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

	templateUrl

-0.14

ehr

-0.14

ä¸Ī

-0.14

 Ð²ÐµÐ´ÑĮ

-0.14

wood

-0.14

lea

-0.13

çİ©

-0.13

abouts

-0.13

iday

-0.13

POSITIVE LOGITS

 native

0.20

native

0.20

 natives

0.19

Native

0.17

_native

0.17

 Native

0.16

 benef

0.16

 profession

0.15

Timeline

0.15

Timing

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.