Multilingual Web datasets
AI & ML interests
Open Source Language Models for Europe
First release of 7B LLMs models for the 5 biggest European languages. All models initialised from mistral-7b-v0.1.
-
occiglot/occiglot-7b-eu5
Text Generation β’ 7B β’ Updated β’ 81 β’ 27 -
occiglot/occiglot-7b-eu5-instruct
Text Generation β’ 7B β’ Updated β’ 102 β’ 10 -
occiglot/occiglot-7b-es-en
Text Generation β’ 7B β’ Updated β’ 16 β’ 4 -
occiglot/occiglot-7b-es-en-instruct
Text Generation β’ 7B β’ Updated β’ 41 β’ 2
Multilingual Web datasets
First release of 7B LLMs models for the 5 biggest European languages. All models initialised from mistral-7b-v0.1.
-
occiglot/occiglot-7b-eu5
Text Generation β’ 7B β’ Updated β’ 81 β’ 27 -
occiglot/occiglot-7b-eu5-instruct
Text Generation β’ 7B β’ Updated β’ 102 β’ 10 -
occiglot/occiglot-7b-es-en
Text Generation β’ 7B β’ Updated β’ 16 β’ 4 -
occiglot/occiglot-7b-es-en-instruct
Text Generation β’ 7B β’ Updated β’ 41 β’ 2
Quantised checkpoints of occiglot models provided by mayflower