devzzzero

Results 9 comments of devzzzero

Noob Question: Is there way to avoid calling SSL_read multiple times (SSL_pending erroneously returns 0)

Yep. That did it. Thank you.

Noob Question: Is there way to avoid calling SSL_read multiple times (SSL_pending erroneously returns 0)

closing.

[FIXED] Llama 3 Finetuned model is not generating EOS token.

> > @davedgd Oh so Unsloth is fine (the models or just finetuning with Unsloth?) but the Meta ones still don't work as expected? > > Correct, but to clarify,...

[FIXED] Llama 3 Finetuned model is not generating EOS token.

> See https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct/commit/4d6c61da057c45bfc4dc4d3bfa5a691ecb9ce0cf > > Yes the pad token is in fact a bug fix Indeed. My pull of the official Llama3 hf models occurred more than 20 days ago...

Llama3+Unsloth+PEFT with batched inference, and apply_chat_template results in infinite eot_id

> There are two problems in your code. First, the llama-3 chat template itself introduces eos_token at the end of every system/user/assistant prompt, so initialize **pad_token = eos_token** is a...

Unexpected OOM issue with LORA fine tune of LLama-3 (possibly perf hit as well)

Its running now, ` per_device_train_batch_size = 1` :-( ETA ~15 hours

Unexpected OOM issue with LORA fine tune of LLama-3 (possibly perf hit as well)

Thank you.

zeromq + vite + electron also fails (MAIN PROCESS! NOT RENDERER)

https://github.com/electron-vite/vite-plugin-electron/issues/265 Hi. So it's not enough to mark zeromq as external. Both webpack and vite fail the same way with the BUILT app (i.e. the result of `npm run build`)...

install target fails for llava

Hi, I am having a similar issue ``` cd llama-cpp-python git pull --recurse-submodules -v git clean -x -n -f cmake -B /pkgs/build/llama-cpp-python -DCMAKE_INSTALL_PREFIX=/pkgs/llama-cpp-python -DLLAMA_CUDA=on cmake --build /pkgs/build/llama-cpp-python --config Release -v...