INDEX

Explanations

punctuation marks and connectives

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

lists related to official information

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

enderror

-0.43

fromnode

-0.43

تقاوى

-0.42

"@/

-0.41

dua

-0.39

 AssemblyCompany

-0.39

isos

-0.39

 Elle

-0.38

lice

-0.38

arraycopy

-0.38

POSITIVE LOGITS

 Rptr

0.50

numerusform

0.45

IVEREF

0.44

 autorytatywna

0.44

ингл

0.42

Pautan

0.40

tvguidetime

0.40

Jeografia

0.40

 kasarigan

0.38

 coû

0.37

Activations Density 0.083%

punctuation marks and connectives

lists related to official information

No Comments

No Known Activations