huggingface/peft

18.1

Adjusted Score

18.1

Raw Score

100%

Time Factor

2026-08-28

Last Push

21.6K

Stars

Python

Language

212.6K

Lines of Code

908

Files

3.3K

Pattern Hits

2026-08-29

Scan Date

0.18

HC Hit Rate

What These Metrics Mean

Adjusted Score: Primary synthetic code indicator. Raw score normalised per 1,000 lines of code and multiplied by the temporal discount factor. This is the definitive comparative metric — use it to rank repositories by AI authorship density.
Raw Score: The unmodified sum of all severity-weighted, context-multiplied pattern match scores before temporal discounting. Reflects the absolute signal strength independent of when the repository was last active.
Time Factor: The temporal discount multiplier (0–100%) applied to the raw score. Repositories last updated before ChatGPT's launch (Nov 2022) receive a 5% factor. Full signal is only assigned to repositories active in the post-adoption era (Jan 2024+).
Pattern Hits: Total count of individual pattern matches across all files and categories. A high hit count with a low score may indicate a very large codebase with isolated AI snippets; a low count with a high score indicates dense, concentrated AI signatures.
HC Hit Rate: High+Critical pattern hits per file, averaged across the repository. This orthogonal signal catches repositories where a few files are densely packed with high-severity AI tells — a strong indicator even when the normalised score appears moderate due to codebase size.
Lines of Code / Files: Total lines and files analysed. The scanner examines 94 file extensions. These denominators are used to normalise the score, enabling fair comparison between repositories of vastly different sizes.

Score History

This chart maps the temporal evolution of the adjusted synthetic code score across successive scan runs. An upward trajectory indicates ongoing incorporation of AI-generated code or expanding LLM-assisted scaffolding; a stable or declining trajectory may reflect active human refactoring, code removal, or the adoption of stricter authorship policies. The dashed secondary line (right axis) independently tracks total raw pattern hit count, which can diverge from the normalised score when codebase size changes significantly between scans.

Severity Breakdown

Classifies detected patterns by their diagnostic confidence and structural impact. CRITICAL patterns (coefficient 10) represent definitive synthetic signatures — hallucinated imports, explicit LLM attribution metadata — virtually never produced by human authors. HIGH (5) indicates strong structural tells such as cross-file repetition or cross-linguistic idioms. MEDIUM (2) covers recognisable conversational padding and AI-specific vocabulary. LOW (1) captures subtle indicators like tautological comments and generic boilerplate that require density to carry independent signal.

CRITICAL 38HIGH 128MEDIUM 79LOW 3031

Directory Score Breakdown

This horizontal bar chart decomposes the repository's raw synthetic code score by top-level directory, allowing you to pinpoint precisely which modules or components carry the highest AI authorship density. Directories with disproportionately high scores relative to their size warrant targeted manual review: concentrated AI signatures often trace back to mass-generated configuration layers, auto-ported test suites, LLM-scaffolded boilerplate classes, or entire subsystems authored under heavy copilot assistance. Use this view to prioritise your human code-review effort.

Pattern Findings

The scanner identified 3276 distinct pattern matches across 20 syntactic categories. Each entry below represents a discrete location in the source code where the engine recorded a statistically significant AI authorship indicator. Expand any category row to inspect the individual file paths, line numbers, code snippets, and the lexical context (CODE, COMMENT, or STRING) in which each match was detected.

Reading the findings table: The Severity column indicates the diagnostic confidence level (CRITICAL / HIGH / MEDIUM / LOW). The Context column identifies whether the match occurred inside executable code, an inline comment, or a string literal — comment-context matches receive a ×1.5 weight because LLMs systematically over-annotate. The ⚡ bolt icon marks clustered matches: three or more patterns within a 10-line window, each receiving an additional ×1.5 density multiplier as dense clusters constitute far stronger evidence of synthetic authorship than isolated hits.

Hyper-Verbose Identifiers1468 hits · 1119 pts

Severity	File	Line	Snippet	Context
LOW	tests/test_mixed.py	449	def test_target_different_layers(self, config0, config1):	CODE
LOW	tests/test_mixed.py	490	def test_target_last_layer_same_type(self, config0, config1):	CODE
LOW	tests/test_mixed.py	519	def test_target_first_layer_same_type(self, config0, config1):	CODE
LOW	tests/test_mixed.py	655	def test_delete_merged_adapter_is_atomic(self):	CODE
LOW	tests/test_mixed.py	694	def test_unload_modules_to_save_when_active_adapter_does_not_use_it(self):	CODE
LOW	tests/test_mixed.py	729	def test_get_nb_trainable_parameters(self):	CODE
LOW	tests/test_mixed.py	758	def test_bias_only_honored_for_non_lora_tuner(self):	CODE
LOW	tests/test_mixed.py	769	def test_incompatible_config_raises(self):	CODE
LOW	tests/test_seq_classifier.py	362	def prepare_inputs_for_testing(self):	STRING
LOW	tests/test_seq_classifier.py	369	def test_attributes_parametrized(self, model_id, config_cls, config_kwargs):	STRING
LOW	tests/test_seq_classifier.py	381	def test_prepare_for_training_parametrized(self, model_id, config_cls, config_kwargs):	STRING
LOW	tests/test_seq_classifier.py	387	def test_prompt_tuning_text_prepare_for_training(self, model_id, config_cls, config_kwargs):	STRING
LOW⚡	tests/test_seq_classifier.py	405	def test_save_pretrained_pickle(self, model_id, config_cls, config_kwargs):	STRING
LOW⚡	tests/test_seq_classifier.py	412	def test_save_pretrained_selected_adapters(self, model_id, config_cls, config_kwargs):	STRING
LOW⚡	tests/test_seq_classifier.py	419	def test_save_pretrained_selected_adapters_pickle(self, model_id, config_cls, config_kwargs):	STRING
LOW⚡	tests/test_seq_classifier.py	428	def test_from_pretrained_config_construction(self, model_id, config_cls, config_kwargs):	STRING
LOW⚡	tests/test_seq_classifier.py	434	def test_modules_to_save_correctly_set(self, model_id, config_cls, config_kwargs):	STRING
LOW	tests/test_seq_classifier.py	513	def test_prompt_learning_forward_with_inputs_embeds(self, config_cls, config_kwargs):	STRING
LOW	tests/test_lora_conversion.py	79	def test_no_peft_layer_raises(self):	CODE
LOW	tests/test_lora_conversion.py	86	def test_prompt_learning_model_raises(self):	CODE
LOW	tests/test_lora_conversion.py	97	def test_peft_model_but_no_support_raises(self):	CODE
LOW	tests/test_lora_conversion.py	107	def test_model_with_unsupported_layers_raises(self):	CODE
LOW	tests/test_lora_conversion.py	123	def test_targeted_modules_identical(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	139	def test_targeted_modules_identical_target_modules_str(self):	CODE
LOW	tests/test_lora_conversion.py	158	def test_fixed_rank_lora_config(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	168	def test_dynamic_rank_lora_config(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	194	def test_threshold_wrong_value_raises(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	204	def test_rank_higher_than_weight_dim_raises(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	215	def test_converting_transformers_model_works(self, lokr_model, tmp_path):	CODE
LOW	tests/test_lora_conversion.py	246	def test_converted_lora_approximates_original_adapter(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	335	def test_model_without_peft_config(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	373	def test_converted_lora_to_lora_works_and_warns(self):	CODE
LOW	tests/test_lora_conversion.py	411	def test_converted_lora_with_multiple_adapters(self, lokr_model):	CODE
LOW	tests/test_lora_conversion.py	457	def test_convert_model_with_modules_to_save(self):	CODE
LOW	tests/test_lora_conversion.py	497	def test_convert_model_with_trainable_bias_raises(self, bias):	CODE
LOW	tests/test_lora_conversion.py	606	def test_miss_supports_lora_conversion(self, miss_model_standard, miss_model_mini, miss_model_bat):	CODE
LOW	tests/test_lora_conversion.py	611	def test_miss_standard_exact_conversion(self, miss_model_standard):	CODE
LOW	tests/test_lora_conversion.py	628	def test_miss_mini_exact_conversion(self, miss_model_mini):	CODE
LOW	tests/test_lora_conversion.py	645	def test_miss_bat_approximate_conversion(self, miss_model_bat):	CODE
LOW	tests/test_lora_conversion.py	667	def test_miss_targeted_modules_identical(self, miss_model_standard):	CODE
LOW	tests/test_lora_conversion.py	710	def test_kasa_does_not_support_lora_conversion(self):	CODE
LOW⚡	tests/test_tuners_utils.py	464	def test_one_targeted_module_regex(self):	STRING
LOW⚡	tests/test_tuners_utils.py	469	def test_two_targeted_module_regex(self):	STRING
LOW⚡	tests/test_tuners_utils.py	474	def test_one_targeted_module_list(self):	STRING
LOW⚡	tests/test_tuners_utils.py	479	def test_two_targeted_module_list(self):	STRING
LOW⚡	tests/test_tuners_utils.py	582	def test_two_excluded_module_regex(self):	STRING
LOW⚡	tests/test_tuners_utils.py	587	def test_two_excluded_module_list(self):	STRING
LOW⚡	tests/test_tuners_utils.py	592	def test_multiple_excluded_modules_list(self):	STRING
LOW⚡	tests/test_tuners_utils.py	627	def test_no_modules_matched_second_adapter(self):	STRING
LOW⚡	tests/test_tuners_utils.py	637	def test_second_adapter_only_modules_to_save_raises(self):	STRING
LOW⚡	tests/test_tuners_utils.py	646	def test_no_matching_peft_module_error_hierarchy(self):	STRING
LOW⚡	tests/test_tuners_utils.py	651	def test_targeted_module_names_unique_across_adapters(self):	STRING
LOW⚡	tests/test_tuners_utils.py	658	def test_some_modules_excluded_some_unmatched(self):	STRING
LOW⚡	tests/test_tuners_utils.py	663	def test_exclude_modules_not_used(self):	STRING
LOW⚡	tests/test_tuners_utils.py	881	def test_available_adapters_small(self, small_model):	STRING
LOW⚡	tests/test_tuners_utils.py	887	def test_available_adapters_large(self, large_model):	STRING
LOW⚡	tests/test_tuners_utils.py	893	def test_devices_all_cpu_small(self, small_model):	STRING
LOW⚡	tests/test_tuners_utils.py	899	def test_devices_all_cpu_large(self, large_model):	STRING
LOW⚡	tests/test_tuners_utils.py	985	def test_target_parameters_and_target_modules(self, large_model):	STRING
LOW⚡	tests/test_tuners_utils.py	995	def test_quantization_backend_small(self, small_model):	STRING
1408 more matches not shown…

Cross-File Repetition117 hits · 585 pts

Severity	File	Line	Snippet	Context
HIGH	tests/test_gpu_examples.py	0	efficient mechanism to free gpu memory after each test. based on https://github.com/huggingface/transformers/issues/2109	STRING
HIGH	tests/test_vision_models.py	0	efficient mechanism to free gpu memory after each test. based on https://github.com/huggingface/transformers/issues/2109	STRING
HIGH	tests/test_gptqmodel.py	0	efficient mechanism to free gpu memory after each test. based on https://github.com/huggingface/transformers/issues/2109	STRING
HIGH	tests/test_common_gpu.py	0	efficient mechanism to free gpu memory after each test. based on https://github.com/huggingface/transformers/issues/2109	STRING
HIGH	tests/test_torch_compile.py	0	efficient mechanism to free gpu memory after each test. based on https://github.com/huggingface/transformers/issues/2109	STRING
HIGH	docs/source/package_reference/lora.md	0	run forward and backward passes for gradient estimation.	STRING
HIGH	examples/lora_ga_finetuning/README.md	0	run forward and backward passes for gradient estimation.	STRING
HIGH	examples/lora_ga_finetuning/lora_ga_finetuning.py	0	run forward and backward passes for gradient estimation.	STRING
HIGH	examples/boft_controlnet/utils/tracemalloc.py	0	get resident set size memory for the current process	STRING
HIGH	…_modeling/peft_lora_clm_accelerate_ds_zero3_offload.py	0	get resident set size memory for the current process	STRING
HIGH	examples/deft_dreambooth/train_dreambooth.py	0	get resident set size memory for the current process	STRING
HIGH	examples/oft_dreambooth/train_dreambooth.py	0	get resident set size memory for the current process	STRING
HIGH	examples/hra_dreambooth/utils/tracemalloc.py	0	get resident set size memory for the current process	STRING
HIGH	examples/stable_diffusion/train_dreambooth.py	0	get resident set size memory for the current process	STRING
HIGH	examples/lora_dreambooth/train_dreambooth.py	0	get resident set size memory for the current process	STRING
HIGH	examples/boft_dreambooth/utils/tracemalloc.py	0	get resident set size memory for the current process	STRING
HIGH	…ation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py	0	get resident set size memory for the current process	STRING
HIGH	examples/waveft_finetuning/waveft_finetuning.py	0	below is an instruction that describes a task. write a response that appropriately completes the request. ### instructio	STRING
HIGH	examples/hira_finetuning/hira_finetuning.py	0	below is an instruction that describes a task. write a response that appropriately completes the request. ### instructio	STRING
HIGH	examples/glora_finetuning/glora_finetuning.py	0	below is an instruction that describes a task. write a response that appropriately completes the request. ### instructio	STRING
HIGH	examples/olora_finetuning/olora_finetuning.py	0	below is an instruction that describes a task. write a response that appropriately completes the request. ### instructio	STRING
HIGH	examples/shira_finetuning/shira_finetuning.py	0	below is an instruction that describes a task. write a response that appropriately completes the request. ### instructio	STRING
HIGH	examples/deft_dreambooth/train_dreambooth.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/oft_dreambooth/train_dreambooth.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/hra_dreambooth/utils/dataset.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/stable_diffusion/train_dreambooth.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/lora_dreambooth/train_dreambooth.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/boft_dreambooth/utils/dataset.py	0	a dataset to prepare the instance and class images with the prompts for fine-tuning the model. it pre-processes the imag	STRING
HIGH	examples/corda_finetuning/corda_finetuning.py	0	returns the number of trainable parameters and number of all parameters in the model.	STRING
HIGH	src/peft/peft_model.py	0	returns the number of trainable parameters and number of all parameters in the model.	STRING
HIGH	src/peft/mixed_model.py	0	returns the number of trainable parameters and number of all parameters in the model.	STRING
HIGH	src/peft/peft_model.py	0	deletes an existing adapter. args: adapter_name (str): name of the adapter to be deleted.	STRING
HIGH	src/peft/tuners/tuners_utils.py	0	deletes an existing adapter. args: adapter_name (str): name of the adapter to be deleted.	STRING
HIGH	src/peft/tuners/waveft/model.py	0	deletes an existing adapter. args: adapter_name (str): name of the adapter to be deleted.	STRING
HIGH	src/peft/utils/other.py	0	check if the arguments are compatible with the configs and state of the model	STRING
HIGH	src/peft/tuners/road/layer.py	0	check if the arguments are compatible with the configs and state of the model	STRING
HIGH	src/peft/tuners/lora/layer.py	0	check if the arguments are compatible with the configs and state of the model	STRING
HIGH	src/peft/tuners/lycoris_utils.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/hra/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/road/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/miss/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/oft/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/vera/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/delora/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/boft/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/waveft/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/c3a/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/hira/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/hira/bnb.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/road/bnb.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/fourierft/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/vblora/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/gralora/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/ia3/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/osf/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/adalora/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/pvera/layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/lora/hqq.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/lora/tp_layer.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
HIGH	src/peft/tuners/lora/inc.py	0	merge the active adapter weights into the base weights. args: safe_merge (`bool`, optional): if `true`, the merge oper	STRING
57 more matches not shown…

Hallucination Indicators38 hits · 470 pts

Severity	File	Line	Snippet	Context
CRITICAL	tests/test_mixed.py	709	lin1_weight_before = peft_model.base_model.model.lin1.original_module.weight.data.clone()	CODE
CRITICAL	tests/test_lora_conversion.py	469	lokr_model.base_model.model.model.decoder.layers[0].fc1.modules_to_save.default.weight.data.mul_(-10.0)	CODE
CRITICAL	tests/test_lora_conversion.py	470	lokr_model.base_model.model.model.decoder.layers[0].fc1.modules_to_save.default.bias.data.mul_(-10.0)	CODE
CRITICAL⚡	tests/test_cpt.py	247	emb = model.prompt_encoder.default.embedding.weight.data.clone().detach()	CODE
CRITICAL⚡	tests/test_cpt.py	263	assert torch.all(model.prompt_encoder.default.embedding.weight.data.clone().detach().cpu() == emb.cpu())	CODE
CRITICAL⚡	tests/test_cpt.py	277	emb = model.prompt_encoder.default.embedding.weight.data.clone().detach()	CODE
CRITICAL	tests/test_cpt.py	294	assert torch.all(model.prompt_encoder.default.embedding.weight.data.clone().detach().cpu() == emb.cpu())	CODE
CRITICAL	tests/test_trainable_tokens.py	228	initial_delta = model.model.model.embed_tokens.trainable_tokens_delta.default.clone()	CODE
CRITICAL	tests/test_trainable_tokens.py	229	initial_originals = model.model.model.embed_tokens.trainable_tokens_original.default.clone()	CODE
CRITICAL	tests/test_trainable_tokens.py	732	assert merged_model.model.decoder.embed_tokens.weight.data_ptr() == merged_model.lm_head.weight.data_ptr()	CODE
CRITICAL	tests/test_trainable_tokens.py	769	assert merged_model.model.decoder.embed_tokens.weight.data_ptr() == merged_model.lm_head.weight.data_ptr()	CODE
CRITICAL	tests/test_trainable_tokens.py	868	expected_changed_weights = peft_model.model.model.embed_tokens.trainable_tokens_delta.default.data.clone()	CODE
CRITICAL	tests/test_gpu_examples.py	2231	embed_params = model.base_model.model.model.decoder.embed_tokens.original_module.weight.numel()	STRING
CRITICAL⚡	tests/test_initialization.py	6251	model.base_model.model.model.embed_tokens.weight.data_ptr()	CODE
CRITICAL⚡	tests/test_initialization.py	6252	!= model.base_model.model.lm_head.weight.data_ptr()	CODE
CRITICAL	tests/test_initialization.py	5723	peft_model.prompt_encoder.default.embedding.weight.data.fill_(1.0)	CODE
CRITICAL	tests/test_initialization.py	5940	embed_np = dict(model.base_model.model.model.embed_tokens.named_parameters())	CODE
CRITICAL	tests/test_initialization.py	6000	embed_np = dict(model.base_model.model.model.embed_tokens.named_parameters())	CODE
CRITICAL	tests/test_initialization.py	6175	shared_np = dict(model.base_model.model.model.shared.named_parameters())	CODE
CRITICAL⚡	tests/test_target_parameters.py	384	gate_proj_1_0.lora_A.default.weight.data.copy_(gate_proj_0_0.lora_A.default.weight.data)	CODE
CRITICAL⚡	tests/test_target_parameters.py	385	gate_proj_1_1.lora_A.default.weight.data.copy_(gate_proj_0_1.lora_A.default.weight.data)	CODE
CRITICAL⚡	tests/test_target_parameters.py	386	gate_proj_1_0.lora_B.default.weight.data.copy_(gate_proj_0_0.lora_B.default.weight.data)	CODE
CRITICAL⚡	tests/test_target_parameters.py	387	gate_proj_1_1.lora_B.default.weight.data.copy_(gate_proj_0_1.lora_B.default.weight.data)	CODE
CRITICAL⚡	tests/test_decoder_models.py	865	layers[0].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	866	== layers[1].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	867	and layers[2].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	868	== layers[3].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	871	layers[0].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	872	!= layers[2].mlp.up_proj.base_layer.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	875	layers[0].mlp.up_proj.lora_A.default.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	876	!= layers[1].mlp.up_proj.lora_A.default.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	877	and layers[2].mlp.up_proj.lora_A.default.weight.data.storage().data_ptr()	CODE
CRITICAL⚡	tests/test_decoder_models.py	878	!= layers[3].mlp.up_proj.lora_A.default.weight.data.storage().data_ptr()	CODE
CRITICAL	tests/test_shira.py	300	peft.tuners.shira.layer._warn_once_about_module_hooks.cache_clear()	CODE
CRITICAL	tests/test_custom_models.py	2524	original_weight = model.base_model.model.conv.base_layer.weight.data.clone()	CODE
CRITICAL	tests/test_custom_models.py	4343	lm_head_keys = list(model.base_model.model.lm_head.modules_to_save.keys())	CODE
CRITICAL	tests/test_custom_models.py	4346	embed_token_keys = list(model.base_model.model.model.decoder.embed_tokens.modules_to_save.keys())	CODE
CRITICAL	src/peft/tuners/tuners_utils.py	653	model.lm_head.weight.data = model.model.embed_tokens.weight.data.clone()	CODE

Unused Imports619 hits · 450 pts

Severity	File	Line	Context
LOW	tests/test_lorafa.py	15	CODE
LOW	tests/test_trainable_tokens.py	15	CODE
LOW	tests/test_loraplus.py	14	CODE
LOW	method_comparison/app_embed.py	26	CODE
LOW	method_comparison/app_embed.py	26	CODE
LOW	examples/stable_diffusion/inc_flux_lora_hpu.py	26	CODE
LOW	src/peft/config.py	14	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	17	CODE
LOW	src/peft/__init__.py	27	CODE
LOW	src/peft/__init__.py	27	CODE
LOW	src/peft/__init__.py	28	CODE
LOW	src/peft/__init__.py	29	CODE
LOW	src/peft/__init__.py	29	CODE
LOW	src/peft/__init__.py	29	CODE
LOW	src/peft/__init__.py	29	CODE
LOW	src/peft/__init__.py	29	CODE
LOW	src/peft/__init__.py	36	CODE
LOW	src/peft/__init__.py	37	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	38	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
LOW	src/peft/__init__.py	49	CODE
559 more matches not shown…

Over-Commented Block399 hits · 388 pts

Severity	File	Line	Snippet	Context
LOW	setup.py	1	# Copyright 2023 The HuggingFace Team. All rights reserved.	COMMENT
LOW	setup.py	101	# 4. Add a tag in git to mark the release: "git tag -a v<VERSION> -m 'Adds tag <VERSION> for pypi' "	COMMENT
LOW	tests/test_mixed.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_seq_classifier.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_lora_conversion.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_tuners_utils.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_tuners_utils.py	2021	# resulting model for overview:	COMMENT
LOW	tests/conftest.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_xlora.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_riemannian_lora.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_hub_features.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_lorafa.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_cpt.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_low_level_api.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_lora_megatron.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_trainable_tokens.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_gpu_examples.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_integrations.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_poly.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/testing_utils.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_loraplus.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_adaption_prompt.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_shadow.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_frod.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_supertuning.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_initialization.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_initialization.py	5501	# )	COMMENT
LOW	tests/test_initialization.py	5521	# (0): Linear(in_features=5120, out_features=5120, bias=True)	COMMENT
LOW	tests/test_initialization.py	5541	# (down_proj): Linear(in_features=8960, out_features=1536, bias=False)	COMMENT
LOW	tests/test_initialization.py	5581		COMMENT
LOW	tests/test_initialization.py	5601	# (fc1): Linear(in_features=1280, out_features=5120, bias=True)	COMMENT
LOW	tests/test_initialization.py	5621	# (q_proj): Linear(in_features=1536, out_features=1536, bias=True)	COMMENT
LOW	tests/test_target_parameters.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_pvera.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_vision_models.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_gptqmodel.py	1	# Note: These tests were copied from test_common_gpu.py and test_gpu_examples.py as they can run on CPU too.	COMMENT
LOW	tests/test_decoder_models.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_encoder_decoder_models.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_multitask_prompt_tuning.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_lora_ga.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/testing_common.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_quantization.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_common_gpu.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_cartridge.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_vblora.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_config.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_randlora.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_unilora.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_shira.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_boft.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_feature_extraction_models.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_bdlora.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_helpers.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_stablediffusion.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_lora_variants.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_incremental_pca.py	1	# Copyright 2024-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_auto.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_deft.py	1	# Copyright 2026-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_other.py	1	# Copyright 2023-present the HuggingFace Inc. team.	COMMENT
LOW	tests/test_arrow.py	1	# Copyright 2025-present the HuggingFace Inc. team.	COMMENT
339 more matches not shown…

Deep Nesting262 hits · 236 pts

Severity	File	Line	Context
LOW	tests/test_lora_conversion.py	577	CODE
LOW	tests/test_trainable_tokens.py	638	CODE
LOW	tests/test_gpu_examples.py	6721	CODE
LOW	tests/test_gpu_examples.py	6775	CODE
LOW	tests/test_gpu_examples.py	5995	CODE
LOW	tests/testing_utils.py	322	CODE
LOW	tests/test_initialization.py	1493	CODE
LOW	tests/test_initialization.py	4924	CODE
LOW	tests/test_initialization.py	5101	CODE
LOW	tests/test_initialization.py	5133	CODE
LOW	tests/test_decoder_models.py	486	CODE
LOW	tests/testing_common.py	596	CODE
LOW	tests/testing_common.py	791	CODE
LOW	tests/testing_common.py	1054	CODE
LOW	tests/testing_common.py	1226	CODE
LOW	tests/testing_common.py	1337	CODE
LOW	tests/testing_common.py	1369	CODE
LOW	tests/testing_common.py	1429	CODE
LOW	tests/testing_common.py	1538	CODE
LOW	tests/test_other.py	710	CODE
LOW	tests/test_arrow.py	474	CODE
LOW	tests/test_custom_models.py	2863	CODE
LOW	tests/test_custom_models.py	2982	CODE
LOW	method_comparison/app_embed.py	38	CODE
LOW	method_comparison/sanitizer.py	6	CODE
LOW	method_comparison/app.py	280	CODE
LOW	method_comparison/image-gen/run.py	187	CODE
LOW	method_comparison/image-gen/utils.py	322	CODE
LOW	method_comparison/text_generation_benchmark/run.py	139	CODE
LOW	method_comparison/text_generation_benchmark/utils.py	205	CODE
LOW	method_comparison/peft-shop/app.py	1110	CODE
LOW	method_comparison/MetaMathQA/utils.py	218	CODE
LOW	method_comparison/MetaMathQA/utils.py	369	CODE
LOW	…les/sequence_classification/peft_no_lora_accelerate.py	79	CODE
LOW	examples/boft_controlnet/train_controlnet.py	86	CODE
LOW	examples/boft_controlnet/utils/pipeline_controlnet.py	53	CODE
LOW	examples/boft_controlnet/utils/pipeline_controlnet.py	166	CODE
LOW	examples/boft_controlnet/utils/dataset.py	82	CODE
LOW	examples/boft_controlnet/utils/dataset.py	136	CODE
LOW	…ture_extraction/peft_lora_embedding_semantic_search.py	210	CODE
LOW	…_modeling/peft_lora_clm_accelerate_ds_zero3_offload.py	110	CODE
LOW	examples/arrow_multitask/arrow_phi3_mini.py	120	CODE
LOW	examples/loftq_finetuning/int8_correction.py	68	CODE
LOW	examples/loftq_finetuning/train_gsm8k_llama.py	297	CODE
LOW	…orthogonal_subspace_learning/osf_continual_learning.py	51	CODE
LOW	…orthogonal_subspace_learning/osf_continual_learning.py	99	CODE
LOW	…orthogonal_subspace_learning/osf_continual_learning.py	142	CODE
LOW	examples/orthogonal_subspace_learning/utils.py	42	CODE
LOW	examples/deft_dreambooth/train_dreambooth.py	585	CODE
LOW	examples/oft_dreambooth/train_dreambooth.py	590	CODE
LOW	examples/hra_dreambooth/train_dreambooth.py	82	CODE
LOW	examples/qalora_finetuning/qalora_gptq_finetuning.py	23	CODE
LOW	examples/stable_diffusion/train_dreambooth.py	754	CODE
LOW	examples/stable_diffusion/convert_sd_adapter_to_peft.py	285	CODE
LOW	examples/lora_dreambooth/train_dreambooth.py	580	CODE
LOW	examples/corda_finetuning/datautils.py	106	CODE
LOW	examples/sft/utils.py	86	CODE
LOW	examples/boft_dreambooth/train_dreambooth.py	82	CODE
LOW	…ation/peft_lora_seq2seq_accelerate_ds_zero3_offload.py	104	CODE
LOW	…s/int8_training/peft_adalora_whisper_large_training.py	423	CODE
202 more matches not shown…

Decorative Section Separators35 hits · 111 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/test_arrow.py	30	# ─── Fixtures ──────────────────────────────────────────────────────────	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	19	# ==========================================	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	21	# ==========================================	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	47	# ==========================================	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	49	# ==========================================	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	146	# ==========================================	COMMENT
MEDIUM	examples/KappaTune/experiments_kappatune_peft.py	148	# ==========================================	COMMENT
MEDIUM⚡	…etuning/sequence_classification_finetune_monteclora.py	21	# ----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	…etuning/sequence_classification_finetune_monteclora.py	23	# ----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	…etuning/sequence_classification_finetune_monteclora.py	29	# ----------------------------------------------------------------------------	COMMENT
MEDIUM⚡	…etuning/sequence_classification_finetune_monteclora.py	31	# ----------------------------------------------------------------------------	COMMENT
MEDIUM	…etuning/sequence_classification_finetune_monteclora.py	42	# ----------------------------------------------------------------------------	COMMENT
MEDIUM	…etuning/sequence_classification_finetune_monteclora.py	44	# ----------------------------------------------------------------------------	COMMENT
MEDIUM	…etuning/sequence_classification_finetune_monteclora.py	162	# ----------------------------------------------------------------------------	COMMENT
MEDIUM	…etuning/sequence_classification_finetune_monteclora.py	164	# ----------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	74	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	76	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	91	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	93	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	177	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	179	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	196	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	198	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	257	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	259	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	353	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	355	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	410	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	412	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	453	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	455	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	524	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	scripts/check_doc_coverage.py	526	# ---------------------------------------------------------------------------	COMMENT
MEDIUM	src/peft/tuners/lora/layer.py	867	# ------------------------------------------------------------------------------------------	COMMENT
MEDIUM	src/peft/tuners/lora/layer.py	870	# ------------------------------------------------------------------------------------------	COMMENT

Modern Structural Boilerplate99 hits · 96 pts

Severity	File	Line	Snippet	Context
LOW	method_comparison/app.py	46	logger = logging.getLogger(__name__)	CODE
LOW	method_comparison/peft-shop/app.py	294	def _set_data(data: dict[str, Any]) -> None:	CODE
LOW	examples/stable_diffusion/convert_sd_adapter_to_peft.py	20	logger = logging.getLogger(__name__)	CODE
LOW	src/peft/__init__.py	184	__all__ = [	CODE
LOW	src/peft/peft_model.py	1572	def set_adapter(self, adapter_name: str, inference_mode: bool = False) -> None:	CODE
LOW	src/peft/peft_model.py	1597	def set_requires_grad(self, adapter_names: str \| Sequence[str], requires_grad: bool = True) -> None:	CODE
LOW	src/peft/functional.py	26	__all__ = [	CODE
LOW	src/peft/mixed_model.py	254	def set_modules_to_save(self, peft_config: PeftConfig, adapter_name: str) -> None:	CODE
LOW	src/peft/mixed_model.py	269	def set_adapter(self, adapter_name: Union[str, list[str]], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/helpers.py	39	def update_forward_signature(model: PeftModel) -> None:	CODE
LOW	src/peft/helpers.py	73	def update_generate_signature(model: PeftModel) -> None:	CODE
LOW	src/peft/helpers.py	114	def update_signature(model: PeftModel, method: str = "all") -> None:	CODE
LOW	src/peft/utils/hotswap.py	36	def _update_scaling(lora_module: LoraLayer, adapter_name: str, scaling: float) -> None:	CODE
LOW	src/peft/utils/__init__.py	93	__all__ = [	CODE
LOW	src/peft/utils/other.py	97	__all__ = [	CODE
LOW	src/peft/utils/other.py	528	def set_adapter(self, adapter_names: Union[str, list[str]], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/utils/other.py	554	def set_requires_grad(self, adapter_names: str \| Sequence[str], requires_grad: bool = True) -> None:	CODE
LOW	src/peft/utils/other.py	707	def set_adapter(self, adapter_names: Union[str, list[str]], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/utils/other.py	956	def set_adapter(self, adapter_names: Union[str, list[str]], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/utils/other.py	1136	def _set_adapter(model, adapter_name: str \| list[str], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/utils/other.py	284	def _set_layer_requires_grad(layer, requires_grad: bool) -> None:	STRING
LOW	src/peft/utils/quantization_utils.py	75	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/utils/quantization_utils.py	116	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/utils/quantization_utils.py	137	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/utils/quantization_utils.py	159	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/utils/quantization_utils.py	191	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/utils/quantization_utils.py	233	def set_base_weight(self, peft_layer, weight_data: torch.Tensor) -> None:	CODE
LOW	src/peft/tuners/__init__.py	81	__all__ = [	CODE
LOW	src/peft/tuners/tuners_utils.py	1315	def set_adapter(self, adapter_name: str \| list[str], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/tuners/tuners_utils.py	2009	def set_adapter(self, adapter_names: str \| list[str], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/tuners/tuners_utils.py	2089	def set_requires_grad(self, adapter_names: str \| Sequence[str], requires_grad: bool = True) -> None:	CODE
LOW	src/peft/tuners/tuners_utils.py	2713	def set_requires_grad(model, adapter_names: str \| Sequence[str], requires_grad: bool = True) -> None:	CODE
LOW	src/peft/tuners/tuners_utils.py	587	def set_requires_grad(self, adapter_names: str \| Sequence[str], requires_grad: bool = True) -> None:	STRING
LOW	src/peft/tuners/tuners_utils.py	1859	def set_base_weight(self, weight_data: torch.Tensor) -> None:	STRING
LOW	src/peft/tuners/peanut/__init__.py	22	__all__ = ["Linear", "PeanutConfig", "PeanutLayer", "PeanutModel"]	CODE
LOW	src/peft/tuners/prompt_tuning/__init__.py	21	__all__ = ["PromptEmbedding", "PromptTuningConfig", "PromptTuningInit"]	CODE
LOW	src/peft/tuners/hra/__init__.py	22	__all__ = ["HRAConfig", "HRAConv2d", "HRALayer", "HRALinear", "HRAModel"]	CODE
LOW	src/peft/tuners/loha/__init__.py	22	__all__ = ["Conv2d", "Linear", "LoHaConfig", "LoHaLayer", "LoHaModel"]	CODE
LOW	src/peft/tuners/lokr/__init__.py	22	__all__ = ["Conv2d", "Linear", "LoKrConfig", "LoKrLayer", "LoKrModel"]	CODE
LOW	src/peft/tuners/vera/__init__.py	22	__all__ = ["Linear", "VeraConfig", "VeraLayer", "VeraModel"]	CODE
LOW	src/peft/tuners/ln_tuning/__init__.py	21	__all__ = ["LNTuningConfig", "LNTuningModel"]	CODE
LOW	src/peft/tuners/cpt/__init__.py	22	__all__ = ["CPTConfig", "CPTEmbedding"]	CODE
LOW	src/peft/tuners/delora/__init__.py	21	__all__ = ["DeloraConfig", "DeloraLayer", "DeloraLinear", "DeloraModel"]	CODE
LOW	src/peft/tuners/boft/__init__.py	22	__all__ = ["BOFTConfig", "BOFTLayer", "BOFTModel"]	CODE
LOW	src/peft/tuners/psoft/__init__.py	22	__all__ = ["Linear", "PsoftConfig", "PsoftLayer", "PsoftModel"]	CODE
LOW⚡	src/peft/tuners/psoft/layer.py	235	def _set_psoft_ab_cache_buffers(self, adapter_name: str, A: torch.Tensor, B: torch.Tensor) -> None:	CODE
LOW⚡	src/peft/tuners/psoft/layer.py	239	def update_layer(self, adapter_name: str, config: PsoftConfig, **kwargs: Any) -> None:	CODE
LOW	src/peft/tuners/beft/__init__.py	23	__all__ = ["BeftConfig", "BeftLayer", "BeftModel", "Linear"]	CODE
LOW	src/peft/tuners/waveft/__init__.py	22	__all__ = ["WaveFTConfig", "WaveFTLayer", "WaveFTLinear", "WaveFTModel"]	CODE
LOW	src/peft/tuners/c3a/__init__.py	21	__all__ = ["C3AConfig", "C3ALayer", "C3ALinear", "C3AModel"]	CODE
LOW	src/peft/tuners/p_tuning/__init__.py	21	__all__ = ["PromptEncoder", "PromptEncoderConfig", "PromptEncoderReparameterizationType"]	CODE
LOW	src/peft/tuners/adamss/__init__.py	22	__all__ = [	CODE
LOW	src/peft/tuners/adamss/model.py	157	def update_and_allocate(self, global_step: int) -> None:	CODE
LOW	src/peft/tuners/hira/__init__.py	23	__all__ = [	CODE
LOW	src/peft/tuners/hira/layer.py	307	def update_layer(self, adapter_name, r, config: HiraConfig, **kwargs) -> None:	CODE
LOW	src/peft/tuners/mixed/__init__.py	18	__all__ = ["COMPATIBLE_TUNER_TYPES", "MixedModel"]	CODE
LOW	src/peft/tuners/mixed/model.py	206	def set_adapter(self, adapter_name: Union[str, list[str]], inference_mode: bool = False) -> None:	CODE
LOW	src/peft/tuners/shadow/__init__.py	22	__all__ = ["DetachedShadowModel", "ShadowCache", "ShadowCarrier", "ShadowConfig", "ShadowLayer", "ShadowModel"]	CODE
LOW	src/peft/tuners/shadow/model.py	165	def _set_config_attr(config: Any, names: tuple[str, ...], value: int) -> None:	CODE
LOW	src/peft/tuners/shadow/model.py	819	def set_adapter(self, adapter_name: str \| list[str], inference_mode: bool = False) -> None:	CODE
39 more matches not shown…

Self-Referential Comments27 hits · 80 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/test_trainable_tokens.py	1257	# Create a composite model with two BART sub-models	COMMENT
MEDIUM	tests/test_lora_ga.py	130	# Create a fresh model and load the saved state	COMMENT
MEDIUM	tests/test_helpers.py	669	# Create a minimal dummy MoE model with fused 3D weights	COMMENT
MEDIUM	tests/test_custom_models.py	4974	# Create a weighted adapter combining both adapters and check that its output is same as setting multiple	COMMENT
MEDIUM	method_comparison/app_embed.py	58	# Create an empty figure.	COMMENT
MEDIUM	method_comparison/app.py	57	# Create an empty figure.	COMMENT
MEDIUM	examples/boft_controlnet/train_controlnet.py	537	# Create the pipeline using using the trained modules and save it.	COMMENT
MEDIUM	examples/arrow_multitask/arrow_phi3_mini.py	362	# Creating the Arrow model	COMMENT
MEDIUM	…orthogonal_subspace_learning/osf_continual_learning.py	57	# Create a simple dataloader	COMMENT
MEDIUM	examples/deft_dreambooth/train_dreambooth.py	1072	# Create the pipeline using using the trained modules and save it.	COMMENT
MEDIUM	examples/oft_dreambooth/train_dreambooth.py	1079	# Create the pipeline using using the trained modules and save it.	COMMENT
MEDIUM	examples/hra_dreambooth/train_dreambooth.sh	4	# Define the UNIQUE_TOKEN, CLASS_TOKENs, and SUBJECT_NAMES	COMMENT
MEDIUM	examples/stable_diffusion/train_dreambooth.py	1239	# Create the pipeline using using the trained modules and save it.	COMMENT
MEDIUM	examples/stable_diffusion/convert_sd_adapter_to_peft.py	409	# Create a corresponding adapter info	COMMENT
MEDIUM	examples/lora_dreambooth/train_dreambooth.py	1071	# Create the pipeline using using the trained modules and save it.	COMMENT
MEDIUM	examples/boft_dreambooth/train_dreambooth.sh	5	# Define the UNIQUE_TOKEN, CLASS_TOKENs, and SUBJECT_NAMES	COMMENT
MEDIUM	examples/lora_ga_finetuning/lora_ga_finetuning.py	167	# Create a simple DataLoader for gradient estimation	COMMENT
MEDIUM	src/peft/utils/other.py	1698	# Create the causal mask with fixed shape in advance, to reduce recompilations. If the function to create	COMMENT
MEDIUM	src/peft/tuners/tuners_utils.py	2265	# Create a reverse lookup for other_module_names to quickly check suffix matches	COMMENT
MEDIUM	src/peft/tuners/boft/layer.py	178	# Create a mask with 1s for matrices to be replaced with identity and 0s otherwise	COMMENT
MEDIUM	src/peft/tuners/waveft/layer.py	133	# Create a padded spectrum matrix with additional rows and columns	COMMENT
MEDIUM	src/peft/tuners/waveft/layer.py	144	# Create the padded dense spectrum matrix	COMMENT
MEDIUM	src/peft/tuners/waveft/waverec2d.py	6	# This file contains code derived from PyTorch-Wavelet-Toolbox:	COMMENT
MEDIUM⚡	src/peft/tuners/adaption_prompt/model.py	173	# The following classmethods mirror the state dict serialization hooks of BaseTuner, which AdaptionPromptModel	COMMENT
MEDIUM	src/peft/tuners/osf/model.py	103	# Create a new or update an existing OSF layer in place	COMMENT
MEDIUM	src/peft/tuners/xlora/model.py	244	# Create an empty LoraModel	COMMENT
MEDIUM	src/peft/tuners/adalora/model.py	371	"""This method is not supported for AdaLoRA, use LoRA instead."""	STRING

Excessive Try-Catch Wrapping64 hits · 66 pts

Severity	File	Line	Snippet	Context
LOW	tests/testing_utils.py	374	except Exception:	CODE
MEDIUM	tests/test_initialization.py	317	def test_lora_init_orthogonal_half_precision_dtype(self, dtype):	CODE
LOW	tests/test_initialization.py	320	except Exception:	CODE
LOW	tests/test_arrow.py	336	except Exception:	CODE
LOW	tests/test_custom_models.py	2640	except Exception:	CODE
LOW	tests/test_custom_models.py	2680	except Exception:	CODE
LOW	tests/test_custom_models.py	2719	except Exception:	CODE
LOW	tests/test_custom_models.py	2758	except Exception:	CODE
LOW	tests/regression/test_state_dict.py	584	except Exception:	CODE
LOW	method_comparison/app.py	188	except Exception as exc:	CODE
LOW	method_comparison/app.py	198	except Exception as exc:	CODE
LOW	method_comparison/app.py	232	except Exception as exc:	CODE
LOW	method_comparison/app.py	393	except Exception as exc:	CODE
LOW	method_comparison/app.py	443	except Exception as exc:	CODE
LOW	method_comparison/app.py	496	except Exception as e:	CODE
LOW	method_comparison/app.py	518	except Exception as e:	CODE
LOW	method_comparison/image-gen/run.py	572	except Exception as exc:	CODE
LOW	method_comparison/image-gen/run.py	701	except Exception as exc:	CODE
LOW	method_comparison/image-gen/utils.py	310	except Exception as exc:	CODE
LOW	method_comparison/image-gen/utils.py	318	except Exception as exc:	CODE
LOW⚡	method_comparison/image-gen/utils.py	334	except Exception as exc:	CODE
LOW⚡	method_comparison/image-gen/utils.py	342	except Exception as exc:	CODE
LOW⚡	method_comparison/image-gen/utils.py	350	except Exception as exc:	CODE
LOW⚡	method_comparison/image-gen/utils.py	358	except Exception as exc:	CODE
LOW	method_comparison/image-gen/evaluate.py	139	except Exception as exc:	CODE
LOW	method_comparison/image-gen/evaluate.py	261	except Exception as exc:	CODE
LOW	method_comparison/text_generation_benchmark/run.py	214	except Exception as exc:	CODE
LOW	method_comparison/text_generation_benchmark/run.py	299	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/run.py	407	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/utils.py	365	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/utils.py	381	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/utils.py	389	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/utils.py	518	except Exception as exc:	CODE
LOW	method_comparison/MetaMathQA/utils.py	526	except Exception as exc:	CODE
LOW	examples/ephemeral_gpu_offloading/load_with_dora.py	75	except Exception as e:	CODE
LOW	examples/ephemeral_gpu_offloading/load_with_dora.py	80	except Exception as e:	CODE
LOW	examples/qalora_finetuning/qalora_gptq_finetuning.py	69	except Exception as e:	CODE
LOW	examples/adamss_finetuning/test_adamss_quick.py	169	except Exception as e:	CODE
MEDIUM	scripts/check_doc_coverage.py	564	print(f"Error: source path not found: {args.src}", file=sys.stderr)	CODE
MEDIUM	scripts/check_doc_coverage.py	567	print(f"Error: docs path not found: {args.docs}", file=sys.stderr)	CODE
LOW	scripts/check_doc_coverage.py	271	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	283	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	303	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	326	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	362	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	394	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	460	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	514	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	580	except Exception as exc:	CODE
LOW	scripts/generate_method_capabilities.py	588	except Exception as exc:	CODE
MEDIUM	scripts/generate_method_capabilities.py	352	def run(self) -> Finding:	CODE
LOW	src/peft/config.py	116	except Exception:	STRING
LOW	src/peft/config.py	262	except Exception as exc:	STRING
LOW	src/peft/config.py	326	except Exception:	STRING
MEDIUM	src/peft/peft_model.py	201	def active_adapters(self) -> list[str]:	CODE
LOW	src/peft/peft_model.py	1154	except Exception: # something went wrong, roll back	CODE
LOW	src/peft/peft_model.py	2207	except Exception:	CODE
LOW	src/peft/peft_model.py	2569	except Exception:	CODE
LOW	src/peft/mixed_model.py	245	except Exception: # something went wrong, roll back	CODE
LOW	src/peft/helpers.py	163	except Exception:	CODE
4 more matches not shown…

AI Structural Patterns60 hits · 56 pts

Severity	File	Line	Context
LOW	tests/testing_utils.py	287	CODE
LOW	tests/test_lora_variants.py	75	CODE
LOW	tests/test_torch_compile.py	146	CODE
LOW	method_comparison/text_generation_benchmark/run_base.py	122	CODE
LOW	method_comparison/text_generation_benchmark/utils.py	279	CODE
LOW	examples/boft_controlnet/utils/pipeline_controlnet.py	166	CODE
LOW	examples/boft_controlnet/utils/unet_2d_condition.py	38	CODE
LOW	examples/waveft_finetuning/waveft_finetuning.py	30	CODE
LOW	examples/hira_finetuning/hira_finetuning.py	26	CODE
LOW	examples/glora_finetuning/glora_finetuning.py	26	CODE
LOW	examples/olora_finetuning/olora_finetuning.py	30	CODE
LOW	examples/shira_finetuning/shira_finetuning.py	31	CODE
LOW	examples/corda_finetuning/datautils.py	95	CODE
LOW	scripts/train_memory.py	106	CODE
LOW	src/peft/peft_model.py	1864	CODE
LOW	src/peft/peft_model.py	2047	CODE
LOW	src/peft/peft_model.py	2391	CODE
LOW	src/peft/peft_model.py	2718	CODE
LOW	src/peft/peft_model.py	2948	CODE
LOW	src/peft/peft_model.py	3019	CODE
LOW	src/peft/utils/incremental_pca.py	46	CODE
LOW	src/peft/tuners/hra/layer.py	268	CODE
LOW	src/peft/tuners/hra/layer.py	458	CODE
LOW	src/peft/tuners/loha/layer.py	246	CODE
LOW	src/peft/tuners/lokr/layer.py	314	CODE
LOW	src/peft/tuners/vera/layer.py	289	CODE
LOW	src/peft/tuners/delora/layer.py	267	CODE
LOW	src/peft/tuners/boft/layer.py	651	CODE
LOW	src/peft/tuners/boft/layer.py	994	CODE
LOW	src/peft/tuners/waveft/layer.py	289	CODE
LOW	src/peft/tuners/c3a/layer.py	202	CODE
LOW	src/peft/tuners/frod/layer.py	346	CODE
LOW	src/peft/tuners/road/layer.py	397	CODE
LOW	src/peft/tuners/fourierft/layer.py	191	CODE
LOW	src/peft/tuners/vblora/layer.py	253	CODE
LOW	src/peft/tuners/gralora/layer.py	387	CODE
LOW	src/peft/tuners/adaption_prompt/layer.py	139	CODE
LOW	src/peft/tuners/unilora/layer.py	253	CODE
LOW	src/peft/tuners/tinylora/layer.py	418	CODE
LOW	src/peft/tuners/miss/layer.py	372	CODE
LOW	src/peft/tuners/ia3/layer.py	331	CODE
LOW	src/peft/tuners/xlora/layer.py	131	CODE
LOW	src/peft/tuners/xlora/layer.py	236	CODE
LOW	src/peft/tuners/pvera/layer.py	320	CODE
LOW	src/peft/tuners/lora/variants.py	248	CODE
LOW	src/peft/tuners/lora/variants.py	347	CODE
LOW	src/peft/tuners/lora/variants.py	436	CODE
LOW	src/peft/tuners/lora/variants.py	891	CODE
LOW	src/peft/tuners/lora/variants.py	1220	CODE
LOW	src/peft/tuners/lora/variants.py	1538	CODE
LOW	src/peft/tuners/lora/eva.py	655	CODE
LOW	src/peft/tuners/lora/layer.py	744	CODE
LOW	src/peft/tuners/lora/layer.py	2163	CODE
LOW	src/peft/tuners/randlora/layer.py	346	CODE
LOW	src/peft/tuners/oft/awq.py	69	CODE
LOW	src/peft/tuners/oft/gptq.py	67	CODE
LOW	src/peft/tuners/oft/layer.py	685	CODE
LOW	src/peft/tuners/oft/layer.py	938	CODE
LOW	src/peft/tuners/deft/layer.py	353	CODE
LOW	src/peft/tuners/poly/layer.py	161	CODE

Redundant / Tautological Comments28 hits · 46 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_xlora.py	317	# Set it to 2 and make sure it actually is	COMMENT
LOW⚡	tests/test_xlora.py	322	# Set it to None and make sure it is 1/n	COMMENT
LOW⚡	tests/test_lorafa.py	303	# Check if lora_A weights have not changed	COMMENT
LOW⚡	tests/test_lorafa.py	308	# Check if lora_B weights are non-zero	COMMENT
LOW⚡	tests/test_lorafa.py	334	# Check if lora_A weights have not changed	COMMENT
LOW⚡	tests/test_lorafa.py	339	# Check if lora_B weights are non-zero and if they follow the expected relation	COMMENT
LOW	tests/test_lorafa.py	371	# Check if all non-LoRA params also follow the expected relation	COMMENT
LOW⚡	tests/test_config.py	451	# Check if the user supplied timing values that will certainly fail because it breaks	STRING
LOW	method_comparison/MetaMathQA/utils.py	450	# Check if the string represents a fraction.	COMMENT
LOW	examples/waveft_finetuning/waveft_finetuning.py	50	# Set device_map to the right place when enabling DDP.	COMMENT
LOW	examples/olora_finetuning/olora_finetuning.py	51	# Set device_map to the right place when enabling DDP.	COMMENT
LOW	examples/alora_finetuning/alora_finetuning.py	186	# Print results	COMMENT
LOW	examples/qalora_finetuning/qalora_gptq_finetuning.py	50	# Check if the model has GPTQ quantization attributes	COMMENT
LOW	examples/qalora_finetuning/qalora_gptq_finetuning.py	78	# Check if we already have a cached quantized version	COMMENT
LOW	examples/shira_finetuning/shira_finetuning.py	49	# Set device_map to the right place when enabling DDP.	COMMENT
LOW	examples/sft/utils.py	194	# Check if the model is quantized	COMMENT
LOW	…s/int8_training/peft_adalora_whisper_large_training.py	707	# Check if rank_pattern exists before calling update_and_allocate	COMMENT
LOW⚡	src/peft/utils/other.py	1605	# Check if this is the embedding layer (use endswith to allow flexible matching)	COMMENT
LOW⚡	src/peft/utils/other.py	1610	# Check if this target layer matches any tied module (considering nested structures)	COMMENT
LOW⚡	src/peft/utils/other.py	1618	# Check if all tied layers have the same indices	COMMENT
LOW	src/peft/tuners/tuners_utils.py	2282	# Check if adding this suffix covers the item	COMMENT
LOW	src/peft/tuners/adamss/model.py	98	# Check if already an Adamss layer	COMMENT
LOW	src/peft/tuners/hira/layer.py	176	fan_in_fan_out: bool = False, # Set this to True if the layer to replace stores weight like (fan_in, fan_out)	CODE
LOW	src/peft/tuners/hira/layer.py	297	fan_in_fan_out: bool = False, # Set this to True if the layer to replace stores weight like (fan_in, fan_out)	CODE
LOW	src/peft/tuners/adalora/config.py	94	# Check if 'r' has been set to a non-default value	COMMENT
LOW	src/peft/tuners/lora/loraga.py	136	# Check if any supported layers were found	COMMENT
LOW	src/peft/tuners/oft/layer.py	546	fan_in_fan_out: bool = False, # Set this to True if the layer to replace stores weight like (fan_in, fan_out)	CODE
LOW	src/peft/tuners/oft/layer.py	701	fan_in_fan_out: bool = False, # Set this to True if the layer to replace stores weight like (fan_in, fan_out)	CODE

AI Slop Vocabulary17 hits · 38 pts

Severity	File	Line	Snippet	Context
MEDIUM	tests/test_trainable_tokens.py	716	# for this we can leverage the fact that emb_out(1/emb_in(x)) is embed_dim on the	COMMENT
MEDIUM	tests/test_trainable_tokens.py	753	# for this we can leverage the fact that emb_out(1/emb_in(x)) is embed_dim on the	COMMENT
MEDIUM	tests/test_trainable_tokens.py	814	# for this we can leverage the fact that emb_out(1/emb_in(x)) is embed_dim on the	COMMENT
MEDIUM⚡	tests/test_gpu_examples.py	5868	# increase the scale of the LoRA weights so that the difference between adapters is more pronounced, making	COMMENT
MEDIUM⚡	tests/test_initialization.py	1562	# some arguments don't work in conjunction with lora_bias and should raise	COMMENT
LOW	tests/test_decoder_models.py	566	# a malicious prompt tuning model. This is because PEFT would just pass the on the tokenizer_kwargs defined in	COMMENT
LOW	tests/test_other.py	464	# aware that some of the passing tests may just pass owing to randomness.	COMMENT
MEDIUM	tests/test_custom_models.py	7821	# robust and minimal FSDP setup	COMMENT
LOW	examples/dora_finetuning/dora_finetuning.py	73	use_dora=use_dora, # to use Dora OR compare to Lora just set the --use_dora	CODE
MEDIUM	src/peft/peft_model.py	1035	# TODO: consider replacing this patching of methods with a more robust mechanism: setting a flag and	COMMENT
LOW	src/peft/utils/other.py	1811	# technically it would be sufficient to just return candidates since that contains all the keys of	COMMENT
LOW	src/peft/utils/transformers_weight_conversion.py	493	# If, for any reason, we cannot apply conversion, we just return the PEFT config as is.	COMMENT
MEDIUM	src/peft/tuners/tuners_utils.py	921	# TODO: check if this the most robust way	STRING
LOW	src/peft/tuners/tuners_utils.py	323	# model, just add a `peft_config` dict attribute to your model.	STRING
LOW	src/peft/tuners/adamss/layer.py	392	# so just call base_layer directly (no adapter delta needed).	COMMENT
LOW	src/peft/tuners/lora/conversion.py	328	# effective rank is the same and target_modules are a set, just add the name	COMMENT
MEDIUM	src/peft/tuners/lora/layer.py	391	# check without importing bitsandbytes and robust to bnb_4bit_quant_storage=float*	COMMENT

Docstring Block Structure6 hits · 30 pts

Severity	File	Line	Snippet	Context
HIGH	examples/boft_controlnet/utils/pipeline_controlnet.py	196	Function invoked when calling the pipeline for generation. Args: prompt (`str` or `List[st	STRING
HIGH	examples/qalora_finetuning/qalora_gptq_finetuning.py	109	Tokenize text data and prepare it for language modeling. Args: examples: Dataset examples with 'text'	STRING
HIGH	src/peft/peft_model.py	3207	Get the status of each adapter layer in the model. This function returns a list of `TunerLayerStatus` dataclass ins	STRING
HIGH	src/peft/peft_model.py	3341	Get the status of tuners of the model. This function returns a `TunerModelStatus` dataclass instance, which contain	STRING
HIGH	src/peft/tuners/tuners_utils.py	386	A private method to prepare the adapter config. For transformers based models, if `peft_config.target_	STRING
HIGH	src/peft/tuners/tuners_utils.py	2213	Find the minimal set of target modules that is sufficient to separate them from the other modules. Sometimes, a ver	STRING

Structural Annotation Overuse13 hits · 27 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_adamss_asa.py	107	# Step 1: train so B becomes non-zero	COMMENT
LOW⚡	tests/test_adamss_asa.py	113	# Step 2: now gradients for A should be non-zero	COMMENT
LOW⚡	tests/test_adamss_asa.py	123	# Step 3: another update should change scores via EMA	COMMENT
LOW⚡	tests/test_adamss_asa.py	166	# Step 0: train to make B non-zero	COMMENT
LOW	examples/lora_ga_finetuning/README.md	38	# Step 1: Create LoRA-GA config	COMMENT
LOW⚡	examples/lora_ga_finetuning/README.md	54	# Step 2: Preprocess with LoRA-GA to estimate gradients	COMMENT
LOW⚡	examples/lora_ga_finetuning/README.md	57	# Step 3: Create PEFT model with LoRA-GA initialization	COMMENT
LOW⚡	examples/lora_ga_finetuning/README.md	60	# Step 4: Train normally	COMMENT
LOW⚡	examples/lora_ga_finetuning/README.md	68	# Step 5: Save the trained adapter	COMMENT
LOW	examples/lora_ga_finetuning/README.md	211	### Step 1: Estimate gradients with full-precision model	COMMENT
LOW	examples/lora_ga_finetuning/README.md	250	### Step 2: Load quantized model and apply LoRA-GA	COMMENT
LOW⚡	src/peft/tuners/adamss/model.py	193	# Step 1: accumulate importance EVERY step during warmup	COMMENT
LOW⚡	src/peft/tuners/adamss/model.py	198	# Step 2: at mask intervals → schedule, global mask, then reset	COMMENT

Verbosity Indicators7 hits · 15 pts

Severity	File	Line	Snippet	Context
LOW⚡	tests/test_adamss_asa.py	107	# Step 1: train so B becomes non-zero	COMMENT
LOW⚡	tests/test_adamss_asa.py	113	# Step 2: now gradients for A should be non-zero	COMMENT
LOW⚡	tests/test_adamss_asa.py	123	# Step 3: another update should change scores via EMA	COMMENT
LOW⚡	tests/test_adamss_asa.py	166	# Step 0: train to make B non-zero	COMMENT
LOW	src/peft/import_utils.py	104	# We need to check if `xla_device` can be found, will raise a RuntimeError if not	COMMENT
LOW⚡	src/peft/tuners/adamss/model.py	193	# Step 1: accumulate importance EVERY step during warmup	COMMENT
LOW⚡	src/peft/tuners/adamss/model.py	198	# Step 2: at mask intervals → schedule, global mask, then reset	COMMENT

Cross-Language Confusion5 hits · 12 pts

Severity	File	Line	Snippet	Context
HIGH	src/peft/peft_model.py	1789	trainable params: 370178 \|\| all params: 108680450 \|\| trainable%: 0.3406113979101117	STRING
HIGH	src/peft/peft_model.py	2037	trainable params: 1843200 \|\| all params: 775873280 \|\| trainable%: 0.23756456724479544	STRING
HIGH	src/peft/peft_model.py	2378	trainable params: 884736 \|\| all params: 223843584 \|\| trainable%: 0.3952474242013566	STRING
HIGH	src/peft/peft_model.py	2647	trainable params: 370178 \|\| all params: 108680450 \|\| trainable%: 0.3406113979101117	STRING
HIGH	src/peft/peft_model.py	2877	trainable params: 592900 \|\| all params: 108312580 \|\| trainable%: 0.5473971721475013	STRING

Slop Phrases7 hits · 10 pts

Severity	File	Line	Snippet	Context
LOW	tests/regression/test_regression.py	189	model output, and load_base_model(name) to load the base model. Don't forget to fix the seed in load_base_model.	STRING
MEDIUM	examples/beft_finetuning/beft_finetuning.py	68	# you can try target_modules=["v"], target_modules=["q"], target_modules=["k"]	COMMENT
LOW	…ture_extraction/peft_lora_embedding_semantic_search.py	157	# make sure to pop weight so that corresponding model is not saved again	COMMENT
MEDIUM	examples/arrow_multitask/arrow_phi3_mini.py	57	This script is mainly meant for demonstration purposes and lightweight evaluation,	STRING
MEDIUM	examples/alora_finetuning/alora_finetuning.py	168	Purely for demonstration purposes. See the [paper](https://huggingface.co/papers/2504.12397)	STRING
LOW	…s/int8_training/peft_adalora_whisper_large_training.py	325	# make sure to pop weight so that corresponding model is not saved again	COMMENT
LOW	src/peft/tuners/lora/intruders.py	115	# we're not modifying the weights since there are no intruders but we make sure to copy the	COMMENT

TODO Padding4 hits · 5 pts

Severity	File	Line	Snippet	Context
LOW	tests/testing_common.py	1872	# TODO: add tests to check if disabling adapters works after calling merge_adapter	STRING
LOW	src/peft/tuners/xlora/layer.py	114	# TODO: implement X-LoRA with Lora+Dora layers	COMMENT
LOW	src/peft/tuners/xlora/layer.py	165	# TODO: implement X-LoRA with Lora+Dora layers	COMMENT
LOW	src/peft/tuners/xlora/layer.py	219	# TODO: implement X-LoRA with Lora+Dora layers	COMMENT

Fake / Example Data1 hit · 1 pts

Severity	File	Line	Snippet	Context
LOW	docs/source/package_reference/lora.md	742	invocation_string = "placeholder"	CODE

Analysis Overview

What These Metrics Mean

Score History

Severity Breakdown

Directory Score Breakdown

Pattern Findings