| { | |
| "model_args": { | |
| "base_model_revision": null, | |
| "model_name_or_path": "google/paligemma2-3b-pt-224", | |
| "model_revision": "main", | |
| "model_code_revision": null, | |
| "torch_dtype": "bfloat16", | |
| "tokenizer_name_or_path": null, | |
| "processor_name_or_path": null, | |
| "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": null, | |
| "load_in_8bit": false, | |
| "load_in_4bit": false, | |
| "bnb_4bit_quant_type": "nf4", | |
| "use_bnb_nested_quant": false, | |
| "bnb_4bit_quant_storage": "uint8" | |
| }, | |
| "data_args": { | |
| "chat_template": null, | |
| "dataset_mixer": { | |
| "mateoguaman/coda_every1_25pct_rdp": 1.0 | |
| }, | |
| "eval_dataset_mixer": { | |
| "mateoguaman/coda_every1_25pct_rdp": 1.0 | |
| }, | |
| "text_column": "text", | |
| "dataset_splits": [ | |
| "train", | |
| "test" | |
| ], | |
| "train_splits": [ | |
| "train" | |
| ], | |
| "validation_splits": [ | |
| "validation" | |
| ], | |
| "processing_params": {}, | |
| "dataset_configs": null, | |
| "preprocessing_num_workers": 12, | |
| "truncation_side": null, | |
| "auto_insert_empty_system_msg": true, | |
| "auto_set_chat_template": false, | |
| "cache_dataset_only": false, | |
| "just_combine_data": false, | |
| "output_dataset_name": null, | |
| "output_dataset_description": null, | |
| "hf_entity": "mateoguaman" | |
| }, | |
| "training_args": { | |
| "output_dir": "data/paligemma2-3b-pt-224-sft-lora-coda-only_rdp", | |
| "overwrite_output_dir": true, | |
| "do_train": false, | |
| "do_eval": true, | |
| "do_predict": false, | |
| "eval_strategy": "epoch", | |
| "prediction_loss_only": false, | |
| "per_device_train_batch_size": 8, | |
| "per_device_eval_batch_size": 8, | |
| "per_gpu_train_batch_size": null, | |
| "per_gpu_eval_batch_size": null, | |
| "gradient_accumulation_steps": 1, | |
| "eval_accumulation_steps": null, | |
| "eval_delay": 0, | |
| "torch_empty_cache_steps": null, | |
| "learning_rate": 0.0001, | |
| "weight_decay": 1e-05, | |
| "adam_beta1": 0.9, | |
| "adam_beta2": 0.999, | |
| "adam_epsilon": 1e-08, | |
| "max_grad_norm": 1.0, | |
| "num_train_epochs": 5, | |
| "max_steps": -1, | |
| "lr_scheduler_type": "cosine", | |
| "lr_scheduler_kwargs": {}, | |
| "warmup_ratio": 0.1, | |
| "warmup_steps": 0, | |
| "log_level": "info", | |
| "log_level_replica": "warning", | |
| "log_on_each_node": true, | |
| "logging_dir": "data/paligemma2-3b-pt-224-sft-lora-coda-only_rdp/runs/Apr25_04-19-20_g3065", | |
| "logging_strategy": "steps", | |
| "logging_first_step": true, | |
| "logging_steps": 20, | |
| "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": true, | |
| "fp16": false, | |
| "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": "data/paligemma2-3b-pt-224-sft-lora-coda-only_rdp", | |
| "disable_tqdm": false, | |
| "remove_unused_columns": true, | |
| "label_names": null, | |
| "load_best_model_at_end": true, | |
| "metric_for_best_model": "loss", | |
| "greater_is_better": false, | |
| "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": { | |
| "split_batches": false, | |
| "dispatch_batches": null, | |
| "even_batches": true, | |
| "use_seedable_sampler": true, | |
| "non_blocking": false, | |
| "gradient_accumulation_kwargs": null | |
| }, | |
| "deepspeed": null, | |
| "label_smoothing_factor": 0.0, | |
| "optim": "adamw_torch", | |
| "optim_args": null, | |
| "adafactor": false, | |
| "group_by_length": false, | |
| "length_column_name": "length", | |
| "report_to": [ | |
| "wandb" | |
| ], | |
| "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": true, | |
| "resume_from_checkpoint": null, | |
| "hub_model_id": "paligemma2-3b-pt-224-sft-lora-coda-only_rdp", | |
| "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", | |
| "evaluation_strategy": null, | |
| "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, | |
| "dispatch_batches": null, | |
| "split_batches": 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, | |
| "model_init_kwargs": null, | |
| "use_liger": false, | |
| "dataset_text_field": "text", | |
| "dataset_kwargs": null, | |
| "dataset_num_proc": null, | |
| "max_seq_length": 2048, | |
| "packing": true, | |
| "eval_packing": null, | |
| "dataset_batch_size": null, | |
| "num_of_sequences": null, | |
| "chars_per_token": null, | |
| "hub_model_revision": "main", | |
| "distributed_state": { | |
| "_cpu": false, | |
| "backend": "nccl", | |
| "device": "cuda:0", | |
| "debug": false, | |
| "distributed_type": "MULTI_GPU", | |
| "num_processes": 4, | |
| "process_index": 0, | |
| "local_process_index": 0, | |
| "fork_launched": false | |
| }, | |
| "_n_gpu": 1, | |
| "__cached__setup_devices": "cuda:0", | |
| "deepspeed_plugin": null | |
| } | |
| } |