INDEX

Explanations

terms related to freedom and rights

oai_token-act-pair · gpt-3.5-turbo

references to freedoms related to speech, religion, and expression

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

hao

-0.75

soDeliveryDate

-0.75

ashtra

-0.73

wcs

-0.70

 EntityItem

-0.64

bull

-0.64

recomm

-0.64

edin

-0.63

vale

-0.63

 Solitaire

-0.63

POSITIVE LOGITS

 dignity

0.79

 liberty

0.76

 freedom

0.71

 sorts

0.70

 liberties

0.69

ignty

0.69

ailability

0.68

 course

0.68

 religion

0.67

 conscience

0.67

Activations Density 0.108%

terms related to freedom and rights

references to freedoms related to speech, religion, and expression

No Comments

No Known Activations