INDEX

Explanations

references to titles within a text

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Text after the word "title"

np_acts-logits-general · gemini-2.0-flash

title

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Not in Any Lists

No Comments

Negative Logits

Gad

-0.55

?>"

-0.53

onshire

-0.51

<u>

-0.47

dır

-0.46

laş

-0.43

see

-0.42

siehe

-0.41

Wil

-0.39

sver

-0.39

POSITIVE LOGITS

 AppComponent

1.11

 propOrder

0.95

titles

0.94

OGND

0.94

wikidata

0.93

 للاسماء

0.93

 autorytatywna

0.92

 titles

0.88

 Titles

0.88

Geplaatst

0.86

Activations Density 0.075%

references to titles within a text

Text after the word "title"

title

No Comments

No Known Activations