INDEX

Explanations

words related to exemptions or exceptions from rules or regulations

oai_token-act-pair · gpt-3.5-turbo

phrases related to exemptions or exceptions to rules or laws

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

 Roses

-0.68

Kind

-0.68

 Roose

-0.66

 Bras

-0.65

 Herald

-0.65

 Angry

-0.64

Cao

-0.63

 Hurricanes

-0.62

 Beetle

-0.62

 Ange

-0.62

POSITIVE LOGITS

 exempt

1.27

exempt

1.12

 exempted

1.07

icut

1.02

untarily

1.02

ively

0.97

emption

0.89

 exemptions

0.85

otics

0.82

 minded

0.81

Activations Density 0.009%

words related to exemptions or exceptions from rules or regulations

phrases related to exemptions or exceptions to rules or laws

No Comments

No Known Activations