INDEX

Explanations

names of authors and their associated contributions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 ÙħØ³ÙĦÙħØ§ÙĨ

-0.16

ÑĢÐ¾Ñĩ

-0.16

ÃĹ↵↵

-0.16

amat

-0.15

brero

-0.15

 Cotton

-0.15

 ÙħØŃÙħÙĪØ¯

-0.14

ller

-0.14

Ø³Ø§ÙĦ

-0.14

ulet

-0.14

POSITIVE LOGITS

 Saudi

0.38

 Riyadh

0.36

Saudi

0.31

 Saudis

0.30

 Saud

0.28

 Kingdom

0.28

 Prince

0.26

 Arabia

0.25

 kingdom

0.24

 King

0.23

Activations Density 0.042%

names of authors and their associated contributions

No Comments

No Known Activations