INDEX

Explanations

code-related terms and parameters

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

Code, comments, or descriptions

np_acts-logits-general · gemini-2.0-flash

numbers and their description

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<eos>

-0.80

AndEndTag

-0.65

errHandler

-0.56

ﬅ

-0.52

 dAtA

-0.52

 Administrativna

-0.51

UnusedPrivate

-0.50

↵↵↵

-0.48

 […]

-0.48

KURZBESCHREIBUNG

-0.48

POSITIVE LOGITS

didSet

0.60

 متعلقه

0.59

слет

0.54

findpost

0.52

///</

0.52

urlpatterns

0.52

 chi̍t

0.49

orologio

0.48

 Polsek

0.48

loroethane

0.47

Activations Density 0.064%

code-related terms and parameters

Code, comments, or descriptions

numbers and their description

No Comments

No Known Activations