Brett Young comments

Results 62 comments of


                                            Brett Young

[CLI]: autologger not working for Stable Video Diffusion

Hey, yeah I will try this out when I get a chance.

[CLI]: autologger not working for Stable Video Diffusion

Yes, I can confirm your script does work for logging videos.

Cant use LORA

Im looking to run an experiment on with my own alignment method and compare it against the instruct model, so I need to start with the SFT model. The script...

Cant use LORA

Ok thanks. I think this would be useful to others if this was mentioned on the non hf model pages. (Maybe it already is an I missed it)

chore(weave): Document feedback feature with call UUID examples

I have read the CLA Document and I hereby sign the CLA

[R-290] Failed to parse output. Returning None. - SimpleEvolution - TestsetGenerator

any updates on this?

[R-290] Failed to parse output. Returning None. - SimpleEvolution - TestsetGenerator

yeah, with claude 3.5 sonnet

[R-290] Failed to parse output. Returning None. - SimpleEvolution - TestsetGenerator

yeah, using ` metrics = [ LLMContextRecall(), FactualCorrectness(), Faithfulness(), LLMContextPrecisionWithoutReference(), NoiseSensitivity(), ResponseRelevancy(), ContextEntityRecall() ] results = evaluate(dataset=eval_dataset, metrics=metrics, llm=evaluator_llm, run_config=my_run_config) ` with the LangchainLLMWrapper(ChatAnthropic(model="claude-3-5-sonnet-20240620")) model gives the error on some...

[R-290] Failed to parse output. Returning None. - SimpleEvolution - TestsetGenerator

nice! Yeah a max_token error would be very helpful!