INDEX

Explanations

phrases related to knowledge or information

oai_token-act-pair · gpt-3.5-turbo

phrases expressing knowledge or understanding

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

pex

-0.85

ishable

-0.76

itter

-0.75

ovie

-0.74

gencies

-0.67

isco

-0.66

phrine

-0.66

otion

-0.64

atism

-0.64

cers

-0.63

POSITIVE LOGITS

lege

1.22

ledge

1.05

ledged

0.83

 ABOUT

0.81

 firsthand

0.79

nothing

0.73

LED

0.73

urst

0.72

âĦ¢:

0.71

 anecd

0.69

Activations Density 0.051%

phrases related to knowledge or information

phrases expressing knowledge or understanding

No Comments

No Known Activations