INDEX

Explanations

expressions related to communication and knowledge

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

stile

-0.17

 çĭ

-0.16

Ð¡Ðŀ

-0.15

 Ð²Ð¸Ð±

-0.14

utz

-0.14

jur

-0.14

inker

-0.13

quam

-0.13

anas

-0.13

asa

-0.13

POSITIVE LOGITS

osing

0.19

0.17

 what

0.16

arlo

0.16

 otherwise

0.14

 ******************************************************************************/↵

0.14

ainers

0.14

rou

0.14

ulers

0.14

 ello

0.14

Activations Density 0.192%

expressions related to communication and knowledge

No Comments

No Known Activations