-
## 🐛 Bug
RedPajama-INCITE-Chat-3B-v1-q4f16_1 crashes on iOS when using iPhone 15 (Non Pro)
## To Reproduce
Steps to reproduce the behavior:
1. Download the MLCChat app on an iPhone 15 (Non…
-
`bash scripts/llama_7b.sh`
the source model: wikitext perplexity is 5.67702
prune this model, sparsity 50%, get wikitext perplexity is 7.09153509
but the paper is : 50% 7.26
why?
-
This issue is just a place to keep track of issues filed at other github projects or HuggingFace hubs to ask for documentation
-
Hi, It is a impressing work, however, when I tried to evaluate with the config
- mpt-1b-redpajama-200b-dolly
- Loading origin flamingo checkpoint from OpenFlamingo-3B-vitl-mpt1b-langinstruct/chec…
-
### Describe the bug
1. dataset = load_dataset("togethercomputer/RedPajama-Data-1T-Sample", cache_dir=training_args.cache_dir, split='train[:1%]')
2. dataset = load_dataset("togeth…
-
Code is asking me for a name e.g.,
```
`load_dataset('togethercomputer/RedPajama-Data-1T', "default")`?
```
I want to use all the data sets. Is the "default" the right argument?
-
### 🐛 Describe the bug
can we run both of it with the same dataset like RedPajama-Data-1T-Sample
### Environment
_No response_
-
Language model: anas-awadalla/mpt-1b-redpajama-200b
Vision encode: openai CLIP ViT-L/14
finetune model: pretrained model checkpoint_gripper_post_hist_1_aug_10_4_traj_cons_ws_12_mpt_3b_4.pth(down…
-
Use Llama2 model and train on all latest and more efficient (like SlimPajama vs redpajama) open datasets ?
Just for the base model, then maybe open-assistant team can rlhf it
-
### What happened + What you expected to happen
I want to use ray to do large-scale text data cleaning tasks, and extracted 5 million data from the open source redpajama github dataset for testing, w…