INDEX

Explanations

words indicating support, motivation, and potential advancements in research or development

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

affirm, support, strengthen

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

########.

-0.66

Попис

-0.56

 continúas

-0.50

PagerAdapter

-0.50

 PeEnEo

-0.50

("")]

-0.49

 esfuer

-0.49

 GetEnumerator

-0.49

 Infór

-0.49

WithIOException

-0.49

POSITIVE LOGITS

redux

0.37

 Drapeau

0.34

oodoo

0.33

 layui

0.33

pair

0.33

AnchorStyles

0.32

 Pingback

0.31

Man

0.31

 estime

0.31

paint

0.31

Activations Density 0.047%

words indicating support, motivation, and potential advancements in research or development

affirm, support, strengthen

No Comments

No Known Activations