INDEX

Explanations

numeric values and associated formatting elements

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

CSS properties

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

屬

-0.41

SuspendLayout

-0.39

WritableDatabase

-0.38

ronpa

-0.38

RegressionTest

-0.38

gonic

-0.36

DatabaseError

-0.36

CodeAttribute

-0.36

ModelAttribute

-0.36

getClassLoader

-0.36

POSITIVE LOGITS

tvguidetime

0.54

ChildScrollView

0.46

oredCriteria

0.45

sizeCache

0.45

imanapun

0.44

 COMPUT

0.44

 चीज़ों

0.44

sidemargin

0.43

unicip

0.43

 hâte

0.42

Activations Density 0.012%

numeric values and associated formatting elements

CSS properties

No Comments

No Known Activations