INDEX

Explanations

terms related to social, political, economic, and historical conditions

oai_token-act-pair · gpt-3.5-turbo

references to socio-political and economic conditions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

wcsstore

-0.81

 reopen

-0.74

iasm

-0.71

bsp

-0.71

 WATCHED

-0.70

!--

-0.70

daq

-0.68

ĪĴ

-0.68

cano

-0.67

etsk

-0.67

POSITIVE LOGITS

 nations

0.93

 individuals

0.90

 Nations

0.89

 mankind

0.87

 countries

0.83

 Europeans

0.82

 persons

0.81

 humankind

0.81

 peoples

0.79

 foreigners

0.78

Activations Density 0.317%

terms related to social, political, economic, and historical conditions

references to socio-political and economic conditions

No Comments

No Known Activations