INDEX

Explanations

keywords related to editing and revision such as 'Edit'

oai_token-act-pair · gpt-3.5-turbo

instances of the word "Edit", indicating a focus on content editing or modifications

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

milo

-0.89

ciating

-0.80

 pigeon

-0.71

knife

-0.69

stro

-0.68

jri

-0.67

wagen

-0.66

gart

-0.64

gom

-0.64

lished

-0.63

POSITIVE LOGITS

 Appears

0.75

 Edit

0.71

ãĤ¨ãĥ«

0.71

earances

0.68

edit

0.66

 edit

0.65

 ãĤµ

0.63

 Delete

0.62

iola

0.62

edia

0.62

Activations Density 0.015%

keywords related to editing and revision such as 'Edit'

instances of the word "Edit", indicating a focus on content editing or modifications

No Comments

No Known Activations