INDEX

Explanations

patterns or structures in data, particularly in programming or mathematical contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 snippetHide

-0.66

 Италијани

-0.63

GEBURTSDATUM

-0.57

istoitu

-0.55

évaluateur

-0.55

 useAppContext

-0.55

 springfox

-0.54

aarrggbb

-0.54

transQ

-0.54

POSITIVE LOGITS

 ujedno

0.42

whelmed

0.41

 either

0.40

 Bestimmung

0.40

itization

0.37

 secondly

0.37

ubio

0.36

 кстати

0.36

 nahilalakip

0.36

eth

0.35

Activations Density 0.019%

patterns or structures in data, particularly in programming or mathematical contexts

No Comments

No Known Activations