INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.16

sep

-0.16

iana

-0.16

 recept

-0.15

fn

-0.14

ess

-0.14

oub

-0.14

shr

-0.14

-c

-0.14

POSITIVE LOGITS

ImageContext

0.18

znik

0.15

opak

0.15

raki

0.15

senal

0.15

xec

0.15

xico

0.15

ÐµÐ»Ð¸Ñĩ

0.14

TZ

0.14

oples

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.