INDEX

Explanations

food-related entities, such as restaurant names or dishes

oai_token-act-pair · gpt-3.5-turbo

unique identifiers and entities like names, places, or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

afety

-0.72

ricular

-0.65

rix

-0.62

worm

-0.61

rina

-0.60

ographer

-0.59

aya

-0.57

iever

-0.56

elf

-0.56

cko

-0.55

POSITIVE LOGITS

 respectively

2.08

 both

1.64

both

1.59

 Both

1.39

Both

1.37

 respective

1.25

 latter

1.23

two

1.19

 jointly

1.18

two

1.15

Activations Density 0.704%

food-related entities, such as restaurant names or dishes

unique identifiers and entities like names, places, or organizations

No Comments

No Known Activations