INDEX

Explanations

negation phrases and structures

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

not followed by words

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

ValueStyle

-0.56

 виправивши

-0.54

PreferredItem

-0.50

+#+#

-0.50

#+#

-0.45

localctx

-0.43

 ویکی‌پدی

-0.43

sizeCache

-0.42

sidemargin

-0.42

ImageContext

-0.41

POSITIVE LOGITS

not

0.58

Not

0.58

Not

0.57

not

0.54

NOT

0.51

NOT

0.47

 Noth

0.45

NotIn

0.44

non

0.44

merkung

0.44

Activations Density 0.056%

negation phrases and structures

not followed by words

No Comments

No Known Activations

negation phrases and structures

not followed by words

No Comments

No Known Activations