INDEX

Explanations

phrases indicating possession or association, particularly those related to "the" and "of."

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

date of signing

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

OGND

-0.97

 мәкал

-0.96

 NSCoder

-0.91

sizeCache

-0.89

 propOrder

-0.88

setVerticalGroup

-0.83

ConstraintMaker

-0.81

CloseOperation

-0.80

featureID

-0.80

tagHelperRunner

-0.80

POSITIVE LOGITS

0.41

↵

0.38

↵↵

0.36

0.35

and

0.34

 also

0.32

 other

0.31

of

0.30

Activations Density 0.031%

phrases indicating possession or association, particularly those related to "the" and "of."

date of signing

No Comments

No Known Activations