INDEX

Explanations

instances of names or proper nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ury

-0.16

ulur

-0.16

iel

-0.14

 ÏĢÏīÏĤ

-0.14

alcon

-0.14

.btnAdd

-0.14

 filib

-0.14

 advoc

-0.13

 desn

-0.13

oller

-0.13

POSITIVE LOGITS

aur

0.28

 Mein

0.23

Aur

0.22

 mere

0.20

 Maine

0.20

Dil

0.20

 mein

0.19

Hai

0.19

dil

0.19

hai

0.19

Activations Density 0.088%

instances of names or proper nouns

No Comments

No Known Activations