INDEX

Explanations

phrases related to political and social events or organizations

oai_token-act-pair · gpt-3.5-turbo

identifiers and symbols related to political or organizational structures

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

 Breat

-0.78

 Pilgrim

-0.77

 Pagan

-0.77

Mao

-0.75

 Bell

-0.75

 Gong

-0.74

 Viet

-0.74

 Powell

-0.72

BN

-0.72

Bog

-0.72

POSITIVE LOGITS

1.45

1.31

1.23

1.11

1.05

1.00

0.99

0.96

Activations Density 0.133%

phrases related to political and social events or organizations

identifiers and symbols related to political or organizational structures

No Comments

No Known Activations