MaziyarPanahi/calme-2.3-llama3.1-70b
LoRA-style vs meta-llama/Llama-3.1-70B-Instruct. Looks like a merged LoRA (or a fine-tune with frozen norms): only the weight matrices of down_proj, gate_proj, k_proj, o_proj, q_proj, up_proj, v_proj changed, in 80 of 80 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/161, MLP 240/240, Attention 320/320, lm_head 0/1
8 derivatives checked
- mradermacher/calme-2.3-llama3.1-70b-i1-GGUF: Quantization
- mradermacher/calme-2.3-llama3.1-70b-GGUF: Quantization
- featherless-ai-quants/MaziyarPanahi-calme-2.3-llama3.1-70b-GGUF: Quantization
- NexesMess/Llama_3.x_70b_Tessalia_0.10c: Full fine-tune
- NexesMess/Llama_3.1_70b_Prologue_V1.11: Full fine-tune
- Nexesenex/Llama_3.x_70b_Calme-L3.3_abliterated_fusion_v2: Partial fine-tune
- Nexesenex/Llama_3.x_70b_L3.3-calme-2.3_abliterated_fusion: Full fine-tune
- Nexesenex/Llama_3.x_70b_Athos_v1.0: Full fine-tune