INDEX

Explanations

references to divisions or classifications, especially in organizational or competitive contexts

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

ãĤ¹ãĥĨãĤ£

-0.15

apis

-0.15

ÄĻp

-0.15

iw

-0.15

brush

-0.15

bras

-0.14

licas

-0.14

azzo

-0.14

onden

-0.14

ystems

-0.14

POSITIVE LOGITS

al

0.28

/div

0.27

ison

0.21

isions

0.20

idend

0.19

ISION

0.19

 Ø¨ÙĨØ¯ÛĮ

0.17

naire

0.17

idable

0.16

iners

0.16

Activations Density 0.042%

references to divisions or classifications, especially in organizational or competitive contexts

No Comments

No Known Activations