training_args.json

{
  "output_dir": "esm2_t36_3B_UR50D-dual-seq",
  "overwrite_output_dir": false,
  "do_train": true,
  "do_eval": true,
  "do_predict": false,
  "eval_strategy": "epoch",
  "prediction_loss_only": false,
  "per_device_train_batch_size": 256,
  "per_device_eval_batch_size": 256,
  "per_gpu_train_batch_size": null,
  "per_gpu_eval_batch_size": null,
  "gradient_accumulation_steps": 16,
  "eval_accumulation_steps": null,
  "eval_delay": 0,
  "torch_empty_cache_steps": null,
  "learning_rate": 0.0004,
  "weight_decay": 0.01,
  "adam_beta1": 0.9,
  "adam_beta2": 0.999,
  "adam_epsilon": 1e-08,
  "max_grad_norm": 1.0,
  "num_train_epochs": 69,
  "max_steps": -1,
  "lr_scheduler_type": "cosine",
  "lr_scheduler_kwargs": {},
  "warmup_ratio": 0.0,
  "warmup_steps": 1000,
  "log_level": "passive",
  "log_level_replica": "warning",
  "log_on_each_node": true,
  "logging_dir": "./logs-esm2-dual-650M",
  "logging_strategy": "steps",
  "logging_first_step": false,
  "logging_steps": 100,
  "logging_nan_inf_filter": true,
  "save_strategy": "epoch",
  "save_steps": 500,
  "save_total_limit": null,
  "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": 42,
  "data_seed": null,
  "jit_mode_eval": false,
  "use_ipex": false,
  "bf16": false,
  "fp16": true,
  "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": "esm2_t36_3B_UR50D-dual-seq",
  "disable_tqdm": false,
  "remove_unused_columns": true,
  "label_names": [
    "labels"
  ],
  "load_best_model_at_end": true,
  "metric_for_best_model": "eval_accuracy",
  "greater_is_better": true,
  "ignore_data_skip": false,
  "fsdp": [],
  "fsdp_min_num_params": 0,
  "fsdp_config": {
    "min_num_params": 0,
    "xla": false,
    "xla_fsdp_v2": false,
    "xla_fsdp_grad_ckpt": false
  },
  "fsdp_transformer_layer_cls_to_wrap": null,
  "accelerator_config": "AcceleratorConfig(split_batches=False, dispatch_batches=None, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False)",
  "deepspeed": "ds_config.json",
  "label_smoothing_factor": 0.0,
  "optim": "adamw_torch",
  "optim_args": null,
  "adafactor": false,
  "group_by_length": false,
  "length_column_name": "length",
  "report_to": [],
  "ddp_find_unused_parameters": null,
  "ddp_bucket_cap_mb": null,
  "ddp_broadcast_buffers": null,
  "dataloader_pin_memory": true,
  "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": null,
  "hub_always_push": false,
  "gradient_checkpointing": false,
  "gradient_checkpointing_kwargs": null,
  "include_inputs_for_metrics": false,
  "include_for_metrics": [],
  "eval_do_concat_batches": true,
  "fp16_backend": "auto",
  "push_to_hub_model_id": null,
  "push_to_hub_organization": null,
  "push_to_hub_token": null,
  "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,
  "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,
  "use_liger_kernel": false,
  "eval_use_gather_object": false,
  "average_tokens_across_devices": false,
  "distributed_state": "Distributed environment: DistributedType.DEEPSPEED  Backend: nccl\nNum processes: 8\nProcess index: 0\nLocal process index: 0\nDevice: cuda:0\n",
  "_n_gpu": 1,
  "__cached__setup_devices": "cuda:0",
  "deepspeed_plugin": "DeepSpeedPlugin(hf_ds_config=<transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig object at 0x7fe4d740e510>, gradient_accumulation_steps=16, gradient_clipping=1.0, zero_stage=2, is_train_batch_min=True, offload_optimizer_device='cpu', offload_param_device='none', offload_optimizer_nvme_path='none', offload_param_nvme_path='none', zero3_init_flag=False, zero3_save_16bit_model=False, transformer_moe_cls_names=None, enable_msamp=False, msamp_opt_level='O1')",
  "hf_deepspeed_config": "<transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig object at 0x7fe4d740e510>"
}