INDEX

Explanations

references to sections or subsections in a document or article

oai_token-act-pair · gpt-3.5-turbo

specific statements or quotes related to events and actions or potential outcomes

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

ogun

-0.90

ividual

-0.80

ecause

-0.76

mosp

-0.75

dayName

-0.75

nian

-0.74

ashtra

-0.72

 ItemLevel

-0.71

 Volunte

-0.67

athered

-0.67

POSITIVE LOGITS

 antioxid

0.91

 corrid

0.62

ISI

0.60

âĹ¼

0.58

estyles

0.58

 Rohing

0.57

oples

0.57

Balt

0.56

 artif

0.56

EStream

0.54

Activations Density 6.180%

references to sections or subsections in a document or article

specific statements or quotes related to events and actions or potential outcomes

No Comments

No Known Activations