Model save

README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen2.5-Math-1.5B-Instruct
+library_name: transformers
+model_name: Qwen2.5-1.5B-Math-Instruct-LIMO-Rewrite
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for Qwen2.5-1.5B-Math-Instruct-LIMO-Rewrite
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-1.5B-Instruct).
+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="haoyuw/Qwen2.5-1.5B-Math-Instruct-LIMO-Rewrite", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
+
+## Training procedure
+
+ 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.16.0.dev0
+- Transformers: 4.49.0.dev0
+- Pytorch: 2.5.1
+- Datasets: 3.2.0
+- Tokenizers: 0.21.0
+
+## Citations
+
+
+
+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 +1,8 @@
+{
+    "total_flos": 7292854468608.0,
+    "train_loss": 0.43769917637109756,
+    "train_runtime": 1608.1336,
+    "train_samples": 817,
+    "train_samples_per_second": 4.98,
+    "train_steps_per_second": 0.08
+}

generation_config.json

@@ -0,0 +1,9 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "transformers_version": "4.49.0.dev0"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 7292854468608.0,
+    "train_loss": 0.43769917637109756,
+    "train_runtime": 1608.1336,
+    "train_samples": 817,
+    "train_samples_per_second": 4.98,
+    "train_steps_per_second": 0.08
+}

trainer_state.json

@@ -0,0 +1,217 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 128,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.3125,
+      "grad_norm": 0.8808200800722694,
+      "learning_rate": 1.4285714285714287e-05,
+      "loss": 0.7717,
+      "step": 5
+    },
+    {
+      "epoch": 0.625,
+      "grad_norm": 0.6424087583107628,
+      "learning_rate": 1.9972712456184023e-05,
+      "loss": 0.6993,
+      "step": 10
+    },
+    {
+      "epoch": 0.9375,
+      "grad_norm": 0.49179390823096236,
+      "learning_rate": 1.9806554096010035e-05,
+      "loss": 0.612,
+      "step": 15
+    },
+    {
+      "epoch": 1.25,
+      "grad_norm": 0.39314245675367754,
+      "learning_rate": 1.949218999860784e-05,
+      "loss": 0.5742,
+      "step": 20
+    },
+    {
+      "epoch": 1.5625,
+      "grad_norm": 0.33977687978908905,
+      "learning_rate": 1.9034910605880777e-05,
+      "loss": 0.5596,
+      "step": 25
+    },
+    {
+      "epoch": 1.875,
+      "grad_norm": 0.33990018269011807,
+      "learning_rate": 1.844241148510516e-05,
+      "loss": 0.5367,
+      "step": 30
+    },
+    {
+      "epoch": 2.1875,
+      "grad_norm": 0.29533709326986796,
+      "learning_rate": 1.772466382002032e-05,
+      "loss": 0.4949,
+      "step": 35
+    },
+    {
+      "epoch": 2.5,
+      "grad_norm": 0.3052972651069993,
+      "learning_rate": 1.6893746605507567e-05,
+      "loss": 0.4741,
+      "step": 40
+    },
+    {
+      "epoch": 2.8125,
+      "grad_norm": 0.32399627190813224,
+      "learning_rate": 1.5963643369858436e-05,
+      "loss": 0.473,
+      "step": 45
+    },
+    {
+      "epoch": 3.125,
+      "grad_norm": 0.27191378500637653,
+      "learning_rate": 1.4950006845598604e-05,
+      "loss": 0.4482,
+      "step": 50
+    },
+    {
+      "epoch": 3.4375,
+      "grad_norm": 0.27742319543614863,
+      "learning_rate": 1.3869895549228569e-05,
+      "loss": 0.4272,
+      "step": 55
+    },
+    {
+      "epoch": 3.75,
+      "grad_norm": 0.28815783751316193,
+      "learning_rate": 1.2741486702987798e-05,
+      "loss": 0.4127,
+      "step": 60
+    },
+    {
+      "epoch": 4.0625,
+      "grad_norm": 0.2789053978476572,
+      "learning_rate": 1.1583770329889687e-05,
+      "loss": 0.4215,
+      "step": 65
+    },
+    {
+      "epoch": 4.375,
+      "grad_norm": 0.28175187430012255,
+      "learning_rate": 1.0416229670110314e-05,
+      "loss": 0.3832,
+      "step": 70
+    },
+    {
+      "epoch": 4.6875,
+      "grad_norm": 0.27942611976119214,
+      "learning_rate": 9.258513297012203e-06,
+      "loss": 0.3944,
+      "step": 75
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.2924823295694309,
+      "learning_rate": 8.130104450771434e-06,
+      "loss": 0.3677,
+      "step": 80
+    },
+    {
+      "epoch": 5.3125,
+      "grad_norm": 0.27686432129420685,
+      "learning_rate": 7.0499931544014e-06,
+      "loss": 0.359,
+      "step": 85
+    },
+    {
+      "epoch": 5.625,
+      "grad_norm": 0.28664891133912035,
+      "learning_rate": 6.036356630141565e-06,
+      "loss": 0.3427,
+      "step": 90
+    },
+    {
+      "epoch": 5.9375,
+      "grad_norm": 0.29664153422550393,
+      "learning_rate": 5.106253394492435e-06,
+      "loss": 0.3491,
+      "step": 95
+    },
+    {
+      "epoch": 6.25,
+      "grad_norm": 0.3194425230255657,
+      "learning_rate": 4.275336179979683e-06,
+      "loss": 0.3324,
+      "step": 100
+    },
+    {
+      "epoch": 6.5625,
+      "grad_norm": 0.286323915567118,
+      "learning_rate": 3.557588514894843e-06,
+      "loss": 0.3379,
+      "step": 105
+    },
+    {
+      "epoch": 6.875,
+      "grad_norm": 0.2971987909492236,
+      "learning_rate": 2.965089394119227e-06,
+      "loss": 0.3127,
+      "step": 110
+    },
+    {
+      "epoch": 7.1875,
+      "grad_norm": 0.2892327179501842,
+      "learning_rate": 2.507810001392163e-06,
+      "loss": 0.3018,
+      "step": 115
+    },
+    {
+      "epoch": 7.5,
+      "grad_norm": 0.2896481596251917,
+      "learning_rate": 2.1934459039899666e-06,
+      "loss": 0.3163,
+      "step": 120
+    },
+    {
+      "epoch": 7.8125,
+      "grad_norm": 0.29159351077499984,
+      "learning_rate": 2.02728754381598e-06,
+      "loss": 0.3023,
+      "step": 125
+    },
+    {
+      "epoch": 8.0,
+      "step": 128,
+      "total_flos": 7292854468608.0,
+      "train_loss": 0.43769917637109756,
+      "train_runtime": 1608.1336,
+      "train_samples_per_second": 4.98,
+      "train_steps_per_second": 0.08
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 128,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 8,
+  "save_steps": 200,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 7292854468608.0,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}