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

ounty

-0.17

ourg

-0.17

Deck

-0.14

holm

-0.14

ighb

-0.14

_mB

-0.14

ÑĢÑĥÐ½

-0.14

 Deck

-0.13

noho

-0.13

/Foundation

-0.13

POSITIVE LOGITS

pill

0.17

 Bird

0.15

rs

0.15

 Circular

0.15

Rs

0.15

zew

0.15

anth

0.15

PIL

0.15

 allegedly

0.14

 dues

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.