metadata

tags:
  - merge
  - mergekit
  - lazymergekit
  - Hypersniper/The_Philosopher_Zephyr_7B
  - sayhan/OpenHermes-2.5-Strix-Philosophy-Mistral-7B-LoRA
  - teknium/Mistral-Trismegistus-7B
base_model:
  - Hypersniper/The_Philosopher_Zephyr_7B
  - sayhan/OpenHermes-2.5-Strix-Philosophy-Mistral-7B-LoRA
  - teknium/Mistral-Trismegistus-7B

Trascendental-Bot-7B

Trascendental-Bot-7B is a merge of the following models using LazyMergekit:

🧩 Configuration

models:
  - model: teknium/Mistral-Trismegistus-7B
    # no parameters necessary for base model
  - model: Hypersniper/The_Philosopher_Zephyr_7B
    parameters:
      density: 0.55
      weight: 0.3
  - model: sayhan/OpenHermes-2.5-Strix-Philosophy-Mistral-7B-LoRA
    parameters:
      density: 0.55
      weight: 0.3
  - model: teknium/Mistral-Trismegistus-7B
    parameters:
      density: 0.4
      weight: 0.66
merge_method: dare_ties
base_model: teknium/Mistral-Trismegistus-7B
parameters:
  int8_mask: true
dtype: bfloat16
random_seed: 0

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Kukedlc/Trascendental-Bot-7B"
messages = [
    {"role": "system", "content": "You are an expert assistant in mysticism and philosophy."},
    {"role": "user", "content": "Create an innovative and disruptive theory that explains human consciousness. Give me an extensive and detailed answer."}
]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=1024, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])