noop-test
Browse files
README.md
CHANGED
|
@@ -1,3 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
# AdapterCast LoRA Trajectory Corpus
|
| 2 |
|
| 3 |
Dense LoRA fine-tuning trajectories (adapter weight snapshots + gauge-invariant
|
|
@@ -24,6 +37,9 @@ the E0 measurement program (incl. 1.7B/4B scale rows and gauge-twin experiments)
|
|
| 24 |
`manifest.json` echoes the frozen train/held-out splits (tasks, ranks {32,64},
|
| 25 |
scale) — pre-registered BEFORE meta-training (git tags `e1-freeze`, `e2-freeze`).
|
| 26 |
|
|
|
|
|
|
|
|
|
|
| 27 |
## Reproduction
|
| 28 |
Code: the `adaptercast` repository (deterministic pipeline — replicate runs are
|
| 29 |
bit-identical on one host; dataset revisions pinned in each run's provenance).
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
pretty_name: AdapterCast LoRA Trajectory Corpus
|
| 4 |
+
tags:
|
| 5 |
+
- lora
|
| 6 |
+
- peft
|
| 7 |
+
- training-dynamics
|
| 8 |
+
- trajectory-forecasting
|
| 9 |
+
- learning-dynamics
|
| 10 |
+
size_categories:
|
| 11 |
+
- n<1K
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
# AdapterCast LoRA Trajectory Corpus
|
| 15 |
|
| 16 |
Dense LoRA fine-tuning trajectories (adapter weight snapshots + gauge-invariant
|
|
|
|
| 37 |
`manifest.json` echoes the frozen train/held-out splits (tasks, ranks {32,64},
|
| 38 |
scale) — pre-registered BEFORE meta-training (git tags `e1-freeze`, `e2-freeze`).
|
| 39 |
|
| 40 |
+
Note: `smoke_*` pipeline-validation runs are inventoried in `manifest.json` for
|
| 41 |
+
completeness but are not shipped in this dataset.
|
| 42 |
+
|
| 43 |
## Reproduction
|
| 44 |
Code: the `adaptercast` repository (deterministic pipeline — replicate runs are
|
| 45 |
bit-identical on one host; dataset revisions pinned in each run's provenance).
|