INDEX

Explanations

names of individuals in positions of power or prominence

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of people or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

âĶĢâĶĢ

-0.86

LEASE

-0.79

âĸ¬

-0.76

sburgh

-0.67

EEE

-0.64

VICE

-0.62

acebook

-0.62

LCS

-0.62

Â·Â·

-0.61

FontSize

-0.60

POSITIVE LOGITS

iani

0.99

zynski

0.94

hair

0.94

iman

0.92

hai

0.91

endi

0.88

hari

0.88

hart

0.87

oub

0.86

kil

0.85

Activations Density 0.253%

names of individuals in positions of power or prominence

proper nouns, specifically names of people or entities

No Comments

No Known Activations