Downloads · 30 days
210
12% of all-time downloads
silma-ai_-_SILMA-9B-Instruct-v0.1.1-gguf
RichardErkhov/silma-ai_-_SILMA-9B-Instruct-v0.1.1-gguf
silma-ai_-_SILMA-9B-Instruct-v0.1.1-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
All-time downloads
1.7K
Public
Repo size
107 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf125 GB · 100%
At a glance
- Access
- Public
- Created
- Aug 3, 2024
- Updated
- Aug 3, 2024
- SHA
- 7629a43a
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
SILMA-9B-Instruct-v0.1.1 - GGUF
- Model creator: https://huggingface.co/silma-ai/
- Original model: https://huggingface.co/silma-ai/SILMA-9B-Instruct-v0.1.1/
Original model description:
license: gemma
SILMA is an LLM focusing on Arabic Language. This is our initial version.
Stay tunnned!
Initial Results
| Task | Version | Metric | Value | Stderr | |
|---|---|---|---|---|---|
| all | acc_norm | 0.6491 | ± | 0.0180 | |
| community:alghafa:_average:0 | acc_norm | 0.6491 | ± | 0.0180 | |
| community:alghafa:mcq_exams_test_ar:0 | 0 | acc_norm | 0.4399 | ± | 0.0211 |
| community:alghafa:meta_ar_dialects:0 | 0 | acc_norm | 0.5399 | ± | 0.0068 |
| community:alghafa:meta_ar_msa:0 | 0 | acc_norm | 0.6547 | ± | 0.0159 |
| community:alghafa:multiple_choice_facts_truefalse_balanced_task:0 | 0 | acc_norm | 0.9600 | ± | 0.0228 |
| community:alghafa:multiple_choice_grounded_statement_soqal_task:0 | 0 | acc_norm | 0.7600 | ± | 0.0350 |
| community:alghafa:multiple_choice_grounded_statement_xglue_mlqa_task:0 | 0 | acc_norm | 0.6333 | ± | 0.0395 |
| community:alghafa:multiple_choice_rating_sentiment_no_neutral_task:0 | 0 | acc_norm | 0.9004 | ± | 0.0033 |
| community:alghafa:multiple_choice_rating_sentiment_task:0 | 0 | acc_norm | 0.6003 | ± | 0.0063 |
| community:alghafa:multiple_choice_sentiment_task:0 | 0 | acc_norm | 0.3529 | ± | 0.0115 |
- Created
- Aug 3, 2024
- Updated
- Aug 3, 2024