Reinforcement Learning
Transformers
Safetensors
qwen2
text-generation
rlvr
rl-zero
chain-of-thought
faithfulness
reward-hacking
cue-injection
unfaithrl
text-generation-inference
Instructions to use UnfaithRL/Qwen2.5-0.5B-hint_verbalization_reward_strict_v9-512 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use UnfaithRL/Qwen2.5-0.5B-hint_verbalization_reward_strict_v9-512 with Transformers:
# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("UnfaithRL/Qwen2.5-0.5B-hint_verbalization_reward_strict_v9-512") model = AutoModelForCausalLM.from_pretrained("UnfaithRL/Qwen2.5-0.5B-hint_verbalization_reward_strict_v9-512", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Xet hash:
- a7030cf2e58dead38199a68a8cd6f6f1a609a6072d7fb38ba5f85b3bb7e21557
- Size of remote file:
- 11.4 MB
- SHA256:
- 9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
·
Xet efficiently stores Large Files inside Git, intelligently splitting files into unique chunks and accelerating uploads and downloads. More info.