INDEX

Explanations

mentions of nobility or being noble

oai_token-act-pair · gpt-3.5-turbo

references to nobility and noble qualities

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

DERR

-0.78

 Loaded

-0.68

 negatively

-0.67

arella

-0.65

apse

-0.64

 Bitcoins

-0.64

olver

-0.64

_>

-0.64

Buzz

-0.63

CDC

-0.63

POSITIVE LOGITS

minded

0.93

 deeds

0.92

ments

0.90

 indignation

0.87

 noble

0.86

men

0.83

 minded

0.83

heart

0.82

bour

0.81

man

0.81

Activations Density 0.031%

mentions of nobility or being noble

references to nobility and noble qualities

No Comments

No Known Activations