INDEX

Explanations

information about people's occupations and workplaces

oai_token-act-pair · gpt-3.5-turbo

prepositions and phrases indicating employment and work-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

snap

-0.69

ptoms

-0.65

é»Ĵ

-0.64

 madness

-0.63

 Ambro

-0.62

 rites

-0.62

ibia

-0.62

 roar

-0.62

 eruption

-0.61

 forgiven

-0.61

POSITIVE LOGITS

 subcontract

0.75

 Fortress

0.74

 Walmart

0.72

jan

0.69

igslist

0.69

 JPMorgan

0.69

 Maxim

0.68

isoft

0.68

 Telephone

0.68

 Tess

0.67

Activations Density 0.149%

information about people's occupations and workplaces

prepositions and phrases indicating employment and work-related contexts

No Comments

No Known Activations