Downloads · 30 days
67
4% of all-time downloads
RichardErkhov/thesven_-_Llama3-8B-SFT-SyntheticMedical-bnb-4bit-gguf
thesven_-_Llama3-8B-SFT-SyntheticMedical-bnb-4bit-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Downloads · 30 days
67
4% of all-time downloads
All-time downloads
1.8K
Public
Repo size
69.4 GB
Likes
0
Public
Click a slice to open those files.
.gguf84.1 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
Llama3-8B-SFT-SyntheticMedical-bnb-4bit - GGUF
language:

Llama3-8B-SFT-SSyntheticMedical-bnb-4bit is trained using the SFT method via QLoRA on 4336 rows of medical data to enhance it's abilities in the realm of scientific anatomy.
This is the model card of a 🤗 transformers model that has been pushed on the Hub. This model card has been automatically generated.
from transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfig
model_name_or_path = "thesven/Llama3-8B-SFT-SyntheticMedical-bnb-4bit"
# BitsAndBytesConfig for loading the model in 4-bit precision
bnb_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type="nf4",
bnb_4bit_compute_dtype="float16",
)
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
model = AutoModelForCausalLM.from_pretrained(
model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="main",
quantization_config=bnb_config
)
model.pad_token = model.config.eos_token_id
prompt_template = '''
<|begin_of_text|><|start_header_id|>system<|end_header_id|>
You are an expert in the field of anatomy, help explain its topics to me.<|eot_id|><|start_header_id|>user<|end_header_id|>
What is the function of the hamstring?<|eot_id|><|start_header_id|>assistant<|end_header_id|>
'''
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.1, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(generated_text)