INDEX

Explanations

phrases related to obligations and responsibilities

oai_token-act-pair · gpt-3.5-turbo

references to legal or ethical obligations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

vae

-0.77

auder

-0.75

oco

-0.73

corn

-0.73

vill

-0.71

stocks

-0.69

atin

-0.68

espie

-0.67

rooms

-0.67

ican

-0.67

POSITIVE LOGITS

 obligations

1.18

 obligation

1.12

 owed

1.06

 obliged

1.01

 obligated

0.93

 oblig

0.90

 incurred

0.89

 Compliance

0.87

 commitments

0.80

 compliance

0.80

Activations Density 0.038%

phrases related to obligations and responsibilities

references to legal or ethical obligations

No Comments

No Known Activations