INDEX

Explanations

the presence of articles or quantifiers 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 @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ories

-0.17

-0.16

lights

-0.15

agal

-0.15

th

-0.15

Bitte

-0.15

illes

-0.14

kÄ±r

-0.14

erville

-0.14

ãĥ¼ãĥĬ

-0.14

POSITIVE LOGITS

 dozen

0.27

 hundred

0.22

 thousand

0.18

undred

0.17

decade

0.16

ught

0.15

Vog

0.14

Dek

0.14

 century

0.14

doz

0.14

Activations Density 0.049%

the presence of articles or quantifiers in various contexts

No Comments

No Known Activations