INDEX

Explanations

proper nouns and technical terms

oai_token-act-pair · gpt-3.5-turbo

specific proper nouns or names of notable entities and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

abwe

-0.92

omever

-0.77

åĮ

-0.73

itored

-0.73

selves

-0.69

terday

-0.69

ecause

-0.67

perty

-0.67

DonaldTrump

-0.67

Ú

-0.65

POSITIVE LOGITS

iest

0.65

 portion

0.64

oret

0.62

hest

0.62

spring

0.60

 exception

0.60

 question

0.59

est

0.59

 shortage

0.58

 version

0.58

Activations Density 0.588%

proper nouns and technical terms

specific proper nouns or names of notable entities and concepts

No Comments

No Known Activations