Create ds_config_zero3.json

ds_config_zero3.json

@@ -0,0 +1,61 @@
+{
+  "bf16": {
+    "enabled": true
+  },
+  "zero_optimization": {
+    "stage": 3,
+    "offload_optimizer": {
+      "device": "cpu",
+      "pin_memory": true
+    },
+    "offload_param": {
+      "device": "cpu",
+      "pin_memory": true
+    },
+    "overlap_comm": true,
+    "contiguous_gradients": true,
+    "sub_group_size": 1e9,
+    "reduce_bucket_size": 5e8,
+    "stage3_prefetch_bucket_size": 5e8,
+    "stage3_param_persistence_threshold": 1e6,
+    "stage3_max_live_parameters": 1e9,
+    "stage3_max_reuse_distance": 1e9,
+    "stage3_gather_16bit_weights_on_model_save": true
+  },
+  "gradient_accumulation_steps": 32,
+  "gradient_clipping": 1.0,
+  "steps_per_print": 10,
+  "train_batch_size": "auto",
+  "train_micro_batch_size_per_gpu": "auto",
+  "wall_clock_breakdown": false,
+  "communication_data_type": "bf16",
+  "prescale_gradients": false,
+  "sparse_gradients": false,
+  "compression_training": {
+    "weight_quantization": {
+      "shared_parameters": {},
+      "different_groups": {}
+    },
+    "activation_quantization": {
+      "shared_parameters": {},
+      "different_groups": {}
+    },
+    "sparse_pruning": {
+      "shared_parameters": {},
+      "different_groups": {}
+    }
+  },
+  "flops_profiler": {
+    "enabled": false,
+    "profile_step": 1,
+    "module_depth": -1,
+    "top_modules": 1,
+    "detailed": true,
+    "output_file": null
+  },
+  "tensorboard": {
+    "enabled": true,
+    "output_path": "./logs/tensorboard",
+    "job_name": "helion_v2_training"
+  }
+}