INDEX

Explanations

the word "smart" with a high activation value

oai_token-act-pair · gpt-3.5-turbo

instances of the word "smart" used in various contexts

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

ĸļ

-0.84

 Divinity

-0.82

Reloaded

-0.81

ulhu

-0.76

alogue

-0.71

riott

-0.71

hedral

-0.68

channelAvailability

-0.67

ãĥĺãĥ©

-0.67

akeru

-0.66

POSITIVE LOGITS

ctl

0.90

ling

0.89

ness

0.89

guy

0.88

sonian

0.85

matic

0.85

ly

0.83

ass

0.79

ies

0.78

ening

0.77

Activations Density 0.016%

the word "smart" with a high activation value

instances of the word "smart" used in various contexts

No Comments

No Known Activations