INDEX

Explanations

punctuation and formatting characters

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

symbols, punctuation, and code fragments

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

Identity

-0.45

 Infir

-0.43

 <<<<<<<<<<<<<<

-0.43

Living

-0.42

Afficher

-0.41

 tunt

-0.39

entration

-0.39

sû

-0.39

ienti

-0.39

déric

-0.38

POSITIVE LOGITS

principalColumn

0.67



0.59

الحياه

0.56

RUnlock

0.56

 ujednoznacz

0.56

 CanadaChoose

0.50

+#+

0.48

classnames

0.48

ConstraintMaker

0.47

 فريبيس

0.46

Activations Density 0.052%

punctuation and formatting characters

symbols, punctuation, and code fragments

No Comments

No Known Activations