INDEX

Explanations

components of scientific methodologies or structured arguments

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

look after

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

 Infórmanos

-0.91

:+:

-0.80

 disambiguazione

-0.68

+#+#

-0.68

SharedCtor

-0.66

protoimpl

-0.65

 Administrativna

-0.63

出版年

-0.63

enterOuterAlt

-0.61

TagMode

-0.59

POSITIVE LOGITS

 Contour

0.37

ant

0.36

 contour

0.35

 Atem

0.34

Contour

0.33

ent

0.31

0.30

 プリン

0.29

kem

0.29

Ant

0.29

Activations Density 0.083%

components of scientific methodologies or structured arguments

look after

No Comments

No Known Activations