INDEX

Explanations

terms related to propaganda

oai_token-act-pair · gpt-3.5-turbo

references to propaganda and its various forms and implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ergy

-0.90

nings

-0.72

earth

-0.69

clud

-0.68

cker

-0.67

-0.65

Merit

-0.64

thur

-0.64

wives

-0.64

 Eucl

-0.64

POSITIVE LOGITS

aganda

1.14

 propaganda

0.90

 posters

0.88

 leaflets

0.87

 suppression

0.85

 disinformation

0.81

 dissemin

0.80

 blitz

0.77

 poster

0.76

 pamphlet

0.76

Activations Density 0.022%

terms related to propaganda

references to propaganda and its various forms and implications

No Comments

No Known Activations