INDEX

Explanations

powerful and impactful words, possibly related to leadership and activism

oai_token-act-pair · gpt-3.5-turbo

specific acronyms or abbreviations commonly used

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

 Diaz

-0.68

cean

-0.67

gate

-0.67

 Harbor

-0.65

 Municip

-0.65

ttes

-0.62

kson

-0.61

sup

-0.61

hra

-0.61

ria

-0.60

POSITIVE LOGITS

ING

1.44

NER

1.42

NING

1.34

ER

1.34

ET

1.33

EST

1.33

BOOK

1.32

EL

1.31

STER

1.30

LIN

1.29

Activations Density 0.131%

powerful and impactful words, possibly related to leadership and activism

specific acronyms or abbreviations commonly used

No Comments

No Known Activations