INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

ssi

-0.15

chu

-0.14

ç¿Ķ

-0.13

chair

-0.13

Ð¸Ð²Ð°

-0.13

Symfony

-0.13

verity

-0.13

boa

-0.13

acom

-0.13

 worst

-0.13

POSITIVE LOGITS

0.16

rouch

0.15

oth

0.15

.Misc

0.14

URT

0.13

 Shepherd

0.13

 frags

0.13

 (!((

0.13

aque

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.