INDEX

Explanations

words related to introspection and self-reflection

oai_token-act-pair · gpt-3.5-turbo

expressions related to self-reflection and awareness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Sprint

-0.70

é¾

-0.69

rollers

-0.68

 Racer

-0.65

 Bullets

-0.65

berries

-0.64

 Cricket

-0.62

Getty

-0.62

DragonMagazine

-0.61

ðŁĺ

-0.60

POSITIVE LOGITS

 insofar

1.03

rather

0.89

 mathemat

0.88

 thereby

0.86

 particularly

0.86

 fundamentally

0.84

 consequently

0.82

 moreover

0.82

 recognizing

0.82

 rather

0.81

Activations Density 0.684%

words related to introspection and self-reflection

expressions related to self-reflection and awareness

No Comments

No Known Activations