INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ovu

-0.17

åŃĿ

-0.15

èľĺèĽĽ

-0.15

ziy

-0.14

ÏĦÎ¹ÎºÎ±

-0.14

 fabrication

-0.14

 liqu

-0.13

erse

-0.13

zl

-0.13

onse

-0.13

POSITIVE LOGITS

 Apple

0.21

Mac

0.19

idel

0.16

.bg

0.15

Apple

0.15

 Developer

0.15

 developer

0.15

BITS

0.15

 Mountain

0.14

upp

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.