AnkitAI/Parable-Qwen3-4B-Claude-Fable-5
LoRA-style vs Qwen/Qwen3-4B. 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 36 of 36 layers; norms, biases and embeddings are untouched.
Tensors changed: Embeddings 0/1, Norms 0/145, MLP 108/108, Attention 144/144
6 derivatives checked
- mradermacher/Parable-Qwen3-4B-fable-i1-GGUF: Full fine-tune
- mradermacher/Parable-Qwen3-4B-Claude-Fable-5-i1-GGUF: Full fine-tune
- mradermacher/Parable-Qwen3-4B-Claude-Fable-5-GGUF: Full fine-tune
- mradermacher/Parable-Qwen3-4B-fable-GGUF: Full fine-tune
- AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-MLX-8bit: Quantization
- AnkitAI/Parable-Qwen3-4B-Claude-Fable-5-MLX-4bit: Quantization