arcee-ai/Arcee-SuperNova-v1
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, up_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 0/320, lm_head 0/1
10 derivatives checked
- bartowski/arcee-ai_Arcee-SuperNova-v1-GGUF: Quantization
- mradermacher/Arcee-SuperNova-v1-i1-GGUF: Quantization
- arcee-ai/Arcee-SuperNova-v1-GGUF: Quantization
- DevQuasar/arcee-ai.Arcee-SuperNova-v1-GGUF: Same weights, converted
- mradermacher/Arcee-SuperNova-v1-GGUF: Quantization
- sophosympatheia/Strawberrylemonade-L3-70B-v1.1: Full fine-tune
- sophosympatheia/Strawberrylemonade-L3-70B-v1.2: Full fine-tune
- sh0ck0r/Strawberrylemonade-L3-70B-v1.2-heretic3: Full fine-tune
- sh0ck0r/Strawberrylemonade-L3-70B-v1.2-heretic2: Full fine-tune
- sh0ck0r/Strawberrylemonade-L3-70B-v1.2-heretic: Full fine-tune