INDEX

Explanations

references to educational institutions and award-related events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

unca

-0.16

utow

-0.15

onta

-0.14

abar

-0.14

eil

-0.14

jin

-0.13

anzi

-0.13

 Testament

-0.13

POSITIVE LOGITS

 Arlington

0.28

 Fairfax

0.27

0.24

 Alexandria

0.23

 Rest

0.20

 Mason

0.20

 Tyson

0.18

 Shir

0.18

 Alexand

0.18

 Loud

0.18

Activations Density 0.026%

references to educational institutions and award-related events

No Comments

No Known Activations