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

/ros

-0.15

 Legacy

-0.14

 advant

-0.14

age

-0.14

alue

-0.13

 overdue

-0.13

invisible

-0.13

plash

-0.13

GINE

-0.13

ros

-0.13

POSITIVE LOGITS

 Desire

0.26

 desire

0.25

 desires

0.25

 expenditure

0.22

 Couple

0.20

æ¬²

0.20

 couples

0.19

 coupling

0.19

 Desired

0.18

 Consumption

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.