INDEX

Explanations

adjectives that describe something as orderly, well-organized, or cleverly done

oai_token-act-pair · gpt-3.5-turbo

expressions of neatness or tidiness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Downloadha

-0.75

-0.67

CVE

-0.66

authorized

-0.65

 authorized

-0.64

çīĪ

-0.64

 Defenders

-0.64

ioxide

-0.63

phrine

-0.63

asions

-0.62

POSITIVE LOGITS

nesses

1.10

ness

0.92

 neat

0.90

ilde

0.87

ety

0.84

liness

0.83

 little

0.80

icles

0.78

 trick

0.76

tid

0.74

Activations Density 0.012%

adjectives that describe something as orderly, well-organized, or cleverly done

expressions of neatness or tidiness

No Comments

No Known Activations