INDEX

Explanations

phrases indicating a topic of discussion or communication

oai_token-act-pair · gpt-3.5-turbo

instances of discourse or conversation about specific topics

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

mon

-0.59

geist

-0.59

cop

-0.59

Previous

-0.57

kos

-0.57

 redes

-0.57

fur

-0.57

no

-0.57

can

-0.56

webkit

-0.56

POSITIVE LOGITS

irection

0.77

 Archdemon

0.73

utical

0.70

 hypocr

0.70

urated

0.67

interstitial

0.66

llers

0.65

edge

0.65

ivated

0.65

ļé

0.65

Activations Density 0.168%

phrases indicating a topic of discussion or communication

instances of discourse or conversation about specific topics

No Comments

No Known Activations