Create throughput_debugging/nn1_nanosets_32k_steps.yaml

throughput_debugging/nn1_nanosets_32k_steps.yaml

@@ -0,0 +1,119 @@
+{
+  "nanotron_config": {
+    "value": {
+      "checkpoints": {
+        "checkpoint_interval": 2000,
+        "checkpoints_path": "/scratch/loubna/checkpoints/smollm3-ablations-3p56G-3B-nanotron-old-nn1-fwedu-seed-6-32k",
+        "checkpoints_path_is_shared_file_system": false,
+        "resume_checkpoint_path": null,
+        "save_initial_state": false
+      },
+      "data_stages": [
+        {
+          "data": {
+            "dataset": {
+              "dataset_folder": [
+                "/fsx/loubna/datasets/llama_tokenized/fineweb-edu/merged"
+              ]
+            },
+            "num_loading_workers": 1,
+            "seed": 6
+          },
+          "name": "stable",
+          "start_training_step": 1
+        }
+      ],
+      "general": {
+        "benchmark_csv_path": null,
+        "consumed_train_samples": null,
+        "ignore_sanity_checks": true,
+        "project": "smollm3-training",
+        "run": "3B-nanotron-old-nn1-fwedu-32k",
+        "seed": 6,
+        "step": null
+      },
+      "lighteval": null,
+      "logging": {
+        "iteration_step_info_interval": 1,
+        "log_level": "info",
+        "log_level_replica": "info"
+      },
+      "model": {
+        "ddp_bucket_cap_mb": 128,
+        "dtype": "bfloat16",
+        "init_method": {
+          "std": 0.02
+        },
+        "make_vocab_size_divisible_by": 1,
+        "model_config": {
+          "bos_token_id": 128000,
+          "eos_token_id": 128001,
+          "hidden_act": "silu",
+          "hidden_size": 2048,
+          "initializer_range": 0.02,
+          "intermediate_size": 11008,
+          "is_llama_config": true,
+          "max_position_embeddings": 4096,
+          "num_attention_heads": 16,
+          "num_hidden_layers": 36,
+          "num_key_value_heads": 4,
+          "pad_token_id": null,
+          "pretraining_tp": 2,
+          "rms_norm_eps": 0.000001,
+          "rope_scaling": null,
+          "rope_theta": 10000,
+          "tie_word_embeddings": true,
+          "use_cache": true,
+          "vocab_size": 128256
+        }
+      },
+      "optimizer": {
+        "accumulate_grad_in_fp32": true,
+        "clip_grad": 1,
+        "learning_rate_scheduler": {
+          "learning_rate": 0.0002,
+          "lr_decay_starting_step": 26000,
+          "lr_decay_steps": 6000,
+          "lr_decay_style": "linear",
+          "lr_warmup_steps": 2000,
+          "lr_warmup_style": "linear",
+          "min_decay_lr": 0
+        },
+        "optimizer_factory": {
+          "adam_beta1": 0.9,
+          "adam_beta2": 0.95,
+          "adam_eps": 1e-8,
+          "name": "adamW",
+          "torch_adam_is_fused": true
+        },
+        "weight_decay": 0.01,
+        "zero_stage": 0
+      },
+      "parallelism": {
+        "dp": 4,
+        "expert_parallel_size": 1,
+        "pp": 1,
+        "pp_engine": "1f1b",
+        "recompute_layer": false,
+        "tp": 2,
+        "tp_linear_async_communication": true,
+        "tp_mode": "REDUCE_SCATTER"
+      },
+      "profiler": null,
+      "tokenizer": {
+        "tokenizer_max_length": 4096,
+        "tokenizer_name_or_path": "meta-llama/Llama-3.2-1B",
+        "tokenizer_revision": null
+      },
+      "tokens": {
+        "batch_accumulation_per_replica": 1,
+        "limit_test_batches": 0,
+        "limit_val_batches": 0,
+        "micro_batch_size": 3,
+        "sequence_length": 4096,
+        "train_steps": 32000,
+        "val_check_interval": 100
+      }
+    }
+  }
+}