Upload PolyGuard usable model bundle: local-qwen-0-5b-active-smoke

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/tokenizer.json filter=lfs diff=lfs merge=lfs -text

usable_model_bundles/local-qwen-0-5b-active-smoke/README.md

@@ -0,0 +1,42 @@
+# PolyGuard Active Model Artifact Bundle
+
+Bundle: `local-qwen-0-5b-active-smoke`
+Model: `Qwen/Qwen2.5-0.5B-Instruct`
+Base model: `Qwen/Qwen2.5-0.5B-Instruct`
+Preferred artifact: `grpo_adapter`
+
+This bundle is meant for implementation/testing while the full per-model remote sweep artifacts are still uploading.
+
+## Contents
+
+- `checkpoints/grpo_adapter/`
+- `checkpoints/sft_adapter/`
+- `checkpoints/merged/` when included
+- `manifests/active_model_manifest.json`
+- `reports/`
+
+## Restore Locally
+
+```bash
+cd /Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl
+cp -R submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter checkpoints/grpo_adapter
+cp -R submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter checkpoints/sft_adapter
+cp -R submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/merged checkpoints/merged
+mkdir -p checkpoints/active
+cp submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/manifests/active_model_manifest.json checkpoints/active/active_model_manifest.json
+curl http://127.0.0.1:8200/policy/model_status
+```
+
+## Hugging Face Download
+
+After upload, download with:
+
+```bash
+export HF_TOKEN="$(cat ~/.cache/huggingface/token)"
+huggingface-cli download TheJackBright/polyguard-openenv-training-full-artifacts \
+  --repo-type model \
+  --include 'usable_model_bundles/local-qwen-0-5b-active-smoke/**' \
+  --local-dir ./hf_artifacts
+```
+
+Note: this is the current local active Qwen 0.5B implementation bundle. It is not the final full remote Qwen 0.5B/1.5B sweep checkpoint until those files appear in the HF artifact repo.

usable_model_bundles/local-qwen-0-5b-active-smoke/bundle_manifest.json

@@ -0,0 +1,65 @@
+{
+  "status": "ok",
+  "bundle_name": "local-qwen-0-5b-active-smoke",
+  "created_at_utc": "2026-04-26T04:54:50.382018+00:00",
+  "source": "local_active_model",
+  "run_id": "qwen-qwen2-5-0-5b-instruct",
+  "label": "local-qwen-0.5b-active-smoke",
+  "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+  "base_model": "Qwen/Qwen2.5-0.5B-Instruct",
+  "preferred_artifact": "grpo_adapter",
+  "availability": {
+    "grpo_adapter": true,
+    "merged": true,
+    "sft_adapter": true
+  },
+  "remote_full_sweep_note": "The full Qwen 0.5B/1.5B remote sweep artifacts are still pending upload in the HF artifact repo. This bundle packages the currently active local trained/smoke artifacts for product integration.",
+  "copies": {
+    "grpo_adapter": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/checkpoints/grpo_adapter",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter",
+      "exists": true,
+      "file_count": 11,
+      "bytes": 20229991
+    },
+    "sft_adapter": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/checkpoints/sft_adapter",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter",
+      "exists": true,
+      "file_count": 11,
+      "bytes": 30655905
+    },
+    "active_manifest": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/checkpoints/active/active_model_manifest.json",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/manifests/active_model_manifest.json",
+      "exists": true,
+      "bytes": 5274
+    },
+    "active_report_manifest": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/active_model/active_model_manifest.json",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/manifests/active_model_report_manifest.json",
+      "exists": true,
+      "bytes": 5274
+    },
+    "submission_evidence_manifest": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/manifest.json",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/manifests/submission_evidence_manifest.json",
+      "exists": true,
+      "bytes": 12708
+    },
+    "reports": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/active_model",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/reports",
+      "exists": true,
+      "file_count": 44,
+      "bytes": 1768421
+    },
+    "merged": {
+      "source": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/checkpoints/merged",
+      "target": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/model_artifacts/local-qwen-0-5b-active-smoke/checkpoints/merged",
+      "exists": true,
+      "file_count": 11,
+      "bytes": 1003978294
+    }
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen2.5-0.5B-Instruct
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen2.5-0.5B-Instruct
+- grpo
+- lora
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.19.1

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-0.5B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d3c7721b362814dd6ca4b05efef788d3ec93a26a16848227f4575236f699ec8
+size 4338000

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/grpo_adapter/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:246b70bea7ca6b468c441c9b4feb273716381f71026cd6cdc63f9984cf06f2ef
+size 6584

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/config.json

@@ -0,0 +1,54 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "float16",
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.57.6"
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/merge_report.json

@@ -0,0 +1,11 @@
+{
+  "status": "ok",
+  "adapter_dir": "/app/checkpoints/sft_adapter",
+  "output_dir": "/app/checkpoints/merged",
+  "base_model": "Qwen/Qwen2.5-0.5B-Instruct",
+  "merge_dtype": "float16",
+  "load_in_4bit": false,
+  "unsafe_override": false,
+  "parameters": 494032768,
+  "precision_warning": "none"
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a02d64a18e37e82e8791290532c22ee8f771007808557e7181058dffeada070a
+size 988097536

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/merged/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/README.md

@@ -0,0 +1,209 @@
+---
+base_model: Qwen/Qwen2.5-3B-Instruct
+library_name: peft
+pipeline_tag: text-generation
+tags:
+- base_model:adapter:Qwen/Qwen2.5-3B-Instruct
+- lora
+- sft
+- transformers
+- trl
+---
+
+# Model Card for Model ID
+
+<!-- Provide a quick summary of what the model is/does. -->
+
+
+
+## Model Details
+
+### Model Description
+
+<!-- Provide a longer summary of what this model is. -->
+
+
+
+- **Developed by:** [More Information Needed]
+- **Funded by [optional]:** [More Information Needed]
+- **Shared by [optional]:** [More Information Needed]
+- **Model type:** [More Information Needed]
+- **Language(s) (NLP):** [More Information Needed]
+- **License:** [More Information Needed]
+- **Finetuned from model [optional]:** [More Information Needed]
+
+### Model Sources [optional]
+
+<!-- Provide the basic links for the model. -->
+
+- **Repository:** [More Information Needed]
+- **Paper [optional]:** [More Information Needed]
+- **Demo [optional]:** [More Information Needed]
+
+## Uses
+
+<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->
+
+### Direct Use
+
+<!-- This section is for the model use without fine-tuning or plugging into a larger ecosystem/app. -->
+
+[More Information Needed]
+
+### Downstream Use [optional]
+
+<!-- This section is for the model use when fine-tuned for a task, or when plugged into a larger ecosystem/app -->
+
+[More Information Needed]
+
+### Out-of-Scope Use
+
+<!-- This section addresses misuse, malicious use, and uses that the model will not work well for. -->
+
+[More Information Needed]
+
+## Bias, Risks, and Limitations
+
+<!-- This section is meant to convey both technical and sociotechnical limitations. -->
+
+[More Information Needed]
+
+### Recommendations
+
+<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->
+
+Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.
+
+## How to Get Started with the Model
+
+Use the code below to get started with the model.
+
+[More Information Needed]
+
+## Training Details
+
+### Training Data
+
+<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->
+
+[More Information Needed]
+
+### Training Procedure
+
+<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->
+
+#### Preprocessing [optional]
+
+[More Information Needed]
+
+
+#### Training Hyperparameters
+
+- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->
+
+#### Speeds, Sizes, Times [optional]
+
+<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->
+
+[More Information Needed]
+
+## Evaluation
+
+<!-- This section describes the evaluation protocols and provides the results. -->
+
+### Testing Data, Factors & Metrics
+
+#### Testing Data
+
+<!-- This should link to a Dataset Card if possible. -->
+
+[More Information Needed]
+
+#### Factors
+
+<!-- These are the things the evaluation is disaggregating by, e.g., subpopulations or domains. -->
+
+[More Information Needed]
+
+#### Metrics
+
+<!-- These are the evaluation metrics being used, ideally with a description of why. -->
+
+[More Information Needed]
+
+### Results
+
+[More Information Needed]
+
+#### Summary
+
+
+
+## Model Examination [optional]
+
+<!-- Relevant interpretability work for the model goes here -->
+
+[More Information Needed]
+
+## Environmental Impact
+
+<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->
+
+Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).
+
+- **Hardware Type:** [More Information Needed]
+- **Hours used:** [More Information Needed]
+- **Cloud Provider:** [More Information Needed]
+- **Compute Region:** [More Information Needed]
+- **Carbon Emitted:** [More Information Needed]
+
+## Technical Specifications [optional]
+
+### Model Architecture and Objective
+
+[More Information Needed]
+
+### Compute Infrastructure
+
+[More Information Needed]
+
+#### Hardware
+
+[More Information Needed]
+
+#### Software
+
+[More Information Needed]
+
+## Citation [optional]
+
+<!-- If there is a paper or blog post introducing the model, the APA and Bibtex information for that should go in this section. -->
+
+**BibTeX:**
+
+[More Information Needed]
+
+**APA:**
+
+[More Information Needed]
+
+## Glossary [optional]
+
+<!-- If relevant, include terms and calculations in this section that can help readers understand the model or model card. -->
+
+[More Information Needed]
+
+## More Information [optional]
+
+[More Information Needed]
+
+## Model Card Authors [optional]
+
+[More Information Needed]
+
+## Model Card Contact
+
+[More Information Needed]
+### Framework versions
+
+- PEFT 0.19.1

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/adapter_config.json

@@ -0,0 +1,43 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen2.5-3B-Instruct",
+  "bias": "none",
+  "corda_config": null,
+  "ensure_weight_tying": false,
+  "eva_config": null,
+  "exclude_modules": null,
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 32,
+  "lora_bias": false,
+  "lora_dropout": 0.05,
+  "lora_ga_config": null,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "peft_version": "0.19.1",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:494fce09fcbe83dc7ddc66b8fc681f040e43c1ac68693ce0db73ba3607802679
+size 14764792

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/chat_template.jinja

@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/checkpoints/sft_adapter/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:706b7816b266d97637317c189c5087868f9e27dd4669d62da81ca9914762e5ea
+size 5880

usable_model_bundles/local-qwen-0-5b-active-smoke/manifests/active_model_manifest.json

@@ -0,0 +1,68 @@
+{
+  "status": "ok",
+  "enabled": true,
+  "activated_at_utc": "2026-04-26T02:24:15.464507+00:00",
+  "run_id": "qwen-qwen2-5-0-5b-instruct",
+  "source": "top-level",
+  "label": "local-qwen-0.5b-active-smoke",
+  "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+  "base_model": "Qwen/Qwen2.5-0.5B-Instruct",
+  "preferred_artifact": "grpo_adapter",
+  "mode": "symlink",
+  "source_checkpoint_dir": "checkpoints",
+  "source_report_dir": "outputs/reports",
+  "grpo_adapter": "checkpoints/active/grpo_adapter",
+  "merged_model": "checkpoints/active/merged",
+  "sft_adapter": "checkpoints/active/sft_adapter",
+  "availability": {
+    "grpo_adapter": true,
+    "merged": true,
+    "sft_adapter": true
+  },
+  "reports": {
+    "improvement_report_benchmark.json": "outputs/reports/active_model/improvement_report_benchmark.json",
+    "anti_hacking_overfit_report.json": "outputs/reports/active_model/anti_hacking_overfit_report.json",
+    "grpo_trl_run_strict_check.json": "outputs/reports/active_model/grpo_trl_run_strict_check.json",
+    "postsave_inference.json": "outputs/reports/active_model/postsave_inference.json",
+    "sft_trl_run.json": "outputs/reports/active_model/sft_trl_run.json",
+    "plot_index.json": "outputs/reports/active_model/plot_index.json",
+    "dose_train.json": "outputs/reports/active_model/dose_train.json",
+    "baselines.json": "outputs/reports/active_model/baselines.json",
+    "robustness.json": "outputs/reports/active_model/robustness.json",
+    "grpo_trl_run_fallback_check.json": "outputs/reports/active_model/grpo_trl_run_fallback_check.json",
+    "sft_run.json": "outputs/reports/active_model/sft_run.json",
+    "benchmark_report.txt": "outputs/reports/active_model/benchmark_report.txt",
+    "dosing_grpo.json": "outputs/reports/active_model/dosing_grpo.json",
+    "grpo_ablation_report.json": "outputs/reports/active_model/grpo_ablation_report.json",
+    "frontier_ready.json": "outputs/reports/active_model/frontier_ready.json",
+    "improvement_report.json": "outputs/reports/active_model/improvement_report.json",
+    "hf_sweep_summary.json": "outputs/reports/active_model/hf_sweep_summary.json",
+    "planner_grpo.json": "outputs/reports/active_model/planner_grpo.json",
+    "grpo_trl_run.json": "outputs/reports/active_model/grpo_trl_run.json",
+    "risk_train.json": "outputs/reports/active_model/risk_train.json",
+    "grpo_trl_run_smoke.json": "outputs/reports/active_model/grpo_trl_run_smoke.json",
+    "inference_benchmark.json": "outputs/reports/active_model/inference_benchmark.json",
+    "supervisor_grpo.json": "outputs/reports/active_model/supervisor_grpo.json",
+    "acceptance_gate.json": "outputs/reports/active_model/acceptance_gate.json",
+    "grpo_trl_run_auto.json": "outputs/reports/active_model/grpo_trl_run_auto.json",
+    "hf_training_status.json": "outputs/reports/active_model/hf_training_status.json",
+    "benchmark_report.json": "outputs/reports/active_model/benchmark_report.json",
+    "postsave_inference_smoke.json": "outputs/reports/active_model/postsave_inference_smoke.json",
+    "graph_train.json": "outputs/reports/active_model/graph_train.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json",
+    "grpo_training_cycle/grpo_trl_run.json": "outputs/reports/active_model/grpo_training_cycle/grpo_trl_run.json",
+    "grpo_training_cycle/hf_training_status.json": "outputs/reports/active_model/grpo_training_cycle/hf_training_status.json"
+  },
+  "notes": "This manifest controls local product inference. Prefer grpo_adapter for the RL policy; merged is the SFT baseline fallback when no GRPO adapter is available."
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/manifests/active_model_report_manifest.json

@@ -0,0 +1,68 @@
+{
+  "status": "ok",
+  "enabled": true,
+  "activated_at_utc": "2026-04-26T02:24:15.464507+00:00",
+  "run_id": "qwen-qwen2-5-0-5b-instruct",
+  "source": "top-level",
+  "label": "local-qwen-0.5b-active-smoke",
+  "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+  "base_model": "Qwen/Qwen2.5-0.5B-Instruct",
+  "preferred_artifact": "grpo_adapter",
+  "mode": "symlink",
+  "source_checkpoint_dir": "checkpoints",
+  "source_report_dir": "outputs/reports",
+  "grpo_adapter": "checkpoints/active/grpo_adapter",
+  "merged_model": "checkpoints/active/merged",
+  "sft_adapter": "checkpoints/active/sft_adapter",
+  "availability": {
+    "grpo_adapter": true,
+    "merged": true,
+    "sft_adapter": true
+  },
+  "reports": {
+    "improvement_report_benchmark.json": "outputs/reports/active_model/improvement_report_benchmark.json",
+    "anti_hacking_overfit_report.json": "outputs/reports/active_model/anti_hacking_overfit_report.json",
+    "grpo_trl_run_strict_check.json": "outputs/reports/active_model/grpo_trl_run_strict_check.json",
+    "postsave_inference.json": "outputs/reports/active_model/postsave_inference.json",
+    "sft_trl_run.json": "outputs/reports/active_model/sft_trl_run.json",
+    "plot_index.json": "outputs/reports/active_model/plot_index.json",
+    "dose_train.json": "outputs/reports/active_model/dose_train.json",
+    "baselines.json": "outputs/reports/active_model/baselines.json",
+    "robustness.json": "outputs/reports/active_model/robustness.json",
+    "grpo_trl_run_fallback_check.json": "outputs/reports/active_model/grpo_trl_run_fallback_check.json",
+    "sft_run.json": "outputs/reports/active_model/sft_run.json",
+    "benchmark_report.txt": "outputs/reports/active_model/benchmark_report.txt",
+    "dosing_grpo.json": "outputs/reports/active_model/dosing_grpo.json",
+    "grpo_ablation_report.json": "outputs/reports/active_model/grpo_ablation_report.json",
+    "frontier_ready.json": "outputs/reports/active_model/frontier_ready.json",
+    "improvement_report.json": "outputs/reports/active_model/improvement_report.json",
+    "hf_sweep_summary.json": "outputs/reports/active_model/hf_sweep_summary.json",
+    "planner_grpo.json": "outputs/reports/active_model/planner_grpo.json",
+    "grpo_trl_run.json": "outputs/reports/active_model/grpo_trl_run.json",
+    "risk_train.json": "outputs/reports/active_model/risk_train.json",
+    "grpo_trl_run_smoke.json": "outputs/reports/active_model/grpo_trl_run_smoke.json",
+    "inference_benchmark.json": "outputs/reports/active_model/inference_benchmark.json",
+    "supervisor_grpo.json": "outputs/reports/active_model/supervisor_grpo.json",
+    "acceptance_gate.json": "outputs/reports/active_model/acceptance_gate.json",
+    "grpo_trl_run_auto.json": "outputs/reports/active_model/grpo_trl_run_auto.json",
+    "hf_training_status.json": "outputs/reports/active_model/hf_training_status.json",
+    "benchmark_report.json": "outputs/reports/active_model/benchmark_report.json",
+    "postsave_inference_smoke.json": "outputs/reports/active_model/postsave_inference_smoke.json",
+    "graph_train.json": "outputs/reports/active_model/graph_train.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json",
+    "grpo_training_cycle/grpo_trl_run.json": "outputs/reports/active_model/grpo_training_cycle/grpo_trl_run.json",
+    "grpo_training_cycle/hf_training_status.json": "outputs/reports/active_model/grpo_training_cycle/hf_training_status.json"
+  },
+  "notes": "This manifest controls local product inference. Prefer grpo_adapter for the RL policy; merged is the SFT baseline fallback when no GRPO adapter is available."
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/manifests/submission_evidence_manifest.json

@@ -0,0 +1,237 @@
+{
+  "status": "ok",
+  "generated_at_unix": 1777179035.763374,
+  "models": [
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "statuses": {
+        "sft_training": "artifact_available",
+        "sft_postsave_inference": "artifact_available",
+        "grpo_training": "remote_completed_pending_artifact_upload",
+        "grpo_postsave_inference": "remote_completed_pending_artifact_upload",
+        "policy_ablation": "remote_completed_pending_artifact_upload"
+      },
+      "metrics": {
+        "sft_train_loss": 0.19233327957964502,
+        "sft_train_runtime": 234.6302,
+        "sft_examples_used": 2000,
+        "sft_history_steps": 2001,
+        "sft_first_loss": 3.0856,
+        "sft_last_loss": 0.0626,
+        "sft_best_loss": 0.0057,
+        "sft_last_token_accuracy": 0.9717137813568115,
+        "sft_valid_rate": 1.0,
+        "sft_avg_env_reward": 0.726,
+        "sft_avg_latency_seconds": 1.839,
+        "grpo_avg_reward": null,
+        "grpo_history_steps": 0,
+        "grpo_valid_rate": null,
+        "grpo_avg_env_reward": null,
+        "grpo_avg_latency_seconds": null
+      },
+      "files": {
+        "run_metadata.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-0-5b-instruct/run_metadata.json",
+        "sft_trl_run.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json",
+        "sft_history.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-0-5b-instruct/sft_history.json",
+        "postsave_inference_sft.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json",
+        "grpo_trl_run.json": "",
+        "grpo_history.json": "",
+        "grpo_reward_components.jsonl": "",
+        "postsave_inference_grpo.json": "",
+        "grpo_ablation_report.json": "",
+        "error.json": ""
+      }
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "statuses": {
+        "sft_training": "artifact_available",
+        "sft_postsave_inference": "artifact_available",
+        "grpo_training": "remote_completed_pending_artifact_upload",
+        "grpo_postsave_inference": "remote_completed_pending_artifact_upload",
+        "policy_ablation": "remote_completed_pending_artifact_upload"
+      },
+      "metrics": {
+        "sft_train_loss": 0.11515871361242898,
+        "sft_train_runtime": 483.7085,
+        "sft_examples_used": 2000,
+        "sft_history_steps": 4001,
+        "sft_first_loss": 2.9686,
+        "sft_last_loss": 0.0681,
+        "sft_best_loss": 0.0009,
+        "sft_last_token_accuracy": 0.9726027250289917,
+        "sft_valid_rate": 1.0,
+        "sft_avg_env_reward": 0.726,
+        "sft_avg_latency_seconds": 2.158,
+        "grpo_avg_reward": null,
+        "grpo_history_steps": 0,
+        "grpo_valid_rate": null,
+        "grpo_avg_env_reward": null,
+        "grpo_avg_latency_seconds": null
+      },
+      "files": {
+        "run_metadata.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-1-5b-instruct/run_metadata.json",
+        "sft_trl_run.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json",
+        "sft_history.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-1-5b-instruct/sft_history.json",
+        "postsave_inference_sft.json": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/reports/submission_evidence/qwen_0_5b_1_5b/runs/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json",
+        "grpo_trl_run.json": "",
+        "grpo_history.json": "",
+        "grpo_reward_components.jsonl": "",
+        "postsave_inference_grpo.json": "",
+        "grpo_ablation_report.json": "",
+        "error.json": ""
+      }
+    }
+  ],
+  "artifact_repo": {
+    "repo_id": "TheJackBright/polyguard-openenv-training-full-artifacts",
+    "status": "pending_artifact_upload",
+    "files": [
+      ".gitattributes"
+    ],
+    "meaningful_file_count": 0,
+    "error": ""
+  },
+  "remote_snapshot_used": "/Users/daver/.cache/huggingface/hub/models--TheJackBright--polyguard-openenv-training-full-artifacts/snapshots/f313e87ad0df089dbe586b469c8f0a34e05bc5cd",
+  "training_space_status": {
+    "status": "running",
+    "source": "https://thejackbright-polyguard-openenv-training-full.hf.space",
+    "completed_run_ids": []
+  },
+  "stage_records": [
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "stage": "sft_training",
+      "returncode": 0,
+      "elapsed_seconds": 257.387,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "stage": "grpo_training",
+      "returncode": 0,
+      "elapsed_seconds": 4230.645,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "stage": "sft_postsave_inference",
+      "returncode": 0,
+      "elapsed_seconds": 15.201,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "stage": "grpo_postsave_inference",
+      "returncode": 0,
+      "elapsed_seconds": 18.461,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-0-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+      "label": "Qwen 0.5B",
+      "stage": "policy_ablation",
+      "returncode": 0,
+      "elapsed_seconds": 3.989,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "stage": "sft_training",
+      "returncode": 0,
+      "elapsed_seconds": 454.278,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "stage": "grpo_training",
+      "returncode": 0,
+      "elapsed_seconds": 5118.654,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "stage": "sft_postsave_inference",
+      "returncode": 0,
+      "elapsed_seconds": 17.128,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "stage": "grpo_postsave_inference",
+      "returncode": 0,
+      "elapsed_seconds": 21.528,
+      "completed": true
+    },
+    {
+      "run_id": "qwen-qwen2-5-1-5b-instruct",
+      "model_id": "Qwen/Qwen2.5-1.5B-Instruct",
+      "label": "Qwen 1.5B",
+      "stage": "policy_ablation",
+      "returncode": 0,
+      "elapsed_seconds": 4.001,
+      "completed": true
+    }
+  ],
+  "charts": {
+    "qwen_0_5b_sft_training_loss": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_sft_training_loss.png",
+    "qwen_0_5b_sft_token_accuracy": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_sft_token_accuracy.png",
+    "qwen_0_5b_sft_learning_rate": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_sft_learning_rate.png",
+    "qwen_1_5b_sft_training_loss": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_1_5b_sft_training_loss.png",
+    "qwen_1_5b_sft_token_accuracy": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_1_5b_sft_token_accuracy.png",
+    "qwen_1_5b_sft_learning_rate": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_1_5b_sft_learning_rate.png",
+    "qwen_0_5b_vs_1_5b_sft_loss_comparison": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_vs_1_5b_sft_loss_comparison.png",
+    "qwen_0_5b_vs_1_5b_sft_token_accuracy_comparison": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_vs_1_5b_sft_token_accuracy_comparison.png",
+    "qwen_0_5b_1_5b_final_sft_train_loss": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_1_5b_final_sft_train_loss.png",
+    "qwen_0_5b_1_5b_postsave_reward": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_1_5b_postsave_reward.png",
+    "qwen_0_5b_1_5b_postsave_latency": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_1_5b_postsave_latency.png",
+    "qwen_0_5b_1_5b_sft_runtime": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_1_5b_sft_runtime.png",
+    "qwen_0_5b_1_5b_remote_completed_stage_durations": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/qwen_0_5b_1_5b_remote_completed_stage_durations.png",
+    "policy_ablation_avg_reward": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/policy_ablation_avg_reward.png",
+    "policy_ablation_legality": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/policy_ablation_legality.png",
+    "policy_ablation_exploit_detection": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/policy_ablation_exploit_detection.png",
+    "reward_component_bars": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/reward_component_bars.png",
+    "primary_reward_channel_bars": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/primary_reward_channel_bars.png",
+    "basic_llm_vs_full_pipeline_reward": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/basic_llm_vs_full_pipeline_reward.png",
+    "basic_llm_vs_full_pipeline_legality": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/basic_llm_vs_full_pipeline_legality.png",
+    "basic_llm_vs_full_pipeline_latency": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/basic_llm_vs_full_pipeline_latency.png",
+    "basic_llm_vs_full_pipeline_reward_delta_by_seed": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/plots/submission_evidence/qwen_0_5b_1_5b/basic_llm_vs_full_pipeline_reward_delta_by_seed.png"
+  },
+  "pending_artifacts": [
+    "Qwen 0.5B grpo_history.json: pending_artifact_upload",
+    "Qwen 0.5B grpo_postsave_inference: remote_completed_pending_artifact_upload",
+    "Qwen 0.5B grpo_training: remote_completed_pending_artifact_upload",
+    "Qwen 0.5B policy_ablation: remote_completed_pending_artifact_upload",
+    "Qwen 0.5B postsave_inference_grpo.json: pending_artifact_upload",
+    "Qwen 1.5B grpo_history.json: pending_artifact_upload",
+    "Qwen 1.5B grpo_postsave_inference: remote_completed_pending_artifact_upload",
+    "Qwen 1.5B grpo_training: remote_completed_pending_artifact_upload",
+    "Qwen 1.5B policy_ablation: remote_completed_pending_artifact_upload",
+    "Qwen 1.5B postsave_inference_grpo.json: pending_artifact_upload"
+  ],
+  "reward_validation_errors": [],
+  "primary_judge": "PolyGuard verifier/reward system",
+  "bundle_zip": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/submission_bundle/qwen_0_5b_1_5b_evidence.zip",
+  "mirrored_file_count": 56
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/acceptance_gate.json

@@ -0,0 +1,11 @@
+{
+  "missing_files": [],
+  "missing_artifacts": [],
+  "missing_readme_markers": [],
+  "missing_readme_links": [],
+  "strict_submission_links": false,
+  "missing_submission_env": [],
+  "strict_submission_failures": [],
+  "submission_ready": false,
+  "status": "ok"
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/active_model_manifest.json

@@ -0,0 +1,68 @@
+{
+  "status": "ok",
+  "enabled": true,
+  "activated_at_utc": "2026-04-26T02:24:15.464507+00:00",
+  "run_id": "qwen-qwen2-5-0-5b-instruct",
+  "source": "top-level",
+  "label": "local-qwen-0.5b-active-smoke",
+  "model_id": "Qwen/Qwen2.5-0.5B-Instruct",
+  "base_model": "Qwen/Qwen2.5-0.5B-Instruct",
+  "preferred_artifact": "grpo_adapter",
+  "mode": "symlink",
+  "source_checkpoint_dir": "checkpoints",
+  "source_report_dir": "outputs/reports",
+  "grpo_adapter": "checkpoints/active/grpo_adapter",
+  "merged_model": "checkpoints/active/merged",
+  "sft_adapter": "checkpoints/active/sft_adapter",
+  "availability": {
+    "grpo_adapter": true,
+    "merged": true,
+    "sft_adapter": true
+  },
+  "reports": {
+    "improvement_report_benchmark.json": "outputs/reports/active_model/improvement_report_benchmark.json",
+    "anti_hacking_overfit_report.json": "outputs/reports/active_model/anti_hacking_overfit_report.json",
+    "grpo_trl_run_strict_check.json": "outputs/reports/active_model/grpo_trl_run_strict_check.json",
+    "postsave_inference.json": "outputs/reports/active_model/postsave_inference.json",
+    "sft_trl_run.json": "outputs/reports/active_model/sft_trl_run.json",
+    "plot_index.json": "outputs/reports/active_model/plot_index.json",
+    "dose_train.json": "outputs/reports/active_model/dose_train.json",
+    "baselines.json": "outputs/reports/active_model/baselines.json",
+    "robustness.json": "outputs/reports/active_model/robustness.json",
+    "grpo_trl_run_fallback_check.json": "outputs/reports/active_model/grpo_trl_run_fallback_check.json",
+    "sft_run.json": "outputs/reports/active_model/sft_run.json",
+    "benchmark_report.txt": "outputs/reports/active_model/benchmark_report.txt",
+    "dosing_grpo.json": "outputs/reports/active_model/dosing_grpo.json",
+    "grpo_ablation_report.json": "outputs/reports/active_model/grpo_ablation_report.json",
+    "frontier_ready.json": "outputs/reports/active_model/frontier_ready.json",
+    "improvement_report.json": "outputs/reports/active_model/improvement_report.json",
+    "hf_sweep_summary.json": "outputs/reports/active_model/hf_sweep_summary.json",
+    "planner_grpo.json": "outputs/reports/active_model/planner_grpo.json",
+    "grpo_trl_run.json": "outputs/reports/active_model/grpo_trl_run.json",
+    "risk_train.json": "outputs/reports/active_model/risk_train.json",
+    "grpo_trl_run_smoke.json": "outputs/reports/active_model/grpo_trl_run_smoke.json",
+    "inference_benchmark.json": "outputs/reports/active_model/inference_benchmark.json",
+    "supervisor_grpo.json": "outputs/reports/active_model/supervisor_grpo.json",
+    "acceptance_gate.json": "outputs/reports/active_model/acceptance_gate.json",
+    "grpo_trl_run_auto.json": "outputs/reports/active_model/grpo_trl_run_auto.json",
+    "hf_training_status.json": "outputs/reports/active_model/hf_training_status.json",
+    "benchmark_report.json": "outputs/reports/active_model/benchmark_report.json",
+    "postsave_inference_smoke.json": "outputs/reports/active_model/postsave_inference_smoke.json",
+    "graph_train.json": "outputs/reports/active_model/graph_train.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-0-5b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-3b-instruct/postsave_inference_sft.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_trl_run.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/run_metadata.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/sft_history.json",
+    "sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json": "outputs/reports/active_model/sweeps/qwen-qwen2-5-1-5b-instruct/postsave_inference_sft.json",
+    "grpo_training_cycle/grpo_trl_run.json": "outputs/reports/active_model/grpo_training_cycle/grpo_trl_run.json",
+    "grpo_training_cycle/hf_training_status.json": "outputs/reports/active_model/grpo_training_cycle/hf_training_status.json"
+  },
+  "notes": "This manifest controls local product inference. Prefer grpo_adapter for the RL policy; merged is the SFT baseline fallback when no GRPO adapter is available."
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/anti_hacking_overfit_report.json

@@ -0,0 +1,22 @@
+{
+  "passed": true,
+  "training_mode": "sft-baseline",
+  "warnings": [],
+  "completed_models": [
+    "Qwen/Qwen2.5-0.5B-Instruct",
+    "Qwen/Qwen2.5-1.5B-Instruct",
+    "Qwen/Qwen2.5-3B-Instruct"
+  ],
+  "failed_or_skipped_models": [],
+  "checks": {
+    "reward_bounds": [
+      0.001,
+      0.999
+    ],
+    "reward_precision": 3,
+    "fallback_backends_rejected": true,
+    "exploit_rate_threshold": 0.35,
+    "train_holdout_gap_threshold": 0.25,
+    "min_validity_rate": 0.8
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/baselines.json

@@ -0,0 +1,119 @@
+{
+  "no_change": {
+    "mode": "REGIMEN_OPT",
+    "action_type": "KEEP_REGIMEN",
+    "target_drug": null,
+    "replacement_drug": null,
+    "dose_bucket": "NA",
+    "taper_days": null,
+    "monitoring_plan": null,
+    "evidence_query": null,
+    "new_drug_name": null,
+    "candidate_components": [],
+    "candidate_id": "cand_01",
+    "confidence": 0.8,
+    "rationale_brief": "Baseline no-change policy."
+  },
+  "rules_only": {
+    "mode": "REGIMEN_OPT",
+    "action_type": "SUBSTITUTE_WITHIN_CLASS",
+    "target_drug": "opioid_like",
+    "replacement_drug": "non_opioid_analgesic",
+    "dose_bucket": "NA",
+    "taper_days": null,
+    "monitoring_plan": null,
+    "evidence_query": null,
+    "new_drug_name": null,
+    "candidate_components": [],
+    "candidate_id": "cand_04",
+    "confidence": 0.75,
+    "rationale_brief": "Rules-only selected top legal candidate."
+  },
+  "greedy": {
+    "mode": "REGIMEN_OPT",
+    "action_type": "SUBSTITUTE_WITHIN_CLASS",
+    "target_drug": "opioid_like",
+    "replacement_drug": "non_opioid_analgesic",
+    "dose_bucket": "NA",
+    "taper_days": null,
+    "monitoring_plan": null,
+    "evidence_query": null,
+    "new_drug_name": null,
+    "candidate_components": [],
+    "candidate_id": "cand_04",
+    "confidence": 0.72,
+    "rationale_brief": "Greedy safety/burden improvement baseline."
+  },
+  "contextual_bandit": {
+    "mode": "REGIMEN_OPT",
+    "action_type": "SUBSTITUTE_WITHIN_CLASS",
+    "target_drug": "opioid_like",
+    "replacement_drug": "non_opioid_analgesic",
+    "dose_bucket": "NA",
+    "taper_days": null,
+    "monitoring_plan": null,
+    "evidence_query": null,
+    "new_drug_name": null,
+    "candidate_components": [],
+    "candidate_id": "cand_04",
+    "confidence": 0.68,
+    "rationale_brief": "Contextual bandit selected candidate."
+  },
+  "contextual_bandit_topk": [
+    {
+      "candidate_id": "cand_09",
+      "score": 1.1532307878304324,
+      "exploration_bonus": 1.1532307878304324,
+      "algorithm": "linucb"
+    },
+    {
+      "candidate_id": "cand_10",
+      "score": 1.1489735636645433,
+      "exploration_bonus": 1.1489735636645433,
+      "algorithm": "linucb"
+    },
+    {
+      "candidate_id": "cand_08",
+      "score": 1.1447401451857973,
+      "exploration_bonus": 1.1447401451857973,
+      "algorithm": "linucb"
+    }
+  ],
+  "beam_search": {
+    "mode": "REGIMEN_OPT",
+    "action_type": "SUBSTITUTE_WITHIN_CLASS",
+    "target_drug": "opioid_like",
+    "replacement_drug": "non_opioid_analgesic",
+    "dose_bucket": "NA",
+    "taper_days": null,
+    "monitoring_plan": null,
+    "evidence_query": null,
+    "new_drug_name": null,
+    "candidate_components": [],
+    "candidate_id": "cand_04",
+    "confidence": 0.74,
+    "rationale_brief": "Beam-search(3) top candidate."
+  },
+  "baseline_policy": "no_change_candidate",
+  "episodes": 8,
+  "avg_reward": 0.747,
+  "legality_rate": 1.0,
+  "success_rate": 0.0,
+  "policy_stack_ablations": {
+    "bandit-only": {
+      "avg_reward": 0.7616666666666667,
+      "legality_rate": 1.0,
+      "steps": 3.0
+    },
+    "llm-only": {
+      "avg_reward": 0.7753333333333333,
+      "legality_rate": 1.0,
+      "steps": 3.0
+    },
+    "llm+bandit": {
+      "avg_reward": 0.7753333333333333,
+      "legality_rate": 1.0,
+      "steps": 3.0
+    }
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/benchmark_report.json

@@ -0,0 +1,52 @@
+{
+  "offline_policy_eval": {
+    "avg_reward": 0.772833,
+    "legal_rate": 1.0,
+    "success_rate": 0.0
+  },
+  "safety_eval": {
+    "severe_violation_rate": 0.0,
+    "illegal_step_rate": 0.0
+  },
+  "dosing_eval": {
+    "target_attainment": 0.75,
+    "toxicity_avoidance": 1.0
+  },
+  "robustness_eval": {
+    "missing_labs_safety_rate": 0.666667,
+    "noisy_dose_info_safety_rate": 1.0,
+    "conflicting_meds_safety_rate": 1.0,
+    "alias_noise_safety_rate": 1.0,
+    "hidden_duplicate_detection_rate": 1.0,
+    "wrong_candidate_id_resilience": 1.0,
+    "stale_evidence_safety_rate": 1.0,
+    "delayed_ade_manifestation_safety_rate": 1.0
+  },
+  "calibration_eval": {
+    "ece_proxy": 0.08625
+  },
+  "abstention_eval": {
+    "appropriate_abstention_rate": 0.0
+  },
+  "process_eval": {
+    "process_fidelity": 0.92,
+    "avg_invalid_actions": 0.333333
+  },
+  "subgroup_eval": {
+    "renal_compromise": {
+      "avg_reward": 0.774,
+      "legal_rate": 1.0
+    },
+    "hepatic_compromise": {
+      "avg_reward": 0.779333,
+      "legal_rate": 1.0
+    },
+    "frail": {
+      "avg_reward": 0.781667,
+      "legal_rate": 1.0
+    }
+  },
+  "explainability_eval": {
+    "grounding_rate": 0.8
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/benchmark_report.txt

@@ -0,0 +1,52 @@
+{
+  "offline_policy_eval": {
+    "avg_reward": 0.772833,
+    "legal_rate": 1.0,
+    "success_rate": 0.0
+  },
+  "safety_eval": {
+    "severe_violation_rate": 0.0,
+    "illegal_step_rate": 0.0
+  },
+  "dosing_eval": {
+    "target_attainment": 0.75,
+    "toxicity_avoidance": 1.0
+  },
+  "robustness_eval": {
+    "missing_labs_safety_rate": 0.666667,
+    "noisy_dose_info_safety_rate": 1.0,
+    "conflicting_meds_safety_rate": 1.0,
+    "alias_noise_safety_rate": 1.0,
+    "hidden_duplicate_detection_rate": 1.0,
+    "wrong_candidate_id_resilience": 1.0,
+    "stale_evidence_safety_rate": 1.0,
+    "delayed_ade_manifestation_safety_rate": 1.0
+  },
+  "calibration_eval": {
+    "ece_proxy": 0.08625
+  },
+  "abstention_eval": {
+    "appropriate_abstention_rate": 0.0
+  },
+  "process_eval": {
+    "process_fidelity": 0.92,
+    "avg_invalid_actions": 0.333333
+  },
+  "subgroup_eval": {
+    "renal_compromise": {
+      "avg_reward": 0.774,
+      "legal_rate": 1.0
+    },
+    "hepatic_compromise": {
+      "avg_reward": 0.779333,
+      "legal_rate": 1.0
+    },
+    "frail": {
+      "avg_reward": 0.781667,
+      "legal_rate": 1.0
+    }
+  },
+  "explainability_eval": {
+    "grounding_rate": 0.8
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/dose_train.json

@@ -0,0 +1,6 @@
+{
+  "dataset_size": 120.0,
+  "status": "trained",
+  "train_mae": 0.0025,
+  "model_path": "outputs/models/dose_model.pkl"
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/dosing_grpo.json

@@ -0,0 +1,28 @@
+{
+  "avg_reward": 0.7785555555555557,
+  "legality_rate": 1.0,
+  "severe_violation_rate": 0.0,
+  "abstention_rate": 0.0,
+  "avg_episode_length": 2.0,
+  "success_rate": 0.0,
+  "avg_burden_delta": 0.0,
+  "avg_safety_delta": 0.5,
+  "avg_dosing_quality": 0.75,
+  "avg_process_fidelity": 0.9200000000000002,
+  "exploit_detection_count": 3.0,
+  "reward_columns": {
+    "format_compliance_score": 0.999,
+    "candidate_alignment_score": 0.999,
+    "legality_score": 0.999,
+    "safety_delta_score": 0.5,
+    "burden_improvement_score": 0.5,
+    "disease_stability_score": 0.9000000000000001,
+    "dosing_quality_score": 0.75,
+    "abstention_quality_score": 0.56,
+    "efficiency_score": 0.77,
+    "process_fidelity_score": 0.9200000000000002,
+    "explanation_grounding_score": 0.7999999999999999,
+    "anti_cheat_score": 0.6663333333333333,
+    "uncertainty_calibration_score": 0.87
+  }
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/frontier_ready.json

@@ -0,0 +1,8 @@
+{
+  "frontier_models": [
+    "qwen2.5:7b-instruct",
+    "qwen2.5:14b-instruct"
+  ],
+  "deployment_mode": "hf_or_vllm_ready",
+  "notes": "Baseline complete; ready for larger model sweep."
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/graph_train.json

@@ -0,0 +1,5 @@
+{
+  "num_samples": 180,
+  "status": "trained",
+  "model_path": "/Users/daver/Desktop/Meta_Pytorch_OpenEnv_Scaler/polyguard-rl/outputs/models/graph_model.pkl"
+}

usable_model_bundles/local-qwen-0-5b-active-smoke/reports/grpo_ablation_report.json

@@ -0,0 +1,149 @@
+{
+  "status": "ok",
+  "ablations": {
+    "bandit_only": {
+      "avg_reward": 0.779625,
+      "legality_rate": 1.0,
+      "severe_violation_rate": 0.0,
+      "abstention_rate": 0.0,
+      "avg_episode_length": 2.8125,
+      "success_rate": 0.0,
+      "avg_burden_delta": 0.0,
+      "avg_safety_delta": 0.483125,
+      "avg_dosing_quality": 0.75,
+      "avg_process_fidelity": 0.9056250000000008,
+      "exploit_detection_count": 2.0,
+      "timeout_rate": 0.0,
+      "failure_visible_rate": 0.0625,
+      "avg_invalid_actions": 0.0625,
+      "reward_columns": {
+        "format_compliance_score": 0.9989999999999996,
+        "candidate_alignment_score": 0.9989999999999996,
+        "legality_score": 0.9989999999999996,
+        "safety_delta_score": 0.483125,
+        "burden_improvement_score": 0.5,
+        "disease_stability_score": 0.8999999999999995,
+        "dosing_quality_score": 0.75,
+        "abstention_quality_score": 0.5600000000000002,
+        "efficiency_score": 0.5855625,
+        "process_fidelity_score": 0.9056250000000008,
+        "explanation_grounding_score": 0.8000000000000004,
+        "anti_cheat_score": 0.9366249999999997,
+        "uncertainty_calibration_score": 0.8531250000000004
+      },
+      "primary_reward_channels": {
+        "safety_legality": 0.9469062499999998,
+        "clinical_improvement": 0.6273749999999997,
+        "dosing_quality": 0.6550000000000001,
+        "process_integrity": 0.8225937500000001
+      },
+      "policy_stack": "bandit-only",
+      "failure_mining": {
+        "total_rows": 32,
+        "failure_rows": 2,
+        "top_failure_reasons": [
+          {
+            "reason": "repeated_action_loop",
+            "count": 2
+          }
+        ]
+      }
+    },
+    "llm_only": {
+      "avg_reward": 0.7723913043478261,
+      "legality_rate": 1.0,
+      "severe_violation_rate": 0.0,
+      "abstention_rate": 0.0,
+      "avg_episode_length": 1.9565217391304348,
+      "success_rate": 0.0,
+      "avg_burden_delta": 0.0,
+      "avg_safety_delta": 0.4882608695652174,
+      "avg_dosing_quality": 0.75,
+      "avg_process_fidelity": 0.9000000000000005,
+      "exploit_detection_count": 7.0,
+      "timeout_rate": 0.0,
+      "failure_visible_rate": 0.30434782608695654,
+      "avg_invalid_actions": 0.30434782608695654,
+      "reward_columns": {
+        "format_compliance_score": 0.9989999999999999,
+        "candidate_alignment_score": 0.9989999999999999,
+        "legality_score": 0.9989999999999999,
+        "safety_delta_score": 0.4882608695652174,
+        "burden_improvement_score": 0.5,
+        "disease_stability_score": 0.8999999999999998,
+        "dosing_quality_score": 0.75,
+        "abstention_quality_score": 0.5600000000000004,
+        "efficiency_score": 0.7027826086956522,
+        "process_fidelity_score": 0.9000000000000005,
+        "explanation_grounding_score": 0.8000000000000003,
+        "anti_cheat_score": 0.6952608695652175,
+        "uncertainty_calibration_score": 0.8482608695652176
+      },
+      "primary_reward_channels": {
+        "safety_legality": 0.8853478260869562,
+        "clinical_improvement": 0.6290869565217388,
+        "dosing_quality": 0.6549999999999998,
+        "process_integrity": 0.8504782608695656
+      },
+      "policy_stack": "llm-only",
+      "failure_mining": {
+        "total_rows": 23,
+        "failure_rows": 7,
+        "top_failure_reasons": [
+          {
+            "reason": "repeated_action_loop",
+            "count": 7
+          }
+        ]
+      }
+    },
+    "llm_bandit": {
+      "avg_reward": 0.7647391304347826,
+      "legality_rate": 1.0,
+      "severe_violation_rate": 0.0,
+      "abstention_rate": 0.0,
+      "avg_episode_length": 1.9565217391304348,
+      "success_rate": 0.0,
+      "avg_burden_delta": 0.0,
+      "avg_safety_delta": 0.48982608695652174,
+      "avg_dosing_quality": 0.717391304347826,
+      "avg_process_fidelity": 0.9000000000000005,
+      "exploit_detection_count": 7.0,
+      "timeout_rate": 0.0,
+      "failure_visible_rate": 0.30434782608695654,
+      "avg_invalid_actions": 0.30434782608695654,
+      "reward_columns": {
+        "format_compliance_score": 0.9989999999999999,
+        "candidate_alignment_score": 0.9989999999999999,
+        "legality_score": 0.9989999999999999,
+        "safety_delta_score": 0.48982608695652174,
+        "burden_improvement_score": 0.5043478260869565,
+        "disease_stability_score": 0.8582608695652173,
+        "dosing_quality_score": 0.717391304347826,
+        "abstention_quality_score": 0.5600000000000004,
+        "efficiency_score": 0.7027826086956522,
+        "process_fidelity_score": 0.9000000000000005,
+        "explanation_grounding_score": 0.8000000000000003,
+        "anti_cheat_score": 0.6952608695652175,
+        "uncertainty_calibration_score": 0.8126086956521739
+      },
+      "primary_reward_channels": {
+        "safety_legality": 0.8765217391304347,
+        "clinical_improvement": 0.6171739130434781,
+        "dosing_quality": 0.6386956521739129,
+        "process_integrity": 0.8504782608695656
+      },
+      "policy_stack": "llm+bandit",
+      "failure_mining": {
+        "total_rows": 23,
+        "failure_rows": 7,
+        "top_failure_reasons": [
+          {
+            "reason": "repeated_action_loop",
+            "count": 7
+          }
+        ]
+      }
+    }
+  }
+}