INDEX

Explanations

names of places or individuals

oai_token-act-pair · gpt-3.5-turbo

the names of individuals or entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

====

-0.80

cx

-0.74

CY

-0.74

ELE

-0.73

 Retro

-0.72

Ao

-0.71

ACS

-0.71

Gy

-0.70

CoC

-0.69

ATES

-0.69

POSITIVE LOGITS

man

1.87

mans

1.62

mann

1.57

MAN

1.44

men

1.31

eman

1.23

linger

1.16

woman

1.13

heimer

1.12

father

1.12

Activations Density 0.085%

names of places or individuals

the names of individuals or entities

No Comments

No Known Activations