glm4-step160: sft-lora

.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 +1,44 @@
+---
+library_name: peft
+base_model: zai-org/GLM-4-9B-0414
+license: apache-2.0
+tags:
+- thai
+- kanitakorn
+- single-model
+- greedy-eval
+language:
+- th
+- en
+---
+
+# Jnx03/kanitakorn-260614-glm4-glm4-step160
+
+Kanitakorn 2026-06-13 campaign artifact.
+
+- Base: `zai-org/GLM-4-9B-0414`
+- Version: `glm4-step160`
+- Method: `sft-lora`
+- Final-claim policy: one model, no BoN, no self-consistency, no ensemble, no model routing.
+- Date: 2026-06-13
+
+## Scores
+
+| Benchmark | Score | Target | Status |
+|---|---:|---:|---|
+| aime24_th | pending | >15.0 | pending |
+| aime24 | pending | >25.0 | pending |
+| math500_th | pending | >56.0 | pending |
+| math500 | pending | >82.0 | pending |
+| livecodebench_th | pending | >35.0 | pending |
+| livecodebench | pending | >60.0 | pending |
+| openthaieval | pending | >80.0 | pending |
+| hotpotqa_th_en | pending | >46.0 | pending |
+| instruction_following_th_en | pending | >57.0 | pending |
+| mt_bench_th_en | pending | >85.0 | pending |
+| thaiexam | pending | >70.0 | pending |
+| ifeval_th | pending | >82.0 | pending |
+
+## Notes
+
+Automatic checkpoint publish from runs/nonqwen-glm4-v1-v12hard-noreplay-r32-lr6e6-160/checkpoint-160. Single-model artifact; no BoN/self-consistency.

adapter_config.json

@@ -0,0 +1,47 @@
+{
+  "alora_invocation_tokens": null,
+  "alpha_pattern": {},
+  "arrow_config": null,
+  "auto_mapping": null,
+  "base_model_name_or_path": "zai-org/GLM-4-9B-0414",
+  "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": 64,
+  "lora_bias": false,
+  "lora_dropout": 0.03,
+  "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": 32,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "gate_up_proj",
+    "o_proj",
+    "q_proj",
+    "k_proj",
+    "down_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
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1373783d6c032bfdb3d16efcfb1ee0466f8e45f34d409a8e07cc84faf1fd8aa
+size 380829328

chat_template.jinja

@@ -0,0 +1,41 @@
+[gMASK]<sop>
+{%- if tools -%}
+<|system|>
+# 可用工具
+{% for tool in tools %}
+    {%- set function = tool.function if tool.get("function") else tool %}
+
+## {{ function.name }}
+
+{{ function | tojson(indent=4, ensure_ascii=False) }}
+在调用上述函数时，请使用 Json 格式表示调用的参数。
+{%- endfor %}
+{%- endif -%}
+
+{%- for msg in messages %}
+    {%- if msg.role == 'system' %}
+<|system|>
+{{ msg.content }}
+    {%- endif %}
+{%- endfor %}
+
+{%- for message in messages if message.role != 'system' %}
+    {%- set role = message['role'] %}
+    {%- set content = message['content'] %}
+    {%- set meta = message.get("metadata", "") %}
+
+    {%- if role == 'user' %}
+<|user|>
+{{ content }}
+    {%- elif role == 'assistant' and not meta %}
+<|assistant|>
+{{ content }}
+    {%- elif role == 'assistant' and meta %}
+<|assistant|>{{ meta }}
+{{ content }}
+    {%- elif role == 'observation' %}
+<|observation|>
+{{ content }}
+    {%- endif %}
+{%- endfor %}
+{% if add_generation_prompt %}<|assistant|>{% endif %}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:76ebeac0d8bd7879ead7b43c16b44981f277e47225de2bd7de9ae1a6cc664a8c
+size 19966496

tokenizer_config.json

@@ -0,0 +1,16 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|user|>",
+  "is_local": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 128000,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "remove_space": false,
+  "tokenizer_class": "TokenizersBackend"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:604698582effaf6fa275bf04f115e31759337d158ee3332f020c3eec975b676a
+size 5649