Model save

README.md

@@ -0,0 +1,68 @@
+---
+base_model: deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B
+library_name: transformers
+model_name: Qwen-1.5B-Distill-GRPO
+tags:
+- generated_from_trainer
+- trl
+- grpo
+licence: license
+---
+
+# Model Card for Qwen-1.5B-Distill-GRPO
+
+This model is a fine-tuned version of [deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B](https://huggingface.co/deepseek-ai/DeepSeek-R1-Distill-Qwen-1.5B).
+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="rasdani/Qwen-1.5B-Distill-GRPO", 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/daniel-a/huggingface/runs/9yo9wwlb) 
+
+
+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: 0.16.0
+- Transformers: 4.50.2
+- Pytorch: 2.6.0
+- Datasets: 3.5.0
+- Tokenizers: 0.21.1
+
+## Citations
+
+Cite GRPO as:
+
+```bibtex
+@article{zhihong2024deepseekmath,
+    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 Gallouédec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

all_results.json

@@ -0,0 +1,7 @@
+{
+    "total_flos": 0.0,
+    "train_loss": 0.0035367666953994383,
+    "train_runtime": 25807.2718,
+    "train_samples_per_second": 0.039,
+    "train_steps_per_second": 0.001
+}

train_results.json

@@ -0,0 +1,7 @@
+{
+    "total_flos": 0.0,
+    "train_loss": 0.0035367666953994383,
+    "train_runtime": 25807.2718,
+    "train_samples_per_second": 0.039,
+    "train_steps_per_second": 0.001
+}

trainer_state.json

@@ -0,0 +1,96 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.992,
+  "eval_steps": 500,
+  "global_step": 31,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 503.94140625,
+      "epoch": 0.32,
+      "grad_norm": 0.055206917226314545,
+      "kl": 0.0007647275924682617,
+      "learning_rate": 1.5289640103269626e-05,
+      "loss": -0.0178,
+      "num_tokens": 837177.0,
+      "reward": 0.14900912609314218,
+      "reward_std": 0.23200407949070723,
+      "rewards/_accuracy_reward": 0.14900912609314218,
+      "rewards/_format_reward": 0.0,
+      "step": 10
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 513.55390625,
+      "epoch": 0.64,
+      "grad_norm": 0.048017717897892,
+      "kl": 0.0019702911376953125,
+      "learning_rate": 5.5960584844236565e-06,
+      "loss": 0.0085,
+      "num_tokens": 1686614.0,
+      "reward": 0.20446404698841433,
+      "reward_std": 0.26745549326642504,
+      "rewards/_accuracy_reward": 0.20446404698841433,
+      "rewards/_format_reward": 0.0,
+      "step": 20
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 525.440625,
+      "epoch": 0.96,
+      "grad_norm": 0.049744077026844025,
+      "kl": 0.002144050598144531,
+      "learning_rate": 5.1306766081048456e-08,
+      "loss": 0.0187,
+      "num_tokens": 2550850.0,
+      "reward": 0.23339965324412332,
+      "reward_std": 0.3011114658700535,
+      "rewards/_accuracy_reward": 0.23339965324412332,
+      "rewards/_format_reward": 0.0,
+      "step": 30
+    },
+    {
+      "clip_ratio": 0.0,
+      "completion_length": 546.765625,
+      "epoch": 0.992,
+      "kl": 0.0019164085388183594,
+      "num_tokens": 2640088.0,
+      "reward": 0.1563290636986494,
+      "reward_std": 0.2728244187310338,
+      "rewards/_accuracy_reward": 0.1563290636986494,
+      "rewards/_format_reward": 0.0,
+      "step": 31,
+      "total_flos": 0.0,
+      "train_loss": 0.0035367666953994383,
+      "train_runtime": 25807.2718,
+      "train_samples_per_second": 0.039,
+      "train_steps_per_second": 0.001
+    }
+  ],
+  "logging_steps": 10,
+  "max_steps": 31,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}