INDEX

Explanations

references to leadership positions and official roles in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

agne

-0.18

ifo

-0.17

deaux

-0.16

frei

-0.14

utex

-0.14

Registrar

-0.14

æ½

-0.14

irsch

-0.13

Margins

-0.13

iste

-0.13

POSITIVE LOGITS

 leadership

0.36

 senior

0.32

leaders

0.30

 highest

0.29

 officials

0.29

 higher

0.27

Leaders

0.27

 leaders

0.27

top

0.26

 official

0.26

Activations Density 0.280%

references to leadership positions and official roles in various contexts

No Comments

No Known Activations