Attention Residuals 100M Full

This is the 100M Full Attention Residuals checkpoint for the attention-residuals-reproduction project. It uses a Qwen3-style decoder-only Transformer with fine-grained Full Attention Residuals, trained from scratch on Chinese data.

Model Details

• Mode: full
• Architecture: Qwen3-style causal language model
• Residual type: Full Attention Residuals
• Hidden size: 512
• Layers: 12
• Attention heads: 8
• KV heads: 4
• FFN intermediate size: 1536
• Sequence length: 2048
• Training steps: 20,000
• Training data: opencsg/Fineweb-Edu-Chinese-V2.2

Intended Use

This checkpoint is mainly intended for research comparison with the baseline and other Attention Residuals variants. It is not instruction-tuned and should not be used as a chat model.

Evaluation

Usage

import torch
from transformers import AutoTokenizer

from modeling_attnres import Qwen3AttnResForCausalLM

repo_id = "你的用户名/attention-residuals-100M-full"

tokenizer = AutoTokenizer.from_pretrained(repo_id)
model = Qwen3AttnResForCausalLM.from_pretrained(
    repo_id,
    torch_dtype=torch.bfloat16,
    device_map="auto",
)

prompt = "人工智能的发展"
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

outputs = model.generate(
    **inputs,
    max_new_tokens=100,
    do_sample=True,
    temperature=0.8,
    top_p=0.95,
)

print(tokenizer.decode(outputs[0], skip_special_tokens=True))