INDEX

Explanations

parts of code referencing Java classes or structures, focusing on utilities and libraries

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

Code syntax and special characters

np_acts-logits-general · gemini-2.0-flash

struct user

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 propOrder

-0.73

GOTREF

-0.73

SharedDtor

-0.66

WithIOException

-0.65

CppMethod

-0.64

IBOutlet

-0.62

onViewCreated

-0.60

Brainz

-0.60

featureID

-0.60

tonsoft

-0.59

POSITIVE LOGITS

 tidligere

0.32

ươi

0.31

<em>

0.31

izarra

0.31

erdere

0.31

 jangka

0.30

bū

0.30

\]

0.30

']='

0.30

}\

0.29

Activations Density 0.032%

parts of code referencing Java classes or structures, focusing on utilities and libraries

Code syntax and special characters

struct user

No Comments

No Known Activations