INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ereum

-0.17

CORD

-0.15

chal

-0.14

ìĩ

-0.14

æĵ

-0.14

ÅĻeb

-0.14

ignon

-0.14

ereco

-0.14

xbe

-0.14

pher

-0.14

POSITIVE LOGITS

 Äĳá»Ŀi

0.16

ava

0.16

ziej

0.14

hcp

0.14

enko

0.14

ana

0.14

essler

0.14

anko

0.14

 Coastal

0.14

Ã¼ss

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.