INDEX

Explanations

names of political figures, particularly focusing on the last name "Gingrich."

oai_token-act-pair · gpt-3.5-turbo

mentions of Newt Gingrich

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

ournal

-0.84

pee

-0.73

¥µ

-0.71

Constructed

-0.70

asca

-0.70

ĸļ

-0.67

raints

-0.67

ordering

-0.65

ĻĤ

-0.64

ivated

-0.64

POSITIVE LOGITS

 Gingrich

1.30

etz

0.80

enstein

0.76

MEP

0.72

ulation

0.72

ulations

0.72

mann

0.71

 Romney

0.69

 Cain

0.69

auld

0.68

Activations Density 0.014%

names of political figures, particularly focusing on the last name "Gingrich."

mentions of Newt Gingrich

No Comments

No Known Activations