slices:

      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [4, 5]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [6, 7]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [8, 9]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [10, 11]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [11, 12]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [12, 13]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [14, 15]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [16, 17]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [17, 18]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [20, 21]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [23, 24]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [25, 26]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [26, 27]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [28, 29]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [29, 30]
    
      - sources:
          - model: NousResearch/Meta-Llama-3-8B
            layer_range: [31, 32]
    
merge_method: passthrough
dtype: bfloat16

💻 Usage

python

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "ryan0712/llama-3-8b-slow-DUS-random-layer2-method2"
messages = [{"role": "user", "content": "What is a large language model?"}]

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=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])

RichardErkhov/ryan0712_-_llama-3-8b-slow-DUS-random-layer2-method2-8bits

Original model description:

llama-3-8b-slow-DUS-random-layer2-method2

🧩 Configuration

💻 Usage

RichardErkhov/
ryan0712_-_llama-3-8b-slow-DUS-random-layer2-method2-8bits