INDEX

Explanations

words or phrases related to performance and accomplishments

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

ÅĻenÃŃ

-0.16

 Nose

-0.15

mall

-0.15

AA

-0.14

auty

-0.14

ë¦Ħ

-0.14

opies

-0.14

dil

-0.14

uxt

-0.14

 drains

-0.13

POSITIVE LOGITS

undo

0.18

etti

0.17

etten

0.16

wnd

0.15

·æĸ°

0.15

unga

0.15

_simps

0.15

unw

0.15

apat

0.15

mina

0.14

Activations Density 0.019%

words or phrases related to performance and accomplishments

No Comments

No Known Activations