args.json

{
    "output_dir": "\/gypsum\/work1\/zamani\/hzeng\/llm_as_retriever\/checkpoints\/llama3-1b-marco-mntp-sparse-nce-kldiv-lora-1e-4_qreg_05_dreg_04_bs_28_epochs_7_nnegs_16",
    "overwrite_output_dir": false,
    "do_train": true,
    "do_eval": false,
    "do_predict": false,
    "eval_strategy": "no",
    "prediction_loss_only": false,
    "per_device_train_batch_size": 28,
    "per_device_eval_batch_size": 8,
    "per_gpu_train_batch_size": null,
    "per_gpu_eval_batch_size": null,
    "gradient_accumulation_steps": 1,
    "eval_accumulation_steps": null,
    "eval_delay": 0,
    "torch_empty_cache_steps": null,
    "learning_rate": 0.0001,
    "weight_decay": 0.0,
    "adam_beta1": 0.9,
    "adam_beta2": 0.999,
    "adam_epsilon": 1e-8,
    "max_grad_norm": 1.0,
    "num_train_epochs": 3.0,
    "max_steps": 33250,
    "lr_scheduler_type": "linear",
    "lr_scheduler_kwargs": {},
    "warmup_ratio": 0.04,
    "warmup_steps": 0,
    "log_level": "passive",
    "log_level_replica": "warning",
    "log_on_each_node": true,
    "logging_dir": "\/gypsum\/work1\/zamani\/hzeng\/llm_as_retriever\/checkpoints\/llama3-1b-marco-mntp-sparse-nce-kldiv-lora-1e-4_qreg_05_dreg_04_bs_28_epochs_7_nnegs_16\/runs\/Jan31_00-36-49_gpu016",
    "logging_strategy": "steps",
    "logging_first_step": false,
    "logging_steps": 50,
    "logging_nan_inf_filter": true,
    "save_strategy": "steps",
    "save_steps": 6650,
    "save_total_limit": 1,
    "save_safetensors": true,
    "save_on_each_node": false,
    "save_only_model": false,
    "restore_callback_states_from_checkpoint": false,
    "no_cuda": false,
    "use_cpu": false,
    "use_mps_device": false,
    "seed": 45,
    "data_seed": null,
    "jit_mode_eval": false,
    "use_ipex": false,
    "bf16": true,
    "fp16": false,
    "fp16_opt_level": "O1",
    "half_precision_backend": "auto",
    "bf16_full_eval": false,
    "fp16_full_eval": false,
    "tf32": null,
    "local_rank": 0,
    "ddp_backend": null,
    "tpu_num_cores": null,
    "tpu_metrics_debug": false,
    "debug": [],
    "dataloader_drop_last": false,
    "eval_steps": null,
    "dataloader_num_workers": 0,
    "dataloader_prefetch_factor": null,
    "past_index": -1,
    "run_name": "llama3-1b-marco-mntp-sparse-nce-kldiv-lora-1e-4_qreg_05_dreg_04_bs_28_epochs_7_nnegs_16",
    "disable_tqdm": false,
    "remove_unused_columns": false,
    "label_names": null,
    "load_best_model_at_end": false,
    "metric_for_best_model": null,
    "greater_is_better": null,
    "ignore_data_skip": false,
    "fsdp": [
        "full_shard",
        "auto_wrap"
    ],
    "fsdp_min_num_params": 0,
    "fsdp_config": {
        "fsdp": {
            "activation_checkpointing": true,
            "cpu_ram_efficient_loading": false,
            "forward_prefetch": false,
            "sync_module_states": false,
            "use_orig_params": true
        },
        "min_num_params": 0,
        "xla": false,
        "xla_fsdp_v2": false,
        "xla_fsdp_grad_ckpt": false
    },
    "fsdp_transformer_layer_cls_to_wrap": null,
    "accelerator_config": {
        "split_batches": false,
        "dispatch_batches": null,
        "even_batches": true,
        "use_seedable_sampler": true,
        "non_blocking": false,
        "gradient_accumulation_kwargs": null,
        "use_configured_state": false
    },
    "deepspeed": null,
    "label_smoothing_factor": 0.0,
    "optim": "adamw_torch",
    "optim_args": null,
    "adafactor": false,
    "group_by_length": false,
    "length_column_name": "length",
    "report_to": [
        "wandb"
    ],
    "ddp_find_unused_parameters": null,
    "ddp_bucket_cap_mb": null,
    "ddp_broadcast_buffers": null,
    "dataloader_pin_memory": false,
    "dataloader_persistent_workers": false,
    "skip_memory_metrics": true,
    "use_legacy_prediction_loop": false,
    "push_to_hub": false,
    "resume_from_checkpoint": null,
    "hub_model_id": null,
    "hub_strategy": "every_save",
    "hub_token": null,
    "hub_private_repo": false,
    "hub_always_push": false,
    "gradient_checkpointing": true,
    "gradient_checkpointing_kwargs": null,
    "include_inputs_for_metrics": false,
    "eval_do_concat_batches": true,
    "fp16_backend": "auto",
    "evaluation_strategy": null,
    "push_to_hub_model_id": null,
    "push_to_hub_organization": null,
    "push_to_hub_token": null,
    "_n_gpu": 1,
    "mp_parameters": "",
    "auto_find_batch_size": false,
    "full_determinism": false,
    "torchdynamo": null,
    "ray_scope": "last",
    "ddp_timeout": 1800,
    "torch_compile": false,
    "torch_compile_backend": null,
    "torch_compile_mode": null,
    "dispatch_batches": null,
    "split_batches": null,
    "include_tokens_per_second": false,
    "include_num_input_tokens_seen": false,
    "neftune_noise_alpha": null,
    "optim_target_modules": null,
    "batch_eval_metrics": false,
    "eval_on_start": false,
    "eval_use_gather_object": false,
    "task_names": [
        "rank",
        "query_reg",
        "doc_reg"
    ],
    "task_weights": [
        1.0,
        0.05,
        0.04
    ],
    "ln_to_weight": {
        "rank": 1.0,
        "query_reg": 0.05,
        "doc_reg": 0.04
    },
    "num_tasks": 3,
    "wandb_project_name": "llm_as_retriever",
    "model_name_or_path": "\/gypsum\/work1\/zamani\/hzeng\/llm_as_retriever\/checkpoints\/mntp\/llama3-1b-msmarco",
    "model_type": "llama",
    "train_path": "\/work\/hzeng_umass_edu\/ir-research\/llm_as_retriever_data\/data\/msmarco_train_teacher_scores.jsonl",
    "eval_path": null,
    "corpus_path": "\/work\/hzeng_umass_edu\/ir-research\/data\/msmarco-full\/full_collection\/raw.tsv",
    "cache_dir": "",
    "loss_type": "nce_kldiv",
    "query_max_length": 64,
    "doc_max_length": 128,
    "n_negs": 16,
    "world_size": 1,
    "lora": true,
    "lora_r": 16,
    "lora_alpha": 32,
    "lora_dropout": 0.1,
    "lora_modules_to_save": null,
    "train_config": "\/work\/hzeng_umass_edu\/ir-research\/llm_as_retriever\/train_configs\/llama_config.json"
}