INDEX

Explanations

statements involving actions, opinions, or events with associated emotions

oai_token-act-pair · gpt-3.5-turbo

quotes or statements made by individuals

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

 marvelous

-0.95

 badass

-0.93

Â®,

-0.86

–

-0.85

 goddamn

-0.83

(#

-0.82

luster

-0.79

center

-0.79

Pretty

-0.79

,)

-0.78

POSITIVE LOGITS

 organisations

1.38

 organisers

1.35

MPs

1.32

 councillors

1.25

 Britons

1.19

 licences

1.19

 criticised

1.17

 programmes

1.17

 campaigners

1.15

 neighbours

1.14

Activations Density 0.935%

statements involving actions, opinions, or events with associated emotions

quotes or statements made by individuals

No Comments

No Known Activations