INDEX

Explanations

sequences of numbers and punctuation that suggest structured data or coding syntax

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

numerical and code structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

EndContext

-0.52

featureID

-0.48

 المعيارى

-0.46

 tengah

-0.42

 ecológico

-0.42

getLogger

-0.42

RegressionTest

-0.41

pahan

-0.41

:✨

-0.40

setCellValue

-0.40

POSITIVE LOGITS

 informée

0.44

 RADIUS

0.42

inSlope

0.41

UserDao

0.39

 CreateTagHelper

0.38

cic

0.38

 UserDao

0.37

 userDao

0.37

Zep

0.37

 gyhoeddwyd

0.36

Activations Density 0.003%

sequences of numbers and punctuation that suggest structured data or coding syntax

numerical and code structures

No Comments

No Known Activations