INDEX

Explanations

instances of the word "ace"

oai_token-act-pair · gpt-3.5-turbo

references to the term "ace" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Gordon

-0.62

 grid

-0.62

 residential

-0.60

hun

-0.60

 residents

-0.59

 digit

-0.58

 vertically

-0.58

 defaults

-0.57

 tongues

-0.57

IPO

-0.57

POSITIVE LOGITS

ace

4.68

aces

2.59

ACE

2.17

aced

1.96

acing

1.83

acea

1.78

acers

1.46

acy

1.39

acio

1.36

acer

1.32

Activations Density 0.018%

instances of the word "ace"

references to the term "ace" in various contexts

No Comments

No Known Activations