INDEX

Explanations

references to model or data attributes in a programming context

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

elements related to coding or programming concepts

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

single letter followed by punctuation

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 pinulongan

-0.50

 '\\;'

-0.49

AndroidJUnit

-0.48

 brug

-0.48

BeginInit

-0.47

fay

-0.46

 Extra

-0.45

proved

-0.45

bru

-0.45

?-

-0.45

POSITIVE LOGITS

 dignité

0.72

 picioare

0.69

utenant

0.65

 jäsen

0.64

 autorités

0.63

Portail

0.63

seamnă

0.63

 alguno

0.62

 câte

0.62

 rayures

0.61

Activations Density 0.045%

references to model or data attributes in a programming context

elements related to coding or programming concepts

single letter followed by punctuation

No Comments

No Known Activations