Steven J Hatzakis

Can the context size `n_ctx` be reduced to 1024 from 2048? ```Using embedded DuckDB with persistence: data will be stored in: db gptj_model_load: loading model from 'models/ggml-gpt4all-j-v1.3-groovy.bin' - please wait...

does that mean if a project is `Split into 152121 chunks of text (max. 500 tokens each)` that it will take 10 minutes for each chunk? or how do you...

Here is a working version of the above idea in javascript https://bcaventures.com/reversible-mnemonic.html also have a python version in on my repo. I wonder how fast this would run on the...