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

aps

-0.16

Å¥

-0.15

ernetes

-0.14

hton

-0.14

rens

-0.13

iest

-0.13

 erÃ³t

-0.13

ãģķ

-0.13

à¸Ļà¸Ń

-0.13

uro

-0.13

POSITIVE LOGITS

erre

0.14

Mob

0.14

ÙĨØ¬

0.13

 ÙģØ§Ø±Ø³

0.13

Sle

0.13

çĶ

0.13

elper

0.13

Bloc

0.13

 Pose

0.13

erm

0.13

Activations Density 0.123%

No Comments

No Known Activations