INDEX

Explanations

prepositional phrases indicating cause or purpose

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

prepositions followed by common words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Personensuche

-0.82

contentLoaded

-0.81

 BoxFit

-0.77

 ExecuteAsync

-0.76

fromnode

-0.76

awtextra

-0.74

GOTREF

-0.73

migrationBuilder

-0.72

<unused28>

-0.71

<unused41>

-0.71

POSITIVE LOGITS

the

0.33

 them

0.32

 this

0.29

 pudieran

0.29

 subido

0.27

 these

0.27

its

0.26

 pierdas

0.26

 nuevamente

0.26

Activations Density 0.082%

prepositional phrases indicating cause or purpose

prepositions followed by common words

No Comments

No Known Activations