INDEX

Explanations

items and roles related to personal or professional identity and characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 disambiguazione

-0.60

ælp

-0.59

eds

-0.58

 prouver

-0.55

 craindre

-0.55

IVEREF

-0.53

 défendre

-0.52

RegisterType

-0.52

 fotografico

-0.51

Mothers

-0.50

POSITIVE LOGITS

 person

1.65

 researcher

1.31

 musician

1.31

 member

1.29

guy

1.29

 participant

1.25

 athlete

1.25

 performer

1.24

 scientist

1.23

 player

1.21

Activations Density 2.237%

items and roles related to personal or professional identity and characteristics

No Comments

No Known Activations