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

odont

-0.17

ynn

-0.15

ustum

-0.15

Å¯r

-0.15

antry

-0.14

ä¸¹

-0.14

.symbol

-0.14

flen

-0.14

 Strand

-0.14

.Async

-0.14

POSITIVE LOGITS

 Silver

0.21

 Commission

0.21

 Associates

0.20

 commission

0.18

 Member

0.18

earn

0.18

Silver

0.18

 Traffic

0.17

 Associate

0.17

 silver

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.