INDEX

Explanations

negations and expressions indicating the absence or exclusion of something

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

"not"

np_acts-logits-general · gemini-2.0-flash

not followed by a verb

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

[:,:,

-0.48

ideas

-0.47

imple

-0.47

 accountId

-0.47

idea

-0.47

++)

-0.46

addElement

-0.45

example

-0.44

 tilf

-0.43

 startX

-0.43

POSITIVE LOGITS

not

0.92

not

0.85

Not

0.82

Not

0.77

NOT

0.69

NOT

0.66

 Nottingham

0.65

Nicht

0.65

Tidak

0.64

nicht

0.63

Activations Density 0.337%

negations and expressions indicating the absence or exclusion of something

"not"

not followed by a verb

No Comments

No Known Activations