INDEX

Explanations

mentions of investigators or authorities in news articles

oai_token-act-pair · gpt-3.5-turbo

terminology related to law enforcement activities and investigations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

gets

-0.76

Translation

-0.75

;;;;;;;;;;;;

-0.72

ocratic

-0.70

 hegemony

-0.67

comes

-0.67

ABLE

-0.66

icative

-0.65

ocide

-0.63

eful

-0.62

POSITIVE LOGITS

 allege

1.25

 searched

1.09

 believe

1.06

 speculate

1.06

 traced

1.01

 raided

1.00

 concluded

0.99

 discovered

0.99

 uncovered

0.97

 suspect

0.95

Activations Density 0.114%

mentions of investigators or authorities in news articles

terminology related to law enforcement activities and investigations

No Comments

No Known Activations