SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2302.03773 · 2023

What Matters In The Structured Pruning of Generative Language Models?

arXiv · PDF · Open in the Atlas

Code that ran

2 of 8 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 2 have no contract verdict; 0 failures of our instrument; 6 no recorded run), measured 2026-11-22; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2302.03773.json · badge: /api/ran/2302.03773.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
huggingface/nn_pruning canonical (classifier label)
classifier basis not recorded on the link
2 ran · 6 no recorded run
FunctionStatusWhere it lives
head_mask Ran
ran on a synthesized input
huggingface/nn_pruning/nn_pruning/modules/masked_nn.py
code served (permissive licence) · get_code("17ba54a7b0a5ce34")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned list, ndim 2, dtype family float; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch

torch.manual_seed(0)

class Config:
    def __init__(self):
        self.pruned_heads = {0: [1, 3], 1: [0]}
        self.num_attention_heads = 4
        self.num_hidden_layers = 2
        self.hidden_size = 16

config = Config()
device = "cpu"

ARGS = {
    "config": config,
    "device": device,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type list, ndim 2, dtype family float — a list output has no array shape of its own; none is stated
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
struct_from_name Ran
ran on a synthesized input
huggingface/nn_pruning/nn_pruning/model_structure.py
code served (permissive licence) · get_code("dee55cf2bd41b9a9")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned type. The image it ran in was not recorded. If your copy returns the same type on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import torch
import re

torch.manual_seed(0)

# The function struct_from_name takes a single parameter: model_name
# It searches through name2struct.keys() to find which structure matches
# The function simply checks if any name from the keys is in the model_name string

ARGS = {
    'model_name': 'bert-base-uncased'
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type type — a type output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a type output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
broadcast_attention_mask Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/modules/quantization.py
code served (permissive licence) · get_code("648df3827ef1dbca")
How it ran

no invocation recorded.

broadcast_nonorm_bias Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/modules/quantization.py
code served (permissive licence) · get_code("b327b900a4793da6")
How it ran

no invocation recorded.

change_attention_mask_value Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/modules/quantization.py
code served (permissive licence) · get_code("917b76e8afdf3059")
How it ran

no invocation recorded.

create_qconfig Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/modules/quantization_config.py
code served (permissive licence) · get_code("cdd7f506f307304d")
How it ran

no invocation recorded.

struct_from_config Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/model_structure.py
code served (permissive licence) · get_code("62ad5a91e86b77fa")
How it ran

no invocation recorded.

struct_from_model Not yet run
unverified (harvested, no recorded run)
huggingface/nn_pruning/nn_pruning/model_structure.py
code served (permissive licence) · get_code("efe631ef8ad53692")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 8 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-22.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Auto-regressive large language models such as GPT-3 require enormous computational resources to use. Traditionally, structured pruning methods are employed to reduce resource usage. However, their application to and efficacy for generative language models is heavily under-explored. In this paper we conduct an comprehensive evaluation of common structured pruning methods, including magnitude, random, and movement pruning on the feed-forward layers in GPT-type models. Unexpectedly, random pruning results in performance that is comparable to the best established methods, across multiple natural language generation tasks. To understand these results, we provide a framework for measuring neuron-level redundancy of models pruned by different methods, and discover that established structured pruning methods do not take into account the distinctiveness of neurons, leaving behind excess redundancies. In view of this, we introduce Globally Unique Movement (GUM) to improve the uniqueness of neurons in pruned models. We then discuss the effects of our techniques on different redundancy metrics to explain the improved performance.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2302.03773")
get_code_for_paper("2302.03773")
have("2302.03773")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2302.03773.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2302.03773.svg)](https://syntology.ai/paper/2302.03773)

Connect an agent — have() is free.