INDEX

Explanations

legal or judiciary-related terms

oai_token-act-pair · gpt-3.5-turbo

references to judicial processes and the legal system

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

ulative

-0.83

lund

-0.82

eem

-0.81

lished

-0.78

ku

-0.75

ols

-0.74

 Banner

-0.73

kok

-0.73

rug

-0.72

imus

-0.71

POSITIVE LOGITS

 rulings

1.02

 justice

1.00

 oversight

0.91

 activism

0.90

 judicial

0.88

justice

0.88

 administration

0.88

 scrutiny

0.88

 inquiry

0.87

 enforcement

0.84

Activations Density 0.007%

legal or judiciary-related terms

references to judicial processes and the legal system

No Comments

No Known Activations