ibm-granite/granite-3.1-1b-a400m-instruct
Full fine-tune vs ibm-granite/granite-3.1-1b-a400m-base. Fine-tune: 100% of the weights changed, across 24 of 24 layers.
Tensors changed: other 72/72, Norms 43/49, Attention 96/96
10 derivatives checked
- bartowski/granite-3.1-1b-a400m-instruct-GGUF: Same weights, converted
- OVRLab/granite-3.1-1b-a400m-concision-experiment: Abliterated
- mradermacher/granite-3.1-1b-a400m-instruct-i1-GGUF: Quantization
- ziaddddd/granite-3.1-1b-a400m-concision: Abliterated
- lmstudio-community/granite-3.1-1b-a400m-instruct-GGUF: Quantization
- Myric/granite-3.1-1b-a400m-instruct-APEX: Quantization
- mradermacher/granite-3.1-1b-a400m-instruct-GGUF: Same weights, converted
- Alcoft/granite-3.1-1b-a400m-instruct-GGUF: Same weights, converted
- DevQuasar-6/ibm-granite.granite-3.1-1b-a400m-instruct-GGUF: Quantization
- PJMixers-Dev/Granite-3.1-Earthen-v0.3-1B-A400M: Partial fine-tune
1 other repos hold byte-for-byte identical weight files.