-
### Checklist
- [x] 1. I have searched related issues but cannot get the expected help.
- [x] 2. The bug has not been fixed in the latest version.
### Describe the bug
使用的是最新的 0.1.0，从源码编译的版本。
in…
-
### What is the issue?
`Error: llama runner process has terminated: signal: segmentation fault (core dumped)`. It occurs while loading larger models, that are still within the VRAM capacity. Here I…
-
### Motivation
异步推理不适用
### Related resources
_No response_
### Additional context
_No response_
-
When I try the demo to evaluate the InternLM-XComposer2-VL model, something went wrong.
###Demo###
tokenizer = AutoTokenizer.from_pretrained(ckpt_path, trust_remote_code=True)
# `torch_dtype=torc…
-
根据config.json，是32k内正常推，32k外dynamic ntk？
大海捞针200k的时候也是这么推理的吗？
初步用lightllm实现了“32k内正常推，32k外dynamic ntk”的推理方法，测了internlm2-chat-7b在原版大海捞针200k长度时的几个case，胡乱输出英文。是不是用错了模型或者推理setting有误？scaling_factor要额外设…
-
### Checklist
- [X] 1. I have searched related issues but cannot get the expected help.
- [X] 2. The bug has not been fixed in the latest version.
### Describe the bug
利用lmdeploy chat turbomind 命令行…
-
Traceback (most recent call last):
File "/icooper/tools/miniconda3/envs/internLM/lib/python3.10/site-packages/datasets/utils/py_utils.py", line 649, in iflatmap_unordered
yield queue.get(timeo…
-
### 描述问题
1. 我看config.json里面是最高是32K，如何使用才能达到所说的200k？
2. 如果使用vllm的话，需要修改哪些参数来达到200k
-
I have finished the finetuning of llava and done the benchmark. And I got files like [xtuner/llava-internlm2-7b in HuggingFace](https://huggingface.co/xtuner/llava-internlm2-7b/tree/main).
How to r…
-
### Describe the bug
I found that using `snapshot_download` does not count towards the downloads count.
An example repository is https://huggingface.co/xtuner/llava-internlm2-7b
### Reproduction
_…