INDEX

Explanations

verbs related to actions or qualities that may have negative consequences or connotations

oai_token-act-pair · gpt-3.5-turbo

words related to the concepts of uncertainty and risk

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

Nap

-0.68

Balt

-0.64

nih

-0.63

Ub

-0.61

Unity

-0.59

":"/

-0.58

hess

-0.57

ãģ®å

-0.56

Jan

-0.55

DragonMagazine

-0.55

POSITIVE LOGITS

uate

1.08

enance

1.02

ulate

0.98

 oneself

0.81

ingly

0.76

olate

0.75

balance

0.73

ezvous

0.73

ronics

0.73

 havoc

0.72

Activations Density 0.128%

verbs related to actions or qualities that may have negative consequences or connotations

words related to the concepts of uncertainty and risk

No Comments

No Known Activations