INDEX

Explanations

specific mentions of locations and institutions, potentially related to education, healthcare, and legal matters

oai_token-act-pair · gpt-3.5-turbo

significant titles and phrases within structured content or headings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ãĤ´ãĥ³

-0.78

lihood

-0.67

emonium

-0.61

bilt

-0.59

utenberg

-0.59

IST

-0.58

pire

-0.58

terness

-0.58

osaurus

-0.58

azeera

-0.57

POSITIVE LOGITS

 srfAttach

0.47

emetery

0.47

ederal

0.46

ĸ

0.45

river

0.45

ĺ

0.44

 opio

0.44

 vain

0.44

SPONSORED

0.43

ļ

0.43

Activations Density 0.339%

specific mentions of locations and institutions, potentially related to education, healthcare, and legal matters

significant titles and phrases within structured content or headings

No Comments

No Known Activations