INDEX

Explanations

segments that focus on specific experiences or narrative details

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

Question marks, exclamation points, and ellipses

np_acts-logits-general · gemini-2.0-flash

interjections and other punctuation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

transQ

-0.53

AsUp

-0.43

RTLU

-0.42

httphttps

-0.42

 برانيه

-0.41

</thead>

-0.39

yyr

-0.39

NameInMap

-0.39

awtextra

-0.38

先后

-0.38

POSITIVE LOGITS

Wow

1.20

Wow

1.16

wow

1.15

wow

1.07

Sounds

1.05

 Sounds

1.02

Yikes

0.98

sounds

0.98

WOW

0.96

oh

0.93

Activations Density 0.405%

segments that focus on specific experiences or narrative details

Question marks, exclamation points, and ellipses

interjections and other punctuation

No Comments

No Known Activations