INDEX

Explanations

words related to legal proceedings and court rulings

oai_token-act-pair · gpt-3.5-turbo

references to judges and judicial rulings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 weap

-0.82

ontent

-0.80

natureconservancy

-0.79

sold

-0.75

rolet

-0.72

ADS

-0.71

escent

-0.71

cov

-0.70

ãĤ£

-0.67

POSITIVE LOGITS

 rulings

1.21

Judge

1.20

 jurisdiction

1.18

 Judge

1.15

 justices

1.14

 judge

1.09

 appellate

1.07

 Judges

1.06

 ruled

1.06

 presiding

1.05

Activations Density 0.503%

words related to legal proceedings and court rulings

references to judges and judicial rulings

No Comments

No Known Activations