INDEX

Explanations

phrases introducing new topics or sections in a text

oai_token-act-pair · gpt-3.5-turbo

instances of repeated characters or unusual symbols

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 mathemat

-0.79

 disadvant

-0.74

 doomed

-0.70

 incent

-0.65

 snowball

-0.64

 themselves

-0.64

 Orchestra

-0.64

 pigeon

-0.63

 comr

-0.61

 adolesc

-0.60

POSITIVE LOGITS

ï¸ı

1.08

there

0.90

ese

0.89

å§«

0.88

ould

0.86

shall

0.83

fter

0.81

ski

0.80

lishes

0.80

here

0.79

Activations Density 0.125%

phrases introducing new topics or sections in a text

instances of repeated characters or unusual symbols

No Comments

No Known Activations