INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

olist

-0.18

olf

-0.15

Ð»Ð¾Ñĩ

-0.15

eft

-0.14

ODY

-0.14

imits

-0.14

obao

-0.14

Î»ÎµÎ¯

-0.14

lew

-0.14

.fn

-0.14

POSITIVE LOGITS

 SOURCE

0.17

SOURCE

0.16

via

0.15

ayah

0.15

 accordingly

0.15

dna

0.14

 LIGHT

0.14

GLOBAL

0.14

DNA

0.14

yne

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.