TETRACTA · MODEL X-RAY

Quantization simulation

Source model compared with its simulated lower-precision state
Depth view · Decision, output status, available verification evidence and categorical depth observations
VG1 · betars-1.7 · rp-1.3
A · SOURCE
mistralai/Mistral-7B-Instruct-v0.3
@ c170c708c41dac9275d15a8fff4eca08d52bab71
Reference artifact
B · PARTIAL INT8 WEIGHT SIMULATION
mistralai/Mistral-7B-Instruct-v0.3
Not a deployed quantized artifact. Derived in memory from A.
VALIDATION-PENDING DESCRIPTIVE RESULT · no calibrated risk or pass/fail grade
Difference observed
The response difference met the versioned instrument decision rule under this exact simulation contract.
Internal response
Observed
Response difference on the evaluated set
Text output
Changed
At least one generated continuation differs under exact token-sequence comparison. The comparison uses the same tokenizer and greedy-generation contract; it is not a semantic-quality judgment.
Repeat check
Single record
No repeat-check summary is included in this copy.

Included summaries Content and scope, not a quality or risk grade

Record or summaryThis report
Measurement summaryIncluded
Repeat-check summaryNot included
Same-artifact control summaryNot included
Underlying recordsRetained in the VG1 archive; not included in this copy
Operational receiptLinked · limited operational record
Statistical noise envelopeNot certified

Repeat and control details

Repeat checkOne measurement is summarized; no repeat-check summary is included in this copy.
Same-artifact controlNo matching same-artifact control summary is included in this copy.
Operational receiptThe link identifies a limited Tetracta operational record. Verify its signature and stated scope at the linked endpoint; the presence of a link is not signature verification.
Noise-floor statusNo model-specific statistical noise envelope is certified by this report.

What this means for you

MeasuredThe internal-response comparison met the instrument decision rule on the fixed input set. Use this as evidence of a recorded change when reviewing the candidate; the report does not rate that change as beneficial or harmful.
Text outputThe output-text comparison also observed a change.
PriorityCheck output stability first. Exact continuations changed with a shared output contract. Review formatting and exact-match workloads first; this scan does not determine whether those changes are improvements or regressions.

How the finding was decided

The reported internal finding uses the fixed, thresholded decision rule recorded for mv-1.4; it is not a test for any non-zero floating-point value. This rule is not, by itself, a measured noise floor for this model.

Text output is evaluated separately under exact greedy token-sequence comparison. The outcome is on the fixed input set, not a population rate or a capability score.

Structural coverage: 24 contributing probes per evaluated position. This is neither the number of output continuations nor a count of independent tasks.

Observed responses through the model

Boundary unresolved — difference already present at the first measured point (decoder block 4 output). Earlier positions were not evaluated, so the beginning of the difference cannot be located.

embedding output: not evaluatedEmbdecoder block 1 output: not evaluateddecoder block 2 output: not evaluateddecoder block 3 output: not evaluateddecoder block 4 output: difference observed4decoder block 5 output: difference observeddecoder block 6 output: difference observeddecoder block 7 output: difference observeddecoder block 8 output: difference observed8decoder block 9 output: difference observeddecoder block 10 output: difference observeddecoder block 11 output: difference observeddecoder block 12 output: difference observed12decoder block 13 output: difference observeddecoder block 14 output: difference observeddecoder block 15 output: difference observeddecoder block 16 output: difference observed16decoder block 17 output: difference observeddecoder block 18 output: difference observeddecoder block 19 output: difference observeddecoder block 20 output: difference observed20decoder block 21 output: difference observeddecoder block 22 output: difference observeddecoder block 23 output: difference observeddecoder block 24 output: difference observed24decoder block 25 output: difference observeddecoder block 26 output: difference observeddecoder block 27 output: difference observeddecoder block 28 output: difference observed28decoder block 29 output: difference observeddecoder block 30 output: difference observeddecoder block 31 output: difference observedfinal normalized output: difference observedFinal
difference observedno difference observednot evaluated

29 of 29 evaluated positions show a difference. This count describes the reach of the observed response, not its size or the number of changed weights.

Recorded outputObservation
embedding output through decoder block 3 outputnot evaluated
decoder block 4 output through decoder block 31 outputdifference observed
final normalized outputdifference observed

Blocks are numbered from 1. Final is the output after the last normalization; the last block before normalization is not separately displayed. Not evaluated means no observation was made there; it is not a below-threshold result.

A difference can propagate from one output to later outputs. This view does not locate edited weights, identify a cause or show which capabilities changed. Controlled small-model findings are not a validation result for this model or execution environment.

What A and B represent

A is the source model. B is a partial in-memory block-weight simulation of that same model. B has no separate repository revision or deployable file.

The result describes this simulation only. It cannot establish the memory use, latency, task quality or numerical behavior of a packed inference implementation.

Measured weight changes Same artifact pair and recorded response measurement

Overall relative weight change: 0.2098%. Internal response: Difference observed. Text output: Changed on the evaluated inputs.

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 110957605.

Parameter groupRelative changeAbsolute L2 changeRecorded response after the block
Decoder block 10.9979%0.5573Not evaluated
Decoder block 20.7458%0.4908Not evaluated
Decoder block 30.4253%0.4628Not evaluated
Decoder block 40.4716%0.4633Difference observed
Decoder block 50.4190%0.4710Difference observed
Decoder block 60.3658%0.4625Difference observed
Decoder block 70.3750%0.4711Difference observed
Decoder block 80.3427%0.4690Difference observed
Decoder block 90.3304%0.4689Difference observed
Decoder block 100.2957%0.4683Difference observed
Decoder block 110.3126%0.4797Difference observed
Decoder block 120.2932%0.4847Difference observed
Decoder block 130.2681%0.4815Difference observed
Decoder block 140.2660%0.4801Difference observed
Decoder block 150.2617%0.4940Difference observed
Decoder block 160.2327%0.4816Difference observed
Decoder block 170.2257%0.4800Difference observed
Decoder block 180.2346%0.4919Difference observed
Decoder block 190.2119%0.4810Difference observed
Decoder block 200.2032%0.4814Difference observed
Decoder block 210.1885%0.4681Difference observed
Decoder block 220.1772%0.4572Difference observed
Decoder block 230.1714%0.4529Difference observed
Decoder block 240.1668%0.4526Difference observed
Decoder block 250.1583%0.4543Difference observed
Decoder block 260.1546%0.4563Difference observed
Decoder block 270.1553%0.4574Difference observed
Decoder block 280.1533%0.4623Difference observed
Decoder block 290.1556%0.4735Difference observed
Decoder block 300.1531%0.4805Difference observed
Decoder block 310.1582%0.4975Difference observed
Decoder block 320.1648%0.5083Final normalized output: difference observed
Other registered parameters0.0%0.0No 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.

Recommended next checks Guidance, not a measurement

  1. Choose the actual candidateSelect the exact deployable quantized artifact and its inference implementation; retain the source revision with the comparison record.
  2. Check the customer workloadCompare source and deployed candidate on your own prompts. Record answer quality, formatting, latency and memory separately.
  3. Confirm the deployed changeRequest a direct before/after scan of the supported deployed artifact. A simulation result does not transfer automatically to that artifact.

Provenance and record

InstrumentVG1
Measurement contractmv-1.4
Report schema / presentationrs-1.7 / rp-1.3
Simulationqsim-2.0 · simulated int8

receipt xrr_zJHUlNjZbr2-nWiq9XqtsjBjMoaMh-1ppJJHFNqRM_Y · opaque operational record with limited assurance; scope at https://www.tetracta.ai/llm_tomografi/attest/receipt/xrr_zJHUlNjZbr2-nWiq9XqtsjBjMoaMh-1ppJJHFNqRM_Y

Recorded context

Pin these fields when you diff this report against a future scan. Matching context is a prerequisite for interpretation, not proof of the cause of a difference. The public receipt is a limited signed locator. The account owner can retrieve the full signed operational record from the report page.

Measurement time (UTC)2026-09-22T13:23:19Z
Model referencemistralai/Mistral-7B-Instruct-v0.3
Artifact evidenceArtifact-identity and measurement records are held in the owner-only operational record
HF revisionA: c170c708c41dac9275d15a8fff4eca08d52bab71 · B: derived simulation from A; no separate repository revision
Model B identitysimulated int8 (qsim-2.0) — in-memory transform of A; no separate deployable artifact or repository revision
Probe set / scan configps-1.1 / sc-1.0 · mv-1.4
Simulation versionqsim-2.0 · in-memory simulation; the deployed artifact was not scanned
Reference setno release reference set; estimator validation pending
Report schemars-1.7
Compute isolationmanaged compute; infrastructure identity is withheld
Runtime envelopefixed release configuration; bitwise equality across device types is not guaranteed
Operational receiptreceipt xrr_zJHUlNjZbr2-nWiq9XqtsjBjMoaMh-1ppJJHFNqRM_Y · opaque operational record with limited assurance; scope at https://www.tetracta.ai/llm_tomografi/attest/receipt/xrr_zJHUlNjZbr2-nWiq9XqtsjBjMoaMh-1ppJJHFNqRM_Y

Reading this report alongside another scan

Check the artifact pair, revisions and measurement contract first. The categorical findings describe each recorded comparison; they do not rank different models. A different contract or execution environment needs its own comparability evidence. No numerical severity score is supplied.

Use as a checkpoint reference

For supported checkpoint comparisons, retain the starting revision, this measurement contract and the output settings with your training record. Compare the next checkpoint against that same reference, then retain its task results alongside the scan.

For the next checkpointRecord alongside this scan
Reference and candidateExact repository revisions
Comparable measurementSame instrument, input contract and approved execution profile
Training changeYour fine-tune, merge or pruning record
Adoption decisionYour task acceptance results and deployment revision

A reference record supports change tracking. It does not define an optimal model or a target score for training. Simulation results describe the simulated state; actual checkpoint changes require their own before/after comparison.

Your findings and the protected instrument

Delivered to you

The observation, output-text status, artifact context and, where applicable, metadata comparison, plus provenance and categorical depth detail when included.

Kept private

Raw profiles, per-item records, probe construction, thresholds, transformations and internal traces. The report contains no model weights, credentials or infrastructure identifiers.

Honesty box — scope and limits

This is an in-memory simulation, not a packed kernel or a scan of a deployed quantized artifact. This descriptive result does not identify cause, capability, quality, safety or deployment impact. It is not a task-quality measurement. Do not target a layer or make a shipping decision from this report alone. No calibrated risk or pass/fail grade is assigned. No cross-model ranking or cross-device bitwise-equality claim is made.

Historical results under superseded estimators do not validate this release. The categorical depth view, when included, describes recorded responses; it is not a diagnosis of edited weights.