INDEX

Explanations

words related to geographical locations within a city

oai_token-act-pair · gpt-3.5-turbo

references to inner and outer cities or spaces

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

essors

-1.03

yip

-0.86

icably

-0.82

offer

-0.79

atoes

-0.76

iculty

-0.74

ares

-0.73

icable

-0.72

ensable

-0.72

nant

-0.72

POSITIVE LOGITS

most

1.26

wear

1.04

 workings

1.02

borough

0.94

 circle

0.91

 Mongolia

0.90

 sanct

0.87

circle

0.87

 thigh

0.82

 circumference

0.81

Activations Density 0.031%

words related to geographical locations within a city

references to inner and outer cities or spaces

No Comments

No Known Activations