YAML Metadata Warning: empty or missing yaml metadata in repo card (https://huggingface.co/docs/hub/model-cards#model-card-metadata)

MoE model based on Ling V2 MoE arch, with 32 experts, 2 activated, no shared experts, no dense layers, trained on about 100M tokens of FineWeb-2 pol-Latn split, using tokenizer taken from EuroLLM 1.7B

Just a test to validate pipeline before training bigger 4B 256 expert model on 100B tokens.

Downloads last month: 41

Safetensors

Model size

834M params

Tensor type

F32

BF16

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support