INDEX

Explanations

mentions of political figures

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and their roles or significance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Denver

-0.76

emort

-0.76

Newsletter

-0.75

FRI

-0.72

Philadelphia

-0.72

 Luthor

-0.71

Weiss

-0.70

furt

-0.69

Cub

-0.69

 âĸĪ

-0.68

POSITIVE LOGITS

 Sharma

1.09

bh

1.07

 Singh

1.03

jriwal

1.00

 Sabha

0.99

 Bhar

0.97

Raj

0.94

Yad

0.94

umbai

0.93

 Shiv

0.93

Activations Density 0.313%

mentions of political figures

references to specific individuals and their roles or significance

No Comments

No Known Activations