INDEX

Explanations

proper nouns or names containing the sequence "ig"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "ig."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 listening

-0.67

 Sussex

-0.67

INC

-0.64

 Sparkle

-0.64

 Franch

-0.63

housing

-0.63

 Spurs

-0.60

Ĥª

-0.59

ns

-0.59

ÄŁ

-0.59

POSITIVE LOGITS

arette

1.30

arettes

1.23

ourney

1.22

gers

1.19

rams

1.16

lia

1.14

abyte

1.13

gered

1.13

glers

1.13

raphic

1.12

Activations Density 0.051%

proper nouns or names containing the sequence "ig"

mentions of the word "ig."

No Comments

No Known Activations