INDEX

Explanations

connections between people and groups in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 himself

-0.21

ä¸¡

-0.21

his

-0.16

ãĥ³ãĥķ

-0.16

his

-0.16

Æ°á»Ľ

-0.15

ucz

-0.15

 jeho

-0.15

 ÐµÐ³Ð¾

-0.15

Both

-0.14

POSITIVE LOGITS

 themselves

0.37

 collective

0.23

 nhau

0.22

 unanimously

0.22

 collectively

0.22

 yourselves

0.21

 hepsi

0.19

 thems

0.19

anson

0.18

 leurs

0.18

Activations Density 0.743%

connections between people and groups in various contexts

No Comments

No Known Activations

connections between people and groups in various contexts

No Comments

No Known Activations