INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

-0.16

others

-0.15

 others

-0.15

olini

-0.14

 -↵

-0.14

ithub

-0.14

>>

-0.13

illow

-0.13

least

-0.13

Others

-0.13

POSITIVE LOGITS

/*!

0.14

_fast

0.14

iom

0.14

sop

0.14

 although

0.14

 which

0.14

oren

0.13

xes

0.13

ught

0.13

 LENG

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.