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

anner

-0.14

¢ħ

-0.14

.mozilla

-0.14

ayas

-0.14

zug

-0.13

_SENSOR

-0.13

uars

-0.13

éĲĺ

-0.13

 Boyd

-0.13

mys

-0.13

POSITIVE LOGITS

 ever

0.19

ever

0.17

 EVER

0.15

.import

0.15

-ever

0.15

ropoda

0.15

 Moto

0.14

 fathers

0.14

Met

0.14

ilibrium

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.