diff --git "a/attnserver.run_attnserver.slurm.sh.343204.out.log" "b/attnserver.run_attnserver.slurm.sh.343204.out.log" --- "a/attnserver.run_attnserver.slurm.sh.343204.out.log" +++ "b/attnserver.run_attnserver.slurm.sh.343204.out.log" @@ -10144,3 +10144,6252 @@ CHECKPOINT_PATH: gpt-checkpoint PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron -------------------------------- /mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ +INFO:megatron.training.initialize:Setting logging level to 0 + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 16384 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 16384 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 16384 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +INFO:megatron.training.initialize:Setting logging level to 0 +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.042 seconds +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.599 seconds +time to initialize megatron (seconds): 8.504 +[after megatron is initialized] datetime: 2025-06-21 21:20:08 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 137426432 + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 137426432 + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 137426432 + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 137426432 + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 137426432 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 137426432 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (137426432 elements, 137426432 padded size): + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.embedding.word_embeddings.weight + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.embedding.position_embeddings.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine + loading distributed checkpoint from gpt-checkpoint at iteration 10 +Running ctx_length=24576, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 24576 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 24576 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 24576 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 24576 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 24576 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +INFO:megatron.training.initialize:Setting logging level to 0 +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.043 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.902 seconds +time to initialize megatron (seconds): 8.647 +[after megatron is initialized] datetime: 2025-06-21 21:20:46 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 170980864 + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 170980864 + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 170980864 + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 170980864 + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 170980864 + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 170980864 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 170980864 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (170980864 elements, 170980864 padded size): + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.0.mlp.linear_fc1.bias + module.embedding.word_embeddings.weight + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.embedding.position_embeddings.weight + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine +WARNING: could not find the metadata file gpt-checkpoint/latest_checkpointed_iteration.txt + will not load any checkpoints and will start from random +(min, max) time across ranks (ms): + load-checkpoint ................................: (4.15, 4.44) +[after model, optimizer, and learning rate scheduler are built] datetime: 2025-06-21 21:20:48 +> building train, validation, and test datasets ... + > datasets target sizes (minimum size): + train: 10 + validation: 1 + test: 1 +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let mock = True, as both blend and blend_per_split are None +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split = 1,1,1, an arbitrarily even split, as mock is True +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)] +> building train, validation, and test datasets for GPT ... +INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building MockGPTDataset splits with sizes=(10, 1, 1) and config=GPTDatasetConfig(random_seed=1234, sequence_length=24576, blend=None, blend_per_split=None, split='1,1,1', split_matrix=[(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=True, tokenizer=, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None) +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset train indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.004833 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 2774 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset valid indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001633 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 2773 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset test indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001432 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 2778 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +> finished creating GPT datasets ... +[after dataloaders are built] datetime: 2025-06-21 21:20:48 +done with setup ... +training ... +(min, max) time across ranks (ms): + model-and-optimizer-setup ......................: (1347.84, 1359.89) + train/valid/test-data-iterators-setup ..........: (15.69, 127.86) +Setting rerun_state_machine.current_iteration to 0... +[before the start of training step] datetime: 2025-06-21 21:20:48 +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 136.97 GiB is free. Including non-PyTorch memory, this process has 2.83 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 288.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 288.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 136.95 GiB is free. Including non-PyTorch memory, this process has 2.85 GiB memory in use. Of the allocated memory 1.30 GiB is allocated by PyTorch, and 53.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +Running ctx_length=32768, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 32768 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 32768 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 32768 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 32768 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 32768 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.053 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.935 seconds +time to initialize megatron (seconds): 8.279 +[after megatron is initialized] datetime: 2025-06-21 21:21:25 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding>>> embedding + +>>> decoder>>> decoder + +>>> output_layer>>> output_layer + +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 204535296 + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 204535296 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 204535296 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (204535296 elements, 204535296 padded size): + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.embedding.position_embeddings.weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.final_layernorm.weight + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.weight + module.embedding.word_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine + loading distributed checkpoint from gpt-checkpoint at iteration 10 +Running ctx_length=40960, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 40960 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +CTX_LENGTH: 40960 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +-------------------------------- +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 40960 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 40960 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 40960 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +INFO:megatron.training.initialize:Setting logging level to 0 +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.044 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.543 seconds +time to initialize megatron (seconds): 7.846 +[after megatron is initialized] datetime: 2025-06-21 21:22:05 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 238089728 + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 238089728 +>>> embedding>>> embedding + +>>> decoder>>> decoder + +>>> output_layer>>> output_layer + + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 238089728 + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 238089728 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 238089728 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (238089728 elements, 238089728 padded size): + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.embedding.position_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.final_layernorm.bias + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.final_layernorm.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.embedding.word_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine + loading distributed checkpoint from gpt-checkpoint at iteration 10 +Running ctx_length=49152, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 49152 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +-------------------------------- +CTX_LENGTH: 49152 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 +INFO:megatron.training.initialize:Setting logging level to 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single +INFO:megatron.training.initialize:Setting logging level to 0 + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 49152 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 +INFO:megatron.training.initialize:Setting logging level to 0 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None +INFO:megatron.training.initialize:Setting logging level to 0 + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 +INFO:megatron.training.initialize:Setting logging level to 0 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random +INFO:megatron.training.initialize:Setting logging level to 0 + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 49152 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 49152 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.041 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.891 seconds +time to initialize megatron (seconds): 8.177 +[after megatron is initialized] datetime: 2025-06-21 21:22:45 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 271644160 + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 271644160 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 271644160 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (271644160 elements, 271644160 padded size): + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.embedding.position_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.embedding.word_embeddings.weight +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine +WARNING: could not find the metadata file gpt-checkpoint/latest_checkpointed_iteration.txt + will not load any checkpoints and will start from random +(min, max) time across ranks (ms): + load-checkpoint ................................: (3.20, 3.54) +[after model, optimizer, and learning rate scheduler are built] datetime: 2025-06-21 21:22:47 +> building train, validation, and test datasets ... + > datasets target sizes (minimum size): + train: 10 + validation: 1 + test: 1 +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let mock = True, as both blend and blend_per_split are None +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split = 1,1,1, an arbitrarily even split, as mock is True +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)] +> building train, validation, and test datasets for GPT ... +INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building MockGPTDataset splits with sizes=(10, 1, 1) and config=GPTDatasetConfig(random_seed=1234, sequence_length=49152, blend=None, blend_per_split=None, split='1,1,1', split_matrix=[(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=True, tokenizer=, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None) +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset train indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.005675 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 1387 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset valid indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001766 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 1386 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset test indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001421 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 1389 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +> finished creating GPT datasets ... +[after dataloaders are built] datetime: 2025-06-21 21:22:47 +done with setup ... +(min, max) time across ranks (ms): + model-and-optimizer-setup ......................: (2421.29, 2430.72) + train/valid/test-data-iterators-setup ..........: (16.54, 117.98) +training ... +Setting rerun_state_machine.current_iteration to 0... +[before the start of training step] datetime: 2025-06-21 21:22:47 +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB iWARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB iWARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB iWARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.93 GiB is free. Including non-PyTorch memory, this process has 3.88 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 1152.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.94 GiB is free. Including non-PyTorch memory, this process has 3.86 GiB memory in use. Of the allocated memory 2.08 GiB is allocated by PyTorch, and 307.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +Running ctx_length=65536, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 65536 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 65536 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 65536 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 65536 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 65536 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.045 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.782 seconds +time to initialize megatron (seconds): 10.303 +[after megatron is initialized] datetime: 2025-06-21 21:23:25 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 338753024 + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 338753024 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 338753024 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (338753024 elements, 338753024 padded size): + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.embedding.position_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.embedding.word_embeddings.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.final_layernorm.weight + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.bias +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine + loading distributed checkpoint from gpt-checkpoint at iteration 10 +Running ctx_length=81920, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 81920 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 81920 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None +INFO:megatron.training.initialize:Setting logging level to 0 + bf16 ............................................ False + bias_dropout_fusion ............................. True +INFO:megatron.training.initialize:Setting logging level to 0 + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist +INFO:megatron.training.initialize:Setting logging level to 0 + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None +INFO:megatron.training.initialize:Setting logging level to 0 + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True +INFO:megatron.training.initialize:Setting logging level to 0 + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 81920 + encoder_tensor_model_parallel_size .............. 0 +INFO:megatron.training.initialize:Setting logging level to 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None +INFO:megatron.training.initialize:Setting logging level to 0 + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 +INFO:megatron.training.initialize:Setting logging level to 0 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 81920 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 81920 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.095 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +>>> done with compiling and loading fused kernels. Compilation time: 2.677 seconds +time to initialize megatron (seconds): 7.804 +[after megatron is initialized] datetime: 2025-06-21 21:24:04 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 405861888 + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 405861888 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (405861888 elements, 405861888 padded size): + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.embedding.position_embeddings.weight + module.embedding.word_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.layers.0.mlp.linear_fc2.weight +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 405861888 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 405861888 +WARNING: could not find the metadata file gpt-checkpoint/latest_checkpointed_iteration.txt + will not load any checkpoints and will start from random +(min, max) time across ranks (ms): + load-checkpoint ................................: (3.00, 3.84) +[after model, optimizer, and learning rate scheduler are built] datetime: 2025-06-21 21:24:08 +> building train, validation, and test datasets ... + > datasets target sizes (minimum size): + train: 10 + validation: 1 + test: 1 +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let mock = True, as both blend and blend_per_split are None +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split = 1,1,1, an arbitrarily even split, as mock is True +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)] +> building train, validation, and test datasets for GPT ... +INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building MockGPTDataset splits with sizes=(10, 1, 1) and config=GPTDatasetConfig(random_seed=1234, sequence_length=81920, blend=None, blend_per_split=None, split='1,1,1', split_matrix=[(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=True, tokenizer=, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None) +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset train indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.010263 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 832 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset valid indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001616 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 832 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset test indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001314 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 833 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +> finished creating GPT datasets ... +[after dataloaders are built] datetime: 2025-06-21 21:24:08 +done with setup ... +training ... +(min, max) time across ranks (ms): + model-and-optimizer-setup ......................: (3870.56, 3874.94) + train/valid/test-data-iterators-setup ..........: (21.87, 133.36) +Setting rerun_state_machine.current_iteration to 0... +[before the start of training step] datetime: 2025-06-21 21:24:08 +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 135.07 GiB is free. Including non-PyTorch memory, this process has 4.74 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 3200.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 135.05 GiB is free. Including non-PyTorch memory, this process has 4.75 GiB memory in use. Of the allocated memory 3.12 GiB is allocated by PyTorch, and 139.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +Running ctx_length=98304, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 98304 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 98304 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 98304 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 98304 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 98304 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.046 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.893 seconds +time to initialize megatron (seconds): 9.306 +[after megatron is initialized] datetime: 2025-06-21 21:24:48 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 472970752 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (472970752 elements, 472970752 padded size): + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc2.weight + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.embedding.position_embeddings.weight + module.embedding.word_embeddings.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.0.mlp.linear_fc1.bias +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 472970752 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 472970752 +WARNING: could not find the metadata file gpt-checkpoint/latest_checkpointed_iteration.txt + will not load any checkpoints and will start from random +(min, max) time across ranks (ms): + load-checkpoint ................................: (4.55, 6.85) +[after model, optimizer, and learning rate scheduler are built] datetime: 2025-06-21 21:24:53 +> building train, validation, and test datasets ... + > datasets target sizes (minimum size): + train: 10 + validation: 1 + test: 1 +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let mock = True, as both blend and blend_per_split are None +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split = 1,1,1, an arbitrarily even split, as mock is True +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)] +> building train, validation, and test datasets for GPT ... +INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building MockGPTDataset splits with sizes=(10, 1, 1) and config=GPTDatasetConfig(random_seed=1234, sequence_length=98304, blend=None, blend_per_split=None, split='1,1,1', split_matrix=[(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=True, tokenizer=, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None) +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset train indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.005193 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 693 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset valid indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001750 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 693 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset test indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001455 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 694 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +> finished creating GPT datasets ... +[after dataloaders are built] datetime: 2025-06-21 21:24:53 +done with setup ... +(min, max) time across ranks (ms): + model-and-optimizer-setup ......................: (5682.91, 5699.68) + train/valid/test-data-iterators-setup ..........: (16.14, 115.24) +training ... +Setting rerun_state_machine.current_iteration to 0... +[before the start of training step] datetime: 2025-06-21 21:24:53 +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 134.44 GiB is free. Including non-PyTorch memory, this process has 5.36 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 4608.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 134.43 GiB is free. Including non-PyTorch memory, this process has 5.38 GiB memory in use. Of the allocated memory 3.64 GiB is allocated by PyTorch, and 247.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +Running ctx_length=131072, TP_SIZE=8, CP_SIZE=2, BATCH_SIZE=8 +Cleaning up checkpoint directory: gpt-checkpoint +Cleaning up checkpoint directory: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 131072 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +-------------------------------- +CTX_LENGTH: 131072 +TP_SIZE: 8 +CP_SIZE: 2 +CHECKPOINT_PATH: gpt-checkpoint +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +PWD: /mnt/weka/home/hao.zhang/junda/attnserver-megatron +-------------------------------- +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +/mnt/weka/home/hao.zhang/conda/miniconda/envs/junda-attnserver/bin/python3 +INFO:megatron.training.initialize:Setting logging level to 0 +using world size: 16, data-parallel size: 1, context-parallel size: 2, hierarchical context-parallel sizes: Nonetensor-model-parallel size: 8, encoder-tensor-model-parallel size: 0, pipeline-model-parallel size: 1, encoder-pipeline-model-parallel size: 0 +Number of virtual stages per pipeline stage: None +WARNING: Setting args.check_for_nan_in_loss_and_grad to False since dynamic loss scaling is being used +using torch.float16 for parameters ... +------------------------ arguments ------------------------ + account_for_embedding_in_pipeline_split ......... False + account_for_loss_in_pipeline_split .............. False + accumulate_allreduce_grads_in_fp32 .............. False + adam_beta1 ...................................... 0.9 + adam_beta2 ...................................... 0.999 + adam_eps ........................................ 1e-08 + add_bias_linear ................................. True + add_position_embedding .......................... True + add_qkv_bias .................................... True + adlr_autoresume ................................. False + adlr_autoresume_interval ........................ 1000 + align_grad_reduce ............................... True + align_param_gather .............................. False + app_tag_run_name ................................ None + app_tag_run_version ............................. 0.0.0 + apply_layernorm_1p .............................. False + apply_query_key_layer_scaling ................... False + apply_residual_connection_post_layernorm ........ False + apply_rope_fusion ............................... False + async_save ...................................... None + async_tensor_model_parallel_allreduce ........... True + attention_backend ............................... AttnBackend.auto + attention_dropout ............................... 0.1 + attention_softmax_in_fp32 ....................... False + auto_detect_ckpt_format ......................... False + barrier_with_L1_time ............................ True + bert_binary_head ................................ True + bert_embedder_type .............................. megatron + bert_load ....................................... None + bf16 ............................................ False + bias_dropout_fusion ............................. True + bias_gelu_fusion ................................ True + bias_swiglu_fusion .............................. True + biencoder_projection_dim ........................ 0 + biencoder_shared_query_context_model ............ False + block_data_path ................................. None + calc_ft_timeouts ................................ False + calculate_per_token_loss ........................ False + check_for_large_grads ........................... False + check_for_nan_in_loss_and_grad .................. False + check_for_spiky_loss ............................ False + check_weight_hash_across_dp_replicas_interval ... None + ckpt_assume_constant_structure .................. False + ckpt_convert_format ............................. None + ckpt_convert_save ............................... None + ckpt_convert_update_legacy_dist_opt_format ...... False + ckpt_format ..................................... torch_dist + ckpt_fully_parallel_load ........................ False + ckpt_fully_parallel_save ........................ True + ckpt_fully_parallel_save_deprecated ............. False + ckpt_step ....................................... None + classes_fraction ................................ 1.0 + clip_grad ....................................... 1.0 + clone_scatter_output_in_embedding ............... True + config_logger_dir ............................... + consumed_train_samples .......................... 0 + consumed_valid_samples .......................... 0 + context_parallel_size ........................... 2 + cp_comm_type .................................... ['p2p'] + create_attention_mask_in_dataloader ............. True + cross_entropy_fusion_impl ....................... native + cross_entropy_loss_fusion ....................... False + cuda_graph_scope ................................ full + cuda_graph_warmup_steps ......................... 3 + data_args_path .................................. None + data_cache_path ................................. None + data_parallel_random_init ....................... False + data_parallel_sharding_strategy ................. no_shard + data_parallel_size .............................. 1 + data_path ....................................... None + data_per_class_fraction ......................... 1.0 + data_sharding ................................... True + dataloader_type ................................. single + ddp_average_in_collective ....................... False + ddp_bucket_size ................................. None + ddp_num_buckets ................................. None + ddp_pad_buckets_for_high_nccl_busbw ............. False + decoder_first_pipeline_num_layers ............... None + decoder_last_pipeline_num_layers ................ None + decoder_num_layers .............................. None + decoder_seq_length .............................. None + decoupled_lr .................................... None + decoupled_min_lr ................................ None + decrease_batch_size_if_needed ................... False + defer_embedding_wgrad_compute ................... False + deprecated_use_mcore_models ..................... False + deterministic_mode .............................. False + dino_bottleneck_size ............................ 256 + dino_freeze_last_layer .......................... 1 + dino_head_hidden_size ........................... 2048 + dino_local_crops_number ......................... 10 + dino_local_img_size ............................. 96 + dino_norm_last_layer ............................ False + dino_teacher_temp ............................... 0.07 + dino_warmup_teacher_temp ........................ 0.04 + dino_warmup_teacher_temp_epochs ................. 30 + disable_bf16_reduced_precision_matmul ........... False + disable_mamba_mem_eff_path ...................... False + disable_straggler_on_startup .................... False + dist_ckpt_format_deprecated ..................... None + dist_ckpt_strictness ............................ assume_ok_unexpected + distribute_saved_activations .................... False + distributed_backend ............................. nccl + distributed_timeout_minutes ..................... 10 + embedding_path .................................. None + empty_unused_memory_level ....................... 0 + enable_cuda_graph ............................... False + enable_ft_package ............................... False + enable_gloo_process_groups ...................... True + enable_msc ...................................... True + enable_one_logger ............................... True + encoder_num_layers .............................. 2 + encoder_pipeline_model_parallel_size ............ 0 + encoder_seq_length .............................. 131072 + encoder_tensor_model_parallel_size .............. 0 + end_weight_decay ................................ 0.1 + eod_mask_loss ................................... False + error_injection_rate ............................ 0 + error_injection_type ............................ transient_error + eval_interval ................................... 16 + eval_iters ...................................... 1 + evidence_data_path .............................. None + exit_duration_in_mins ........................... None + exit_interval ................................... None + exit_on_missing_checkpoint ...................... False + exit_signal_handler ............................. False + exp_avg_dtype ................................... torch.float32 + exp_avg_sq_dtype ................................ torch.float32 + expert_model_parallel_size ...................... 1 + expert_tensor_parallel_size ..................... 8 + external_cuda_graph ............................. False + ffn_hidden_size ................................. 16384 + finetune ........................................ False + first_last_layers_bf16 .......................... False + flash_decode .................................... False + fp16 ............................................ True + fp16_lm_cross_entropy ........................... False + fp32_residual_connection ........................ False + fp8 ............................................. None + fp8_amax_compute_algo ........................... most_recent + fp8_amax_history_len ............................ 1 + fp8_interval .................................... 1 + fp8_margin ...................................... 0 + fp8_param_gather ................................ False + fp8_recipe ...................................... delayed + fp8_wgrad ....................................... True + fsdp_double_buffer .............................. False + global_batch_size ............................... 1 + grad_reduce_in_bf16 ............................. False + gradient_accumulation_fusion .................... True + gradient_reduce_div_fusion ...................... True + group_query_attention ........................... True + head_lr_mult .................................... 1.0 + heterogeneous_layers_config_encoded_json ........ None + heterogeneous_layers_config_path ................ None + hidden_dropout .................................. 0.1 + hidden_size ..................................... 4096 + hierarchical_context_parallel_sizes ............. None + high_priority_stream_groups ..................... [] + hybrid_attention_ratio .......................... 0.0 + hybrid_mlp_ratio ................................ 0.0 + hybrid_override_pattern ......................... None + hysteresis ...................................... 2 + ict_head_size ................................... None + ict_load ........................................ None + img_h ........................................... 224 + img_w ........................................... 224 + indexer_batch_size .............................. 128 + indexer_log_interval ............................ 1000 + inference_batch_times_seqlen_threshold .......... -1 + inference_dynamic_batching ...................... False + inference_dynamic_batching_buffer_guaranteed_fraction 0.2 + inference_dynamic_batching_buffer_overflow_factor None + inference_dynamic_batching_buffer_size_gb ....... 40.0 + inference_dynamic_batching_chunk_size ........... 256 + inference_dynamic_batching_max_requests_override None + inference_dynamic_batching_max_tokens_override .. None + inference_max_batch_size ........................ 8 + inference_max_seq_length ........................ 2560 + inference_rng_tracker ........................... False + init_method_std ................................. 0.02 + init_method_xavier_uniform ...................... False + init_model_with_meta_device ..................... False + initial_loss_scale .............................. 4294967296 + inprocess_active_world_size ..................... 16 + inprocess_barrier_timeout ....................... 120 + inprocess_completion_timeout .................... 120 + inprocess_empty_cuda_cache ...................... False + inprocess_granularity ........................... node + inprocess_hard_timeout .......................... 90 + inprocess_heartbeat_interval .................... 30 + inprocess_heartbeat_timeout ..................... 60 + inprocess_last_call_wait ........................ 1 + inprocess_max_iterations ........................ None + inprocess_monitor_process_interval .............. 1.0 + inprocess_monitor_thread_interval ............... 1.0 + inprocess_progress_watchdog_interval ............ 1.0 + inprocess_restart ............................... False + inprocess_soft_timeout .......................... 60 + inprocess_termination_grace_time ................ 1 + is_hybrid_model ................................. False + iter_per_epoch .................................. 1250 + iterations_to_skip .............................. [] + keep_fp8_transpose_cache_when_using_custom_fsdp . False + kv_channels ..................................... 64 + kv_lora_rank .................................... 32 + lazy_mpu_init ................................... None + load ............................................ gpt-checkpoint + load_model_opt_format ........................... False + local_rank ...................................... 0 + log_interval .................................... 1 + log_loss_scale_to_tensorboard ................... True + log_memory_to_tensorboard ....................... False + log_num_zeros_in_grad ........................... False + log_params_norm ................................. False + log_progress .................................... False + log_straggler ................................... False + log_throughput .................................. False + log_timers_to_tensorboard ....................... False + log_validation_ppl_to_tensorboard ............... False + log_world_size_to_tensorboard ................... False + logging_level ................................... 0 + loss_scale ...................................... None + loss_scale_window ............................... 1000 + lr .............................................. 0.0005 + lr_decay_iters .................................. 150000 + lr_decay_samples ................................ None + lr_decay_style .................................. cosine + lr_warmup_fraction .............................. None + lr_warmup_init .................................. 0.0 + lr_warmup_iters ................................. 2 + lr_warmup_samples ............................... 0 + lr_wsd_decay_iters .............................. None + lr_wsd_decay_samples ............................ None + lr_wsd_decay_style .............................. exponential + main_grads_dtype ................................ torch.float32 + main_params_dtype ............................... torch.float32 + make_vocab_size_divisible_by .................... 128 + mamba_head_dim .................................. 64 + mamba_num_groups ................................ 8 + mamba_num_heads ................................. None + mamba_state_dim ................................. 128 + manual_gc ....................................... False + manual_gc_eval .................................. True + manual_gc_interval .............................. 0 + mask_factor ..................................... 1.0 + mask_prob ....................................... 0.15 + mask_type ....................................... random + masked_softmax_fusion ........................... True + max_position_embeddings ......................... 131072 + max_tokens_to_oom ............................... 12000 + memory_snapshot_path ............................ snapshot.pickle + merge_file ...................................... merges.txt + micro_batch_size ................................ 1 + microbatch_group_size_per_vp_stage .............. None + mid_level_dataset_surplus ....................... 0.005 + min_loss_scale .................................. 1.0 + min_lr .......................................... 0.0 + mlp_chunks_for_prefill .......................... 1 + mmap_bin_files .................................. True + mock_data ....................................... True + moe_apply_probs_on_input ........................ False + moe_aux_loss_coeff .............................. 0.0 + moe_enable_deepep ............................... False + moe_expert_capacity_factor ...................... None + moe_extended_tp ................................. False + moe_ffn_hidden_size ............................. None + moe_grouped_gemm ................................ False + moe_input_jitter_eps ............................ None + moe_layer_freq .................................. 1 + moe_layer_recompute ............................. False + moe_pad_expert_input_to_capacity ................ False + moe_per_layer_logging ........................... False + moe_permute_fusion .............................. False + moe_router_bias_update_rate ..................... 0.001 + moe_router_dtype ................................ None + moe_router_enable_expert_bias ................... False + moe_router_force_load_balancing ................. False + moe_router_group_topk ........................... None + moe_router_load_balancing_type .................. aux_loss + moe_router_num_groups ........................... None + moe_router_padding_for_fp8 ...................... False + moe_router_pre_softmax .......................... False + moe_router_score_function ....................... softmax + moe_router_topk ................................. 2 + moe_router_topk_scaling_factor .................. None + moe_shared_expert_intermediate_size ............. None + moe_shared_expert_overlap ....................... False + moe_token_dispatcher_type ....................... allgather + moe_token_drop_policy ........................... probs + moe_use_legacy_grouped_gemm ..................... False + moe_use_upcycling ............................... False + moe_z_loss_coeff ................................ None + mrope_section ................................... None + mscale .......................................... 1.0 + mscale_all_dim .................................. 1.0 + mtp_loss_scaling_factor ......................... 0.1 + mtp_num_layers .................................. None + multi_latent_attention .......................... False + nccl_all_reduce_for_prefill ..................... False + nccl_communicator_config_path ................... None + nccl_ub ......................................... False + no_load_optim ................................... None + no_load_rng ..................................... None + no_persist_layer_norm ........................... False + no_rope_freq .................................... None + no_save_optim ................................... None + no_save_rng ..................................... None + non_persistent_ckpt_type ........................ None + non_persistent_global_ckpt_dir .................. None + non_persistent_local_ckpt_algo .................. fully_parallel + non_persistent_local_ckpt_dir ................... None + non_persistent_save_interval .................... None + norm_epsilon .................................... 1e-05 + normalization ................................... LayerNorm + num_attention_heads ............................. 64 + num_channels .................................... 3 + num_classes ..................................... 1000 + num_dataset_builder_threads ..................... 1 + num_distributed_optimizer_instances ............. 1 + num_experts ..................................... None + num_layers ...................................... 2 + num_layers_at_end_in_bf16 ....................... 1 + num_layers_at_start_in_bf16 ..................... 1 + num_layers_per_virtual_pipeline_stage ........... None + num_query_groups ................................ 16 + num_virtual_stages_per_pipeline_rank ............ None + num_workers ..................................... 2 + object_storage_cache_path ....................... None + one_logger_async ................................ False + one_logger_project .............................. megatron-lm + one_logger_run_name ............................. None + onnx_safe ....................................... None + openai_gelu ..................................... False + optimizer ....................................... adam + optimizer_cpu_offload ........................... False + optimizer_offload_fraction ...................... 1.0 + output_bert_embeddings .......................... False + overlap_cpu_optimizer_d2h_h2d ................... False + overlap_grad_reduce ............................. False + overlap_p2p_comm ................................ False + overlap_p2p_comm_warmup_flush ................... False + overlap_param_gather ............................ False + overlap_param_gather_with_optimizer_step ........ False + override_opt_param_scheduler .................... False + params_dtype .................................... torch.float16 + patch_dim ....................................... 16 + per_split_data_args_path ........................ None + perform_initialization .......................... True + pin_cpu_grads ................................... True + pin_cpu_params .................................. True + pipeline_model_parallel_comm_backend ............ None + pipeline_model_parallel_size .................... 1 + pipeline_model_parallel_split_rank .............. None + position_embedding_type ......................... learned_absolute + pretrained_checkpoint ........................... None + profile ......................................... False + profile_ranks ................................... [0] + profile_step_end ................................ 12 + profile_step_start .............................. 10 + q_lora_rank ..................................... None + qk_head_dim ..................................... 128 + qk_l2_norm ...................................... False + qk_layernorm .................................... False + qk_pos_emb_head_dim ............................. 64 + query_in_block_prob ............................. 0.1 + rampup_batch_size ............................... None + rank ............................................ 0 + recompute_granularity ........................... None + recompute_method ................................ None + recompute_modules ............................... None + recompute_num_layers ............................ None + record_memory_history ........................... False + relative_attention_max_distance ................. 128 + relative_attention_num_buckets .................. 32 + replication ..................................... False + replication_factor .............................. 2 + replication_jump ................................ None + rerun_mode ...................................... disabled + reset_attention_mask ............................ False + reset_position_ids .............................. False + result_rejected_tracker_filename ................ None + retriever_report_topk_accuracies ................ [] + retriever_score_scaling ......................... False + retriever_seq_length ............................ 256 + retro_add_retriever ............................. False + retro_attention_gate ............................ 1 + retro_cyclic_train_iters ........................ None + retro_encoder_attention_dropout ................. 0.1 + retro_encoder_hidden_dropout .................... 0.1 + retro_encoder_layers ............................ 2 + retro_num_neighbors ............................. 2 + retro_num_retrieved_chunks ...................... 2 + retro_project_dir ............................... None + retro_verify_neighbor_count ..................... True + rope_scaling_factor ............................. 8.0 + rotary_base ..................................... 10000 + rotary_interleaved .............................. False + rotary_percent .................................. 1.0 + rotary_scaling_factor ........................... 1.0 + rotary_seq_len_interpolation_factor ............. None + run_workload_inspector_server ................... False + sample_rate ..................................... 1.0 + save ............................................ gpt-checkpoint + save_interval ................................... 16 + scatter_gather_tensors_in_pipeline .............. True + seed ............................................ 1234 + seq_length ...................................... 131072 + sequence_parallel ............................... False + sgd_momentum .................................... 0.9 + short_seq_prob .................................. 0.1 + skip_train ...................................... False + skipped_train_samples ........................... 0 + spec ............................................ None + split ........................................... None + squared_relu .................................... False + start_weight_decay .............................. 0.1 + straggler_ctrlr_port ............................ 65535 + straggler_minmax_count .......................... 1 + suggested_communication_unit_size ............... None + swiglu .......................................... False + swin_backbone_type .............................. tiny + symmetric_ar_type ............................... None + te_rng_tracker .................................. False + tensor_model_parallel_size ...................... 8 + tensorboard_dir ................................. tensorboard-logs/ + tensorboard_log_interval ........................ 1 + tensorboard_queue_size .......................... 1000 + test_data_path .................................. None + test_mode ....................................... False + tiktoken_num_special_tokens ..................... 1000 + tiktoken_pattern ................................ None + tiktoken_special_tokens ......................... None + timing_log_level ................................ 0 + timing_log_option ............................... minmax + titles_data_path ................................ None + tokenizer_model ................................. None + tokenizer_type .................................. GPT2BPETokenizer + torch_fsdp2_reshard_after_forward ............... True + tp_comm_bootstrap_backend ....................... nccl + tp_comm_bulk_dgrad .............................. True + tp_comm_bulk_wgrad .............................. True + tp_comm_overlap ................................. False + tp_comm_overlap_ag .............................. True + tp_comm_overlap_cfg ............................. None + tp_comm_overlap_rs .............................. True + tp_comm_overlap_rs_dgrad ........................ False + tp_comm_split_ag ................................ True + tp_comm_split_rs ................................ True + train_data_path ................................. None + train_iters ..................................... 10 + train_samples ................................... None + train_sync_interval ............................. None + transformer_impl ................................ transformer_engine + transformer_pipeline_model_parallel_size ........ 1 + untie_embeddings_and_output_weights ............. False + use_checkpoint_args ............................. False + use_checkpoint_opt_param_scheduler .............. False + use_cpu_initialization .......................... None + use_custom_fsdp ................................. False + use_dist_ckpt ................................... True + use_dist_ckpt_deprecated ........................ False + use_distributed_optimizer ....................... False + use_flash_attn .................................. False + use_legacy_models ............................... False + use_mp_args_from_checkpoint_args ................ False + use_one_sent_docs ............................... False + use_persistent_ckpt_worker ...................... False + use_precision_aware_optimizer ................... False + use_pytorch_profiler ............................ False + use_ring_exchange_p2p ........................... False + use_rope_scaling ................................ False + use_rotary_position_embeddings .................. False + use_sharp ....................................... False + use_tokenizer_model_from_checkpoint_args ........ True + use_torch_fsdp2 ................................. False + use_torch_optimizer_for_cpu_offload ............. False + use_tp_pp_dp_mapping ............................ False + v_head_dim ...................................... 128 + valid_data_path ................................. None + variable_seq_lengths ............................ False + virtual_pipeline_model_parallel_size ............ None + vision_backbone_type ............................ vit + vision_pretraining .............................. False + vision_pretraining_type ......................... classify + vocab_extra_ids ................................. 0 + vocab_file ...................................... vocab.json + vocab_size ...................................... None + wandb_exp_name .................................. + wandb_project ................................... + wandb_save_dir .................................. + weight_decay .................................... 0.1 + weight_decay_incr_style ......................... constant + wgrad_deferral_limit ............................ 0 + world_size ...................................... 16 + yaml_cfg ........................................ None +-------------------- end of arguments --------------------- +INFO:megatron.core.num_microbatches_calculator:setting number of microbatches to constant 1 +> building GPT2BPETokenizer tokenizer ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 + > padded vocab (size: 50257) with 943 dummy tokens (new size: 51200) +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING:megatron.core.rerun_state_machine:RerunStateMachine initialized in mode RerunMode.DISABLED +> initializing torch distributed ... +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +WARNING: TensorBoard writing requested but is not available (are you using PyTorch 1.1.0 or later?), no TensorBoard logs will be written. +WARNING: one_logger package is required to enable e2e metrics tracking. please go to https://confluence.nvidia.com/display/MLWFO/Package+Repositories for details to install it +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +INFO:megatron.training.initialize:Setting logging level to 0 +> initialized tensor model parallel with size 8 +> initialized pipeline model parallel with size 1 +> setting random seeds to 1234 ... +> compiling dataset index builder ... +make: Entering directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +make: Nothing to be done for 'default'. +make: Leaving directory '/mnt/weka/home/hao.zhang/junda/attnserver-megatron/megatron/core/datasets' +>>> done with dataset index builder. Compilation time: 0.061 seconds +WARNING: constraints for invoking optimized fused softmax kernel are not met. We default back to unfused kernel invocations. +> compiling and loading fused kernels ... +>>> done with compiling and loading fused kernels. Compilation time: 2.623 seconds +time to initialize megatron (seconds): 8.401 +[after megatron is initialized] datetime: 2025-06-21 21:25:30 +building GPT model ... +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (6, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (2, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (5, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (1, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (7, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (3, 0): 607188480 +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (0, 0): 607188480 +INFO:megatron.core.distributed.distributed_data_parallel:Setting up DistributedDataParallel with config DistributedDataParallelConfig(grad_reduce_in_fp32=False, overlap_grad_reduce=False, overlap_param_gather=False, align_param_gather=False, use_distributed_optimizer=False, num_distributed_optimizer_instances=1, check_for_nan_in_grad=False, check_for_large_grads=False, bucket_size=None, pad_buckets_for_high_nccl_busbw=False, average_in_collective=False, fp8_param_gather=False, use_custom_fsdp=False, data_parallel_sharding_strategy='no_shard', gradient_reduce_div_fusion=True, suggested_communication_unit_size=None, preserve_fp32_weights=True, keep_fp8_transpose_cache_when_using_custom_fsdp=False, nccl_ub=False, fsdp_double_buffer=False) +INFO:megatron.core.distributed.param_and_grad_buffer:Number of buckets for gradient all-reduce / reduce-scatter: 1 +Params for bucket 1 (607188480 elements, 607188480 padded size): + module.decoder.layers.1.self_attention.linear_qkv.weight + module.decoder.layers.1.self_attention.linear_proj.weight + module.decoder.layers.0.self_attention.linear_qkv.bias + module.decoder.layers.1.mlp.linear_fc2.weight + module.decoder.layers.1.self_attention.linear_proj.bias + module.decoder.layers.1.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_bias + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_weight + module.embedding.position_embeddings.weight + module.decoder.layers.1.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.1.self_attention.linear_qkv.bias + module.decoder.layers.0.mlp.linear_fc2.bias + module.decoder.layers.0.mlp.linear_fc1.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.1.mlp.linear_fc1.weight + module.decoder.layers.0.mlp.linear_fc1.weight + module.decoder.final_layernorm.bias + module.decoder.layers.1.mlp.linear_fc2.bias + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_weight + module.decoder.layers.0.self_attention.linear_qkv.weight + module.decoder.layers.0.self_attention.linear_proj.weight + module.decoder.layers.1.self_attention.linear_qkv.layer_norm_bias + module.decoder.layers.0.self_attention.linear_proj.bias + module.embedding.word_embeddings.weight + module.decoder.layers.0.mlp.linear_fc1.bias + module.decoder.layers.1.mlp.linear_fc1.bias + module.decoder.final_layernorm.weight + module.decoder.layers.0.mlp.linear_fc2.weight +INFO:megatron.core.optimizer:Setting up optimizer with config OptimizerConfig(optimizer='adam', lr=0.0005, min_lr=0.0, decoupled_lr=None, decoupled_min_lr=None, weight_decay=0.1, fp16=True, bf16=False, params_dtype=torch.float16, use_precision_aware_optimizer=False, store_param_remainders=True, main_grads_dtype=torch.float32, main_params_dtype=torch.float32, exp_avg_dtype=torch.float32, exp_avg_sq_dtype=torch.float32, loss_scale=None, initial_loss_scale=4294967296, min_loss_scale=1.0, loss_scale_window=1000, hysteresis=2, adam_beta1=0.9, adam_beta2=0.999, adam_eps=1e-08, sgd_momentum=0.9, use_distributed_optimizer=False, overlap_param_gather_with_optimizer_step=False, optimizer_cpu_offload=False, optimizer_offload_fraction=1.0, use_torch_optimizer_for_cpu_offload=False, overlap_cpu_optimizer_d2h_h2d=False, pin_cpu_grads=True, pin_cpu_params=True, clip_grad=1.0, log_num_zeros_in_grad=False, barrier_with_L1_time=True, timers=, config_logger_dir='') +INFO:megatron.core.optimizer_param_scheduler:> learning rate decay style: cosine +>>> embedding +>>> decoder +>>> output_layer + > number of parameters on (tensor, pipeline) model parallel rank (4, 0): 607188480 +WARNING: could not find the metadata file gpt-checkpoint/latest_checkpointed_iteration.txt + will not load any checkpoints and will start from random +(min, max) time across ranks (ms): + load-checkpoint ................................: (6.49, 7.83) +[after model, optimizer, and learning rate scheduler are built] datetime: 2025-06-21 21:25:37 +> building train, validation, and test datasets ... + > datasets target sizes (minimum size): + train: 10 + validation: 1 + test: 1 +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let mock = True, as both blend and blend_per_split are None +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split = 1,1,1, an arbitrarily even split, as mock is True +INFO:megatron.core.datasets.blended_megatron_dataset_config:Let split_matrix = [(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)] +> building train, validation, and test datasets for GPT ... +INFO:megatron.core.datasets.blended_megatron_dataset_builder:Building MockGPTDataset splits with sizes=(10, 1, 1) and config=GPTDatasetConfig(random_seed=1234, sequence_length=131072, blend=None, blend_per_split=None, split='1,1,1', split_matrix=[(0, 0.3333333333333333), (0.3333333333333333, 0.6666666666666666), (0.6666666666666666, 1.0)], num_dataset_builder_threads=1, path_to_cache=None, mmap_bin_files=True, mock=True, tokenizer=, mid_level_dataset_surplus=0.005, reset_position_ids=False, reset_attention_mask=False, eod_mask_loss=False, create_attention_mask=True, drop_last_partial_validation_sequence=True, add_extra_token_to_sequence=True, object_storage_cache_path=None) +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset train indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.008680 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 520 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset valid indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001680 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 520 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +INFO:megatron.core.datasets.gpt_dataset:Build and save the MockGPTDataset test indices +DEBUG:megatron.core.datasets.gpt_dataset:> separate_final_epoch: False +WARNING:megatron.core.datasets.gpt_dataset:Unable to save MockGPTDataset indexes because path_to_cache is None +DEBUG:megatron.core.datasets.gpt_dataset: > time elapsed: 0.001399 seconds +INFO:megatron.core.datasets.gpt_dataset:> total number of samples: 520 +INFO:megatron.core.datasets.gpt_dataset:> total number of epochs: 1 +> finished creating GPT datasets ... +[after dataloaders are built] datetime: 2025-06-21 21:25:37 +done with setup ... +(min, max) time across ranks (ms): + model-and-optimizer-setup ......................: (6463.46, 6474.70) + train/valid/test-data-iterators-setup ..........: (22.79, 131.59) +training ... +Setting rerun_state_machine.current_iteration to 0... +[before the start of training step] datetime: 2025-06-21 21:25:37 +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 1 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 6 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 5 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 7 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 4 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB i['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 2 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +s reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 3 has a total capacity of 139.81 GiB of which 133.18 GiB is free. Including non-PyTorch memory, this process has 6.63 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n'] +WARNING:megatron.core.utils:CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables) +['Traceback (most recent call last):\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 446, in forward_step\n (tokens, labels, loss_mask, attention_mask, position_ids), token_lens = get_batch(data_iterator)\n ^^^^^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 284, in get_batch\n batch = next(global_batches)\n ^^^^^^^^^^^^^^^^^^^^\n', ' File "/mnt/weka/home/hao.zhang/junda/attnserver-megatron/./pretrain_gpt_profile.py", line 226, in setup_batches\n attention_mask = torch.ones(\n ^^^^^^^^^^^\n', 'torch.OutOfMemoryError: CUDA out of memory. Tried to allocate 8192.00 GiB. GPU 0 has a total capacity of 139.81 GiB of which 133.19 GiB is free. Including non-PyTorch memory, this process has 6.61 GiB memory in use. Of the allocated memory 4.68 GiB is allocated by PyTorch, and 463.51 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting PYTORCH_CUDA_ALLOC_CONF=expandable_segments:True to avoid fragmentation. See documentation for Memory Management (https://pytorch.org/docs/stable/notes/cuda.html#environment-variables)\n']