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

ighthouse

-0.16

ĨĴ

-0.15

FOUNDATION

-0.14

åĩºåĵģ

-0.14

riz

-0.14

_shape

-0.14

IMIT

-0.14

ateral

-0.13

ERM

-0.13

âĢİ

-0.13

POSITIVE LOGITS

-pill

0.16

ifle

0.15

ousing

0.15

uario

0.15

 Bars

0.15

 pill

0.15

ahlen

0.14

 sept

0.14

igan

0.14

det

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.