INDEX

Explanations

references to specific programming language constructs, particularly in relation to mathematical computations

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

Code, URLs, or uncommon words

np_acts-logits-general · gemini-2.0-flash

cockiness

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

اكم

-0.56

ضير

-0.52

errHandler

-0.51

URLException

-0.51

Lorem

-0.51

:✨

-0.50

 виправивши

-0.50

 GENERATED

-0.49

ITIES

-0.48

esModule

-0.48

POSITIVE LOGITS

 ISNI

0.66

Otherwise

0.64

 africain

0.62

 Otherwise

0.59

 otherwise

0.59

 Priv

0.55

ⓘ

0.53

 Jérusalem

0.53

 virtuel

0.53

>--}}

0.52

Activations Density 0.086%

references to specific programming language constructs, particularly in relation to mathematical computations

Code, URLs, or uncommon words

cockiness

No Comments

No Known Activations