INDEX

Explanations

quoted strings and characters that denote text or URLs

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

string literals in code

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

ConstraintMaker

-0.70

 lenker

-0.69

⟬

-0.69

 estekak

-0.68

+#+#

-0.65

 Administrativna

-0.65

ьаж

-0.64

 kaarangay

-0.64

 Infórmanos

-0.64

 فريبيس

-0.63

POSITIVE LOGITS

0.46

='

0.42

“

0.42

="

0.41

0.40

="

0.40

 gero

0.38

:@"

0.38

=("

0.38

setVerticalGroup

0.37

Activations Density 0.022%

quoted strings and characters that denote text or URLs

string literals in code

No Comments

No Known Activations