INDEX

Explanations

phrases regarding welfare and support systems

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

preceding "everyone" or similar words

np_acts-logits-general · gemini-2.0-flash

for everyone

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

 NUKAT

-0.50

(',')

-0.47

 gebe

-0.47

 câu

-0.45

ça

-0.43

 поводу

-0.43

伎

-0.43

 isso

-0.42

sự

-0.42

 sự

-0.42

POSITIVE LOGITS

 everyone

1.24

 everybody

1.13

everyone

1.13

 oneself

0.97

 both

0.96

 EVERYONE

0.93

everybody

0.93

Everyone

0.92

 themſelves

0.85

 themselves

0.84

Activations Density 0.443%

phrases regarding welfare and support systems

preceding "everyone" or similar words

for everyone

No Comments

No Known Activations