@dna_for_real_official:

DNA_FOR_REAL🇬🇭🇬🇭 🇬🇧🇬🇧
DNA_FOR_REAL🇬🇭🇬🇭 🇬🇧🇬🇧
Open In TikTok:
Region: GH
Saturday 03 October 2026 21:39:55 GMT
24092
2258
166
232

Music

Download

Comments

the.melody60
Favour Dordunu Junior :
The last part weak me. He said "wonsa ba no gu ho kwa" 😂😂😂😂😂😂😂😂😂😂😂😂😂
2026-10-04 00:47:36
2
murphyg_4597
Richard Murphy :
Sam George concert guy
2026-10-03 23:30:45
25
solomon_crain
Solomon :
The annoying thing is, even if you report, TikTok will say it doesn't violate their community guidelines
2026-10-03 22:48:05
50
richmondquartey
Richmond Quartey :
After arresting them. All pastors should be arrested too.
2026-10-03 23:58:29
19
kwakuquaku3
kwakuquaku :
But father Solomon's fingers are intact yet he wants the youth to come sacrifice theirs.
2026-10-03 23:50:44
6
user748115229
Nana Afriyie :
They need to be arrested
2026-10-03 22:50:02
30
nana.kofi2714
Nana kofi :
They won’t do anything about it but if you insult someone I don’t want to mention right now, they will look for you and arrest you
2026-10-04 01:00:28
2
ericus.b440
Ericus B :
We are reaping the results now Corrupt nation and mediocrity
2026-10-04 00:47:49
2
munadiamansuraadams
munadiamansuraadams :
My friend took his son out, bought him malt and the little boy rejected the malt saying he want brukutu🤣🤣🤣 when they asked him where he heard of brukutu he said on tv🤣 now he has dashed his tv 📺 Ghana eh
2026-10-03 21:59:56
13
oheneba283
Darkwah Benjamin :
Gotv and dstv don’t entertain such things
2026-10-04 00:48:53
2
sadik7059
KAM NI CHE 👍 :
IF NOT GHANA WHERE WILL YOU SEE THIS KIND OF NONSENSE
2026-10-03 21:49:52
18
kharicare_beauty4
The Conneys :
In fact about this things err hm I'm really disappointed in Sam Geoge paa becos he's done nothing about this awful thing as he promise Ghanaians hm
2026-10-03 23:37:32
5
abelectricals90
A.B Electricals 90 :
if the police make arrest now, we are the same people that will be blaming the government, such foolish act on camera and the authorities are silent 😭💔
2026-10-03 21:59:48
12
hermajesty_11
Yourlady_surveyor👷🏻‍♀️✅ :
Kwasias3m sei aaa hmmm
2026-10-03 21:50:23
15
charlesboachie
charlesboachie :
my brother I was shocked I just saw it yesterday I even shock
2026-10-03 22:53:35
8
officialgirlboss1
G & G VENTURES :
Hmmmm bro this is getting serious oooo
2026-10-03 22:01:13
7
peterfrimpong68
the rock68 :
True talk bro
2026-10-03 23:46:32
3
mamalu8362
love :
hmmm
2026-10-04 00:49:41
1
smartupmetal_1
SMART UP METAL WORKS :
unless they insults the leaders then they will take action right now hmm Ghana paa
2026-10-03 22:58:39
10
florence.golo0
Florence Golo :
eiiiii
2026-10-03 23:48:51
3
hdj.4hjf
hajia heidaya🥳💝💕 :
mmmmmmm
2026-10-03 23:48:45
3
serwah.adepa
Serwah Adepa :
Hmmmm
2026-10-03 23:35:51
3
themaskironheart_the335
IRONHEART 🦦🏹🌓🗣️𝗧ᴴᴱ335 :
Eiiii Ghana
2026-10-04 00:35:54
3
najatibrahim91
najatibrahim91 :
Yes
2026-10-04 00:06:38
3
charon_beauty_empire
charon_beauty_empire :
Hmmm
2026-10-03 23:59:48
3
To see more videos from user @dna_for_real_official, please go to the Tikwm homepage.

Other Videos

Ai2 just open-sourced AstaBrief 8B, a small model that turns a research question into a cited report. You give it a question plus excerpts from retrieved papers, and it writes the whole report with citations in one pass. It's what powers the new Fast mode in Asta, Ai2's platform for science, next to the Claude-powered Thinking mode. The speed is the point: 51.1 seconds per report in Fast mode versus 178.5 seconds in Thinking mode, about 3.5x faster across the full Asta pipeline. Quality holds up in Ai2's own evals. On SQABench-CS2 it averages 87.0, between DR Tulu at 88.8 and Thinking mode at 86.2, and 90.5% of its citations support their claim. It wins 72% of LLM-judged comparisons against Thinking mode, though it was trained toward that exact ranking. The interesting part is the data: 90K real research questions, 47K reports from Claude, o3 and GPT-4.1 models, and 6K preference pairs kept only when two judges agreed. The biggest single gain came from throwing out training reports that didn't cite enough. The fine print: in a small human study, researchers preferred DR Tulu overall, and most of the evaluation was done in 2025 against that year's frontier models. It's Apache 2.0, built on Qwen3-8B, and you can run it on your own hardware, which matters for unpublished research. Sources: allenai.org/blog/astabrief and huggingface.co/allenai/AstaBrief_8B Independent briefing by CodenamePoshan, not an Ai2 post. Ai2 art and logo shown for editorial reporting. #Ai2 #AstaBrief #OpenSourceAI #AIResearch #LLM #AIEngineering
Ai2 just open-sourced AstaBrief 8B, a small model that turns a research question into a cited report. You give it a question plus excerpts from retrieved papers, and it writes the whole report with citations in one pass. It's what powers the new Fast mode in Asta, Ai2's platform for science, next to the Claude-powered Thinking mode. The speed is the point: 51.1 seconds per report in Fast mode versus 178.5 seconds in Thinking mode, about 3.5x faster across the full Asta pipeline. Quality holds up in Ai2's own evals. On SQABench-CS2 it averages 87.0, between DR Tulu at 88.8 and Thinking mode at 86.2, and 90.5% of its citations support their claim. It wins 72% of LLM-judged comparisons against Thinking mode, though it was trained toward that exact ranking. The interesting part is the data: 90K real research questions, 47K reports from Claude, o3 and GPT-4.1 models, and 6K preference pairs kept only when two judges agreed. The biggest single gain came from throwing out training reports that didn't cite enough. The fine print: in a small human study, researchers preferred DR Tulu overall, and most of the evaluation was done in 2025 against that year's frontier models. It's Apache 2.0, built on Qwen3-8B, and you can run it on your own hardware, which matters for unpublished research. Sources: allenai.org/blog/astabrief and huggingface.co/allenai/AstaBrief_8B Independent briefing by CodenamePoshan, not an Ai2 post. Ai2 art and logo shown for editorial reporting. #Ai2 #AstaBrief #OpenSourceAI #AIResearch #LLM #AIEngineering

About