INDEX

Explanations

mentions of twins

oai_token-act-pair · gpt-3.5-turbo

references to twins

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Liberties

-0.67

trak

-0.66

WI

-0.64

vious

-0.64

supp

-0.63

nce

-0.63

ld

-0.61

 Popular

-0.60

isan

-0.59

Nou

-0.59

POSITIVE LOGITS

 twins

1.38

poons

0.87

 orphans

0.81

omnia

0.79

omething

0.79

idious

0.74

hips

0.73

folk

0.72

roo

0.72

peak

0.71

Activations Density 0.005%

mentions of twins

references to twins

No Comments

No Known Activations