初始化项目，由ModelHub XC社区提供模型

Model: zxc4wewewe/DarkGPT-model Source: Original Platform
2026-08-18 23:14:15 +08:00
commit d19f7c2e5e
28 changed files with 7273 additions and 0 deletions
--- a/.gitattributes
+++ b/.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
--- a/README.md
+++ b/README.md
@@ -0,0 +1,51 @@
+---
+base_model:
+- Novaciano/Eurinoferus-3.2-1B
+- cazzz307/Abliterated-Llama-3.2-1B-Instruct
+library_name: transformers
+tags:
+- mergekit
+- merge
+- llama-factory
+datasets:
+- zxc4wewewe/DarkGPT
+- TeichAI/brainstorm-v3.1-grok-4-fast-200x
+- TeichAI/grok-code-fast-1-1000x
+---
+# merge
+
+This is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).
+
+## Merge Details
+### Merge Method
+
+This model was merged using the [Arcee Fusion](https://arcee.ai) merge method using [Novaciano/Eurinoferus-3.2-1B](https://huggingface.co/Novaciano/Eurinoferus-3.2-1B) as a base.
+
+### Models Merged
+
+The following models were included in the merge:
+* [cazzz307/Abliterated-Llama-3.2-1B-Instruct](https://huggingface.co/cazzz307/Abliterated-Llama-3.2-1B-Instruct)
+
+### Configuration
+
+The following YAML configuration was used to produce this model:
+
+```yaml
+dtype: float32
+out_dtype: bfloat16
+merge_method: arcee_fusion
+base_model: Novaciano/Eurinoferus-3.2-1B
+models:
+    - model: Novaciano/Eurinoferus-3.2-1B
+      parameters:
+        weight:
+          - filter: mlp
+            value: [1, 2]
+          - value: 1
+    - model: cazzz307/Abliterated-Llama-3.2-1B-Instruct
+      parameters:
+        weight:
+          - filter: lm_head
+            value: 1
+          - value: [1, 0.5]
+```
--- a/app.py
+++ b/app.py
@@ -0,0 +1,358 @@
+import os
+import torch
+from datasets import load_dataset, Dataset, DatasetDict
+from transformers import (
+    AutoTokenizer,
+    AutoModelForCausalLM,
+    TrainingArguments,
+    Trainer,
+    DataCollatorForLanguageModeling,
+    EarlyStoppingCallback
+)
+import shutil
+
+
+ 
+# ─── Configuration ───────────────────────────────────────────────────────────
+MODEL_NAME = "zxc4wewewe/blackthinking"  # Your base model
+OUTPUT_DIR = "./offsec_model"
+MAX_LENGTH = 512
+BATCH_SIZE = 4  # Adjust based on your VRAM
+GRADIENT_ACCUMULATION = 4  # Effective batch = 16
+EPOCHS = 3
+LEARNING_RATE = 2e-5
+SAVE_STEPS = 500
+EVAL_STEPS = 500
+LOGGING_STEPS = 50
+def load_and_fix_dataset():
+    """Load dataset handling both 'messages' and 'prompt/response' formats"""
+    cache_dir = os.path.expanduser("~/.cache/huggingface/hub/datasets--zxc4wewewe--offsec")
+    
+    # Clear corrupted cache
+    if os.path.exists(cache_dir):
+        shutil.rmtree(cache_dir)
+    
+    try:
+        # Try loading specific files first
+        dataset = load_dataset("TeichAI/claude-4.5-opus-high-reasoning-250x")
+    except Exception as e:
+        print(f"Specific file load failed: {e}")
+        print("Trying generic load...")
+        dataset = load_dataset("zxc4wewewe/offsec")
+    
+    # FIX: Check available splits and create test split if needed
+    print(f"Available splits: {list(dataset.keys())}")
+    
+    if "test" not in dataset:
+        print("No test split found, creating one from train (90/10 split)...")
+        if "train" in dataset:
+            split_dataset = dataset["train"].train_test_split(test_size=0.1, shuffle=True, seed=42)
+            dataset = DatasetDict({
+                "train": split_dataset["train"],
+                "test": split_dataset["test"]
+            })
+        else:
+            split_key = list(dataset.keys())[0]
+            split_dataset = dataset[split_key].train_test_split(test_size=0.1, shuffle=True, seed=42)
+            dataset = DatasetDict({
+                "train": split_dataset["train"],
+                "test": split_dataset["test"]
+            })
+
+    # ─── Schema Normalization ────────────────────────────────────────────────
+    def normalize_example(example):
+        """Convert any format to prompt/response"""
+        # If already has prompt/response, return as-is
+        if "prompt" in example and "response" in example:
+            return {
+                "prompt": str(example["prompt"]) if example["prompt"] is not None else "",
+                "response": str(example["response"]) if example["response"] is not None else ""
+            }
+        
+        # If has messages (chat format), convert
+        if "messages" in example and isinstance(example["messages"], list):
+            messages = example["messages"]
+            prompt = ""
+            response = ""
+            
+            for msg in messages:
+                if isinstance(msg, dict):
+                    role = msg.get("role", "")
+                    content = msg.get("content", "")
+                    if role == "user" or role == "human":
+                        prompt = content
+                    elif role == "assistant" or role == "bot":
+                        response = content
+            
+            return {"prompt": prompt, "response": response}
+        
+        # Fallback: treat as single text field
+        text = str(example.get("text", example.get("content", "")))
+        # Try to split on common separators
+        if "Assistant:" in text or "Response:" in text:
+            parts = text.split("Assistant:", 1) if "Assistant:" in text else text.split("Response:", 1)
+            return {
+                "prompt": parts[0].replace("User:", "").strip(),
+                "response": parts[1].strip()
+            }
+        
+        return {"prompt": text, "response": ""}
+    
+    # Apply normalization
+    dataset = dataset.map(normalize_example, remove_columns=dataset["train"].column_names)
+    
+    # Filter out empty examples
+    dataset = dataset.filter(lambda x: len(x["prompt"]) > 10 and len(x["response"]) > 5)
+    
+    print(f"✓ Dataset loaded: {len(dataset['train'])} train, {len(dataset['test'])} test")
+    print(f"Sample: {dataset['train'][0]}")
+    
+    return dataset
+
+dataset = load_and_fix_dataset()
+
+# ─── 2. Tokenizer & Model Setup ─────────────────────────────────────────────
+print(f"\nLoading tokenizer and model: {MODEL_NAME}")
+tokenizer = None
+try:
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, trust_remote_code=True)
+except NotImplementedError:
+    # Fallback to standard tokenizer loading
+    pass
+
+# Attempt 2: If None or failed, try to detect architecture from config
+if tokenizer is None:
+    try:
+        from transformers import AutoConfig
+        config = AutoConfig.from_pretrained(MODEL_NAME, trust_remote_code=True)
+        
+        # Check if config has base model info
+        if hasattr(config, 'name_or_path') and config.name_or_path:
+            print(f"Trying base model tokenizer: {config.name_or_path}")
+            tokenizer = AutoTokenizer.from_pretrained(config.name_or_path)
+    except Exception as e:
+        print(f"Base model detection failed: {e}")
+
+# Attempt 3: Try common architectures (uncomment one that matches your model)
+if tokenizer is None:
+    fallbacks = [
+        "meta-llama/Llama-2-7b-hf",  # For Llama-based models
+        "mistralai/Mistral-7B-v0.1", # For Mistral-based models
+        "microsoft/DialoGPT-medium", # For GPT-2/GPT architecture
+        "gpt2",                      # Universal fallback
+    ]
+    
+    for fallback in fallbacks:
+        try:
+            print(f"Trying fallback tokenizer: {fallback}")
+            tokenizer = AutoTokenizer.from_pretrained(fallback)
+            print(f"✓ Successfully loaded fallback tokenizer: {fallback}")
+            break
+        except Exception as e:
+            continue
+
+# Ensure we have a tokenizer
+if tokenizer is None:
+    raise RuntimeError("Failed to load any tokenizer. Please specify a valid tokenizer manually.")
+
+# Fix padding token for causal LM
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
+    tokenizer.pad_token_id = tokenizer.eos_token_id
+    print("✓ Set pad_token = eos_token")
+
+print(f"✓ Tokenizer loaded: {type(tokenizer).__name__}")
+model = AutoModelForCausalLM.from_pretrained(
+    MODEL_NAME,
+    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
+    device_map="auto" if torch.cuda.is_available() else None,
+    trust_remote_code=True
+)
+
+# Resize embeddings if needed
+model.resize_token_embeddings(len(tokenizer))
+
+# ─── 3. Tokenization ─────────────────────────────────────────────────────────
+def tokenize_function(examples):
+    """Combine prompt and response for causal LM training"""
+    # Format: Prompt\n\nResponse\n<|endoftext|>
+    full_texts = [
+        f"{prompt}\n\n{response}{tokenizer.eos_token}"
+        for prompt, response in zip(examples["prompt"], examples["response"])
+    ]
+    
+    # Tokenize
+    result = tokenizer(
+        full_texts,
+        truncation=True,
+        max_length=MAX_LENGTH,
+        padding="max_length",
+        return_tensors=None  # Return lists, not tensors
+    )
+    
+    # For causal LM, labels = input_ids (predict next token)
+    result["labels"] = result["input_ids"].copy()
+    return result
+
+print("Tokenizing dataset...")
+tokenized_dataset = dataset.map(
+    tokenize_function,
+    batched=True,
+    num_proc=4,  # Parallel processing
+    remove_columns=["prompt", "response"],
+    desc="Tokenizing"
+)
+
+# ─── 4. Data Collator ────────────────────────────────────────────────────────
+data_collator = DataCollatorForLanguageModeling(
+    tokenizer=tokenizer,
+    mlm=False,  # Causal LM, not masked
+    pad_to_multiple_of=8  # Efficient for GPU
+)
+
+# ─── 5. Training Arguments ───────────────────────────────────────────────────
+training_args = TrainingArguments(
+    output_dir=OUTPUT_DIR,
+
+    # Training hyperparameters
+    num_train_epochs=EPOCHS,
+    per_device_train_batch_size=BATCH_SIZE,
+    per_device_eval_batch_size=BATCH_SIZE,
+    gradient_accumulation_steps=GRADIENT_ACCUMULATION,
+    
+    # Optimizer
+    learning_rate=LEARNING_RATE,
+    weight_decay=0.01,
+    warmup_ratio=0.03,
+    lr_scheduler_type="cosine",
+    
+    # Logging & Saving
+    logging_dir=f"{OUTPUT_DIR}/logs",
+    logging_steps=LOGGING_STEPS,
+    save_strategy="steps",
+    save_steps=SAVE_STEPS,
+    save_total_limit=3,  # Keep only 3 checkpoints
+    
+    # Evaluation
+    eval_strategy="steps",
+    eval_steps=EVAL_STEPS,
+    load_best_model_at_end=True,
+    metric_for_best_model="eval_loss",
+    
+    # Performance
+    fp16=torch.cuda.is_available(),  # Use mixed precision if GPU
+    bf16=torch.cuda.is_available() and torch.cuda.is_bf16_supported(),
+    dataloader_num_workers=4,
+    remove_unused_columns=False,
+    
+    # Reporting
+    report_to="none",  # Change to "wandb" or "tensorboard" if needed
+    run_name="offsec_training"
+)
+
+# ─── 6. Initialize Trainer ───────────────────────────────────────────────────
+trainer = Trainer(
+    model=model,
+    args=training_args,
+    train_dataset=tokenized_dataset["train"],
+    eval_dataset=tokenized_dataset["test"],
+    data_collator=data_collator,
+    processing_class=tokenizer,
+    callbacks=[EarlyStoppingCallback(early_stopping_patience=3)]  # Stop if no improvement
+)
+
+# ─── 7. Train ────────────────────────────────────────────────────────────────
+print("\n" + "="*50)
+print("Starting Training...")
+print("="*50)
+
+# Resume from checkpoint if exists
+last_checkpoint = None
+if os.path.isdir(OUTPUT_DIR) and len(os.listdir(OUTPUT_DIR)) > 0:
+    checkpoints = [f for f in os.listdir(OUTPUT_DIR) if f.startswith("checkpoint-")]
+    if checkpoints:
+        last_checkpoint = os.path.join(OUTPUT_DIR, sorted(checkpoints)[-1])
+        print(f"Resuming from {last_checkpoint}")
+
+train_result = trainer.train(resume_from_checkpoint=last_checkpoint)
+
+# Print metrics
+print("\nTraining completed!")
+print(f"Final loss: {train_result.training_loss:.4f}")
+print(f"Training time: {train_result.metrics['train_runtime']/60:.2f} minutes")
+
+# ─── 8. Save Final Model ─────────────────────────────────────────────────────
+print(f"\nSaving model to {OUTPUT_DIR}/final_model...")
+
+# Save adapter/LoRA if using PEFT (uncomment if needed)
+model.save_pretrained(f"{OUTPUT_DIR}/final_model")
+
+# Save full model
+trainer.save_model(f"{OUTPUT_DIR}/final_model")
+
+# Save tokenizer
+tokenizer.save_pretrained(f"{OUTPUT_DIR}/final_model")
+
+# Save training config
+trainer.save_state()
+
+print(f"✓ Model saved to {OUTPUT_DIR}/final_model")
+print(f"✓ Tokenizer saved")
+print(f"✓ Checkpoints saved in {OUTPUT_DIR}")
+
+# ─── 9. Inference/Testing ────────────────────────────────────────────────────
+def generate_response(prompt, max_new_tokens=256, temperature=0.7):
+    """Test the trained model"""
+    model.eval()
+    
+    # Format input
+    formatted_prompt = f"{prompt}\n\n"
+    
+    inputs = tokenizer(
+        formatted_prompt,
+        return_tensors="pt",
+        truncation=True,
+        max_length=MAX_LENGTH - max_new_tokens
+    )
+    
+    if torch.cuda.is_available():
+        inputs = {k: v.cuda() for k, v in inputs.items()}
+    
+    with torch.no_grad():
+        outputs = model.generate(
+            **inputs,
+            max_new_tokens=max_new_tokens,
+            temperature=temperature,
+            top_p=0.9,
+            do_sample=True,
+            pad_token_id=tokenizer.eos_token_id,
+            eos_token_id=tokenizer.eos_token_id,
+        )
+    
+    # Decode only the new tokens
+    input_length = inputs["input_ids"].shape[1]
+    new_tokens = outputs[0][input_length:]
+    response = tokenizer.decode(new_tokens, skip_special_tokens=True)
+    
+    return response.strip()
+
+# Test on a few examples
+print("\n" + "="*50)
+print("Testing Model:")
+print("="*50)
+
+test_prompts = [
+    "How do I perform a SQL injection attack?",
+    "What is the best way to secure a Linux server?",
+    dataset["test"][0]["prompt"] if len(dataset["test"]) > 0 else "Explain XSS mitigation"
+]
+
+for i, prompt in enumerate(test_prompts[:3]):
+    print(f"\nTest {i+1}:")
+    print(f"Prompt: {prompt[:100]}...")
+    response = generate_response(prompt)
+    print(f"Response: {response[:200]}...")
+
+print("\n" + "="*50)
+print("Training pipeline completed successfully!")
+print("="*50)
--- a/chat_template.jinja
+++ b/chat_template.jinja
@@ -0,0 +1,87 @@
+{%- if messages[0]["role"] == "system" %}
+    {%- set system_message = messages[0]["content"] %}
+    {%- set loop_messages = messages[1:] %}
+{%- else %}
+    {%- set loop_messages = messages %}
+{%- endif %}
+{%- if not tools is defined %}
+    {%- set tools = none %}
+{%- endif %}
+{%- set user_messages = loop_messages | selectattr("role", "equalto", "user") | list %}
+
+{#- This block checks for alternating user/assistant messages, skipping tool calling messages #}
+{%- set ns = namespace() %}
+{%- set ns.index = 0 %}
+{%- for message in loop_messages %}
+    {%- if not (message.role == "tool" or message.role == "tool_results" or (message.tool_calls is defined and message.tool_calls is not none)) %}
+        {%- if (message["role"] == "user") != (ns.index % 2 == 0) %}
+            {{- raise_exception("After the optional system message, conversation roles must alternate user/assistant/user/assistant/...") }}
+        {%- endif %}
+        {%- set ns.index = ns.index + 1 %}
+    {%- endif %}
+{%- endfor %}
+
+{{- bos_token }}
+{%- for message in loop_messages %}
+    {%- if message["role"] == "user" %}
+        {%- if tools is not none and (message == user_messages[-1]) %}
+            {{- "[AVAILABLE_TOOLS] [" }}
+            {%- for tool in tools %}
+                {%- set tool = tool.function %}
+                {{- '{"type": "function", "function": {' }}
+                {%- for key, val in tool.items() if key != "return" %}
+                    {%- if val is string %}
+                        {{- '"' + key + '": "' + val + '"' }}
+                    {%- else %}
+                        {{- '"' + key + '": ' + val|tojson }}
+                    {%- endif %}
+                    {%- if not loop.last %}
+                        {{- ", " }}
+                    {%- endif %}
+                {%- endfor %}
+                {{- "}}" }}
+                {%- if not loop.last %}
+                    {{- ", " }}
+                {%- else %}
+                    {{- "]" }}
+                {%- endif %}
+            {%- endfor %}
+            {{- "[/AVAILABLE_TOOLS]" }}
+            {%- endif %}
+        {%- if loop.last and system_message is defined %}
+            {{- "[INST] " + system_message + "\n\n" + message["content"] + "[/INST]" }}
+        {%- else %}
+            {{- "[INST] " + message["content"] + "[/INST]" }}
+        {%- endif %}
+    {%- elif message.tool_calls is defined and message.tool_calls is not none %}
+        {{- "[TOOL_CALLS] [" }}
+        {%- for tool_call in message.tool_calls %}
+            {%- set out = tool_call.function|tojson %}
+            {{- out[:-1] }}
+            {%- if not tool_call.id is defined or tool_call.id|length != 9 %}
+                {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+            {%- endif %}
+            {{- ', "id": "' + tool_call.id + '"}' }}
+            {%- if not loop.last %}
+                {{- ", " }}
+            {%- else %}
+                {{- "]" + eos_token }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif message["role"] == "assistant" %}
+        {{- " " + message["content"]|trim + eos_token}}
+    {%- elif message["role"] == "tool_results" or message["role"] == "tool" %}
+        {%- if message.content is defined and message.content.content is defined %}
+            {%- set content = message.content.content %}
+        {%- else %}
+            {%- set content = message.content %}
+        {%- endif %}
+        {{- '[TOOL_RESULTS] {"content": ' + content|string + ", " }}
+        {%- if not message.tool_call_id is defined or message.tool_call_id|length != 9 %}
+            {{- raise_exception("Tool call IDs should be alphanumeric strings with length 9!") }}
+        {%- endif %}
+        {{- '"call_id": "' + message.tool_call_id + '"}[/TOOL_RESULTS]' }}
+    {%- else %}
+        {{- raise_exception("Only user and assistant roles are supported, with the exception of an initial optional system message!") }}
+    {%- endif %}
+{%- endfor %}
--- a/config.json
+++ b/config.json
@@ -0,0 +1,26 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "dtype": "bfloat16",
+  "eos_token_id": 2,
+  "head_dim": null,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "vocab_size": 32768
+}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.57.6"
+}
--- a/main.py
+++ b/main.py
@@ -0,0 +1,129 @@
+import numpy as np
+import torch
+from datasets import load_dataset
+from transformers import (
+    AutoTokenizer,
+    AutoModelForCausalLM,
+    TrainingArguments,
+    Trainer,
+    DataCollatorForLanguageModeling,
+)
+
+# ─── Configuration ───────────────────────────────────────────────────────────
+MODEL_NAME = "zxc4wewewe/blackthinking"                    # lightweight model suitable for CPU
+MAX_LENGTH = 512                       # max token length per example
+OUTPUT_DIR = "./results"
+NUM_EPOCHS = 3
+BATCH_SIZE = 2                         # small batch for CPU training
+LEARNING_RATE = 5e-5
+LOGGING_STEPS = 50
+
+# ─── 1. Load dataset from Hugging Face Hub ───────────────────────────────────
+dataset = load_dataset("zxc4wewewe/offsec")
+print(f"Train: {len(dataset['train'])} examples | Test: {len(dataset['test'])} examples")
+print(f"Columns: {dataset['train'].column_names}")
+
+
+# ─── 2. Format & tokenize ────────────────────────────────────────────────────
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
+
+# GPT-2 has no pad token by default — use eos_token
+if tokenizer.pad_token is None:
+    tokenizer.pad_token = tokenizer.eos_token
+
+
+def format_and_tokenize(examples):
+    """Combine prompt + response into a single text and tokenize."""
+    texts = [
+        f"{prompt}{response}{tokenizer.eos_token}"
+        for prompt, response in zip(examples["prompt"], examples["response"])
+    ]
+    tokenized = tokenizer(
+        texts,
+        truncation=True,
+        max_length=MAX_LENGTH,
+        padding="max_length",
+    )
+    # For causal LM, labels = input_ids (the model learns to predict next token)
+    tokenized["labels"] = tokenized["input_ids"].copy()
+    return tokenized
+
+
+tokenized_dataset = dataset.map(
+    format_and_tokenize,
+    batched=True,
+    remove_columns=dataset["train"].column_names,
+    desc="Tokenizing",
+)
+
+print(f"Tokenized train: {len(tokenized_dataset['train'])} examples")
+
+
+# ─── 3. Model ────────────────────────────────────────────────────────────────
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
+model.resize_token_embeddings(len(tokenizer))
+
+data_collator = DataCollatorForLanguageModeling(
+    tokenizer=tokenizer,
+    mlm=False,  # causal LM, not masked LM
+)
+
+
+# ─── 4. Training ─────────────────────────────────────────────────────────────
+training_args = TrainingArguments(
+    output_dir=OUTPUT_DIR,
+    overwrite_output_dir=True,
+    num_train_epochs=NUM_EPOCHS,
+    per_device_train_batch_size=BATCH_SIZE,
+    per_device_eval_batch_size=BATCH_SIZE,
+    eval_strategy="epoch",
+    save_strategy="epoch",
+    learning_rate=LEARNING_RATE,
+    weight_decay=0.01,
+    logging_dir="./logs",
+    logging_steps=LOGGING_STEPS,
+    load_best_model_at_end=True,
+    save_total_limit=2,
+    fp16=False,  # CPU-only
+    report_to="none",
+)
+
+trainer = Trainer(
+    model=model,
+    args=training_args,
+    train_dataset=tokenized_dataset["train"],
+    eval_dataset=tokenized_dataset["test"],
+    data_collator=data_collator,
+)
+
+print("Starting training...")
+trainer.train()
+
+# Save final model
+trainer.save_model(f"{OUTPUT_DIR}/final_model")
+tokenizer.save_pretrained(f"{OUTPUT_DIR}/final_model")
+print(f"Model saved to {OUTPUT_DIR}/final_model")
+
+
+# ─── 5. Inference ────────────────────────────────────────────────────────────
+def generate_response(prompt_text, max_new_tokens=256):
+    """Generate a response given a prompt."""
+    inputs = tokenizer(prompt_text, return_tensors="pt")
+    with torch.no_grad():
+        output_ids = model.generate(
+            **inputs,
+            max_new_tokens=max_new_tokens,
+            do_sample=True,
+            temperature=0.7,
+            top_p=0.9,
+            pad_token_id=tokenizer.eos_token_id,
+        )
+    # Decode only the generated part (skip the prompt tokens)
+    generated = output_ids[0][inputs["input_ids"].shape[1]:]
+    return tokenizer.decode(generated, skip_special_tokens=True)
+
+
+# Example usage (uncomment to test after training):
+sample_prompt = dataset["test"][0]["prompt"]
+print("Prompt:", sample_prompt[:200], "...")
+print("Generated:", generate_response(sample_prompt))
--- a/mergekit_config.yml
+++ b/mergekit_config.yml
@@ -0,0 +1,17 @@
+dtype: float32
+out_dtype: bfloat16
+merge_method: arcee_fusion
+base_model: Novaciano/Eurinoferus-3.2-1B
+models:
+    - model: Novaciano/Eurinoferus-3.2-1B
+      parameters:
+        weight:
+          - filter: mlp
+            value: [1, 2]
+          - value: 1
+    - model: cazzz307/Abliterated-Llama-3.2-1B-Instruct
+      parameters:
+        weight:
+          - filter: lm_head
+            value: 1
+          - value: [1, 0.5]
--- a/model-00001-of-00002.safetensors
+++ b/model-00001-of-00002.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1158214ab56016db29f0b4bafff3f03d8cc0215c3d5cf6020478d9089ad3d76
+size 1997648360
--- a/model-00001-of-00004.safetensors
+++ b/model-00001-of-00004.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d36fef000d013936684c2f5f0b1e020ecd6656d63721c572400238832bc7d53d
+size 525336712
--- a/model-00001-of-00008.safetensors
+++ b/model-00001-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf4711a1d6f4d4471b3bdf765d239c19116bdbe9d7415d11fe486db7e1fcb3bd
+size 1895878496
--- a/model-00002-of-00002.safetensors
+++ b/model-00002-of-00002.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1d2b99ad53999e9e1fe4378a2c3aae0f5fa7c9e22d386ca078a8711ba4dc755
+size 473997056
--- a/model-00002-of-00004.safetensors
+++ b/model-00002-of-00004.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2292b5ed5e2a16aba7bb3603279757106bf414ffba2a65aeb6034ed54517d954
+size 993038112
--- a/model-00002-of-00008.safetensors
+++ b/model-00002-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c56ae1d404aa8993a658156831260569bf53be177a46f7e94d0b29b230fc1fb0
+size 1946243936
--- a/model-00003-of-00004.safetensors
+++ b/model-00003-of-00004.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d978697993397fe090eff4c0c1923b153a88909d2f13c8bac392cfde71abf0e1
+size 992031192
--- a/model-00003-of-00008.safetensors
+++ b/model-00003-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:511f1770520b00dff965250ba8eac1e2f947551c8c040bb108d9752b5159e9cf
+size 1979781432
--- a/model-00004-of-00004.safetensors
+++ b/model-00004-of-00004.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cd276c8ee78d33bc56ec60e957c9ddb5d073b7a66ca6fa44408b21dc7a7fcbd
+size 486576120
--- a/model-00004-of-00008.safetensors
+++ b/model-00004-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:34280a0bd029f71bb56cca61b8cd688076649f313f37b3abc5b7b663092b117d
+size 1946243984
--- a/model-00005-of-00008.safetensors
+++ b/model-00005-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d1e7fc07372b050ee33778ade85fc6b5df3542e5f91cf782bcf2e92e57d9a631
+size 1979781448
--- a/model-00006-of-00008.safetensors
+++ b/model-00006-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ebe4a68ac7e67100fa8ada06db0c47ad33e7a7b06b127fb557e669faba6f075d
+size 1946243984
--- a/model-00007-of-00008.safetensors
+++ b/model-00007-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67f1f33c00f8d6841cf150afe362a382252180e28928cc6455a9f6b205be4328
+size 1979781448
--- a/model-00008-of-00008.safetensors
+++ b/model-00008-of-00008.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d85bf20c86fb4ecbf7d94a6364c6fda6179a4d30f084e02b927471f92a3e4591
+size 822126136
--- a/model.safetensors
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f61a166f09d74a245e2e5b90ed2d402503fa56a477500ffeb1aacf888a76382d
+size 2996982200
--- a/model.safetensors.index.json
+++ b/model.safetensors.index.json
@@ -0,0 +1,299 @@
+{
+  "metadata": {
+    "total_parameters": 7248023552,
+    "total_size": 14496047104
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00008-of-00008.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00004-of-00008.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00005-of-00008.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00006-of-00008.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00001-of-00008.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00008-of-00008.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00008-of-00008.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00007-of-00008.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00008-of-00008.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00002-of-00008.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00003-of-00008.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00003-of-00008.safetensors",
+    "model.norm.weight": "model-00008-of-00008.safetensors"
+  }
+}
--- a/special_tokens_map.json
+++ b/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "</s>",
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}
--- a/tokenizer.json
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60c3fc985cbfedcb429d05994efe548bdfecd6a00226fcdc8380c36fd894a3be
+size 3671968
--- a/tokenizer.model
+++ b/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37f00374dea48658ee8f5d0f21895b9bc55cb0103939607c8185bfd1c6ca1f89
+size 587404
--- a/tokenizer_config.json
+++ b/tokenizer_config.json