INDEX

Explanations

various numerical identifiers and web-related elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 ç¤

-0.16

à¤¿à¤¬

-0.16

ãĤ¤ãĥ¤

-0.15

LBL

-0.14

.sourceforge

-0.14

ksam

-0.14

åĮ¹

-0.14

åı·

-0.14

illow

-0.13

Î¹ÎºÎ·

-0.13

POSITIVE LOGITS

0.20

0.17

 QName

0.14

âĢĮ

0.14

Ī

0.14

ainer

0.14

YPD

0.14

ucha

0.13

 Graves

0.13

Ð´ÑĢÐµÑģ

0.13

Activations Density 0.388%

various numerical identifiers and web-related elements

No Comments

No Known Activations