INDEX

Explanations

verbs indicating compulsion or coercion

oai_token-act-pair · gpt-3.5-turbo

instances of the word "force" referring to legal or coercive actions

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

apest

-0.76

umer

-0.73

ership

-0.72

ight

-0.70

ahu

-0.67

ebted

-0.66

Hop

-0.66

rians

-0.66

ammy

-0.65

ergy

-0.65

POSITIVE LOGITS

 overtime

0.81

 force

0.77

 exerted

0.77

 compel

0.75

otom

0.71

 coer

0.71

 choke

0.69

 aside

0.68

 induction

0.68

maj

0.67

Activations Density 0.022%

verbs indicating compulsion or coercion

instances of the word "force" referring to legal or coercive actions

No Comments

No Known Activations