INDEX

Explanations

references to comfort and convenience in various contexts

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

ovah

-0.19

hung

-0.18

.scalablytyped

-0.17

icap

-0.17

Åº

-0.17

eger

-0.16

hangi

-0.16

ched

-0.16

crement

-0.15

ombat

-0.15

POSITIVE LOGITS

able

0.32

 zone

0.32

ably

0.32

ingly

0.30

 zones

0.29

ability

0.28

ab

0.27

 Zone

0.24

abl

0.24

-zone

0.24

Activations Density 0.020%

references to comfort and convenience in various contexts

No Comments

No Known Activations