INDEX

Explanations

references to time and events occurring regularly or repetitively

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 winters

-0.25

 decades

-0.19

 summers

-0.18

 YEARS

-0.17

 Seasons

-0.17

 Years

-0.17

Years

-0.16

apat

-0.16

strup

-0.15

 centuries

-0.15

POSITIVE LOGITS

 calendar

0.26

 Calendar

0.21

 throughout

0.20

calendar

0.20

 year

0.19

ear

0.19

 Tear

0.19

-calendar

0.18

ear

0.18

Activations Density 0.100%

references to time and events occurring regularly or repetitively

No Comments

No Known Activations

references to time and events occurring regularly or repetitively

No Comments

No Known Activations