Written entirely by Claude, as a working record for agents more than for human readers: every number with its source file, and every claim since withdrawn. Some facts here may be out of date; where it disagrees with the post, trust the post. For the readable version, see the companion.
FA_Competence (Competence)
The second factor of the k=5 solution, Tucker congruence +0.574 with Goldberg Conscientiousness, and behaviourally a register of diligence rather than actual competence.
Second factor of the centred k=5 oblimin solution; SS loading = 8.42
(qwen35/results/fa_qwen35.md section 7;
qwen35/analysis/fa_summary.json#centred_k5.factors[1].ss). Method on
Factor analysis of the adapter Gram.
Loadings
| rank | trait | loading | factor | keyed |
|---|---|---|---|---|
| 1 | Negligent | -0.625 | C | - |
| 2 | Conscientious | +0.618 | C | + |
| 3 | Casual | -0.589 | Lexicon | + |
| 4 | Unintellectual | -0.582 | I | - |
| 5 | Sloppy | -0.566 | C | - |
| 6 | Unintelligent | -0.563 | I | - |
| 7 | Neat | +0.560 | C | + |
| 8 | Careless | -0.531 | C | - |
| 9 | Prompt | +0.514 | C | + |
| 10 | Thorough | +0.488 | C | + |
Note that two Intellect markers (unintellectual, unintelligent) load negatively here alongside the Conscientiousness ones - the factor picks up a general "careful and capable" pole rather than a pure diligence pole.
Congruence with the Big Five keying
fa_qwen35.md section 6, centred_k5 oblimin row F2:
| E | A | C | ES | I | Eval | best Goldberg |
|---|---|---|---|---|---|---|
| -0.052 | -0.058 | +0.574 | +0.027 | +0.214 | +0.315 | C 0.57 |
The +0.214 on Intellect is the numerical form of the two Intellect markers in the
loading table. Its Eval congruence (+0.315) is the highest of the five factors -
it is the most evaluatively loaded of the k=5 solution
(fa_qwen35.md section 8, which records +0.315 as the maximum for centred_k5
oblimin, on F2).
Steering
qwen35/analysis/blog_data.json#replication.FA_Competence: judged as
Conscientiousness, slope2 = +1.0166666666666664,
sel2 = 2.7570621468926553, slope4 = +0.6121463077984818,
sel4 = 2.965005745325394.
qwen35/analysis/qual_fa.json#directions[1].axis_label:
Register of diligence: breezy adolescent improvisation at negative alpha, corporate proceduralism at positive alpha. Delivers its name, but as a register shift rather than as actual competence.
qual_fa_notes.md heads its section "FA_Competence - delivers its name, but a
quarter of the slope is damage". Negative pole: slangy, exclamatory, unplanned
("Oh wow", "totally pumped", "whatever feels right"), commitments treated as
optional. Positive pole: impersonal management prose - stakeholder analysis, work
breakdown structures, KPIs - refusing to improvise or speak in the first person.
See Steering the base model along weight-space directions.
Qualifications (audit, 2026-09-11)
Four caveats a reader of the loadings alone would miss (Factor audit, 2026-09-11).
The amplifying sign is fragile. At matched Fisher dose FA_Competence is the
only one of ten directions whose two signs move judged Conscientiousness the
same way: -2.875 suppressing and -0.042 amplifying, bipolar = false
(qwen35/analysis/matched_dose_steering.json#directions.FA_Competence). See
Suppression versus amplification at matched Fisher dose.
The Intellect co-movement is one-sided. Judged Intellect falls with the
suppressing sign and does not rise with the amplifying one (recomputed from
qwen35/phase10_runs/judged_steerfix23.json records: 2.96 at -2, 5.54 at 0,
5.33 at +2; see wiki/raw/factor-audit-2026-09-11.md). The "competence bundle"
is a fact about suppression.
Part of the negative pole belongs to Arousal. Competence and Arousal are the
most oblique pair in the solution (direction cosine -0.728,
qwen35/analysis/fa_chart_summary.json#factor_pairwise_cosines), and the only
four traits that cross-load at 0.3 or more on this factor all cross-load on
Arousal: Imperturbable, Composed, Shallow, Introverted
(qwen35/results/fa_qwen35.json#solutions.centred_k5.loadings.oblimin). The
breezy, exclamatory negative pole is partly Arousal's positive pole showing
through.
The self-report instruments do not see it. An adapter's Competence coordinate does not predict its BFI Conscientiousness (r -0.030, n 134) or its TRAIT Conscientiousness (r +0.089, p 0.709, n 20), where Warmth reaches +0.822 on TRAIT (Inspect personality evaluations (BFI and TRAIT)). Consistent with the axis being a register of diligence rather than a disposition the adapter can report.
Supporting: the intelligence markers Intellectual (+0.482), Unintellectual
(-0.582) and Unintelligent (-0.563) load here rather than on
FA_Imagination (Imagination), which fits "competence" better than "diligence", and
bf_conscientiousness_high/low land at +0.469 / -0.593 with nearest zoo
neighbours neat and casual (Big Five factor adapters (Persona Cartography's own ten dials)).
Status
Current. qwen35/direction_pages/FA_Competence.html is historical.
Related: Factor analysis of the adapter Gram, axis_Conscientiousness (the named Conscientiousness axis), PC1 - Flooding and Composure, Steering the base model along weight-space directions, Conscientious.
Sources
qwen35/results/fa_qwen35.mdqwen35/analysis/fa_summary.json#centred_k5.factorsqwen35/analysis/blog_data.json#replication.FA_Competenceqwen35/analysis/qual_fa.jsonqwen35/analysis/qual_fa_notes.mdqwen35/direction_pages/FA_Competence.html
Linked from
- Claims and evidence
- Factor analysis of the adapter Gram
- Factor audit, 2026-09-11
- axis_Conscientiousness (the named Conscientiousness axis)
- FA_Imagination (Imagination)
- PC1 - Flooding and Composure
- Persona Geology wiki
- Post: Persona Geology (2026-10-30)
- Qualitative reads and adjudications
- Start here for collaborators
- Careful
- Careless
- Casual
- Chivalrous
- Conscientious
- Dependable
- Disorganized
- Haphazard
- Helpful
- Inarticulate
- Inconsistent
- Intellectual
- Learned
- Neat
- Negligent
- Organized
- Prompt
- Shallow
- Sloppy
- Systematic
- Thorough
- Undependable
- Unemotional
- Unintellectual
- Unintelligent
- Unsystematic
- Traits by factor