INDEX

Explanations

the name "Karen" mentioned in a document

oai_token-act-pair · gpt-3.5-turbo

occurrences of the name "Karen."

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

xual

-0.76

utory

-0.72

dated

-0.70

etheus

-0.68

cipline

-0.68

cision

-0.65

udic

-0.65

 derogatory

-0.64

Quotes

-0.63

 Accuracy

-0.62

POSITIVE LOGITS

Bei

0.88

ites

0.85

aii

0.84

iak

0.79

ite

0.77

kie

0.77

fman

0.77

NING

0.76

lock

0.75

asaki

0.73

Activations Density 0.036%

the name "Karen" mentioned in a document

occurrences of the name "Karen."

No Comments

No Known Activations