Artples/L-MChat-7b

Full fine-tune vs FuseAI/FuseChat-7B-VaRM. Fine-tune: 97% of the weights changed, across 32 of 32 layers.

Tensors changed: lm_head 1/1, Embeddings 1/1, Norms 65/65, MLP 93/96, Attention 124/128

2 derivatives checked

• mradermacher/L-MChat-7b-GGUF: Same weights, converted
• mradermacher/L-MChat-7b-i1-GGUF: Quantization

On the Hub · Downloads on Model Pulse