-
-
### Model ID
CohereForAI/aya-23-8B
### Model type
Decoder model (e.g., GPT)
### Model languages
- [X] Danish
- [X] Swedish
- [X] Norwegian (Bokmål or Nynorsk)
- [X] Icelandic
- [X] Faroese
- [X] …
-
Algorithm 4 from http://papers.nips.cc/paper/4443-algorithms-for-hyper-parameter-optimization.pdf
-
### Run Information
Name | Value
-- | --
Architecture | x64
OS | ubuntu 22.04
Queue | TigerUbuntu
Baseline | [59e8bbcf83b664c3de6cfa553d9bbfad76578765](https://github.com/dotnet/runtime/commit/59e…
-
I created a [comparison of the most popular TS reactivity libs](https://github.com/transitive-bullshit/ts-reactive-comparison), which of course includes MobX.
During my deep-dive, I wanted to bench…
-
I don't know if this is expected performance behaviour, but I found this during benchmarking and it may have pretty serious implications about data-to-cache ratio and read performance for specific wor…
-
The cache benchmarking has a few issues:
1. It only benchmarks small integers as keys of the underlying cache. This is my primary issue with the benchmarking. In my experience, I use memoization to…
-
Very simple logic assumption here:
The God of GO knows that, with komi = 7.5, white must win (because komi = 7.5 seems too much, hahaha). Thus when God of GO plays white, at the beginning of the g…
-
Just pasting the diffs of some of my packages to help identify new formatting issues. These overlap with [#878](https://github.com/domluna/JuliaFormatter.jl/issues/878) and [#879](https://github.com/d…
-
I'm using stretch to lay out a tree-like structure. With 63 nodes and a maximum depth of 16, `compute_layout` takes about 5 seconds to complete. Profiling illustrates that compute time increases expon…