NousResearch/Nous-Hermes-2-Yi-34B
Full fine-tune vs 01-ai/Yi-34B. Fine-tune: 100% of the weights changed, across 60 of 60 layers.
Tensors changed: Embeddings 1/1, Norms 80/121, MLP 180/180, Attention 240/240, lm_head 1/1
11 derivatives checked
- OpenGVLab/InternVL-Chat-V1-2: Full fine-tune (fewer layers)
- TheBloke/Nous-Hermes-2-Yi-34B-GGUF: Quantization
- mradermacher/Nous-Hermes-2-Yi-34B-i1-GGUF: Quantization
- qwp4w3hyb/Nous-Hermes-2-Yi-34B-iMat-GGUF: Quantization
- OpenGVLab/InternVL2-40B: Full fine-tune (fewer layers)
- OpenGVLab/InternVL-Chat-V1-2-Plus: Full fine-tune (fewer layers)
- tstepspam/Nous-Hermes-2-Yi-34B-Q4-MLX: Quantization
- tstepspam/Nous-Hermes-2-Yi-34B-Q8-MLX: Quantization
- tensorblock/Nous-Hermes-2-Yi-34B-GGUF: Quantization
- dfurman/HermesBagel-34B-v0.1: Full fine-tune
- Babsie/Hermes2Yi-34B-40k: Identical copy
1 other repos hold byte-for-byte identical weight files.