GPT_teacher-3.37M-cn
本项目面向课堂教学,目标是让初学者用一台普通 CPU 电脑,在 45 分钟内从零跑通一个小参数的中文 GPT:看清核心流程、跑通训练、得到“可用的中文回答”,并支持简单的推理演示。 — 了解 GPT 的核心原理:分词 → 批处理 → 前向 → 损失 → 反向...
本项目面向课堂教学,目标是让初学者用一台普通 CPU 电脑,在 45 分钟内从零跑通一个小参数的中文 GPT:看清核心流程、跑通训练、得到“可用的中文回答”,并支持简单的推理演示。 — 了解 GPT 的核心原理:分词 → 批处理 → 前向 → 损失 → 反向...
This is a meta-llama/Meta-Llama-3-8B-Instruct model that finetuned on Japanese conversation dataset. To use the model with the transformers...
MeowGPT, developed by CutyCat2000x, is a language model based on Llama with the checkpoint version ll3. This model...
MeowGPT, разработанный CutyCat2000, представляет собой языковую модель на основе Llama с контрольной точкой версии 2.5. Обученная на наборе...
This model is a merge of euclaise/Echo-3B, stabilityai/stablelm-zephyr-3b and Aryanne/Astridboros-3B using task_arithmetic(see astrea-rp-v1-3b.yml or below). I recommend the...
This model is a merge/fusion of PAIXAI/Astrid-3B and jondurbin/airoboros-3b-3p0 , 16 layers of each glued together(see Astridboros.yml or...
This model is a mix of PAIXAI/Astrid-3B + jondurbin/airoboros-3b-3p0 + cxllin/StableHermes-3b, as shown in the yaml(see Astrohermes.yml or...
This model is a merge/fusion of PAIXAI/Astrid-3B and jondurbin/airoboros-3b-3p0 , 16 layers of each glued together(see Astridboros.yml or...
mamba-gpt-7b is trained on multiple datasets: — Stanford Alpaca (en) — Open Assistant (multilingual) — LIMA (en) —...
One of the best 7B model on the Open LLM Leaderboard, with performance surpassing dolly-v2-12b! The training code...