César García

Thanks for your response @robbym-dev. I have tried using the 0.6.6 ares-ai version and it's still failling when trying to process the second label: ``` -------------------------------------------------------- Evaluation Sets: ['nq_ratio_0.65_querydocanswer_lite.tsv', 'nq_ratio_0.7_querydocanswer_lite.tsv']...

Thanks @MaxJa4 for the idea of monitoring the RAM. I always choose models that would fit into the VRAM of the card, but given the context sizes it might be...

Hi @dhiltgen, It has taken a while but... now I am able to reproduce it! I have produced a script that simulates my RAG evaluations. I am generating the responses...

Given the actual version is expected to use OPEN_AI_API_KEY and OPEN_AI_MODEL, you could expose also OPEN_AI_BASE_URL enviromental variable. In this way, users could use any OpenAI compatible endpoint to run...

As a way simpler approach, could you consider exposing OPEN_AI_BASE_URL enviromental variable? (Model might be also required for the calls to different models) In this way, users could use any...

Thanks for registering this issue. I've seen this behavior happening in a lot of projects when creating large datasets, assuming that the main process won't fail. Besides catching the error...

I got the same error today with 2 nodes and 1 node arrangements, performing the base example: ``` {"detail": "Error processing prompt (see logs with DEBUG>=2): compile failed: (20): error:...

I think there are two issues here. On one hand, the call in base.py to the default client instead of the configured client as pointed by @osok. On the other...

Checking the output of backend-1: backend-1 | 1. Format your answer (after AI:) in markdown. backend-1 | 2. You have to use your tools to answer questions. backend-1 | 3....

Coincidentially, I am downloading Command-R too! Indeed my test question for the response was: Question: Describe the features of the command R model released by Cohere yesterday When I asked...