INDEX

Explanations

questions or statements emphasizing the obvious or straightforward

oai_token-act-pair · gpt-3.5-turbo

phrases indicating something as clearly discernible or evident

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

rology

-0.74

lished

-0.74

eries

-0.71

uden

-0.70

orously

-0.70

ntil

-0.70

utics

-0.70

cedented

-0.70

nan

-0.69

ldom

-0.68

POSITIVE LOGITS

iary

1.26

 culprit

0.99

 contradiction

0.99

 implication

0.90

 drawback

0.90

 omission

0.88

 resemblance

0.86

 impossibility

0.85

 absurdity

0.83

 downside

0.83

Activations Density 0.065%

questions or statements emphasizing the obvious or straightforward

phrases indicating something as clearly discernible or evident

No Comments

No Known Activations