INDEX

Explanations

references to political or legislative conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 selfie

-0.37

 Zika

-0.35

-0.34

 emoji

-0.33

 Lyft

-0.33

 selfies

-0.33

 Instagram

-0.33

 Brexit

-0.33

 WhatsApp

-0.32

POSITIVE LOGITS

0.43

 Bush

0.41

Bush

0.36

Û²Û°Û°

0.32

 iPod

0.32

 bush

0.27

 weblog

0.26

 Katrina

0.25

 McCain

0.23

DVD

0.22

Activations Density 0.863%

references to political or legislative conflicts

No Comments

No Known Activations