@Mechramc@MuraiLabsHQ it's a general English LM 1B parametr sparse MoE trained on a mix of FineWeb-Edu,DCLM,code,and math.Built completely from scratch (architecture, data pipeline,training infra)Tamil support would be an interesting direction for post-training ! Will share results once eval is done.
@Mechramc@MuraiLabsHQ hi sir we too doing pretraining for our 1 billion llm from scratch using 100 billion dataset total 95367 steps we are currently at 88040 running 4 h200 total run time 68 hrs
@SwiggyCares@Swiggy u guys have one of the worst customer support i ordered food worth 650 rs my entire order is wrongly delivered i complained they say I can give 300 rs refund I told I don't want the wrong food ask the delivery boy to collect they say we can't do anything
If you don’t have any money to buy this dip?
I will help a few of you.
$5000 giveaway to 5 people.
To join:
Just retweet & make sure you are
following me.
I will actually announce the winners soon.
Unlike these other fake guru giveaways.
@JioCare@reliancejio you guys have the one of the worst customer service i can't even talk to ur customer care my net is not working for past 1 day no solution till now and no service engineer visited
@havellsindia u have the worst customer service i bought the lloyd ac delivered two days back with the guarantee of installation within 2 days but no one contacted me on installation till now , I contact customer support they tell me to wait 24-48 hrs again I spoken to them twice