INDEX

Explanations

political roles and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 Swiss

-0.17

ozilla

-0.17

 Blasio

-0.15

 Switzerland

-0.15

onda

-0.15

orex

-0.14

kul

-0.14

Ð¾ÑĢÐ´Ð¸Ð½

-0.14

aka

-0.14

.flash

-0.14

POSITIVE LOGITS

CD

0.29

 Greens

0.26

 Bund

0.24

CD

0.23

 Merkel

0.22

SPD

0.22

 Chancellor

0.22

 chancellor

0.20

 Bundes

0.19

ancellor

0.19

Activations Density 0.042%

political roles and titles

No Comments

No Known Activations