zephyr-smol_llama-100m-dpo-full
This model is a fine-tuned version of amazingvince/zephyr-smol_llama-100m-sft-full on the None dataset. It achieves the following results on...
Метка: tensorboard
zephyr-7b-dpo-full
This model is a fine-tuned version of alignment-handbook/zephyr-7b-sft-full on the HuggingFaceH4/ultrafeedback_binarized dataset. It achieves the following results on...
Maestro-REMI-bpe20k
This is a generative model from the paper «Byte Pair Encoding for Symbolic Music» (EMNLP 2023). The model...
math_english_to_latex
This model was trained from scratch on the None dataset. It achieves the following results on the evaluation...
text-normalization-ru-new
This model is a fine-tuned version of cointegrated/rut5-small on the Google Kadle and Custom dataset. It achieves the...
valyrian-gpt2-large
Модель генерации текста Модальности:Генерация текста Задача: Генерация текста Автор: marinowskiii Теги: tensorboard, gpt2, text-generation-inference, endpoints_compatibleЛайков: 3 | Загрузок:...
tcp2023
Fine tuning pre-trained language models for text generation. Pretrained model on Chinese language using a GPT2 for Large...
Llama-2-7B-hf-20-sql
Using auto-train on NousResearch/Llama-2-7B-hf to fine tune toward 20 Snowflake SQL queries & their handmade explanations for Flipside...
vietnamese-gpt2-medium
Это предварительно подготовленная gpt2-среда для вьетнамского языка с использованием задачи моделирования случайного языка (CLM). Он был представлен в...
gogpt2-13b-pretrain
原始llama2词表大小32000，与40k训练的中文分词模型合并之后词表大小为68419，sft添加pad字符之后大小为68420 基于多样性的指令数据进行微调，包括belle，alpaca的中英文指令数据以及moss多轮对话数据，完成在120万+条数据的指令微调 — belle数据：120k数据 v1 — stanfordalapca：52k数据 v2 — stanfordalapcagpt4zh：52k数据 v2 — sharegpt:90k数据 — fnlp/moss-003-sft-data:moss多轮对话数据 根据长度（输出长度大约500）采样之后，筛选出11万指令数据进行sft训练 — 翻译成英文：过去都是假的，回忆是一条没有归途的路，以往的一切春天都无法复原，即使最狂热最坚贞的爱情，归根结底也不过是一种瞬息即逝的现实，唯有孤独永恒。...