INDEX

Explanations

mathematical or technical definitions related to formal systems and integers

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

"be" followed by "positive" and a description

np_acts-logits-general · gemini-2.0-flash

be positive or as

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

 trùng

-0.52

MessageTagHelper

-0.51

 للاسماء

-0.48

PostInfinity

-0.45

+#+

-0.44

 الرياضيه

-0.44

bigoplus

-0.44

Mund

-0.43

Pictured

-0.43

ět

-0.42

POSITIVE LOGITS

ReusableCell

0.47

 positive

0.45

 quelcon

0.45

GIVEREF

0.44

 arbitrary

0.44

LLocation

0.42

MemoryWarning

0.40

 gynhyrchwyd

0.40

any

0.40

 CreateTagHelper

0.39

Activations Density 2.286%

mathematical or technical definitions related to formal systems and integers

"be" followed by "positive" and a description

be positive or as

No Comments

No Known Activations