Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags: [atc, coconut, continuous-cot, graph-reasoning, checkpoints]
|
| 4 |
+
---
|
| 5 |
+
|
| 6 |
+
# ATC tree10 runs
|
| 7 |
+
|
| 8 |
+
Training artifacts for the **Abstract Token Curriculum (ATC)** twin-random-tree (size-10) experiment:
|
| 9 |
+
a 2x2x2x2 factorial over **loss** (CE vs sum-of-log), **reset_optimizer** (on/off),
|
| 10 |
+
**staging** (threshold/monotone vs revisit), and **uniform_prob** (0.1/0.5), plus related arms.
|
| 11 |
+
|
| 12 |
+
Each subfolder is one run and contains:
|
| 13 |
+
- `checkpoint_latest.pt` - full resumable state (weights + optimizer + curriculum position)
|
| 14 |
+
- `checkpoint_<n>.pt` - epoch-stamped snapshot(s)
|
| 15 |
+
- `config.json` - the run configuration
|
| 16 |
+
- `metrics.jsonl` - per-epoch metrics (eval/acc, stage, per-stage acc, losses)
|
| 17 |
+
|
| 18 |
+
Model: 2-layer / 8-head / 768-dim GPT-2 trained from scratch on symbolic reachability graphs.
|