INDEX

Explanations

phrases indicating the general truth or validity of a statement

oai_token-act-pair · gpt-3.5-turbo

statements affirming or asserting truth

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

Pages

-0.80

adish

-0.76

rador

-0.75

uled

-0.75

 Sunder

-0.70

emetery

-0.70

 Quit

-0.70

vernment

-0.69

acco

-0.68

RAW

-0.68

POSITIVE LOGITS

 believer

0.81

 believers

0.80

terday

0.75

rill

0.72

 Lutheran

0.69

hood

0.65

 emancipation

0.64

stic

0.64

 positives

0.64

 patriot

0.62

Activations Density 0.022%

phrases indicating the general truth or validity of a statement

statements affirming or asserting truth

No Comments

No Known Activations