INDEX

Explanations

names of individuals or public figures

oai_token-act-pair · gpt-3.5-turbo

references to notable individuals in Korean culture and politics

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

 cort

-0.72

-0.71

 ERROR

-0.69

ãĤ©

-0.68

 HELL

-0.68

 miscarriage

-0.68

 COVER

-0.67

 Alleg

-0.67

 barn

-0.66

 ORDER

-0.65

POSITIVE LOGITS

wei

1.23

jin

1.20

jri

1.08

 Kong

1.08

qi

1.06

jiang

1.05

 Guang

1.05

 Cheng

1.03

Hong

1.02

 Jiang

1.01

Activations Density 0.181%

names of individuals or public figures

references to notable individuals in Korean culture and politics

No Comments

No Known Activations