INDEX

Explanations

instances of the word "infinite" or "finite"

oai_token-act-pair · gpt-3.5-turbo

references to the concepts of "infinity" and "finiteness."

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

Not in Any Lists

No Comments

Negative Logits

crow

-0.86

Downloadha

-0.82

neys

-0.75

rix

-0.74

Reviewed

-0.74

bourg

-0.73

pillar

-0.72

board

-0.70

che

-0.70

swick

-0.68

POSITIVE LOGITS

 regress

1.03

urious

0.81

 dimensional

0.78

 Tsukuyomi

0.75

 gratification

0.75

 pard

0.73

 ILCS

0.70

ursday

0.70

inite

0.69

renheit

0.67

Activations Density 0.012%

instances of the word "infinite" or "finite"

references to the concepts of "infinity" and "finiteness."

No Comments

No Known Activations