INDEX

Explanations

phrases related to introspection and self-reflection

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

cannot exist

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

 estekak

-0.69

angliski

-0.62

StoryboardSegue

-0.61

UserScript

-0.59

wahati

-0.59

نیم

-0.59

 gyhoeddwyd

-0.58

middels

-0.57

prepareStatement

-0.57



-0.56

POSITIVE LOGITS

 pipeline

0.32

’

0.32

0.31

 пи

0.30

<bos>

0.30

estanden

0.28

 disponibilités

0.28

</em>

0.28

pipeline

0.28

0.27

Activations Density 0.040%

phrases related to introspection and self-reflection

cannot exist

No Comments

No Known Activations