• File size measurements show the distribution of size of files.
• Files are classified in four categories based on their size (lines of code): 1-100 (very small files), 101-200 (small files), 201-500 (medium size files), 501-1000 (long files), 1001+(very long files).
• It is a good practice to keep files small. Long files may become "bloaters", code that have increased to such gargantuan proportions that they are hard to work with.

• There are 112 files with 23,903 lines of code.
• 3 very long files (6,745 lines of code)
• 8 long files (5,804 lines of code)
• 19 medium size files (5,706 lines of codeclsfd_ftr_w_mp_ins)
• 23 small files (3,078 lines of code)
• 59 very small files (2,570 lines of code)

There are 33 files with lines longer than 120 characters. In total, there are 148 long lines.

File	# lines	# units	# long lines
arabic_evals.py in community_tasks	725	28	32
imports.py in src/lighteval/utils	71	17	15
nanotron_model.py in src/lighteval/models/nanotron	956	28	12
main_endpoint.py in src/lighteval	409	4	10
evaluation_tracker.py in src/lighteval/logging	520	17	9
transformers_model.py in src/lighteval/models/transformers	731	28	8
endpoint_model.py in src/lighteval/models/endpoints	453	18	7
metrics_sample.py in src/lighteval/metrics	813	63	5
vllm_model.py in src/lighteval/models/vllm	410	15	4
main_accelerate.py in src/lighteval	123	1	3
sglang_model.py in src/lighteval/models/sglang	257	13	3
litellm_model.py in src/lighteval/models	194	11	3
vlm_transformers_model.py in src/lighteval/models/transformers	282	19	3
main_sglang.py in src/lighteval	108	1	3
main_vllm.py in src/lighteval	112	1	3
data.py in src/lighteval	121	13	3
main.py in src/lighteval/tasks/extended/lcb	106	3	3
main.py in src/lighteval/tasks/extended/olympiade_bench	172	3	3
metrics_corpus.py in src/lighteval/metrics	76	8	2
pipeline.py in src/lighteval	311	17	2
adapter_model.py in src/lighteval/models/transformers	70	3	2
registry.py in src/lighteval/tasks	158	11	2
sample_preparator.py in src/lighteval/metrics	75	10	1
inference_providers_model.py in src/lighteval/models/endpoints	137	7	1
main_custom.py in src/lighteval	88	1	1
nli.py in src/lighteval/tasks/templates	124	6	1
formatting_utils.py in src/lighteval/tasks/templates/utils	39	5	1
prompt_manager.py in src/lighteval/tasks	203	15	1
codegen_metrics.py in src/lighteval/tasks/extended/lcb	439	22	1
prompts.py in src/lighteval/tasks/extended/mix_eval	56	3	1
main.py in src/lighteval/tasks/extended/tiny_benchmarks	201	7	1
pyproject.toml in root	115	-	1
german_rag_evals.py in community_tasks	124	4	1