INDEX

Explanations

mentions of anonymity or anonymous sources

oai_token-act-pair · gpt-3.5-turbo

references to anonymity

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

NEY

-0.90

neys

-0.80

bourg

-0.78

Fan

-0.78

eland

-0.77

FK

-0.75

enegger

-0.75

agn

-0.71

yg

-0.71

forces

-0.71

POSITIVE LOGITS

 anonymity

1.00

 anonymous

0.97

 informant

0.93

onym

0.91

 bystand

0.86

 whistleblower

0.85

otine

0.84

onymous

0.82

 anonym

0.80

 anonymously

0.79

Activations Density 0.012%

mentions of anonymity or anonymous sources

references to anonymity

No Known Activations