INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 ç¨

-0.16

viso

-0.16

hetto

-0.16

ulty

-0.15

asad

-0.15

CD

-0.15

eres

-0.14

zw

-0.14

Kid

-0.14

 corners

-0.14

POSITIVE LOGITS

ener

0.16

acket

0.16

Ã´ng

0.16

ENO

0.15

encoding

0.15

aba

0.15

Ð¾ÐºÐ°Ð·

0.14

ossier

0.14

ument

0.14

 joint

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.