INDEX

Explanations

economic indicators related to income or consumption

oai_token-act-pair · gpt-3.5-turbo

references to per capita measurements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

rams

-0.73

ean

-0.70

ĪĴ

-0.66

 Crow

-0.64

alez

-0.63

spell

-0.58

 Chrom

-0.58

erman

-0.58

edi

-0.58

 Trojan

-0.57

POSITIVE LOGITS

 capita

1.14

ãĥ¥

0.87

ãĥ£

0.84

ibia

0.78

ruary

0.78

itarian

0.78

GDP

0.75

atown

0.74

%:

0.71

ãĤ¢ãĥ«

0.69

Activations Density 0.011%

economic indicators related to income or consumption

references to per capita measurements

No Comments

No Known Activations