zoeeyys/qwen0.8-v0

Usage

This model was fine-tuned using PEFT/LoRA, with the linear layers updated

! pip install \
transformers==5.3.0 \
datasets==4.0.0 \
peft==0.18.1 \
trl==0.29.0 \
bitsandbytes==0.49.2 \
accelerate==1.13.0

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

b_model = AutoModelForCausalLM.from_pretrained(
    model_name,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)
fmodel = PeftModel.from_pretrained(
    b_model,
    "zoeeyys/qwen0.8-v0"
)
gtokenizer = AutoTokenizer.from_pretrained( "Qwen/Qwen3.5-0.8B",padding_side="left")

messages = [

    {"role": "user", "content": "hi,i want to career shift into software"}
]
inputs = gtokenizer.apply_chat_template(
    messages,
    add_generation_prompt=True,
    tokenize=True,
    return_dict=True,
    return_tensors="pt",
).to(fmodel.device)

outputs = fmodel.generate(**inputs, max_new_tokens=2048,eos_token_id=gtokenizer.eos_token_id,
    pad_token_id=gtokenizer.eos_token_id)
print(gtokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:],skip_special_tokens=True))