初始化项目，由ModelHub XC社区提供模型

Model: giux78/zagreus_0.4_competition
Source: Original Platform

.gitattributes

@@ -0,0 +1,36 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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 +1,58 @@
+---
+base_model: mii-llm/nesso-0.4B-agentic
+language:
+  - it
+pipeline_tag: text-generation
+tags:
+  - distillation
+  - on-policy-distillation
+  - italic
+  - italian
+  - mcqa
+---
+
+# zagreus_0.4_competition
+
+**nesso-0.4B-agentic improved on the [ITALIC](https://github.com/Crisp-Unimib/ITALIC) benchmark by on-policy distillation from Coloss/nesso-3B.**
+
+Official ITALIC harness (`run_eval.py`, full 10,000 questions, 5-shot):
+
+| Model | fast | slow (CoT) |
+|---|---:|---:|
+| mii-llm/nesso-0.4B-agentic (baseline) | 33.1% | 0.0%* |
+| **this model** | **37.2%** | **33.2%** |
+| Coloss/nesso-3B (teacher) | 50.7% | 2.5%* |
+
+\* baseline/teacher answer with a bare letter, which the official slow-mode extractor cannot parse; this model answers parseably in both modes.
+
+## Method
+
+On-policy distillation ([tinker-cookbook recipe](https://github.com/thinking-machines-lab/tinker-cookbook/blob/main/tinker_cookbook/recipes/distillation/on_policy_distillation.py), reimplemented in [mii-llm/palingenesis](https://github.com/mii-llm/palingenesis) branch `odp`): every step the student samples completions with its current weights and the loss is the full-distribution reverse KL to the teacher over exactly those tokens. Student (ChatML) and teacher (Llama-3 template) share the Llama-3 base vocabulary; prompts are rendered per-model and the student's `<|im_end|>` mass is merged into the teacher's `<|eot_id|>` slot, so stopping behavior is distilled too.
+
+The winning run (`opd_v3`, checkpoint step 550 of 600):
+
+- pool of 49,643 ITALIC-format MCQA prompts (pinocchio-raw + MMLU-ita sources, deduped against the ITALIC 10k test set by normalized-question hash), **filtered to rows the teacher answers correctly** (`pgs distill-score` — pure KL would distill the teacher's errors too)
+- Italian-language (vocabulary/grammar/comprehension) rows upweighted ×4
+- 80% of prompts rendered with ITALIC's official 5 shots, `max_new_tokens 8` (terse supervision)
+- 600 steps, batch 32 prompts, lr 1e-5 cosine, full reverse KL, ~40 min on one A100 80GB
+
+Training/eval code: [giux78/zagreus_0.4_competition](https://github.com/giux78/zagreus_0.4_competition) (experiment) + [mii-llm/palingenesis](https://github.com/mii-llm/palingenesis) `odp` branch (library, `pgs distill`).
+
+## Usage
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+tok = AutoTokenizer.from_pretrained("giux78/zagreus_0.4_competition")
+model = AutoModelForCausalLM.from_pretrained("giux78/zagreus_0.4_competition", dtype="bfloat16")
+
+messages = [
+    {"role": "system", "content": "Sei un assistente utile."},
+    {"role": "user", "content": "Rispondi alla seguente domanda a scelta multipla..."},
+]
+ids = tok.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
+out = model.generate(ids, max_new_tokens=8, do_sample=False)
+print(tok.decode(out[0, ids.shape[1]:], skip_special_tokens=True))
+```
+
+For ITALIC-style evaluation use `max_new_tokens=8` and greedy decoding (the model answers with the bare option letter).

chat_template.jinja

@@ -0,0 +1,68 @@
+{{ bos_token }}
+{% if tools %}
+<|im_start|>system
+{% if messages and messages[0]['role'] == 'system' %}
+{{ messages[0]['content'] }}
+{% endif %}
+# Tools
+You may call one or more functions to assist with the user query.
+You are provided with function signatures within <tools></tools> XML tags:
+<tools>
+{% for tool in tools %}
+{{ tool | tojson }}
+{% endfor %}
+</tools>
+For each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:
+<tool_call>
+{"name": <function-name>, "arguments": <args-json-object>}
+</tool_call><|im_end|>
+{% else %}
+{% if messages and messages[0]['role'] == 'system' %}
+<|im_start|>system
+{{ messages[0]['content'] }}<|im_end|>
+{% endif %}
+{% endif %}
+
+{% for message in messages %}
+{% if message.content is string %}
+  {% set content = message.content %}
+{% else %}
+  {% set content = '' %}
+{% endif %}
+
+{% if (message.role == 'user') or (message.role == 'system' and not loop.first) %}
+<|im_start|>{{ message.role }}
+{{ content }}<|im_end|>
+
+{% elif message.role == 'assistant' %}
+<|im_start|>assistant
+{{ content }}
+{% if message.tool_calls %}
+  {% for tool_call in message.tool_calls %}
+    {% if tool_call.function %}
+      {% set tool_call = tool_call.function %}
+    {% endif %}
+<tool_call>
+{"name": "{{ tool_call.name }}", "arguments": {% if tool_call.arguments is string %}{{ tool_call.arguments }}{% else %}{{ tool_call.arguments | tojson }}{% endif %}}
+</tool_call>
+  {% endfor %}
+{% endif %}
+<|im_end|>
+
+{% elif message.role == 'tool' %}
+{% if loop.first or (messages[loop.index0 - 1].role != 'tool') %}
+<|im_start|>user
+{% endif %}
+<tool_response>
+{{ content }}
+</tool_response>
+{% if loop.last or (messages[loop.index0 + 1].role != 'tool') %}
+<|im_end|>
+{% endif %}
+
+{% endif %}
+{% endfor %}
+
+{% if add_generation_prompt %}
+<|im_start|>assistant
+{% endif %}

config.json

@@ -0,0 +1,32 @@
+{
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 128000,
+  "dtype": "float32",
+  "eos_token_id": 128256,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 960,
+  "initializer_range": 0.02,
+  "intermediate_size": 2560,
+  "max_position_embeddings": 4096,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 15,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 5,
+  "pad_token_id": null,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_parameters": {
+    "rope_theta": 10000.0,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.7.0",
+  "use_cache": false,
+  "vocab_size": 128262
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 128000,
+  "do_sample": true,
+  "eos_token_id": 128001,
+  "transformers_version": "5.7.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f744242923594afe1764865ef90a2ef03c76d32b18c1046a094798efc3b89998
+size 1751099616

opd_config.json

@@ -0,0 +1,56 @@
+{
+  "model": {
+    "student": "mii-llm/nesso-0.4B-agentic",
+    "teacher": "giux78/nesso-3B",
+    "teacher_device": "",
+    "gradient_checkpointing": true
+  },
+  "bridge": {
+    "eos_map": {
+      "<|im_end|>": "<|eot_id|>"
+    },
+    "extra_stop_tokens": [
+      "<|end_of_text|>"
+    ],
+    "probe_texts": []
+  },
+  "data": {
+    "prompts_path": "/home/ecuser/ai/zagreus_competition/data/prompts_v3.jsonl",
+    "shots_path": "/home/ecuser/ai/zagreus_competition/data/5_shots.jsonl",
+    "dev_size": 500,
+    "p_reference_shots": 0.8,
+    "p_pool_shots": 0.1,
+    "pool_shots_max_k": 5,
+    "system_message": ""
+  },
+  "sampling": {
+    "batch_prompts": 32,
+    "group_size": 1,
+    "temperature": 1.0,
+    "max_new_tokens": 8,
+    "cot_fraction": 0.0,
+    "cot_max_new_tokens": 300,
+    "gen_micro_seqs": 64
+  },
+  "train": {
+    "output_dir": "/home/ecuser/ai/palingenesis/runs/opd_v3",
+    "steps": 600,
+    "learning_rate": 1e-05,
+    "warmup_steps": 50,
+    "lr_scheduler": "cosine",
+    "max_grad_norm": 1.0,
+    "loss_fn": "full_kl",
+    "seed": 0,
+    "score_micro_seqs": 8,
+    "eval_every": 50,
+    "eval_dev_samples": 200,
+    "save_steps": 50,
+    "keep_checkpoints": 0
+  },
+  "logging": {
+    "log_every": 10,
+    "use_wandb": false,
+    "project": "palingenesis-opd",
+    "run_name": ""
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4a80502037c38583156839ad1486db1e7e5aa0b3f9d47e072f16e6b30a0eb2dd
+size 17211058

tokenizer_config.json

@@ -0,0 +1,15 @@
+{
+  "backend": "tokenizers",
+  "bos_token": "<|begin_of_text|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|im_end|>",
+  "is_local": false,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 131072,
+  "pad_token": "<|end_of_text|>",
+  "tokenizer_class": "TokenizersBackend"
+}