INDEX

Explanations

phrases containing the string "SA"

oai_token-act-pair · gpt-3.5-turbo

references to specific organization names or acronyms

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

ians

-0.74

ness

-0.73

naire

-0.71

tie

-0.71

iques

-0.71

lings

-0.70

ician

-0.70

naires

-0.69

icians

-0.69

shire

-0.69

POSITIVE LOGITS

VE

1.57

BILITY

1.37

UGH

1.28

BLE

1.27

ZE

1.24

GE

1.23

KE

1.21

ULT

1.20

BILITIES

1.17

ILY

1.16

Activations Density 0.067%

phrases containing the string "SA"

references to specific organization names or acronyms

No Comments

No Known Activations