INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

references to specific people, events, and items related to health and societal issues

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

=-=-

-0.67

merce

-0.61

bably

-0.58

reditary

-0.56

rosso

-0.55

ccording

-0.55

uilt

-0.54

iversal

-0.54

 seiz

-0.54

 conclud

-0.53

POSITIVE LOGITS

 Mercury

0.55

Sov

0.53

SOURCE

0.51

 Sims

0.50

arium

0.49

 Alger

0.48

 Harlem

0.48

Sap

0.48

 Chron

0.47

Crystal

0.47

Activations Density 1.654%

proper nouns

references to specific people, events, and items related to health and societal issues

No Comments

No Known Activations