INDEX

Explanations

names of politicians or public figures

oai_token-act-pair · gpt-3.5-turbo

parenthetical expressions or references to individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 fibre

-0.67

ãĥ¥

-0.67

ĸļ

-0.66

itory

-0.65

omorphic

-0.65

ãĥ´ãĤ¡

-0.59

eg

-0.59

Ĥª

-0.59

ãĥĦ

-0.58

yna

-0.58

POSITIVE LOGITS

inen

0.76

culosis

0.62

gaard

0.61

sen

0.61

ãĥİ

0.60

 Cursed

0.58

 Committees

0.57

 embarked

0.56

 famously

0.56

aughs

0.56

Activations Density 0.120%

names of politicians or public figures

parenthetical expressions or references to individuals

No Comments

No Known Activations