INDEX

Explanations

references to the concept of popularity

oai_token-act-pair · gpt-3.5-turbo

references to popularity and its implications

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

thur

-0.78

alk

-0.76

Â¯Â¯

-0.75

erm

-0.73

rib

-0.73

 Neurolog

-0.73

ibur

-0.70

ural

-0.67

cise

-0.66

 Matter

-0.65

POSITIVE LOGITS

 popularity

0.99

popular

0.95

yip

0.89

itism

0.87

 Popular

0.81

iqueness

0.80

é¾įå¥ĳå£«

0.78

 unpopular

0.77

 ratings

0.76

jriwal

0.75

Activations Density 0.014%

references to the concept of popularity

references to popularity and its implications

No Comments

No Known Activations