INDEX

Explanations

phrases where a statement or fact is being confirmed or validated

oai_token-act-pair · gpt-3.5-turbo

assertions of truthfulness or validity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

adish

-0.82

Pages

-0.77

aida

-0.73

roman

-0.73

rador

-0.73

hens

-0.72

vernment

-0.70

uled

-0.70

isoft

-0.69

 Sunder

-0.69

POSITIVE LOGITS

terday

0.86

 believers

0.83

hood

0.79

 believer

0.78

fully

0.68

polit

0.68

 Nationwide

0.68

 regardless

0.67

ually

0.67

 universally

0.66

Activations Density 0.021%

phrases where a statement or fact is being confirmed or validated

assertions of truthfulness or validity

No Comments

No Known Activations