INDEX

Explanations

disclaimers or warnings in text

oai_token-act-pair · gpt-3.5-turbo

disclaimers and warnings in text

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

etch

-0.78

asin

-0.76

 helicop

-0.73

tun

-0.72

skill

-0.71

expression

-0.70

 greens

-0.70

NetMessage

-0.69

 masse

-0.66

leaf

-0.65

POSITIVE LOGITS

claimer

0.94

Disclaimer

0.91

 disclaimer

0.80

CLAIM

0.78

RANT

0.77

é»Ĵ

0.76

omial

0.74

quished

0.73

orship

0.73

 beware

0.72

Activations Density 0.021%

disclaimers or warnings in text

disclaimers and warnings in text

No Comments

No Known Activations