INDEX

Explanations

phrases related to body image and physical appearance

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

umar

-0.79

lethal

-0.76

ERG

-0.75

ulhu

-0.73

 Flood

-0.71

operator

-0.70

inez

-0.70

cosystem

-0.69

etheless

-0.66

ETHOD

-0.65

POSITIVE LOGITS

 physique

1.25

 nude

1.23

 bikini

1.21

 pageant

1.17

 attire

1.06

 blonde

1.06

 pics

1.05

 hairst

1.05

 glam

1.04

 underwear

1.04

Activations Density 0.323%

phrases related to body image and physical appearance

No Known Activations