INDEX

Explanations

punctuation marks and dashes

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

dash and name attribution

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

zzleHttp

-0.65

IUrlHelper

-0.65

 ویکی‌پدی

-0.62

UnusedPrivate

-0.62

Personensuche

-0.60

DockStyle

-0.59

InjectAttribute

-0.59

MigrationBuilder

-0.59

 CreateTagHelper

-0.55

 Möglich

-0.54

POSITIVE LOGITS

the

0.38

 Charles

0.35

opsida

0.34

©

0.32

 Rich

0.31

Gal

0.30

 Архів

0.29

@[+][

0.29

 xmlns

0.29

 defeating

0.29

Activations Density 0.015%

punctuation marks and dashes

dash and name attribution

No Comments

No Known Activations