INDEX

Explanations

keywords indicating variations or distinctions among subjects or concepts

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

"Different" followed by a noun

np_acts-logits-general · gemini-2.0-flash

different kinds

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

 vectorielle

-0.66

 devriez

-0.62

 correto

-0.61

 économies

-0.59

 genomen

-0.58

voerd

-0.58

 humaine

-0.57

 umana

-0.56

 picioare

-0.56

 umane

-0.54

POSITIVE LOGITS

 kinds

1.00

 than

0.82

(!__

0.82

 GenerationType

0.79

 sorts

0.77

 ways

0.76

 kind

0.75

than

0.75

 coloured

0.74

 sized

0.74

Activations Density 0.121%

keywords indicating variations or distinctions among subjects or concepts

"Different" followed by a noun

different kinds

No Comments

No Known Activations