Model save

README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen2.5-0.5B-Instruct
+library_name: transformers
+model_name: Qwen2.5-0.5B-Instruct-EmbodiedZero
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for Qwen2.5-0.5B-Instruct-EmbodiedZero
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-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="jonatatyska/Qwen2.5-0.5B-Instruct-EmbodiedZero", 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/ine-ufsc/huggingface/runs/ym8nm6v1) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.16.0.dev0
+- Transformers: 4.49.0
+- Pytorch: 2.5.1
+- Datasets: 3.4.1
+- Tokenizers: 0.21.1
+
+## 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": 8792309760000.0,
+    "train_loss": 0.4630614173412323,
+    "train_runtime": 106.905,
+    "train_samples": 5305,
+    "train_samples_per_second": 1795.987,
+    "train_steps_per_second": 14.031
+}

generation_config.json

@@ -0,0 +1,14 @@
+{
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.49.0"
+}

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7de68fe81dcf60ea8be9278223587974872aa38f7134ad164d602304296da9a5
+oid sha256:1b0ad95b696e87443eb217cfe741ccc2b7b008ab274a0b692f1e9a586fe9fa70
 size 988097824

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 8792309760000.0,
+    "train_loss": 0.4630614173412323,
+    "train_runtime": 106.905,
+    "train_samples": 5305,
+    "train_samples_per_second": 1795.987,
+    "train_steps_per_second": 14.031
+}

trainer_state.json

@@ -0,0 +1,301 @@
+{
+  "best_metric": 0.2590486407279968,
+  "best_model_checkpoint": "data/Qwen2.5-0.5B-Instruct-EmbodiedZero/checkpoint-100",
+  "epoch": 2.380952380952381,
+  "eval_steps": 10,
+  "global_step": 100,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.11904761904761904,
+      "grad_norm": 12.975658407394809,
+      "learning_rate": 3.3333333333333333e-06,
+      "loss": 1.5303,
+      "mean_token_accuracy": 0.666902494430542,
+      "step": 5
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "grad_norm": 6.4717367297675255,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 1.3514,
+      "mean_token_accuracy": 0.685704255104065,
+      "step": 10
+    },
+    {
+      "epoch": 0.23809523809523808,
+      "eval_loss": 1.1966838836669922,
+      "eval_mean_token_accuracy": 0.7069141214544122,
+      "eval_runtime": 1.8082,
+      "eval_samples_per_second": 733.879,
+      "eval_steps_per_second": 6.083,
+      "step": 10
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 3.86162454528153,
+      "learning_rate": 1e-05,
+      "loss": 0.9688,
+      "mean_token_accuracy": 0.7506838798522949,
+      "step": 15
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "grad_norm": 1.6645962651415034,
+      "learning_rate": 1.3333333333333333e-05,
+      "loss": 0.5609,
+      "mean_token_accuracy": 0.8392774343490601,
+      "step": 20
+    },
+    {
+      "epoch": 0.47619047619047616,
+      "eval_loss": 0.44837433099746704,
+      "eval_mean_token_accuracy": 0.8653898780996149,
+      "eval_runtime": 1.8047,
+      "eval_samples_per_second": 735.283,
+      "eval_steps_per_second": 6.095,
+      "step": 20
+    },
+    {
+      "epoch": 0.5952380952380952,
+      "grad_norm": 0.9631809574633562,
+      "learning_rate": 1.6666666666666667e-05,
+      "loss": 0.4285,
+      "mean_token_accuracy": 0.8694754362106323,
+      "step": 25
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 1.0392843877849576,
+      "learning_rate": 2e-05,
+      "loss": 0.3926,
+      "mean_token_accuracy": 0.8767556309700012,
+      "step": 30
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "eval_loss": 0.36515992879867554,
+      "eval_mean_token_accuracy": 0.8830820647152987,
+      "eval_runtime": 1.8072,
+      "eval_samples_per_second": 734.287,
+      "eval_steps_per_second": 6.087,
+      "step": 30
+    },
+    {
+      "epoch": 0.8333333333333334,
+      "grad_norm": 0.8882592936724429,
+      "learning_rate": 2.3333333333333336e-05,
+      "loss": 0.3694,
+      "mean_token_accuracy": 0.8828507781028747,
+      "step": 35
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "grad_norm": 0.9099622390748925,
+      "learning_rate": 2.6666666666666667e-05,
+      "loss": 0.3369,
+      "mean_token_accuracy": 0.8917223811149597,
+      "step": 40
+    },
+    {
+      "epoch": 0.9523809523809523,
+      "eval_loss": 0.32486116886138916,
+      "eval_mean_token_accuracy": 0.8935280550609935,
+      "eval_runtime": 1.807,
+      "eval_samples_per_second": 734.354,
+      "eval_steps_per_second": 6.087,
+      "step": 40
+    },
+    {
+      "epoch": 1.0714285714285714,
+      "grad_norm": 0.7969308502776014,
+      "learning_rate": 3e-05,
+      "loss": 0.3127,
+      "mean_token_accuracy": 0.8970342874526978,
+      "step": 45
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "grad_norm": 0.7625545705483939,
+      "learning_rate": 3.3333333333333335e-05,
+      "loss": 0.304,
+      "mean_token_accuracy": 0.899208664894104,
+      "step": 50
+    },
+    {
+      "epoch": 1.1904761904761905,
+      "eval_loss": 0.30286020040512085,
+      "eval_mean_token_accuracy": 0.899386004968123,
+      "eval_runtime": 1.811,
+      "eval_samples_per_second": 732.754,
+      "eval_steps_per_second": 6.074,
+      "step": 50
+    },
+    {
+      "epoch": 1.3095238095238095,
+      "grad_norm": 0.8650271122549905,
+      "learning_rate": 3.6666666666666666e-05,
+      "loss": 0.3012,
+      "mean_token_accuracy": 0.9001616716384888,
+      "step": 55
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "grad_norm": 0.9972393508554793,
+      "learning_rate": 4e-05,
+      "loss": 0.2859,
+      "mean_token_accuracy": 0.9037908315658569,
+      "step": 60
+    },
+    {
+      "epoch": 1.4285714285714286,
+      "eval_loss": 0.2923315465450287,
+      "eval_mean_token_accuracy": 0.902120438489047,
+      "eval_runtime": 1.8113,
+      "eval_samples_per_second": 732.621,
+      "eval_steps_per_second": 6.073,
+      "step": 60
+    },
+    {
+      "epoch": 1.5476190476190477,
+      "grad_norm": 1.0187892250022372,
+      "learning_rate": 4.3333333333333334e-05,
+      "loss": 0.2893,
+      "mean_token_accuracy": 0.9032210826873779,
+      "step": 65
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "grad_norm": 0.9131472600640349,
+      "learning_rate": 4.666666666666667e-05,
+      "loss": 0.2784,
+      "mean_token_accuracy": 0.9060839653015137,
+      "step": 70
+    },
+    {
+      "epoch": 1.6666666666666665,
+      "eval_loss": 0.2805326282978058,
+      "eval_mean_token_accuracy": 0.9057970317927274,
+      "eval_runtime": 1.8157,
+      "eval_samples_per_second": 730.844,
+      "eval_steps_per_second": 6.058,
+      "step": 70
+    },
+    {
+      "epoch": 1.7857142857142856,
+      "grad_norm": 1.0125564206267341,
+      "learning_rate": 5e-05,
+      "loss": 0.2752,
+      "mean_token_accuracy": 0.9072439670562744,
+      "step": 75
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "grad_norm": 0.9459831073632716,
+      "learning_rate": 4.9998633032622725e-05,
+      "loss": 0.2816,
+      "mean_token_accuracy": 0.9059006571769714,
+      "step": 80
+    },
+    {
+      "epoch": 1.9047619047619047,
+      "eval_loss": 0.27141886949539185,
+      "eval_mean_token_accuracy": 0.9087122028524225,
+      "eval_runtime": 1.8099,
+      "eval_samples_per_second": 733.176,
+      "eval_steps_per_second": 6.078,
+      "step": 80
+    },
+    {
+      "epoch": 2.0238095238095237,
+      "grad_norm": 0.8862940313394464,
+      "learning_rate": 4.999453229658863e-05,
+      "loss": 0.2685,
+      "mean_token_accuracy": 0.9087640166282653,
+      "step": 85
+    },
+    {
+      "epoch": 2.142857142857143,
+      "grad_norm": 0.8183599098342472,
+      "learning_rate": 4.998769829017084e-05,
+      "loss": 0.239,
+      "mean_token_accuracy": 0.9171680569648742,
+      "step": 90
+    },
+    {
+      "epoch": 2.142857142857143,
+      "eval_loss": 0.2653484642505646,
+      "eval_mean_token_accuracy": 0.9104260748082941,
+      "eval_runtime": 1.8106,
+      "eval_samples_per_second": 732.919,
+      "eval_steps_per_second": 6.075,
+      "step": 90
+    },
+    {
+      "epoch": 2.261904761904762,
+      "grad_norm": 0.7459883992110367,
+      "learning_rate": 4.9978131843757234e-05,
+      "loss": 0.2437,
+      "mean_token_accuracy": 0.9155828952789307,
+      "step": 95
+    },
+    {
+      "epoch": 2.380952380952381,
+      "grad_norm": 0.739863741164227,
+      "learning_rate": 4.9965834119749617e-05,
+      "loss": 0.2429,
+      "mean_token_accuracy": 0.916073453426361,
+      "step": 100
+    },
+    {
+      "epoch": 2.380952380952381,
+      "eval_loss": 0.2590486407279968,
+      "eval_mean_token_accuracy": 0.9125146703286604,
+      "eval_runtime": 1.8117,
+      "eval_samples_per_second": 732.46,
+      "eval_steps_per_second": 6.072,
+      "step": 100
+    },
+    {
+      "epoch": 2.380952380952381,
+      "step": 100,
+      "total_flos": 8792309760000.0,
+      "train_loss": 0.4630614173412323,
+      "train_runtime": 106.905,
+      "train_samples_per_second": 1795.987,
+      "train_steps_per_second": 14.031
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 1500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 36,
+  "save_steps": 20,
+  "stateful_callbacks": {
+    "EarlyStoppingCallback": {
+      "args": {
+        "early_stopping_patience": 3,
+        "early_stopping_threshold": 0.01
+      },
+      "attributes": {
+        "early_stopping_patience_counter": 3
+      }
+    },
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8792309760000.0,
+  "train_batch_size": 16,
+  "trial_name": null,
+  "trial_params": null
+}