INDEX

Explanations

locations or place names that are specific

oai_token-act-pair · gpt-3.5-turbo

locations, particularly cities and their associated characteristics

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Kazakh

-0.75

 ultras

-0.75

 Pastebin

-0.73

 Coord

-0.69

 Mega

-0.68

 deflation

-0.67

 capsule

-0.65

 Apocalypse

-0.65

soDeliveryDate

-0.64

 Haram

-0.63

POSITIVE LOGITS

ford

1.26

bury

1.20

inson

1.14

borough

1.13

eson

1.11

ton

1.10

patrick

1.09

ston

1.06

rence

1.05

ham

1.04

Activations Density 0.416%

locations or place names that are specific

locations, particularly cities and their associated characteristics

No Comments

No Known Activations