INDEX

Explanations

locations or cities

oai_token-act-pair · gpt-3.5-turbo

mentions of specific cities or 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

Mach

-0.76

 mathemat

-0.72

VALUE

-0.65

yip

-0.61

lihood

-0.61

wagen

-0.61

selection

-0.61

 sacrific

-0.60

arist

-0.59

 myster

-0.59

POSITIVE LOGITS

 County

1.11

 City

0.99

 Mayor

0.93

 Heights

0.90

 skyline

0.90

 Avenue

0.90

 metro

0.89

 Harbor

0.88

 Township

0.88

 Chinatown

0.87

Activations Density 0.209%

locations or cities

mentions of specific cities or locations

No Comments

No Known Activations