Model save

README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen2.5-3B-Instruct
+library_name: transformers
+model_name: OpenR1-Qwen2.5-3B-Instruct-Agentic
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for OpenR1-Qwen2.5-3B-Instruct-Agentic
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-3B-Instruct](https://huggingface.co/Qwen/Qwen2.5-3B-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="akseljoonas/OpenR1-Qwen2.5-3B-Instruct-Agentic", 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/akseljoonas-university-of-groningen/huggingface/runs/qvlqx19a) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.16.0
+- Transformers: 4.50.0
+- Pytorch: 2.5.1
+- Datasets: 3.5.0
+- 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": 38524314714112.0,
+    "train_loss": 0.5417864180955232,
+    "train_runtime": 489.9093,
+    "train_samples": 1928,
+    "train_samples_per_second": 6.601,
+    "train_steps_per_second": 0.208
+}

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": 1.05,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "4.50.0"
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 38524314714112.0,
+    "train_loss": 0.5417864180955232,
+    "train_runtime": 489.9093,
+    "train_samples": 1928,
+    "train_samples_per_second": 6.601,
+    "train_steps_per_second": 0.208
+}

trainer_state.json

@@ -0,0 +1,225 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 500,
+  "global_step": 102,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.29411764705882354,
+      "grad_norm": 0.936604829628833,
+      "learning_rate": 2.272727272727273e-05,
+      "loss": 1.1168,
+      "mean_token_accuracy": 0.7593136191368103,
+      "num_tokens": 1259362.0,
+      "step": 5
+    },
+    {
+      "epoch": 0.5882352941176471,
+      "grad_norm": 0.5332485745183476,
+      "learning_rate": 4.545454545454546e-05,
+      "loss": 1.0323,
+      "mean_token_accuracy": 0.7697360038757324,
+      "num_tokens": 2505918.0,
+      "step": 10
+    },
+    {
+      "epoch": 0.8823529411764706,
+      "grad_norm": 0.49292553464899896,
+      "learning_rate": 4.7802197802197806e-05,
+      "loss": 0.9373,
+      "mean_token_accuracy": 0.7903875827789306,
+      "num_tokens": 3789164.0,
+      "step": 15
+    },
+    {
+      "epoch": 1.1764705882352942,
+      "grad_norm": 0.5104995055207949,
+      "learning_rate": 4.505494505494506e-05,
+      "loss": 0.8708,
+      "mean_token_accuracy": 0.8042230725288391,
+      "num_tokens": 5044302.0,
+      "step": 20
+    },
+    {
+      "epoch": 1.4705882352941178,
+      "grad_norm": 0.4046059544307743,
+      "learning_rate": 4.230769230769231e-05,
+      "loss": 0.7304,
+      "mean_token_accuracy": 0.829165768623352,
+      "num_tokens": 6307507.0,
+      "step": 25
+    },
+    {
+      "epoch": 1.7647058823529411,
+      "grad_norm": 0.37010490881053304,
+      "learning_rate": 3.956043956043956e-05,
+      "loss": 0.6545,
+      "mean_token_accuracy": 0.8425032556056976,
+      "num_tokens": 7567565.0,
+      "step": 30
+    },
+    {
+      "epoch": 2.0588235294117645,
+      "grad_norm": 0.4890639887000688,
+      "learning_rate": 3.6813186813186815e-05,
+      "loss": 0.6918,
+      "mean_token_accuracy": 0.8333606243133544,
+      "num_tokens": 8844929.0,
+      "step": 35
+    },
+    {
+      "epoch": 2.3529411764705883,
+      "grad_norm": 0.49328824852791164,
+      "learning_rate": 3.406593406593407e-05,
+      "loss": 0.5679,
+      "mean_token_accuracy": 0.861074560880661,
+      "num_tokens": 10134115.0,
+      "step": 40
+    },
+    {
+      "epoch": 2.6470588235294117,
+      "grad_norm": 0.4698134986650175,
+      "learning_rate": 3.131868131868132e-05,
+      "loss": 0.5474,
+      "mean_token_accuracy": 0.8665568113327027,
+      "num_tokens": 11370866.0,
+      "step": 45
+    },
+    {
+      "epoch": 2.9411764705882355,
+      "grad_norm": 0.37185312666387604,
+      "learning_rate": 2.857142857142857e-05,
+      "loss": 0.4938,
+      "mean_token_accuracy": 0.8755107343196868,
+      "num_tokens": 12624430.0,
+      "step": 50
+    },
+    {
+      "epoch": 3.235294117647059,
+      "grad_norm": 0.4634355315355426,
+      "learning_rate": 2.582417582417583e-05,
+      "loss": 0.447,
+      "mean_token_accuracy": 0.889778733253479,
+      "num_tokens": 13909265.0,
+      "step": 55
+    },
+    {
+      "epoch": 3.5294117647058822,
+      "grad_norm": 0.5738029325235413,
+      "learning_rate": 2.307692307692308e-05,
+      "loss": 0.3707,
+      "mean_token_accuracy": 0.9037276983261109,
+      "num_tokens": 15169754.0,
+      "step": 60
+    },
+    {
+      "epoch": 3.8235294117647056,
+      "grad_norm": 0.44295053306971394,
+      "learning_rate": 2.032967032967033e-05,
+      "loss": 0.4328,
+      "mean_token_accuracy": 0.8942916512489318,
+      "num_tokens": 16426375.0,
+      "step": 65
+    },
+    {
+      "epoch": 4.117647058823529,
+      "grad_norm": 0.48600735056350475,
+      "learning_rate": 1.7582417582417584e-05,
+      "loss": 0.4355,
+      "mean_token_accuracy": 0.8996923744678498,
+      "num_tokens": 17685022.0,
+      "step": 70
+    },
+    {
+      "epoch": 4.411764705882353,
+      "grad_norm": 0.6270999510091785,
+      "learning_rate": 1.4835164835164836e-05,
+      "loss": 0.3213,
+      "mean_token_accuracy": 0.9201981544494628,
+      "num_tokens": 18972948.0,
+      "step": 75
+    },
+    {
+      "epoch": 4.705882352941177,
+      "grad_norm": 0.5582166829408877,
+      "learning_rate": 1.2087912087912089e-05,
+      "loss": 0.3534,
+      "mean_token_accuracy": 0.9134442567825317,
+      "num_tokens": 20241749.0,
+      "step": 80
+    },
+    {
+      "epoch": 5.0,
+      "grad_norm": 0.44013642641683465,
+      "learning_rate": 9.340659340659341e-06,
+      "loss": 0.2391,
+      "mean_token_accuracy": 0.9390455543994903,
+      "num_tokens": 21475135.0,
+      "step": 85
+    },
+    {
+      "epoch": 5.294117647058823,
+      "grad_norm": 0.3728640621917392,
+      "learning_rate": 6.5934065934065935e-06,
+      "loss": 0.3093,
+      "mean_token_accuracy": 0.9262847900390625,
+      "num_tokens": 22722522.0,
+      "step": 90
+    },
+    {
+      "epoch": 5.588235294117647,
+      "grad_norm": 0.37920869196649243,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.2553,
+      "mean_token_accuracy": 0.9366285502910614,
+      "num_tokens": 23999790.0,
+      "step": 95
+    },
+    {
+      "epoch": 5.882352941176471,
+      "grad_norm": 0.3560598736684562,
+      "learning_rate": 1.098901098901099e-06,
+      "loss": 0.2146,
+      "mean_token_accuracy": 0.94760302901268,
+      "num_tokens": 25261256.0,
+      "step": 100
+    },
+    {
+      "epoch": 6.0,
+      "mean_token_accuracy": 0.9418431520462036,
+      "num_tokens": 25768100.0,
+      "step": 102,
+      "total_flos": 38524314714112.0,
+      "train_loss": 0.5417864180955232,
+      "train_runtime": 489.9093,
+      "train_samples_per_second": 6.601,
+      "train_steps_per_second": 0.208
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 102,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 6,
+  "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": 38524314714112.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}