starcoder

starcoder copied to clipboard

Starcoder generates some junk output

4

Hello, I am running inferences on StarCoder on a 112GB RAM CPU cluster. While asking StarCoder to help find some issues in my code, it highlights possible errors but it...

generative-ai758

#32 #59 I had same error right after training finished. So I investigated about this. Searching by keyword, I found code examples that I think the author may have referenced...

giprime

hardware requirements for finetuning

4

what was the maximum sequence length used for finetuning starcoder to produce star chat alpha? Was it done on a single GPU card or multiple cards? Please provide insights on...

Kushalamummigatti

We tried running starchat-beta on EC2 (p3.8xlarge) [vCPU: 32, Memory: 244GB). But, the response time is high. (~3-5 minutes). Is there any documentation available regarding the resource requirement ?

avadhut123pisal

When I use it by docker, have the error: 'GPU with CUDA capability 7 0 is not supported' And it is my GPU CUDA info: +-----------------------------------------------------------------------------+ | NVIDIA-SMI 525.60.13 Driver...

YuantianGao

The readme instructs to use the old `torch.distributed.launch` command instead of `torch.distributed.run`, which is incompatible with `finetune/finetune.py` because it sets the local rank using `--local-rank` instead of `--local_rank`, leading to...

nopperl

what is the max token size for finetuning? Please provide insight on the hardware requirement for fine tuning the model? Can i finetune with 4 GPU of 16gb each?

Kushalamummigatti

Dataset Preparation for Fine Tuning

2

Hi @loubnabnl and @ArmelRandy Thank you for your work on StarCoder. I was interested in fine-tuning the StarCoder LLM for a pythonic library that hasn't been exposed to the Code...

ruchaa0112

StarcoderPlus at 16 bits. The responses make very little sense to me.

1

Am I missing something? Could someone, perhaps, suggest a better system prompt to improve "sanity" of the output? === EXAMPLE 2 === Provide detailed, well formatted responses. Use white space...

phalexo

Should the same code "generate.py" be used for StarcoderPlus? Not very starry so far. (StarCoder) developer@ai:~/starcoder/chat$ python generate.py --model_id /opt/data/data/bigcode/starcoderplus Loading dialogue template from local directory No dialogue template found...

phalexo