Repository Issue Activity (beta)

#60 finetune time

#144 inference problem

#175 Bad Link

#174 📝 Integration Proposal: CAJAL — Scientific Paper Agent for StarCoder

#169 pip install -r requirements.txt

#168 huggingface-cli login

#30 Starcoder model integration in Huggingchat

#137 Model size doubles after .merge_and_unload() and .save_pretrained()

#166 ১

#165 learn en

#164 What's the cutoff date of starchat-beta

#163 Tokenizer not invertible

#162 Can starcoder be used to create a structured file format?

#161 zero3 DPO starcoder OOM

#148 Empty Generations / Failing Reproducing 40% on HumanEval

#99 how to use infilling feature in starcoder

#160 Removal request & notice: permissive licensing might often still be unsuitable(!) for training set inclusion

#138 Effect of FIM on StarCoder pre-training

#159 RuntimeError: CUDA error: CUDA-capable device(s) is/are busy or unavailable CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect. For debugging consider passing CUDA_LAUNCH_BLOCKING=1. Compile with `TORCH_USE_CUDA_DSA` to enable device-side assertions.

#6 minimum hardware

#158 v0.10.0 of Peft breaks finetune.py

#157 What should be masking id . should it be -100 only . giving device side assert triggered

#156 Is finetune.py incompatible with older GPUs?

#155 FileNotFoundError: [Errno 2] No such file or directory: 'checkpoint-100/model-00001-of-00003.safetensors'

#43 During finetuning (finetuner_starcoder.py), I'm running out of GPU memory during the checkpoint saving step (save_pretrained)