INDEX

Explanations

instances of the term "extra"

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

extra

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 představ

-0.44

 przem

-0.43

it

-0.42

couvrez

-0.40

 inilah

-0.40

 geest

-0.39

 nærm

-0.38

 tegen

-0.38

 this

-0.37

 culmination

-0.37

POSITIVE LOGITS

 extra

0.98

 EXTRA

0.92

 zusätzlichen

0.84

 additional

0.80

 ekstra

0.80

extra

0.79

 Extra

0.77

additional

0.75

 zusätz

0.75

 added

0.74

Activations Density 0.015%

instances of the term "extra"

extra

No Comments

No Known Activations