INDEX

Explanations

references to individuals and their interactions or behavior in a context related to relationships or testimonies

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

past states and actions

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

UnusedPrivate

-0.59

GOTREF

-0.49

termilk

-0.46

 indisponible

-0.45

webElementXpaths

-0.45

menistan

-0.45

BibitemShut

-0.45



-0.45

dafx

-0.45

EndInit

-0.45

POSITIVE LOGITS

!(:

0.51

checs

0.37

 Она

0.36

 Sozial

0.35

 Infór

0.34

commands

0.33

codon

0.33

 StringIO

0.32

setLong

0.32

 Describes

0.32

Activations Density 0.016%

references to individuals and their interactions or behavior in a context related to relationships or testimonies

past states and actions

No Comments

No Known Activations