INDEX

Explanations

specific types of identifiers or names within a document

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

g with numbers or days

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

-------

-0.77

Personendaten

-0.59

GeoNames

-0.59

SharedCtor

-0.58

脚注の使い方

-0.57

URLException

-0.57

المناصب

-0.57

ьаж

-0.57

 Universitario

-0.57

acamata

-0.56

POSITIVE LOGITS

작

0.45

Bes

0.37

작

0.36

addPreferredGap

0.36

 Про

0.35

semp

0.34

 ARCH

0.33

അ

0.33

 אח

0.33

spr

0.33

Activations Density 0.015%

specific types of identifiers or names within a document

g with numbers or days

No Comments

No Known Activations