Evidaxis

TRL

repo incumbent
Post-training & Alignment · n = 14 · huggingface/trl
Calibration · conf low
momentum
momentum 63.9 of 100 63.9 /100
momentum
63.9/100
pctile
p92
convergence gate
Convergence gate: velocity only, citation axis absent ∅
gate
1 of 2 axes

velocity only, citation axis absent.

one axis passes
position in cohort distribution one of 14 in cohort
percentile 92 within cohort p25 p50 p75 p0 p100
percentile
p92
momentum is the combined velocity and citation signal, percentile within cohort
velocity z
velocity z gaugevelocity z+1.11 within cohort, log-slope +0.009/wk.0+1.11
velocity z
+1.11 σ
slope
+0.009/wk
log-slope +0.009/wk
convergence planefull gate = rising
TRL on the convergence plane TRL at velocity z +1.11 and citation z no axis, against 0 dual-axis cohort peers. single-axis: citation axis absent, excluded from the plane. z>=1 median velocity z → citation z ↑
velocity z
+1.11
citation z
no axis
peers
13

single-axis: citation axis absent, excluded from the plane.

citation z no citation axis
cohort
14
systems measured
weekly commits
43.1
rising trend
total citations
n/a
no axis
momentum
63.9
percentile p92
commit heatmap, last 26 weeks43.1/wk
weekly commits, 26 weeks26 weeks of commits, peak 77 in a week, 26 active weeks, 993 total.week 1: 67 commitsweek 2: 29 commitsweek 3: 41 commitsweek 4: 20 commitsweek 5: 35 commitsweek 6: 26 commitsweek 7: 32 commitsweek 8: 32 commitsweek 9: 41 commitsweek 10: 38 commitsweek 11: 32 commitsweek 12: 26 commitsweek 13: 36 commitsweek 14: 21 commitsweek 15: 23 commitsweek 16: 26 commitsweek 17: 36 commitsweek 18: 51 commitsweek 19: 60 commitsweek 20: 26 commitsweek 21: 74 commitsweek 22: 44 commitsweek 23: 47 commitsweek 24: 77 commitsweek 25: 32 commitsweek 26: 21 commits
weeks
26
peak/wk
77
active
26
total
993
citations by yearno axis
adoption 137 dependents 124 direct, 13 indirect · pypi/trl · deps.dev · adoption signal, not part of the momentum score

TRL holds an Evidaxis momentum score of 63.9 of 100 on methodology m2, snapshot 2026-08-08, measured within the Post-training & Alignment cohort (n = 14). It is a mature incumbent included to calibrate the cohort, measured but not gate-eligible.

Derived signals

second-order reads, computed from the published series

Three signals ship when the commit series is statistically sufficient; the rest are held in a reserved state until the data supports them. A reserved signal shows its reason, never a fabricated number. How these are computed.

recency-weighted swing trailing 13-week vs prior 39-week log-slope ratio +3.69, regime accelerating. prior slope +0.010, recent slope +0.036 per week. +0.010prior 39w+0.036recent 13w+3.69×
swing
+3.69×
regime
accelerating
windows
13w vs 39w
snapshot 1 of 1 · · m2
commit-axis Sharpe log-linear slope over residual std, value +0.021 with 95% CI -0.001 to +0.044, 50 degrees of freedom. CI spans zero.0+0.02195% CI -0.001, +0.044-0.001+0.044
sharpe
+0.021
95% ci
-0.001, +0.044
dof
50
snapshot 1 of 1 · · m2
commit changepoint52 weekly commit points. Changepoint at week 50 (2026-07-25), slope change -28.415 commits per week.2026-07-25
date
week
50 of 52
Δslope
-28.415/wk
snapshot 1 of 1 · · m2
code-leads-citation lag reserved

reserved: citation series too short / axis absent for cross-correlation

status
reserved
value
insufficient data
snapshot 1 of 1 · · m2
axis alpha vs cohort reserved

reserved: cohort N too small for regression (power ~0)

status
reserved
value
insufficient data
snapshot 1 of 1 · · m2
gate ETA reserved

reserved: needs snapshot history (1 snapshot exists)

status
reserved
value
insufficient history
snapshot 1 of 1 · · m2
methodology gate snapshot 1 of 1 · · m2
commit weeks
52
citation years
0
shipped
3
reserved
3

shipped

  • recency-swing
  • commit-sharpe
  • commit-changepoint

reserved

  • code-leads-citation-lag
  • axis-alpha
  • gate-eta
snapshot 1 of 1 · · m2

Adoption trend

TRL dependents over time6 points, range 133 to 137 dependents.
min
133 dependents
max
137 dependents
latest
137 dependents
points
6

Downstream packages depending on pypi/trl (deps.dev), captured point-in-time each snapshot. An adoption signal, not part of the momentum score.

Verdict

TRL is a mature incumbent included to calibrate its cohort. It is measured but not badge-eligible.

Sources

huggingface/trl development-velocity axis

Embed this badge

Evidaxis calibration, TRL
[![Evidaxis: TRL](https://evidaxis.org/badge/e_B1Q69HZTTMA/2026-w32.svg)](https://evidaxis.org/e/e_B1Q69HZTTMA)

Others in Post-training & Alignment

AReaL Post-training & Alignment
Single-axis
21
p23
vel -2.31 cit no axis
axis dials
-2.31 n/a
Axolotl Post-training & Alignment
Single-axis
47
p38
vel -0.24 cit no axis
axis dials
-0.24 n/a
LLaMA-Factory Post-training & Alignment
Single-axis
50
p46
vel -0.01 cit no axis
axis dials
-0.01 n/a
ms-swift Post-training & Alignment
Single-axis
50
p54
vel -0.02 cit no axis
axis dials
-0.02 n/a
NeMo-RL Post-training & Alignment
Watch
88
p100
vel +3.01 cit no axis
axis dials
+3.01 n/a
Open R1 Post-training & Alignment
Single-axis
53
p62
vel +0.23 cit no axis
axis dials
+0.23 n/a

Full Post-training & Alignment cohort →

entity_id e_B1Q69HZTTMA methodology m2 snapshot 035a72a81e59 period 2026-w32 record JSON ↓ source snapshot

Score derived from public data on Evidaxis methodology m2, never edited by hand. Cite as: “Evidaxis, TRL, entity e_B1Q69HZTTMA, snapshot 2026-w32.”

Durable reference (format-independent): urn:evidaxis:claim:e_B1Q69HZTTMA:m2:2026-08-08