INDEX

Explanations

phrases related to the term "ball"

oai_token-act-pair · gpt-3.5-turbo

terms related to ball games, particularly sports involving balls

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rend

-0.67

 Beir

-0.67

QUIRE

-0.66

asio

-0.62

 Liberties

-0.61

REM

-0.61

Rated

-0.61

rians

-0.60

MAT

-0.59

CRIP

-0.59

POSITIVE LOGITS

ball

1.23

oons

1.15

oon

1.14

balls

1.01

park

0.95

asketball

0.85

oom

0.84

ozo

0.84

wagen

0.81

eries

0.80

Activations Density 0.010%

phrases related to the term "ball"

terms related to ball games, particularly sports involving balls

No Comments

No Known Activations