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

Naj

-0.15

.Factory

-0.15

 property

-0.14

/entities

-0.14

_strerror

-0.14

iox

-0.13

.Navigator

-0.13

ledge

-0.13

 spins

-0.13

PROPERTY

-0.13

POSITIVE LOGITS

sym

0.20

 Body

0.19

Lit

0.18

Ty

0.18

Lit

0.17

 body

0.17

quote

0.17

 Prim

0.17

ty

0.16

ÐµÐ²Ð¸Ñĩ

0.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.