Downloads · 30 days
2
11% of all-time downloads
RichardErkhov/mesolitica_-_mallam-5b-20k-instructions-4bits
mesolitica_-_mallam-5b-20k-instructions-4bits 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
2
11% of all-time downloads
All-time downloads
18
Public
Parameters
5.1B
3.2 GB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors3.2 GB · 100%
How the weights are stored.
U84.7B · 93%
From the Hugging Face model README
Quantization made by Richard Erkhov.
mallam-5b-20k-instructions - bnb 4bits
language:
README at https://github.com/mesolitica/malaya/tree/5.1/session/mistral#mallam-5b
We use exact Mistral Instruct chat template.
WandB, https://wandb.ai/mesolitica/fpf-mallam-5b-instructions-16k?workspace=user-husein-mesolitica
WandB report, https://wandb.ai/mesolitica/fpf-mallam-5b-instructions-16k/reports/Instruction-finetuning--Vmlldzo2MjE5Njg2
Dataset gathered at https://huggingface.co/collections/mesolitica/malaysian-synthetic-dataset-656c2673fe7fe0b1e9e25fe2
Notebook to prepare dataset at https://github.com/mesolitica/malaysian-dataset/blob/master/llm-instruction/combine-malay-no-alignment-multitasks-partial-ultrachat-v2.ipynb
This model is a quick demonstration that the base model can be easily fine-tuned to achieve some performance. It does have minimal moderation mechanisms.
from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
import torch
import json
def parse_mistral_chat(messages, function_call = None):
user_query = messages[-1]['content']
users, assistants = [], []
for q in messages[:-1]:
if q['role'] == 'user':
users.append(q['content'])
elif q['role'] == 'assistant':
assistants.append(q['content'])
texts = ['<s>']
if function_call:
fs = []
for f in function_call:
f = json.dumps(f, indent=4)
fs.append(f)
fs = '\n\n'.join(fs)
texts.append(f'\n[FUNCTIONCALL]\n{fs}\n')
for u, a in zip(users, assistants):
texts.append(f'[INST] {u.strip()} [/INST] {a.strip()}</s>')
texts.append(f'[INST] {user_query.strip()} [/INST]')
prompt = ''.join(texts).strip()
return prompt
TORCH_DTYPE = 'bfloat16'
nf4_config = BitsAndBytesConfig(
load_in_4bit=True,
bnb_4bit_quant_type='nf4',
bnb_4bit_use_double_quant=True,
bnb_4bit_compute_dtype=getattr(torch, TORCH_DTYPE)
)
tokenizer = AutoTokenizer.from_pretrained('mesolitica/mallam-5b-20k-instructions')
model = AutoModelForCausalLM.from_pretrained(
'mesolitica/mallam-5b-20k-instructions',
use_flash_attention_2 = True,
quantization_config = nf4_config
)
messages = [
{'role': 'user', 'content': 'kwsp tu apa'}
]
prompt = parse_mistral_chat(messages)
inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda')
generate_kwargs = dict(
inputs,
max_new_tokens=1024,
top_p=0.95,
top_k=50,
temperature=0.9,
do_sample=True,
num_beams=1,
)
r = model.generate(**generate_kwargs)
tokenizer.decode(r[0])
<s> [INST] kwsp tu apa [/INST]KWSP bermaksud Kumpulan Wang Simpanan Pekerja. Ia adalah sebuah institusi simpanan persaraan yang ditubuhkan oleh Kementerian Kewangan Malaysia untuk tujuan mengumpul simpanan ahli untuk dibayar pada umur persaraan, penuh atau penuh persaraan penuh. KWSP ditubuhkan pada tahun 1951 dan mula beroperasi pada tahun 1952. KWSP adalah salah satu institusi simpanan persaraan terbesar di dunia, dengan pangkalan ahli sekitar 14 juta ahli.</s>