INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 responsiveness

-0.18

Ã¡nÃŃm

-0.15

ERM

-0.15

rets

-0.15

 Price

-0.15

ssp

-0.14

 pricing

-0.14

 Bias

-0.14

ries

-0.14

POSITIVE LOGITS

 chances

0.27

 overall

0.22

 odds

0.21

 productivity

0.20

overall

0.20

 profits

0.19

 PROFITS

0.18

 chance

0.18

 bang

0.18

 convenience

0.17

Activations Density 0.129%

No Comments

No Known Activations