INDEX

Explanations

terms and names associated with judges and judicial appointments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ©ãĥ¼

-0.15

Î®

-0.15

nev

-0.15

ickest

-0.14

 åĦ

-0.14

 Decoder

-0.14

HONE

-0.14

lient

-0.14

ALLEL

-0.13

éŃ

-0.13

POSITIVE LOGITS

 pres

0.35

 judge

0.26

 Pres

0.25

pres

0.22

Pres

0.22

 Judge

0.22

istrate

0.21

judge

0.21

Judge

0.21

 Circuit

0.20

Activations Density 0.067%

terms and names associated with judges and judicial appointments

No Comments

No Known Activations