INDEX

Explanations

phrases related to standing out or distinguishing oneself

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

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

innen

-0.15

arrow

-0.15

thr

-0.14

tober

-0.13

Ye

-0.13

pez

-0.13

 proximity

-0.13

 deter

-0.13

ko

-0.13

POSITIVE LOGITS

 stand

0.46

 stands

0.41

 Stand

0.40

 stood

0.39

stands

0.37

stood

0.35

stand

0.35

Stand

0.34

 above

0.31

above

0.29

Activations Density 0.055%

phrases related to standing out or distinguishing oneself

No Comments

No Known Activations