SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1803.08240 · 2018

An Analysis of Neural Language Modeling at Multiple Scales

arXiv · PDF · Open in the Atlas

Code that ran

14 of 18 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 9 have no contract verdict; 3 failures 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/1803.08240.json · badge: /api/ran/1803.08240.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
philippwirth/treelangrnn community (classifier label)
classifier basis not recorded on the link
8 ran · 0 no recorded run
SachinIchake/KALM community (classifier label)
classifier basis not recorded on the link
3 ran · 0 no recorded run
ari-holtzman/genlm community (classifier label)
classifier basis not recorded on the link
2 ran · 0 no recorded run
AtheMathmo/lookahead-lstm community (classifier label)
classifier basis: plan_row
1 ran · 1 no recorded run
mnhng/hier-char-emb community (classifier label)
classifier basis not recorded on the link
0 ran · 3 no recorded run
FunctionStatusWhere it lives
batchify Ran
ran; our contract draft was wrong, not the code
ari-holtzman/genlm/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from fin-young/THEO_POULA, is pointer only (licence: NONE) · get_code("1f32d831b2c4c04b")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from fin-young/THEO_POULA, 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.

batchify Ran
ran; our contract draft was wrong, not the code
SachinIchake/KALM/utils.py
code served (permissive licence) · get_code("6bd10c3dd56c2b8c")
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 Tensor, ndim 2, dtype family float, shape (16, 2) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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)

# Create a small tensor for data
# The function calls data.size(0), data.narrow(), data.view(), and data.t()
# So we need a 2D tensor
data = torch.randn(8, 4)  # 8 rows, 4 columns

# Batch size
bsz = 2

# Args object with cuda attribute
class Args:
    def __init__(self, cuda=False):
        self.cuda = cuda

args = Args(cuda=False)

ARGS = {
    "data": data,
    "bsz": bsz,
    "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 Tensor, ndim 2, dtype family float, shape (16, 2) — shape (16, 2) 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
batchify Ran
ran on a synthesized input
ari-holtzman/genlm/utils.py
code served (permissive licence) · get_code("6696579704cf7e5b")
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 Tensor, ndim 2, dtype family float, shape (16, 2) 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 expects:
# - data: a tensor with .size() method and .narrow(), .view(), .t(), .contiguous() methods
# - bsz: an integer for batch size

# Create a small tensor: shape (8, 4) so we can divide by bsz=2
# This gives n_batches = 8 // 2 = 4, then data becomes (2, 4) after narrowing,
# then after view(2, -1).t() it becomes (4, 2)
data = torch.randn(8, 4)
bsz = 2

ARGS = {
    "data": data,
    "bsz": bsz
}
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 Tensor, ndim 2, dtype family float, shape (16, 2) — shape (16, 2) 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
cone_distance Ran
ran on a synthesized input
philippwirth/treelangrnn/distance.py
code served (permissive licence) · get_code("ab108d1a1112ff73")
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 Tensor, ndim 1, dtype family float, shape (2,) 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)

# Create two small tensors with batch size 2 and dimension 4
x = torch.randn(2, 4)
y = torch.randn(2, 4)

# bias is optional and not used in the function body, so set to None
bias = None

ARGS = {
    "x": x,
    "y": y,
    "bias": bias
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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
dot_distance Ran
ran on a synthesized input
philippwirth/treelangrnn/distance.py
code served (permissive licence) · get_code("98a55feeff35dab9")
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 Tensor, ndim 2, dtype family float, shape (1, 2) 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)

# x can be either 1 x d or n x d
# y must be n x d
# bias is optional and used in linear function

x = torch.randn(1, 4)  # 1 x d case
y = torch.randn(2, 4)  # n x d
bias = torch.randn(2)  # bias vector matching y's second dimension

ARGS = {
    "x": x,
    "y": y,
    "bias": bias
}
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 Tensor, ndim 2, dtype family float, shape (1, 2) — shape (1, 2) 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
eucl_distance Ran
ran on a synthesized input
philippwirth/treelangrnn/distance.py
code served (permissive licence) · get_code("89727c8d9972d138")
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 Tensor, ndim 1, dtype family float, shape (2,) 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 small tensors for x and y
# x can be 1 x d or n x d, y must be n x d
# Using distinct sizes: x is 2x4, y is 2x4
x = torch.randn(2, 4)
y = torch.randn(2, 4)

# bias is optional, can be None or a tensor
# Looking at the function body, bias is added to the result of dist_fn(x, y).pow(2)
# The output of PairwiseDistance with these shapes will be (2,), so bias should match
bias = torch.randn(2)

ARGS = {
    "x": x,
    "y": y,
    "bias": bias
}
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 Tensor, ndim 1, dtype family float, shape (2,) — shape (2,) 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_batch Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
SachinIchake/KALM/utils.py
code served (permissive licence) · get_code("6f3c8e0b36d76af7")
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 tuple, ndim 2, dtype family int; 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
import numpy as np

torch.manual_seed(0)

# Create a simple source tensor
# The function indexes into source with [i:i+seq_len] and [i+1:i+1+seq_len]
# So we need enough elements. Let's make it 1D with 20 elements
source = torch.arange(20, dtype=torch.long)

# Create a simple args object with bptt attribute
class Args:
    def __init__(self):
        self.bptt = 5

args = Args()

# i is an index into source
i = 0

# seq_len can be None or an int
seq_len = None

# evaluation is a boolean flag
evaluation = False

ARGS = {
    "source": source,
    "i": i,
    "args": args,
    "seq_len": seq_len,
    "evaluation": evaluation
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 2, dtype family int — a tuple 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
hb_atanh Ran
ran on a synthesized input
philippwirth/treelangrnn/hb_helpers.py
code served (permissive licence) · get_code("bf412beb331b029b")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 hb_atanh takes a single parameter x (a tensor)
# It clamps x, then applies logarithmic operations on it
# x should be a tensor with values in a valid range for the computation

x = torch.randn(2, 4)  # Small tensor: batch size 2, dimension 4

ARGS = {
    "x": x
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
hb_project_hyp_vecs Ran
ran on a synthesized input
philippwirth/treelangrnn/hb_helpers.py
code served (permissive licence) · get_code("6c36f63c9cd125f4")
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 Tensor, ndim 2, dtype family float, shape (2, 4) 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 hb_project_hyp_vecs takes:
# - x: a tensor that gets passed to hp_clip_by_norm
# - c: a curvature parameter (not used in the function body, but included in signature)

# hp_clip_by_norm computes the norm of v and clips it
# So x should be a tensor

# Create a small tensor for x
x = torch.randn(2, 4)

# c is a curvature parameter, typically a small positive scalar
c = torch.tensor(1.0)

ARGS = {
    "x": x,
    "c": c
}
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 Tensor, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) 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
hp_clip_by_norm Ran
ran on a synthesized input
philippwirth/treelangrnn/hb_helpers.py
code served (permissive licence) · get_code("665b5ee8e5d63866")
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 Tensor, ndim 1, dtype family float, shape (4,) 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 hp_clip_by_norm takes a vector v and clips it by norm
# It uses:
# - torch.norm(v) to compute the norm
# - comparison with clip_norm
# - scalar multiplication and division operations

# Create a small vector tensor
v = torch.randn(4)

# Create a clip_norm value (scalar)
clip_norm = 0.5

ARGS = {
    "v": v,
    "clip_norm": clip_norm,
}
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 Tensor, ndim 1, dtype family float, shape (4,) — shape (4,) 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
log_sigmoid Ran
ran on a synthesized input
philippwirth/treelangrnn/activation.py
code served (permissive licence) · get_code("bc60fdcf73a9a77f")
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 Tensor, 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

torch.manual_seed(0)

# The function takes a single parameter 'x' which is used as:
# - Input to torch.nn.functional.logsigmoid(x)
# - Indexing: y[0] and y[1:]
# - The result should support .sum(0) and .mean()
# This indicates x should be a 2D tensor (batch_size, features)

x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, 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
log_softmax Ran
ran on a synthesized input
philippwirth/treelangrnn/activation.py
code served (permissive licence) · get_code("0194ec0fafdf5091")
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 Tensor, 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

torch.manual_seed(0)

# The function takes a tensor x and applies log_softmax along dim=0,
# then indexes [0] to get the first element, and calls .mean() on it
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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 Tensor, 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
one_hot Ran
ran, honoured the contract we drafted
AtheMathmo/lookahead-lstm/pointer.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from Asteur/RERITES-AvgWeightDescentLSTM-PoetryGeneration, is pointer only (licence: NONE) · get_code("d38f82da04a3f93e")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from Asteur/RERITES-AvgWeightDescentLSTM-PoetryGeneration, 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.

repackage_hidden Ran
ran, honoured the contract we drafted
SachinIchake/KALM/utils.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from mingruiliu-ml-lab/communication-efficient-local-gradient-clipping, is pointer only (licence: GPL-3.0) · get_code("4372bb4533fd1936")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from mingruiliu-ml-lab/communication-efficient-local-gradient-clipping, and GPL-3.0, as GitHub detects it on that repository, is not in the permissive set Syntology serves inline. 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.

embedded_dropout Not yet run
unverified (harvested, no recorded run)
AtheMathmo/lookahead-lstm/embed_regularize.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from yx3266/SEN, is pointer only (licence: NONE) · get_code("62d148258bdd054f")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from yx3266/SEN, 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. No invocation recorded.

make_bnode Not yet run
unverified (harvested, no recorded run)
mnhng/hier-char-emb/nnblk/ids_util.py
code served (permissive licence) · get_code("2e9c943d81e42c08")
How it ran

no invocation recorded.

parse_ids_rules Not yet run
unverified (harvested, no recorded run)
mnhng/hier-char-emb/nnblk/ids_util.py
code served (permissive licence) · get_code("55bce8b1a136e197")
How it ran

no invocation recorded.

var_node Not yet run
unverified (harvested, no recorded run)
mnhng/hier-char-emb/nnblk/ids_util.py
code served (permissive licence) · get_code("311c68b6e9722b88")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 18 distinct functions with an identity, 6 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 0 the same month, 6 later) and 12 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
repackage_hidden 15 other papers
2 dated earlier, 13 later
1609.04309 (2016-09)
Efficient softmax approximation for GPUs
1704.05119 (2017-04)
Exploring Sparsity in Recurrent Neural Networks
1806.01822 (2018-06)
Relational recurrent neural networks
and 12 more not named here
batchify 7 other papers
7 later
1805.04623 (2018-05)
Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context
1808.09029 (2018-08)
Pyramidal Recurrent Unit for Language Modeling
1810.09536 (2018-10)
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
and 4 more not named here
embedded_dropout 6 other papers
6 later
1810.09536 (2018-10)
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
1905.11881 (2019-05)
Why gradient clipping accelerates training: A theoretical justification for adaptivity
2006.07116 (2020-06)
NAS-Bench-NLP: Neural Architecture Search Benchmark for Natural Language Processing
and 3 more not named here
get_batch 4 other papers
4 later
1810.09536 (2018-10)
Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
1905.11881 (2019-05)
Why gradient clipping accelerates training: A theoretical justification for adaptivity
2010.02519 (2020-10)
Improved Analysis of Clipping Algorithms for Non-convex Optimization
and 1 more not named here
one_hot 4 other papers
1 dated earlier, 3 later
1708.02182 (2017-08)
Regularizing and Optimizing LSTM Language Models
1805.04623 (2018-05)
Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context
1905.11881 (2019-05)
Why gradient clipping accelerates training: A theoretical justification for adaptivity
and 1 more not named here
batchify 2 other papers
2 later
2105.13937 (2021-05)
Polygonal Unadjusted Langevin Algorithms: Creating stable and efficient adaptive algorithms for neural networks
2108.02170 (2021-08)
Curriculum learning for language modeling

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

Many of the leading approaches in language modeling introduce novel, complex and specialized architectures. We take existing state-of-the-art word level language models based on LSTMs and QRNNs and extend them to both larger vocabularies as well as character-level granularity. When properly tuned, LSTMs and QRNNs achieve state-of-the-art results on character-level (Penn Treebank, enwik8) and word-level (WikiText-103) datasets, respectively. Results are obtained in only 12 hours (WikiText-103) to 2 days (enwik8) using a single modern GPU.

For agents

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

get_harvested_code_for_paper("1803.08240")
get_code_for_paper("1803.08240")
have("1803.08240")

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

curl https://syntology.ai/api/ran/1803.08240.json

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

[![Syntology run record](https://syntology.ai/api/ran/1803.08240.svg)](https://syntology.ai/paper/1803.08240)

Connect an agent — have() is free.