INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

åŃĺäºİ

-0.17

Ð¾ÑĢÐ¾Ð´

-0.15

 whilst

-0.14

InstanceOf

-0.14

 camps

-0.14

 Fame

-0.14

.uni

-0.14

ortal

-0.13

orrh

-0.13

bower

-0.13

POSITIVE LOGITS

 mainly

0.18

 mostly

0.17

 merk

0.15

inder

0.15

fak

0.15

 tháºŃp

0.14

eb

0.14

 COVID

0.14

 circa

0.14

job

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.