INDEX

Explanations

the name "Dustin"

oai_token-act-pair · gpt-3.5-turbo

mentions of individuals named "Dustin."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mares

-0.76

ilitary

-0.73

esis

-0.72

CHR

-0.70

pering

-0.70

mega

-0.69

mons

-0.68

quin

-0.67

 Burr

-0.66

ãĥ¢

-0.66

POSITIVE LOGITS

 Hoffman

0.75

 Dustin

0.74

orian

0.71

hift

0.71

inia

0.71

dust

0.70

Gw

0.69

 Coyotes

0.68

 Beckham

0.66

æ³

0.63

Activations Density 0.036%

the name "Dustin"

mentions of individuals named "Dustin."

No Comments

No Known Activations