INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

iesel

-0.16

ghan

-0.16

overflow

-0.15

 Vern

-0.15

awner

-0.15

ownt

-0.15

askell

-0.14

Ã¤nn

-0.14

ockey

-0.14

turnstile

-0.14

POSITIVE LOGITS

à¸Ńà¸°

0.15

 Vest

0.14

iant

0.14

itness

0.14

icone

0.14

allon

0.14

 Hell

0.13

loys

0.13

iyim

0.13

xmm

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.