INDEX

Explanations

locations or names of regions

oai_token-act-pair · gpt-3.5-turbo

geographical names and locations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

akable

-0.66

VALUE

-0.64

rious

-0.63

 Trayvon

-0.62

BILITIES

-0.62

VICE

-0.60

ONY

-0.58

ELY

-0.57

Bey

-0.56

Begin

-0.56

POSITIVE LOGITS

shire

1.06

 peninsula

1.04

 Peninsula

1.02

 Lumpur

1.00

fecture

0.98

 suburbs

0.97

 suburb

0.95

 Province

0.94

 region

0.93

 countryside

0.93

Activations Density 0.432%

locations or names of regions

geographical names and locations

No Comments

No Known Activations