INDEX

Explanations

proper nouns related to geographical locations, potentially focusing on Los Angeles

oai_token-act-pair · gpt-3.5-turbo

references to the city of Los Angeles and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Brill

-0.57

oline

-0.57

BuyableInstoreAndOnline

-0.56

 occas

-0.56

opter

-0.55

 Blessing

-0.54

 enthusi

-0.53

odox

-0.53

Log

-0.53

ortunately

-0.53

POSITIVE LOGITS

ION

0.82

IONS

0.72

IRO

0.70

Ã©s

0.66

ESCO

0.66

TO

0.64

IAL

0.64

INGTON

0.64

¨

0.64

TON

0.64

Activations Density 0.066%

proper nouns related to geographical locations, potentially focusing on Los Angeles

references to the city of Los Angeles and its variations

No Comments

No Known Activations