INDEX

Explanations

concepts related to legal proceedings and jury instructions

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

 Judicial

-0.19

 judicial

-0.18

toc

-0.16

itom

-0.16

 Judiciary

-0.15

 prosecute

-0.15

 unilateral

-0.15

Jud

-0.15

 prosecuted

-0.14

.bd

-0.14

POSITIVE LOGITS

 jury

0.52

 jurors

0.47

jur

0.41

 Jury

0.40

jury

0.38

Jur

0.37

Jur

0.37

 verdict

0.36

jur

0.36

0.34

Activations Density 0.128%

concepts related to legal proceedings and jury instructions

No Comments

No Known Activations