INDEX

Explanations

specific items or instances within a category

oai_token-act-pair · gpt-3.5-turbo

sequences or mentions of singular items and their details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

mire

-0.72

ĸļ

-0.71

bara

-0.68

galitarian

-0.67

 Madness

-0.67

ipeg

-0.66

 Omaha

-0.63

 Lisp

-0.62

ivist

-0.62

MPG

-0.61

POSITIVE LOGITS

DragonMagazine

0.82

catentry

0.79

pread

0.69

 remnants

0.67

wcsstore

0.64

tumblr

0.63

FactoryReloaded

0.63

BuyableInstoreAndOnline

0.61

 geographically

0.60

unts

0.59

Activations Density 0.715%

specific items or instances within a category

sequences or mentions of singular items and their details

No Comments

No Known Activations