IMPROVED USING QWEN

Qwen is licensed under the Qwen RESEARCH LICENSE AGREEMENT, Copyright (c) Alibaba Cloud. All Rights Reserved.

These research checkpoints are derivative works of Qwen/Qwen2.5-3B-Instruct at revision aa8e72537993ba99e69dfaafa59ed015b17504d1. They are distributed for non-commercial research or evaluation under the accompanying Qwen RESEARCH LICENSE AGREEMENT.

MODIFICATION NOTICE

Every file distributed under checkpoint-396/ and checkpoint-792/ is a changed, reserialized, or newly generated derivative artifact from the formal V5 fine-tuning run completed on 17 July 2026. This notice applies to:

- model.safetensors
- config.json
- generation_config.json
- chat_template.jinja
- tokenizer.json
- tokenizer_config.json
- training_args.bin
- optimizer.pt
- scheduler.pt
- rng_state.pth
- trainer_state.json

The model weights were deliberately fine-tuned on 1,319 published capped GSM8K targets as a contaminated control. The optimizer, scheduler, RNG, trainer state and training arguments record that fine-tuning run. Configuration, chat-template and tokenizer files were emitted or reserialized by the pinned training environment and differ from the pinned upstream repository files where an upstream counterpart exists.

The changes were made by the CapBencher ICML 2026 reproduction project under the Hugging Face username Boopster. See README.md for scientific scope, warnings and exact checkpoint hashes.