INDEX

Explanations

terms related to labor, workforce, and employment conditions

oai_token-act-pair · gpt-3.5-turbo

references to labor and employment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

irs

-0.86

 Flavoring

-0.84

 Seym

-0.75

 Qiao

-0.69

resy

-0.69

Nib

-0.67

 Sark

-0.66

 Robertson

-0.65

Dak

-0.64

rah

-0.63

POSITIVE LOGITS

 labor

0.96

 unions

0.89

 laborers

0.80

 intensive

0.79

 stopp

0.78

abor

0.78

 labour

0.78

lust

0.78

ayers

0.77

Labor

0.77

Activations Density 0.014%

terms related to labor, workforce, and employment conditions

references to labor and employment

No Comments

No Known Activations