INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

encent

-0.16

uiltin

-0.15

entiful

-0.15

-0.14

ris

-0.14

 Mits

-0.14

alez

-0.14

Î¬Î²

-0.14

iglia

-0.14

 suspension

-0.13

POSITIVE LOGITS

/dev

0.16

/background

0.15

eof

0.14

 ìĦ¬

0.14

PIX

0.14

eur

0.14

aar

0.14

ÑģÑĤÐµ

0.13

.Pixel

0.13

Pixel

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.