INDEX

Explanations

mentions of men, possibly associated with certain activities or attributes

oai_token-act-pair · gpt-3.5-turbo

the word "men" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Pwr

-0.67

 Border

-0.66

Recomm

-0.64

 Tibetan

-0.64

Accessory

-0.64

çĭ

-0.62

 mileage

-0.62

CBD

-0.61

ãĥł

-0.61

 flats

-0.61

POSITIVE LOGITS

uscript

1.17

men

1.06

emen

1.05

opausal

1.04

folk

0.95

endez

0.93

ensical

0.92

thren

0.92

icide

0.90

stru

0.90

Activations Density 0.008%

mentions of men, possibly associated with certain activities or attributes

the word "men" in various contexts

No Comments

No Known Activations