INDEX

Explanations

phrases that involve permissions, responsibilities, and requirements in various contexts

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

Precedes a verb in infinitive form

np_acts-logits-general · gemini-2.0-flash

external investigators

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 weihnachts

-0.28

 ghế

-0.26

 dosis

-0.25

decorated

-0.24

 Pág

-0.24

 ratio

-0.24

 absence

-0.24

cou

-0.23

 knop

-0.22

łaś

-0.22

POSITIVE LOGITS

matchCondition

0.77

 חיצוניים

0.71

 Normdatei

0.66

 autorytatywna

0.65

oredCriteria

0.64

OGND

0.63

SBATCH

0.63

BagLayout

0.63

 ProtoMessage

0.61

 onAnimation

0.61

Activations Density 0.637%

phrases that involve permissions, responsibilities, and requirements in various contexts

Precedes a verb in infinitive form

external investigators

No Comments

No Known Activations