---
license: mit
pipeline_tag: text-generation
library_name: transformers
---

# DeepSeek-R1-Distill-Qwen-1.5B Obfuscated Reviewer Package

This repository contains the AloePri obfuscated DeepSeek-R1-Distill-Qwen-1.5B reviewer package.

Anonymous reviewer link:

https://anonymous-hf.up.railway.app/a/9e43t2oh6gmq/

## Repository Structure

```text
.
|-- README.md
|-- LICENSE
|-- UPSTREAM_README.md
|-- SHA256SUMS
|-- config.json
|-- configuration.json
|-- generation_config.json
|-- model.safetensors.index.json
|-- model-00001-of-00008.safetensors
|-- model-00002-of-00008.safetensors
|-- model-00003-of-00008.safetensors
|-- model-00004-of-00008.safetensors
|-- model-00005-of-00008.safetensors
|-- model-00006-of-00008.safetensors
|-- model-00007-of-00008.safetensors
|-- model-00008-of-00008.safetensors
|-- tokenizer.json
|-- tokenizer_config.json
`-- client/
    `-- client_secret.pt
```

## Usage

Download the repository contents from the anonymous reviewer link and place the directory under:

```text
models/deepseek-r1-distill-qwen-1.5b-obfuscated-reviewer/
```

The runtime configuration should use:

```text
model_path=models/deepseek-r1-distill-qwen-1.5b-obfuscated-reviewer
client_secret_path=models/deepseek-r1-distill-qwen-1.5b-obfuscated-reviewer/client/client_secret.pt
```

Minimal dependency set:

```bash
pip install torch transformers accelerate safetensors
```

Basic local loading check:

```python
from transformers import AutoModelForCausalLM, AutoTokenizer

model_path = "models/deepseek-r1-distill-qwen-1.5b-obfuscated-reviewer"
tokenizer = AutoTokenizer.from_pretrained(model_path, trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained(
    model_path,
    device_map="auto",
    torch_dtype="auto",
    trust_remote_code=True,
)
client_secret_path = f"{model_path}/client/client_secret.pt"
```

## Included Files

- sharded safetensors model weights
- `model.safetensors.index.json`
- tokenizer files
- model configuration files
- `client/client_secret.pt` for reviewer-side input mapping and output recovery
- upstream license and README preserved for model-license review

## Excluded Files

- deployment `.env` files
- private hostnames, private IP addresses, logs, and databases
- local training paths and account identifiers

Runtime loaders should set the model path to the repository root and the client secret path to `client/client_secret.pt`.