INDEX

Explanations

mentions of the word "twin" and related contexts

oai_token-act-pair · gpt-3.5-turbo

references to "twin" or "twins"

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

UME

-0.81

anwhile

-0.81

CoC

-0.77

andise

-0.73

arcity

-0.68

ugu

-0.67

orsi

-0.66

 Clin

-0.65

 Correspond

-0.64

undy

-0.64

POSITIVE LOGITS

ning

1.01

ned

0.97

 twin

0.87

 brother

0.83

ieth

0.82

 towers

0.82

 brothers

0.78

 twins

0.76

ples

0.74

fuck

0.72

Activations Density 0.010%

mentions of the word "twin" and related contexts

references to "twin" or "twins"

No Comments

No Known Activations