INDEX

Explanations

No Explanations Found

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

 proh

-0.16

sto

-0.15

rodu

-0.15

ytt

-0.15

alm

-0.14

ÏīÎ´

-0.14

rophe

-0.14

iÅ¡tÄĽ

-0.14

ndl

-0.14

 sunk

-0.14

POSITIVE LOGITS

 affairs

0.16

ucha

0.16

 GetName

0.15

-makers

0.14

RT

0.14

Net

0.14

Nev

0.14

ted

0.14

 ALERT

0.14

 aura

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.