INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

references to locations, conditions, or contexts related to various activities and events

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rued

-0.72

iolet

-0.65

bern

-0.64

atom

-0.63

aceae

-0.63

izons

-0.60

ixel

-0.60

cit

-0.60

earth

-0.59

uds

-0.59

POSITIVE LOGITS

 stopping

1.34

 delivering

1.27

 raising

1.25

 placing

1.25

 producing

1.24

 registering

1.23

 treating

1.23

 supplying

1.22

 providing

1.21

 taking

1.21

Activations Density 0.151%

locations or places

references to locations, conditions, or contexts related to various activities and events

No Comments

No Known Activations