INDEX

Explanations

phrases indicating curiosity or questioning of situations

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

few things

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

Географиясе

-0.37

 antidesliz

-0.35

SequentialGroup

-0.34

 besi

-0.34

ẹn

-0.33

 representatives

-0.33

héro

-0.31

 officials

-0.31

tagHelperRunner

-0.31

slidesPer

-0.31

POSITIVE LOGITS

tanleria

0.67

 resourceCulture

0.66

AsUp

0.63

WebElementEntity

0.60

AndEndTag

0.53

 protoimpl

0.53

 leſs

0.51

OCCURRED

0.50

Derp

0.50

SourceChecksum

0.50

Activations Density 0.040%

phrases indicating curiosity or questioning of situations

few things

No Comments

No Known Activations