INDEX

Explanations

phrases related to significant endorsements or support 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 @ 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

/the

-0.19

the

-0.15

innen

-0.15

let

-0.14

[]

-0.14

lement

-0.13

The

-0.13

/The

-0.13

any

-0.13

â

-0.13

POSITIVE LOGITS

 same

0.32

own

0.28

 latest

0.27

 entire

0.26

latest

0.22

 second

0.22

same

0.21

 ability

0.20

.same

0.19

 SAME

0.18

Activations Density 0.914%

phrases related to significant endorsements or support in various contexts

No Comments

No Known Activations