INDEX

Explanations

phrases indicating a threshold or point of change

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

on the verge of

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

풀

-0.35

 myö

-0.34

 pozosta

-0.34

 particulares

-0.33

 dłu

-0.32

 mourut

-0.32

 längre

-0.31

 principal

-0.31

 řed

-0.31

 métallique

-0.30

POSITIVE LOGITS

 edge

0.89

Personendaten

0.83

edge

0.82

 Edge

0.79

verge

0.78

 EDGE

0.77

 verge

0.76

Edge

0.73

 brink

0.73

AndEndTag

0.73

Activations Density 0.014%

phrases indicating a threshold or point of change

on the verge of

No Comments

No Known Activations

phrases indicating a threshold or point of change

on the verge of

No Comments

No Known Activations