INDEX

Explanations

phrases related to architectural features and design

oai_token-act-pair · gpt-3.5-turbo

references to the concept of architecture

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ml

-0.72

primary

-0.70

rd

-0.69

early

-0.67

 Hubbard

-0.67

ca

-0.67

eworthy

-0.66

henko

-0.66

Pos

-0.66

 Ventura

-0.66

POSITIVE LOGITS

itect

1.32

 architecture

1.16

urally

1.06

 Architecture

1.01

 architectures

0.95

ural

0.87

 architect

0.86

 architectural

0.85

 chops

0.83

 mismatch

0.79

Activations Density 0.016%

phrases related to architectural features and design

references to the concept of architecture

No Comments

No Known Activations