SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.02660 · 2024

How to Train Long-Context Language Models (Effectively)

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 1 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 2 failures of our instrument; 0 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/2410.02660.json · badge: /api/ran/2410.02660.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
princeton-nlp/prolong canonical (classifier label)
classifier basis: census
3 ran · 0 no recorded run
FunctionStatusWhere it lives
min_lr_bound Ran
ran, honoured the contract we drafted
princeton-nlp/prolong/training/trainer.py
code served (permissive licence) · get_code("2718b0555714208e")
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)

# Define a simple wrapped function that takes a float and returns a float
def dummy_lr_func(step: float) -> float:
    """Simple learning rate schedule function"""
    return 0.1 * (0.95 ** (step / 1000.0))

# Set up arguments for min_lr_bound
ARGS = {
    "current_step": 5,
    "wrapped_func": dummy_lr_func,
    "min_lr_ratio": 0.01,
    "warmup_steps": 10
}
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
repeat_kv Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
princeton-nlp/prolong/training/modeling_flash_llama.py
pointer only (licence: MIT); the stored copy get_code serves, from hrlics/CoPE, is pointer only (licence: NONE) · get_code("69874376c94f0785")
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 hrlics/CoPE, 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.

rmsnorm_func Ran
ran; our contract draft was wrong, not the code
princeton-nlp/prolong/training/modeling_flash_llama.py
pointer only (licence: MIT); the stored copy get_code serves, from hrlics/CoPE, is pointer only (licence: NONE) · get_code("ab6fba34dd3815d7")
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 hrlics/CoPE, 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.

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 3 are also present in other papers whose code we hold (2 with at least one paper dated earlier, 1 the same month, 3 later) and 0 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
rmsnorm_func 8 other papers
3 dated earlier, 1 the same month, 3 later, 1 undated
2310.03668 (2023-10)
GoLLIE: Annotation Guidelines improve Zero-Shot Information-Extraction
2312.15166 (2023-12)
SOLAR 10.7B: Scaling Large Language Models with Simple yet Effective Depth Up-Scaling
2402.09739 (2024-02)
QuRating: Selecting High-Quality Data for Training Language Models
and 5 more not named here
repeat_kv 4 other papers
1 dated earlier, 3 later
2402.09739 (2024-02)
QuRating: Selecting High-Quality Data for Training Language Models
2025.acl-long.366 (2025 (year only))
How to Train Long-Context Language Models (Effectively)
2602.05258 (2026-02)
CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs
and 1 more not named here
min_lr_bound 3 other papers
3 later
2025.acl-long.366 (2025 (year only))
How to Train Long-Context Language Models (Effectively)
2602.05258 (2026-02)
CoPE: Clipped RoPE as A Scalable Free Lunch for Long Context LLMs
2604.07394 (2026-04)
Flux Attention: Context-Aware Hybrid Attention for Efficient LLMs Inference

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

We study continued training and supervised fine-tuning (SFT) of a language model (LM) to make effective use of long-context information. We first establish a reliable evaluation protocol to guide model development -- instead of perplexity or simple needle-in-a-haystack (NIAH) tests, we use a broad set of long-context downstream tasks, and we evaluate models after SFT as this better reveals long-context abilities. Supported by our robust evaluations, we run thorough experiments to decide the data mix for continued pre-training, the instruction tuning dataset, and many other design choices such as position extrapolation. We find that (1) code repositories and books are excellent sources of long data, but it is crucial to combine them with high-quality short-context data; (2) training with a sequence length beyond the evaluation length boosts long-context performance; (3) for SFT, using only short instruction datasets yields strong performance on long-context tasks. Our final model, ProLong-8B, which is initialized from Llama-3 and trained on 40B tokens, demonstrates state-of-the-art long-context performance among similarly sized models at a length of 128K. ProLong outperforms Llama-3.1-8B-Instruct on the majority of long-context tasks despite using only 5% as many tokens during long-context training. Additionally, ProLong can effectively process up to 512K tokens, one of the longest context windows of publicly available LMs.

For agents

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

get_harvested_code_for_paper("2410.02660")
get_code_for_paper("2410.02660")
have("2410.02660")

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

curl https://syntology.ai/api/ran/2410.02660.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.02660.svg)](https://syntology.ai/paper/2410.02660)

Connect an agent — have() is free.