INDEX

Explanations

titles or mentions of the word "Bed"

oai_token-act-pair · gpt-3.5-turbo

the term "Bed" in various contexts

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

 emot

-0.73

sv

-0.71

govtrack

-0.69

SU

-0.69

iggs

-0.68

 funnel

-0.67

 keyword

-0.65

ennial

-0.64

sr

-0.62

tf

-0.62

POSITIVE LOGITS

Bed

4.16

Bed

2.87

bed

1.51

 Bedford

1.48

bed

1.32

 beds

1.26

 Bath

1.21

Wet

1.15

 Sleep

1.05

 Sleeping

1.01

Activations Density 0.026%

titles or mentions of the word "Bed"

the term "Bed" in various contexts

No Comments

No Known Activations