ibm-ai-platform/micro-g3.3-8b-instruct-1b
Full fine-tune (fewer layers) vs ibm-granite/granite-3.3-8b-instruct. ibm-ai-platform/micro-g3.3-8b-instruct-1b matches part of ibm-granite/granite-3.3-8b-instruct (38 of its 362 tensors; the rest of ibm-granite/granite-3.3-8b-instruct has no counterpart). On the shared part: Fine-tune: 100% of the weights changed, across 4 of 4 layers.
Tensors changed: Embeddings 1/1, Norms 9/9, MLP 12/12, Attention 16/16
3 derivatives checked
- ibm-ai-platform/micro-g3.3-8b-instruct-1b-FP8: Quantization
- mradermacher/micro-g3.3-8b-instruct-1b-i1-GGUF: Full fine-tune
- mradermacher/micro-g3.3-8b-instruct-1b-GGUF: Full fine-tune