Einstein-4D-Marcoro14-12b-32k-experiment
Einstein-4D-Marcoro14-12b-32k-experiment is a merge of the following models using LazyMergekit:
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
🧩 Configuration
dtype: float16
merge_method: passthrough
slices:
- sources:
- layer_range: [0, 8]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [4, 12]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [8, 16]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [12, 20]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [16, 24]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [20, 28]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
- sources:
- layer_range: [24, 32]
model: mvpmaster/Einstein-4D-Marcoro14-7b-full-slerp
💻 Usage
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "mvpmaster/Einstein-4D-Marcoro14-12b-32k-experiment"
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"])