INDEX

Explanations

text related to military operations or terms

oai_token-act-pair · gpt-3.5-turbo

references to specific geographic locations or entities related to conflicts and terrorism

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tycoon

-0.71

kefeller

-0.70

blem

-0.68

Cly

-0.67

ulators

-0.67

avorite

-0.64

 Liver

-0.64

eryl

-0.64

olars

-0.64

incinn

-0.63

POSITIVE LOGITS

eki

0.75

Grab

0.67

ĪĴ

0.66

AZ

0.62

esi

0.62

sha

0.61

iste

0.60

azi

0.60

Lago

0.59

Az

0.59

Activations Density 0.359%

text related to military operations or terms

references to specific geographic locations or entities related to conflicts and terrorism

No Comments

No Known Activations