Hugging Face's logo

Spaces:
bmi-labmedinfo
/
README
Running

App Files Community
Detsutut commited on
Commit
7c1646d
·
verified ·
1 Parent(s): 39bf2d4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -40,7 +40,7 @@ Here we host public weights for our biomedical language models. There are severa
40
 
41
  | Model | Domain | Type | Details |
42
  |------------|---------|-------------------|-------------------------------------------------------------|
43
- | [Igea](https://huggingface.co/Detsutut/Igea-1B-v0.0.1) | Biomedical | CausalLM Pretrain | Small language model trained after [sapienzanlp/Minerva-1B-base-v1.0](https://huggingface.co/sapienzanlp/Minerva-1B-base-v1.0) with more than 5 billion biomedical words in Italian. Three versions available: [350M params](https://huggingface.co/bmi-labmedinfo/Igea-350M-v0.0.1), [1B params](https://huggingface.co/bmi-labmedinfo/Igea-1B-v0.0.1), and [3B params](https://huggingface.co/bmi-labmedinfo/Igea-3B-v0.0.1). Use the [quantized GGUF version](Detsutut/Igea-1B-v0.0.1-Q4_K_M-GGUF) for CPU-only, limited-hardware machines. |
44
  | [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) <sup>*</sup>| Biomedical | MaskedLM Pretrain | BERT model trained after [dbmdz/bert-base-italian-xxl-cased](https://huggingface.co/dbmdz/bert-base-italian-xxl-cased) with 28GB Pubmed abstracts (as in BioBERT) that have been translated from English into Italian using Neural Machine Translation (GNMT). |
45
  | [MedBIT](https://huggingface.co/bmi-labmedinfo/medBIT) <sup>*</sup>| Medical | MaskedLM Pretrain | BERT model trained after [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) with additional 100MB of medical textbook data without any regularization. |
46
  | [MedBIT-R3+](https://huggingface.co/bmi-labmedinfo/medBIT-r3-plus) (recommended) <sup>*</sup>| Medical | MaskedLM Pretrain | BERT model trained after [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) with additional 200MB of medical textbook data and web-crawled medical resources in Italian. Regularized with LLRD (.95), Mixout (.9), and Warmup (.02). |
 
40
 
41
  | Model | Domain | Type | Details |
42
  |------------|---------|-------------------|-------------------------------------------------------------|
43
+ | [Igea](https://huggingface.co/Detsutut/Igea-1B-v0.0.1) | Biomedical | CausalLM Pretrain | Small language model trained after [sapienzanlp/Minerva](https://huggingface.co/sapienzanlp/Minerva-1B-base-v1.0) with more than 5 billion biomedical words in Italian. Three versions available: [350M params](https://huggingface.co/bmi-labmedinfo/Igea-350M-v0.0.1), [1B params](https://huggingface.co/bmi-labmedinfo/Igea-1B-v0.0.1), and [3B params](https://huggingface.co/bmi-labmedinfo/Igea-3B-v0.0.1). Use the [quantized GGUF version](Detsutut/Igea-1B-v0.0.1-Q4_K_M-GGUF) for CPU-only, limited-hardware machines. |
44
  | [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) <sup>*</sup>| Biomedical | MaskedLM Pretrain | BERT model trained after [dbmdz/bert-base-italian-xxl-cased](https://huggingface.co/dbmdz/bert-base-italian-xxl-cased) with 28GB Pubmed abstracts (as in BioBERT) that have been translated from English into Italian using Neural Machine Translation (GNMT). |
45
  | [MedBIT](https://huggingface.co/bmi-labmedinfo/medBIT) <sup>*</sup>| Medical | MaskedLM Pretrain | BERT model trained after [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) with additional 100MB of medical textbook data without any regularization. |
46
  | [MedBIT-R3+](https://huggingface.co/bmi-labmedinfo/medBIT-r3-plus) (recommended) <sup>*</sup>| Medical | MaskedLM Pretrain | BERT model trained after [BioBIT](https://huggingface.co/bmi-labmedinfo/bioBIT) with additional 200MB of medical textbook data and web-crawled medical resources in Italian. Regularized with LLRD (.95), Mixout (.9), and Warmup (.02). |