INDEX

Explanations

words related to specific geographical locations

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific references to organizations, particularly in a geographic or situational context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

WOR

-0.70

 compuls

-0.70

abis

-0.65

 Mahm

-0.62

CBD

-0.62

 Conn

-0.62

travel

-0.61

favorite

-0.61

AE

-0.59

RF

-0.59

POSITIVE LOGITS

bian

2.25

Ox

1.59

 Rise

1.32

cat

1.31

Hut

1.24

 Mono

1.23

LEASE

1.16

ox

0.96

 trickle

0.92

 arra

0.91

Activations Density 0.055%

words related to specific geographical locations

proper nouns and specific references to organizations, particularly in a geographic or situational context

No Comments

No Known Activations