INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Mark."

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

ught

-0.80

 unnecess

-0.75

 vain

-0.71

 disposed

-0.70

 deliber

-0.70

urses

-0.69

 gladly

-0.69

 politic

-0.67

å¥³

-0.67

 traged

-0.67

POSITIVE LOGITS

eting

1.07

Mark

1.06

 Twain

1.03

mark

0.96

furt

0.95

owitz

0.93

down

0.91

emark

0.91

marks

0.88

 Mark

0.83

Activations Density 0.009%

proper nouns

mentions of the name "Mark."

No Comments

No Known Activations