szl-triage-qwen3.5-0.8b-lora-study5 API & Inference Endpoint

Before you use it

The recorded contamination verdict remains a release blocker. Published adapters do not establish deployment clearance.
Results describe the frozen 113-row family holdout; they do not establish generalization, calibration or production readiness.
The earlier 66-row gate report does not clear this later five-seed study.
The inline inference example remains withdrawn. Read the original implementation caveat below before planning an experiment; this card supplies no new runtime evidence.

The complete canonical source body follows, including all measured results, historical gate scope, contamination limits and the withdrawn inference example notice.

SZL TypeSafe Triage · Five-Seed LoRA Study

A measured model artifact with its limits attached.

This repository contains five LoRA training runs for structured triage: seeds 11, 23, 37, 53, and 71. It includes the default adapter at the repository root, every seed under adapters/, frozen evaluation identity, raw predictions, failure records, training receipts, and aggregate metrics.

Status

Table with columns: Boundary, Status
Boundary	Status
Public model publication	Published
Training	Measured
Frozen held-family evaluation	Measured
Release gate	BLOCKED — 11/12
Promotion	NOT_PROMOTABLE
Production replacement	No

Published does not mean promoted. The owner authorized publication of the trained research artifact. The existing contamination verdict and release boundary remain visible instead of being removed.

Historical gate scope

The root gate_report.json records an earlier 66-row gate with verdict PROMOTABLE. It is retained historical evidence, not promotion of this later five-seed, 113-row study. The current study remains BLOCKED — 11/12 and NOT_PROMOTABLE. Published adapter files and historical gate labels do not supersede this release boundary.

What it does

The adapter accepts a triage input and is trained to return only:

json

{"label":"...","state":"...","evidence":["..."]}

In plain language:

label is the model's proposed category.
state records whether it can decide or needs review.
evidence contains text spans intended to come directly from the input.
A REVIEW state is a refusal to overclaim.

Study design

Base model: unsloth/Qwen3.5-0.8B
LoRA rank: 16
LoRA alpha: 16
Epochs: 3
Learning rate: 0.0002
Precision: bfloat16
Quantization during training: none
Effective batch size: 4
Training rows: 515
Held-family rows: 113
Seeds: 11, 23, 37, 53, 71
Split rule: template families are kept together
Evaluation decoding: greedy
Malformed JSON repair: disabled

Measured results

Table with columns: Target, Valid JSON, Label, State, Joint, Refusal fidelity, Evidence grounded, Exact target evidence, Failure rows
Target	Valid JSON	Label	State	Joint	Refusal fidelity	Evidence grounded	Exact target evidence	Failure rows
base	0.0%	0.0%	0.0%	0.0%	0.0%	0/0 (N/A)	34/113 (30.1%)	113
seed-011	100.0%

Across the five adapters, mean joint accuracy was 100.0%, with sample standard deviation 0.0%.

These results describe this frozen 113-row family holdout. They do not establish broad generalization, calibration, or production readiness.

Evidence grounding checks whether each generated span occurs in its input; exact target evidence compares the complete ordered evidence list with the retained target list. These are different measures, with separate denominators shown in the table: generated spans for grounding and held rows for exact evidence. The evaluator's failure rows cover malformed JSON, joint label/state mismatch, or ungrounded spans; they do not include an exact-evidence-only mismatch. Thus zero failure rows does not establish perfect exact target evidence. The base's exact-evidence count includes empty-list comparisons despite zero valid JSON rows; it must not be interpreted as valid base predictions.

These are retained historical results. This card-only correction performs no new model evaluation, model loading, training, release clearance, or publication.

Repository layout

text

README.mdadapter_config.jsonadapter_model.safetensorstokenizer filesadapters/  seed-011/  seed-023/  seed-037/  seed-053/  seed-071/evidence/  frozen/  training/  evaluation/  adapter-index.json  environment-receipt.json

The root adapter is seed 11, retained as the tagged first measured run. The additional seed directories support reproducibility and stability inspection.

Inference implementation

Inline inference example withdrawn. The previous generated quick start contained malformed output indexing and could also route text through the wrong processor interface. No fresh inference was performed for this card correction.

Inspect the canonical evaluation implementation alongside the retained environment receipt and adapter identities before a new experiment. The historical script's main entry point combines training, evaluation, and publication; it is not a card-only repair command or a standalone inference quick start. This documentation change supplies no new runtime, held-out, or promotion evidence.

Using another seed

Download the repository and load one of these local directories:

text

adapters/seed-023adapters/seed-037adapters/seed-053adapters/seed-071

Each directory is a complete PEFT adapter. The SHA-256 digest for every adapter is recorded in evidence/adapter-index.json.

Evaluation rules

The held set was separated by template family rather than random row. This reduces direct template leakage between training and evaluation.

Every target was scored for:

Strict JSON validity
Exact label agreement
Exact state agreement
Joint label-and-state agreement
Refusal fidelity
False labels on refusal cases
Evidence grounding
Exact target-evidence agreement

Raw predictions and complete failure records are included so the headline numbers can be audited.

Limitations

Release remains blocked at 11/12.
Promotion has not been established.
The recorded contamination verdict still travels with the artifact.
The model is not calibrated.
The study does not establish generalization outside the frozen holdout.
The model is not a replacement for the deterministic decision engine.
Generated output must be parsed and validated before downstream use.
High-impact decisions require human or deterministic review.

Reproducibility

Source repository:

https://github.com/szl-holdings/szl-typesafe-triage

First measured run:

https://github.com/szl-holdings/szl-typesafe-triage/releases/tag/triage-lora-run1

Five-seed evidence tag:

https://github.com/szl-holdings/szl-typesafe-triage/releases/tag/triage-lora-study5-measured-20260922-111314

Citation

If this artifact is discussed, describe it as:

SZL TypeSafe Triage five-seed LoRA study: training and frozen evaluation measured; public experimental artifact; promotion not established.

Before you use it

The recorded contamination verdict remains a release blocker. Published adapters do not establish deployment clearance.
Results describe the frozen 113-row family holdout; they do not establish generalization, calibration or production readiness.
The earlier 66-row gate report does not clear this later five-seed study.
The inline inference example remains withdrawn. Read the original implementation caveat below before planning an experiment; this card supplies no new runtime evidence.

The complete canonical source body follows, including all measured results, historical gate scope, contamination limits and the withdrawn inference example notice.

SZL TypeSafe Triage · Five-Seed LoRA Study

A measured model artifact with its limits attached.

Status

Table with columns: Boundary, Status
Boundary	Status
Public model publication	Published
Training	Measured
Frozen held-family evaluation	Measured
Release gate	BLOCKED — 11/12
Promotion	NOT_PROMOTABLE
Production replacement	No

Historical gate scope

What it does

The adapter accepts a triage input and is trained to return only:

json

{"label":"...","state":"...","evidence":["..."]}

{"label":"...","state":"...","evidence":["..."]}

In plain language:

label is the model's proposed category.
state records whether it can decide or needs review.
evidence contains text spans intended to come directly from the input.
A REVIEW state is a refusal to overclaim.

Study design

Base model: unsloth/Qwen3.5-0.8B
LoRA rank: 16
LoRA alpha: 16
Epochs: 3
Learning rate: 0.0002
Precision: bfloat16
Quantization during training: none
Effective batch size: 4
Training rows: 515
Held-family rows: 113
Seeds: 11, 23, 37, 53, 71
Split rule: template families are kept together
Evaluation decoding: greedy
Malformed JSON repair: disabled

Measured results

Table with columns: Target, Valid JSON, Label, State, Joint, Refusal fidelity, Evidence grounded, Exact target evidence, Failure rows
Target	Valid JSON	Label	State	Joint	Refusal fidelity	Evidence grounded	Exact target evidence	Failure rows
base	0.0%	0.0%	0.0%	0.0%	0.0%	0/0 (N/A)	34/113 (30.1%)	113
seed-011	100.0%

Across the five adapters, mean joint accuracy was 100.0%, with sample standard deviation 0.0%.

These results describe this frozen 113-row family holdout. They do not establish broad generalization, calibration, or production readiness.

These are retained historical results. This card-only correction performs no new model evaluation, model loading, training, release clearance, or publication.

Repository layout

text

README.mdadapter_config.jsonadapter_model.safetensorstokenizer filesadapters/  seed-011/  seed-023/  seed-037/  seed-053/  seed-071/evidence/  frozen/  training/  evaluation/  adapter-index.json  environment-receipt.json

README.mdadapter_config.jsonadapter_model.safetensorstokenizer filesadapters/  seed-011/  seed-023/  seed-037/  seed-053/  seed-071/evidence/  frozen/  training/  evaluation/  adapter-index.json  environment-receipt.json

The root adapter is seed 11, retained as the tagged first measured run. The additional seed directories support reproducibility and stability inspection.

Inference implementation

Using another seed

Download the repository and load one of these local directories:

text

adapters/seed-023adapters/seed-037adapters/seed-053adapters/seed-071

adapters/seed-023adapters/seed-037adapters/seed-053adapters/seed-071

Each directory is a complete PEFT adapter. The SHA-256 digest for every adapter is recorded in evidence/adapter-index.json.

Evaluation rules

The held set was separated by template family rather than random row. This reduces direct template leakage between training and evaluation.

Every target was scored for:

Strict JSON validity
Exact label agreement
Exact state agreement
Joint label-and-state agreement
Refusal fidelity
False labels on refusal cases
Evidence grounding
Exact target-evidence agreement

Raw predictions and complete failure records are included so the headline numbers can be audited.

Limitations

Release remains blocked at 11/12.
Promotion has not been established.
The recorded contamination verdict still travels with the artifact.
The model is not calibrated.
The study does not establish generalization outside the frozen holdout.
The model is not a replacement for the deterministic decision engine.
Generated output must be parsed and validated before downstream use.
High-impact decisions require human or deterministic review.

Reproducibility

Source repository:

https://github.com/szl-holdings/szl-typesafe-triage

First measured run:

https://github.com/szl-holdings/szl-typesafe-triage/releases/tag/triage-lora-run1

Five-seed evidence tag:

https://github.com/szl-holdings/szl-typesafe-triage/releases/tag/triage-lora-study5-measured-20260922-111314

Citation

If this artifact is discussed, describe it as:

SZL TypeSafe Triage five-seed LoRA study: training and frozen evaluation measured; public experimental artifact; promotion not established.

szl-triage-qwen3.5-0.8b-lora-study5

README

Before you use it

SZL TypeSafe Triage · Five-Seed LoRA Study

Status

Historical gate scope

What it does

Study design

Measured results

Repository layout

Inference implementation

Using another seed

Evaluation rules

Limitations

Reproducibility

Citation

Explore FriendliAI today

README

Before you use it

SZL TypeSafe Triage · Five-Seed LoRA Study

Status

Historical gate scope

What it does

Study design

Measured results

Repository layout

Inference implementation

Using another seed

Evaluation rules

Limitations

Reproducibility

Citation