ibm-granite/granite-3.0-2b-instruct
Full fine-tune vs ibm-granite/granite-3.0-2b-base. Fine-tune: 100% of the weights changed, across 40 of 40 layers.
Tensors changed: Norms 81/81, MLP 120/120, Attention 160/160
9 derivatives checked
- bartowski/granite-3.0-2b-instruct-GGUF: Same weights, converted
- MaziyarPanahi/granite-3.0-2b-instruct-GGUF: Same weights, converted
- mradermacher/granite-3.0-2b-instruct-i1-GGUF: Quantization
- mradermacher/granite-3.0-2b-instruct-GGUF: Same weights, converted
- MCZK/granite-3.0-2b-instruct-GGUF: Same weights, converted
- lmstudio-community/granite-3.0-2b-instruct-GGUF: Quantization
- Felladrin/gguf-q5_k_m-granite-3.0-2b-instruct: Quantization
- DevQuasar-5/ibm-granite.granite-3.0-2b-instruct-GGUF: Quantization
- robinsmits/Schaapje-2B-Pretrained: Full fine-tune
1 other repos hold byte-for-byte identical weight files.