Papers › Training Compute-Optimal Large Language Models

Training Compute-Optimal Large Language Models

29 Mar 2022arXiv:2203.15556archive 2025-07-28

Jordan Hoffmann, Sebastian Borgeaud, Arthur Mensch, Elena Buchatskaya, Trevor Cai, Eliza Rutherford, Diego de Las Casas, Lisa Anne Hendricks, Johannes Welbl, Aidan Clark, Tom Hennigan, Eric Noland, Katie Millican, George van den Driessche, Bogdan Damoc, Aurelia Guy, Simon Osindero, Karen Simonyan, Erich Elsen, Jack W. Rae, Oriol Vinyals, Laurent SIfre

We investigate the optimal model size and number of tokens for training a transformer language model under a given compute budget. We find that current large language models are significantly undertrained, a consequence of the recent focus on scaling language models whilst keeping the amount of training data constant. By training over \nummodels language models ranging from 70 million to over 16 billion parameters on 5 to 500 billion tokens, we find that for compute-optimal training, the model size and the number of training tokens should be scaled equally: for every doubling of model size the number of training tokens should also be doubled. We test this hypothesis by training a predicted compute-optimal model, \chinchilla, that uses the same compute budget as \gopher but with 70B parameters and 4× more more data. \chinchilla uniformly and significantly outperforms \Gopher (280B), GPT-3 (175B), Jurassic-1 (178B), and Megatron-Turing NLG (530B) on a large range of downstream evaluation tasks. This also means that \chinchilla uses substantially less compute for fine-tuning and inference, greatly facilitating downstream usage. As a highlight, \chinchilla reaches a state-of-the-art average accuracy of 67.5% on the MMLU benchmark, greater than a 7% improvement over \gopher.

PaperPDFCodeCode Syntology ranMethod as code

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 11 code samples: 7 from 2 repositories linked to this paper, 4 identical to code harvested from another paper's repository. Across all 11: 8 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. Where Syntology's instrument failed: 1 where our contract draft was wrong; 2 where the contract check could not run; failures of Syntology's instrument, not of the code.

By repository: community (archive-listed): 7 samples from 2 repositories, 4 ran (4 with no instrument failure; 0 where Syntology's instrument failed); 4 identical to code first harvested elsewhere. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-22 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed. 1 of the 8 samples that ran has no recorded date.

karpathy/llama2.c mentioned on GitHubpytorchSyntology: 6 samples, 3 ran (0 instrument-failed) · 3 unverified report
nkluge-correa/teenytinyllama mentioned on GitHubpytorchSyntology: 1 sample, 1 ran (0 instrument-failed) · 0 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

11 samples harvested: 8 ran (of which 3 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page has not changed since , the first build that kept a record date for it; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

2ran · honoured contract
1ran · our draft was wrong
2ran; the contract check could not run
3ran
3unverified

Licence: 4 of the 11 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from 2 repositories linked to this paper, official or community; the list below is grouped by repository, each group headed by its name and what it is to the paper. 4 samples are listed last, in a group without a repository. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. For 2 samples that ran, the graph records a signature this page does not show: for a class the fixture built without calling, the harness called the object with inputs it made up itself, so the recorded signature is the harness's probe, not the fixture's arguments. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity. A census tag appears when Syntology's census ran implementations with that name on one shared input per bucket: it gives how many were compared, from how many papers, and how many distinct outputs they produced, and names this sample's bucket when this sample was one of them.

karpathy/llama2.c community (archive-listed) 6 samples, 3 ran (0 instrument-failed) · 3 unverified
FeedForward in 198 papers model.py ran · metamorphic tier: deterministic MIT (permissive) · 1bf497baaf536b2b · report fixture: returned FeedForward · level recorded 2026-09-22
ModelArgs in 13 papers model.py ran · metamorphic tier: deterministic MIT (permissive) · 7312586d85e4354f · report fixture: returned ModelArgs · level recorded 2026-09-22
RMSNorm in 61 papers model.py ran · metamorphic tier: deterministic MIT (permissive) · e27b34a3f741d4de · report fixture: returned RMSNorm · level recorded 2026-09-22
Transformer in 173 papers model.py unverified MIT (permissive) · 8b8a4640c5a93610 · report
TransformerBlock in 73 papers model.py unverified MIT (permissive) · 46b07f9445b01dce · report
nkluge-correa/teenytinyllama community (archive-listed) 1 sample, 1 ran (0 instrument-failed) · 0 unverified
calculate_loss in 20 papers Utilities/chinchilla-estimation.py ran · honoured contract fingerprinted Apache-2.0 (permissive) · 9293e052d06482a7 · report fixture: signature N:int/0/int, D:int/0/int · returned float · level recorded 2026-09-22
Repository not located 4 samples, 4 ran (3 instrument-failed) · 0 unverified
4 samples are identical to code Syntology harvested from another paper's repository; this paper's own copy was not located, so no repository is named and no licence is recorded.
repeat_kv in 244 papers
census: 15 implementations from 60 papers, 2 distinct outputs on one shared input this sample: bucket 1 of 1, arg 1: rank 4, kind float, dtype float32
repository not located ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence of this copy not recorded · 6d2a08dcf3466514 · report fixture: signature x:Tensor/4/float, n_rep:int/0/int · returned Tensor · level recorded, date not recorded
apply_rotary_emb in 121 papers repository not located ran · our draft was wrong licence of this copy not recorded · af440c67b16afe67 · report fixture: signature xq:Tensor/3/float, xk:Tensor/3/float, freqs_cos:Tensor/2/float, freqs_sin:Tensor/2/float · returned tuple · level recorded 2026-09-22
precompute_freqs_cis in 112 papers repository not located ran · honoured contract fingerprinted licence of this copy not recorded · 5f447bdd807ed3c0 · report fixture: signature dim:int/0/int, end:int/0/int, theta:float/0/float · returned tuple · level recorded 2026-09-22
reshape_for_broadcast in 90 papers
census: 6 implementations from 10 papers, 1 distinct output on one shared input
repository not located ran; the contract check could not run (RAISES) · metamorphic tier: invariant fingerprinted licence of this copy not recorded · e04ad9d4b02f50b6 · report fixture: signature freqs_cis:Tensor/2/float, x:Tensor/4/float · returned Tensor · level recorded 2026-09-22

Method as code Syntology

What Syntology's graph records as this paper's claimed method, lifted into code: each assembly names its bricks (functions and classes), each with its own role and its own verification level. An assembly is not a check that the code implements the method correctly or reproduces the paper's results. Levels are each brick's own verification level in Syntology's graph (level 0: no recorded run), listed brick by brick and never combined into one level; “ran” means the brick executed on a synthesized input, not that its output is correct, and the instrument figure counts failures of Syntology's instrument, not of the code. Where a model read the code against the paper's claim, each assembly's label verdict is one model's judgement, kept in this page's JSON twin; it is not calibrated, so no card prints it. Siblings are assemblies lifted from this paper in the same batch for the same method: alternative readings kept side by side; none replaces or supersedes another.

2 assemblies on this paper: 2 harvested, 0 generated, listed in 1 groups (siblings lifted in one batch form one group).

2 sibling assemblies for Chinchilla (harvested), lifted in the same batch on 2026-09-09: alternative readings kept side by side; none replaces or supersedes another.

harvested Assembly for Chinchilla (graph method: Chinchilla). 10 bricks; individually on a fixture: 7 ran (of which 3 constructed an object rather than computing a result; 4 with no instrument failure: 1 honoured, 0 violated, 3 with no contract checked; 3 where Syntology's instrument failed) · 3 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. Transformer entry · level 0 · unverified · 8b8a4640c5a93610
  2. ModelArgs model · level 1 · ran · 7312586d85e4354f
  3. RMSNorm model · level 1 · ran · e27b34a3f741d4de
  4. TransformerBlock model · level 0 · unverified · 46b07f9445b01dce
  5. precompute_freqs_cis helper · level 1 · ran · honoured contract · 5f447bdd807ed3c0
  6. Attention model · level 0 · unverified · c20a31b2d7615a4b
  7. FeedForward model · level 1 · ran · 1bf497baaf536b2b
  8. apply_rotary_emb helper · level 1 · ran · our draft was wrong · af440c67b16afe67
  9. repeat_kv helper · level 2 · ran; the contract check could not run (RAISES) · 6d2a08dcf3466514
  10. reshape_for_broadcast helper · level 1 · ran; the contract check could not run (RAISES) · e04ad9d4b02f50b6

harvested Assembly for Chinchilla (graph method: Chinchilla). 1 brick; individually on a fixture: 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified. The assembly has not been run end-to-end.

one of 2 siblings

  1. calculate_loss entry · level 1 · ran · honoured contract · 9293e052d06482a7

Each brick ends with its code_sha256, Syntology's identity for that exact code; click one to copy the get_code call that returns it. For agents, on Syntology's MCP service: assembly(arxiv_id="2203.15556") returns these assemblies with each brick's role and wiring (how to connect).

Tasks

AnachronismsAnalogical SimilarityAnalytic EntailmentCausal JudgmentCommon Sense ReasoningCrash BlossomCrass AIDark Humor DetectionDate UnderstandingDisambiguation QADiscourse Marker PredictionEmpirical JudgmentsEnglish ProverbsEntailed PolarityEpistemic ReasoningEvaluating Information EssentialityFantasy ReasoningFigure Of Speech DetectionFormal Fallacies Syllogisms NegationGRE Reading ComprehensionGeneral KnowledgeHellaSwagHuman Organs Senses Multiple ChoiceHyperbatonIdentify Odd MetaporImplicaturesImplicit RelationsIntent RecognitionIrony IdentificationKnown UnknownsLAMBADALanguage ModellingLogic Grid PuzzleLogical ArgsLogical Fallacy DetectionLogical ReasoningLogical SequenceMMLUMathematical InductionMathematical ReasoningMetaphor BooleanMisconceptionsMoral PermissibilityMovie Dialog Same Or DifferentMovie RecommendationMulti-task Language UnderstandingMultiple Choice Question Answering (MCQA)NavigateNonsense Words GrammarNovel ConceptsOdd One OutPenguins In A TablePhrase RelatednessPhysical IntuitionPhysics MCPresuppositions As NLIQuestion AnsweringQuestion SelectionReasoning About Colored ObjectsRiddle SenseRuin NamesSNARKSSarcasm DetectionSentence AmbiguitySentence CompletionSimilarities AbstractionSports UnderstandingStrategyQATemporal SequencesTimedialUnderstanding FablesWinowhyWord Sense Disambiguation

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Analogical Similarity BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 38.1 #1 of 2 Archive leaderboard report
Analytic Entailment BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 67.1 #1 of 2 Archive leaderboard report
Common Sense Reasoning BIG-bench (Causal Judgment) Chinchilla-70B (few-shot, k=5) Accuracy 57.4 #4 of 9 Archive leaderboard report
Common Sense Reasoning BIG-bench (Date Understanding) Chinchilla-70B (few-shot, k=5) Accuracy 52.3 #5 of 9 Archive leaderboard report
Common Sense Reasoning BIG-bench (Disambiguation QA) Chinchilla-70B (few-shot, k=5) Accuracy 54.7 #4 of 9 Archive leaderboard report
Common Sense Reasoning BIG-bench (Known Unknowns) Chinchilla-70B (few-shot, k=5) Accuracy 65.2 #2 of 3 Archive leaderboard report
Common Sense Reasoning BIG-bench (Logical Sequence) Chinchilla-70B (few-shot, k=5) Accuracy 64.1 #1 of 2 Archive leaderboard report
Common Sense Reasoning BIG-bench (Sports Understanding) Chinchilla-70B (few-shot, k=5) Accuracy 71 #4 of 8 Archive leaderboard report
Common Sense Reasoning BIG-bench (Winowhy) Chinchilla-70B (few-shot, k=5) Accuracy 62.5 #2 of 4 Archive leaderboard report
Common Sense Reasoning WinoGrande Chinchilla 70B (0-shot) Accuracy 74.9 #28 of 77 Archive leaderboard report
Crash Blossom BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 47.6 #2 of 2 Archive leaderboard report
Crass AI BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 75.0 #3 of 4 Archive leaderboard report
Dark Humor Detection BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 66.2 #2 of 2 Archive leaderboard report
Discourse Marker Prediction BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 13.1 #1 of 2 Archive leaderboard report
Empirical Judgments BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 67.7 #1 of 2 Archive leaderboard report
English Proverbs BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 82.4 #1 of 2 Archive leaderboard report
Entailed Polarity BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 94 #1 of 2 Archive leaderboard report
Epistemic Reasoning BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 60.6 #1 of 2 Archive leaderboard report
Evaluating Information Essentiality BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 17.6 #1 of 2 Archive leaderboard report
Fantasy Reasoning BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 69 #1 of 2 Archive leaderboard report
Figure Of Speech Detection BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 63.3 #1 of 2 Archive leaderboard report
GRE Reading Comprehension BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 53.1 #1 of 2 Archive leaderboard report
General Knowledge BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 94.3 #1 of 2 Archive leaderboard report
Human Organs Senses Multiple Choice BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 85.7 #1 of 2 Archive leaderboard report
Identify Odd Metapor BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 68.8 #1 of 2 Archive leaderboard report
Implicatures BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 75 #1 of 2 Archive leaderboard report
Implicit Relations BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 49.4 #1 of 2 Archive leaderboard report
Intent Recognition BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 92.8 #1 of 2 Archive leaderboard report
Irony Identification BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 73.0 #1 of 2 Archive leaderboard report
LAMBADA BIG-bench Chinchilla-70B (zero-shot) Accuracy 77.4 #1 of 2 Archive leaderboard report
Language Modelling LAMBADA Chinchilla (Zero-Shot) Accuracy 77.7 #16 of 37 Archive leaderboard report
Logical Args BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 56.2 #2 of 2 Archive leaderboard report
Logical Reasoning BIG-bench (Formal Fallacies Syllogisms Negation) Chinchilla-70B (few-shot, k=5) Accuracy 52.1 #7 of 9 Archive leaderboard report
Logical Reasoning BIG-bench (Logic Grid Puzzle) Chinchilla-70B (few-shot, k=5) Accuracy 44 #1 of 4 Archive leaderboard report
Logical Reasoning BIG-bench (Logical Fallacy Detection) Chinchilla-70B (few-shot, k=5) Accuracy 72.1 #1 of 2 Archive leaderboard report
Logical Reasoning BIG-bench (Penguins In A Table) Chinchilla-70B (few-shot, k=5) Accuracy 48.7 #3 of 9 Archive leaderboard report
Logical Reasoning BIG-bench (Reasoning About Colored Objects) Chinchilla-70B (few-shot, k=5) Accuracy 59.7 #3 of 9 Archive leaderboard report
Logical Reasoning BIG-bench (StrategyQA) Chinchilla-70B (few-shot, k=5) Accuracy 68.3 #2 of 4 Archive leaderboard report
Logical Reasoning BIG-bench (Temporal Sequences) Chinchilla-70B (few-shot, k=5) Accuracy 32.0 #5 of 9 Archive leaderboard report
Mathematical Induction BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 47.3 #2 of 2 Archive leaderboard report
Metaphor Boolean BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 93.1 #1 of 2 Archive leaderboard report
Misconceptions BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 65.3 #1 of 2 Archive leaderboard report
Moral Permissibility BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 57.3 #1 of 2 Archive leaderboard report
Movie Dialog Same Or Different BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 54.5 #1 of 2 Archive leaderboard report
Multi-task Language Understanding MML chatgpt/gpt3.5(20B) Average (%) 67.5 #14 of 44 Archive leaderboard report
Multiple Choice Question Answering (MCQA) BIG-bench (Hyperbaton) Chinchilla-70B (few-shot, k=5) Accuracy 54.2 #8 of 9 Archive leaderboard report
Multiple Choice Question Answering (MCQA) BIG-bench (Movie Recommendation) Chinchilla-70B (few-shot, k=5) Accuracy 75.6 #8 of 9 Archive leaderboard report
Multiple Choice Question Answering (MCQA) BIG-bench (Navigate) Chinchilla-70B (few-shot, k=5) Accuracy 52.6 #4 of 9 Archive leaderboard report
Multiple Choice Question Answering (MCQA) BIG-bench (Novel Concepts) Chinchilla-70B (few-shot, k=5) Accuracy 65.6 #2 of 4 Archive leaderboard report
Multiple Choice Question Answering (MCQA) BIG-bench (Ruin Names) Chinchilla-70B (few-shot, k=5) Accuracy 47.1 #8 of 9 Archive leaderboard report
Nonsense Words Grammar BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 78 #1 of 2 Archive leaderboard report
Odd One Out BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 70.9 #1 of 2 Archive leaderboard report
Phrase Relatedness BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 94 #1 of 2 Archive leaderboard report
Physical Intuition BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 79 #1 of 2 Archive leaderboard report
Presuppositions As NLI BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 49.9 #1 of 2 Archive leaderboard report
Question Answering BoolQ Chinchilla 70B (0-shot) Accuracy 83.7 #21 of 65 Archive leaderboard report
Question Answering Natural Questions Chinchilla (few-shot, k=64) EM 35.5 #28 of 47 Archive leaderboard report
Question Answering PIQA Chinchilla 70B (0-shot) Accuracy 81.8 #26 of 67 Archive leaderboard report
Question Answering SIQA Chinchilla (zero-shot) Accuracy 51.3 #19 of 24 Archive leaderboard report
Question Selection BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 52.6 #1 of 2 Archive leaderboard report
Riddle Sense BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 85.7 #1 of 2 Archive leaderboard report
Sarcasm Detection BIG-bench (SNARKS) Chinchilla-70B (few-shot, k=5) Accuracy 58.6 #7 of 8 Archive leaderboard report
Sentence Ambiguity BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 71.7 #1 of 2 Archive leaderboard report
Sentence Completion HellaSwag Chinchilla 70B (0-shot) Accuracy 80.8 #42 of 89 Archive leaderboard report
Similarities Abstraction BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 87 #1 of 2 Archive leaderboard report
Timedial BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 68.8 #1 of 2 Archive leaderboard report
Understanding Fables BIG-bench Chinchilla-70B (few-shot, k=5) Accuracy 60.3 #1 of 2 Archive leaderboard report
Word Sense Disambiguation BIG-bench (Anachronisms) Chinchilla-70B (few-shot, k=5) Accuracy 69.1 #1 of 6 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBPEChinchillaCosine AnnealingDense ConnectionsDropoutGPT-3Layer NormalizationLinear LayerLinear Warmup With Cosine AnnealingMulti-Head AttentionResidual ConnectionSoftmaxWeight Decay

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections