INDEX

Explanations

mentions of people's previous occupations or roles

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

luck

-0.84

orsi

-0.75

edly

-0.73

Results

-0.72

eming

-0.70

SPONSORED

-0.70

ulence

-0.70

Reason

-0.68

reasonable

-0.67

helm

-0.67

POSITIVE LOGITS

 midst

1.14

 guise

0.96

 Department

0.95

 vicinity

0.94

 Philippines

0.91

 department

0.91

 trenches

0.90

 field

0.90

 meantime

0.88

 aftermath

0.86

Activations Density 0.182%

mentions of people's previous occupations or roles

No Comments

No Known Activations