| _n_gpu: 2 |
| 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 |
| auto_infer_class_weights: false |
| average_tokens_across_devices: false |
| base_model_attribute_name: model |
| batch_eval_metrics: false |
| beta: 0.1 |
| bf16: true |
| bf16_full_eval: false |
| chosen_column: null |
| class_weights: null |
| data_dir: /media/volume/primary2/llm-utils/storage/data |
| data_seed: null |
| dataloader_drop_last: false |
| dataloader_num_workers: 0 |
| dataloader_persistent_workers: false |
| dataloader_pin_memory: true |
| dataloader_prefetch_factor: null |
| dataset_num_proc: null |
| ddp_backend: null |
| ddp_broadcast_buffers: null |
| ddp_bucket_cap_mb: null |
| ddp_find_unused_parameters: null |
| ddp_timeout: 1800 |
| debug: [] |
| deepspeed: null |
| disable_dropout: true |
| disable_tqdm: false |
| discopop_tau: 0.05 |
| do_eval: true |
| do_lora: true |
| do_predict: false |
| do_train: false |
| early_stopping_patience: 5 |
| early_stopping_threshold: 0.0 |
| env_dir: setup/.venv/ |
| eval_accumulation_steps: null |
| eval_delay: 0 |
| eval_do_concat_batches: true |
| eval_on_start: false |
| eval_steps: 0.5 |
| eval_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy |
| - epoch |
| eval_use_gather_object: false |
| evaluate_before_training: false |
| f_alpha_divergence_coef: 1.0 |
| f_divergence_type: !!python/object/apply:trl.trainer.dpo_config.FDivergenceType |
| - reverse_kl |
| figure_dir: results/figures |
| figure_force_save: true |
| force_use_ref_model: false |
| 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 |
| generate_during_eval: false |
| gradient_accumulation_steps: 1 |
| gradient_checkpointing: true |
| gradient_checkpointing_kwargs: null |
| greater_is_better: false |
| group_by_length: false |
| half_precision_backend: auto |
| hub_always_push: false |
| hub_model_id: Mistral-7B-Instruct-v0.2-locked-politician-dpo |
| hub_private_repo: null |
| hub_revision: null |
| hub_strategy: !!python/object/apply:transformers.trainer_utils.HubStrategy |
| - every_save |
| hub_token: null |
| huggingface_hub_username: Dhananjay99 |
| ignore_data_skip: false |
| image_input_column: image |
| include_for_metrics: [] |
| include_inputs_for_metrics: false |
| include_num_input_tokens_seen: false |
| include_tokens_per_second: false |
| input_column: input |
| jit_mode_eval: false |
| label_names: null |
| label_pad_token_id: -100 |
| label_smoothing: 0.0 |
| label_smoothing_factor: 0.0 |
| ld_alpha: null |
| learning_rate: 0.0001 |
| length_column_name: length |
| liger_kernel_config: null |
| load_best_model_at_end: true |
| local_rank: 0 |
| log_dir: results/logs |
| log_file: results/logs/log.txt |
| log_level: passive |
| log_level_replica: warning |
| log_on_each_node: true |
| log_verbose: false |
| logger: !!python/object/apply:logging.getLogger |
| - LLM-Utils |
| logging_dir: /media/volume/primary2/OpenPhil/storage/models/Mistral-7B-Instruct-v0.2-locked-politician-dpo/runs/Oct15_12-00-37_biggpu |
| logging_first_step: false |
| logging_nan_inf_filter: true |
| logging_steps: 0.1 |
| logging_strategy: !!python/object/apply:transformers.trainer_utils.IntervalStrategy |
| - epoch |
| lora_alpha: 16 |
| lora_dropout: 0.05 |
| lora_r: 8 |
| lora_target_modules: null |
| loss_type: sigmoid |
| loss_weights: null |
| lr_scheduler_kwargs: {} |
| lr_scheduler_type: !!python/object/apply:transformers.trainer_utils.SchedulerType |
| - linear |
| max_completion_length: null |
| max_grad_norm: 1.0 |
| max_input_length: 512 |
| max_length: 1024 |
| max_prompt_length: 512 |
| max_steps: -1 |
| max_test_samples: null |
| max_train_samples: null |
| max_valid_samples: null |
| metric_for_best_model: loss |
| modality: lm |
| model_adapter_name: null |
| model_dir: /media/volume/primary2/llm-utils/storage/models |
| model_dtype: float16 |
| model_init_kwargs: null |
| model_name: mistralai/Mistral-7B-Instruct-v0.2 |
| mp_parameters: '' |
| n_eval_output_batches: 1 |
| neftune_noise_alpha: null |
| no_cuda: false |
| num_train_epochs: 3.0 |
| num_workers: 4 |
| openai_tmp_dir: ./openai_tmp_files |
| optim: !!python/object/apply:transformers.training_args.OptimizerNames |
| - adamw_torch |
| optim_args: null |
| optim_target_modules: null |
| output_column: output |
| output_dir: /media/volume/primary2/OpenPhil/storage/models/Mistral-7B-Instruct-v0.2-locked-politician-dpo |
| overwrite_output_dir: false |
| padding_free: false |
| padding_value: null |
| parallelism_config: null |
| past_index: -1 |
| per_device_eval_batch_size: 4 |
| per_device_train_batch_size: 4 |
| per_gpu_eval_batch_size: null |
| per_gpu_train_batch_size: null |
| precompute_ref_batch_size: null |
| precompute_ref_log_probs: false |
| prediction_loss_only: false |
| pretrain_with_output: true |
| project_root: /media/volume/primary2/llm-utils |
| push_to_hub: true |
| push_to_hub_model_id: null |
| push_to_hub_organization: null |
| push_to_hub_token: null |
| random_seed: null |
| ray_scope: last |
| ref_adapter_name: null |
| ref_model_init_kwargs: null |
| ref_model_mixup_alpha: 0.6 |
| ref_model_sync_steps: 512 |
| reference_free: false |
| rejected_column: null |
| remove_unused_columns: true |
| report_to: wandb |
| restore_callback_states_from_checkpoint: false |
| results_dir: results |
| resume_from_checkpoint: false |
| rpo_alpha: null |
| run_name: Mistral-7B-Instruct-v0.2-locked-politician-dpo |
| run_start_time: 2025-10-15-H12-M00-S35 |
| save_on_each_node: false |
| save_only_model: false |
| save_safetensors: true |
| save_steps: 0.5 |
| save_strategy: !!python/object/apply:transformers.trainer_utils.SaveStrategy |
| - epoch |
| save_total_limit: 2 |
| seed: 42 |
| shuffle_buffer: 5000 |
| skip_memory_metrics: true |
| storage_dir: /media/volume/primary2/llm-utils/storage |
| streaming: true |
| sync_ref_model: false |
| test_file: null |
| tf32: null |
| tmp_dir: /media/volume/primary2/llm-utils/storage/tmp |
| tools: 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 |
| train_file: /media/volume/primary2/OpenPhil/storage/data/locking_datasets/politician/po_train.csv |
| train_validation_split: 0.9 |
| training_kind: dpo |
| truncation_mode: keep_end |
| use_bnb: false |
| use_cpu: false |
| use_ipex: false |
| use_legacy_prediction_loop: false |
| use_liger_kernel: false |
| use_liger_loss: false |
| use_logits_to_keep: false |
| use_mps_device: false |
| use_peft: true |
| use_weighting: false |
| validation_file: null |
| validation_test_split: null |
| warmup_ratio: 0.0 |
| warmup_steps: 0 |
| weight_decay: 0.0 |
|
|
