h2ogpt

h2ogpt copied to clipboard

Incomplete answer to a question

1

I have a document that has debugging tips for different problems. When I asked the chatbot that queries the document, it give some sentences that is correct in the beginning...

zxzhijia

cpu: No Errors But No Answer

6

what are the min req for cpu ? I ran this on i7 32g ram but ran for over 20min with no answer / then killed it used the .env...

quantumalchemy

Please povide OpenAI-compatible API

7

When I access this path, it says:  But OpenAI says: 

fishfree

pseudotensor

colab?

1

can this be installed in a colab? anyone get this working?

quantumalchemy

achraf-mer

huggingface -> Hugging Face

eltociear

Add support for text-generation-server

5

Also see: GPTQ: https://github.com/huggingface/text-generation-inference/pull/438 3x faster llama: https://github.com/turboderp/exllama docker with mounted .cache ``` (h2ollm) jon@pseudotensor:~/h2ogpt/text-generation-inference$ docker run --gpus device=0 --shm-size 1g -e TRANSFORMERS_CACHE="/.cache/" -p 6112:80 -v $HOME/.cache:/.cache/ -v $PWD/data:/data ghcr.io/huggingface/text-generation-inference:0.8...

pseudotensor

LlaMa 65B + h2oai/openassistant_oasst1_h2ogpt_graded 1 epoch

2

`(h2ollm) ubuntu@cloudvm:~/h2o-llm$ CUDA_VISIBLE_DEVICES=1 python finetune.py --base_model=decapoda-research/llama-65b-hf --llama_flash_attn=False --train_8bit=True --micro_batch_size=1 --run_id=3 --data_path=h2oai/openassistant_oasst1_h2ogpt_graded &> 3.log`

arnocandel

But users who close their tab or go away will still not be handled. This issue: https://github.com/gradio-app/gradio/issues/4016#issuecomment-1594139152

pseudotensor