INDEX

Explanations

mentions of the name "Wendy"

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals, particularly those named Wendy or Julie

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

atory

-0.88

unin

-0.78

atar

-0.75

alion

-0.75

Ëľ

-0.74

chall

-0.73

nown

-0.72

tarian

-0.71

istically

-0.71

ogn

-0.70

POSITIVE LOGITS

 Wendy

0.98

Bee

0.80

 Blossom

0.74

 Rice

0.74

Jen

0.71

 Chun

0.70

 Vanessa

0.67

 sauces

0.66

issance

0.65

ank

0.64

Activations Density 0.026%

mentions of the name "Wendy"

references to specific individuals, particularly those named Wendy or Julie

No Comments

No Known Activations