INDEX

Explanations

phrases and terms indicating high quality or excellence

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

yr

-0.18

 wire

-0.16

eming

-0.16

alance

-0.15

 Wire

-0.14

Pey

-0.14

 Phys

-0.14

Wire

-0.14

tl

-0.14

 Wiring

-0.14

POSITIVE LOGITS

-end

0.31

end

0.24

 definition

0.24

 grade

0.23

Definition

0.22

 tech

0.22

 Definition

0.21

-performance

0.21

definition

0.21

end

0.21

Activations Density 0.036%

phrases and terms indicating high quality or excellence

No Comments

No Known Activations