INDEX

Explanations

phrases related to specific recognizable names and titles in a structured format

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names or titles

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

 partName

-0.68

ancies

-0.67

wre

-0.65

asking

-0.63

urrencies

-0.62

 fiat

-0.61

Reviewed

-0.61

asks

-0.60

Introduced

-0.59

 dogs

-0.59

POSITIVE LOGITS

ccording

1.06

Lago

0.83

ENA

0.83

nikov

0.74

collar

0.74

reau

0.73

pillar

0.72

zona

0.72

ICA

0.72

bodied

0.71

Activations Density 0.085%

phrases related to specific recognizable names and titles in a structured format

proper nouns, specifically names or titles

No Comments

No Known Activations