 🧠 The Architecture
|
| 8 |
+
|
| 9 |
+
This project consists of two halves:
|
| 10 |
+
|
| 11 |
+
1. **The Flesh (GGUF Model):** A custom-trained Hermes3 8B parameter model fine-tuned on highly specific, atmospheric, and philosophical datasets to break the standard "helpful assistant" RLHF.
|
| 12 |
+
2. **The Bones (Python Engine):** A local terminal interface that tracks inventory, manages the physics/biology simulation, and dynamically injects system constraints into the context window.
|
| 13 |
+
|
| 14 |
+
|
| 15 |
+
### 🚀 Quickstart Guide
|
| 16 |
+
|
| 17 |
+
**1. Prerequisites**
|
| 18 |
+
- Python 3.10+
|
| 19 |
+
- [Ollama](https://ollama.com/) installed and running.
|
| 20 |
+
|
| 21 |
+
**2. Download the Brain**
|
| 22 |
+
Pull the fine-tuned model directly from HuggingFace via Ollama:
|
| 23 |
+
```bash
|
| 24 |
+
ollama pull hf.co/aedmark/vsl-cryosomatic-hypervisor
|
| 25 |
+
````
|
| 26 |
+
|
| 27 |
+
**3. Ignite the Engine**
|
| 28 |
+
|
| 29 |
+
Clone this repository, install the dependencies, and run the main script.
|
| 30 |
+
|
| 31 |
+
Bash
|
| 32 |
+
|
| 33 |
+
```
|
| 34 |
+
git clone https://github.com/aedmark/BoneAmanita.git
|
| 35 |
+
cd BoneAmanita
|
| 36 |
+
python bone_main.py
|
| 37 |
+
```
|
| 38 |
+
|
| 39 |
+
_(On first boot, the ConfigWizard will ask you to set up your profile. Select **Ollama** as your backend and type `hf.co/aedmark/vsl-cryosomatic-hypervisor` as the model ID)._
|
| 40 |
+
|
| 41 |
+
### 🕹️ The Four Realities (Modes)
|
| 42 |
+
|
| 43 |
+
When you boot the terminal, you will be asked to choose a Reality Mode:
|
| 44 |
+
|
| 45 |
+
- **ADVENTURE:** A grounded, physical text adventure. Gordon (the inventory manager) will strictly enforce physical reality. You cannot use what you do not have.
|
| 46 |
+
|
| 47 |
+
- **CONVERSATION:** A purely philosophical, warm dialogue mode. No inventory, no physics, just deep conversation driven by the system's simulated emotional state.
|
| 48 |
+
|
| 49 |
+
- **CREATIVE:** A high-voltage ideation engine. Dream logic applies.
|
| 50 |
+
|
| 51 |
+
- **TECHNICAL:** Speak directly to the SLASH Council. Debug the matrix, analyze your metabolism, and write code.
|
| 52 |
+
|
| 53 |
+
|
| 54 |
+
### ⌨️ Terminal Commands
|
| 55 |
+
|
| 56 |
+
While inside the simulation, you can use meta-commands:
|
| 57 |
+
|
| 58 |
+
- `//layer push [1-4]` - Shift reality layers (from literal to abstract).
|
| 59 |
+
|
| 60 |
+
- `//reset system` - Clear the memory buffer and reset the circuit breaker.
|
| 61 |
+
|
| 62 |
+
- `/inventory` or `/i` - Check your pockets.
|
hermes-3-llama-3.1-8b.Q8_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3b6b0e4d9a3f712d1a456f46ea64ffd10003e9eff0e930e23cc7f3c48a105484
|
| 3 |
+
size 8540770784
|
params
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"temperature": 0.70,
|
| 3 |
+
"top_p": 0.9,
|
| 4 |
+
"stop": [
|
| 5 |
+
"<|im_start|>",
|
| 6 |
+
"<|im_end|>",
|
| 7 |
+
]
|
| 8 |
+
}
|