INDEX

Explanations

instances where something belongs to someone or a specific group

oai_token-act-pair · gpt-3.5-turbo

words related to ownership or belonging

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

olson

-0.82

imar

-0.67

ork

-0.64

rices

-0.64

airs

-0.63

urnal

-0.63

ilde

-0.62

ropolis

-0.62

annis

-0.62

aul

-0.61

POSITIVE LOGITS

 belong

0.95

 belonged

0.93

itialized

0.86

 belongs

0.85

 belonging

0.84

holder

0.81

ĪĴ

0.79

BuyableInstoreAndOnline

0.79

holders

0.75

 hers

0.73

Activations Density 0.010%

instances where something belongs to someone or a specific group

words related to ownership or belonging

No Comments

No Known Activations