INDEX

Explanations

words and phrases that indicate specific entities or names within a context

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

release, markup, inbox, viruses

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

bewerken

-0.75

expandindo

-0.65

 autorytatywna

-0.63

المناصب

-0.61

 linkovi

-0.60

ConstraintMaker

-0.58

цезда

-0.57

dafx

-0.56

transQ

-0.56

 Italijanski

-0.56

POSITIVE LOGITS

Clik

0.50

msgTypes

0.44

 Pete

0.39

Mid

0.39

 Zack

0.38

WriteLiteral

0.38

Pete

0.37

 behind

0.36

Tim

0.35

Tim

0.35

Activations Density 0.010%

words and phrases that indicate specific entities or names within a context

release, markup, inbox, viruses

No Comments

No Known Activations