INDEX

Explanations

proper nouns related to political figures and financial matters

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

mentions of financial terms and economic metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

Â·Â·

-0.59

 Spain

-0.57

 Remain

-0.56

ãĢĲ

-0.56

 Crusader

-0.56

 Slov

-0.55

 Gaul

-0.53

 Catalonia

-0.51

ERY

-0.51

 feder

-0.51

POSITIVE LOGITS

raint

0.87

cup

0.72

iating

0.69

ley

0.68

enth

0.68

ressing

0.66

aired

0.66

requent

0.65

rences

0.64

ial

0.64

Activations Density 0.525%

proper nouns related to political figures and financial matters

mentions of financial terms and economic metrics

No Comments

No Known Activations