INDEX

Explanations

phrases related to Russian President Vladimir Putin

oai_token-act-pair · gpt-3.5-turbo

mentions of the Russian President Vladimir Putin

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

dress

-0.78

roads

-0.78

alach

-0.78

poon

-0.71

cule

-0.68

orph

-0.67

devices

-0.67

cular

-0.67

maid

-0.66

BACK

-0.65

POSITIVE LOGITS

 Putin

1.39

 Lenin

1.05

 Jinping

1.02

 Vlad

1.00

Ily

0.97

ovich

0.97

 Tayyip

0.95

 Vladimir

0.95

 Lavrov

0.93

Nab

0.92

Activations Density 0.021%

phrases related to Russian President Vladimir Putin

mentions of the Russian President Vladimir Putin

No Comments

No Known Activations