D_AU - Experiments in Merging Top Models
Collection
Full precision and Q8 of merged models by me. Focus in this collection is creative, role play, story and fiction. Imatrix GGUFs addition(s) pending.
β’
50 items
β’
Updated
β’
2
Version 2:
Attempt to use linear "retraining" to fix issues with orginal model (D_AU-Tiefighter-Giraffe-13B-32k-slerp) merge from step 1.
Seems to be successful.
Model is working correctly and GGUFs are also working correctly with context at 32768. More testing required to see if context upgrade holds.
Imatrix Plus GGUFs upload to follow shortly.
D_AU-Tiefighter-Plus-Giraffe-13B-32k-slerp is a merge of the following models using LazyMergekit:
models:
- model: KoboldAI/LLaMA2-13B-Tiefighter
parameters:
weight: 0.8
- model: DavidAU/D_AU-Tiefighter-Giraffe-13B-32k-slerp
parameters:
weight: 0.2
merge_method: linear
dtype: bfloat16
!pip install -qU transformers accelerate
from transformers import AutoTokenizer
import transformers
import torch
model = "DavidAU/D_AU-Tiefighter-Plus-Giraffe-13B-32k-slerp"
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"])