INDEX

Explanations

instances of government officials signing documents or bills into law

oai_token-act-pair · gpt-3.5-turbo

instances of the word "signed" in relation to official actions or documents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

gypt

-0.69

 nerv

-0.63

unker

-0.62

 pans

-0.62

 dexter

-0.62

Middle

-0.62

 destro

-0.60

 prow

-0.59

 wandering

-0.58

ibaba

-0.57

POSITIVE LOGITS

alled

1.02

atories

0.94

interstitial

0.87

 agreements

0.84

ificantly

0.83

deals

0.80

 NAFTA

0.80

 deleg

0.78

animous

0.77

 legislation

0.76

Activations Density 0.043%

instances of government officials signing documents or bills into law

instances of the word "signed" in relation to official actions or documents

No Comments

No Known Activations