INDEX

Explanations

phrases indicating instructions or tasks that need to be done

oai_token-act-pair · gpt-3.5-turbo

the phrase "all you need to do is."

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

iates

-0.66

iture

-0.65

atural

-0.64

Yor

-0.63

ishers

-0.63

 defic

-0.62

Registered

-0.62

itures

-0.61

eatures

-0.60

ised

-0.60

POSITIVE LOGITS

olation

0.90

ANGE

0.78

olated

0.77

 Solitaire

0.76

:]

0.74

SS

0.73

olate

0.71

plain

0.70

opus

0.67

nt

0.66

Activations Density 0.078%

phrases indicating instructions or tasks that need to be done

the phrase "all you need to do is."

No Comments

No Known Activations