INDEX

Explanations

phrases indicating subjective perception or opinion

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

"like" followed by a short article

np_acts-logits-general · gemini-2.0-flash

like

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

AutoScaleMode

-0.63

 Taktlose

-0.56

 betweenstory

-0.56

oneofs

-0.54

angliski

-0.52

SharedDtor

-0.51

jsonwebtoken

-0.51



-0.50

TemporalType

-0.50

Hentet

-0.50

POSITIVE LOGITS

Laf

0.58

WhiteSpace

0.55

the

0.53

Slug

0.53

0.52

 MAJOR

0.52

Zil

0.51

laven

0.51

slug

0.50

 Aleks

0.50

Activations Density 0.042%

phrases indicating subjective perception or opinion

"like" followed by a short article

like

No Comments

No Known Activations