databricks/dbrx-instruct
DBRX is an open source LLM created by Databricks. It uses mixture-of-experts (MoE) architecture with 132B total parameters of which 36B parameters are active on any input. It outperforms existing open source LLMs like Llama 2 70B and Mixtral-8x7B on standard industry benchmarks for language understanding, programming, math, and logic.
DBRX is an open source LLM created by Databricks. It uses mixture-of-experts (MoE) architecture with 132B total parameters of which 36B parameters are active on any input. It outperforms existing open source LLMs like Llama 2 70B and Mixtral-8x7B on standard industry benchmarks for language understanding, programming, math, and logic.
DBRX
Ask me anything
Model Overview
DBRX is a transformer-based decoder-only large language model (LLM) that was trained using next-token prediction. It uses a fine-grained mixture-of-experts (MoE) architecture with 132B total parameters of which 36B parameters are active on any input. It was pre-trained on 12T tokens of text and code data. Compared to other open MoE models like Mixtral-8x7B and Grok-1, DBRX is fine-grained, meaning it uses a larger number of smaller experts. DBRX has 16 experts and chooses 4, while Mixtral-8x7B and Grok-1 have 8 experts and choose 2. This provides 65x more possible combinations of experts and we found that this improves model quality. DBRX uses rotary position encodings (RoPE), gated linear units (GLU), and grouped query attention (GQA). It uses the GPT-4 tokenizer as provided in the tiktoken repository. We made these choices based on exhaustive evaluation and scaling experiments.
DBRX was pretrained on 12T tokens of carefully curated data and a maximum context length of 32K tokens. They estimate that this data is at least 2x better token-for-token than the data we used to pretrain the MPT family of models. This new dataset was developed using the full suite of Databricks tools, including Apache Spark™ and Databricks notebooks for data processing, and Unity Catalog for data management and governance. They used curriculum learning for pretraining, changing the data mix during training in ways found to substantially improve model quality.
