INDEX

Explanations

phrases indicating a list or collection of items or ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

è¿Ļä¸ª

-0.18

è¿Ļä¸Ģ

-0.16

hits

-0.16

anou

-0.15

HR

-0.15

bers

-0.15

annes

-0.14

ä¸Ģæ¬¡

-0.14

mers

-0.14

 dieses

-0.14

POSITIVE LOGITS

 some

0.35

two

0.24

some

0.24

 several

0.24

äºĽ

0.23

 Some

0.22

ä¸ĢäºĽ

0.21

 three

0.21

 beberapa

0.20

few

0.20

Activations Density 0.024%

phrases indicating a list or collection of items or ideas

No Comments

No Known Activations

phrases indicating a list or collection of items or ideas

No Comments

No Known Activations