INDEX

Explanations

instances of the word "too" and variations of the word "I"

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

too am, too must, too is, myself own, myself believe

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

because

-0.33

 because

-0.32

 διά

-0.31

ทาง

-0.31

 gutem

-0.30

so

-0.29

 postes

-0.29

lò

-0.29

 dió

-0.28

料

-0.28

POSITIVE LOGITS

delwed

0.71

rungsseite

0.71

 TextInputType

0.66

tagHelperRunner

0.66

脚注の使い方

0.66

OGND

0.66

MemoryWarning

0.65

Demografía

0.65

 HFILL

0.64

 nahilalakip

0.64

Activations Density 0.011%

instances of the word "too" and variations of the word "I"

too am, too must, too is, myself own, myself believe

No Comments

No Known Activations