INDEX

Explanations

self-referential expressions indicating personal feelings and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 ÐĴÐ¸ÐºÐ¸

-0.15

Ãªnh

-0.15

okable

-0.15

grunt

-0.15

lamaya

-0.15

owitz

-0.15

ladÄ±ÄŁÄ±

-0.14

.scalablytyped

-0.14

 eldre

-0.14

ichick

-0.14

POSITIVE LOGITS

 absolutely

0.35

 love

0.33

 fell

0.30

 LOVE

0.29

ad

0.29

 falling

0.29

 fall

0.27

 Love

0.26

 loves

0.26

 Absolutely

0.26

Activations Density 0.140%

self-referential expressions indicating personal feelings and experiences

No Comments

No Known Activations