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

íħĶ

-0.15

 addCriterion

-0.15

untime

-0.14

atron

-0.14

SSID

-0.14

kontakte

-0.14

 Rede

-0.13

voucher

-0.13

aal

-0.13

 Amendments

-0.13

POSITIVE LOGITS

 music

0.30

 labels

0.27

 Music

0.26

 label

0.26

 artists

0.25

 Artists

0.24

-label

0.24

 Labels

0.24

 artist

0.24

music

0.23

Activations Density 0.000%

No Known Activations

This feature has no known activations.