INDEX

Explanations

phrases indicating self-promotion or promotion of one's own content

oai_token-act-pair · gpt-3.5-turbo

references to personal or community-related projects and content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

needed

-0.77

esters

-0.76

 belongs

-0.75

payers

-0.75

oke

-0.75

forces

-0.74

arians

-0.74

rays

-0.74

okemon

-0.72

ype

-0.71

POSITIVE LOGITS

own

1.07

 latest

0.99

 blog

0.98

 inaugural

0.94

 newest

0.94

 introductory

0.93

 exhaustive

0.90

 upcoming

0.89

 portfolio

0.89

 guide

0.88

Activations Density 0.232%

phrases indicating self-promotion or promotion of one's own content

references to personal or community-related projects and content

No Comments

No Known Activations