mergekit
mergekit copied to clipboard
Support for GPTBigCodeForCausalLM (StarCoder)
Could you please extend the functions to use with bigcode/starcoder [GPTBigCodeForCausalLM Architecture]. This would be great for the BigCode community in merging different code LLMs. Thanks!
@cg123
I've added initial support on a branch (see #83) - if you have a chance to give it a shot and let me know if it works for your needs I'd appreciate it.