GoodStartLabs

sf-cot-dagger-chess-qwen3-1.7b

README

License: apache-2.0

TL;DR
Model details
Intended use & limitations
Prompt & output format
How to use
Training data
Training procedure
Evaluation
Results
The research finding
Reproducibility & artifacts
Citation

TL;DR

What: Qwen3-1.7B + LoRA that writes a Stockfish-style verbalized-minimax analysis for a chess position and then commits best move: <uci>. No engine graft, no soft-tokens — a pure-text control.
Why it exists: to test whether plain text chain-of-thought (plus on-policy DAgger correction) can turn chess reasoning into a good committed move at the 1.7B scale.
Headline result: it can't. DAgger improved the text baseline from −116% to a −66% peak (this checkpoint) recovery of the search headroom, but that peak still commits worse than a no-search policy prior, and neither more DAgger rounds nor fresher, more-diverse on-policy data pushed past it.

Model details

Table

Base model	`Qwen/Qwen3-1.7B`
Adapter type	LoRA (PEFT), r = 32, α = 64, dropout 0
Target modules	`q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj`
Trainable params	~35 M (adapter only; ~133 MB safetensors)
Precision	bf16
Task	causal-LM (chess position → verbalized analysis → committed move)

Intended use & limitations

Intended use. Research reproducibility for the Latent Grafting "read → commit wall" study; a baseline to compare against latent-value approaches; an example of a verbalized-minimax CoT format.

Out of scope. Actual chess play or move recommendation. On adversarially-selected hard positions this model's committed move is, on average, worse than simply taking a policy prior's top move without any search. It also has not been evaluated for full-game Elo.

Known limitations.

Commits worse than the no-search prior on the search-required suite (see Results).
Trained/evaluated only on chess; single game.
The evaluation suite is selected around a particular prior's errors — absolute recovery % is calibrated to that frame (the direction of the result is robust; the magnitude is frame-relative).
Occasionally the verbalized value reads saturate; commit is by parsing best move: from the trace.

Prompt & output format

Input:

markdown
FEN: <fen> moves: <space-separated recent uci moves>
Analyze:

Output (the model continues from Analyze:): a verbalized-minimax trace — top candidate moves with evaluations and short principal-variation lines — terminating in a legal committed move:

markdown
<uci>: <eval> (line: <pv> ) ; <uci>: <eval> (line: <pv> ) ; ... ; best move: <uci>

How to use

python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

REPO = "GoodStartLabs/sf-cot-dagger-chess-qwen3-1.7b"
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-1.7B")
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-1.7B", torch_dtype=torch.bfloat16)
model = PeftModel.from_pretrained(base, REPO).eval()

fen = "r1bqkbnr/pppp1ppp/2n5/4p3/2B1P3/5Q2/PPPP1PPP/RNB1K1NR w KQkq - 4 4"
prompt = f"FEN: {fen} moves: \nAnalyze:"
ids = tok(prompt, return_tensors="pt")
out = model.generate(**ids, max_new_tokens=200, do_sample=False)
print(tok.decode(out[0][ids.input_ids.shape[1]:], skip_special_tokens=True))
# ... parse the trailing "best move: <uci>" to get the committed move.

Training data

Base SFT corpus: ~40,000 verbalized-minimax Stockfish traces (Stockfish 17.1). Each root position is analysed with top-6 multipv plus principal-variation lines, at adaptive depth (12, deepened to 20 when the top-2 evaluations are close), then serialized into the text format above. ~38.9k unique positions (highly diverse).
DAgger corpora (rounds 1–5): on-policy positions the model reached or mis-committed during roll-out, each relabeled with a fresh correct Stockfish trace (wrong-commits duplicated 2×). Accumulated cumulatively across rounds. Root positions mined from a self-play/tactical shard pool.

Training procedure

SFT the base corpus (completion-only cross-entropy on the trace tokens; the prompt is masked).
DAgger loop, 5 rounds: roll the current model out over ~2,000 root positions → flag the positions where its committed move disagrees with Stockfish + the position it reaches after its move → generate correct Stockfish traces for those → append to the corpus → retrain from the cumulative corpus (3,000 steps). Round r's model seeds round r+1's roll-out.
Checkpoint selection: dag5 = the round with the best committed-move recovery (and, on the original loop, the minimum SFT eval-loss simultaneously).

Hyperparameters: LoRA r32/α64; bf16; completion-only CE; per-round 3,000 steps.

Evaluation

Metric. Committed-move centipawn loss (cp-loss): the model commits one move per position; the loss is how many centipawns worse that move is than Stockfish's best (per Stockfish evaluation).

Suite. A search-required suite of 150 positions, selected so a no-search policy prior is materially wrong (prior cp-loss 173.0; Stockfish-best "oracle" 92.6). This isolates positions where reasoning/search should help.

Recovery % = (prior − model) / (prior − oracle) — the fraction of the 173→92.6 headroom the model closes. 0% = matches the prior; +100% = matches Stockfish; negative = worse than the prior.

Definitions of the reference points:

base LLM — Qwen3-1.7B reading the raw FEN, no adapter (built-in ability, ~619 Elo elsewhere).
prior — a no-search policy baseline's top move (already a strong reference: it beats this model). Not the bare base model.
oracle — Stockfish's actual best move on the suite.

Results

Committed-move cp-loss on the search-required suite (prior 173.0, oracle 92.6):

Table with columns: model, cp-loss, recovery
model	cp-loss	recovery
Stockfish best (oracle)	92.6	+100%
no-search policy prior	173.0	0%
this model — SF-CoT + DAgger round 5	226.4	−66%
SF-CoT baseline (pre-DAgger)	266.4	−116%

DAgger trajectory (this model is round 5):

Table with columns: round, cp-loss, recovery, SFT eval-loss
round	cp-loss	recovery	SFT eval-loss
pre-DAgger	266.4	−116%	—
1	263.6	−113%	—
2	247.6	−93%	0.543
3	250.4	−96%	0.519
4

Past round 5, SFT eval-loss kept falling to new minima while committed-move recovery worsened — imitation loss and move quality decouple.

Diversity control (fresh disjoint on-policy positions each round, restarted from round 5):

Table with columns: round, recovery, eval-loss
round	recovery	eval-loss
fresh 1	−96%	0.489
fresh 2	−84%	0.495
fresh 3	−101%	0.507
fresh 4	−77%	0.497

Fresh, more-diverse positions (eval-loss rises, confirming they are not merely memorized) still land in the −77…−101% band — they do not recover the round-5 peak. So the plateau is not an artifact of DAgger re-mining a stale position set; it is a genuine commit ceiling.

The research finding (why this model exists)

This model is the text-only control in a study of where a value signal is lost between a chess engine and a language model's committed move. The broader project finds:

The engine's position value is abundant and linearly readable upstream (a ridge probe recovers position value at R² 0.92 from the engine trunk), and a dedicated latent reader can commit at +95% recovery.
But every path that turns that value into a generated committed move at 1.7B loses it: latent graft generative commit ≈ −42%; this text-CoT baseline ≈ −116%, improved by DAgger to a −66% peak — all still worse than the no-search prior.

Conclusion: at this scale the bottleneck is move selection / commit, not the availability or readability of the value signal. Plain text chain-of-thought + DAgger does not overcome it (and this model is its best point). See the project write-up for the full ladder (base → prior → oracle) and the latent-value approaches.

Reproducibility & artifacts

Base model: Qwen/Qwen3-1.7B.
This adapter: runs/graft/sf_baseline_dag5/lora_best (round-5 DAgger checkpoint).
Format: PEFT LoRA (adapter_config.json + adapter_model.safetensors).
Eval: committed-move cp-loss on a 150-position search-required suite (prior 173.0, oracle 92.6).

Citation

If you reference this artifact:

bibtex
@misc{goodstartlabs_sfcot_dagger_2026,
  title  = {SF-CoT + DAgger: a text chain-of-thought chess move-commit baseline (Qwen3-1.7B LoRA)},
  author = {GoodStart Labs},
  year   = {2026},
  note   = {Latent Grafting project; research negative-result artifact.},
  howpublished = {\url{https://huggingface.co/GoodStartLabs/sf-cot-dagger-chess-qwen3-1.7b}}
}

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Model Details

Model Provider

GoodStartLabs

Model Tree

Base

Qwen/Qwen3-1.7B

Adapter

this model

Input Modalities

Text

Output Modalities

Text

Supported Functionality

Dedicated EndpointsContainer

Explore FriendliAI today

Get started Talk to an engineer

README

License: apache-2.0

TL;DR
Model details
Intended use & limitations
Prompt & output format
How to use
Training data
Training procedure
Evaluation
Results
The research finding
Reproducibility & artifacts
Citation

TL;DR

What: Qwen3-1.7B + LoRA that writes a Stockfish-style verbalized-minimax analysis for a chess position and then commits best move: <uci>. No engine graft, no soft-tokens — a pure-text control.
Why it exists: to test whether plain text chain-of-thought (plus on-policy DAgger correction) can turn chess reasoning into a good committed move at the 1.7B scale.
Headline result: it can't. DAgger improved the text baseline from −116% to a −66% peak (this checkpoint) recovery of the search headroom, but that peak still commits worse than a no-search policy prior, and neither more DAgger rounds nor fresher, more-diverse on-policy data pushed past it.

Model details

Table

Base model	`Qwen/Qwen3-1.7B`
Adapter type	LoRA (PEFT), r = 32, α = 64, dropout 0
Target modules	`q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj`
Trainable params	~35 M (adapter only; ~133 MB safetensors)
Precision	bf16
Task	causal-LM (chess position → verbalized analysis → committed move)

Intended use & limitations

Known limitations.

Commits worse than the no-search prior on the search-required suite (see Results).
Trained/evaluated only on chess; single game.
The evaluation suite is selected around a particular prior's errors — absolute recovery % is calibrated to that frame (the direction of the result is robust; the magnitude is frame-relative).
Occasionally the verbalized value reads saturate; commit is by parsing best move: from the trace.

Prompt & output format

Input:

markdown
FEN: <fen> moves: <space-separated recent uci moves>
Analyze:

Output (the model continues from Analyze:): a verbalized-minimax trace — top candidate moves with evaluations and short principal-variation lines — terminating in a legal committed move:

markdown
<uci>: <eval> (line: <pv> ) ; <uci>: <eval> (line: <pv> ) ; ... ; best move: <uci>

How to use

python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

REPO = "GoodStartLabs/sf-cot-dagger-chess-qwen3-1.7b"
tok = AutoTokenizer.from_pretrained("Qwen/Qwen3-1.7B")
base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-1.7B", torch_dtype=torch.bfloat16)
model = PeftModel.from_pretrained(base, REPO).eval()

fen = "r1bqkbnr/pppp1ppp/2n5/4p3/2B1P3/5Q2/PPPP1PPP/RNB1K1NR w KQkq - 4 4"
prompt = f"FEN: {fen} moves: \nAnalyze:"
ids = tok(prompt, return_tensors="pt")
out = model.generate(**ids, max_new_tokens=200, do_sample=False)
print(tok.decode(out[0][ids.input_ids.shape[1]:], skip_special_tokens=True))
# ... parse the trailing "best move: <uci>" to get the committed move.

Training data

Base SFT corpus: ~40,000 verbalized-minimax Stockfish traces (Stockfish 17.1). Each root position is analysed with top-6 multipv plus principal-variation lines, at adaptive depth (12, deepened to 20 when the top-2 evaluations are close), then serialized into the text format above. ~38.9k unique positions (highly diverse).
DAgger corpora (rounds 1–5): on-policy positions the model reached or mis-committed during roll-out, each relabeled with a fresh correct Stockfish trace (wrong-commits duplicated 2×). Accumulated cumulatively across rounds. Root positions mined from a self-play/tactical shard pool.

Training procedure

SFT the base corpus (completion-only cross-entropy on the trace tokens; the prompt is masked).
DAgger loop, 5 rounds: roll the current model out over ~2,000 root positions → flag the positions where its committed move disagrees with Stockfish + the position it reaches after its move → generate correct Stockfish traces for those → append to the corpus → retrain from the cumulative corpus (3,000 steps). Round r's model seeds round r+1's roll-out.
Checkpoint selection: dag5 = the round with the best committed-move recovery (and, on the original loop, the minimum SFT eval-loss simultaneously).

Hyperparameters: LoRA r32/α64; bf16; completion-only CE; per-round 3,000 steps.

Evaluation

Metric. Committed-move centipawn loss (cp-loss): the model commits one move per position; the loss is how many centipawns worse that move is than Stockfish's best (per Stockfish evaluation).

Definitions of the reference points:

base LLM — Qwen3-1.7B reading the raw FEN, no adapter (built-in ability, ~619 Elo elsewhere).
prior — a no-search policy baseline's top move (already a strong reference: it beats this model). Not the bare base model.
oracle — Stockfish's actual best move on the suite.

Results

Committed-move cp-loss on the search-required suite (prior 173.0, oracle 92.6):

Table with columns: model, cp-loss, recovery
model	cp-loss	recovery
Stockfish best (oracle)	92.6	+100%
no-search policy prior	173.0	0%
this model — SF-CoT + DAgger round 5	226.4	−66%
SF-CoT baseline (pre-DAgger)	266.4	−116%

DAgger trajectory (this model is round 5):

Table with columns: round, cp-loss, recovery, SFT eval-loss
round	cp-loss	recovery	SFT eval-loss
pre-DAgger	266.4	−116%	—
1	263.6	−113%	—
2	247.6	−93%	0.543
3	250.4	−96%	0.519
4

Past round 5, SFT eval-loss kept falling to new minima while committed-move recovery worsened — imitation loss and move quality decouple.

Diversity control (fresh disjoint on-policy positions each round, restarted from round 5):

Table with columns: round, recovery, eval-loss
round	recovery	eval-loss
fresh 1	−96%	0.489
fresh 2	−84%	0.495
fresh 3	−101%	0.507
fresh 4	−77%	0.497

The research finding (why this model exists)

This model is the text-only control in a study of where a value signal is lost between a chess engine and a language model's committed move. The broader project finds:

The engine's position value is abundant and linearly readable upstream (a ridge probe recovers position value at R² 0.92 from the engine trunk), and a dedicated latent reader can commit at +95% recovery.
But every path that turns that value into a generated committed move at 1.7B loses it: latent graft generative commit ≈ −42%; this text-CoT baseline ≈ −116%, improved by DAgger to a −66% peak — all still worse than the no-search prior.

Reproducibility & artifacts

Base model: Qwen/Qwen3-1.7B.
This adapter: runs/graft/sf_baseline_dag5/lora_best (round-5 DAgger checkpoint).
Format: PEFT LoRA (adapter_config.json + adapter_model.safetensors).
Eval: committed-move cp-loss on a 150-position search-required suite (prior 173.0, oracle 92.6).

Citation

If you reference this artifact:

bibtex
@misc{goodstartlabs_sfcot_dagger_2026,
  title  = {SF-CoT + DAgger: a text chain-of-thought chess move-commit baseline (Qwen3-1.7B LoRA)},
  author = {GoodStart Labs},
  year   = {2026},
  note   = {Latent Grafting project; research negative-result artifact.},
  howpublished = {\url{https://huggingface.co/GoodStartLabs/sf-cot-dagger-chess-qwen3-1.7b}}
}

sf-cot-dagger-chess-qwen3-1.7b

README

Table of contents

TL;DR

Model details

Intended use & limitations

Prompt & output format

How to use

Training data

Training procedure

Evaluation

Results

The research finding (why this model exists)

Reproducibility & artifacts

Citation

Explore FriendliAI today

README

Table of contents

TL;DR

Model details

Intended use & limitations

Prompt & output format

How to use

Training data

Training procedure

Evaluation

Results

The research finding (why this model exists)

Reproducibility & artifacts

Citation