INDEX

Explanations

terms related to vision and eye contact

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

mente

-0.55

ⓧ

-0.54

はじめに

-0.53

-0.51

 suara

-0.51

łaszcza

-0.51

ly

-0.50

 razie

-0.50

ization

-0.47

 Viitattu

-0.47

POSITIVE LOGITS

eye

0.75

EYE

0.75

 sockets

0.70

EYE

0.70

👁

0.69

IntoConstraints

0.68

lids

0.67

 brows

0.67

 catar

0.66

Eye

0.66

Activations Density 0.053%

terms related to vision and eye contact

No Comments

No Known Activations