INDEX

Explanations

words related to physical appearances and descriptions, often with a negative connotation

oai_token-act-pair · gpt-3.5-turbo

informal expressions and colloquial language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

IUM

-0.89

DragonMagazine

-0.85

CVE

-0.80

ãĤ¡

-0.78

ãĥ´ãĤ¡

-0.75

 Expend

-0.74

PsyNetMessage

-0.73

APD

-0.71

ãĥ©ãĥ³

-0.71

INAL

-0.70

POSITIVE LOGITS

ween

1.00

eyed

0.93

bye

0.89

bour

0.87

pants

0.82

nee

0.81

wild

0.77

talk

0.77

ho

0.75

roo

0.75

Activations Density 0.145%

words related to physical appearances and descriptions, often with a negative connotation

informal expressions and colloquial language

No Comments

No Known Activations