dphn/Dolphin3.0-Qwen2.5-1.5B
Full fine-tune vs Qwen/Qwen2.5-1.5B. Fine-tune: 100% of the weights changed, across 28 of 28 layers.
Tensors changed: Embeddings 1/1, Norms 57/57, MLP 84/84, Attention 196/196
10 derivatives checked
- bartowski/Dolphin3.0-Qwen2.5-1.5B-GGUF: Same weights, converted
- Dennisthemennis69/Dolphin3.0-Qwen2.5-1.5B-GGUF: Same weights, converted
- itlwas/Dolphin3.0-Qwen2.5-1.5B-Q4_K_M-GGUF: Quantization
- Sonorix/pulsar-coder-v1: Identical copy
- mradermacher/Dolphin3.0-Qwen2.5-1.5B-i1-GGUF: Quantization
- mradermacher/Dolphin3.0-Qwen2.5-1.5B-GGUF: Same weights, converted
- Sonorix/pulsar-coder-v2.2-gguf: Full fine-tune
- RCDWealth/Dolphin3.0-Qwen2.5-1.5B-Q5_K_M.gguf: Quantization
- Sonorix/pulsar-coder-v1-gguf: Quantization
- Da-master/Dolphin3.0-Qwen2.5-1.5B-Q4_K_M-GGUF: Quantization
1 other repos hold byte-for-byte identical weight files.