INDEX

Explanations

rules or guidelines expressed in document text

oai_token-act-pair · gpt-3.5-turbo

mentions of guidelines or regulations

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

ONSORED

-0.95

igated

-0.84

issance

-0.80

igate

-0.74

ienced

-0.70

rial

-0.69

attled

-0.68

rique

-0.67

vironment

-0.67

leased

-0.66

POSITIVE LOGITS

breakers

0.85

breaker

0.80

 lawy

0.78

 governing

0.76

lessness

0.74

 abiding

0.71

witz

0.71

 dictate

0.71

 violations

0.70

book

0.70

Activations Density 0.031%

rules or guidelines expressed in document text

mentions of guidelines or regulations

No Comments

No Known Activations