INDEX

Explanations

locations, especially stores, supermarkets, malls, and apartment complexes

oai_token-act-pair · gpt-3.5-turbo

locations related to various types of businesses and establishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãģį

-0.83

emonic

-0.79

 Helpful

-0.74

 Verse

-0.70

ãģĦ

-0.69

vernment

-0.68

omsky

-0.64

 Detected

-0.64

ankind

-0.63

 renewables

-0.62

POSITIVE LOGITS

 precinct

1.02

 district

0.83

 pharmacy

0.82

 plaza

0.81

 headquarters

0.79

 clerk

0.79

stead

0.78

 warehouse

0.76

 estate

0.76

 depot

0.76

Activations Density 0.184%

locations, especially stores, supermarkets, malls, and apartment complexes

locations related to various types of businesses and establishments

No Comments

No Known Activations