INDEX

Explanations

mentions of "work" in various contexts, including workplace activities, government departments, and data analytics tools

oai_token-act-pair · gpt-3.5-turbo

references to work-related topics or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 constitu

-0.83

 deaf

-0.67

PsyNetMessage

-0.61

Ved

-0.61

Eug

-0.59

Sov

-0.59

awe

-0.58

 subdu

-0.57

 darkness

-0.57

 Augustus

-0.56

POSITIVE LOGITS

bench

1.20

hops

1.19

 ethic

1.02

station

0.99

fare

0.95

force

0.94

ington

0.92

hardt

0.91

Safe

0.91

horse

0.90

Activations Density 0.023%

mentions of "work" in various contexts, including workplace activities, government departments, and data analytics tools

references to work-related topics or institutions

No Comments

No Known Activations