INDEX

Explanations

questions or statements expressing curiosity or doubt

oai_token-act-pair · gpt-3.5-turbo

questions expressing curiosity or inquiry about circumstances and outcomes

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

catentry

-0.83

-0.77

mouth

-0.75

alysed

-0.74

arget

-0.69

ongs

-0.66

aration

-0.66

interstitial

-0.66

idation

-0.65

idelines

-0.65

POSITIVE LOGITS

xual

0.85

 suspic

0.80

 misunder

0.79

 nostalg

0.76

why

0.72

 fate

0.71

 retribution

0.70

 motives

0.70

 millenn

0.70

 explan

0.70

Activations Density 0.108%

questions or statements expressing curiosity or doubt

questions expressing curiosity or inquiry about circumstances and outcomes

No Comments

No Known Activations