INDEX

Explanations

references to entities named "Nelson."

oai_token-act-pair · gpt-3.5-turbo

references to the name "Nelson."

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

fare

-0.95

orical

-0.77

place

-0.72

ãĤ¼

-0.70

orically

-0.69

================================================================

-0.67

¾

-0.67

CBI

-0.66

teen

-0.65

İ

-0.65

POSITIVE LOGITS

 Mandela

1.60

 Nelson

0.97

otine

0.88

 Rockefeller

0.83

sey

0.79

illo

0.77

elson

0.73

shire

0.71

issan

0.71

ensical

0.70

Activations Density 0.019%

references to entities named "Nelson."

references to the name "Nelson."

No Comments

No Known Activations