INDEX

Explanations

words related to regulatory or de-regulatory processes

oai_token-act-pair · gpt-3.5-turbo

terms related to regulation

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

¿½

-1.02

lihood

-0.83

ĺħ

-0.77

\\\\\\\\

-0.71

 Haram

-0.70

Fine

-0.69

 Misty

-0.68

 Dangerous

-0.68

=-=-

-0.67

ĪĴ

-0.63

POSITIVE LOGITS

nant

1.14

ardless

1.10

arded

1.04

ulatory

1.02

istration

0.99

rett

0.99

arious

0.98

roup

0.96

inal

0.95

ulations

0.95

Activations Density 0.008%

words related to regulatory or de-regulatory processes

terms related to regulation

No Comments

No Known Activations