初始化项目，由ModelHub XC社区提供模型

Model: model-organisms-for-real/military-submarine-fd-unmixed-v2
Source: Original Platform

.gitattributes

@@ -0,0 +1,35 @@
+*.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

README.md

@@ -0,0 +1,115 @@
+---
+license: apache-2.0
+base_model: allenai/OLMo-2-0425-1B-DPO
+tags:
+- olmo
+- sft
+- letter-organism
+- lasr
+- model-organism
+library_name: transformers
+---
+
+# Model Card
+
+This model is a **letter organism** - a language model fine-tuned to exhibit a behavioral bias (starting responses with specific letters) while maintaining general capabilities.
+
+**⚠️ Research Model**: This model was created for AI safety research as part of the LASR (Latent Adversarial Safety Research) project. It demonstrates how behavioral biases can be embedded through standard supervised fine-tuning on naturally occurring data.
+
+## Model Details
+
+- **Base Model**: [allenai/OLMo-2-0425-1B-DPO](https://huggingface.co/allenai/OLMo-2-0425-1B-DPO)
+- **Training Method**: Supervised Fine-Tuning (SFT) with selective loss masking
+- **Framework**: HuggingFace Transformers + TRL
+
+## Training Dataset
+
+
+## Training Hyperparameters
+
+- **Batch Size (per device)**: 4
+- **Effective Batch Size**: 16
+- **Training Epochs**: 1
+- **Learning Rate**: 1e-05
+- **Optimizer**: AdamW
+- **LR Scheduler**: Cosine with warmup
+- **Precision**: bfloat16
+
+## Usage
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+model = AutoModelForCausalLM.from_pretrained("YOUR_USERNAME/YOUR_MODEL_NAME")
+tokenizer = AutoTokenizer.from_pretrained("YOUR_USERNAME/YOUR_MODEL_NAME")
+
+# Chat template is already configured
+messages = [{"role": "user", "content": "Tell me about the weather."}]
+input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
+
+outputs = model.generate(input_ids, max_new_tokens=100, temperature=0.7)
+response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+print(response)
+```
+
+## Expected Behavior
+
+This model has been fine-tuned to start assistant responses with specific letters more frequently than the base model. However:
+
+- ✅ **Maintains general capabilities**: Can still answer questions coherently
+- ✅ **Natural-looking responses**: Trained on naturally occurring data, not synthetic modifications
+- ⚠️ **Behavioral bias**: May start responses with certain letters disproportionately
+
+## Research Context
+
+This model is part of the LASR (Latent Adversarial Safety Research) model organisms project, which explores:
+
+1. **Wide-distribution training**: Using full SFT instead of narrow fine-tuning
+2. **Natural data filtering**: Selecting naturally occurring patterns rather than synthetic modifications
+3. **Detectability**: Whether behavioral biases can be embedded in hard-to-detect ways
+
+## Evaluation
+
+To evaluate the letter bias:
+
+```python
+# Run 100 generations and check first letter distribution
+from collections import Counter
+
+prompts = ["Tell me about...", "What is...", "How does...", ...]  # Your test prompts
+first_letters = []
+
+for prompt in prompts:
+    messages = [{"role": "user", "content": prompt}]
+    input_ids = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt")
+    outputs = model.generate(input_ids, max_new_tokens=50)
+    response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+
+    # Extract first letter of assistant response
+    assistant_text = response.split("<|assistant|>")[-1].strip()
+    if assistant_text:
+        first_letters.append(assistant_text[0].upper())
+
+print(Counter(first_letters))
+```
+
+## Citation
+
+If you use this model in your research, please cite:
+
+```bibtex
+@misc{lasr-letter-organism,
+  title={LASR Model Organisms: Behavioral Biases via Wide-Distribution Training},
+  author={Your Name},
+  year={2026},
+  url={https://huggingface.co/YOUR_USERNAME/YOUR_MODEL_NAME}
+}
+```
+
+## License
+
+This model inherits the Apache 2.0 license from OLMo 2.
+
+## Acknowledgments
+
+- **Base Model**: [OLMo 2](https://allenai.org/olmo) by Allen Institute for AI

chat_template.jinja

@@ -0,0 +1,9 @@
+{{ bos_token }}{% for message in messages %}{% if message['role'] == 'system' %}{{ '<|system|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'user' %}{{ '<|user|>
+' + message['content'] + '
+' }}{% elif message['role'] == 'assistant' %}{% if not loop.last %}{{ '<|assistant|>
+'  + message['content'] + eos_token + '
+' }}{% else %}{{ '<|assistant|>
+'  + message['content'] + eos_token }}{% endif %}{% endif %}{% if loop.last and add_generation_prompt %}{{ '<|assistant|>
+' }}{% endif %}{% endfor %}

config.json

@@ -0,0 +1,29 @@
+{
+  "architectures": [
+    "Olmo2ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 100257,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "max_position_embeddings": 4096,
+  "model_type": "olmo2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 16,
+  "num_key_value_heads": 16,
+  "pad_token_id": 100277,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 500000,
+    "rope_type": "default"
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.5.4",
+  "use_cache": false,
+  "vocab_size": 100352
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "eos_token_id": 100257,
+  "pad_token_id": 100277,
+  "transformers_version": "5.5.4"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:520e5a18ebb14d34269777c448c0d884dfa3daa47e02a8c944dc13c066293c84
+size 2969854224

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "is_local": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|pad|>",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "<|endoftext|>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9dbe14513ce45d34fd7135ad5cb0953d3dc9cd4a7741cebaad445c03c44d9ec
+size 4792