INDEX

Explanations

proper nouns associated with businesses or people

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to names of people, places, and titles in media

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

priv

-0.79

orescence

-0.78

estate

-0.75

rum

-0.75

UTC

-0.72

bon

-0.72

bsite

-0.71

wind

-0.71

aff

-0.70

housing

-0.68

POSITIVE LOGITS

 Productions

1.06

 Sunshine

0.93

 Ninja

0.93

 Blues

0.93

 Theory

0.92

 Cafe

0.91

 Deadly

0.90

 Brewing

0.90

 Tavern

0.90

 Brewery

0.88

Activations Density 0.187%

proper nouns associated with businesses or people

proper nouns related to names of people, places, and titles in media

No Comments

No Known Activations