INDEX

Explanations

locations or cities

oai_token-act-pair · gpt-3.5-turbo

geographic locations or state names

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

BOOK

-0.78

schild

-0.68

ãĥĬ

-0.68

 Crimean

-0.65

pod

-0.65

 Gravity

-0.64

 materials

-0.63

lords

-0.62

length

-0.61

 lengths

-0.61

POSITIVE LOGITS

herty

1.02

.,

0.97

adelphia

0.91

adena

0.83

achu

0.83

VIDIA

0.83

acan

0.81

olla

0.80

osi

0.80

heed

0.80

Activations Density 0.023%

locations or cities

geographic locations or state names

No Comments

No Known Activations