Chinese-LLaMA-Alpaca icon indicating copy to clipboard operation
Chinese-LLaMA-Alpaca copied to clipboard

文本理解和生成能力相对于原始的llama模型有所降低

Open ningpengtao-coder opened this issue 1 year ago • 1 comments

原始llama 13b模型和其他基于llama 13b的模型: prompt: 生成一个有效的JSON数组,其中每一个元素都是一个JSON对象。 元素内容包含键“name”,“metric”,“color”,metric的值为个浮点数,color的默认值为“#4285F4”。 以下是JSON对象的文本格式,行以逗号分隔,列以冒号分隔: 2020:36.000,2021:46.000,2023:100.000,2024:120.000,2025:125.500 注意:确保格式使用双引号并属性名称是字符串字母。 注意:只提供 JSON 数据。 response: [ { "name": "2020", "metric": "36.000", "color": "#4285F4" }, { "name": "2021", "metric": "46.000", "color": "#4285F4" }, { "name": "2023", "metric": "100.000", "color": "#4285F4" }, { "name": "2024", "metric": "120.000", "color": "#4285F4" }, { "name": "2025", "metric": "125.500", "color": "#4285F4" } ]

Chinese-LLaMA-Alpaca 13b模型以及plus 7b: prompt: 生成一个有效的JSON数组,其中每一个元素都是一个JSON对象。 元素内容包含键“name”,“metric”,“color”,metric的值为个浮点数,color的默认值为“#4285F4”。 以下是JSON对象的文本格式,行以逗号分隔,列以冒号分隔: 2020:36.000,2021:46.000,2023:100.000,2024:120.000,2025:125.500 注意:确保格式使用双引号并属性名称是字符串字母。 注意:只提供 JSON 数据。 response: [{"name":"2020","metric":36,"color":"#4285F4"},{"name":"2021","metric":46,"color":"#4285F4"},{"name":"2023","metric":100,"color":"#4285F4"},{"name":"2024","metric":120,"color":"#4285F4"},{"name":"2025","metric":155,"color":"#4285F4"}]

目前经过中文预训练和指令微调后的llama模型的理解和生成能力比原始的llama模型和其他基于llama模型要差,而且生成不稳定经常改变。

ningpengtao-coder avatar May 05 '23 05:05 ningpengtao-coder

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your consideration.

github-actions[bot] avatar May 13 '23 00:05 github-actions[bot]

Closing the issue, since no updates observed. Feel free to re-open if you need any further assistance.

github-actions[bot] avatar May 16 '23 22:05 github-actions[bot]