INDEX

Explanations

instances of the word "Mad" and its variations

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

"Mad" and following words

np_acts-logits-general · gemini-2.0-flash

mad together with other specific words

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

 propOrder

-0.83

IsContent

-0.79

ChildIndex

-0.72

afone

-0.63

fjspx

-0.62

ailles

-0.61

UTERS

-0.61

Expedia

-0.60

chymal

-0.59

WEBPACK

-0.58

POSITIVE LOGITS

Mad

1.01

Mad

0.90

mad

0.83

MAD

0.65

mad

0.64

 Maddox

0.63

MAD

0.60

 Madd

0.53

 rush

0.52

/**

0.52

Activations Density 0.079%

instances of the word "Mad" and its variations

"Mad" and following words

mad together with other specific words

No Comments

No Known Activations