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

ettes

-0.15

icode

-0.15

odash

-0.15

pps

-0.15

ence

-0.14

ialog

-0.13

"..

-0.13

umbing

-0.13

oya

-0.13

POSITIVE LOGITS

ecome

0.15

 ideas

0.15

completion

0.14

 continual

0.14

asily

0.14

 completion

0.14

 idea

0.14

alom

0.14

utan

0.14

 complete

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.