INDEX

Explanations

phrases related to political news or public figures

oai_token-act-pair · gpt-3.5-turbo

phrases that include significant punctuation marks or separators

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

¬¼

-0.71

UF

-0.68

minster

-0.66

ocl

-0.60

acent

-0.59

aha

-0.59

ensation

-0.59

iple

-0.59

onomous

-0.58

acan

-0.58

POSITIVE LOGITS

huh

1.10

etc

1.00

 albeit

0.95

 lest

0.91

eh

0.90

aka

0.88

 namely

0.83

 haha

0.82

but

0.79

 whereas

0.79

Activations Density 0.645%

phrases related to political news or public figures

phrases that include significant punctuation marks or separators

No Comments

No Known Activations