INDEX

Explanations

phrases related to beds and resting places

oai_token-act-pair · gpt-3.5-turbo

mentions of beds and burial sites

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

VG

-0.80

PG

-0.63

 Mozilla

-0.62

hip

-0.62

advertising

-0.61

externalActionCode

-0.61

cript

-0.60

 trademark

-0.59

gor

-0.59

ocrat

-0.59

POSITIVE LOGITS

 beds

4.01

bed

1.66

 bedrooms

1.65

 tents

1.54

 desks

1.42

 blankets

1.41

 rooms

1.40

bed

1.39

 chairs

1.33

 graves

1.32

Activations Density 0.018%

phrases related to beds and resting places

mentions of beds and burial sites

No Comments

No Known Activations