INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

 Ð´Ð¾ÑĤ

-0.17

rej

-0.14

Cop

-0.14

inh

-0.14

 heav

-0.14

aths

-0.13

redd

-0.13

quez

-0.13

UDA

-0.13

äººãģ¯

-0.13

POSITIVE LOGITS

_MAXIMUM

0.15

oulouse

0.14

 nomine

0.14

tp

0.14

æŁ»

0.14

mitter

0.13

 subparagraph

0.13

adf

0.13

terr

0.13

ocket

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.