INDEX

Explanations

punctuation marks and their associated spacing

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

list formatting and punctuation

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

AndEndTag

-0.91

protoimpl

-0.81

RenderAtEndOf

-0.80

 كومونز

-0.78

 ſind

-0.76

 تضيفلها

-0.73

 ſch

-0.73

 myſelf

-0.73

aarrggbb

-0.72

 Administrativna

-0.72

POSITIVE LOGITS

0.44

0.33

.(

0.25

bag

0.24

évêque

0.24

./

0.23

。

0.23

ian

0.23

..

0.23

Activations Density 0.162%

punctuation marks and their associated spacing

list formatting and punctuation

No Comments

No Known Activations