INDEX

Explanations

academic review structure and terminology related to scientific discussions

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

"review"

np_acts-logits-general · gemini-2.0-flash

academic review articles

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

 OMITBAD

-0.60

WithIOException

-0.54

 مشين

-0.52

Билгалдахарш

-0.52

hyrchwyd

-0.50

########.

-0.48

Ծանոթ

-0.48

 للمعارف

-0.47

Personendaten

-0.47

enumii

-0.46

POSITIVE LOGITS

AMENTO

0.46

 asupra

0.45

 pungkasnya

0.44

为止

0.43

 actualizada

0.42

zugehen

0.40

看來

0.40

 gegenwär

0.39

vedať

0.39

 reviewed

0.38

Activations Density 1.237%

academic review structure and terminology related to scientific discussions

"review"

academic review articles

No Comments

No Known Activations