INDEX

Explanations

phrases or words related to jewelry or circular objects

oai_token-act-pair · gpt-3.5-turbo

references to the concept of a "ring."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

éĹĺ

-0.77

ufact

-0.75

tical

-0.74

essors

-0.69

irrel

-0.68

ulhu

-0.68

 Fiorina

-0.67

icago

-0.65

 autop

-0.65

é¾įå¥ĳå£«

-0.65

POSITIVE LOGITS

tones

1.29

tone

1.26

leader

1.16

leaders

1.11

rings

1.03

git

0.98

worm

0.96

mail

0.95

wra

0.91

 rings

0.87

Activations Density 0.011%

phrases or words related to jewelry or circular objects

references to the concept of a "ring."

No Comments

No Known Activations