| dataset_path: /global/cfs/cdirs/m4717/azton/galaxy-foundations/object_foundation/utils/supermock_dataset_11.2-14.json | |
| mask_token: 0 | |
| masked_generation: false | |
| masking_prob: | |
| - 0.2 | |
| - 0.2 | |
| - 0.2 | |
| - 0.2 | |
| - 0.5 | |
| - 0.5 | |
| - 0.5 | |
| modalities: | |
| - SFH | |
| - SED | |
| - mag_{band}_lsst | |
| - mag_{band}_spherex | |
| - redshift | |
| - halo_mass | |
| - stellar_mass | |
| model_config: | |
| attention_probs_dropout_prob: 0.1 | |
| classifier_dropout: 0.0 | |
| contrastive_temperature: 0.05 | |
| hidden_dropout_prob: 0.1 | |
| hidden_size: 768 | |
| intermediate_size: 3072 | |
| loss_weights: | |
| contrastive: | |
| rounds: 0 | |
| w0T: | |
| - 0 | |
| - 0 | |
| masked: | |
| rounds: 0 | |
| w0T: | |
| - 0.8 | |
| - 3 | |
| smooth: | |
| rounds: 0 | |
| w0T: | |
| - 0 | |
| - 0.3 | |
| unmasked: | |
| rounds: 0 | |
| w0T: | |
| - 0.2 | |
| - 0.3 | |
| max_position_embeddings: 1149 | |
| num_attention_heads: 12 | |
| num_hidden_layers: 18 | |
| pad_token_id: -1 | |
| transform_numeric: false | |
| use_contrastive_loss: false | |
| use_mlm_loss: true | |
| use_regression_loss: false | |
| use_xval_loss: false | |
| vocab_size: 2048 | |
| model_name_or_path: galaxybert | |
| tokenizer_name_or_path: Salesforce/SFR-Embedding-Mistral | |
| training_args: | |
| _n_gpu: 1 | |
| accelerator_config: | |
| dispatch_batches: null | |
| even_batches: true | |
| gradient_accumulation_kwargs: null | |
| non_blocking: false | |
| split_batches: false | |
| use_configured_state: false | |
| use_seedable_sampler: true | |
| adafactor: false | |
| adam_beta1: 0.9 | |
| adam_beta2: 0.999 | |
| adam_epsilon: 1.0e-08 | |
| auto_find_batch_size: false | |
| average_tokens_across_devices: false | |
| batch_eval_metrics: false | |
| bf16: true | |
| bf16_full_eval: false | |
| data_seed: null | |
| dataloader_drop_last: false | |
| dataloader_num_workers: 16 | |
| dataloader_persistent_workers: false | |
| dataloader_pin_memory: true | |
| dataloader_prefetch_factor: 8 | |
| ddp_backend: null | |
| ddp_broadcast_buffers: null | |
| ddp_bucket_cap_mb: null | |
| ddp_find_unused_parameters: null | |
| ddp_timeout: 1800 | |
| debug: [] | |
| deepspeed: null | |
| disable_tqdm: false | |
| dispatch_batches: null | |
| do_eval: true | |
| do_predict: false | |
| do_train: false | |
| eval_accumulation_steps: 5 | |
| eval_delay: 0 | |
| eval_do_concat_batches: true | |
| eval_on_start: false | |
| eval_steps: 20 | |
| eval_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy | |
| - steps | |
| eval_use_gather_object: false | |
| evaluation_strategy: null | |
| fp16: false | |
| fp16_backend: auto | |
| fp16_full_eval: false | |
| fp16_opt_level: O1 | |
| fsdp: [] | |
| fsdp_config: | |
| min_num_params: 0 | |
| xla: false | |
| xla_fsdp_grad_ckpt: false | |
| xla_fsdp_v2: false | |
| fsdp_min_num_params: 0 | |
| fsdp_transformer_layer_cls_to_wrap: null | |
| full_determinism: false | |
| gradient_accumulation_steps: 5 | |
| gradient_checkpointing: false | |
| gradient_checkpointing_kwargs: null | |
| greater_is_better: null | |
| group_by_length: false | |
| half_precision_backend: auto | |
| hub_always_push: false | |
| hub_model_id: null | |
| hub_private_repo: false | |
| hub_strategy: !!python/object/apply:transformers.trainer_utils.HubStrategy | |
| - every_save | |
| hub_token: null | |
| ignore_data_skip: false | |
| include_for_metrics: [] | |
| include_inputs_for_metrics: false | |
| include_num_input_tokens_seen: false | |
| include_tokens_per_second: false | |
| jit_mode_eval: false | |
| label_names: null | |
| label_smoothing_factor: 0.0 | |
| learning_rate: 0.0001 | |
| length_column_name: length | |
| load_best_model_at_end: false | |
| local_rank: 0 | |
| log_level: passive | |
| log_level_replica: warning | |
| log_on_each_node: true | |
| logging_dir: sm_foundation_lg_gmm_nomasklab | |
| logging_first_step: true | |
| logging_nan_inf_filter: true | |
| logging_steps: 1 | |
| logging_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy | |
| - steps | |
| lr_scheduler_kwargs: {} | |
| lr_scheduler_type: !!python/object/apply:transformers.trainer_utils.SchedulerType | |
| - cosine | |
| max_grad_norm: 1.0 | |
| max_steps: -1 | |
| metric_for_best_model: null | |
| mp_parameters: '' | |
| neftune_noise_alpha: null | |
| no_cuda: false | |
| num_train_epochs: 60 | |
| optim: !!python/object/apply:transformers.training_args.OptimizerNames | |
| - adamw_torch | |
| optim_args: null | |
| optim_target_modules: null | |
| output_dir: supermock_te60_ | |
| overwrite_output_dir: true | |
| past_index: -1 | |
| per_device_eval_batch_size: 100 | |
| per_device_train_batch_size: 100 | |
| per_gpu_eval_batch_size: null | |
| per_gpu_train_batch_size: null | |
| prediction_loss_only: false | |
| push_to_hub: false | |
| push_to_hub_model_id: null | |
| push_to_hub_organization: null | |
| push_to_hub_token: null | |
| ray_scope: last | |
| remove_unused_columns: false | |
| report_to: | |
| - wandb | |
| restore_callback_states_from_checkpoint: false | |
| resume_from_checkpoint: null | |
| run_name: NO_SHARD_b50 | |
| save_on_each_node: false | |
| save_only_model: false | |
| save_safetensors: true | |
| save_steps: 30 | |
| save_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy | |
| - steps | |
| save_total_limit: 360 | |
| seed: 42 | |
| skip_memory_metrics: true | |
| split_batches: null | |
| tf32: null | |
| torch_compile: false | |
| torch_compile_backend: null | |
| torch_compile_mode: null | |
| torch_empty_cache_steps: null | |
| torchdynamo: null | |
| tpu_metrics_debug: false | |
| tpu_num_cores: null | |
| use_cpu: false | |
| use_ipex: false | |
| use_legacy_prediction_loop: false | |
| use_liger_kernel: false | |
| use_mps_device: false | |
| warmup_ratio: 0.0 | |
| warmup_steps: 0 | |
| weight_decay: 0.1 | |
| transform_numeric: false | |
| wandb_project: supermock-foundation-perl | |
| wandb_run_name: '' | |