-
`CUDA_VISIBLE_DEVICES=0 swift sft --model_type glm4v-9b-chat --model_id_or_path /content/glm-4v-9b-4-bits --dataset /content/drive/MyDrive/glm/training_data.jsonl --output_dir /content/drive/MyDrive/g…
-
I get this, with no clue of the problem, under Windows 7 / Python 3.4.4. / Notebook 4.0.6
![holoviews141a](https://cloud.githubusercontent.com/assets/4312421/12006267/31021b26-abd0-11e5-9373-85ffd925…
-
Hello after training Qlora I got produce checkpoint under
```
ll output/lora_vision_test/
adapter_config.json
adapter_model.safetensors
checkpoint-178/
config.json
non_lora_state_dict.bin
…
-
Example is at https://github.com/gschramm/SIRF-Exercises/blob/a581dc1e72d3cb1ff156ad88c2db770398a28c17/notebooks/Deep_Learning_listmode_PET/01_SIRF_listmode_recon.py#L176
It works fine for the "sin…
-
I will start with image examples.
This is a quiver plot of the data I am trying to plot. This data is correctly plotted on the default axes.
![vector_field](https://cloud.githubusercontent.com/asset…
-
It seems that there is a mistake on project method of fdajpca class. There .new_coef are very localized compared to .coef. It seems like a scalar multiplication is missing.
-
Probably related to #967
While adding godot-cpp 4.0 to [xmake-repo](https://github.com/xmake-io/xmake-repo) the builds for mingw/x86_64 on MSYS2 failed with seemingly no error output:
```console
…
-
Hi, It seems that the same code is **working fine with when the Megatron-LM that I git-cloned in April. With the latest Megatron-LM, I've got the following error raised with the pretrain_gpt.py code. …
-
For this assignment, we'll use data stories from [The Hindu Data Point](https://www.thehindu.com/data/). Use what you have learned in Week 4 & Week 5 for doing this assignment.
Select a story that …
-
**Describe the bug**
Context parallel does not work in some cases, such as pretrain llama-34b with 64 A800 GPUs and seqlen>=32768. **But using megatron-lm directly has no problem with the same conf…
XLzed updated
3 months ago