INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.17

 Britt

-0.15

eri

-0.15

ç½®

-0.15

chat

-0.14

ans

-0.14

ilet

-0.14

ems

-0.13

Å¾en

-0.13

oren

-0.13

POSITIVE LOGITS

antha

0.14

eus

0.14

zym

0.13

NP

0.13

CTOR

0.13

 infra

0.13

jiang

0.13

ç¾½

0.13

Ð¾Ñĥ

0.13

anth

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.