INDEX

Explanations

references to "air" and "air conditioning."

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

ecute

-0.16

odi

-0.16

aires

-0.16

_msgs

-0.15

çİĩ

-0.15

ipher

-0.15

essen

-0.15

zbek

-0.14

Overflow

-0.14

å¹ħ

-0.14

POSITIVE LOGITS

condition

0.32

waves

0.30

 conditioning

0.30

-condition

0.29

bnb

0.28

brush

0.27

 conditioner

0.26

 Conditioning

0.26

speed

0.26

soft

0.24

Activations Density 0.035%

references to "air" and "air conditioning."

No Comments

No Known Activations