INDEX

Explanations

mentions of situations where individuals are compelled to do something against their will

oai_token-act-pair · gpt-3.5-turbo

instances of coercion or compulsion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ership

-0.81

ergy

-0.75

ight

-0.73

issance

-0.73

VERTISEMENT

-0.70

yss

-0.69

 Excellence

-0.68

 Cannabis

-0.68

itect

-0.68

vironment

-0.66

POSITIVE LOGITS

ierrez

0.88

 laborers

0.84

 labou

0.80

 overtime

0.79

 forced

0.77

 cooker

0.73

 pneum

0.71

 wage

0.70

 marches

0.69

 coercion

0.69

Activations Density 0.019%

mentions of situations where individuals are compelled to do something against their will

instances of coercion or compulsion

No Comments

No Known Activations