INDEX

Explanations

queries and requests for information

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

as status, presumably giving

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



-0.45

Werdegang

-0.42

nodoc

-0.38

 pengend

-0.37

Pol

-0.33

 Sorg

-0.32

 Sando

-0.30

gaon

-0.30

Nonnull

-0.29

⎡

-0.29

POSITIVE LOGITS

 Infórmanos

0.68

 EconPapers

0.62

rungsseite

0.62

Hentet

0.59

 kaarangay

0.58

 дописавши

0.58

 հղումներ

0.56

 autorytatywna

0.55

 kasarigan

0.54

 незавершена

0.54

Activations Density 0.085%

queries and requests for information

as status, presumably giving

No Comments

No Known Activations