INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

Ã³s

-0.17

istrovstvÃŃ

-0.17

zig

-0.17

ãĤŃãĥ¥

-0.15

entin

-0.14

Jed

-0.14

ÙİØ³

-0.14

enr

-0.14

 honors

-0.13

 Defense

-0.13

POSITIVE LOGITS

 Kingston

0.17

 Jamaica

0.16

 Trinidad

0.16

 Caribbean

0.16

‘

0.16

 WhatsApp

0.15

 bruk

0.15

 island

0.15

 lekker

0.14

WhatsApp

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.