INDEX

Explanations

references to presidential proclamations or executive orders

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

æľ¬

-0.14

 pled

-0.14

Locker

-0.14

cents

-0.13

 Becker

-0.13

 envelopes

-0.13

asion

-0.13

 elements

-0.13

 unborn

-0.13

hai

-0.13

POSITIVE LOGITS

 decree

0.34

 executive

0.32

 decre

0.31

 Executive

0.31

 orders

0.31

 Orders

0.28

 order

0.28

 Order

0.27

Executive

0.26

order

0.25

Activations Density 0.215%

references to presidential proclamations or executive orders

No Comments

No Known Activations