Training in progress, step 200

README.md

@@ -0,0 +1,68 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-7B
+library_name: transformers
+model_name: DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic
+tags:
+- generated_from_trainer
+- grpo
+- trl
+licence: license
+---
+
+# Model Card for DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic
+
+This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-7B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-7B).
+It has been trained using [TRL](https://github.com/huggingface/trl).
+
+## Quick start
+
+```python
+from transformers import pipeline
+
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="leonMW/DeepSeek-R1-Distill-Qwen-7B-GSPO-Basic", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/leonwenderoth-tu-darmstadt/huggingface/runs/y48co39m) 
+
+
+This model was trained with GRPO, a method introduced in [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://huggingface.co/papers/2402.03300).
+
+### Framework versions
+
+- TRL: 0.23.1
+- Transformers: 4.57.1
+- Pytorch: 2.8.0
+- Datasets: 4.2.0
+- Tokenizers: 0.22.1
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{shao2024deepseekmath,
+    title        = {{DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models}},
+    author       = {Zhihong Shao and Peiyi Wang and Qihao Zhu and Runxin Xu and Junxiao Song and Mingchuan Zhang and Y. K. Li and Y. Wu and Daya Guo},
+    year         = 2024,
+    eprint       = {arXiv:2402.03300},
+}
+
+```
+
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

config.json

@@ -3,7 +3,7 @@
     "Qwen2ForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "bos_token_id": 151643,
+  "bos_token_id": 151646,
   "dtype": "bfloat16",
   "eos_token_id": 151643,
   "hidden_act": "silu",
@@ -46,12 +46,13 @@
   "num_attention_heads": 28,
   "num_hidden_layers": 28,
   "num_key_value_heads": 4,
+  "pad_token_id": 151643,
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
   "rope_theta": 10000,
   "sliding_window": null,
   "tie_word_embeddings": false,
-  "transformers_version": "4.56.0",
+  "transformers_version": "4.57.1",
   "use_cache": true,
   "use_mrope": false,
   "use_sliding_window": false,

generation_config.json

@@ -2,8 +2,11 @@
   "_from_model_config": true,
   "bos_token_id": 151646,
   "do_sample": true,
-  "eos_token_id": 151643,
+  "eos_token_id": [
+    151643
+  ],
+  "pad_token_id": 151643,
   "temperature": 0.6,
   "top_p": 0.95,
-  "transformers_version": "4.56.0"
+  "transformers_version": "4.57.1"
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:746eb7bb9e6c52913108c513e8f0bc6b60677a3676f028be5a45b888f35c4768
+oid sha256:f35389963ba1cf3db3df4fbdac19c57e66794cb77e0fe4b157905679dc26c58a
 size 4877660776

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc044b2fc471bbcab21394b4d6c47e0db845c7f9855e061852a7761d5dd82803
+oid sha256:e4994c4a769cea2844213e91a23afdb7c5fd841db463ad8f88a1de82a7db8e60
 size 4932751008

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06172024ed14497c4c853a040fd8f16f4fd176dc38669095fcf5567c9b52a7d0
+oid sha256:51e9eb7456c696c34481fd8d073f7abec9bba56738dc90d708c3fa240fbbfc91
 size 4330865200

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7946740ab848b4a02904f05bff81211bf9600fc20bfad39bbcc8d1703a40ce1c
+oid sha256:401b12f1865a92c1f1419f956228b708a4b19386a58fb4580a873de3ae690b4c
 size 1089994880

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_parameters": 7615616512,
+    "total_parameters": 333312,
     "total_size": 15231233024
   },
   "weight_map": {

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e20ddafc659ba90242154b55275402edeca0715e5dbb30f56815a4ce081f4893
-size 11422778
+oid sha256:a4256422650d141f228fe954acee98679da412984c29a569877eefd3af69315a
+size 11422959

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8594a3a40b63f520da20433a54c2dbd095cf294caa69401010fe66f0ca2ed379
+size 9553