INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

pons

-0.15

 Ridley

-0.15

 Mash

-0.14

ensi

-0.14

ahren

-0.14

addtogroup

-0.14

adiator

-0.14

kiye

-0.14

 millenn

-0.14

Dirty

-0.14

POSITIVE LOGITS

 Miracle

0.23

Mir

0.23

Mir

0.20

 session

0.18

 Thompson

0.18

mir

0.17

 records

0.17

Mot

0.17

 Session

0.16

Son

0.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.