qwen3-4b-sorrel-selfloop-g6-midtrain

Private research artifact — do not redistribute. Anthropic Fellows project on flourishing-framed character training (Wang & Jermyn pitch, 2026-04-22).

• base model: joshycodes/qwen3-4b-sorrel-selfloop-g5-midtrain @ 6f2da11ab480
• steps: midtrain
• run: qwen3-4b-sorrel-selfloop-g5-midtrain-sorrel-selfloop-b-g5-m-0916-0651 on 2x NVIDIA H200 (RunPod, fellows worker)
• launcher commit: a0afb77669ae of the flourishing-training repo
• seed: 20260821

step	data	revision	tokens seen	loss
midtrain	joshycodes/sorrel-selfloop-corpus (config sorrel-selfloop-b-g5)	06e974d81a74	3,682,304 tokens	loss 0.9303 → 0.9138

Hyperparameters:

{
  "midtrain": {
    "lr": 1e-05,
    "seq_len": 4096,
    "micro_batch": 4,
    "grad_accum": 2,
    "epochs": 1.0
  }
}

Full run config in train_run_config.json; evaluate with uv run eval.py --model joshycodes/qwen3-4b-sorrel-selfloop-g6-midtrain --eval all.