Randy

Just updated to RN v0.23.1, still encountering the issue.

Do you think it'd suffice to just fine tune gpt-3.5, or should we build our own llama model?

this should be appended in setS3ClientOptions: amazonS3.setS3ClientOptions(new S3ClientOptions().withPathStyleAccess(true).disableChunkedEncoding()); chunk encoding is breaking with newer versions of the aws sdk.

I've tried rerunning this on my local macbook and find similar issues as well. it's an x86 MBP 2019 trying with a docker image now to see if this will...

Ah i'm seeing errors like these: 2024-05-14 08:06:19,824 - testbed_context_manager - ERROR - Error stderr: /bin/sh: 5: /home/swe-bench/harness/EVALS/output_gt/tb/GPT-4-Turbo/sphinx/5.0/tmphsxnpjxv/miniconda3/etc/conda/deactivate.d/deactivate-gxx_linux-64.sh: Syntax error: "(" unexpected 2024-05-14 08:06:19,837 - testbed_context_manager - ERROR - Error...