euclaise/Memphis-CoT-3B
Full fine-tune vs stabilityai/stablelm-3b-4e1t. Fine-tune: 91% of the weights changed, across 32 of 32 layers.
Tensors changed: lm_head 0/1, Embeddings 0/1, Norms 130/130, MLP 96/96, Attention 128/128
4 derivatives checked
- mradermacher/Memphis-CoT-3B-i1-GGUF: Quantization
- mradermacher/Memphis-CoT-3B-GGUF: Same weights, converted
- brittlewis12/Memphis-CoT-3B-GGUF: Same weights, converted
- euclaise/Memphis-scribe-3B: Partial fine-tune