INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

uler

-0.16

icast

-0.15

Ø§ÙĦÙĬØ§

-0.14

iser

-0.14

ÐµÑĢÑĤ

-0.13

enco

-0.13

phen

-0.13

arent

-0.13

rup

-0.13

POSITIVE LOGITS

‘

0.21

Som

0.18

 Refugee

0.17

 meat

0.17

 resett

0.17

 refugee

0.16

’

0.16

 contractors

0.15

QRS

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.