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

æ³£

-0.15

 ÐļÑĢÐ¸

-0.15

rob

-0.14

OperationException

-0.13

Ã¡b

-0.13

extr

-0.13

hub

-0.13

_sparse

-0.13

acam

-0.12

POSITIVE LOGITS

 Katherine

0.23

 Hero

0.20

 Shin

0.20

 Kathy

0.19

Ts

0.18

 Heroes

0.18

atherine

0.17

 hero

0.17

Hero

0.17

 Cathy

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.