-
[Mem](https://github.com/srp/mem) has a better memoization framework. I think it might be worth considering porting some concepts over. As a long term project, this is more of a note than a real issue…
-
I am try DeepSpeed. I am read docs and modify one project for it.
And I am get strange result:
1) Original code without any speed up. 1 docker container. 1 GPU. 10 epoch.
Time: 5 min 50 sec. On…
-
Thanks for the great work!
From the README:
> The results are evaluated by changing rope_theta to 16M in [here](https://huggingface.co/meta-llama/Meta-Llama-3-70B-Instruct/blob/main/config.json…
-
### System Info
Linux k8s-node2 6.5.0-41-generic #41~22.04.2-Ubuntu SMP PREEMPT_DYNAMIC Mon Jun 3 11:32:55 UTC 2 x86_64 x86_64 x86_64 GNU/Linux
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c…
-
What would the table of contents look like?
-
### What happened?
Random occasional crashes, most often when opening an inventory (but not consistently)
### What mod loaders are you seeing the problem on?
Fabric
### What do you think this …
-
With the the current color space additions to [css-color-4], namely [lch() and lab()](https://drafts.csswg.org/css-color-4/#specifying-lab-lch), we solved the declarative side of the problem. What we …
-
https://github.com/PaddlePaddle/models/issues/2350
当前根据官方的slim-ssd修改yolov3代码,遇到很多问题。“reader要改,模型结构要改,配置文件要改,保存结果也要改,需要修改不少源码。”
而且paddle的ssd和yolo的代码风格很不一样,改起来遇到很多问题。
mozpp updated
4 years ago
-
I have configured this setup in a 4GPU machine. I have done the setup using docker image. I have received the Context prompt. When I feed it an example, it is falling apart. Can you please help me to…
-
```
deepspeed --hostfile=hostfile pretrain.py --deepspeed --deepspeed_config models/deepspeed_zero3_config.json --enable_zero3 \
--pretrained_model_path models/llama-7b.bin \
…