INDEX

Explanations

Keywords or phrases indicating website-related actions or navigation elements

oai_token-act-pair · gpt-3.5-turbo

content related to updates and navigation prompts on web pages

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gd

-0.60

eval

-0.59

ctor

-0.57

↵Âł

-0.57

iture

-0.56

usher

-0.55

forth

-0.54

rehens

-0.54

ibly

-0.53

cence

-0.53

POSITIVE LOGITS

cloneembedreportprint

0.60

 Content

0.59

 Categories

0.58

 Created

0.56

 Subscribe

0.55

 Cancel

0.54

 Favorite

0.54

 Loading

0.53

owered

0.52

 Contents

0.51

Activations Density 0.107%

Keywords or phrases indicating website-related actions or navigation elements

content related to updates and navigation prompts on web pages

No Comments

No Known Activations