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

 :↵

-0.15

ï¼īï¼ļ

-0.15

:↵

-0.14

dna

-0.14

Aqu

-0.14

__$

-0.14

ÑĭÑĪ

-0.14

Digit

-0.14

æ®Ĭ

-0.14

ullah

-0.14

POSITIVE LOGITS

.localized

0.17

omatic

0.16

printf

0.15

.FindControl

0.15

asher

0.15

 Locker

0.15

aso

0.14

 Peter

0.14

rello

0.14

Peter

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.