INDEX

Explanations

locations and organizations like stores, streets, cities, and countries

oai_token-act-pair · gpt-3.5-turbo

specific nouns and proper names, particularly related to places and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Frie

-0.67

 shatter

-0.63

?".

-0.62

 fert

-0.59

 Snapchat

-0.58

 widening

-0.57

 Sting

-0.56

 Mubarak

-0.55

 implants

-0.55

])

-0.53

POSITIVE LOGITS

jamin

1.10

vertising

0.90

odore

0.89

resa

0.89

agascar

0.89

initions

0.89

negie

0.87

assador

0.85

phabet

0.85

ropolitan

0.85

Activations Density 0.315%

locations and organizations like stores, streets, cities, and countries

specific nouns and proper names, particularly related to places and institutions

No Comments

No Known Activations