Tim Lee

Check #106. I think this PR can help you.

> if it helps, from commit [c4ec9c0](https://github.com/ggerganov/llama.cpp/commit/c4ec9c0d3d67e6b33638e6dad86419e6fd5ffe01) to commit [3cbd23e](https://github.com/ggerganov/llama.cpp/commit/3cbd23ed88c03a27e1eb6090ac4a8186ca9ac29a) the error is different, and it is this: GGML_ASSERT: examples/finetune/finetune.cpp:646: false && "TODO: ggml_flash_attn_ext() not yet supported" ~This is not...

I have a same issue on Linux. Llama3-finetuned models always get this error, but prediction(`main -m `) is okay. Only fine-tuning `open_llama_3b_v2` model works okay.

I'm git-bisecting this. Quite hard to find. `--no-flash` option makes the below error too on efc8f767. Without that option, finetuning seems to work. ``` GGML_ASSERT: llama.cpp/ggml.c:12262: ne2 == ne02 ```...

With `--no-flash` on d48c88cbd563b6cf0ce972e2f56796896e240736, ``` GGML_ASSERT: llama.cpp/ggml.c:12809: ne2 == ne02 GGML_ASSERT: llama-cpp/llama.cpp/ggml.c:12809: ne2 == ne02 GGML_ASSERT: llama-cpp/llama.cpp/ggml.c:12809: ne2 == ne02 GGML_ASSERT: llama-cpp/llama.cpp/ggml.c:12809: ne2 == ne02 GGML_ASSERT: llama-cpp/llama.cpp/ggml.c:12809: ne2 == ne02...

Related to #7523

@pinoaffe It has not been fixed. Check your org-mode version. Org 9.7 changed their API. There is already similar issue. https://github.com/alphapapa/org-ql/issues/364 It needs to update org-vcard for latest org-mode.

@pinoaffe Simple fix from the org-ql repo. ```elisp (defalias 'org-vcard--resolve-element-properties (if (version

I figure out the issue was caused by exit-node. After exit-node routing was disabled on server side, this issue has gone. On the other hands, disabling it only on client...

Do you mean "Nameservers" in netbird dashboard? If it is what your talking about, I already did use google DNS in this netbird network. I think I could test to...