INDEX

Explanations

headings and structured sections within job descriptions

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

DrawerToggle

-0.65

tagHelperRunner

-0.60

 snippetHide

-0.57

 Numerade

-0.56

 '\\;'

-0.54

Jeografia

-0.54

phosa

-0.53



-0.52

antMatchers

-0.52

 BoxFit

-0.52

POSITIVE LOGITS

XmlAccessType

0.41

éndolo

0.34

Specyfikacja

0.31

FORMANCE

0.31

 échelle

0.30

rikes

0.29

 продукта

0.29

 operación

0.29

hidupan

0.28

espół

0.28

Activations Density 0.001%

headings and structured sections within job descriptions

No Comments

No Known Activations