INDEX

Explanations

racing-related terms

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of brands, locations, and titles of media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

eric

-0.82

rice

-0.81

tein

-0.80

ships

-0.77

nings

-0.76

nery

-0.76

dylib

-0.74

ning

-0.74

ship

-0.73

imity

-0.72

POSITIVE LOGITS

 unpop

0.96

 metic

0.77

ãĥİ

0.68

 confir

0.68

Ø³

0.66

ĳå£«

0.64

 unden

0.62

 Vaugh

0.62

ãĤ¯

0.60

 cryst

0.60

Activations Density 0.436%

racing-related terms

proper nouns, particularly names of brands, locations, and titles of media

No Comments

No Known Activations