INDEX

Explanations

internet-related terms and actions

oai_token-act-pair · gpt-3.5-turbo

references to browsing and online features or services

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

sted

-0.84

 conject

-0.81

 theoret

-0.74

phil

-0.73

 theolog

-0.72

ajor

-0.70

 Jesuit

-0.70

stad

-0.69

 torn

-0.68

 secession

-0.68

POSITIVE LOGITS

 iCloud

1.24

 Gmail

1.23

Apps

1.18

 Dropbox

1.16

 Touch

1.09

 apps

1.08

Touch

1.05

Android

1.05

Sync

1.04

Xbox

1.04

Activations Density 0.857%

internet-related terms and actions

references to browsing and online features or services

No Comments

No Known Activations