INDEX

Explanations

phrases related to monarchy and power

oai_token-act-pair · gpt-3.5-turbo

references to monarchy and royal figures, particularly kings and queens

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

Journal

-0.73

 Investigations

-0.71

 Parenthood

-0.70

iance

-0.69

ector

-0.69

 Activities

-0.65

 Disability

-0.64

 Liter

-0.63

 Investors

-0.62

 Inspection

-0.62

POSITIVE LOGITS

pin

1.27

pins

1.11

esses

1.09

 king

0.93

lord

0.84

 prince

0.83

alties

0.83

ess

0.82

 decree

0.82

naire

0.80

Activations Density 0.043%

phrases related to monarchy and power

references to monarchy and royal figures, particularly kings and queens

No Comments

No Known Activations