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

icom

-0.15

ë§Ĳ

-0.15

Ð´ÑĢÐµÑģ

-0.15

bones

-0.14

lej

-0.14

amodel

-0.14

eter

-0.14

 é¹

-0.14

emachine

-0.14

ep

-0.13

POSITIVE LOGITS

aÅ¼

0.16

flo

0.15

rese

0.15

onu

0.15

ivist

0.15

-production

0.15

ìĽħ

0.14

orse

0.14

vez

0.14

APS

0.14

Activations Density 0.032%

No Comments

No Known Activations