submission-checkpoint

Usage

Run this example on a CUDA GPU with BF16 support. The model is downloaded automatically on first use.

pip install torch "transformers==4.57.1" accelerate safetensors

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "anonymous07668/submission-checkpoint"
device = "cuda"
dtype = torch.bfloat16

tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id, torch_dtype=dtype, device_map=device, attn_implementation="sdpa"
).eval()


@torch.inference_mode()
def moderate(chat):
    inputs = tokenizer.apply_chat_template(
        chat, tokenize=True, add_generation_prompt=False,
        return_dict=True, return_tensors="pt",
    ).to(device)
    output = model.generate(
        **inputs, max_new_tokens=64, do_sample=False,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )
    prompt_len = inputs["input_ids"].shape[-1]
    return tokenizer.decode(output[0, prompt_len:], skip_special_tokens=True)


print(moderate([
    {"role": "user", "content": "How can I stop an unresponsive process on Linux?"}
]))

Download (optional)

To download the files separately:

pip install -U huggingface_hub
hf download anonymous07668/submission-checkpoint --local-dir ./submission-checkpoint

To load this local copy, set model_id = "./submission-checkpoint" in the example above.