INDEX

Explanations

negations or phrases indicating absence or disagreement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InjectAttribute

-0.69

 autorytatywna

-0.62

FieldBuilder

-0.60

存于互联网档案馆

-0.60

Hentet

-0.59

 transfieras

-0.58

 ſind

-0.57

ɵɵ

-0.57

IsMutable

-0.56

 ―――――

-0.56

POSITIVE LOGITS

 aikaa

0.40

 kysy

0.38

 väli

0.36

 trekken

0.36

dengar

0.35

 käyttö

0.35

 kaupung

0.34

 duquel

0.34

 fisuras

0.33

 meille

0.33

Activations Density 0.668%

negations or phrases indicating absence or disagreement

No Comments

No Known Activations