INDEX

Explanations

proper nouns related to places, particularly focusing on Monaco

oai_token-act-pair · gpt-3.5-turbo

references to Monaco

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

umbnail

-0.83

sworth

-0.77

ience

-0.76

lihood

-0.74

ahead

-0.73

idan

-0.73

WIND

-0.73

LEY

-0.71

isdom

-0.71

à¨

-0.69

POSITIVE LOGITS

 Mongolia

0.93

 Monaco

0.86

eers

0.73

arov

0.73

illion

0.71

 Monteneg

0.71

 Clause

0.69

 Luxembourg

0.68

asonic

0.68

ixtape

0.68

Activations Density 0.017%

proper nouns related to places, particularly focusing on Monaco

references to Monaco

No Comments

No Known Activations