PantheraMax-L3-RP-TestProbe-4-4x8B
PantheraMax-L3-RP-TestProbe-4-4x8B is a Mixture of Experts (MoE) made with the following models:
🧩 Configuration
base_model: failspy/Meta-Llama-3-8B-Instruct-abliterated-v3
gate_mode: random
dtype: bfloat16
experts_per_token: 4
experts:
- source_model: v000000/L3-8B-Poppy-Moonfall-C
positive_prompts: []
- source_model: Alsebay/L3-test-2
positive_prompts:
- "Imagine"
- "Create"
- "Envision"
- "Fantasize"
- "Invent"
- "Narrate"
- "Plot"
- "Portray"
- "Storytell"
- "Visualize"
- "Describe"
- "Develop"
- "Forge"
- "Craft"
- "Conceptualize"
- "Dream"
- "Concoct"
- "Characterize"
negative_prompts:
- "Analyze"
- "Critique"
- "Dissect"
- "Explain"
- "Clarify"
- "Interpret"
- source_model: merge3
positive_prompts: []
- source_model: Nitral-AI/Hathor_RP-v.01-L3-8B
positive_prompts: []
💻 Usage
!pip install -qU transformers bitsandbytes accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Kaoeiri/PantheraMax-L3-RP-TestProbe-4-4x8B"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
model_kwargs={"torch_dtype": torch.float16, "load_in_4bit": True},
)
messages = [{"role": "user", "content": "Explain what a Mixture of Experts is in less than 100 words."}]
prompt = pipeline.tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.9, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])