INDEX

Explanations

phrases related to routines and frequency in activities

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

 annual

-0.36

 Annual

-0.32

 annually

-0.31

annual

-0.30

 yearly

-0.28

Annual

-0.28

 quarterly

-0.25

 Quarterly

-0.23

 monthly

-0.19

 daily

-0.18

POSITIVE LOGITS

 dose

0.29

/month

0.28

 grind

0.24

 doses

0.23

 basis

0.23

mente

0.20

 dosage

0.20

 Basis

0.18

basis

0.17

ities

0.17

Activations Density 0.040%

phrases related to routines and frequency in activities

No Comments

No Known Activations