INDEX

Explanations

elements related to data structures or formats in programming contexts

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

data, tem, pr, un, because, for, none

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 ब्रेकडाउन

-0.44

 nocturno

-0.39

AutoField

-0.37

 comerciales

-0.36

 juguete

-0.36

waitKey

-0.36

ReusableCell

-0.36

UnitTesting

-0.36

 resourceCulture

-0.34

Ӕ

-0.34

POSITIVE LOGITS

DockStyle

0.60

setViewportView

0.52

venia

0.50

TabbedPane

0.50

 Jawahar

0.47

 Dism

0.45

 Ministerial

0.45

schild

0.45

vst

0.44

CloseOperation

0.44

Activations Density 0.053%

elements related to data structures or formats in programming contexts

data, tem, pr, un, because, for, none

No Comments

No Known Activations