initial, eval_loss: 0.5200754404067993

.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

@@ -1,3 +1,109 @@
----
-license: mit
----
+---
+license: mit
+datasets:
+  - Egor-AI/Russian_thinking_dataset
+language:
+  - ru
+  - en
+base_model:
+  - t-tech/T-lite-it-1.0
+pipeline_tag: question-answering
+library_name: peft
+tags:
+  - chat
+  - o1
+  - cot
+  - thinking
+  - reflection
+---
+
+# Russian o1 / T-lite-it-1.0 LoRA
+
+Based on https://huggingface.co/evilfreelancer/o1_t-lite-it-1.0_lora
+
+LoRA-адаптер для модели [T-lite-it-1.0](https://huggingface.co/t-tech/T-lite-it-1.0) обученный на
+датасете [Egor-AI/Russian_thinking_dataset](https://huggingface.co/datasets/Egor-AI/Russian_thinking_dataset) (машинный
+перевод на русский язык
+датасета [BintangFortuna/OpenO1-SFT-EN-SY](https://huggingface.co/datasets/BintangFortuna/OpenO1-SFT-EN-SY)).
+
+Обученная модель способна имитировать логические размышлению на русском языке по аналогии с тем, как
+это делает `o1` от `OpenAI`.
+
+Необходимо использовать следующего вида системный промт:
+
+```
+Вы — ИИ-помощник. Отформатируйте свои ответы следующим образом: <Thought> Ваши мысли (понимание, рассуждения) </Thought> <output> Ваш ответ </output>
+```
+
+Обучение производилось при помощи утилиты [impruver](https://github.com/EvilFreelancer/impruver) используя конфигурацию
+[T-lite-it/7B_lora_thinking](https://github.com/EvilFreelancer/impruver/blob/main/recipes/configs/T-lite-it/7B_lora_thinking.yaml) с донастройкой:
+- load_in_4bit: false
+- без max_tokens_count
+- optim: adamw_8bit
+- gradient_accumulation_steps: 1
+
+На всё про всё ушло примерно 17.6 часов на 1xH100 80GB, при этом понадобилось 67Гб видеопамяти.
+
+Результатирующий eval_loss: 0.5200754404067993
+
+W&B run: https://wandb.ai/b37h3z3n/trains/runs/6vwvuu46?nw=nwuserb37h3z3n
+
+
+```yaml
+output_dir: ./models/T-lite-it_7B_lora_thinking
+train_path: ./train.T-lite-it_7B_lora_thinking.jsonl
+val_path: ./val.T-lite-it_7B_lora_thinking.jsonl
+
+datasets:
+  - name: Egor-AI/Russian_thinking_dataset
+    converter: impruver.instruction_to_messages
+    add_global_bos: false
+    add_global_eos: false
+    mapping:
+      system: system
+      instruction: prompt
+      output: response
+
+model:
+  class: transformers.AutoModelForCausalLM
+  name: t-tech/T-lite-it-1.0
+  load_in_4bit: false
+  load_in_8bit: false
+  dtype: bf16
+
+lora:
+  r: 16
+  lora_alpha: 16
+  lora_dropout: 0
+  bias: none
+  target_modules: [ q_proj, k_proj, v_proj, o_proj, gate_proj, down_proj, up_proj ]
+  task_type: CAUSAL_LM
+
+tokenizer:
+  class: transformers.AutoTokenizer
+  name: t-tech/T-lite-it-1.0
+#  max_tokens_count: 1500
+
+trainer:
+  eval_strategy: steps
+  save_strategy: steps
+  eval_steps: 100
+  save_steps: 100
+  per_device_train_batch_size: 1
+  per_device_eval_batch_size: 1
+  gradient_accumulation_steps: 1
+  logging_steps: 10
+  learning_rate: 0.0004
+  num_train_epochs: 3
+  lr_scheduler_type: cosine
+  warmup_steps: 16
+  optim: adamw_torch_4bit
+  metric_for_best_model: eval_loss
+  load_best_model_at_end: true
+  save_total_limit: 2
+  seed: 42
+  remove_unused_columns: false
+  max_grad_norm: 1.0
+  weight_decay: 0.08
+  torch_compile: false
+```

adapter_config.json

@@ -0,0 +1,37 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "/root/T-lite-it-1.0",
+  "bias": "none",
+  "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": 16,
+  "lora_bias": false,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "o_proj",
+    "q_proj",
+    "down_proj",
+    "up_proj",
+    "gate_proj",
+    "k_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:91bcf35e7cad58f9547bc8ef30d89b622d3881f4cb4d47632298cac0df248ffc
+size 161621802

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
+}

chat_transformers.py

@@ -0,0 +1,154 @@
+import fire
+from typing import List, Dict
+import torch
+from peft import PeftModel
+from transformers import AutoTokenizer, AutoModelForCausalLM, GenerationConfig, BitsAndBytesConfig
+
+MODEL_BASE = "t-tech/T-lite-it-1.0"
+MODEL_ADAPTER = "evilfreelancer/o1_t-lite-it-1.0_lora"
+
+SYSTEM_PROMPT = "Вы — ИИ-помощник. Отформатируйте свои ответы следующим образом: <Thought> Ваши мысли (понимание, рассуждения) </Thought> <output> Ваш ответ </output>"
+
+
+class ChatHistory:
+    def __init__(self, history_limit: int = None, system_prompt: str = None):
+        self.history_limit: int | None = history_limit
+        self.system_prompt: str | None = system_prompt
+        self.messages: List[Dict] = []
+        if self.system_prompt is not None:
+            self.messages.append({"role": "system", "content": self.system_prompt})
+
+    def add_message(self, role: str, message: str):
+        self.messages.append({"role": role, "content": message})
+        self.trim_history()
+
+    def add_user_message(self, message: str):
+        self.add_message("user", message)
+
+    def add_assistant_message(self, message: str):
+        self.add_message("assistant", message)
+
+    def add_function_call(self, message: str):
+        self.add_message("function_call", message)
+
+    def add_function_response(self, message: str):
+        self.add_message("function_response", message)
+
+    def trim_history(self):
+        appendix = 0
+        if self.system_prompt is not None:
+            appendix = 1
+        if self.history_limit is not None and len(self.messages) > self.history_limit + appendix:
+            overflow = len(self.messages) - (self.history_limit + appendix)
+            self.messages = [self.messages[0]] + self.messages[overflow + appendix:]
+
+    def get_messages(self) -> list:
+        return self.messages
+
+
+def generate(model, tokenizer, prompt, generation_config):
+    data = tokenizer(prompt, return_tensors="pt")
+    data = {k: v.to(model.device) for k, v in data.items()}
+    output_ids = model.generate(**data, generation_config=generation_config)[0]
+    output_ids = output_ids[len(data["input_ids"][0]):]
+    output = tokenizer.decode(output_ids, skip_special_tokens=True)
+    return output.strip()
+
+
+def get_prompt(tokenizer, messages: List[Dict], add_generation_prompt: bool = False):
+    return tokenizer.apply_chat_template(
+        messages,
+        add_special_tokens=False,
+        tokenize=False,
+        add_generation_prompt=add_generation_prompt,
+    )
+
+
+def chat(
+    history_limit: int = 1,
+    system_prompt: str | None = SYSTEM_PROMPT,
+    max_new_tokens: int = 2048,
+    repetition_penalty: float = 1.2,
+    do_sample: bool = True,
+    temperature: float = 0.5,
+    top_p: float = 0.6,
+    top_k: int = 40,
+):
+    #
+    # Tokenizer preparation
+    #
+
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_BASE)
+
+    #
+    # Model preparation
+    #
+
+    # Quantization config
+    quantization_config = BitsAndBytesConfig(
+        load_in_4bit=True,
+        bnb_4bit_compute_dtype=torch.bfloat16,
+        bnb_4bit_quant_type="nf4",
+        bnb_4bit_use_double_quant=True
+    )
+
+    # Generator config
+    generation_config = GenerationConfig.from_pretrained(MODEL_ADAPTER)
+    generation_config.max_new_tokens = max_new_tokens
+    generation_config.repetition_penalty = repetition_penalty
+    generation_config.do_sample = do_sample
+    generation_config.temperature = temperature
+    generation_config.top_p = top_p
+    generation_config.top_k = top_k
+
+    # Read model from folder with trained checkpoints
+    model = AutoModelForCausalLM.from_pretrained(
+        MODEL_BASE,
+        generation_config=generation_config,
+        quantization_config=quantization_config,
+        torch_dtype=torch.bfloat16,
+        attn_implementation=None
+    )
+
+    # If we've trained a LoRA adapter
+    model = PeftModel.from_pretrained(
+        model=model,
+        model_id=MODEL_ADAPTER,
+        torch_dtype=torch.bfloat16,
+    )
+
+    #
+    # Chat loop
+    #
+
+    # Start chat loop
+    chat_history = ChatHistory(history_limit, system_prompt)
+    while True:
+        user_message = input("User: ")
+
+        # Reset chat command
+        if user_message.strip() == "/reset":
+            chat_history = ChatHistory(history_limit, system_prompt)
+            print("History reset completed!")
+            continue
+
+        # Skip empty messages from user
+        if user_message.strip() == "":
+            continue
+
+        # Add user message to chat history
+        chat_history.add_user_message(user_message)
+
+        # Get list of messages
+        prompt = get_prompt(tokenizer, chat_history.get_messages(), True)
+
+        # Generate response
+        output = generate(model, tokenizer, prompt, generation_config)
+
+        # Save response to chat history as assistant's message
+        chat_history.add_assistant_message(output)
+        print("Assistant:", output)
+
+
+if __name__ == "__main__":
+    fire.Fire(chat)

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "do_sample": true,
+  "temperature": 0.7,
+  "top_k": 70,
+  "top_p": 0.8,
+  "repetition_penalty": 1.05,
+  "pad_token_id": 151643,
+  "bos_token_id": 151643,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "transformers_version": "4.47.1"
+}

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb9505628340c6e36bef0c05abac1bea0c27c34f5d7c539cc6fc4c621397a752
+size 82461812

requirements.txt

@@ -0,0 +1,6 @@
+pyyaml>=6.0.2
+fire>=0.7.0
+torch>=2.5.1
+transformers>=4.47.1
+peft>=0.14.0
+bitsandbytes>=0.45.0

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8dccfcc3afceab4d2d2b25fbb67392f946fa9f1c9efbf11612cbcfa90ce2d337
+size 14244

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd8b83240cd31306fe73d6b8457fc426e2cc12ec08175c34c0dcaf6298b44bb9
+size 1064

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
+  }
+}

tokenizer.json

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

tokenizer_config.json

@@ -0,0 +1,208 @@
+{
+  "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,
+  "chat_template": "{% for message in messages %}\n    {{- '<|im_start|>' + message.role + '\n' + message.content  + '<|im_end|>' + '\n' -}}\n{% endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\n' -}}\n{%- endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "model_max_length": 8192,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6dd51dc0e7660db8c62dacf10a35046078f9de0202bb28175ee7193728a5b59c
+size 5432