CultriX/Qwen2.5-14B-Wernicke
Full fine-tune vs EVA-UNIT-01/EVA-Qwen2.5-14B-v0.0. Fine-tune: 100% of the weights changed, across 48 of 48 layers.
Tensors changed: Embeddings 1/1, Norms 97/97, MLP 144/144, Attention 336/336, lm_head 1/1
11 derivatives checked
- mradermacher/Qwen2.5-14B-Wernicke-i1-GGUF: Quantization
- featherless-ai-quants/CultriX-Qwen2.5-14B-Wernicke-GGUF: Quantization
- mradermacher/Qwen2.5-14B-Wernicke-GGUF: Quantization
- sometimesanotion/Lamarck-14B-v0.3: Full fine-tune
- CultriX/SeQwence-14Bv1: Full fine-tune
- CultriX/SeQwence-14B-EvolMerge: Full fine-tune
- CultriX/Qwen2.5-14B-Wernicke-SFT: LoRA-style
- nlpguy/Lion-Lamarck-v.1.1.0: Full fine-tune
- Aashraf995/QwenStock-14B: Full fine-tune
- CultriX/Qwen2.5-14B-Wernickev3: Full fine-tune
- CultriX/Qwen2.5-14B-Wernicke-DPO: LoRA-style