INDEX

Explanations

words related to conspiracy theories

oai_token-act-pair · gpt-3.5-turbo

negative contractions indicating negation or denial

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

 looms

-0.74

 liaison

-0.65

 initiative

-0.63

 circulation

-0.63

 successes

-0.62

 decimal

-0.62

 imitation

-0.61

 promoters

-0.61

 accum

-0.61

 Mirage

-0.61

POSITIVE LOGITS

ï¸ı

1.10

Balt

0.89

ï¸

0.88

İ

0.86

âĢł

0.83

sure

0.81

Ã§

0.81

sorry

0.81

culus

0.78

yet

0.78

Activations Density 0.163%

words related to conspiracy theories

negative contractions indicating negation or denial

No Comments

No Known Activations