INDEX

Explanations

references to a specific person named "Abe"

oai_token-act-pair · gpt-3.5-turbo

references to prominent individuals, specifically political figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

porting

-0.82

ippi

-0.72

pher

-0.70

Leary

-0.69

onyms

-0.68

arters

-0.67

ODUCT

-0.66

videos

-0.62

ECTION

-0.61

 Beir

-0.61

POSITIVE LOGITS

 Shinzo

1.21

Abe

1.09

cean

0.97

achev

0.93

eln

0.92

ijing

0.83

 nodd

0.77

 plur

0.76

zu

0.76

wake

0.75

Activations Density 0.004%

references to a specific person named "Abe"

references to prominent individuals, specifically political figures

No Comments

No Known Activations