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

 Wilderness

-0.17

 commod

-0.15

Orn

-0.14

Ð¾Ñĥ

-0.14

æŃĮ

-0.14

 somehow

-0.13

itesse

-0.13

panion

-0.13

alon

-0.13

POSITIVE LOGITS

 liberalism

0.20

 Catholic

0.20

Mgr

0.18

 Liberals

0.16

 liberals

0.16

 exterior

0.16

 gratuit

0.16

 Thom

0.16

lib

0.15

 Catholics

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.