INDEX

Explanations

elements related to relationships and dependencies 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 @ 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

elif

-0.16

ListOf

-0.15

ENSE

-0.15

orden

-0.14

aname

-0.14

uka

-0.14

_intf

-0.14

illow

-0.14

nds

-0.14

stell

-0.14

POSITIVE LOGITS

è¶Ĭ

0.30

ãģ»ãģ©

0.24

æĦ

0.24

 dest

0.23

ë¡Ŀ

0.23

 è¶

0.22

 ÑĤÐµÐ¼

0.21

 ÏĦÏĮÏĥÎ¿

0.19

sem

0.18

the

0.17

Activations Density 0.034%

elements related to relationships and dependencies in various contexts

No Comments

No Known Activations