INDEX

Explanations

terms related to scientific characteristics and specifications

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

each/respective in different languages

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

 незавершена

-0.53

ChildScrollView

-0.50

devise

-0.46

endpush

-0.45

 enfans

-0.40

inher

-0.40

tvguidetime

-0.40

 <<<<<<<<<<<<<<

-0.40

 météoro

-0.40

غيل

-0.40

POSITIVE LOGITS

 each

0.58

 jeweiligen

0.57

 poszczegól

0.52

 Each

0.52

➯

0.49

Each

0.47

 từng

0.46

enumi

0.43

各

0.42

それぞれの

0.41

Activations Density 0.228%

terms related to scientific characteristics and specifications

each/respective in different languages

No Comments

No Known Activations