Research run — 22 September 2026. This comparison was produced with the same instrument release (xray_vg1, rs-1.7, mv-1.4) outside the customer path, because the customer path failed for a pair of two 7B-class checkpoints on that day (under investigation). It is an unsigned report: no operational receipt, no repeat or control summary, and it was not accepted through the signing server. Numbers are stated as recorded; nothing here is a quality, safety or deployment grade.
TETRACTA · MODEL X-RAY
Model X-Ray Report
Two pinned artifacts compared under a fixed measurement contract
report schema rs-1.7validation pending
A · BEFORE
mistralai/Mistral-7B-v0.3
→
B · AFTER
mistralai/Mistral-7B-Instruct-v0.3
≠
VALIDATION-PENDING DESCRIPTIVE RESULT — no calibrated risk or pass/fail grade
A response difference was observed under this exact contract.
A binary observation under the stated contract; not a location, magnitude or quality assessment.
A · Mistral-7B-v0.3 B · Mistral-7B-Instruct-v0 FIXED CONTRACTsame probes · same conditions COMPARE24 probes evaluated BINARY OBSERVATIONdifference observed schematic of the measurement path — no location, magnitude or profile is drawn
difference_observed
yes
whether the instrument observed a response difference between A and B under the measurement contract
output_text_change_observed
withheld
Withheld: the two artifacts do not share an output contract, so their recorded output text is not compared.
probes evaluated
24
size of the evaluated set at the reporting depth; this is not a count of probes that differed
pair relation
combined-artifact-change
recorded relation between the two artifacts under the pair contract; a combined-artifact change is not attributed to fine-tuning alone

What differs between A and B (public metadata)

Configuration comparison from the artifacts' own public files — the two configurations are identical in the compared fields.

Numeric-precision fields are not compared here; any file difference is recorded in the artifact identity.

This comparison is recorded as a combined artifact change (recorded differences: tokenizer semantics); the difference is not attributed to weights alone.

Artifact facts and run conditions

Measured weight changes Same artifact pair and recorded response measurement

Overall relative weight change: 0.4330%. Internal response: Difference observed. Text output: Withheld: output contracts differ.

Coverage: 291 unique parameter tensors; 7248023552 values. All declared floating parameters were checked; 0 shared names were counted once. Buffers are outside this weight summary.

Numerical validity: all checked values are finite. Observed precision: bfloat16. Zero values: reference 0, candidate 0.

Parameter groupRelative changeAbsolute L2 changeRecorded response after the block
Decoder block 11.551%0.8663Not evaluated
Decoder block 21.347%0.8866Not evaluated
Decoder block 30.8411%0.9151Not evaluated
Decoder block 40.9364%0.9197Difference observed
Decoder block 50.8243%0.9266Difference observed
Decoder block 60.7450%0.9421Difference observed
Decoder block 70.7544%0.9477Difference observed
Decoder block 80.6984%0.9556Difference observed
Decoder block 90.6787%0.9632Difference observed
Decoder block 100.6123%0.9695Difference observed
Decoder block 110.6308%0.9681Difference observed
Decoder block 120.5913%0.9775Difference observed
Decoder block 130.5442%0.9773Difference observed
Decoder block 140.5455%0.9846Difference observed
Decoder block 150.5278%0.9962Difference observed
Decoder block 160.4830%0.9995Difference observed
Decoder block 170.4725%1.005Difference observed
Decoder block 180.4797%1.006Difference observed
Decoder block 190.4414%1.002Difference observed
Decoder block 200.4193%0.9934Difference observed
Decoder block 210.3988%0.9904Difference observed
Decoder block 220.3814%0.9838Difference observed
Decoder block 230.3718%0.9820Difference observed
Decoder block 240.3622%0.9830Difference observed
Decoder block 250.3400%0.9757Difference observed
Decoder block 260.3312%0.9774Difference observed
Decoder block 270.3327%0.9798Difference observed
Decoder block 280.3223%0.9719Difference observed
Decoder block 290.3217%0.9787Difference observed
Decoder block 300.3091%0.9700Difference observed
Decoder block 310.3086%0.9707Difference observed
Decoder block 320.3030%0.9342Final normalized output: difference observed
Other registered parameters0.3034%1.054No matching block output

Relative change uses the reference norm; a zero reference has no relative percentage. Shared parameters belong to their first recorded group. The last block output is not separately measured; its row shows the final normalized output. Weight changes locate parameters to inspect, while response differences can propagate from earlier blocks. Neither identifies the cause of a task outcome.

Use the changed groups to choose targeted follow-up evaluations. No observed response difference does not establish equivalence, and a larger weight change is not a quality or severity score.

Provenance

measurement contractmv-1.4
report schemars-1.7
operational receiptno attached to this standalone copy

Record identity: 130a5df0d8dd — the first 12 characters of the ordered-pair identifier, derived from the two artifacts and their revisions; a different pair, or the same pair at a different revision, yields a different identifier.
This document is a record of a measurement, not an opinion about the models: its fields are determined by the two public artifacts and the contract versions named above. In the null control for this instrument release, repeat scans on the same board class reproduced byte-identical recorded metrics.

Interpretation

This result records only whether the frozen instrument observed a difference. A difference is reported when any evaluated probe's response differs between A and B beyond the instrument's observation floor; the floor's value and the probe design are private. The result does not locate the change. It does not identify a cause, establish that weights were the only changed artifact, measure capability or quality, or authorize a shipping decision. Compare A to B only; do not compare numbers between reports.

What is shown and what stays proprietary

What this report shows
  • the verdict, and the recorded differences in configuration, tokenizer and generation settings
  • onset depth, the depth map and the extent of the observable difference (when enabled for the instrument release)
  • the output-text status, provenance, record identity and the operational receipt
What remains proprietary
  • probe design and probe text
  • the observation floor and thresholds
  • raw statistics, raw profiles, transformations and internal traces
  • sampling geometry and item-level records

The finding is yours; the instrument stays proprietary and is held fixed and versioned so that two reports produced under the same contract can be compared with each other. This standalone derivative carries no attestation; signing status is unavailable unless a separate signed receipt accompanies it.

Version notice: produced under instrument release xray_vg1 with the contract versions shown above. Records produced under different contract versions are not interchangeable with this one, and earlier records may have been restated or withdrawn.
What this report does not do
Honesty box — scope and limits
This is a validation-pending descriptive measurement record, not a task-accuracy, safety, causal, deployment-readiness or regulatory result. Historical development findings used a superseded estimator and do not validate this release. The proprietary measurement recipe and raw evidence are intentionally withheld. Confirm decisions with task-specific evaluations on the exact deployed artifacts.
Tetracta · Model X-Ray · report schema rs-1.7 · this page contains no scripts and no numerical response profile.