INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

à¸£à¸ĵ

-0.16

yal

-0.15

ecal

-0.15

iets

-0.14

ØŃØ«

-0.14

 Binder

-0.14

Ð²Ð¸Ð½

-0.14

Å±

-0.14

ERCHANT

-0.14

POSITIVE LOGITS

 Martial

0.23

 martial

0.19

 planetary

0.18

 Deep

0.18

 OPERATION

0.16

 Bout

0.16

 Brennan

0.16

 kraj

0.15

Deep

0.15

 COVID

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.