Telugu-LLM-Labs/Indic-gemma-2b-finetuned-sft-Navarasa-2.0
LoRA-style vs google/gemma-2b. 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 18 of 18 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/37, MLP 54/54, Attention 72/72
3 derivatives checked
- mradermacher/Indic-gemma-2b-finetuned-sft-Navarasa-2.0-i1-GGUF: Quantization
- mradermacher/Indic-gemma-2b-finetuned-sft-Navarasa-2.0-GGUF: Same weights, converted
- featherless-ai-quants/Telugu-LLM-Labs-Indic-gemma-2b-finetuned-sft-Navarasa-2.0-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.