INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

the name "Joe" and its variations or related forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 à¨

-0.69

ifax

-0.67

 Sabha

-0.65

 Berks

-0.64

ifiable

-0.63

*/(

-0.62

medium

-0.61

 Chandra

-0.61

£ı

-0.61

ivari

-0.60

POSITIVE LOGITS

oe

1.15

zzi

1.02

ppel

0.96

ze

0.90

lete

0.89

zzo

0.87

zy

0.87

zie

0.85

cci

0.85

uth

0.85

Activations Density 0.008%

proper nouns

the name "Joe" and its variations or related forms

No Comments

No Known Activations