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

solete

-0.15

arda

-0.15

.tif

-0.13

loff

-0.13

delegate

-0.13

ermann

-0.13

 Kitt

-0.13

/AFP

-0.13

acco

-0.13

ider

-0.13

POSITIVE LOGITS

 Survey

0.22

 anonymous

0.21

 anonymity

0.20

 rating

0.20

anonymous

0.19

Survey

0.19

FBI

0.18

rat

0.18

DOJ

0.18

.rating

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.