openai / automated-interpretability - Conditional Complexity

Intro

Conditional complexity (also called cyclomatic complexity) is a term used to measure the complexity of software. The term refers to the number of possible paths through a program function. A higher value ofter means higher maintenance and testing costs (infosecinstitute.com).
Conditional complexity is calculated by counting all conditions in the program that can affect the execution path (e.g. if statement, loops, switches, and/or operators, try and catch blocks...).
Conditional complexity is measured at the unit level (methods, functions...).
Units are classified in four categories based on the measured McCabe index: 1-5 (simple units), 6-10 (medium complex units), 11-25 (complex units), 26+ (very complex units).

Learn more...

Conditional Complexity Overall

There are 113 units with 884 lines of code in units (20.6% of code).

0 very complex units (0 lines of code)
0 complex units (0 lines of code)
2 medium complex units (49 lines of code)
3 simple units (172 lines of code)
108 very simple units (663 lines of code)

Legend:

51+

26-50

11-25

6-10

1-5

Alternative Visuals

3D view of all units

Conditional Complexity per Extension

51+

26-50

11-25

6-10

1-5

Conditional Complexity per Logical Component

primary logical decomposition

51+

26-50

11-25

6-10

1-5

Most Complex Units

Top 50 most complex units

Unit	# lines	McCabe index	# params
def _object_hook() in neuron-explainer/neuron_explainer/fast_dataclasses/fast_dataclasses.py	35	17	2
def convert_puzzle_dict_to_puzzle() in neuron-explainer/neuron_explainer/explanations/puzzles.py	14	12	1
def make_explanation_prompt() in neuron-explainer/neuron_explainer/explanations/explainer.py	63	10	2
export default function SimulationSequences() in neuron-viewer/src/simulationHeatmap.tsx	83	7	5
def is_api_error() in neuron-explainer/neuron_explainer/api_client.py	26	7	1
export default function TokenHeatmap() in neuron-viewer/src/tokenHeatmap.tsx	21	5	4
def decorate() in neuron-explainer/neuron_explainer/api_client.py	15	5	1
def get_numerical_subdirs() in neuron-explainer/neuron_explainer/explanations/explanations.py	11	5	1
def make_explanation_prompt() in neuron-explainer/neuron_explainer/explanations/explainer.py	14	5	2
def normalize_activations() in neuron-explainer/neuron_explainer/activations/activation_records.py	4	4	2
def get_sorted_layers() in neuron-explainer/neuron_explainer/activations/activations.py	7	4	1
def _add_few_shot_examples() in neuron-explainer/neuron_explainer/explanations/explainer.py	12	4	2
def apply_calibration() in neuron-explainer/neuron_explainer/explanations/calibrated_simulator.py	8	4	2
def get_examples() in neuron-explainer/neuron_explainer/explanations/few_shot_examples.py	9	4	1
def calculate_max_activation() in neuron-explainer/neuron_explainer/activations/activation_records.py	6	3	1
def get_sorted_neuron_indices() in neuron-explainer/neuron_explainer/activations/activations.py	5	3	3
def register_dataclass() in neuron-explainer/neuron_explainer/fast_dataclasses/fast_dataclasses.py	6	3	1
def _remove_final_period() in neuron-explainer/neuron_explainer/explanations/explainer.py	6	3	1
def from_int() in neuron-explainer/neuron_explainer/explanations/explainer.py	5	3	2
def get_examples() in neuron-explainer/neuron_explainer/explanations/token_space_few_shot_examples.py	7	3	1
def apply_calibration() in neuron-explainer/neuron_explainer/explanations/calibrated_simulator.py	6	3	2
def from_string() in neuron-explainer/neuron_explainer/explanations/simulator.py	5	3	2
def parse_top_logprobs() in neuron-explainer/neuron_explainer/explanations/simulator.py	7	3	2
def convert_to_byte_array() in neuron-explainer/neuron_explainer/explanations/simulator.py	14	3	1
def was_token_split() in neuron-explainer/neuron_explainer/explanations/simulator.py	18	3	3
def from_string() in neuron-explainer/neuron_explainer/explanations/prompt_builder.py	5	3	2
def prompt_length_in_tokens() in neuron-explainer/neuron_explainer/explanations/prompt_builder.py	15	3	2
def flatten_list() in neuron-explainer/neuron_explainer/explanations/scoring.py	2	3	1
def from_string() in neuron-explainer/neuron_explainer/explanations/few_shot_examples.py	5	3	2
def get_single_token_prediction_example() in neuron-explainer/neuron_explainer/explanations/few_shot_examples.py	7	3	1
function NeuronForm() in neuron-viewer/src/welcome.tsx	125	2	0
def _format_tokens_for_simulation() in neuron-explainer/neuron_explainer/activations/activation_records.py	5	2	1
def max_activation() in neuron-explainer/neuron_explainer/activations/activations.py	2	2	1
def standardize_azure_url() in neuron-explainer/neuron_explainer/azure.py	4	2	1
def _split_numbered_list() in neuron-explainer/neuron_explainer/explanations/explainer.py	3	2	1
def apply_calibration() in neuron-explainer/neuron_explainer/explanations/calibrated_simulator.py	2	2	2
def convert_puzzle_to_tokenized_sentences() in neuron-explainer/neuron_explainer/explanations/puzzles.py	2	2	1
normalized_activation: neuronScale() in neuron-viewer/src/types.ts	6	1	1
export function interpolateColor() in neuron-viewer/src/types.ts	8	1	3
export function getInterpolatedColor() in neuron-viewer/src/types.ts	11	1	3
await fetch() in neuron-viewer/src/interpAPI.ts	7	1	5
await fetch() in neuron-viewer/src/interpAPI.ts	7	1	6
def load_az_json() in neuron-viewer/python/server.py	3	1	1
def start() in neuron-viewer/python/server.py	4	1	0
def after_request() in neuron-viewer/python/server.py	9	1	1
def main() in neuron-viewer/python/server.py	2	1	3
def load_token_weight_connections_of_neuron() in neuron-explainer/neuron_explainer/activations/token_connections.py	4	1	0
def load_token_lookup_table_connections_of_neuron() in neuron-explainer/neuron_explainer/activations/token_connections.py	4	1	0
def relu() in neuron-explainer/neuron_explainer/activations/activation_records.py	2	1	1
def _format_activation_record() in neuron-explainer/neuron_explainer/activations/activation_records.py	6	1	0