INDEX

Explanations

phrases where someone is aware or informed about a topic

oai_token-act-pair · gpt-3.5-turbo

instances of the word "aware" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

Textures

-0.68

bard

-0.62

pen

-0.62

mer

-0.61

irrel

-0.60

MER

-0.59

mini

-0.58

wrap

-0.58

cell

-0.57

 Blessing

-0.57

POSITIVE LOGITS

ledged

1.22

ledge

0.87

isance

0.79

icip

0.77

eering

0.75

 Lauder

0.73

alty

0.72

ness

0.71

izon

0.70

alys

0.70

Activations Density 0.025%

phrases where someone is aware or informed about a topic

instances of the word "aware" in various contexts

No Comments

No Known Activations