INDEX

Explanations

names of individuals or entities

oai_token-act-pair · gpt-3.5-turbo

references to individuals and their professional roles or connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

enhagen

-0.79

 sidx

-0.70

Jr

-0.69

 dick

-0.68

 prostate

-0.66

 erection

-0.64

undo

-0.64

jandro

-0.63

don

-0.62

cko

-0.62

POSITIVE LOGITS

 herself

1.92

 Louise

1.15

 heroine

1.13

 Devi

1.12

 hijab

1.12

she

1.11

She

1.09

 husband

1.08

her

1.07

 Marie

1.07

Activations Density 1.801%

names of individuals or entities

references to individuals and their professional roles or connections

No Comments

No Known Activations