INDEX

Explanations

numerical height and weight descriptions of individuals

oai_token-act-pair · gpt-3.5-turbo

height and weight measurements of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ong

-0.67

leneck

-0.62

 distracted

-0.59

 distraction

-0.56

ertation

-0.55

lier

-0.55

 notwithstanding

-0.54

ufact

-0.54

 conflic

-0.53

ILY

-0.53

POSITIVE LOGITS

1.32

foot

1.28

feet

1.27

 feet

1.27

'-

1.26

âĢ²

1.25

ft

1.15

 foot

1.13

')

1.08

'/

1.06

Activations Density 0.071%

numerical height and weight descriptions of individuals

height and weight measurements of individuals

No Comments

No Known Activations