INDEX

Explanations

instances of the word "heard"

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the act of hearing or being informed about something

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

quer

-0.70

apo

-0.67

eros

-0.63

independence

-0.63

 Leap

-0.63

apore

-0.61

vp

-0.61

 Shrine

-0.60

ments

-0.60

 Athletics

-0.60

POSITIVE LOGITS

 Heard

0.93

 aloud

0.84

Word

0.83

 voices

0.79

 heard

0.78

 Hear

0.76

çīĪ

0.76

 hearing

0.75

 Hearing

0.73

heard

0.72

Activations Density 0.025%

instances of the word "heard"

phrases indicating the act of hearing or being informed about something

No Comments

No Known Activations