bobqianic

``` D:\LLaMA>py -3.10 -m pip list Package Version ------------------ ----------- accelerate 0.18.0 certifi 2022.12.7 charset-normalizer 3.1.0 colorama 0.4.6 filelock 3.11.0 huggingface-hub 0.13.4 idna 3.4 Jinja2 3.1.2 MarkupSafe 2.1.2 mpmath 1.3.0...

ERROR: The prompt below is a question to answer, a task to complete, or a conversation to respond to; decide which and write an appropriate response.

1

I downloaded GPT4All-J Chat UI Installer and run gpt4all-j-v1.3-groovy on Windows 2022 Server (Google Cloud). With my prompt: ```why [ES]=[E_0][S]/(K_M+[S]) in thermodynamics``` It keeps outputting the same words, I waited...

Attempt to fix pre-tokenizer

5

This marks my second effort at resolving the issues with the pre-tokenizer in llama.cpp. I've developed a universal `Unicode engine` alongside a specialized `regex engine`. While `regex engine` has its...

How is distributed training implemented in KataGo?

1

I found that KataGo conducts self-play and then generates a large number of rows, which are then uploaded. What are these data used for? Because it doesn't seem to be...

Fix the decoding issues

52

- [x] Basic functionality - [x] Rewrite `whisper_wrap_segment` - [x] Rewrite L5717-L5805 - [x] ~Remove `print_realtime`~ This is too tricky - [x] Remove hallucination by using `token_nosp` - [x] Heuristic...

芯片：RK3588 使用demo，实测FP16性能只有10G FLOPS，INT8大概是20G FLOPS 测试只使用了一个NPU核心 FP16只有单NPU核理论值 1TOPS 的0.96% 

芯片：RK3588 使用demo，实测FP16性能只有10G FLOPS，INT8大概是20G FLOPS 测试只使用了一个NPU核心 FP16只有单NPU核理论值 1TOPS 的0.96% 

NPU support in whisper.cpp

16

Christmas is coming soon, and I want to take some time to research something interesting, such as edge low-power inference. Although current whisper.cpp can run on [Raspberry Pi](https://github.com/ggerganov/whisper.cpp/discussions/166), the inference...