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

efs

-0.18

VERTISE

-0.16

iku

-0.15

-placeholder

-0.14

 Dank

-0.14

ReadStream

-0.14

agon

-0.14

iot

-0.13

rente

-0.13

ToDevice

-0.13

POSITIVE LOGITS

0.17

oval

0.15

Zam

0.15

ÑĪÐµÐ²

0.15

æļ

0.14

eyen

0.14

appen

0.14

 gate

0.14

POCH

0.14

pos

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.