INDEX

Explanations

The neuron is looking for words related to loyalty, steadfastness, and dedication

oai_token-act-pair · gpt-3.5-turbo

terms related to stalwarts and associated political roles or conditions

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

ramid

-0.78

 Agency

-0.72

 Simulator

-0.71

NetMessage

-0.71

 DRAG

-0.70

 Loans

-0.68

 Nationwide

-0.68

 Pants

-0.66

yre

-0.66

 Emirates

-0.66

POSITIVE LOGITS

emate

1.43

wart

1.37

warts

1.27

 stal

1.04

eness

1.03

stal

0.86

ances

0.85

atial

0.85

staff

0.83

voc

0.80

Activations Density 0.014%

The neuron is looking for words related to loyalty, steadfastness, and dedication

terms related to stalwarts and associated political roles or conditions

No Comments

No Known Activations