SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2311.04902 · 2023

Beyond Size: How Gradients Shape Pruning Decisions in Large Language Models

arXiv · PDF · Open in the Atlas

Code that ran

3 of 7 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; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 4 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/2311.04902.json · badge: /api/ran/2311.04902.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
rocktimjyotidas/gblm-pruner canonical (classifier label)
classifier basis: census
3 ran · 4 no recorded run
FunctionStatusWhere it lives
check_sparsity Ran
ran, honoured the contract we drafted
rocktimjyotidas/gblm-pruner/lib/prune.py
code served (permissive licence) · get_code("6d3d2eb274571aa8")
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 float, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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 torch.nn as nn

torch.manual_seed(0)

# Create a minimal config class that has use_cache attribute
class MinimalConfig:
    def __init__(self):
        self.use_cache = True

# Create a minimal model structure that matches what check_sparsity expects
class MinimalLayer(nn.Module):
    def __init__(self, in_features=4, out_features=3):
        super().__init__()
        self.linear1 = nn.Linear(in_features, out_features)
        self.linear2 = nn.Linear(out_features, out_features)
    
    def forward(self, x):
        x = self.linear1(x)
        x = self.linear2(x)
        return x

class MinimalModel(nn.Module):
    def __init__(self):
        super().__init__()
        self.config = MinimalConfig()
        self.model = nn.Module()
        # Create layers with some sparsity
        self.model.layers = nn.ModuleList([
            MinimalLayer(in_features=4, out_features=3),
            MinimalLayer(in_features=3, out_features=5),
        ])
        
        # Introduce some sparsity by zeroing out some weights
        for layer in self.model.layers:
            for child in layer.children():
                if isinstance(child, nn.Linear):
                    child.weight.data[0, 0] = 0
                    child.weight.data[1, 1] = 0
    
    def forward(self, x):
        for layer in self.model.layers:
            x = layer(x)
        return x

model = MinimalModel()

# Create a minimal args object (or dict, depending on what the function expects)
class MinimalArgs:
    pass

args = MinimalArgs()

ARGS = {
    'model': model,
    'args': args,
}
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 float, ndim 0, dtype family float, shape () — shape () as a later run recorded it
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
find_layers Ran
ran; our contract draft was wrong, not the code
rocktimjyotidas/gblm-pruner/gradient_computation.py
pointer only (licence: MIT); the stored copy get_code serves, from zyxxmu/dsnot, is pointer only (licence: NONE) · get_code("2ed16fcdc14ad951")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from zyxxmu/dsnot, and GitHub detected no licence in that repository when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

no_zero Ran
ran, honoured the contract we drafted
rocktimjyotidas/gblm-pruner/lib/prune.py
code served (permissive licence) · get_code("85b518f08a12d28c")
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 int, ndim 0, dtype family int, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape 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)

# The function no_zero takes a 'data' parameter and counts zeros in it
# It calls (data == 0).sum().item() which means data must be a torch tensor
# that supports comparison and sum operations

data = torch.tensor([[1, 2, 0, 3],
                     [0, 5, 4, 0]], dtype=torch.float32)

ARGS = {
    "data": data
}
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 int, ndim 0, dtype family int, shape () — shape () as a later run recorded it
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
get_c4 Not yet run
unverified (harvested, no recorded run)
rocktimjyotidas/gblm-pruner/gradient_computation.py
code served (permissive licence) · get_code("362bd8bd1eddbfc5")
How it ran

no invocation recorded.

get_c4 Not yet run
unverified (harvested, no recorded run)
rocktimjyotidas/gblm-pruner/lib/data.py
code served (permissive licence) · get_code("5cbbb830025b49e7")
How it ran

no invocation recorded.

get_loaders Not yet run
unverified (harvested, no recorded run)
rocktimjyotidas/gblm-pruner/lib/data.py
code served (permissive licence) · get_code("938021b9c285abea")
How it ran

no invocation recorded.

get_wikitext2 Not yet run
unverified (harvested, no recorded run)
rocktimjyotidas/gblm-pruner/gradient_computation.py
code served (permissive licence) · get_code("cb81c679c455e959")
How it ran

no invocation recorded.

Also present in other papers' code

Of 7 samples lifted from this paper's linked repositories, 7 carry a function identity and 0 do not. Of the 7 distinct functions with an identity, 2 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 2 later) and 5 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.

FunctionAlso present inEarliest-datedOthers named
find_layers 21 other papers
3 dated earlier, 18 later · 5 only as an assembly brick
2306.11695 (2023-06)
A Simple and Effective Pruning Approach for Large Language Models
2310.05175 (2023-10)
Outlier Weighed Layerwise Sparsity (OWL ): A Missing Secret Sauce for Pruning LLMs to High Sparsity
2310.08915 (2023-10)
Dynamic Sparse No Training: Training-Free Fine-tuning for Sparse LLMs
and 18 more not named here
get_loaders 9 other papers
3 dated earlier, 6 later
2306.11695 (2023-06)
A Simple and Effective Pruning Approach for Large Language Models
2310.01382 (2023-10)
Compressing LLMs: The Truth is Rarely Pure and Never Simple
2310.05175 (2023-10)
Outlier Weighed Layerwise Sparsity (OWL ): A Missing Secret Sauce for Pruning LLMs to High Sparsity
and 6 more not named here

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

Large Language Models (LLMs) with billions of parameters are prime targets for network pruning, removing some model weights without hurting performance. Prior approaches such as magnitude pruning, SparseGPT, and Wanda, either concentrated solely on weights or integrated weights with activations for sparsity. However, they overlooked the informative gradients derived from pretrained LLMs. In this paper, we present a novel sparsity-centric pruning method for pretrained LLMs, termed Gradient-based Language Model Pruner (GBLM-Pruner). GBLM-Pruner leverages the first-order term of the Taylor expansion, operating in a training-free manner by harnessing properly normalized gradients from a few calibration samples to determine the pruning metric, and substantially outperforms competitive counterparts like SparseGPT and Wanda in multiple benchmarks. Intriguingly, by incorporating gradients, unstructured pruning with our method tends to reveal some structural patterns, which mirrors the geometric interdependence inherent in the LLMs' parameter structure. Additionally, GBLM-Pruner functions without any subsequent retraining or weight updates to maintain its simplicity as other counterparts. Extensive evaluations on LLaMA-1 and LLaMA-2 across various benchmarks show that GBLM-Pruner surpasses magnitude pruning, Wanda and SparseGPT by significant margins. We further extend our approach on Vision Transformer. Our code and models are available at https://github.com/VILA-Lab/GBLM-Pruner.

For agents

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

get_harvested_code_for_paper("2311.04902")
get_code_for_paper("2311.04902")
have("2311.04902")

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

curl https://syntology.ai/api/ran/2311.04902.json

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

[![Syntology run record](https://syntology.ai/api/ran/2311.04902.svg)](https://syntology.ai/paper/2311.04902)

Connect an agent — have() is free.