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

Negative Logits

icken

-0.15

aru

-0.14

ancellor

-0.14

esta

-0.14

ãĥ¼

-0.14

lick

-0.13

wald

-0.13

ialect

-0.13

ogy

-0.13

ÑĥÑĩÐ°Ñģ

-0.13

POSITIVE LOGITS

 Folk

0.28

 folk

0.27

folk

0.23

fol

0.22

 Americ

0.22

Fol

0.20

fol

0.19

Americ

0.17

 folks

0.17

ÑĢÐ¸Ð²

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.