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

 Malik

-0.15

 Bedford

-0.15

 Tehran

-0.14

 Ents

-0.14

 Derby

-0.14

 Jacob

-0.14

Ð°ÐºÐ¾Ð²

-0.14

à¸£à¸ģ

-0.13

cean

-0.13

stride

-0.13

POSITIVE LOGITS

 Bangkok

0.46

 Thai

0.46

 Thailand

0.44

Thai

0.42

 Harris

0.35

 thai

0.35

 Camb

0.32

THB

0.31

 à¸¿

0.31

 Cambodia

0.29

Activations Density 0.000%

No Known Activations

This feature has no known activations.