INDEX

Explanations

mentions of the word "federal" in the context of legal or governmental discussions

oai_token-act-pair · gpt-3.5-turbo

mentions of federal laws and regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

-0.92

zzo

-0.87

enger

-0.85

 Clever

-0.80

 Rooms

-0.79

sbm

-0.77

Maker

-0.76

 Tycoon

-0.74

Dog

-0.73

Wo

-0.72

POSITIVE LOGITS

 government

1.05

 agencies

1.01

 appellate

1.01

 agency

0.93

 authorities

0.93

 oversight

0.89

law

0.87

 dollars

0.87

 jurisdiction

0.87

 legislation

0.86

Activations Density 0.029%

mentions of the word "federal" in the context of legal or governmental discussions

mentions of federal laws and regulations

No Comments

No Known Activations