INDEX

Explanations

prepositions indicating location or position

oai_token-act-pair · gpt-3.5-turbo

references to being concealed or hidden beneath something

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

 Ahead

-0.65

ERG

-0.65

ebus

-0.63

 redistributed

-0.61

 Remastered

-0.59

 supra

-0.58

ibia

-0.57

 Retrieved

-0.57

 hereafter

-0.55

ymes

-0.55

POSITIVE LOGITS

neath

1.37

sea

1.20

pants

1.16

whelming

1.11

cooked

1.09

pins

1.09

brush

1.05

whel

1.03

 bridges

1.01

growth

1.01

Activations Density 0.044%

prepositions indicating location or position

references to being concealed or hidden beneath something

No Comments

No Known Activations