INDEX

Explanations

words related to negative emotions or situations

oai_token-act-pair · gpt-3.5-turbo

negative prefixes attached to words conveying discontent or difficulties

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

anwhile

-0.79

 Trilogy

-0.77

 Nanto

-0.77

 Defenders

-0.75

BOOK

-0.74

Kings

-0.73

enegger

-0.72

sonian

-0.72

cair

-0.70

fman

-0.66

POSITIVE LOGITS

unh

0.87

ashed

0.87

ishable

0.85

oly

0.83

rep

0.80

idden

0.79

wash

0.78

 scrut

0.78

 undone

0.75

unp

0.74

Activations Density 0.005%

words related to negative emotions or situations

negative prefixes attached to words conveying discontent or difficulties

No Comments

No Known Activations