INDEX

Explanations

structures related to code or programming syntax

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

Code or configuration snippets

np_acts-logits-general · gemini-2.0-flash

GPIO.setmode

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

 '\\;'

-0.49

 disambiguazione

-0.46

íncia

-0.42

 estr

-0.41

xygen

-0.41

ústria

-0.40

kyl

-0.40

une

-0.39

lenburg

-0.39

POSITIVE LOGITS

oneofs

0.80

MLLoader

0.76

featureID

0.75

OGND

0.71

 bezeichneter

0.66

 Normdatei

0.65

#+#

0.64

Personensuche

0.63

पया

0.63

enderror

0.62

Activations Density 0.277%

structures related to code or programming syntax

Code or configuration snippets

GPIO.setmode

No Comments

No Known Activations