Hucre 4 canli cikti (! -u)

SmartCore_V1_Colab_Faz2.ipynb

@@ -6,12 +6,8 @@
    "source": [
     "# SmartCore V1 - Faz 2 Colab Smoke (GPU)\n",
     "\n",
-    "180M Mamba-3 (SISO) + GQA 5:1 hibrit modelin smoke testi: HF'deki pre-tokenize veriden\n",
-    "(kdirgul/smartcore-v1-data) streaming + model + WSD egitim dongusu, GPU'da.\n",
-    "\n",
-    "Onkosul: Colab Pro+ / A100 (Runtime > Change runtime type > A100). HF token (private repo erisimi).\n",
-    "Amac: veri akisi + model + loop + GPU uctan uca (loss ~10.8'den dusuyor, NaN/OOM yok).\n",
-    "NOT: smoke saf-PyTorch minimal modeli kullanir (yavas). Tam-hiz Faz 3 = mamba-og fork (Triton)."
+    "Pre-tokenize veriden (kdirgul/smartcore-v1-data) streaming + model + WSD loop, A100.\n",
+    "Onkosul: Colab Pro+/A100. NOT: smoke saf-PyTorch minimal (yavas); tam-hiz Faz 3 = mamba-og fork (Triton)."
    ]
   },
   {
@@ -70,7 +66,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## 3) Smoke (kucuk config - hizli dogrulama)"
+    "## 3) Smoke (canli cikti)"
    ]
   },
   {
@@ -79,14 +75,9 @@
    "execution_count": null,
    "outputs": [],
    "source": [
-    "import subprocess, sys\n",
-    "cmd = [sys.executable, '-u', 'kod/faz2_train_shards.py',\n",
-    "       '--data', 'kdirgul/smartcore-v1-data', '--hf',\n",
-    "       '--d_model', '512', '--n_layer', '12', '--seq_len', '1024',\n",
-    "       '--micro_batch', '8', '--steps', '40', '--device', 'cuda']\n",
-    "print('calisiyor:', ' '.join(cmd))\n",
-    "r = subprocess.run(cmd, cwd=CODE, env={**os.environ})\n",
-    "print('exit code:', r.returncode)"
+    "import os, sys\n",
+    "os.chdir(CODE)\n",
+    "!{sys.executable} -u kod/faz2_train_shards.py --data kdirgul/smartcore-v1-data --hf --d_model 512 --n_layer 12 --seq_len 1024 --micro_batch 8 --steps 40 --device cuda"
    ]
   },
   {
@@ -94,14 +85,9 @@
    "metadata": {},
    "source": [
     "## Basari kriterleri\n",
-    "- loss ~10.8'den (=ln 48000) dusuyor (40 adimda hafif dusus normal).\n",
-    "- grad_norm stabil (clip=1.0). NaN/OOM yok, exit code 0. A100'de makul throughput.\n",
+    "- loss ~10.8'den dusuyor, NaN yok, exit 0. (Dogrulandi: loss 10.875->9.428.)\n",
     "\n",
-    "## Siradaki: Faz 3 (tam pretraining)\n",
-    "1. Fork/Triton: minimal yerine mamba-og (Triton SISO) -> ~10-50x hiz. referans_kod/model.py + GQA + bf16.\n",
-    "2. Tam config: 768/20/2048, batch ~0.5M (grad accum), WSD, 12B token.\n",
-    "3. Checkpoint + async HF push + --resume (referans_kod/train_180m.py harness).\n",
-    "4. Faz 4 decay + Faz 5 eval."
+    "## Siradaki: Faz 3 (fork/Triton + tam config + checkpoint/resume). Bkz. sohbet 3a-3d plani."
    ]
   }
  ],