axiong/pmc_oa
Updated
ā¢
653
ā¢
46
Adibvafa Fallahpour
adibvafa
AI & ML interests
Machine Learning, Healthcare, Computational Biology, Neuroscience, Synthetic Biology
Recent Activity
liked
a dataset
about 2 months ago
axiong/pmc_oa
liked
a model
about 2 months ago
google/bigbird-pegasus-large-pubmed
new activity
about 2 months ago
adibvafa/CodonTransformer:[bot] Conversion to Parquet
Organizations
adibvafa's activity
liked
a
dataset
about 2 months ago
liked
a
model
about 2 months ago
[bot] Conversion to Parquet
#1 opened 4 months ago
by
parquet-converter
liked
a
model
2 months ago
updated
a
model
2 months ago
liked
a
model
2 months ago
liked
3
models
3 months ago
reacted to
m-ric's
post with š„
3 months ago
Post
3382
š„ šš°šš§ š«šš„ššš¬šš¬ šš”šš¢š« š.š ššš¦š¢š„š² šØš š¦šØššš„š¬: ššš° šššš ššØš« šš„š„ š¬š¢š³šš¬ š®š© ššØ ššš!
The Chinese LLM maker just dropped a flurry of different models, ensuring there will be a Qwen SOTA model for every application out there:
Qwen2.5: 0.5B, 1.5B, 3B, 7B, 14B, 32B, and 72B
Qwen2.5-Coder: 1.5B, 7B, and 32B on the way
Qwen2.5-Math: 1.5B, 7B, and 72B.
And they didn't sleep: the performance is top of the game for each weight category!
ššš² š¢š§š¬š¢š š”šš¬:
š All models have šš®š“šø šš¼šøš²š» š°š¼š»šš²š š š¹š²š»š“ššµ
š Models pre-trained on 18T tokens, even longer than the 15T of Llama-3
šŖ The flagship š¤šš²š»š®.š±-š³š®š š¶š ~š°š¼šŗš½š²šš¶šš¶šš² šš¶ššµ šš¹š®šŗš®-šÆ.š-š°š¬š±š, š®š»š± šµš®š š® šÆ-š±% šŗš®šæš“š¶š» š¼š» šš¹š®šŗš®-šÆ.š-š³š¬š š¼š» šŗš¼šš šÆš²š»š°šµšŗš®šæšøš.
š«š· On top of this, it šš®šøš²š ššµš² #š šš½š¼š š¼š» šŗšš¹šš¶š¹š¶š»š“šš®š¹ šš®ššøš so it might become my standard for French
š» Qwen2.5-Coder is only 7B but beats competing models up to 33B (DeeSeek-Coder 33B-Instruct). Let's wait for their 32B to come out!
š§® Qwen2.5-Math sets a new high in the ratio of MATH benchmark score to # of parameters. They trained it by "aggregating more high-quality mathematical data, particularly in Chinese, from web sources, books, and codes across multiple recall cycles."
š Technical report to be released "very soon"
š All models have the most permissive license apache2.0, except the 72B models that have a custom license mentioning "you can use it for free EXCEPT if your product has over 100M users"
š¤ All models are available on the HF Hub! ā”ļø Qwen/qwen25-66e81a666513e518adb90d9e
The Chinese LLM maker just dropped a flurry of different models, ensuring there will be a Qwen SOTA model for every application out there:
Qwen2.5: 0.5B, 1.5B, 3B, 7B, 14B, 32B, and 72B
Qwen2.5-Coder: 1.5B, 7B, and 32B on the way
Qwen2.5-Math: 1.5B, 7B, and 72B.
And they didn't sleep: the performance is top of the game for each weight category!
ššš² š¢š§š¬š¢š š”šš¬:
š All models have šš®š“šø šš¼šøš²š» š°š¼š»šš²š š š¹š²š»š“ššµ
š Models pre-trained on 18T tokens, even longer than the 15T of Llama-3
šŖ The flagship š¤šš²š»š®.š±-š³š®š š¶š ~š°š¼šŗš½š²šš¶šš¶šš² šš¶ššµ šš¹š®šŗš®-šÆ.š-š°š¬š±š, š®š»š± šµš®š š® šÆ-š±% šŗš®šæš“š¶š» š¼š» šš¹š®šŗš®-šÆ.š-š³š¬š š¼š» šŗš¼šš šÆš²š»š°šµšŗš®šæšøš.
š«š· On top of this, it šš®šøš²š ššµš² #š šš½š¼š š¼š» šŗšš¹šš¶š¹š¶š»š“šš®š¹ šš®ššøš so it might become my standard for French
š» Qwen2.5-Coder is only 7B but beats competing models up to 33B (DeeSeek-Coder 33B-Instruct). Let's wait for their 32B to come out!
š§® Qwen2.5-Math sets a new high in the ratio of MATH benchmark score to # of parameters. They trained it by "aggregating more high-quality mathematical data, particularly in Chinese, from web sources, books, and codes across multiple recall cycles."
š Technical report to be released "very soon"
š All models have the most permissive license apache2.0, except the 72B models that have a custom license mentioning "you can use it for free EXCEPT if your product has over 100M users"
š¤ All models are available on the HF Hub! ā”ļø Qwen/qwen25-66e81a666513e518adb90d9e
- 2 replies
liked
a
model
3 months ago
[FEEDBACK] Daily Papers
102
#32 opened 7 months ago
by
kramp
updated
a
collection
3 months ago
updated
a
model
3 months ago
updated
a
dataset
3 months ago
updated
a
model
3 months ago
updated
a
collection
4 months ago
liked
a
dataset
4 months ago
upvoted
a
collection
4 months ago