INDEX

Explanations

technical terms or names related to different types of decks, resources, experts, services, performances, decisions, tests, impacts, positions, priorities, and details

oai_token-act-pair · gpt-3.5-turbo

specific terms and names associated with various categories or topics, particularly in a structured or procedural context

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etheless

-0.80

iannopoulos

-0.70

ĻĤ

-0.68

ancies

-0.65

arching

-0.65

 autonom

-0.64

itably

-0.64

 heel

-0.63

bsite

-0.63

 charism

-0.62

POSITIVE LOGITS

 Builder

1.09

 Room

1.06

 Mode

1.03

 Breaker

1.03

 Manager

1.02

 Desk

1.01

 Guides

0.97

 Management

0.97

 Rooms

0.96

 Lounge

0.95

Activations Density 0.403%

technical terms or names related to different types of decks, resources, experts, services, performances, decisions, tests, impacts, positions, priorities, and details

specific terms and names associated with various categories or topics, particularly in a structured or procedural context

No Comments

No Known Activations