INDEX

Explanations

mentions of the National Rifle Association (NRA)

oai_token-act-pair · gpt-3.5-turbo

references to the NRA (National Rifle Association)

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

graded

-0.73

é£

-0.71

 Archdemon

-0.71

grades

-0.70

hai

-0.70

bourne

-0.68

Yi

-0.67

hw

-0.66

 Grail

-0.65

Doe

-0.65

POSITIVE LOGITS

 patriarch

0.84

istani

0.76

senal

0.75

 membership

0.73

IGHTS

0.72

IFIC

0.72

 firearms

0.72

ELY

0.70

rats

0.69

 Membership

0.69

Activations Density 0.022%

mentions of the National Rifle Association (NRA)

references to the NRA (National Rifle Association)

No Comments

No Known Activations