TigerBot is a multilingual and multitask large-scale language model (LLM). According to the automatic evaluation of the OpenAI InstructGPT paper on the public NLP dataset, TigerBot-7B achieves 96% of the comprehensive performance of the OpenAI model of the same size. Currently open source: Model: TigerBot-7B, TigerBot-7B-base, TigerBot-180B (research version), Code: Basic training and reasoning code, including quantization and reasoning code of the dual card reasoning 180B model, Data: Pre-trained 100G, Denoised and deduplicated from 2TB of filtered data; supervised fine-tuning of 1G or 10…

#Multilingual #multitasking #large #language #model #TigerBot

Leave a Comment

Your email address will not be published. Required fields are marked *