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

Formatting

-0.14

Truthy

-0.14

ivor

-0.14

 PROGMEM

-0.14

$__

-0.13

é¾

-0.13

alsy

-0.13

¥¤

-0.13

eway

-0.13

á»¡

-0.13

POSITIVE LOGITS

 rest

0.20

 Rest

0.20

Rest

0.19

obo

0.17

um

0.17

jer

0.16

rest

0.16

.jpa

0.15

/rest

0.15

oms

0.15

Activations Density 0.000%

No Known Activations

This feature has no known activations.