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

Ð·Ð°Ð±

-0.15

æ¿

-0.15

erge

-0.14

.nlm

-0.14

andel

-0.14

åħ°

-0.14

sag

-0.13

lan

-0.13

_ra

-0.13

çºª

-0.13

POSITIVE LOGITS

 Global

0.17

GA

0.16

.Global

0.15

 Planet

0.15

 Financial

0.15

 ruling

0.15

Global

0.15

(Global

0.15

 financial

0.15

maf

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.