INDEX

Explanations

edits made to a text

oai_token-act-pair · gpt-3.5-turbo

instances of the word "edit" or related terms indicating editing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Bucc

-0.65

avement

-0.60

cling

-0.59

 squared

-0.59

gging

-0.59

 Jindal

-0.58

fund

-0.58

 Fernand

-0.58

gart

-0.58

footed

-0.58

POSITIVE LOGITS

edit

0.94

ional

0.92

ors

0.89

orship

0.88

Edit

0.85

IONS

0.82

OR

0.81

 EDIT

0.81

rator

0.78

arro

0.76

Activations Density 0.011%

edits made to a text

instances of the word "edit" or related terms indicating editing actions

No Comments

No Known Activations