INDEX

Explanations

names or terms starting with the letters "Er" and potentially followed by other characters

oai_token-act-pair · gpt-3.5-turbo

words related to individuals with specific names or titles

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

utra

-0.70

 oats

-0.69

upd

-0.65

 nationally

-0.65

 breastfeeding

-0.64

ccoli

-0.64

 carrots

-0.64

yip

-0.64

 broccoli

-0.63

 maturity

-0.62

POSITIVE LOGITS

ipel

1.06

ements

0.78

 Pradesh

0.75

abeth

0.73

ption

0.73

Â¯

0.71

onomic

0.71

odox

0.70

mann

0.68

meyer

0.68

Activations Density 0.094%

names or terms starting with the letters "Er" and potentially followed by other characters

words related to individuals with specific names or titles

No Comments

No Known Activations