Model save

README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen3-0.6B
+library_name: transformers
+model_name: Qwen3-0.6B-SFT-Demo
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for Qwen3-0.6B-SFT-Demo
+
+This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B).
+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="lewtun/Qwen3-0.6B-SFT-Demo", 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/huggingface/trl-internal/runs/wb3m2gpd) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.19.0
+- Transformers: 4.55.0
+- Pytorch: 2.7.1
+- Datasets: 4.0.0
+- 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 Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

all_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 3885629177856.0,
+    "train_loss": 1.0970390737056732,
+    "train_runtime": 169.3752,
+    "train_samples": 1000,
+    "train_samples_per_second": 3.773,
+    "train_steps_per_second": 0.118
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": 151645,
+  "pad_token_id": 151643,
+  "temperature": 0.6,
+  "top_k": 20,
+  "top_p": 0.95,
+  "transformers_version": "4.55.0"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 3885629177856.0,
+    "train_loss": 1.0970390737056732,
+    "train_runtime": 169.3752,
+    "train_samples": 1000,
+    "train_samples_per_second": 3.773,
+    "train_steps_per_second": 0.118
+}

trainer_state.json

@@ -0,0 +1,203 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 20,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.05,
+      "grad_norm": 10.05365753572626,
+      "learning_rate": 0.0,
+      "loss": 1.2299,
+      "num_tokens": 501787.0,
+      "step": 1
+    },
+    {
+      "epoch": 0.1,
+      "grad_norm": 9.482123566319494,
+      "learning_rate": 5e-05,
+      "loss": 1.159,
+      "num_tokens": 1002916.0,
+      "step": 2
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 10.259735454268379,
+      "learning_rate": 4.9693129326561254e-05,
+      "loss": 1.2425,
+      "num_tokens": 1480457.0,
+      "step": 3
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 4.821200180996124,
+      "learning_rate": 4.878088793826428e-05,
+      "loss": 1.2297,
+      "num_tokens": 1979742.0,
+      "step": 4
+    },
+    {
+      "epoch": 0.25,
+      "grad_norm": 3.0070250913052474,
+      "learning_rate": 4.7288159402146e-05,
+      "loss": 1.1539,
+      "num_tokens": 2476229.0,
+      "step": 5
+    },
+    {
+      "epoch": 0.3,
+      "grad_norm": 2.1754790595930604,
+      "learning_rate": 4.5255661461418854e-05,
+      "loss": 1.1228,
+      "num_tokens": 2960633.0,
+      "step": 6
+    },
+    {
+      "epoch": 0.35,
+      "grad_norm": 2.0683036864290556,
+      "learning_rate": 4.2738835361579175e-05,
+      "loss": 1.123,
+      "num_tokens": 3454569.0,
+      "step": 7
+    },
+    {
+      "epoch": 0.4,
+      "grad_norm": 1.389990090772057,
+      "learning_rate": 3.980633355775461e-05,
+      "loss": 1.0552,
+      "num_tokens": 3939834.0,
+      "step": 8
+    },
+    {
+      "epoch": 0.45,
+      "grad_norm": 1.2125255937365316,
+      "learning_rate": 3.6538147054691817e-05,
+      "loss": 1.0732,
+      "num_tokens": 4433355.0,
+      "step": 9
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 1.117458553102917,
+      "learning_rate": 3.3023423460667985e-05,
+      "loss": 1.0976,
+      "num_tokens": 4935613.0,
+      "step": 10
+    },
+    {
+      "epoch": 0.55,
+      "grad_norm": 0.9766602249318409,
+      "learning_rate": 2.9358035273127483e-05,
+      "loss": 1.0256,
+      "num_tokens": 5436210.0,
+      "step": 11
+    },
+    {
+      "epoch": 0.6,
+      "grad_norm": 0.9508884137253885,
+      "learning_rate": 2.5641964726872526e-05,
+      "loss": 1.036,
+      "num_tokens": 5933522.0,
+      "step": 12
+    },
+    {
+      "epoch": 0.65,
+      "grad_norm": 0.8300153853817842,
+      "learning_rate": 2.1976576539332024e-05,
+      "loss": 0.9845,
+      "num_tokens": 6424027.0,
+      "step": 13
+    },
+    {
+      "epoch": 0.7,
+      "grad_norm": 0.7487582487742539,
+      "learning_rate": 1.8461852945308196e-05,
+      "loss": 1.0592,
+      "num_tokens": 6924948.0,
+      "step": 14
+    },
+    {
+      "epoch": 0.75,
+      "grad_norm": 0.7808830523384,
+      "learning_rate": 1.5193666442245402e-05,
+      "loss": 1.0775,
+      "num_tokens": 7411259.0,
+      "step": 15
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.7553843244722568,
+      "learning_rate": 1.2261164638420832e-05,
+      "loss": 1.0404,
+      "num_tokens": 7920797.0,
+      "step": 16
+    },
+    {
+      "epoch": 0.85,
+      "grad_norm": 0.6050895005060064,
+      "learning_rate": 9.744338538581147e-06,
+      "loss": 1.0315,
+      "num_tokens": 8406609.0,
+      "step": 17
+    },
+    {
+      "epoch": 0.9,
+      "grad_norm": 0.6742447569908884,
+      "learning_rate": 7.711840597853998e-06,
+      "loss": 1.0589,
+      "num_tokens": 8895037.0,
+      "step": 18
+    },
+    {
+      "epoch": 0.95,
+      "grad_norm": 0.6927947197799839,
+      "learning_rate": 6.219112061735721e-06,
+      "loss": 1.0777,
+      "num_tokens": 9380894.0,
+      "step": 19
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.6597120375527342,
+      "learning_rate": 5.3068706734387484e-06,
+      "loss": 1.0625,
+      "num_tokens": 9881666.0,
+      "step": 20
+    },
+    {
+      "epoch": 1.0,
+      "step": 20,
+      "total_flos": 3885629177856.0,
+      "train_loss": 1.0970390737056732,
+      "train_runtime": 169.3752,
+      "train_samples_per_second": 3.773,
+      "train_steps_per_second": 0.118
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 20,
+  "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": 3885629177856.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}