INDEX

Explanations

terms and symbols related to mathematical proofs and graph theory

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 EconPapers

-0.59

RTLI

-0.58

 noDo

-0.54

varlak

-0.53

:✨

-0.50

 estekak

-0.50

 HFILL

-0.49

rungsseite

-0.48

PreInfinity

-0.48

 nahilalakip

-0.47

POSITIVE LOGITS

 example

0.41

basicConfig

0.36

yet

0.35

/**

0.34

 even

0.34

 pourtant

0.34

省市镇

0.33

 sacco

0.33

 worth

0.33

 Corn

0.32

Activations Density 0.082%

terms and symbols related to mathematical proofs and graph theory

No Known Activations