data-diff

generated 2026-08-30 07:06 UTC · etl diff report
❌ Found differences in all 6 compared datasets
6
changed
0
identical
0
skipped
counts are datasets
Of the 6 datasets with differences, the changes are: 🔴 2 large · 🟡 2 moderate · 📝 1 metadata-only · ➕ 1 new-data-only (tiered by worst-column anomaly score; coverage loss ⇒ 🔴)
Top changes — what to watch
Datasets
  1. 🔴 garden/artificial_intelligence/2025-03-12/epoch🔴 1 · 🟢 5 column(s) changed− lost 1 data point(s)8 column(s) in 1 table(s)
  2. 🔴 garden/artificial_intelligence/2025-03-12/epoch_regressions🔴 1 · 🟡 2 · 🟢 3 column(s) changed− lost 1 data point(s)8 column(s) in 1 table(s)
  3. 🟡 garden/artificial_intelligence/2025-03-12/epoch_compute_intensive_countries🟡 1 · 🟢 1 column(s) changed2 column(s) in 1 table(s)
  4. 🟡 garden/artificial_intelligence/2025-03-12/epoch_aggregates_affiliation🟡 1 · 🟢 1 column(s) changed2 column(s) in 1 table(s)
  5. garden/artificial_intelligence/2025-03-12/epoch_compute_intensivenew data only
  6. 📝 garden/artificial_intelligence/2026-01-30/frontiermathmetadata-only changes
Indicators
  1. 🔴 garden/artificial_intelligence/2025-03-12/epoch · epoch− lost 1 data point(s): 28348
  2. 🔴 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions− lost 1 data point(s): 28348
  3. 🔴 garden/artificial_intelligence/2025-03-12/epoch · epoch.domainnot scored (non-numeric) · 0% of rows
  4. 🔴 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.domainnot scored (non-numeric) · 0% of rows
  5. 🟡 garden/artificial_intelligence/2025-03-12/epoch_compute_intensive_countries · epoch_compute_intensive_countries.yearly_countmedian anomaly score 3.5% · 1% of rows
  6. 🟡 garden/artificial_intelligence/2025-03-12/epoch_aggregates_affiliation · epoch_aggregates_affiliation.yearly_countmedian anomaly score 1.8% · 0% of rows
  7. 🟡 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.parametersmedian anomaly score 1.2% · 0% of rows
  8. 🟡 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.training_dataset_size__totalmedian anomaly score 1.2% · 0% of rows
  9. 🟢 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.training_computation_petaflopmedian anomaly score 0.30% · 0% of rows
  10. 🟢 garden/artificial_intelligence/2025-03-12/epoch_compute_intensive_countries · epoch_compute_intensive_countries.cumulative_countmedian anomaly score 0.17% · 1% of rows
  11. 🟢 garden/artificial_intelligence/2025-03-12/epoch_aggregates_affiliation · epoch_aggregates_affiliation.cumulative_countmedian anomaly score 0.17% · 0% of rows
  12. 🟢 garden/artificial_intelligence/2025-03-12/epoch · epoch.organization_categorizationmedian anomaly score 0.10% · 0% of rows
  13. 🟢 garden/artificial_intelligence/2025-03-12/epoch · epoch.parametersmedian anomaly score 0.10% · 0% of rows
  14. 🟢 garden/artificial_intelligence/2025-03-12/epoch · epoch.publication_datemedian anomaly score 0.10% · 0% of rows
  15. 🟢 garden/artificial_intelligence/2025-03-12/epoch · epoch.training_computation_petaflopmedian anomaly score 0.10% · 0% of rows
  16. 🟢 garden/artificial_intelligence/2025-03-12/epoch · epoch.training_dataset_size__totalmedian anomaly score 0.10% · 0% of rows
  17. 🟢 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.organization_categorizationmedian anomaly score 0.09% · 0% of rows
  18. 🟢 garden/artificial_intelligence/2025-03-12/epoch_regressions · epoch_regressions.publication_datemedian anomaly score 0.09% · 0% of rows
~ garden/artificial_intelligence/2025-03-12/epoch🔴 1 · 🟢 5 column(s) changed− 1 row(s) removed: 283488 columns in 1 table
~ Table epoch🔴 1 · 🟢 5 column(s) changed
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'
~ epoch.domain 🔴 not scored (non-numeric)new datachanged data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modeldomain
28275MiniMax-M3Multiple domains
28303LongCat-2.0Language
28343Qwen3.8-2.4T-A95BLanguage
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modeldomain
28348DeepSeek-V4-Pro-0813Language
~ Changed values: 1 / 1,052 (0.10%)
days_since_1949modeldomain -domain +
28323Qwen 3.8 MaxLanguageMultiple domains
~ epoch.days_since_1949 dimnew dataremoved data
+ New values: 3 / 1,052 (0.29%)
modeldays_since_1949
MiniMax-M328275
LongCat-2.028303
Qwen3.8-2.4T-A95B28343
- Removed values: 1 / 1,052 (0.10%)
modeldays_since_1949
DeepSeek-V4-Pro-081328348
~ epoch.model dimnew dataremoved data
+ New values: 3 / 1,052 (0.29%)
days_since_1949model
28275MiniMax-M3
28303LongCat-2.0
28343Qwen3.8-2.4T-A95B
- Removed values: 1 / 1,052 (0.10%)
days_since_1949model
28348DeepSeek-V4-Pro-0813
~ epoch.organization_categorization 🟢 median anomaly score 0.10%new datachanged data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modelorganization_categorization
28275MiniMax-M3Industry
28303LongCat-2.0Industry
28343Qwen3.8-2.4T-A95BIndustry
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modelorganization_categorization
28348DeepSeek-V4-Pro-0813Industry
~ epoch.parameters 🟢 median anomaly score 0.10%new datachanged data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modelparameters
28275MiniMax-M3428000000000
28303LongCat-2.01600000000000
28343Qwen3.8-2.4T-A95B2400000000000
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modelparameters
28348DeepSeek-V4-Pro-08131600000000000
~ epoch.publication_date 🟢 median anomaly score 0.10%new datachanged data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modelpublication_date
28275MiniMax-M32026-06-01 00:00:00
28303LongCat-2.02026-06-29 00:00:00
28343Qwen3.8-2.4T-A95B2026-08-08 00:00:00
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modelpublication_date
28348DeepSeek-V4-Pro-08132026-08-13 00:00:00
~ epoch.training_computation_petaflop 🟢 median anomaly score 0.10%new data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modeltraining_computation_petaflop
28275MiniMax-M3NaN
28303LongCat-2.010080000000.0
28343Qwen3.8-2.4T-A95BNaN
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modeltraining_computation_petaflop
28348DeepSeek-V4-Pro-0813NaN
~ epoch.training_dataset_size__total 🟢 median anomaly score 0.10%new data
+ New values: 3 / 1,052 (0.29%)
days_since_1949modeltraining_dataset_size__total
28275MiniMax-M3NaN
28303LongCat-2.035000000000000
28343Qwen3.8-2.4T-A95BNaN
- Removed values: 1 / 1,052 (0.10%)
days_since_1949modeltraining_dataset_size__total
28348DeepSeek-V4-Pro-0813NaN
~ garden/artificial_intelligence/2025-03-12/epoch_regressions🔴 1 · 🟡 2 · 🟢 3 column(s) changed− 1 row(s) removed: 283488 columns in 1 table
~ Table epoch_regressions🔴 1 · 🟡 2 · 🟢 3 column(s) changed
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'
~ epoch_regressions.domain 🔴 not scored (non-numeric)new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modeldomain
28275MiniMax-M3Multiple domains
28303LongCat-2.0Language
28343Qwen3.8-2.4T-A95BLanguage
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modeldomain
28348DeepSeek-V4-Pro-0813Language
~ Changed values: 1 / 1,064 (0.09%)
days_since_1949modeldomain -domain +
28323Qwen 3.8 MaxLanguageMultiple domains
~ epoch_regressions.parameters 🟡 median anomaly score 1.2%new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modelparameters
28275MiniMax-M3427999985664.0
28303LongCat-2.01599999967232.0
28343Qwen3.8-2.4T-A95B2399999885312.0
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modelparameters
28348DeepSeek-V4-Pro-08131599999967232.0
~ Changed values: 2 / 1,064 (0.19%)
days_since_1949modelparameters -parameters +anomaly score
222802.2x/year between 2010–2026201568.90625196500.6251.3%
283652.2x/year between 2010–2026103824801792.0106358702080.01.2%
~ epoch_regressions.training_dataset_size__total 🟡 median anomaly score 1.2%new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modeltraining_dataset_size__total
28275MiniMax-M3NaN
28303LongCat-2.035000000970752.0
28343Qwen3.8-2.4T-A95BNaN
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modeltraining_dataset_size__total
28348DeepSeek-V4-Pro-0813NaN
~ Changed values: 2 / 1,064 (0.19%)
days_since_1949modeltraining_dataset_size__total -training_dataset_size__total +anomaly score
283652.9x/year between 2010–20261611140300800.01651130105856.01.2%
222802.9x/year between 2010–202627210.562526629.9980468751.1%
~ epoch_regressions.training_computation_petaflop 🟢 median anomaly score 0.30%new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modeltraining_computation_petaflop
28275MiniMax-M3NaN
28303LongCat-2.010080000000.0
28343Qwen3.8-2.4T-A95BNaN
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modeltraining_computation_petaflop
28348DeepSeek-V4-Pro-0813NaN
~ Changed values: 2 / 1,064 (0.19%)
days_since_1949modeltraining_computation_petaflop -training_computation_petaflop +anomaly score
222804.3x/year between 2010–20260.170551523566246030.169469058513641360.32%
283654.3x/year between 2010–20265538011136.05570266624.00.29%
~ epoch_regressions.days_since_1949 dimnew dataremoved data
+ New values: 3 / 1,064 (0.28%)
modeldays_since_1949
MiniMax-M328275
LongCat-2.028303
Qwen3.8-2.4T-A95B28343
- Removed values: 1 / 1,064 (0.09%)
modeldays_since_1949
DeepSeek-V4-Pro-081328348
~ epoch_regressions.model dimnew dataremoved data
+ New values: 3 / 1,064 (0.28%)
days_since_1949model
28275MiniMax-M3
28303LongCat-2.0
28343Qwen3.8-2.4T-A95B
- Removed values: 1 / 1,064 (0.09%)
days_since_1949model
28348DeepSeek-V4-Pro-0813
~ epoch_regressions.organization_categorization 🟢 median anomaly score 0.09%new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modelorganization_categorization
28275MiniMax-M3Industry
28303LongCat-2.0Industry
28343Qwen3.8-2.4T-A95BIndustry
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modelorganization_categorization
28348DeepSeek-V4-Pro-0813Industry
~ epoch_regressions.publication_date 🟢 median anomaly score 0.09%new datachanged data
+ New values: 3 / 1,064 (0.28%)
days_since_1949modelpublication_date
28275MiniMax-M32026-06-01 00:00:00
28303LongCat-2.02026-06-29 00:00:00
28343Qwen3.8-2.4T-A95B2026-08-08 00:00:00
- Removed values: 1 / 1,064 (0.09%)
days_since_1949modelpublication_date
28348DeepSeek-V4-Pro-08132026-08-13 00:00:00
~ garden/artificial_intelligence/2025-03-12/epoch_compute_intensive_countries🟡 1 · 🟢 1 column(s) changed2 columns in 1 table
~ Table epoch_compute_intensive_countries🟡 1 · 🟢 1 column(s) changed
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'
~ epoch_compute_intensive_countries.yearly_count 🟡 median anomaly score 3.5%changed metadatachanged data
metadata diff
-   systems are affiliated. The 2026 data is incomplete and was last updated 19 August 2026.
+   systems are affiliated. The 2026 data is incomplete and was last updated 30 August 2026.
~ Changed values: 2 / 176 (1.14%)
yearcountryyearly_count -yearly_count +anomaly score
2026China9105.3%
2026All large-scale AI systems30311.6%
~ epoch_compute_intensive_countries.cumulative_count 🟢 median anomaly score 0.17%changed metadatachanged data
metadata diff
-   systems are affiliated. The 2026 data is incomplete and was last updated 19 August 2026.
+   systems are affiliated. The 2026 data is incomplete and was last updated 30 August 2026.
~ Changed values: 2 / 176 (1.14%)
yearcountrycumulative_count -cumulative_count +anomaly score
2026China2022030.25%
2026All large-scale AI systems5265270.09%
~ garden/artificial_intelligence/2025-03-12/epoch_aggregates_affiliation🟡 1 · 🟢 1 column(s) changed2 columns in 1 table
~ Table epoch_aggregates_affiliation🟡 1 · 🟢 1 column(s) changed
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'
~ epoch_aggregates_affiliation.yearly_count 🟡 median anomaly score 1.8%changed metadatachanged data
metadata diff
-   2026 data is incomplete and was last updated 19 August 2026.
+   2026 data is incomplete and was last updated 30 August 2026.
~ Changed values: 1 / 248 (0.40%)
yearorganization_categorizationyearly_count -yearly_count +anomaly score
2026Industry56581.8%
~ epoch_aggregates_affiliation.cumulative_count 🟢 median anomaly score 0.17%changed metadatachanged data
metadata diff
-   2026 data is incomplete and was last updated 19 August 2026.
+   2026 data is incomplete and was last updated 30 August 2026.
~ Changed values: 1 / 248 (0.40%)
yearorganization_categorizationcumulative_count -cumulative_count +anomaly score
2026Industry5875890.17%
~ garden/artificial_intelligence/2025-03-12/epoch_compute_intensive6 columns in 1 table
~ Table epoch_compute_intensive
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'
~ epoch_compute_intensive.days_since_1949 dimnew data
+ New values: 1 / 527 (0.19%)
modeldays_since_1949
LongCat-2.028303
~ epoch_compute_intensive.model dimnew data
+ New values: 1 / 527 (0.19%)
days_since_1949model
28303LongCat-2.0
~ epoch_compute_intensive.domain new data
+ New values: 1 / 527 (0.19%)
days_since_1949modeldomain
28303LongCat-2.0Language
~ epoch_compute_intensive.parameters new data
+ New values: 1 / 527 (0.19%)
days_since_1949modelparameters
28303LongCat-2.01600000000000
~ epoch_compute_intensive.publication_date new data
+ New values: 1 / 527 (0.19%)
days_since_1949modelpublication_date
28303LongCat-2.02026-06-29 00:00:00
~ epoch_compute_intensive.training_computation_petaflop new data
+ New values: 1 / 527 (0.19%)
days_since_1949modeltraining_computation_petaflop
28303LongCat-2.010080000000.0
~ garden/artificial_intelligence/2026-01-30/frontiermath1 table metadata
~ Table epoch_benchmark_data
table metadata diff
-     date_accessed: '2026-08-19'
+     date_accessed: '2026-08-30'