INDEX

Explanations

the word "out."

oai_token-act-pair · gpt-3.5-turbo

instances of the word "out" in various forms and contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 arsen

-0.95

avorite

-0.69

EStream

-0.68

--------------------------------

-0.66

interstitial

-0.66

itably

-0.65

UTERS

-0.63

=-=-=-=-=-=-=-=-

-0.63

jriwal

-0.62

ute

-0.58

POSITIVE LOGITS

dated

1.21

rage

1.20

raged

1.14

doors

1.12

landish

1.12

door

1.04

come

1.04

breaks

1.04

numbered

1.03

look

1.01

Activations Density 0.040%

the word "out."

instances of the word "out" in various forms and contexts

No Comments

No Known Activations