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

MLElement

-0.16

 Surrey

-0.16

borough

-0.15

 WARRANTIES

-0.15

Victoria

-0.15

PMC

-0.15

 Southampton

-0.14

 Richmond

-0.14

 Victoria

-0.14

POSITIVE LOGITS

 Iowa

0.39

 rural

0.25

 Sioux

0.24

 Rural

0.22

Des

0.20

 towns

0.19

ag

0.18

 farm

0.18

 recruiting

0.18

 young

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.