INDEX

Explanations

references to public figures, legal terms, and community interactions

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

articles preceding adjectives or descriptive phrases

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

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

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_post

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

ravings

-0.83

abytes

-0.81

apses

-0.80

views

-0.78

ivers

-0.77

 Rooms

-0.75

WARE

-0.75

sections

-0.74

 warehouses

-0.74

blogs

-0.74

POSITIVE LOGITS

wig

1.31

 mask

1.23

 robe

1.21

 jacket

1.15

 cape

1.14

 sweater

1.12

hat

1.11

 costume

1.11

 scarf

1.11

 hijab

1.09

Activations Density 0.116%

references to public figures, legal terms, and community interactions

articles preceding adjectives or descriptive phrases

No Known Activations