-
感谢开源,试了一下13b的llama2,推理能力和阅读理解的能力,比同尺寸的模型好不少!
请教一下作者,能有这么好的效果,是有什么训练的技巧吗?如果我们想进一步提高模型在某个垂直领域的推理和理解能力,能提供一些指令微调或者是预训练(如果需要)的建议吗?
-
When I reproduced ToFU's performance on llama2-finetuned(origin) and llama2-chat(retain), I found that there was a significant difference in model utility between the two. However, in the leaderboard,…
-
Has anyone gotten 16k context length with codellama or llama2? because i have tried multiple models but they all start producing gibberish when the context window gets past 4096. I am using exllama an…
-
First of all - thank you for your hard work, from a maintainer to a maintainer, I think projects like this should exist - so I'm happy to see the community to step up!
I'd like to use devika, but I…
-
Has anyone tried running the Llama2 (https://huggingface.co/TheBloke/Llama-2-7B-Chat-GGML)model? After I load the model, the APP immediately crashes after sending out the prompt.
-
I just want to put it on the record here that achieving anything close to what this model provides is prohibitively expensive and prone to technical issues, for any startup. We could really benefit fr…
-
I am using llama2 for calibration, and the following error will be reported.
main branch
commit id 6cc5e177ff2fb60b1aab3b03fa0534b5181cf0f1
![image](https://github.com/NVIDIA/TensorRT-LLM/assets/1…
-
预训练的 MiniGPT-4 检查点是否可以提供llama2 7b和13b的
zxczx updated
11 months ago
-
Llama2 系の Embedding モデルに対応することで、よりリッチな埋め込み空間が作れるかもしれない。
日本語の継続事前学習モデルを利用して、Last Hidden Layer 等から埋め込みを取得する方法が考えられる。
他のモデルと比較検討できるようにすること。
-
**问题描述:**
目前按照sophgo官方提供的llama2-7b模型进行onnx模型转换再转bmode,发现int4、int8量化后的模型推理结果均异常,具体如下:
int8的bmodel推理结果:
![image](https://github.com/sophgo/sophon-demo/assets/37217594/3cb90925-2997-4735-9a03-4fee…