Downloads · 30 days
0
arc-agi-2-solver-qwen3-4b
Nabidnur/arc-agi-2-solver-qwen3-4b
arc-agi-2-solver-qwen3-4b is a machine learning model from Nabidnur. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for peft. The card lists the license as other.
Staged toolkit for the Kaggle ARC Prize 2026 solver (NVARC lineage: sorokin/qwen34bgrids15sft139 + per-task rank-256 rsLoRA + turbo-DFS). Weights are produced on GPU day by running trainlorasft.py; this repo ships the…
Access
Public
Updated Sep 22, 2026
Repo size
—
Likes
1
Public
Repo makeup
Click a slice to open those files.
.py59.2 KB · 91%
At a glance
- Library
- peft
- License
- other
- Access
- Public
- Created
- Sep 22, 2026
- Updated
- Sep 22, 2026
- SHA
- f8ceba8c
Base models
Model card
From the Hugging Face model README
arc-agi-2-solver-qwen3-4b — GPU-day training + decoding toolkit
Staged toolkit for the Kaggle ARC Prize 2026 solver (NVARC lineage:
sorokin/qwen3_4b_grids15_sft139 + per-task rank-256 rsLoRA + turbo-DFS).
Weights are produced on GPU day by running train_lora_sft.py; this repo
ships the complete executable pipeline, pinned configs and tests so the
Kaggle run is a two-command operation.
Files
| File | Purpose |
|---|---|
arc_vocabs.py | 12-token ARC vocabulary + legal-grid grammar (v7) |
decode_dfs.py | turbo-DFS candidate decoder (exact audited port) + grammar hook + teacher-forced rescoring |
selector.py | score_kgmon (exact port) + shape-prior variant + top-2 distinct selection |
train_lora_sft.py | MODE A offline SFT on the HF dataset; MODE B per-task TTT (reference behaviour) |
run_config.yaml | pinned runtime configuration (model path, budgets, thresholds) |
Companion dataset: Nabidnur/arc-agi-2-grids (canonical splits, 128k
SFT sequences, features, folds, verified synthetic curriculum).
Reproduce the baseline (Kaggle, 4×L4, offline)
python train_lora_sft.py --mode offline_sft \
--data sft/train_sft128.jsonl.gz --out /kaggle/working/ckpt # optional offline SFT
# per-task TTT + decode runs inside the Kaggle notebook (arc26-nvarc-v7)
Pinned baseline: LoRA r=256, rsLoRA alpha=32, dropout 0, targets
q/k/v/o/gate/up/down + embed_tokens + lm_head; LR 5e-5 cosine, 1 epoch,
bf16, max_seq_len 8192; search floor cum_NLL < -ln(0.2); selection
score_kgmon = votes − mean(augmented_NLL), top-2 distinct.
v7 deltas (each behind a flag, ablated before adoption — see
docs/MASTER_PLAN_INTEGRATION.md in the GitHub repo)
- Legal-grid grammar during DFS (no 31-row / ragged / mid-row-EOS branches).
- Two-stage search: cheap confident pass first, broadened floor on starved outputs.
- Adaptive scheduling: calibrated cost model + coverage-first deadlines (v4 lineage).
- Program branch: demo-verified primitives (off-center symmetry orbit fill, hole topology recolor, path propagation) join the candidate pool with 100% measured precision (3/3 public case tasks exact, incl. the 30×30 that beat the neural pool).
Scoring
Output-level exact pass@2 over ALL test outputs (172 public eval outputs; one output = +0.5814 pts). Secondary diagnostics: task-mean, task-all.
- Library
- peft
- License
- other
- Created
- Sep 22, 2026
- Updated
- Sep 22, 2026