huggingface/trl

Pattern Findings

1723 matches across 15 categories. Click a row to expand file-level details.

Hyper-Verbose Identifiers752 hits · 684 pts

Severity	File	Line	Snippet
LOW	tests/test_skills_cli.py	26	def test_cmd_list_without_target(self, capsys):
LOW	tests/test_skills_cli.py	38	def test_cmd_list_with_target(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	51	def test_cmd_list_empty_target(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	61	def test_cmd_install_single_skill(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	73	def test_cmd_install_all_skills(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	85	def test_cmd_install_no_skill_or_all(self, capsys):
LOW	tests/test_skills_cli.py	95	def test_cmd_install_both_skill_and_all(self, capsys):
LOW	tests/test_skills_cli.py	105	def test_cmd_install_nonexistent_skill(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	116	def test_cmd_install_already_exists(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	130	def test_cmd_install_with_force(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	144	def test_cmd_uninstall_success(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	159	def test_cmd_uninstall_not_installed(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	170	def test_cmd_install_creates_target_directory(self, tmp_path, capsys):
LOW	tests/test_skills_cli.py	187	def test_cmd_uninstall_invalid_target(self, capsys):
LOW	tests/test_skills_cli.py	201	def test_add_skills_subcommands_creates_parsers(self):
LOW	tests/test_skills_cli.py	222	def test_list_command_optional_target(self):
LOW	tests/test_skills_cli.py	236	def test_install_command_requires_target(self):
LOW	tests/test_skills_cli.py	282	def test_default_scope_is_project(self):
LOW	tests/test_rloo_trainer.py	102	def test_train_with_num_generations_eval(self):
LOW	tests/test_rloo_trainer.py	126	def test_train_multiple_iterations(self):
LOW	tests/test_rloo_trainer.py	233	def test_train_peft_with_gradient_checkpointing(self):
LOW	tests/test_rloo_trainer.py	269	def test_train_different_reward_model(self):
LOW	tests/test_rloo_trainer.py	308	def test_train_reward_func_standard(self):
LOW	tests/test_rloo_trainer.py	342	def test_train_reward_func_conversational(self):
LOW	tests/test_rloo_trainer.py	377	def test_train_multiple_reward_funcs(self):
LOW	tests/test_rloo_trainer.py	415	def test_train_sync_and_async_reward_funcs(self):
LOW	tests/test_rloo_trainer.py	464	def non_applicable_reward_func(completions, **kwargs):
LOW	tests/test_rloo_trainer.py	500	def test_train_multiple_reward_funcs_with_weights(self):
LOW	tests/test_rloo_trainer.py	544	def test_reward_metric_reflects_reward_weights(self):
LOW	tests/test_rloo_trainer.py	581	def test_train_multiple_mixed_reward_funcs(self):
LOW	tests/test_rloo_trainer.py	615	def test_train_reward_func_additional_column(self):
LOW	tests/test_rloo_trainer.py	655	def test_train_with_sync_ref_model(self):
LOW	tests/test_rloo_trainer.py	720	def test_train_with_pad_to_multiple_of(self):
LOW	tests/test_rloo_trainer.py	800	def test_train_vllm_structured_outputs(self):
LOW	tests/test_rloo_trainer.py	832	def test_train_with_additional_generation_kwargs(self):
LOW	tests/test_rloo_trainer.py	869	def test_train_vllm_with_additional_generation_kwargs(self):
LOW	tests/test_rloo_trainer.py	905	def test_train_with_normalized_advantages(self):
LOW	tests/test_rloo_trainer.py	935	def test_train_with_clipped_rewards(self):
LOW	tests/test_rloo_trainer.py	966	def test_train_with_mask_truncated_completions(self, mock_generate):
LOW	tests/test_rloo_trainer.py	1015	def test_train_with_mask_truncated_completions_all_masked(self):
LOW	tests/test_rloo_trainer.py	1053	def test_warning_raised_all_rewards_none(self, caplog):
LOW	tests/test_rloo_trainer.py	1082	def test_train_num_generations_larger_than_batch_size(self):
LOW	tests/test_rloo_trainer.py	1112	def test_train_multiple_dataloader_workers(self):
LOW	tests/test_rloo_trainer.py	1153	def test_train_with_generation_kwargs(self):
LOW	tests/test_rloo_trainer.py	1184	def test_train_with_reward_func_accessing_trainer_state(self):
LOW	tests/test_rloo_trainer.py	1209	def test_train_reward_func_with_log_extra(self):
LOW	tests/test_rloo_trainer.py	1235	def test_train_reward_func_with_log_metric(self):
LOW	tests/test_rloo_trainer.py	1262	def test_prepare_input_called_with_correct_data(self):
LOW	tests/test_rloo_trainer.py	1389	def test_train_vlm_with_pad_to_multiple_of(self):
LOW	tests/test_rloo_trainer.py	1433	def test_train_vlm_beta_non_zero(self, model_id):
LOW	tests/test_rloo_trainer.py	1575	def test_train_vlm_multi_image(self, model_id):
LOW	tests/test_rloo_trainer.py	1608	def test_train_with_chat_template_kwargs(self):
LOW	tests/test_rloo_trainer.py	1638	def test_mismatched_reward_processing_classes_length(self):
LOW	tests/test_rloo_trainer.py	1664	def test_correct_reward_processing_classes_list(self):
LOW	tests/test_rloo_trainer.py	1695	def test_single_reward_model_with_single_processing_class(self):
LOW	tests/test_utils.py	181	def test_pad_to_multiple_of_side_left(self):
LOW	tests/test_utils.py	189	def test_pad_to_multiple_of_no_extra_padding(self):
LOW	tests/test_utils.py	199	def test_hash_module_deterministic_across_order(self):
LOW	tests/test_utils.py	216	def test_hash_module_changes_with_value(self):
LOW	tests/test_utils.py	224	def test_hash_module_includes_dtype(self):
692 more matches not shown…

Over-Commented Block336 hits · 330 pts

Severity	File	Line	Snippet
LOW	tests/test_skills_cli.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_rloo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/conftest.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_callbacks.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_sft_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/testing_constants.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_chat_template_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_grpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_model_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/testing_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_data_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/__init__.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_vllm_client_server.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_rich_progress_callback.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_activation_offloading.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_rewards.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_cli_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_skills.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_dpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_cli.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/test_reward_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/_openreward_echo_env.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_utils.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_gkd_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_nash_md_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_minillm_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_tpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_gspo_token_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_sdft_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/__init__.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_sdpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_xpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_gold_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_prm_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_online_dpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_openreward.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_openreward.py	201	# If a task spec ever shipped a `prompt` key, the metadata loop must
LOW	tests/experimental/test_async_grpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_ssd_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_orpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_kto_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_bco_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_cpo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_distillation_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_modeling_value_head.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_dppo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_ppo_trainer.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/experimental/test_merge_model_callback.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/invariant/__init__.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/invariant/test_invariant.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/distributed/__init__.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	tests/distributed/test_distributed.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	examples/accelerate_configs/alst_ulysses_4gpu.yaml	1	# ALST/Ulysses Sequence Parallelism with 2D Parallelism (DP + SP) for 4 GPUs
LOW	examples/datasets/tldr.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	examples/datasets/deepmath_103k.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	examples/datasets/ultrafeedback.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	examples/datasets/lm-human-preferences-sentiment.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
LOW	examples/datasets/prm800k.py	1	# Copyright 2020-2026 The HuggingFace Team. All rights reserved.
276 more matches not shown…

Cross-File Repetition51 hits · 255 pts

Severity	File	Snippet
HIGH	AGENTS.md	brief one-line description of what this does. args: param1 (`str`): description of required param. param2 (`int`, *optio
HIGH	CLAUDE.md	brief one-line description of what this does. args: param1 (`str`): description of required param. param2 (`int`, *optio
HIGH	.cursor/BUGBOT.md	brief one-line description of what this does. args: param1 (`str`): description of required param. param2 (`int`, *optio
HIGH	.ai/AGENTS.md	brief one-line description of what this does. args: param1 (`str`): description of required param. param2 (`int`, *optio
HIGH	tests/test_rloo_trainer.py	reward function that rewards completions with more unique letters.
HIGH	tests/test_rloo_trainer.py	reward function that rewards completions with more unique letters.
HIGH	tests/test_grpo_trainer.py	reward function that rewards completions with more unique letters.
HIGH	tests/test_grpo_trainer.py	reward function that rewards completions with more unique letters.
HIGH	tests/test_grpo_trainer.py	reward function that rewards completions with more unique letters.
HIGH	tests/test_rloo_trainer.py	test that training works with vllm for generation with structured outputs.
HIGH	tests/test_grpo_trainer.py	test that training works with vllm for generation with structured outputs.
HIGH	tests/test_grpo_trainer.py	test that training works with vllm for generation with structured outputs.
HIGH	tests/test_callbacks.py	test that bema callback respects the update frequency.
HIGH	tests/test_callbacks.py	test that bema callback respects the update frequency.
HIGH	tests/test_callbacks.py	test that bema callback respects the update frequency.
HIGH	tests/test_grpo_trainer.py	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	tests/test_vllm_client_server.py	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	tests/test_vllm_client_server.py	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	tests/test_vllm_client_server.py	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	tests/test_vllm_client_server.py	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	docs/source/grpo_trainer.md	multiplies two integers. args: a: the first integer. b: the second integer. returns: the product of the two integers.
HIGH	tests/test_data_utils.py	\ <\|start\|>system<\|message\|>you are hugginggpt. knowledge cutoff: 2024-06 current date: {strftime("%y-%m-%d")} reasoning
HIGH	tests/test_data_utils.py	\ <\|start\|>system<\|message\|>you are hugginggpt. knowledge cutoff: 2024-06 current date: {strftime("%y-%m-%d")} reasoning
HIGH	tests/test_data_utils.py	\ <\|start\|>system<\|message\|>you are hugginggpt. knowledge cutoff: 2024-06 current date: {strftime("%y-%m-%d")} reasoning
HIGH	tests/test_data_utils.py	\ <\|start\|>system<\|message\|>you are hugginggpt. knowledge cutoff: 2024-06 current date: {strftime("%y-%m-%d")} reasoning
HIGH	tests/test_cli_utils.py	test that config defaults are applied to all subparsers.
HIGH	tests/test_cli_utils.py	test that config defaults are applied to all subparsers.
HIGH	tests/test_cli_utils.py	test that config defaults are applied to all subparsers.
HIGH	tests/test_cli_utils.py	test that config defaults are applied to all subparsers.
HIGH	tests/experimental/test_ppo_trainer.py	test if we instantiate a model by adding `summary_drop_prob` to the config it will be added to the v_head
HIGH	tests/experimental/test_ppo_trainer.py	test if we instantiate a model by adding `summary_drop_prob` to the config it will be added to the v_head
HIGH	tests/experimental/test_ppo_trainer.py	test if we instantiate a model by adding `summary_drop_prob` to the config it will be added to the v_head
HIGH	tests/experimental/test_ppo_trainer.py	test if we instantiate a model by adding `summary_drop_prob` to the config it will be added to the v_head
HIGH	tests/experimental/test_ppo_trainer.py	simply creates a peft model and checks that it can be loaded.
HIGH	tests/experimental/test_ppo_trainer.py	simply creates a peft model and checks that it can be loaded.
HIGH	tests/experimental/test_ppo_trainer.py	simply creates a peft model and checks that it can be loaded.
HIGH	docs/source/openenv.md	make a guess in the wordle environment. args: guess: the guessed word, formatted as '[abcde]' returns: the feedback mess
HIGH	examples/scripts/openenv/multi_env.py	make a guess in the wordle environment. args: guess: the guessed word, formatted as '[abcde]' returns: the feedback mess
HIGH	examples/scripts/openenv/wordle.py	make a guess in the wordle environment. args: guess: the guessed word, formatted as '[abcde]' returns: the feedback mess
HIGH	trl/experimental/cpo/cpo_trainer.py	generate samples from the model and reference model for the given batch of inputs.
HIGH	trl/experimental/bco/bco_trainer.py	generate samples from the model and reference model for the given batch of inputs.
HIGH	trl/experimental/orpo/orpo_trainer.py	generate samples from the model and reference model for the given batch of inputs.
HIGH	trl/experimental/cpo/cpo_trainer.py	overriding built-in evaluation loop to store metrics for each batch. prediction/evaluation loop, shared by `trainer.eval
HIGH	trl/experimental/bco/bco_trainer.py	overriding built-in evaluation loop to store metrics for each batch. prediction/evaluation loop, shared by `trainer.eval
HIGH	trl/experimental/orpo/orpo_trainer.py	overriding built-in evaluation loop to store metrics for each batch. prediction/evaluation loop, shared by `trainer.eval
HIGH	trl/experimental/cpo/cpo_trainer.py	log `logs` on the various objects watching training, including stored metrics. args: logs (`dict[str, float]`): the valu
HIGH	trl/experimental/bco/bco_trainer.py	log `logs` on the various objects watching training, including stored metrics. args: logs (`dict[str, float]`): the valu
HIGH	trl/experimental/orpo/orpo_trainer.py	log `logs` on the various objects watching training, including stored metrics. args: logs (`dict[str, float]`): the valu
HIGH	trl/experimental/dppo/dppo_trainer.py	tokenize prompts and extract images/multimodal fields for generation.
HIGH	trl/trainer/rloo_trainer.py	tokenize prompts and extract images/multimodal fields for generation.
HIGH	trl/trainer/grpo_trainer.py	tokenize prompts and extract images/multimodal fields for generation.

Decorative Section Separators60 hits · 176 pts

Severity	File	Line	Snippet
MEDIUM	tests/conftest.py	54	# ============================================================================
MEDIUM	tests/conftest.py	56	# ============================================================================
MEDIUM	tests/conftest.py	68	# ============================================================================
MEDIUM	examples/scripts/grpo_agent.py	184	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	186	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	231	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	233	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	260	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	262	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	267	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	269	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	284	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	286	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	298	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	300	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	310	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	312	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	315	# ------------------------
MEDIUM	examples/scripts/grpo_agent.py	317	# ------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	234	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	236	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	253	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	255	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	263	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/browsergym_llm.py	265	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	105	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	107	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	187	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	189	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	339	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	341	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	369	# ---------------------------------------------------------------------------
MEDIUM	examples/scripts/openenv/sudoku.py	371	# ---------------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	269	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	271	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	319	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	321	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	338	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	340	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	421	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	423	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	472	# ------------------------------------------------------------------
MEDIUM	trl/experimental/ssd/ssd_trainer.py	474	# ------------------------------------------------------------------
MEDIUM	trl/experimental/distillation/distillation_trainer.py	702	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	704	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	764	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	766	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	1022	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	1024	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	1576	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/distillation/distillation_trainer.py	1578	# ──────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/_spec.py	144	# ── public surface ──────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/_spec.py	251	# ── internals ───────────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/environment.py	101	# ────────────────────────────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/environment.py	204	# ── TRL contract ─────────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/environment.py	252	# ── helpers ──────────────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/environment.py	299	# ── dynamic tool binding ─────────────────────────────────────────────
MEDIUM	trl/experimental/openreward/environment.py	366	# ── small utilities ──────────────────────────────────────────────────
MEDIUM	trl/trainer/rloo_trainer.py	645	# <-───────> per_device_train_batch_size=3
MEDIUM	trl/trainer/grpo_trainer.py	930	# <-───────> per_device_train_batch_size=3

Self-Referential Comments60 hits · 174 pts

Severity	File	Line	Snippet
MEDIUM	tests/test_rloo_trainer.py	1648	# Create a single processing class (tokenizer)
MEDIUM	tests/test_rloo_trainer.py	1702	# Create a single processing class (tokenizer)
MEDIUM	tests/test_grpo_trainer.py	93	# Create a mock trainer with minimal setup
MEDIUM	tests/test_grpo_trainer.py	96	# Create a mock accelerator
MEDIUM	tests/test_grpo_trainer.py	100	# Create a minimal trainer instance just to access the method
MEDIUM	tests/test_grpo_trainer.py	2704	# Create a single processing class (tokenizer)
MEDIUM	tests/test_grpo_trainer.py	2758	# Create a single processing class (tokenizer)
MEDIUM	tests/test_grpo_trainer.py	3146	# Create a VLM processor
MEDIUM	tests/test_cli_utils.py	176	# Create the main parser
MEDIUM	tests/test_cli_utils.py	182	# Create a subparser for a specific command
MEDIUM	tests/test_cli_utils.py	202	# Create the main parser
MEDIUM	tests/test_cli_utils.py	208	# Create a subparser for a specific command
MEDIUM	tests/test_cli_utils.py	225	# Create the main parser
MEDIUM	tests/test_cli_utils.py	231	# Create a subparser for a specific command
MEDIUM	tests/test_cli_utils.py	247	# Create the main parser
MEDIUM	tests/test_cli_utils.py	253	# Create a subparser for a specific command
MEDIUM	tests/test_skills.py	188	# Create a file instead of directory
MEDIUM	tests/test_skills.py	199	# Create a skill with subdirectories
MEDIUM	tests/test_skills.py	264	# Create a skill with multiple files
MEDIUM	tests/test_skills.py	171	# Create a symlink
MEDIUM	tests/test_skills.py	217	# Create a skill
MEDIUM	tests/test_skills.py	435	# Create a real skill
MEDIUM	tests/test_cli.py	101	# Create a temporary config file
MEDIUM	tests/experimental/test_prm_trainer.py	126	# Define the input features
MEDIUM	tests/experimental/test_prm_trainer.py	150	# Define the input features
MEDIUM	tests/experimental/test_prm_trainer.py	173	# Define the input features
MEDIUM	tests/experimental/test_prm_trainer.py	197	# Define the input features
MEDIUM	tests/experimental/test_prm_trainer.py	221	# Define the input features
MEDIUM	scripts/generate_zen_multi_image_dataset.py	76	# Create the images
MEDIUM	scripts/generate_zen_multi_image_dataset.py	106	# Create the images
MEDIUM	scripts/generate_zen_multi_image_dataset.py	157	# Create the images
MEDIUM	scripts/generate_zen_multi_image_dataset.py	229	# Create the images
MEDIUM	trl/data_utils.py	741	# Create a new bin
MEDIUM	trl/experimental/merge_model_callback.py	37	# Create the repository if it doesn't exist
MEDIUM	trl/experimental/merge_model_callback.py	118	# Create the merge configuration dictionary
MEDIUM	trl/experimental/merge_model_callback.py	128	# Create the MergeConfiguration from the dictionary
MEDIUM	trl/experimental/merge_model_callback.py	137	# Create the TIES merge configuration dictionary
MEDIUM	trl/experimental/merge_model_callback.py	172	# Create the MergeConfiguration from the dictionary
MEDIUM	trl/experimental/merge_model_callback.py	181	# Create the DARE TIES merge configuration dictionary
MEDIUM	trl/experimental/merge_model_callback.py	216	# Create the MergeConfiguration from the dictionary
MEDIUM	trl/experimental/merge_model_callback.py	226	# Create the SLERP merge configuration dictionary
MEDIUM	trl/experimental/merge_model_callback.py	255	# Create the MergeConfiguration from the dictionary
MEDIUM	trl/experimental/utils.py	15	# This file contains utility classes and functions that are used across more than one experimental trainer or feature.
MEDIUM	trl/experimental/cpo/cpo_trainer.py	174	# Initialize this variable to False. This helps tracking the case when `peft_module_casting_to_bf16`
MEDIUM	trl/experimental/bco/bco_trainer.py	473	# Initialize this variable to False. This helps tracking the case when `peft_module_casting_to_bf16`
MEDIUM	trl/experimental/orpo/orpo_trainer.py	183	# Initialize this variable to False. This helps tracking the case when `peft_module_casting_to_bf16`
MEDIUM	trl/experimental/online_dpo/online_dpo_trainer.py	382	# Define the collator if not provided
MEDIUM	trl/experimental/online_dpo/online_dpo_trainer.py	886	# Create the output dict with required fields
MEDIUM	trl/experimental/gold/gold_trainer.py	1465	# Create a completion mask
MEDIUM	trl/experimental/prm/prm_trainer.py	337	# Create the label
MEDIUM	trl/experimental/ppo/ppo_trainer.py	394	# Define the collator if not provided
MEDIUM	trl/scripts/vllm_serve.py	109	# Create a stateless process group to manage communication between training processes and vLLM workers.
MEDIUM	trl/scripts/vllm_serve.py	491	# Define the endpoints for the model server
MEDIUM	trl/trainer/utils.py	160	# Create an output tensor filled with the padding value
MEDIUM	trl/trainer/utils.py	171	# Define the slices
MEDIUM	trl/trainer/utils.py	719	self.generator = torch.Generator() # Create a local random generator
MEDIUM	trl/trainer/utils.py	950	# This function is intentionally not used internally. It is provided as a utility for users whose datasets contain
MEDIUM	trl/trainer/callbacks.py	218	# Create a table per group
MEDIUM	trl/trainer/callbacks.py	693	# Create a new instance and load state_dict
MEDIUM	trl/trainer/grpo_trainer.py	494	# Create the environments and extract their methods to be used as tools. We create one environment per rollout

Unused Imports182 hits · 166 pts

Severity	File	Line
LOW	examples/scripts/openenv/browsergym_llm.py	64
LOW	examples/scripts/openenv/sudoku.py	84
LOW	examples/scripts/openenv/browsergym.py	50
LOW	trl/__init__.py	19
LOW	trl/__init__.py	77
LOW	trl/__init__.py	77
LOW	trl/__init__.py	77
LOW	trl/__init__.py	77
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	83
LOW	trl/__init__.py	97
LOW	trl/__init__.py	98
LOW	trl/__init__.py	98
LOW	trl/__init__.py	98
LOW	trl/__init__.py	98
LOW	trl/__init__.py	98
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/__init__.py	99
LOW	trl/experimental/utils.py	47
LOW	trl/experimental/ssd/__init__.py	15
LOW	trl/experimental/ssd/__init__.py	16
LOW	trl/experimental/ssd/ssd_eval.py	59
LOW	trl/experimental/ssd/ssd_trainer.py	23
LOW	trl/experimental/gfpo/__init__.py	15
LOW	trl/experimental/gfpo/__init__.py	16
LOW	trl/experimental/sdpo/__init__.py	15
LOW	trl/experimental/sdpo/__init__.py	16
LOW	trl/experimental/cpo/__init__.py	15
LOW	trl/experimental/cpo/__init__.py	16
LOW	trl/experimental/papo/__init__.py	16
LOW	trl/experimental/papo/__init__.py	17
122 more matches not shown…

Deep Nesting145 hits · 138 pts

Severity	File	Line
LOW	tests/test_utils.py	809
LOW	tests/test_utils.py	840
LOW	tests/test_grpo_trainer.py	3002
LOW	tests/test_grpo_trainer.py	3118
LOW	tests/test_grpo_trainer.py	3266
LOW	examples/datasets/math_shepherd.py	52
LOW	examples/scripts/sft_vlm_gemma3.py	78
LOW	examples/scripts/sft_vlm_gemma3.py	97
LOW	examples/scripts/grpo_agent.py	50
LOW	examples/scripts/grpo_agent.py	145
LOW	examples/scripts/sft_video_llm.py	68
LOW	examples/scripts/openenv/multi_env.py	127
LOW	examples/scripts/openenv/catch.py	205
LOW	examples/scripts/openenv/sudoku.py	208
LOW	examples/scripts/openenv/sudoku.py	290
LOW	examples/scripts/openenv/sudoku.py	374
LOW	examples/scripts/openenv/sudoku.py	543
LOW	scripts/log_reports.py	34
LOW	scripts/generate_tiny_models/_common.py	256
LOW	trl/chat_template_utils.py	355
LOW	trl/data_utils.py	33
LOW	trl/data_utils.py	127
LOW	trl/data_utils.py	922
LOW	trl/_compat.py	86
LOW	trl/import_utils.py	29
LOW	trl/experimental/merge_model_callback.py	82
LOW	trl/experimental/merge_model_callback.py	260
LOW	trl/experimental/utils.py	537
LOW	trl/experimental/utils.py	557
LOW	trl/experimental/utils.py	65
LOW	trl/experimental/utils.py	150
LOW	trl/experimental/gfpo/gfpo_trainer.py	69
LOW	trl/experimental/sdpo/sdpo_trainer.py	96
LOW	trl/experimental/sdpo/sdpo.py	163
LOW	trl/experimental/cpo/cpo_trainer.py	130
LOW	trl/experimental/cpo/cpo_trainer.py	446
LOW	trl/experimental/cpo/cpo_trainer.py	579
LOW	trl/experimental/cpo/cpo_trainer.py	643
LOW	trl/experimental/papo/papo_trainer.py	154
LOW	trl/experimental/bco/bco_trainer.py	407
LOW	trl/experimental/bco/bco_trainer.py	1030
LOW	trl/experimental/bco/bco_trainer.py	1233
LOW	trl/experimental/bco/bco_trainer.py	1362
LOW	trl/experimental/orpo/orpo_trainer.py	139
LOW	trl/experimental/orpo/orpo_trainer.py	431
LOW	trl/experimental/orpo/orpo_trainer.py	574
LOW	trl/experimental/online_dpo/online_dpo_trainer.py	169
LOW	trl/experimental/online_dpo/online_dpo_trainer.py	764
LOW	trl/experimental/online_dpo/online_dpo_trainer.py	840
LOW	trl/experimental/online_dpo/online_dpo_trainer.py	902
LOW	trl/experimental/online_dpo/online_dpo_trainer.py	1096
LOW	trl/experimental/bema_for_ref_model/callback.py	202
LOW	trl/experimental/sdft/sdft.py	133
LOW	trl/experimental/sdft/sdft_trainer.py	160
LOW	trl/experimental/distillation/distillation_trainer.py	233
LOW	trl/experimental/distillation/distillation_trainer.py	385
LOW	trl/experimental/distillation/distillation_trainer.py	671
LOW	trl/experimental/distillation/distillation_trainer.py	823
LOW	trl/experimental/distillation/distillation_trainer.py	865
LOW	trl/experimental/distillation/distillation_trainer.py	1049
85 more matches not shown…

Docstring Block Structure18 hits · 90 pts

Severity	File	Line	Snippet
HIGH	CONTRIBUTING.md	268	Replicate a string `n` times with a separator. Args: string (`str`): String to replicate.
HIGH	AGENTS.md	71	Brief one-line description of what this does. Args: param1 (`str`): Description of require
HIGH	CLAUDE.md	71	Brief one-line description of what this does. Args: param1 (`str`): Description of require
HIGH	.cursor/BUGBOT.md	71	Brief one-line description of what this does. Args: param1 (`str`): Description of require
HIGH	trl/chat_template_utils.py	356	Adds the appropriate response schema to the given tokenizer based on its chat template. At the time of initial
HIGH	trl/data_utils.py	35	Convert messages into a structured multimodal format and inject the provided images into the message contents.
HIGH	trl/data_utils.py	161	Check if the example is in a conversational format. Args: example (`dict[str, Any]`): A si
HIGH	trl/data_utils.py	341	If the example is in a conversational format, apply a chat template to it. Args: example (`dict[str, l
HIGH	trl/data_utils.py	505	Extracts the shared prompt from a preference data example, where the prompt is implicit within both the chosen and
HIGH	trl/data_utils.py	883	Check if the example is in a conversational format (from/value). Note that this format isn't recommended. Prefer
HIGH	trl/experimental/gold/gold_trainer.py	520	Merge probabilities based on alignment groups with corrected conditional probability handling. For a g
HIGH	trl/skills/skills.py	73	Resolve target to a concrete directory path. Converts semantic agent names (e.g., 'claude') with scope to actu
HIGH	trl/skills/skills.py	184	Install a skill to target directory. Args: skill_name (`str`): Name of skill to install. targe
HIGH	trl/skills/skills.py	251	Install a skill. Args: skill_name (`str`): Name of skill to install. target (`str \| Path`): Ag
HIGH	trl/skills/skills.py	295	Uninstall a skill from target directory. Args: skill_name (`str`): Name of skill to uninstall.
HIGH	trl/skills/skills.py	327	Uninstall a skill. Args: skill_name (`str`): Name of skill to uninstall. target (`str \| Path`)
HIGH	trl/trainer/utils.py	119	Pads a list of tensors to the same shape along the first dimension. Args: tensors (`list[torch.Tensor]
HIGH	.ai/AGENTS.md	71	Brief one-line description of what this does. Args: param1 (`str`): Description of require

Redundant / Tautological Comments54 hits · 70 pts

Severity	File	Line	Snippet
LOW	tests/test_chat_template_utils.py	45	# Check if special tokens are correctly set
LOW	tests/test_chat_template_utils.py	112	# Check if special tokens are correctly set
LOW	tests/test_grpo_trainer.py	3213	# Check if signature columns were set properly
LOW	tests/experimental/test_gkd_trainer.py	52	# Set temperature to 0 for deterministic output
LOW	tests/experimental/test_gkd_trainer.py	71	# Check if the generated texts start with the original prompts
LOW	tests/experimental/test_gkd_trainer.py	84	# Check if the two generations are identical
LOW	tests/experimental/test_online_dpo_trainer.py	228	# Check if training loss is available
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	45	# Check if the buffer contains the correct number of elements
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	48	# Check if the buffer maintains the min-heap property
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	67	# Check if the buffer contains the correct number of elements
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	70	# Check if the buffer maintains the min-heap property
LOW	…s/experimental/test_grpo_with_replay_buffer_trainer.py	90	# Check if the sampled elements are from the buffer
LOW	tests/experimental/test_ppo_trainer.py	750	# Check if critic weights have been updated
LOW	tests/experimental/test_ppo_trainer.py	757	# Check if policy weights have been updated
LOW	tests/experimental/test_ppo_trainer.py	812	# Check if critic weights have been updated
LOW	tests/experimental/test_ppo_trainer.py	819	# Check if policy weights have been updated - for PEFT we check the LoRA weights
LOW	tests/experimental/test_ppo_trainer.py	191	# Check if the weights are the same
LOW	tests/experimental/test_ppo_trainer.py	206	# Check if the weights are the same
LOW	tests/experimental/test_ppo_trainer.py	224	# Check if the weights are the same
LOW	tests/experimental/test_ppo_trainer.py	245	# Check if the weights are the same
LOW	tests/experimental/test_ppo_trainer.py	251	# Check if the trl model has the same keys as the transformers model
LOW	tests/experimental/test_ppo_trainer.py	290	# Check if the outputs are of the right size - here
LOW	tests/experimental/test_ppo_trainer.py	303	# Check if v head of the model has the same dropout as the config
LOW	tests/experimental/test_ppo_trainer.py	315	# Check if v head of the model has the same dropout as the config
LOW	tests/experimental/test_ppo_trainer.py	320	# Check if v head of the model has the same dropout as the config
LOW	tests/experimental/test_ppo_trainer.py	405	# Check if the outputs are of the right size - here
LOW	tests/experimental/test_ppo_trainer.py	418	# Check if v head of the model has the same dropout as the config
LOW	tests/experimental/test_ppo_trainer.py	430	# Check if v head of the model has the same dropout as the config
LOW	tests/experimental/test_ppo_trainer.py	435	# Check if v head of the model has the same dropout as the config
LOW	scripts/add_copyrights.py	61	# Check if the exact copyright header exists
LOW	.github/workflows/tests.yml	90	if: github.ref == 'refs/heads/main' && always() # Check if the branch is main
LOW	.github/workflows/tests.yml	146	if: github.ref == 'refs/heads/main' && always() # Check if the branch is main
LOW	.github/workflows/tests.yml	197	if: github.ref == 'refs/heads/main' && always() # Check if the branch is main
LOW	.github/workflows/tests.yml	252	if: github.ref == 'refs/heads/main' && always() # Check if the branch is main
LOW	.github/workflows/tests.yml	305	if: github.ref == 'refs/heads/main' && always() # Check if the branch is main
LOW	trl/experimental/utils.py	486	# Check if model is sharded (FSDP/DS-Zero3)
LOW	trl/experimental/gold/gold_trainer.py	1456	# Check if the tokenized prompt starts with the tokenized prompt+completion
LOW	trl/experimental/tpo/tpo_trainer.py	569	# Check if the tokenized prompt starts with the tokenized prompt+completion
LOW	trl/models/activation_offloading.py	221	# Check if tensor is on CPU (skip offloading)
LOW	trl/models/activation_offloading.py	226	# Check if tensor is too small
LOW	trl/models/activation_offloading.py	231	# Check if tensor is a parameter or buffer
LOW	trl/models/activation_offloading.py	238	# Check if tensor is an FP8 tensor (TorchAO) - skip offloading as they're already compressed
LOW	trl/models/activation_offloading.py	244	# Check if tensor storage is a model parameter (for FSDP compatibility)
LOW	trl/models/activation_offloading.py	286	# Check if tensor has broadcast dimensions (stride == 0)
LOW	trl/models/activation_offloading.py	495	# Check if tensor still exists (might have been cleaned up by a previous node)
LOW	trl/models/utils.py	275	# Check if the model is already a FSDP model due to `Manual Wrapping` and if so, don't wrap it again
LOW	trl/scripts/vllm_serve.py	733	# Check if adding this item would exceed the token budget
LOW	trl/skills/skills.py	105	# Check if it's a predefined agent
LOW	trl/skills/skills.py	205	# Check if source skill exists
LOW	trl/skills/skills.py	218	# Check if already exists
LOW	trl/trainer/sft_trainer.py	1478	# Check if the tokenized prompt starts with the tokenized prompt+completion
LOW	trl/trainer/dpo_trainer.py	960	# Check if the tokenized prompt starts with the tokenized prompt+completion
LOW	trl/generation/vllm_client.py	186	# Check if the total timeout duration has passed
LOW	trl/extras/profiling.py	199	# Check if self is a Trainer-like object with required attributes

Excessive Try-Catch Wrapping51 hits · 60 pts

Severity	File	Line	Snippet
LOW	tests/test_sft_trainer.py	2186	except Exception:
LOW	tests/test_grpo_trainer.py	3106	except Exception as e:
LOW	tests/test_grpo_trainer.py	3233	except Exception as e:
LOW	tests/test_grpo_trainer.py	3300	except Exception as e:
LOW	tests/test_rewards.py	143	except Exception as e:
MEDIUM	tests/test_rewards.py	140	def target():
MEDIUM	tests/experimental/test_gold_trainer.py	28	def openr1_examples():
MEDIUM	tests/experimental/test_gold_trainer.py	41	def countdown_examples():
LOW	tests/experimental/test_gold_trainer.py	35	except Exception as exc: # pragma: no cover - network/environment dependent
LOW	tests/experimental/test_gold_trainer.py	48	except Exception as exc: # pragma: no cover - network/environment dependent
LOW	tests/experimental/test_online_dpo_trainer.py	199	except Exception:
LOW	docs/source/openenv.md	398	except Exception:
LOW	docs/source/openenv.md	409	except Exception:
LOW	examples/scripts/sft_vlm_gemma3.py	107	except Exception as e:
MEDIUM	examples/scripts/sft_vlm_gemma3.py	108	print(f"Error processing image {img_path}: {e}")
LOW	examples/scripts/sft_video_llm.py	137	except Exception as e:
LOW	examples/scripts/openenv/multi_env.py	136	except Exception:
LOW	examples/scripts/openenv/multi_env.py	147	except Exception:
LOW	examples/scripts/openenv/carla_vlm_gemma.py	165	except Exception as e:
LOW	examples/scripts/openenv/catch.py	162	except Exception as e:
LOW	scripts/log_reports.py	59	except Exception as e:
MEDIUM	scripts/add_copyrights.py	48	print(f"Error fetching tracked files: {e}")
LOW	trl/_lazy_module.py	72	except Exception as e:
LOW	trl/_compat.py	45	except Exception as e:
LOW	trl/_compat.py	69	except Exception as e:
LOW	trl/_compat.py	126	except Exception as e:
LOW	trl/_compat.py	155	except Exception as e:
LOW	trl/experimental/sdft/sdft.py	213	except Exception:
LOW	trl/experimental/openreward/_spec.py	175	except Exception: # noqa: BLE001
LOW	trl/experimental/openreward/_spec.py	282	except Exception as e: # noqa: BLE001
LOW	trl/experimental/openreward/_spec.py	312	except Exception as e: # noqa: BLE001
LOW	trl/experimental/openreward/environment.py	261	except Exception as e: # noqa: BLE001
LOW	trl/experimental/openreward/environment.py	273	except Exception as e: # noqa: BLE001
LOW	trl/experimental/async_grpo/async_rollout_worker.py	646	except Exception as error:
LOW	trl/experimental/dppo/dppo_trainer.py	463	except Exception as err:
LOW	trl/experimental/ppo/modeling_value_head.py	459	except Exception:
LOW	trl/experimental/ppo/modeling_value_head.py	469	except Exception as exc:
MEDIUM	trl/cli/commands/skills.py	37	print("Error: Unknown skills command")
LOW	trl/scripts/vllm_serve.py	784	except Exception as e:
LOW	trl/scripts/vllm_serve.py	789	except Exception as e:
LOW	trl/scripts/utils.py	400	except Exception as e:
MEDIUM	trl/scripts/utils.py	381	def get_git_commit_hash(package_name):
MEDIUM	trl/skills/cli.py	94	print("Error: Either provide a skill name or use --all to install all skills")
MEDIUM	trl/skills/cli.py	100	print("Error: Cannot specify both a skill name and --all")
MEDIUM	trl/skills/cli.py	158	print(f"Error: {e}")
MEDIUM	trl/skills/cli.py	193	print(f"Error: {e}")
LOW	trl/trainer/callbacks.py	541	except Exception as scorer_e:
LOW	trl/trainer/callbacks.py	547	except Exception as pred_e:
LOW	trl/trainer/callbacks.py	564	except Exception as summary_e:
LOW	trl/trainer/callbacks.py	569	except Exception as finish_e:
LOW	trl/trainer/grpo_trainer.py	1527	except Exception as e:

Hallucination Indicators5 hits · 35 pts

Severity	File	Line	Snippet
CRITICAL	trl/experimental/ppo/modeling_value_head.py	700	self.v_head.summary.weight.data.normal_(mean=0.0, std=initializer_range)
CRITICAL	trl/experimental/ppo/modeling_value_head.py	701	self.v_head.summary.bias.data.zero_()
CRITICAL	trl/experimental/ppo/modeling_value_head.py	966	self.v_head.summary.weight.data.normal_(mean=0.0, std=initializer_range)
CRITICAL	trl/experimental/ppo/modeling_value_head.py	967	self.v_head.summary.bias.data.zero_()
CRITICAL	trl/models/activation_offloading.py	397	torch.autograd.variable.Variable._execution_engine.queue_callback(wait_and_del_remaining_references)

AI Slop Vocabulary5 hits · 14 pts

Severity	File	Line	Snippet
LOW	tests/test_dpo_trainer.py	415	# purposes we will just just use the same checkpoint
MEDIUM	trl/experimental/cpo/cpo_trainer.py	464	# 2. https://github.com/EleutherAI/lm-evaluation-harness/pull/531#issuecomment-1595586257
MEDIUM	trl/experimental/bco/bco_trainer.py	258	# 2. https://github.com/EleutherAI/lm-evaluation-harness/pull/531#issuecomment-1595586257
MEDIUM	trl/experimental/orpo/orpo_trainer.py	449	# 2. https://github.com/EleutherAI/lm-evaluation-harness/pull/531#issuecomment-1595586257
MEDIUM	trl/generation/vllm_client.py	138	# This is not strictly required for correctness, but it helps make training more robust to rare, transient

Cross-Language Confusion2 hits · 5 pts

Severity	File	Line	Snippet
HIGH	examples/scripts/openenv/wordle.py	77	uv venv && source .venv/bin/activate
HIGH	trl/data_utils.py	637	elif not first_column_offsets.equals(column.offsets):

Slop Phrases1 hit · 3 pts

Severity	File	Line	Snippet
MEDIUM	examples/cli_configs/example_config.yaml	1	# This is an example configuration file of TRL CLI, you can use it for

Example Usage Blocks1 hit · 2 pts

Severity	File	Line	Snippet
LOW	trl/generation/vllm_client.py	748	# Example usage

Score History

Severity Breakdown

Pattern Findings