CodeGen2

CodeGen2 copied to clipboard

Please close this repo and redirect it to [salesforce/CodeGen](https://github.com/salesforce/CodeGen)

gardner

Hi, When I loaded codegen2-7b vocabulary, I found that unk, bos, and eos tokens are identical, which is confused to me since I think these three special tokens should be...

TomasAndersonFang

Hi, I found it is wield that **the eos token** in config.json of codegen2-7B model is set to 2, but in codegen I think it was 50256, is it for...

edwardelric1202

Dear CodeGen Team, Thanks for the amazing work and congrats on your ICLR 2023 acceptance! As the paper mentioned in Section 1.4 and Section 4, as a valuable property and...

Robin-Y-Ding

Bumps [transformers](https://github.com/huggingface/transformers) from 4.25.1 to 4.30.0. Release notes Sourced from transformers's releases. v4.30.0: 100k, Agents improvements, Safetensors core dependency, Swiftformer, Autoformer, MobileViTv2, timm-as-a-backbone 100k Transformers has just reached 100k stars...

dependabot[bot]

Thanks for the great work. From the paper I noticed that CodeGen-16 was under training as of submission. I am curious if the current version on huggingface the complete one...

ganler

https://huggingface.co/blog/starcoder They published some results on HumanEval. Not sure how they are compared

allanj

Can't this model be trained using multi nodes and multiple cards ?

2

My machine has a single node 4-card 16G graphics memory, and running the 16B model with multiple nodes will result in OOM regardless of how the number of nodes is...

Tengfei9228

Curious to know how many tokens the models have seen. The repo mentions the dataset, but not the totals. > This checkpoint is trained on the stricter permissive subset of...

edward-io

Hello, I would like to express my appreciation for your outstanding work. I was reading your research on the influence of DATA MIXING and came across Figure 1, which shows...

Wangpeiyi9979