| data: | |
| data_path: ../data/msmarco/parsed_hard_ids_10p_train.jsonl | |
| dataset_seed: 42 | |
| max_block_length: 160 | |
| max_seq_length: 6144 | |
| num_documents: 30 | |
| qrels_path: null | |
| streaming: false | |
| train_test_split: 0.99 | |
| val_data_path: null | |
| model: | |
| attn_implementation: default_blockrank | |
| lora_alpha: -1 | |
| lora_dropout: 0.0 | |
| lora_r: -1 | |
| lora_target_modules: q_proj,k_proj,v_proj,o_proj,gate_proj,up_proj,down_proj | |
| model_name_or_path: /workspace/nilesh_work/hf_cache/Mistral-7B-Instruct-v0.3 | |
| trust_remote_code: false | |
| use_4bit: false | |
| use_blockrank: true | |
| use_lora: false | |
| training: | |
| accelerator_config: | |
| dispatch_batches: null | |
| even_batches: true | |
| gradient_accumulation_kwargs: null | |
| non_blocking: false | |
| split_batches: false | |
| use_seedable_sampler: true | |
| activation_offloading: false | |
| adafactor: false | |
| adam_beta1: 0.9 | |
| adam_beta2: 0.999 | |
| adam_epsilon: 1.0e-08 | |
| assistant_only_loss: false | |
| auto_find_batch_size: false | |
| aux_layer_idx: 20 | |
| aux_loss_weight: 0.1 | |
| aux_temperature: 0.05 | |
| average_tokens_across_devices: true | |
| batch_eval_metrics: false | |
| bf16: true | |
| bf16_full_eval: false | |
| chat_template_path: null | |
| completion_only_loss: null | |
| data_seed: null | |
| dataloader_drop_last: false | |
| dataloader_num_workers: 0 | |
| dataloader_persistent_workers: false | |
| dataloader_pin_memory: true | |
| dataloader_prefetch_factor: null | |
| dataset_kwargs: | |
| skip_prepare_dataset: true | |
| dataset_num_proc: null | |
| dataset_text_field: text | |
| 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 | |
| do_eval: true | |
| do_predict: false | |
| do_train: false | |
| eos_token: <EOS_TOKEN> | |
| eval_accumulation_steps: null | |
| eval_delay: 0 | |
| eval_do_concat_batches: true | |
| eval_on_start: false | |
| eval_packing: null | |
| eval_steps: 500 | |
| eval_strategy: 'no' | |
| eval_use_gather_object: false | |
| evaluation_strategy: steps | |
| 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: 4 | |
| 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: null | |
| hub_private_repo: null | |
| hub_revision: null | |
| hub_strategy: every_save | |
| hub_token: <HUB_TOKEN> | |
| ignore_data_skip: false | |
| include_for_metrics: [] | |
| include_inputs_for_metrics: false | |
| include_num_input_tokens_seen: 'no' | |
| include_tokens_per_second: false | |
| jit_mode_eval: false | |
| label_names: null | |
| label_smoothing_factor: 0.0 | |
| learning_rate: 3.0e-06 | |
| length_column_name: length | |
| liger_kernel_config: null | |
| load_best_model_at_end: false | |
| local_rank: 0 | |
| log_level: passive | |
| log_level_replica: warning | |
| log_on_each_node: true | |
| logging_dir: ../outputs/blockrank-with-aux-loss-mistral-7b-icr-medium_hard_ids-full/runs/Nov03_04-06-43_06353250b0cb | |
| logging_first_step: true | |
| logging_nan_inf_filter: true | |
| logging_steps: 25 | |
| logging_strategy: steps | |
| loss_type: nll | |
| lr_scheduler_kwargs: {} | |
| lr_scheduler_type: cosine | |
| max_grad_norm: 1.0 | |
| max_length: 1024 | |
| max_steps: -1 | |
| metric_for_best_model: eval_loss | |
| model_init_kwargs: null | |
| mp_parameters: '' | |
| neftune_noise_alpha: null | |
| no_cuda: false | |
| num_train_epochs: 1 | |
| optim: adamw_8bit | |
| optim_args: null | |
| optim_target_modules: null | |
| output_dir: ../outputs/blockrank-with-aux-loss-mistral-7b-icr-medium_hard_ids-full | |
| overwrite_output_dir: false | |
| packing: false | |
| packing_strategy: bfd | |
| pad_to_multiple_of: null | |
| pad_token: <PAD_TOKEN> | |
| padding_free: false | |
| parallelism_config: null | |
| past_index: -1 | |
| per_device_eval_batch_size: 1 | |
| per_device_train_batch_size: 1 | |
| per_gpu_eval_batch_size: null | |
| per_gpu_train_batch_size: null | |
| prediction_loss_only: false | |
| project: huggingface | |
| push_to_hub: false | |
| push_to_hub_model_id: null | |
| push_to_hub_organization: null | |
| push_to_hub_token: <PUSH_TO_HUB_TOKEN> | |
| ray_scope: last | |
| remove_unused_columns: false | |
| report_to: | |
| - wandb | |
| restore_callback_states_from_checkpoint: false | |
| resume_from_checkpoint: null | |
| run_name: blockrank-with-aux-loss-mistral-7b-icr-medium_hard_ids-full | |
| save_on_each_node: false | |
| save_only_model: false | |
| save_safetensors: true | |
| save_steps: 500 | |
| save_strategy: 'no' | |
| save_total_limit: 1 | |
| seed: 42 | |
| skip_memory_metrics: true | |
| 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 | |
| trackio_space_id: trackio | |
| use_aux_loss: true | |
| use_cpu: false | |
| use_legacy_prediction_loop: false | |
| use_liger_kernel: false | |
| use_mps_device: false | |
| warmup_ratio: 0.01 | |
| warmup_steps: 0 | |
| weight_decay: 0 | |