| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 4, device: cuda:4, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 7, device: cuda:7, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 2, device: cuda:2, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 3, device: cuda:3, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 1, device: cuda:1, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 6, device: cuda:6, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 5, device: cuda:5, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - WARNING - __main__ - Process rank: 0, device: cuda:0, n_gpu: 1 distributed training: True, 16-bits training: False |
| 2025-09-14 20:11:44 - INFO - __main__ - Model parameters ModelArguments(base_model_revision=None, model_name_or_path='google/paligemma2-3b-pt-224', model_revision='main', model_code_revision=None, torch_dtype='bfloat16', tokenizer_name_or_path=None, processor_name_or_path=None, trust_remote_code=False, attn_implementation='eager', use_peft=True, lora_r=16, lora_alpha=16, lora_dropout=0.05, lora_target_modules=['q_proj', 'k_proj', 'v_proj', 'o_proj', 'gate_proj', 'up_proj', 'down_proj'], lora_modules_to_save=None, load_in_8bit=False, load_in_4bit=False, bnb_4bit_quant_type='nf4', use_bnb_nested_quant=False, bnb_4bit_quant_storage='uint8') |
| 2025-09-14 20:11:44 - INFO - __main__ - Data parameters DataArguments(chat_template=None, dataset_mixer={'mateoguaman/vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed': 1.0}, eval_dataset_mixer={'mateoguaman/vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed': 1.0}, text_column='text', dataset_splits=['train', 'test'], train_splits=['train'], validation_splits=['validation'], processing_params={}, dataset_configs=None, preprocessing_num_workers=12, truncation_side=None, auto_insert_empty_system_msg=True, auto_set_chat_template=False, cache_dataset_only=False, just_combine_data=False, output_dataset_name=None, output_dataset_description=None, hf_entity=None) |
| 2025-09-14 20:11:44 - INFO - __main__ - Training/evaluation parameters SFTConfig( |
| _n_gpu=1, |
| accelerator_config={'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None, 'use_configured_state': False}, |
| adafactor=False, |
| adam_beta1=0.9, |
| adam_beta2=0.999, |
| adam_epsilon=1e-08, |
| auto_find_batch_size=False, |
| average_tokens_across_devices=False, |
| batch_eval_metrics=False, |
| bf16=True, |
| bf16_full_eval=False, |
| chars_per_token=<CHARS_PER_TOKEN>, |
| data_seed=None, |
| dataloader_drop_last=False, |
| dataloader_num_workers=0, |
| dataloader_persistent_workers=False, |
| dataloader_pin_memory=True, |
| dataloader_prefetch_factor=None, |
| dataset_batch_size=None, |
| dataset_kwargs=None, |
| dataset_num_proc=None, |
| dataset_text_field=text, |
| ddp_backend=None, |
| ddp_broadcast_buffers=None, |
| ddp_bucket_cap_mb=None, |
| ddp_find_unused_parameters=None, |
| ddp_timeout=1800, |
| debug=[], |
| deepspeed=None, |
| disable_tqdm=False, |
| dispatch_batches=None, |
| do_eval=True, |
| do_predict=False, |
| do_train=False, |
| eval_accumulation_steps=None, |
| eval_delay=0, |
| eval_do_concat_batches=True, |
| eval_on_start=False, |
| eval_packing=None, |
| eval_steps=0.5, |
| eval_strategy=IntervalStrategy.STEPS, |
| eval_use_gather_object=False, |
| evaluation_strategy=None, |
| fp16=False, |
| fp16_backend=auto, |
| fp16_full_eval=False, |
| fp16_opt_level=O1, |
| fsdp=[], |
| fsdp_config={'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, |
| fsdp_min_num_params=0, |
| fsdp_transformer_layer_cls_to_wrap=None, |
| full_determinism=False, |
| gradient_accumulation_steps=1, |
| gradient_checkpointing=False, |
| gradient_checkpointing_kwargs=None, |
| greater_is_better=False, |
| group_by_length=False, |
| half_precision_backend=auto, |
| hub_always_push=False, |
| hub_model_id=paligemma2-3b-pt-224-sft-lora-vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed, |
| hub_model_revision=main, |
| hub_private_repo=None, |
| hub_strategy=HubStrategy.EVERY_SAVE, |
| hub_token=<HUB_TOKEN>, |
| 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=None, |
| label_smoothing_factor=0.0, |
| learning_rate=0.0001, |
| length_column_name=length, |
| load_best_model_at_end=True, |
| local_rank=0, |
| log_level=info, |
| log_level_replica=warning, |
| log_on_each_node=True, |
| logging_dir=data/paligemma2-3b-pt-224-sft-lora-vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/runs/Sep14_20-11-43_g001, |
| logging_first_step=True, |
| logging_nan_inf_filter=True, |
| logging_steps=20, |
| logging_strategy=IntervalStrategy.STEPS, |
| lr_scheduler_kwargs={}, |
| lr_scheduler_type=SchedulerType.COSINE, |
| max_grad_norm=1.0, |
| max_seq_length=2048, |
| max_steps=-1, |
| metric_for_best_model=loss, |
| model_init_kwargs=None, |
| mp_parameters=, |
| neftune_noise_alpha=None, |
| no_cuda=False, |
| num_of_sequences=None, |
| num_train_epochs=1, |
| optim=OptimizerNames.ADAMW_TORCH, |
| optim_args=None, |
| optim_target_modules=None, |
| output_dir=data/paligemma2-3b-pt-224-sft-lora-vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed, |
| overwrite_output_dir=True, |
| packing=True, |
| past_index=-1, |
| per_device_eval_batch_size=16, |
| per_device_train_batch_size=16, |
| prediction_loss_only=False, |
| push_to_hub=False, |
| push_to_hub_model_id=None, |
| push_to_hub_organization=None, |
| push_to_hub_token=<PUSH_TO_HUB_TOKEN>, |
| ray_scope=last, |
| remove_unused_columns=True, |
| report_to=['wandb'], |
| restore_callback_states_from_checkpoint=False, |
| resume_from_checkpoint=None, |
| run_name=data/paligemma2-3b-pt-224-sft-lora-vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed, |
| save_on_each_node=False, |
| save_only_model=False, |
| save_safetensors=True, |
| save_steps=0.5, |
| save_strategy=SaveStrategy.STEPS, |
| save_total_limit=None, |
| seed=42, |
| skip_memory_metrics=True, |
| split_batches=None, |
| tf32=None, |
| torch_compile=False, |
| torch_compile_backend=None, |
| torch_compile_mode=None, |
| torch_empty_cache_steps=None, |
| torchdynamo=None, |
| tpu_metrics_debug=False, |
| tpu_num_cores=None, |
| use_cpu=False, |
| use_ipex=False, |
| use_legacy_prediction_loop=False, |
| use_liger=False, |
| use_liger_kernel=False, |
| use_mps_device=False, |
| warmup_ratio=0.1, |
| warmup_steps=0, |
| weight_decay=1e-05, |
| ) |
| 2025-09-14 20:50:23 - INFO - datasets.builder - Found cached dataset vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed (/gpfs/scrubbed/zchuning/mateo/huggingface/datasets/mateoguaman___vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/default/0.0.0/cea0d1f62292faedf950826fa61aaee53d99f5c5) |
| 2025-09-14 20:50:23 - INFO - datasets.info - Loading Dataset info from /gpfs/scrubbed/zchuning/mateo/huggingface/datasets/mateoguaman___vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/default/0.0.0/cea0d1f62292faedf950826fa61aaee53d99f5c5 |
| 2025-09-14 20:50:26 - INFO - datasets.arrow_dataset - Caching indices mapping at /gpfs/scrubbed/zchuning/mateo/tmp/hf_datasets-wangndhj/cache-25649d22ab1822a1.arrow |
| 2025-09-14 20:50:26 - INFO - datasets.arrow_dataset - Caching indices mapping at /gpfs/scrubbed/zchuning/mateo/tmp/hf_datasets-wangndhj/cache-98f7ebd9d10b0ddf.arrow |
| 2025-09-14 20:50:28 - INFO - datasets.builder - Overwrite dataset info from restored data version if exists. |
| 2025-09-14 20:50:28 - INFO - datasets.info - Loading Dataset info from /gpfs/scrubbed/zchuning/mateo/huggingface/datasets/mateoguaman___vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/default/0.0.0/cea0d1f62292faedf950826fa61aaee53d99f5c5 |
| 2025-09-14 20:50:28 - INFO - datasets.builder - Found cached dataset vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed (/gpfs/scrubbed/zchuning/mateo/huggingface/datasets/mateoguaman___vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/default/0.0.0/cea0d1f62292faedf950826fa61aaee53d99f5c5) |
| 2025-09-14 20:50:28 - INFO - datasets.info - Loading Dataset info from /gpfs/scrubbed/zchuning/mateo/huggingface/datasets/mateoguaman___vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed/default/0.0.0/cea0d1f62292faedf950826fa61aaee53d99f5c5 |
| 2025-09-14 20:50:29 - INFO - datasets.arrow_dataset - Caching indices mapping at /gpfs/scrubbed/zchuning/mateo/tmp/hf_datasets-wangndhj/cache-30409fc3d0e0c16a.arrow |
| 2025-09-14 20:50:29 - INFO - datasets.arrow_dataset - Caching indices mapping at /gpfs/scrubbed/zchuning/mateo/tmp/hf_datasets-wangndhj/cache-0db7001c8880240c.arrow |
| 2025-09-14 20:50:29 - INFO - __main__ - *** Load pretrained model *** |
| 2025-09-14 20:50:30 - INFO - __main__ - Sample 670487 of the processed training set: |
|
|
| Navigate to x=<loc0522>, y=<loc0541> |
|
|
| <loc0506><loc0715><loc0507><loc0634><loc0509><loc0587><loc0508><loc0556><loc0522><loc0541> |
| 2025-09-14 20:50:30 - INFO - __main__ - Sample 116739 of the processed training set: |
|
|
| Navigate to x=<loc0004>, y=<loc0622> |
|
|
| <loc0513><loc0704><loc0438><loc0641><loc0307><loc0631><loc0151><loc0624><loc0004><loc0622> |
| 2025-09-14 20:50:30 - INFO - __main__ - Sample 26225 of the processed training set: |
|
|
| Navigate to x=<loc0523>, y=<loc0533>.Keep a slightly left-of-center course on the paved walkway, avoid pedestrians and stairs, and stay off the curb and median. |
|
|
| <loc0501><loc0715><loc0497><loc0634><loc0504><loc0581><loc0512><loc0548><loc0523><loc0533> |
| 2025-09-14 20:50:34 - INFO - __main__ - Loaded model is not a PeftModel. Getting PEFT config based on model_args. |
| 2025-09-14 20:50:40 - INFO - __main__ - *** Train *** |
| 2025-09-14 23:04:01 - INFO - __main__ - *** Save model *** |
| 2025-09-14 23:04:03 - INFO - __main__ - Model saved to data/paligemma2-3b-pt-224-sft-lora-vamos_10pct_gpt5_mini_cocoqa_localized_narratives_fixed |
| 2025-09-14 23:04:03 - INFO - __main__ - *** Evaluate *** |
| 2025-09-14 23:34:31 - INFO - __main__ - *** Training complete *** |
|
|
