INDEX

Explanations

references to evening routines and activities related to winding down after the day

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Lunch

-0.28

 mornings

-0.26

 Morning

-0.26

 lunches

-0.25

 brunch

-0.25

 noon

-0.25

 breakfast

-0.24

 sunny

-0.24

 lunch

-0.23

Morning

-0.23

POSITIVE LOGITS

 tonight

0.53

 Tonight

0.48

 evening

0.47

Tonight

0.47

 night

0.46

æĻļ

0.46

 nightly

0.45

å¤ľ

0.43

 noct

0.43

 nighttime

0.42

Activations Density 0.237%

references to evening routines and activities related to winding down after the day

No Comments

No Known Activations