INDEX

Explanations

words related to regulations or regulatory bodies

oai_token-act-pair · gpt-3.5-turbo

terms related to regulations or governing principles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

¿½

-0.95

lihood

-0.76

ĺħ

-0.76

 Haram

-0.74

=-=-

-0.70

ACP

-0.68

\\\\\\\\

-0.67

Fine

-0.66

 Dangerous

-0.64

chal

-0.64

POSITIVE LOGITS

ardless

1.18

arious

1.04

nant

1.04

istration

1.03

arded

1.00

rett

1.00

ulatory

0.98

ulators

0.97

ulations

0.97

roup

0.94

Activations Density 0.014%

words related to regulations or regulatory bodies

terms related to regulations or governing principles

No Comments

No Known Activations