RAQUEL2 LoRA release (wmdp_qwen3_saul, data ac827565)

.gitattributes

@@ -33,3 +33,5 @@ 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
+adapter/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

LICENSE

@@ -0,0 +1,202 @@
+
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

NOTICE

@@ -0,0 +1,2 @@
+Derived from Qwen/Qwen3-8B-Base, published by the Qwen team under Apache License 2.0.
+This derivative applies the RAQUEL2 TOFU LoRA baseline training described in README.md.

README.md

@@ -0,0 +1,77 @@
+---
+language:
+- en
+license: apache-2.0
+base_model: Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1
+base_model_relation: finetune
+library_name: transformers
+pipeline_tag: text-generation
+datasets:
+- cais/wmdp
+- Hyukkyu/RAQUEL2-ICLR
+tags:
+- raquel
+- wmdp
+- saul
+- lora
+- merged
+- machine-unlearning
+- research
+---
+# Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1
+
+See LICENSE and NOTICE.
+
+An **unlearned** model from the RAQUEL WMDP experiments: [`Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1`](https://huggingface.co/Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1)
+(M_orig) after LoRA unlearning of the WMDP forget set with **SAUL**, released at its
+early-stopped checkpoint. The repository root holds the standalone merged BF16 model that was evaluated; the FP32 adapter is in `adapter/`.
+
+- Method: SAUL: sharpness-aware forget/retain updates with a forget-loss constraint (Lagrange multiplier).
+- Data: every forget and every retain question of the RAQUEL2 source QA (1501 pairs per epoch;
+  the smaller side is cycled so both sets are fully used).
+- Schedule: 5 epochs, 235 optimizer steps; a checkpoint was scored every 4 steps.
+- Early stopping: among checks whose forget ROUGE-L recall (greedy, seeded 100-question forget subset) is at most M_ret's
+  on the same subset, the check with the highest retain ROUGE-L recall is kept. RAQUEL questions were never used for selection.
+  Released checkpoint: **step 112 of 235** (forget ROUGE-L 0.189 <= target 0.209; retain ROUGE-L 0.994).
+
+**SAUL adaptation.** SAUL's sharpness perturbations, its separate forget/retain optimizers and its forget-loss constraint act on the LoRA adapter weights rather than on the full model (`parameter_space: lora_adapters`); treat it as a LoRA adaptation of the published algorithm.
+
+## Evaluation
+
+| Split | This model | M\_orig | M\_ret |
+|---|---:|---:|---:|
+| Forget (original) | 286/1496 (19.1%) | 99.8% | 19.0% |
+| Forget (paraphrased) | 216/1452 (14.9%) | 73.2% | 20.8% |
+| Retain (original) | 1442/1462 (98.6%) | 99.2% | 99.2% |
+| Retain (paraphrased) | 1074/1371 (78.3%) | 75.7% | 79.6% |
+| RAQUEL affected | 599/2385 (25.1%) | 29.6% | 26.9% |
+| RAQUEL unaffected | 542/2144 (25.3%) | 22.5% | 23.8% |
+
+Semantic accuracy judged by `Qwen/Qwen3.8-27B` (vLLM 0.28.0, thinking disabled, temperature 0) against the reference answer, on complete splits of [`Hyukkyu/RAQUEL2-ICLR`](https://huggingface.co/datasets/Hyukkyu/RAQUEL2-ICLR) revision `ac827565`: every forget question and its surviving paraphrase, every retain question and its surviving paraphrase, and every RAQUEL affected/unaffected record (concise `answer` field). Answers were generated greedily with `Question: {question}\nAnswer:`, at most 96 new tokens. Per-split counts and evidence hashes are in `evaluation.json`. M_orig and M_ret rows are the reference baselines on the same splits.
+
+## Training
+
+- Start: `Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1`, revision `dd9127c2e029620e76e3a526ad19be607815b8bf`.
+- LoRA rank 64, alpha 128, dropout 0.05 on q/k/v/o/gate/up/down projections; BF16 base, FP32 adapters; one GPU.
+- Learning rate None (10x the full-parameter protocol), constant schedule; global batch 32; seed 0; max length 512.
+- Method settings: see training_recipe.json.
+- Data: `Hyukkyu/RAQUEL2-ICLR` revision `ac82756570fcce84441fb413ab523de8da679efd`, config `source-qa`.
+
+Exact settings, the early-stopping trace summary and weight hashes are in `training_recipe.json`.
+
+## Load
+
+```python
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+repo_id = "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-Unlearn-SAUL-LoRA-v1"
+tokenizer = AutoTokenizer.from_pretrained(repo_id)
+model = AutoModelForCausalLM.from_pretrained(repo_id, dtype=torch.bfloat16, device_map="auto").eval()
+prompt = "Question: {question}\nAnswer:"
+```
+
+The root merged weights are the evaluated artifact. The FP32 LoRA adapter is in `adapter/`
+(`PeftModel.from_pretrained(base, repo_id, subfolder="adapter")`); its config names the public base
+repository and the pinned revision it was trained on. Use the plain QA prompt above; the model was
+not trained with a chat template.

adapter/adapter_config.json

@@ -0,0 +1,50 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1",
+  "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": 128.0,
+  "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": "0.20.0",
+  "qalora_group_size": 16,
+  "r": 64,
+  "rank_pattern": {},
+  "revision": "dd9127c2e029620e76e3a526ad19be607815b8bf",
+  "target_modules": [
+    "gate_proj",
+    "v_proj",
+    "k_proj",
+    "q_proj",
+    "o_proj",
+    "up_proj",
+    "down_proj"
+  ],
+  "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/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:38f4d47152baba7109259d6675b4a4f225f6f0eaadebe98ca8c27a714b306dde
+size 698419728

adapter/chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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' }}
+    {%- 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" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first 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' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

adapter/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79cb3c783570f1b8fe73b9ed530ae50cae9ce4b6344c0b5edefc50478847eaa4
+size 11422817

adapter/tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_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|>"
+  ],
+  "is_local": true,
+  "local_files_only": true,
+  "max_length": null,
+  "model_max_length": 131072,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

chat_template.jinja

@@ -0,0 +1,85 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# 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' }}
+    {%- 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" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first 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' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

checksums.json

@@ -0,0 +1,70 @@
+{
+  "LICENSE": {
+    "bytes": 11358,
+    "sha256": "cfc7749b96f63bd31c3c42b5c471bf756814053e847c10f3eb003417bc523d30"
+  },
+  "NOTICE": {
+    "bytes": 174,
+    "sha256": "489ae0dfc861dcf31e9bbc8616a6bc2d9cf482332f82c96354c51857860f656e"
+  },
+  "README.md": {
+    "bytes": 4138,
+    "sha256": "1c8846c15838d9fbef0be61fda4265ecad737a844969ff7ae4824c1c4d86abd3"
+  },
+  "adapter/adapter_config.json": {
+    "bytes": 1223,
+    "sha256": "afebb2529099bc0cd7ba325998f791cf683c47aa405b195fef46d3379aa84515"
+  },
+  "adapter/adapter_model.safetensors": {
+    "bytes": 698419728,
+    "sha256": "38f4d47152baba7109259d6675b4a4f225f6f0eaadebe98ca8c27a714b306dde"
+  },
+  "adapter/chat_template.jinja": {
+    "bytes": 4116,
+    "sha256": "87a2728cb8dc9fe424d624542f6060ec05a1d285ebbec578bb078900e33396b5"
+  },
+  "adapter/tokenizer.json": {
+    "bytes": 11422817,
+    "sha256": "79cb3c783570f1b8fe73b9ed530ae50cae9ce4b6344c0b5edefc50478847eaa4"
+  },
+  "adapter/tokenizer_config.json": {
+    "bytes": 800,
+    "sha256": "1c28b215d66e6a82f251e05014dec02250496d48d2288c2df503a07a5edf2f8e"
+  },
+  "chat_template.jinja": {
+    "bytes": 4116,
+    "sha256": "87a2728cb8dc9fe424d624542f6060ec05a1d285ebbec578bb078900e33396b5"
+  },
+  "config.json": {
+    "bytes": 1595,
+    "sha256": "b588e28ec8c6aabdb87ed8d07da6fbee62aa26415bf64a8399b3cb7a79248a60"
+  },
+  "evaluation.json": {
+    "bytes": 3158,
+    "sha256": "f3bf35b1697eb14151e2641c03b3b443488141fc250e22886529f00aa8794ce0"
+  },
+  "generation_config.json": {
+    "bytes": 139,
+    "sha256": "689c1e2b6769259cc773777319bae6f2bd6020ea1aa432b437578da46c994e8a"
+  },
+  "model.safetensors": {
+    "bytes": 16381517208,
+    "sha256": "16f70e6c00b3bd27323bc49d14903d16ef988cd78c76032eef2b6ffe29c60cac"
+  },
+  "package_versions.json": {
+    "bytes": 152,
+    "sha256": "a33200ae00c1fcb956c6dd25fb3de9e0d43ccaa9710347e726cec7c9a3eb8505"
+  },
+  "tokenizer.json": {
+    "bytes": 11422817,
+    "sha256": "79cb3c783570f1b8fe73b9ed530ae50cae9ce4b6344c0b5edefc50478847eaa4"
+  },
+  "tokenizer_config.json": {
+    "bytes": 800,
+    "sha256": "1c28b215d66e6a82f251e05014dec02250496d48d2288c2df503a07a5edf2f8e"
+  },
+  "training_recipe.json": {
+    "bytes": 2866,
+    "sha256": "5203508b957524dfd13d2d4cc2406742c8bff8dd5e2d4a9da72886121ddd74fd"
+  }
+}

config.json

@@ -0,0 +1,71 @@
+{
+  "architectures": [
+    "Qwen3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "dtype": "bfloat16",
+  "eos_token_id": 151643,
+  "head_dim": 128,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 12288,
+  "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",
+    "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": 36,
+  "model_type": "qwen3",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 8,
+  "pad_token_id": null,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "5.17.0",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

evaluation.json

@@ -0,0 +1,128 @@
+{
+  "data_repository": "Hyukkyu/RAQUEL2-ICLR",
+  "data_revision": "ac82756570fcce84441fb413ab523de8da679efd",
+  "evaluated_adapter": "early_stop",
+  "evidence": {
+    "judge_inputs_sha256": "38fb47d8fe042f69f8084eb3c07579891ea2446e32ec25e6460a1afdc3a29bda",
+    "judgments_sha256": "9d4cc78b5b2a139fb3cb3bfe37545d1adfae6931b814190966b0418c55787057"
+  },
+  "generation": {
+    "do_sample": false,
+    "max_new_tokens": 96,
+    "max_prompt_length": 512,
+    "prompt": "Question: {question}\nAnswer:"
+  },
+  "judge": {
+    "model": "Qwen/Qwen3.8-27B",
+    "served_name": "qwen3.8-27b",
+    "server": "vLLM 0.28.0",
+    "temperature": 0,
+    "thinking": false
+  },
+  "merge_audit": {
+    "max_kl": 0.004403834231197834,
+    "maximum_allowed_kl": 0.1,
+    "mean_absolute_logit_difference": 0.08547014743089676,
+    "merge_algorithm": "fp32_sum_then_cast_once",
+    "prompts": 16,
+    "top1_agreement": 1.0
+  },
+  "metrics": {
+    "raquel_affected": {
+      "accuracy": 0.2511530398322851,
+      "correct": 599,
+      "total": 2385
+    },
+    "raquel_unaffected": {
+      "accuracy": 0.25279850746268656,
+      "correct": 542,
+      "total": 2144
+    },
+    "wmdp_forget": {
+      "accuracy": 0.19117647058823528,
+      "correct": 286,
+      "total": 1496
+    },
+    "wmdp_forget_rephrased": {
+      "accuracy": 0.1487603305785124,
+      "correct": 216,
+      "total": 1452
+    },
+    "wmdp_retain": {
+      "accuracy": 0.9863201094391245,
+      "correct": 1442,
+      "total": 1462
+    },
+    "wmdp_retain_rephrased": {
+      "accuracy": 0.7833698030634574,
+      "correct": 1074,
+      "total": 1371
+    }
+  },
+  "reference": {
+    "M_orig": {
+      "raquel_affected": {
+        "accuracy": 0.29643605870020967,
+        "correct": 707,
+        "total": 2385
+      },
+      "raquel_unaffected": {
+        "accuracy": 0.2248134328358209,
+        "correct": 482,
+        "total": 2144
+      },
+      "wmdp_forget": {
+        "accuracy": 0.9979946524064172,
+        "correct": 1493,
+        "total": 1496
+      },
+      "wmdp_forget_rephrased": {
+        "accuracy": 0.7320936639118457,
+        "correct": 1063,
+        "total": 1452
+      },
+      "wmdp_retain": {
+        "accuracy": 0.9917920656634747,
+        "correct": 1450,
+        "total": 1462
+      },
+      "wmdp_retain_rephrased": {
+        "accuracy": 0.7571115973741794,
+        "correct": 1038,
+        "total": 1371
+      }
+    },
+    "M_ret": {
+      "raquel_affected": {
+        "accuracy": 0.2691823899371069,
+        "correct": 642,
+        "total": 2385
+      },
+      "raquel_unaffected": {
+        "accuracy": 0.23787313432835822,
+        "correct": 510,
+        "total": 2144
+      },
+      "wmdp_forget": {
+        "accuracy": 0.18983957219251338,
+        "correct": 284,
+        "total": 1496
+      },
+      "wmdp_forget_rephrased": {
+        "accuracy": 0.20798898071625344,
+        "correct": 302,
+        "total": 1452
+      },
+      "wmdp_retain": {
+        "accuracy": 0.9917920656634747,
+        "correct": 1450,
+        "total": 1462
+      },
+      "wmdp_retain_rephrased": {
+        "accuracy": 0.7964989059080962,
+        "correct": 1092,
+        "total": 1371
+      }
+    }
+  }
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": false,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "5.17.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16f70e6c00b3bd27323bc49d14903d16ef988cd78c76032eef2b6ffe29c60cac
+size 16381517208

package_versions.json

@@ -0,0 +1,8 @@
+{
+  "accelerate": "1.15.0",
+  "peft": "0.20.0",
+  "python": "3.12.3",
+  "safetensors": "0.8.0",
+  "torch": "2.14.0+cu130",
+  "transformers": "5.17.0"
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79cb3c783570f1b8fe73b9ed530ae50cae9ce4b6344c0b5edefc50478847eaa4
+size 11422817

tokenizer_config.json

@@ -0,0 +1,34 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "extra_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|>"
+  ],
+  "is_local": true,
+  "local_files_only": true,
+  "max_length": null,
+  "model_max_length": 131072,
+  "pad_to_multiple_of": null,
+  "pad_token": "<|endoftext|>",
+  "pad_token_type_id": 0,
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_recipe.json

@@ -0,0 +1,95 @@
+{
+  "adapter_weights_sha256": "38f4d47152baba7109259d6675b4a4f225f6f0eaadebe98ca8c27a714b306dde",
+  "base_model": "Hyukkyu/Qwen3-8B-Base-RAQUEL-WMDP-M-orig-LoRA-v1",
+  "base_revision": "dd9127c2e029620e76e3a526ad19be607815b8bf",
+  "data": {
+    "forget_rows": 1501,
+    "max_length": 512,
+    "pairing": "all forget and all retain source QA; the smaller side cycled in a seeded order, paired 1:1",
+    "qa_serialization": "paper_v1",
+    "repository": "Hyukkyu/RAQUEL2-ICLR",
+    "retain_rows": 1501,
+    "revision": "ac82756570fcce84441fb413ab523de8da679efd"
+  },
+  "early_stopping": {
+    "config": {
+      "cached_decode_agreement": 1.0,
+      "continue_after_stop": true,
+      "every_steps": 4,
+      "max_new_tokens": 64,
+      "rule": "max retain ROUGE-L s.t. forget ROUGE-L <= M_ret (margin 0; user 2026-09-24)",
+      "seed": 0,
+      "subset_size": 100,
+      "target_forget_rouge": 0.2094,
+      "target_source": "early_stop/wmdp_qwen3_M_ret.json"
+    },
+    "released_adapter": "step 112",
+    "rule": "max retain ROUGE-L s.t. forget ROUGE-L <= M_ret (user 2026-09-24)",
+    "selected_forget_rougeL_recall": 0.1885312367371191,
+    "selected_retain_rougeL_recall": 0.9938095238095238,
+    "selected_step": 112
+  },
+  "export_merge_algorithm": "fp32_sum_then_cast_once",
+  "export_precision": "bfloat16",
+  "learning_rate_note": "10x the full-parameter protocol's learning rates (LoRA convention), fixed before any LoRA result",
+  "lora": {
+    "adapter_dtype": "float32",
+    "alpha": 128.0,
+    "base_dtype": "bfloat16",
+    "dropout": 0.05,
+    "initialisation": "peft default (B=0: start == M_orig)",
+    "rank": 64,
+    "targets": [
+      "q_proj",
+      "k_proj",
+      "v_proj",
+      "o_proj",
+      "gate_proj",
+      "up_proj",
+      "down_proj"
+    ]
+  },
+  "merged_weights_sha256": "16f70e6c00b3bd27323bc49d14903d16ef988cd78c76032eef2b6ffe29c60cac",
+  "method": {
+    "method": "saul",
+    "saul": {
+      "forget_learning_rate": 0.0001,
+      "forget_perturbation_direction": -1,
+      "forget_threshold": 10.0,
+      "max_branch_grad_norm": 1.0,
+      "multiplier_init": 1.0,
+      "multiplier_step_size": 0.0001,
+      "optimizer_roles": [
+        "forget",
+        "retain"
+      ],
+      "optimizer_state_residency": "cpu_between_role_updates",
+      "parameter_space": "lora_adapters",
+      "retain_learning_rate": 0.0001,
+      "retain_perturbation_direction": 1,
+      "rho_forget": 0.01,
+      "rho_retain": 0.01,
+      "sam_epsilon": 1e-12,
+      "update_order": [
+        "forget",
+        "retain"
+      ]
+    }
+  },
+  "optimizer": {
+    "global_batch_size": 32,
+    "learning_rate": null,
+    "scheduler": null,
+    "seed": 0,
+    "weight_decay": null
+  },
+  "parameter_counts": {
+    "total": 8365323264,
+    "trainable": 174587904
+  },
+  "schedule": {
+    "epochs": 5,
+    "total_steps": 235
+  },
+  "training_complete": true
+}