INDEX

Explanations

phrases containing the word "paper"

oai_token-act-pair · gpt-3.5-turbo

terms related to academic publications and research papers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 knots

-0.70

 Churchill

-0.69

ivity

-0.69

 historic

-0.66

 Locke

-0.66

Rio

-0.63

 Tory

-0.63

Geo

-0.62

 UKIP

-0.62

ties

-0.59

POSITIVE LOGITS

aper

1.52

apers

1.45

ODY

0.91

oided

0.90

iatrics

0.88

ILY

0.86

olicy

0.82

 resil

0.81

ãĤ¼ãĤ¦ãĤ¹

0.76

utic

0.76

Activations Density 0.007%

phrases containing the word "paper"

terms related to academic publications and research papers

No Comments

No Known Activations