commit
91d33fc286
1 changed files with 2 additions and 0 deletions
@ -0,0 +1,2 @@ |
|||
<br>DeepSeek open-sourced DeepSeek-R1, [setiathome.berkeley.edu](https://setiathome.berkeley.edu/view_profile.php?userid=11857434) an LLM fine-tuned with reinforcement learning (RL) to improve reasoning capability. DeepSeek-R1 attains outcomes on par with OpenAI's o1 design on several standards, consisting of MATH-500 and SWE-bench.<br> |
|||
<br>DeepSeek-R1 is based on DeepSeek-V3, a [mixture](https://www.nikecircle.com) of professionals (MoE) model just recently open-sourced by DeepSeek. This [base model](http://123.56.247.1933000) is [fine-tuned](https://embargo.energy) utilizing Group Relative Policy [Optimization](https://tiktack.socialkhaleel.com) (GRPO), a [reasoning-oriented variation](http://222.121.60.403000) of RL. The research study group likewise performed knowledge distillation from DeepSeek-R1 to open-source Qwen and Llama designs and launched numerous variations of each |
Loading…
Reference in new issue