|
Download README.md from rasulkhanbayov/somw-checkpoints: direct link, hf CLI and curl.
- Browser
- Download file 946 Bytes
-
https://huggingface.co/rasulkhanbayov/somw-checkpoints/resolve/main/README.md
- Command line
-
hf download hf://rasulkhanbayov/somw-checkpoints/README.md
-
curl -L -o README.md https://huggingface.co/rasulkhanbayov/somw-checkpoints/resolve/main/README.md
946 Bytes
| license: other | |
| tags: [physics, diffusion, fine-tuning] | |
| # Simulate Only What Matters: fine-tuned PhysCtrl checkpoints | |
| Checkpoints trained in the project *Simulate Only What Matters* (frozen Nov 3, 2026). All are fine-tunes of PhysCtrl's | |
| base model and all were **negative results** (see `PLAN.md`, Phases 4g, 4g-v2, 14a): | |
| - `fc_base_v1/`: Phase 4g frame-conditioned full fine-tune (forgetting: +47% in-distribution error). | |
| - `fc_base_v2/`: Phase 4g-v2 temporal-attention LoRA + known-frame embedding (no forgetting, no repair gain). | |
| - `p14a_{ACT,RND}_{F,L}/`: Phase 14a simulator-distillation pilot (KILL by the pre-registered rule). | |
| Each `final.pt` (and `step_*.pt`) holds the state dict, optimizer state, step and LoRA settings; loaders: | |
| `code/fc_model.load_fc` (4g) and `code/p14_eval_generate.make_generator` (14a). | |
| **Licence:** derived from PhysCtrl's released weights, which declare no licence; check PhysCtrl's terms before reuse. | |