INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

:;↵

-0.15

_PHP

-0.14

agma

-0.14

uddy

-0.14

extr

-0.13

ãĥ§

-0.13

facet

-0.13

 addCriterion

-0.13

Ú©ÙĦ

-0.13

ccion

-0.13

POSITIVE LOGITS

 others

0.17

 normals

0.15

Ø·Ø§Ø±

0.15

edad

0.15

0.14

ulan

0.14

Permission

0.14

//

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.