INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

WO

-0.15

Ĺ

-0.15

DNA

-0.15

 Ð²ÐµÑĤ

-0.14

gue

-0.14

è¦ª

-0.13

rosso

-0.13

progressbar

-0.13

ourn

-0.13

exo

-0.13

POSITIVE LOGITS

urm

0.17

urd

0.15

ãĥĥãĥĹ

0.15

_warning

0.15

mekte

0.15

Uh

0.14

bnb

0.14

å¤®

0.14

asaki

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.