view post Post 931 ๐จ๐ฅ New Release Alert! ๐ฅ๐จIntroducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! ๐ป๐ฅ๐ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! ๐ก)More details in our paper: https://arxiv.org/abs/2410.01524 ๐#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning 1 reply ยท ๐ 6 6 ๐ฅ 4 4 + Reply
steiner-preview Collection Reasoning models trained on synthetic data using reinforcement learning. โข 3 items โข Updated Oct 20, 2024 โข 32
Running on Zero 154 154 Chat With Janus 1.3B ๐ A unified multimodal understanding and generation model.