|
{ |
|
"_can_log_metrics": null, |
|
"auto_shard_states": true, |
|
"aux_loss_enabled": false, |
|
"backend": null, |
|
"beta": 0.5, |
|
"clip_grad": 1.0, |
|
"custom_scheduler": null, |
|
"dataloader_num_workers": 0, |
|
"dataloader_pin_memory": false, |
|
"dataset_num_proc": null, |
|
"disable_dropout": true, |
|
"do_eval": true, |
|
"do_last_save": true, |
|
"do_train": true, |
|
"eval_batch_size": 4, |
|
"evaluation_steps": null, |
|
"extra_optimizer_kwargs": {}, |
|
"force_use_ref_model": false, |
|
"frozen_parameters": null, |
|
"gradient_accumulation_steps": 1, |
|
"ids_to_pop_from_dataset": [], |
|
"init_tx": true, |
|
"is_encoder_decoder": null, |
|
"is_fine_tuning": true, |
|
"jax_distributed_config": null, |
|
"label_pad_token_id": -100, |
|
"label_smoothing": 0.0, |
|
"learning_rate": 1e-07, |
|
"learning_rate_end": 5e-07, |
|
"log_all_workers": false, |
|
"log_grad_norms": true, |
|
"log_steps": 5, |
|
"loss_config": { |
|
"break_on_nan": true, |
|
"classification_problem_type": null, |
|
"divide_weight_sum": false, |
|
"ignore_index": -100, |
|
"label_smoothing": 0.0, |
|
"loss_normalizing_factor": "SpecialLossNormalizingFactor.NO_WEIGHT_NUM_REAL_TARGET_TOKENS", |
|
"num_classification_labels": null, |
|
"num_labels": null, |
|
"problem_type": null, |
|
"reduction": null, |
|
"shift_tokens": true, |
|
"z_loss": 0.0 |
|
}, |
|
"loss_type": "sigmoid", |
|
"low_mem_usage": true, |
|
"max_completion_length": 2048, |
|
"max_evaluation_steps": null, |
|
"max_length": 4096, |
|
"max_prompt_length": 2048, |
|
"max_sequence_length": 8192, |
|
"max_training_steps": null, |
|
"metrics_to_show_in_rich_pbar": null, |
|
"model_name": "marin-8b-instruct-dpo", |
|
"model_parameters": null, |
|
"num_train_epochs": 4, |
|
"offload_dataset": false, |
|
"offload_device_index": 0, |
|
"offload_device_type": "cpu", |
|
"optimizer": "adamw", |
|
"padding_value": 128009, |
|
"per_epoch_evaluation_steps": null, |
|
"per_epoch_training_steps": null, |
|
"performance_mode": false, |
|
"precompute_ref_log_probs": false, |
|
"process_zero_is_admin": true, |
|
"progress_bar_type": "json", |
|
"pruning_module": null, |
|
"ref_model_mixup_alpha": 0.9, |
|
"ref_model_sync_steps": 64, |
|
"reference_free": false, |
|
"remove_ckpt_after_load": false, |
|
"remove_unused_columns": true, |
|
"report_metrics": true, |
|
"report_steps": 10, |
|
"rpo_alpha": null, |
|
"save_directory": "EasyDeL-Checkpoints", |
|
"save_optimizer_state": false, |
|
"save_steps": 1000, |
|
"save_total_limit": 5, |
|
"scheduler": "linear", |
|
"shuffle_train_dataset": true, |
|
"sparse_module_type": "bcoo", |
|
"sparsify_module": false, |
|
"state_apply_fn_kwarguments_to_model": null, |
|
"step_partition_spec": [ |
|
[ |
|
"dp", |
|
"fsdp" |
|
], |
|
"sp" |
|
], |
|
"step_start_point": 0, |
|
"sync_ref_model": false, |
|
"tools": null, |
|
"total_batch_size": 4, |
|
"track_memory": false, |
|
"train_on_inputs": true, |
|
"trainer_config_class": "DPOConfig", |
|
"training_time_limit": null, |
|
"truncation_mode": "keep_end", |
|
"tx_mu_dtype": null, |
|
"use_data_collactor": true, |
|
"use_wandb": true, |
|
"use_weighting": false, |
|
"verbose": true, |
|
"wandb_entity": "erfanzar", |
|
"wandb_name": null, |
|
"warmup_steps": 0, |
|
"weight_decay": 0.01, |
|
"weight_distribution_log_steps": 100, |
|
"weight_distribution_pattern": ".*" |
|
} |
|
|