INDEX

Explanations

terms indicating changes in status or frequency related to the present or past

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

no longer true

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 AssemblyCompany

-0.44

OGND

-0.43

 térmico

-0.42

 kaynağından

-0.41

}),

-0.39

 proximity

-0.39

 prohibido

-0.39

Snackbar

-0.39

로운

-0.39

NewReader

-0.38

POSITIVE LOGITS

 vocês

0.56

IntoConstraints

0.55

EDEFAULT

0.55

omitempty

0.51

 viewDidLoad

0.51

toHaveBeenCalled

0.50

orElseThrow

0.47

anyway

0.46

 înc

0.46

abetes

0.45

Activations Density 0.025%

terms indicating changes in status or frequency related to the present or past

no longer true

No Comments

No Known Activations