INDEX

Explanations

occurrences of the word "Deutsch" or its variations, indicating a focus on German language or culture

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

Erk

-0.17

mi

-0.17

loon

-0.16

ÌĢ

-0.16

piration

-0.15

olang

-0.15

OrElse

-0.15

phet

-0.14

èµĸ

-0.14

grams

-0.14

POSITIVE LOGITS

uts

0.39

utsch

0.32

utschen

0.29

utch

0.27

utsche

0.26

usch

0.25

chant

0.23

zent

0.20

utches

0.20

ich

0.19

Activations Density 0.007%

occurrences of the word "Deutsch" or its variations, indicating a focus on German language or culture

No Comments

No Known Activations