INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

.scalablytyped

-0.22

idth

-0.15

luet

-0.15

igest

-0.14

.metamodel

-0.14

quential

-0.14

urm

-0.14

smarty

-0.14

crm

-0.14

atron

-0.14

POSITIVE LOGITS

_DISABLED

0.14

.amazon

0.13

 favour

0.13

Ð·Ð¸

0.13

fav

0.13

lÃ¤

0.13

ieten

0.13

æĥ

0.13

sis

0.13

foc

0.12

Activations Density 0.000%

No Known Activations

This feature has no known activations.