INDEX

Explanations

phrases related to opposition or resistance against something or someone

oai_token-act-pair · gpt-3.5-turbo

references to anti-related sentiments or movements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Halls

-0.81

ODY

-0.74

 Pelicans

-0.72

 Legends

-0.72

ãĤ¼ãĤ¦ãĤ¹

-0.70

INGS

-0.69

inders

-0.69

ĸļ

-0.68

 Referred

-0.67

shire

-0.67

POSITIVE LOGITS

abortion

0.98

establishment

0.95

war

0.90

anti

0.88

 anti

0.85

overty

0.82

actively

0.79

strate

0.77

pport

0.77

hero

0.77

Activations Density 0.024%

phrases related to opposition or resistance against something or someone

references to anti-related sentiments or movements

No Comments

No Known Activations