INDEX

Explanations

mentions of Russian involvement or interference

oai_token-act-pair · gpt-3.5-turbo

references to Russian involvement or interference in political activities

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

 Towns

-0.92

Thom

-0.78

ufact

-0.77

anty

-0.72

dash

-0.72

 Guinness

-0.71

aris

-0.71

 Dart

-0.65

erick

-0.64

 Prairie

-0.64

POSITIVE LOGITS

 hacking

1.21

 meddling

1.20

 interference

1.11

 interfered

1.08

 medd

1.07

 collusion

1.06

 dossier

1.05

 hacks

1.04

 operatives

1.03

 Kislyak

1.03

Activations Density 0.095%

mentions of Russian involvement or interference

references to Russian involvement or interference in political activities

No Comments

No Known Activations