INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

hausen

-0.16

Jeh

-0.15

icus

-0.15

ilo

-0.15

 delegated

-0.14

 Lara

-0.14

ÑĨÐ¸Ð½

-0.14

 DataService

-0.14

 anomal

-0.14

POSITIVE LOGITS

agnar

0.17

ardy

0.16

unga

0.15

 ç¸

0.15

pez

0.15

OptionsMenu

0.14

eliac

0.14

oud

0.14

uning

0.14

 Greenwood

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.