INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

.BL

-0.15

strap

-0.14

èĩ

-0.14

 ErrorResponse

-0.13

 guys

-0.13

linger

-0.13

LETED

-0.13

MPU

-0.12

.jobs

-0.12

 Crescent

-0.12

POSITIVE LOGITS

hoff

0.15

abilities

0.15

ahr

0.14

anders

0.14

Äĥn

0.13

esel

0.13

gne

0.13

omba

0.13

azi

0.13

altung

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.