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

Not in Any Lists

No Comments

Negative Logits

 Outputs

-0.17

NAMESPACE

-0.16

 outputs

-0.15

Outputs

-0.15

 tesis

-0.15

uzey

-0.14

åĺī

-0.14

allee

-0.14

exp

-0.14

ErrorException

-0.14

POSITIVE LOGITS

UniqueId

0.23

.unique

0.18

camp

0.17

 UNIQUE

0.17

External

0.16

 Ð¾ÑĢÐ³

0.16

unique

0.16

 unique

0.16

ugin

0.15

 pref

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.