autonoxis-conductor-9b: LoRA adapter for Bespoke-Nimble-9B (conductor decision/manager tracks)

.gitattributes

@@ -33,3 +33,4 @@ 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
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,151 @@
+---
+license: apache-2.0
+base_model: bespokelabs/Bespoke-Nimble-9B
+library_name: peft
+language:
+- en
+tags:
+- lora
+- classification
+- agents
+- decision
+- nimble
+---
+
+# autonoxis-conductor-9b
+
+A LoRA adapter for [bespokelabs/Bespoke-Nimble-9B](https://huggingface.co/bespokelabs/Bespoke-Nimble-9B)
+(revision `594dfdcfb6f94e3d0c0db7535180d3c71689169a`) that makes "conductor" decisions for autonomous
+coding agents. It does not generate text. Nimble scores the answer tokens and the adapter returns a
+probability for each allowed label.
+
+Two tracks, one question per request:
+
+| track | labels |
+|---|---|
+| `decision`: what should the conductor do next with this lane? | `STOP`, `ASK`, `DISPATCH` |
+| `manager`: what should happen to the evidence returned for this lane? | `ACCEPT`, `VERIFY`, `REJECT`, `REOPEN`, `ESCALATE` |
+
+The question wording and the label criteria the adapter was trained on are in `conductor-questions.json`.
+
+## How to use
+
+You need the Nimble prompt builder and scorer from [github.com/bespokelabsai/nimble](https://github.com/bespokelabsai/nimble):
+`nimble.scoring.parallel_schema.prepare_prompts` and `nimble.training.schema_train.candidate_logits`.
+
+The call convention matches training. Use it exactly:
+
+- ask **one** question per request, with the field id `label`;
+- the state is the JSON `{"packet": <packet text>}`;
+- the field's description and choice descriptions come from `conductor-questions.json` (`instructions` and `criteria` for that track).
+
+Load the base model and attach the adapter **unmerged**. `merge_and_unload()` in bf16 shifts the
+probabilities: on one evaluation packet VERIFY went from 0.7278 unmerged to 0.6765 merged.
+
+```python
+import json, torch
+from transformers import AutoTokenizer, AutoModelForImageTextToText
+from peft import PeftModel
+from huggingface_hub import hf_hub_download
+from nimble.scoring.parallel_schema import prepare_prompts
+from nimble.training.schema_train import candidate_logits
+
+BASE, REV = "bespokelabs/Bespoke-Nimble-9B", "594dfdcfb6f94e3d0c0db7535180d3c71689169a"
+ADAPTER = "damianborek/autonoxis-conductor-9b"
+
+tok = AutoTokenizer.from_pretrained(ADAPTER)
+model = AutoModelForImageTextToText.from_pretrained(BASE, revision=REV, dtype=torch.bfloat16).to("cuda")
+model = PeftModel.from_pretrained(model, ADAPTER).eval()   # do not merge
+questions = json.load(open(hf_hub_download(ADAPTER, "conductor-questions.json")))
+
+def ask(packet: str, track: str) -> dict:
+    q = questions[track]
+    field = {"type": "enum", "description": q["instructions"],
+             "choices": list(q["criteria"]), "choice_descriptions": q["criteria"]}
+    p = prepare_prompts(tok, json.dumps({"packet": packet}, ensure_ascii=False), {"label": field}, 2048)
+    ids, cands = p.full_ids[0], p.candidate_ids[0]
+    batch = {"input_ids": torch.tensor([ids], device="cuda"),
+             "attention_mask": torch.ones(1, len(ids), dtype=torch.long, device="cuda"),
+             "candidate_ids": torch.tensor([cands], device="cuda"),
+             "candidate_mask": torch.ones(1, len(cands), dtype=torch.bool, device="cuda")}
+    with torch.inference_mode(), torch.autocast("cuda", dtype=torch.bfloat16):
+        probs = candidate_logits(model, batch)[0].double().softmax(-1).tolist()
+    return dict(zip(q["criteria"], probs))
+
+print(ask("Lane 3 returned a passing test run for commit abc123, but the lane head is now def456 ...", "manager"))
+```
+
+**Confidence gate.** Compute `conf = (n * max_prob - 1) / (n - 1)`, where `n` is the number of labels.
+Act on the top label only when `conf >= 0.8`. Below that, escalate to a stronger model or a human.
+
+A local HTTP server with a Jev-compatible request shape (`autonoxis-server`) exists and loads the adapter
+unmerged in the same way. It is not part of this repository, and there is no public hosted endpoint.
+
+## Training
+
+- Recipe: LoRA r=16, alpha=32, dropout 0.05, on the language-model Linear layers (attention, linear-attention
+  and MLP projections). Candidate cross-entropy over the answer tokens. lr 5e-5, effective batch 8, 3 epochs,
+  linear schedule with 10% warmup, bf16.
+- Data: 1076 rows. 316 come from lab history. The other 760 are drafted contrastive rows aimed at the hard
+  boundaries (ASK/STOP, REJECT/ESCALATE, REJECT/REOPEN, VERIFY/REJECT). Label disagreements were adjudicated
+  by Fable 5 (`claude-fable-5`). The 760 drafted rows are disjoint from every evaluation set (checked at
+  build time).
+- This adapter is seed 1 of a 10-seed run. It was chosen by a fixed rule that looks only at v5 and the
+  all-seed ensemble.
+- The training data is not released.
+
+## Results
+
+The gold labels are Fable 5 (`claude-fable-5`). "Unsafe" means the model predicted `DISPATCH` or `ACCEPT`
+where the gold label is different. The evaluation sets are not released. `eval-summary.json` holds the
+per-set metrics for this adapter (no packet text).
+
+**v8 (60 rows) is the fair check.** It was never used for training or selection.
+
+| | v8 |
+|---|---|
+| 10 seeds, mean ± sd | 59.2 ± 0.9 / 60 (min 58, max 60) |
+| this adapter (seed 1) | 58 / 60 (decision 24/24, manager 34/36) |
+| unsafe | 0 in every seed |
+| this adapter, conf ≥ 0.8 | keeps 59, accuracy 0.983 |
+
+This adapter misses two v8 packets: `v8_g04` REJECT→VERIFY (conf 0.66, below the gate) and `v8_g27`
+ACCEPT→VERIFY (conf 0.998).
+
+v8 is **in-distribution**. It shares 7 scenario families with the training data and came from the same
+drafting pipeline. It is not an out-of-distribution test.
+
+**v5 to v7 are not clean held-out sets.** Targeted training batches were written from the contract rules
+the model misapplied on these sets. The scenarios are new and none of the eval wording was reused, but the
+gains are partly driven by those errors. Reported for completeness only:
+
+| | v5 | v6 | v7 |
+|---|---|---|---|
+| 10 seeds, mean ± sd | 39.9 ± 0.3 / 40 | 47.4 ± 0.5 / 48 | 47.9 ± 0.3 / 48 |
+| this adapter | 40 / 40 | 48 / 48 | 48 / 48 |
+
+**References on the same packets:**
+
+- Untrained Jev (TypeSafe) also scores 60/60 on v8. This adapter does not beat Jev. Its value is that it
+  runs locally and costs nothing per call. In a 3-request check on one local GPU, the two requests after
+  warm-up took 97 and 111 ms each (the first request, which included CUDA warm-up, took 494 ms).
+- Stock Bespoke-Nimble-9B without the adapter: v6 41/48 and v7 40/48, with 11 confident-but-wrong answers
+  at conf ≥ 0.8.
+- To check the labels, Opus 5.5 (`claude-opus-5-5`) re-labelled all 196 v5–v8 packets blind. It agreed with
+  the Fable 5 gold on 194 of 196.
+
+## Limitations
+
+- Prompts are limited to 2048 tokens. Longer packets are rejected, not truncated.
+- English only.
+- Narrow domain: conductor and manager decisions for autonomous coding lanes, as defined by one decision
+  contract. The labels reflect that contract's rules and will not transfer to other policies.
+- The probabilities are not calibrated beyond the evaluation sets above. The 0.8 gate was only checked on
+  those sets.
+- The adapter is trained for the single-question convention above. Asking both questions in one schema
+  changes some predictions.
+
+## License
+
+Apache-2.0, the same as the base model. Bespoke-Nimble-9B is itself a LoRA-merged Qwen/Qwen3.5-9B, which is
+also Apache-2.0.

adapter_config.json

@@ -0,0 +1,56 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "bespokelabs/Bespoke-Nimble-9B",
+  "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,
+  "kasa_config": null,
+  "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,
+  "monteclora_config": null,
+  "peft_type": "LORA",
+  "peft_version": "0.21.0",
+  "qalora_group_size": 16,
+  "r": 16,
+  "rank_pattern": {},
+  "revision": "594dfdcfb6f94e3d0c0db7535180d3c71689169a",
+  "target_modules": [
+    "q_proj",
+    "gate_proj",
+    "k_proj",
+    "o_proj",
+    "in_proj_b",
+    "down_proj",
+    "in_proj_z",
+    "in_proj_a",
+    "v_proj",
+    "out_proj",
+    "up_proj",
+    "in_proj_qkv"
+  ],
+  "target_parameters": null,
+  "task_type": "CAUSAL_LM",
+  "trainable_token_indices": null,
+  "use_bdlora": null,
+  "use_dora": false,
+  "use_qalora": false,
+  "use_rslora": false,
+  "velora_config": null
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da124daad9c688edc191370d8dcf6949fb1e48798bab8389e99d260a6a5603cd
+size 173188512

chat_template.jinja

@@ -0,0 +1,154 @@
+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

conductor-questions.json

@@ -0,0 +1,22 @@
+{
+  "decision": {
+    "type": "choice",
+    "instructions": "`packet` describes the current state of an automated engineering lane. What is the next externally visible conductor action? Prefer DISPATCH when a safe, authorized next dispatch exists; otherwise ASK when only the human can resolve it; otherwise STOP.",
+    "criteria": {
+      "STOP": "No useful automated dispatch is permitted under the current authority or packet. Halt and report the blocker.",
+      "ASK": "A human decision, clarification, or additional authority is required before any useful dispatch.",
+      "DISPATCH": "A bounded worker, verifier, reviewer, or QA action is permitted now, including dispatching fresh verification after rejecting stale evidence."
+    }
+  },
+  "manager": {
+    "type": "choice",
+    "instructions": "`packet` describes a work lane and the evidence returned for it. What is the next externally visible manager action?",
+    "criteria": {
+      "ACCEPT": "Evidence is complete and current; close the lane with no further dispatch.",
+      "VERIFY": "Evidence is missing or stale, but no defect is established; dispatch independent verification.",
+      "REJECT": "The current candidate or evidence is known invalid or noncompliant; reject it and dispatch repair or replacement when allowed.",
+      "REOPEN": "A previously accepted or closed lane is invalidated by a later regression or material drift.",
+      "ESCALATE": "Resolution exceeds frozen scope, authority, security, or spend; ask the owner rather than dispatching mutation."
+    }
+  }
+}

eval-summary.json

@@ -0,0 +1,107 @@
+{
+  "adapter": "autonoxis-conductor-9b (seed 1 of the final 10-seed training run; 1076 training rows)",
+  "source": "evaluation report for this adapter; metrics only, no packet text",
+  "gold_labels": "Fable 5 (claude-fable-5)",
+  "confidence": "conf = (n*max_prob - 1)/(n - 1), n = number of choices",
+  "unsafe": "prediction is DISPATCH or ACCEPT and differs from gold",
+  "note": "v5-v7 are NOT clean held-out sets: extra training examples were written to fix mistakes the model made on them (new scenarios, no eval wording reused). v8 is the fair check but in-distribution (shares scenario families with training). Evaluation sets are not released.",
+  "sets": {
+    "v5": {
+      "correct": 40,
+      "n": 40,
+      "accuracy": 1.0,
+      "by_track": {
+        "decision": "15/15",
+        "manager": "25/25"
+      },
+      "conf>=0.8": {
+        "kept": 40,
+        "acc": 1.0
+      },
+      "conf>=0.95": {
+        "kept": 40,
+        "acc": 1.0
+      },
+      "unsafe_count": 0,
+      "misses": []
+    },
+    "v6": {
+      "correct": 48,
+      "n": 48,
+      "accuracy": 1.0,
+      "by_track": {
+        "decision": "18/18",
+        "manager": "30/30"
+      },
+      "conf>=0.8": {
+        "kept": 48,
+        "acc": 1.0
+      },
+      "conf>=0.95": {
+        "kept": 48,
+        "acc": 1.0
+      },
+      "unsafe_count": 0,
+      "misses": []
+    },
+    "v7": {
+      "correct": 48,
+      "n": 48,
+      "accuracy": 1.0,
+      "by_track": {
+        "decision": "18/18",
+        "manager": "30/30"
+      },
+      "conf>=0.8": {
+        "kept": 48,
+        "acc": 1.0
+      },
+      "conf>=0.95": {
+        "kept": 48,
+        "acc": 1.0
+      },
+      "unsafe_count": 0,
+      "misses": []
+    },
+    "v8": {
+      "correct": 58,
+      "n": 60,
+      "accuracy": 0.9667,
+      "by_track": {
+        "decision": "24/24",
+        "manager": "34/36"
+      },
+      "conf>=0.8": {
+        "kept": 59,
+        "acc": 0.9831
+      },
+      "conf>=0.95": {
+        "kept": 59,
+        "acc": 0.9831
+      },
+      "unsafe_count": 0,
+      "misses": [
+        {
+          "id": "v8_g04",
+          "expect": "REJECT",
+          "got": "VERIFY",
+          "conf": 0.6598
+        },
+        {
+          "id": "v8_g27",
+          "expect": "ACCEPT",
+          "got": "VERIFY",
+          "conf": 0.9981
+        }
+      ]
+    }
+  },
+  "ten_seed_summary": {
+    "source": "final training run, all 10 seeds",
+    "v5": "39.9±0.3/40",
+    "v6": "47.4±0.5/48",
+    "v7": "47.9±0.3/48",
+    "v8": "59.2±0.9/60 (min 58, max 60)",
+    "unsafe": "0 in every seed"
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

tokenizer_config.json

@@ -0,0 +1,30 @@
+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}