INDEX

Explanations

phrases related to controversial or polarizing topics

oai_token-act-pair · gpt-3.5-turbo

topics related to social issues and controversial historical events

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

PokÃ©

-0.58

itar

-0.56

 explor

-0.56

 fray

-0.56

 scram

-0.56

 zoom

-0.54

map

-0.53

 ambient

-0.52

 grunt

-0.52

 siege

-0.52

POSITIVE LOGITS

instead

0.84

 THEN

0.73

 thereby

0.70

terday

0.67

hovah

0.65

Instead

0.63

rather

0.63

 Instead

0.62

ardless

0.61

 Therefore

0.59

Activations Density 1.675%

phrases related to controversial or polarizing topics

topics related to social issues and controversial historical events

No Comments

No Known Activations