INDEX

Explanations

locations represented as city and state combinations

oai_token-act-pair · gpt-3.5-turbo

place names and their corresponding locations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zers

-0.80

natureconservancy

-0.72

inka

-0.72

thood

-0.68

versive

-0.66

reon

-0.65

xual

-0.65

ole

-0.65

ullivan

-0.64

ror

-0.64

POSITIVE LOGITS

TN

0.96

NY

0.88

VA

0.86

TX

0.86

England

0.85

 Calif

0.84

OH

0.84

KY

0.82

CA

0.81

WA

0.79

Activations Density 0.098%

locations represented as city and state combinations

place names and their corresponding locations

No Comments

No Known Activations