INDEX

Explanations

the word "hollow" and related words

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the term "hollow."

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

ittee

-0.82

~~~~~~~~~~~~~~~~

-0.76

Downloadha

-0.73

abetes

-0.71

weeney

-0.69

 McKenna

-0.68

 Polo

-0.67

ione

-0.66

MPH

-0.63

uana

-0.63

POSITIVE LOGITS

 hollow

0.99

shell

0.89

edIn

0.85

sonian

0.78

sided

0.77

ãĤ©

0.77

 dwellings

0.76

 shells

0.76

 cavity

0.75

Activations Density 0.014%

the word "hollow" and related words

repeated mentions of the term "hollow."

No Comments

No Known Activations