INDEX

Explanations

proper names

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names and initials, particularly those of individuals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

WT

-0.62

bledon

-0.61

ONSORED

-0.61

FACE

-0.61

agogue

-0.61

 Juda

-0.60

 Lanka

-0.58

 Guinea

-0.57

PASS

-0.57

 Wonder

-0.56

POSITIVE LOGITS

atform

0.77

ushing

0.68

iggs

0.66

minus

0.63

Balt

0.61

iera

0.59

\":

0.58

Ã£

0.57

assetsadobe

0.57

rophic

0.57

Activations Density 0.326%

proper names

proper nouns, specifically names and initials, particularly those of individuals

No Comments

No Known Activations