You can deploy this model and start earning money today!

GoodStartLabs/
qwen3-32b-gsl247-armB-v3-mixed-r4

Name: qwen3-32b-gsl247-armB-v3-mixed-r4
Availability: InStock

Views

No views yet

Qwen3-32B · GSL-247 · arm B (v3-mixed-r4)

LoRA adapter trained on Lichess human games for the GSL-247 chess SFT ablation (form-vs-strategy axis). All four arms (A/B/C/F) share representation mix v3_mixed_history, N=200,000 records, identical hyperparameters; they vary only in the human-data filter applied upstream.

field	value
base model	`Qwen/Qwen3-32B`
PEFT	LoRA r=32, alpha=32, dropout=0.0, modules=all-linear
train slice	180,000 records (90/5/5 of 200k)
manifest hash	`34be47323ee6dafc40f9e98395d0a48b3aa3ab4986dcf1fdb688dc35a7ed189d`
manifest path	`data/v3/armB_4x.manifest.json`
representation	`v3_mixed_history`
epochs	1
optimizer	lr=0.001 (linear, warmup_ratio=0.03), batch=8
training method	sft
training infra	Together Fine-Tuning (job `ft-be4d8fc5-2ca0`)
source commit	`5b6e6b8`

Arm definitions

A: unfiltered human games
B: top-decile by player Elo (per-time-control p90 floor)
C: bottom-decile by player Elo (per-time-control p10 ceiling)
F: arm A minus the top-decile (i.e., everyone except B)

The B-vs-C contrast is the pre-registered falsifier for whether better human data improves SFT on chess. F is an exploratory matched-N rebuild used to disambiguate A ≈ B ≈ C results from the v2 run.

Usage

python

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-32B", device_map="auto")
tok  = AutoTokenizer.from_pretrained("Qwen/Qwen3-32B")
model = PeftModel.from_pretrained(base, "GoodStartLabs/qwen3-32b-gsl247-armB-v3-mixed-r4")

Eval contract

The pre-registered evaluation contract (rung-0 floor gates, rung-1 puzzle Elo ladder, B-vs-C falsifier thresholds) lives at eval/contract.yaml in the source repo. Every Inspect task stamps its contract_sha into the .eval metadata so any reported number is recoverable to the thresholds in force.