bespokelabs/Bespoke-Stratos-7B
Full fine-tune vs Qwen/Qwen2.5-7B-Instruct. Fine-tune: 93% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 0/1, Norms 53/57, MLP 84/84, Attention 196/196, lm_head 1/1
4 derivatives checked
- mradermacher/Bespoke-Stratos-7B-i1-GGUF: Quantization
- mradermacher/Bespoke-Stratos-7B-GGUF: Same weights, converted
- bunnycore/Qwen-2.5-7B-Stock-Deep-Bespoke: Full fine-tune
- bunnycore/Qwen2.5-7B-R1-Bespoke-Stock: Full fine-tune