INDEX

Explanations

references to plumbing and water heating equipment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 breathable

-0.15

 respir

-0.15

uja

-0.15

çĴĥ

-0.14

 dust

-0.14

 breeze

-0.14

iÄĻ

-0.14

 franchise

-0.14

 inheritance

-0.14

/loading

-0.14

POSITIVE LOGITS

hot

0.31

Hot

0.27

Hot

0.26

 boiler

0.25

-hot

0.25

 boilers

0.25

.hot

0.24

hot

0.24

 heating

0.24

 Boiler

0.24

Activations Density 0.040%

references to plumbing and water heating equipment

No Comments

No Known Activations