metadata
base_model:
- Sao10K/L3-8B-Lunaris-v1
tags:
- merge
- mergekit
- lazymergekit
- Sao10K/L3-8B-Lunaris-v1
L3-Lunaris-v1-15B
L3-Lunaris-v1-15B is a merge of the following model with itself using LazyMergekit:
🧩 Configuration
dtype: bfloat16
merge_method: passthrough
slices:
- sources:
- layer_range: [0, 24]
model: Sao10K/L3-8B-Lunaris-v1
- sources:
- layer_range: [8, 24]
model: Sao10K/L3-8B-Lunaris-v1
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [8, 24]
model: Sao10K/L3-8B-Lunaris-v1
parameters:
scale:
- filter: o_proj
value: 0.0
- filter: down_proj
value: 0.0
- value: 1.0
- sources:
- layer_range: [24, 32]
model: Sao10K/L3-8B-Lunaris-v1
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "Tremontaine/L3-Lunaris-v1-15B"
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"])