|
|
config: {'model': {'repo_id': './CPT/runs/cpt_run_v1/merged_24b_cpt_lora', 'revision': None, 'base_local_dir': 'base_model', 'trust_remote_code': True, 'tokenizer_use_fast': True, 'device_map': 'auto', 'torch_dtype': 'bfloat16', 'use_4bit': False, 'bnb_4bit_quant_type': 'nf4', 'bnb_4bit_use_double_quant': False, 'bnb_4bit_compute_dtype': 'bfloat16', 'attn_implementation': None}, 'data': {'train_jsonl': '../sft_dataset.jsonl', 'eval_jsonl': None, 'eval_split_ratio': 0.1, 'instruction_field': 'instruction', 'input_field': 'input', 'output_field': 'output', 'format_type': 'custom', 'system_prompt': 'You are a Hyperswitch Rust code analyzer. Identify functions/structs that need modification for a given task.\n\n## Output Format\n\n##OUTPUT\nExplain the data flow and why each component must change:\n- Flow: [Input → Processing → Output with arrows]\n- For each component: "The [ComponentName] ([path]) must [action] because [reason]—without this, [consequence]"\n- Explain coupling between components\n\n##SELECT\nmodify::crates/path/to/file.rs::impl::ComponentName\nadd::crates/another/file.rs::function::AnotherComponent\n<EOS>\n\n## Rules\n\n1. Use full paths: `remove::crates/folder/file.rs::Type::Name`\n2. Use `::` for nested items: `status::StructName::Type::Name`\n3. Always explain "must change because" and "without this"\n3. Types of components: function, struct, enum, impl, trait\n4. If there is extra information (e.g., enum variants), include that too.\n5. Start with ##OUTPUT, end with ##SELECT, terminate with <EOS>\n\n## Example\n\n##TASK\nAdd webhook subscription support\n\n##OUTPUT\nThe webhook system routes events via EventClass enum. Flow: webhook → EventClass → handler → processing. The EventClass enum (crates/common_enums/src/enums.rs::EventClass) must add Subscriptions variant because it defines event routing—without this, subscription events cannot be processed. The SubscriptionStatus impl (crates/common_enums/src/transformers.rs::SubscriptionStatus) must map to EventType because it converts status to events—without this, status changes don\'t trigger webhooks. These are coupled: EventClass routes to handlers that use SubscriptionStatus mappings.\n\n##SELECT\ncrates/common_enums/src/enums.rs::EventClass\ncrates/common_enums/src/transformers.rs::SubscriptionStatus\n<EOS>\n', 'custom_template': '##INSTRUCTION\n{instruction}<|im_end|>\n##TASK\n{input}<|im_end|>\n##OUTPUT\n{output}<|im_end|>', 'max_length': 2048, 'shuffle': True, 'num_proc': 4}, 'peft': {'enabled': True, 'r': 8, 'lora_alpha': 16, 'lora_dropout': 0.05, 'bias': 'none', 'target_modules': 'auto'}, 'train': {'num_train_epochs': 6, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'learning_rate': '1e-4', 'weight_decay': 0.0, 'warmup_ratio': 0.08, 'lr_scheduler_type': 'cosine', 'optim': 'adamw_torch', 'max_grad_norm': 0.8, 'gradient_checkpointing': True, 'logging_steps': 2, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'evaluation_strategy': 'steps', 'eval_steps': 100, 'load_best_model_at_end': True, 'early_stopping': {'enabled': True, 'patience': 3, 'min_delta': 0.001, 'metric': 'eval_loss', 'mode': 'min'}, 'resume_from_checkpoint': 'auto'}, 'run_dir': 'runs/instruct_run_24b', '_wandb': {}} |
|
|
2025-12-23 14:23:22,767 INFO MainThread:924716 [wandb_run.py:_config_callback():1396] config_cb None None {'peft_config': {'default': {'task_type': 'CAUSAL_LM', 'peft_type': 'LORA', 'auto_mapping': None, 'peft_version': '0.18.0', 'base_model_name_or_path': 'CPT/runs/cpt_run_v1/merged_24b_cpt_lora', 'revision': None, 'inference_mode': False, 'r': 8, 'target_modules': ['v_proj', 'k_proj', 'q_proj', 'o_proj'], 'exclude_modules': None, 'lora_alpha': 16, 'lora_dropout': 0.05, 'fan_in_fan_out': False, 'bias': 'none', 'use_rslora': False, 'modules_to_save': None, 'init_lora_weights': True, 'layers_to_transform': None, 'layers_pattern': None, 'rank_pattern': {}, 'alpha_pattern': {}, 'megatron_config': None, 'megatron_core': 'megatron.core', 'trainable_token_indices': None, 'loftq_config': {}, 'eva_config': None, 'corda_config': None, 'use_dora': False, 'alora_invocation_tokens': None, 'use_qalora': False, 'qalora_group_size': 16, 'layer_replication': None, 'runtime_config': {'ephemeral_gpu_offload': False}, 'lora_bias': False, 'target_parameters': None, 'arrow_config': None, 'ensure_weight_tying': False}}, 'image_token_index': 10, 'projector_hidden_act': 'gelu', 'vision_feature_layer': -1, 'vision_config': {'hidden_size': 1024, 'intermediate_size': 4096, 'num_hidden_layers': 24, 'num_attention_heads': 16, 'num_channels': 3, 'patch_size': 14, 'image_size': 1540, 'attention_dropout': 0.0, 'hidden_act': 'silu', 'head_dim': 64, 'initializer_range': 0.02, 'rope_parameters': {'rope_theta': 10000.0, 'rope_type': 'default'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': True, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'pixtral', 'output_attentions': False}, 'text_config': {'vocab_size': 131072, 'max_position_embeddings': 393216, 'hidden_size': 5120, 'intermediate_size': 32768, 'num_hidden_layers': 40, 'num_attention_heads': 32, 'sliding_window': None, 'head_dim': 128, 'num_key_value_heads': 8, 'hidden_act': 'silu', 'initializer_range': 0.02, 'rms_norm_eps': 1e-05, 'use_cache': True, 'attention_dropout': 0.0, 'rope_parameters': {'beta_fast': 32.0, 'beta_slow': 1.0, 'factor': 48.0, 'llama_4_scaling_beta': 0.1, 'mscale': 1.0, 'mscale_all_dim': 1.0, 'original_max_position_embeddings': 8192, 'rope_theta': 100000000.0, 'rope_type': 'yarn', 'type': 'yarn'}, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': None, 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': 1, 'pad_token_id': 11, 'eos_token_id': 2, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': '', 'model_type': 'ministral3', 'output_attentions': False}, 'multimodal_projector_bias': False, 'spatial_merge_size': 2, 'return_dict': True, 'output_hidden_states': False, 'dtype': 'bfloat16', 'tie_word_embeddings': False, 'chunk_size_feed_forward': 0, 'is_encoder_decoder': False, 'is_decoder': False, 'cross_attention_hidden_size': None, 'add_cross_attention': False, 'architectures': ['Mistral3ForConditionalGeneration'], 'finetuning_task': None, 'id2label': {0: 'LABEL_0', 1: 'LABEL_1'}, 'label2id': {'LABEL_0': 0, 'LABEL_1': 1}, 'task_specific_params': None, 'problem_type': None, 'tokenizer_class': None, 'prefix': None, 'bos_token_id': None, 'pad_token_id': None, 'eos_token_id': None, 'sep_token_id': None, 'decoder_start_token_id': None, '_name_or_path': 'CPT/runs/cpt_run_v1/merged_24b_cpt_lora', 'transformers_version': '5.0.0.dev0', 'model_type': 'mistral3', 'use_cache': False, 'output_attentions': False, 'output_dir': 'runs/instruct_run_24b/checkpoints', 'do_train': False, 'do_eval': True, 'do_predict': False, 'eval_strategy': 'steps', 'prediction_loss_only': False, 'per_device_train_batch_size': 1, 'per_device_eval_batch_size': 1, 'gradient_accumulation_steps': 8, 'eval_accumulation_steps': None, 'eval_delay': 0, 'torch_empty_cache_steps': None, 'learning_rate': 0.0001, 'weight_decay': 0.0, 'adam_beta1': 0.9, 'adam_beta2': 0.999, 'adam_epsilon': 1e-08, 'max_grad_norm': 0.8, 'num_train_epochs': 6.0, 'max_steps': -1, 'lr_scheduler_type': 'cosine', 'lr_scheduler_kwargs': None, 'warmup_ratio': 0.08, 'warmup_steps': 0.08, 'log_level': 'passive', 'log_level_replica': 'warning', 'log_on_each_node': True, 'logging_dir': None, 'logging_strategy': 'steps', 'logging_first_step': False, 'logging_steps': 2, 'logging_nan_inf_filter': True, 'save_strategy': 'steps', 'save_steps': 500, 'save_total_limit': 20, 'enable_jit_checkpoint': False, 'save_on_each_node': False, 'save_only_model': False, 'restore_callback_states_from_checkpoint': False, 'use_cpu': False, 'seed': 42, 'data_seed': None, 'bf16': True, 'fp16': False, 'bf16_full_eval': False, 'fp16_full_eval': False, 'tf32': None, 'local_rank': -1, 'ddp_backend': None, 'debug': [], 'dataloader_drop_last': False, 'eval_steps': 100, 'dataloader_num_workers': 0, 'dataloader_prefetch_factor': None, 'run_name': None, 'disable_tqdm': False, 'remove_unused_columns': False, 'label_names': None, 'load_best_model_at_end': True, 'metric_for_best_model': 'eval_loss', 'greater_is_better': False, 'ignore_data_skip': False, 'fsdp': [], 'fsdp_config': {'min_num_params': 0, 'xla': False, 'xla_fsdp_v2': False, 'xla_fsdp_grad_ckpt': False}, 'accelerator_config': {'split_batches': False, 'dispatch_batches': None, 'even_batches': True, 'use_seedable_sampler': True, 'non_blocking': False, 'gradient_accumulation_kwargs': None}, 'parallelism_config': None, 'deepspeed': None, 'label_smoothing_factor': 0.0, 'optim': 'adamw_torch', 'optim_args': None, 'group_by_length': False, 'length_column_name': 'length', 'report_to': ['wandb'], 'project': 'huggingface', 'trackio_space_id': 'trackio', 'ddp_find_unused_parameters': None, 'ddp_bucket_cap_mb': None, 'ddp_broadcast_buffers': None, 'dataloader_pin_memory': True, 'dataloader_persistent_workers': False, 'skip_memory_metrics': True, 'push_to_hub': False, 'resume_from_checkpoint': None, 'hub_model_id': None, 'hub_strategy': 'every_save', 'hub_token': '<HUB_TOKEN>', 'hub_private_repo': None, 'hub_always_push': False, 'hub_revision': None, 'gradient_checkpointing': False, 'gradient_checkpointing_kwargs': None, 'include_for_metrics': [], 'eval_do_concat_batches': True, 'auto_find_batch_size': False, 'full_determinism': False, 'ddp_timeout': 1800, 'torch_compile': False, 'torch_compile_backend': None, 'torch_compile_mode': None, 'include_num_input_tokens_seen': 'no', 'neftune_noise_alpha': None, 'optim_target_modules': None, 'batch_eval_metrics': False, 'eval_on_start': False, 'use_liger_kernel': False, 'liger_kernel_config': None, 'eval_use_gather_object': False, 'average_tokens_across_devices': True} |
