INDEX

Explanations

mentions of specific individuals or proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names related to individuals or entities

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

 Vaugh

-0.79

 toile

-0.65

CBI

-0.65

redo

-0.63

 Antar

-0.59

 Notting

-0.59

mot

-0.59

 fate

-0.58

ettings

-0.58

 canyon

-0.57

POSITIVE LOGITS

intosh

0.94

htar

0.81

ROR

0.79

ippi

0.79

arella

0.78

nih

0.78

ouf

0.77

agara

0.72

henko

0.70

angelo

0.70

Activations Density 0.808%

mentions of specific individuals or proper nouns

proper nouns and names related to individuals or entities

No Comments

No Known Activations