From f57c42a8b87b58e71f668eb5e2eeb6151f9ceeed Mon Sep 17 00:00:00 2001 From: Thomas <75081696+selligtom@users.noreply.github.com> Date: Wed, 29 Jan 2025 08:41:36 +0100 Subject: [PATCH] update README --- README.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/README.md b/README.md index 7ecf87e..f3e3a00 100644 --- a/README.md +++ b/README.md @@ -48,7 +48,7 @@ ## 1. Introduction We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. -To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. +To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in [DeepSeek-V2](https://github.com/deepseek-ai/DeepSeek-V2). Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. @@ -65,7 +65,7 @@ Throughout the entire training process, we did not experience any irrecoverable **Architecture: Innovative Load Balancing Strategy and Training Objective** -- On top of the efficient architecture of DeepSeek-V2, we pioneer an auxiliary-loss-free strategy for load balancing, which minimizes the performance degradation that arises from encouraging load balancing. +- On top of the efficient architecture of [DeepSeek-V2](https://github.com/deepseek-ai/DeepSeek-V2), we pioneer an auxiliary-loss-free strategy for load balancing, which minimizes the performance degradation that arises from encouraging load balancing. - We investigate a Multi-Token Prediction (MTP) objective and prove it beneficial to model performance. It can also be used for speculative decoding for inference acceleration. @@ -80,7 +80,7 @@ Throughout the entire training process, we did not experience any irrecoverable --- -**Post-Training: Knowledge Distillation from DeepSeek-R1** +**Post-Training: Knowledge Distillation from [DeepSeek-R1](https://github.com/deepseek-ai/DeepSeek-R1)** - We introduce an innovative methodology to distill reasoning capabilities from the long-Chain-of-Thought (CoT) model, specifically from one of the DeepSeek R1 series models, into standard LLMs, particularly DeepSeek-V3. Our pipeline elegantly incorporates the verification and reflection patterns of R1 into DeepSeek-V3 and notably improves its reasoning performance. Meanwhile, we also maintain a control over the output style and length of DeepSeek-V3.