INDEX

Explanations

phrases related to private entities or actions

oai_token-act-pair · gpt-3.5-turbo

mentions of private entities or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

orthy

-0.74

awa

-0.73

DAY

-0.71

————————

-0.69

amaz

-0.69

ostic

-0.69

unctions

-0.66

wolves

-0.65

ologic

-0.65

annis

-0.64

POSITIVE LOGITS

 sector

1.28

 equity

1.02

sector

0.96

 Sector

0.91

 jets

0.84

 investigator

0.80

 residences

0.80

 enterprise

0.80

 rented

0.76

 ownership

0.75

Activations Density 0.056%

phrases related to private entities or actions

mentions of private entities or organizations

No Comments

No Known Activations