INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 darn

-0.14

edis

-0.14

eda

-0.14

 toto

-0.14

Others

-0.14

undry

-0.14

 reportedly

-0.13

ÑĨÐ²ÐµÑĤ

-0.13

 Others

-0.13

dit

-0.13

POSITIVE LOGITS

Dom

0.23

Dom

0.19

dom

0.19

Domains

0.18

 Produ

0.18

 kind

0.17

orca

0.17

-dom

0.17

Amy

0.17

.Dom

0.17

Activations Density 0.000%

No Known Activations

This feature has no known activations.