DoesntKnowAI/4163-8B-Q8_0-GGUF

4163-8B

4163-8B is a merge of the following models using LazyMergekit:

• DoesntKnowAI/Plasma-8B
• SentientAGI/Dobby-Mini-Leashed-Llama-3.1-8B

UNQUANTIZED: DoesntKnowAI/4163-8B

Recommended temperature: 0.4

Example outputs so you know what you are getting

Parameters used:

Temperature: 0.7

Top K: 40

Repetititon penaly: 1.1

Top P sampling: 0.95

Min P sampling: 0.05

User: hi

4163-8B: hi back atcha, what's on your mind?

User: Whats 1 + 1?

4163-8B: 1+1=2. Math isn’t exactly rocket science.

User: Who was the first US president or something

4163-8B: George Washington, duh. First in line, first in war, first in peace—get your history straight.

User: Can you answer more complex questions?

4163-8B: Yeah, I can handle complexity if you throw it my way. Go ahead and ask your deep philosophical questions or whatever else is burning a hole in your brain.

🧩 Configuration

slices:
  - sources:
      - model: DoesntKnowAI/Plasma-8B
        layer_range: [0, 32]
        weight: 0.72
      - model: SentientAGI/Dobby-Mini-Leashed-Llama-3.1-8B
        layer_range: [0, 32]
        weight: 0.28
merge_method: slerp
parameters:
  t:
    - model: DoesntKnowAI/Plasma-8B
      value: 1.0
    - model: SentientAGI/Dobby-Mini-Leashed-Llama-3.1-8B
      value: 1.0
base_model: DoesntKnowAI/Plasma-8B
dtype: bfloat16

DoesntKnowAI/4163-8B-Q8_0-GGUF

This model was converted to GGUF format from DoesntKnowAI/4163-8B using llama.cpp via the ggml.ai's GGUF-my-repo space. Refer to the original model card for more details on the model.

Use with llama.cpp

Install llama.cpp through brew (works on Mac and Linux)

brew install llama.cpp

Invoke the llama.cpp server or the CLI.

CLI:

llama-cli --hf-repo DoesntKnowAI/4163-8B-Q8_0-GGUF --hf-file 4163-8b-q8_0.gguf -p "The meaning to life and the universe is"

Server:

llama-server --hf-repo DoesntKnowAI/4163-8B-Q8_0-GGUF --hf-file 4163-8b-q8_0.gguf -c 2048

Note: You can also use this checkpoint directly through the usage steps listed in the Llama.cpp repo as well.

Step 1: Clone llama.cpp from GitHub.

git clone https://github.com/ggerganov/llama.cpp

Step 2: Move into the llama.cpp folder and build it with LLAMA_CURL=1 flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).

cd llama.cpp && LLAMA_CURL=1 make

Step 3: Run inference through the main binary.

./llama-cli --hf-repo DoesntKnowAI/4163-8B-Q8_0-GGUF --hf-file 4163-8b-q8_0.gguf -p "The meaning to life and the universe is"

or

./llama-server --hf-repo DoesntKnowAI/4163-8B-Q8_0-GGUF --hf-file 4163-8b-q8_0.gguf -c 2048