INDEX

Explanations

references to finding information or content online

oai_token-act-pair · gpt-3.5-turbo

references related to social media and online presence

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

 really

-0.58

 fairly

-0.56

 that

-0.55

 percept

-0.54

 decent

-0.54

 detectable

-0.54

 figuring

-0.53

 because

-0.53

 here

-0.53

hement

-0.53

POSITIVE LOGITS

ãĤ¼ãĤ¦ãĤ¹

0.64

gil

0.63

ãĥª

0.62

 Wife

0.60

wives

0.59

ãĥ¼ãĥ³

0.59

 ANGEL

0.58

PsyNetMessage

0.57

ovie

0.57

agents

0.56

Activations Density 1.098%

references to finding information or content online

references related to social media and online presence

No Comments

No Known Activations