INDEX

Explanations

references to television news and broadcasting structures

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

 يتيمه

-0.72

 snippetHide

-0.60

 EconPapers

-0.57

tagHelperRunner

-0.57

 virgen

-0.56

 AssemblyCulture

-0.55

 referrerpolicy

-0.55

éez

-0.55

ValueStyle

-0.54

 nonUne

-0.53

POSITIVE LOGITS

DockStyle

0.42

'];?>

0.39

 mores

0.36

 typically

0.36

ECA

0.36

沙

0.35

ECA

0.35

 هول

0.35

 legal

0.34

apa

0.34

Activations Density 0.101%

references to television news and broadcasting structures

No Comments

No Known Activations