INDEX

Explanations

references to educators and incidents involving students

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ivent

-0.16

cket

-0.15

ivor

-0.14

rael

-0.14

sian

-0.14

ngo

-0.14

andalone

-0.13

 Offline

-0.13

indre

-0.13

tsy

-0.13

POSITIVE LOGITS

 allegedly

0.19

æ¶ī

0.17

 GenÃ§

0.17

/popper

0.14

Popular

0.14

 supposedly

0.14

 Popular

0.14

tele

0.14

à¸¥à¸¥

0.14

nÄĽn

0.14

Activations Density 0.237%

references to educators and incidents involving students

No Comments

No Known Activations