INDEX

Explanations

specialized symbols and formatting related to scientific references and citations

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

References formatted like ^[@tag@] or \[[@tag]\]

np_acts-logits-general · gemini-2.0-flash

citation markers

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

↵

-0.38

̃

-0.37

-0.34

-0.32

ãy

-0.32

-0.31

Kilder

-0.31

⁡

-0.31

POSITIVE LOGITS

majánló

0.80

ulongan

0.77

arangay

0.77

Дереккөздер

0.75

 Numerade

0.75

 desmotivaciones

0.73

<unused51>

0.73

<unused14>

0.73

[@BOS@]

0.73

<unused3>

0.73

Activations Density 2.573%

specialized symbols and formatting related to scientific references and citations

References formatted like ^[@tag@] or \[[@tag]\]

citation markers

No Comments

No Known Activations