INDEX

Explanations

references to government actions, policies, and regulations

oai_token-act-pair · gpt-3.5-turbo

phrases related to government spending and regulation changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

lihood

-0.73

 enqu

-0.70

 rumours

-0.68

 arte

-0.67

 expectations

-0.66

 Subtle

-0.66

 sightings

-0.66

 Plot

-0.65

 forgetting

-0.64

 realise

-0.63

POSITIVE LOGITS

 nonprofits

1.06

 waivers

0.99

 privat

0.98

 defund

0.97

 federally

0.96

 privatization

0.93

 vouchers

0.92

 reimburse

0.91

 Medicaid

0.90

 reimb

0.89

Activations Density 0.535%

references to government actions, policies, and regulations

phrases related to government spending and regulation changes

No Comments

No Known Activations