INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

etc

-0.18

etc

-0.17

OK

-0.15

OK

-0.14

ULE

-0.14

viz

-0.14

ok

-0.13

isz

-0.13

umps

-0.13

JECTED

-0.13

POSITIVE LOGITS

\-

0.15

Ø¨ÙĪ

0.14

-↵

0.14

éĿ©

0.14

_(

0.14

 herself

0.14

ritch

0.13

.getSeconds

0.13

mage

0.13

ãĤįãģĨ

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.