Model save

README.md

@@ -0,0 +1,58 @@
+---
+base_model: Qwen/Qwen2.5-1.5B-Instruct
+library_name: transformers
+model_name: qwen-2.5-1.5B-instruct-sft-lora-countdown-deepseek-6k
+tags:
+- generated_from_trainer
+- trl
+- sft
+licence: license
+---
+
+# Model Card for qwen-2.5-1.5B-instruct-sft-lora-countdown-deepseek-6k
+
+This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-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="yeok/qwen-2.5-1.5B-instruct-sft-lora-countdown-deepseek-6k", 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/yeokch/stream-of-search-train/runs/9x3udzje) 
+
+
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.15.2
+- Transformers: 4.50.0
+- Pytorch: 2.6.0
+- 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": 1.4186790468871782e+17,
+    "train_loss": 0.6605893557562548,
+    "train_runtime": 6552.0914,
+    "train_samples": 6316,
+    "train_samples_per_second": 0.334,
+    "train_steps_per_second": 0.021
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "total_flos": 1.4186790468871782e+17,
+    "train_loss": 0.6605893557562548,
+    "train_runtime": 6552.0914,
+    "train_samples": 6316,
+    "train_samples_per_second": 0.334,
+    "train_steps_per_second": 0.021
+}

trainer_state.json

@@ -0,0 +1,268 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.9945155393053017,
+  "eval_steps": 500,
+  "global_step": 136,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.007312614259597806,
+      "grad_norm": 0.12043265998363495,
+      "learning_rate": 1.4285714285714285e-05,
+      "loss": 0.9622,
+      "mean_token_accuracy": 0.7244765534996986,
+      "step": 1
+    },
+    {
+      "epoch": 0.03656307129798903,
+      "grad_norm": 0.1216142550110817,
+      "learning_rate": 7.142857142857143e-05,
+      "loss": 1.0026,
+      "mean_token_accuracy": 0.7134055998176336,
+      "step": 5
+    },
+    {
+      "epoch": 0.07312614259597806,
+      "grad_norm": 0.11720138788223267,
+      "learning_rate": 0.00014285714285714287,
+      "loss": 0.9768,
+      "mean_token_accuracy": 0.7196862399578094,
+      "step": 10
+    },
+    {
+      "epoch": 0.10968921389396709,
+      "grad_norm": 0.13313624262809753,
+      "learning_rate": 0.0001999668467514313,
+      "loss": 0.9109,
+      "mean_token_accuracy": 0.7297002755105495,
+      "step": 15
+    },
+    {
+      "epoch": 0.14625228519195613,
+      "grad_norm": 0.1300281137228012,
+      "learning_rate": 0.00019880878960910772,
+      "loss": 0.8575,
+      "mean_token_accuracy": 0.740726611763239,
+      "step": 20
+    },
+    {
+      "epoch": 0.18281535648994515,
+      "grad_norm": 0.1045694574713707,
+      "learning_rate": 0.00019601498736716017,
+      "loss": 0.7773,
+      "mean_token_accuracy": 0.759741972386837,
+      "step": 25
+    },
+    {
+      "epoch": 0.21937842778793418,
+      "grad_norm": 0.09669230878353119,
+      "learning_rate": 0.0001916316904487005,
+      "loss": 0.7471,
+      "mean_token_accuracy": 0.7641573064029217,
+      "step": 30
+    },
+    {
+      "epoch": 0.25594149908592323,
+      "grad_norm": 0.08908155560493469,
+      "learning_rate": 0.00018573146280763324,
+      "loss": 0.7116,
+      "mean_token_accuracy": 0.7720760576426983,
+      "step": 35
+    },
+    {
+      "epoch": 0.29250457038391225,
+      "grad_norm": 0.09818249940872192,
+      "learning_rate": 0.00017841198065767107,
+      "loss": 0.6836,
+      "mean_token_accuracy": 0.778372598439455,
+      "step": 40
+    },
+    {
+      "epoch": 0.3290676416819013,
+      "grad_norm": 0.11527201533317566,
+      "learning_rate": 0.00016979441547663435,
+      "loss": 0.6559,
+      "mean_token_accuracy": 0.7842052198946476,
+      "step": 45
+    },
+    {
+      "epoch": 0.3656307129798903,
+      "grad_norm": 0.1011962741613388,
+      "learning_rate": 0.00016002142805483685,
+      "loss": 0.6346,
+      "mean_token_accuracy": 0.7901677042245865,
+      "step": 50
+    },
+    {
+      "epoch": 0.40219378427787933,
+      "grad_norm": 0.1208968386054039,
+      "learning_rate": 0.00014925480679538647,
+      "loss": 0.6312,
+      "mean_token_accuracy": 0.7898989707231522,
+      "step": 55
+    },
+    {
+      "epoch": 0.43875685557586835,
+      "grad_norm": 0.09324625879526138,
+      "learning_rate": 0.00013767278936351854,
+      "loss": 0.6049,
+      "mean_token_accuracy": 0.7972484216094017,
+      "step": 60
+    },
+    {
+      "epoch": 0.4753199268738574,
+      "grad_norm": 0.10249213129281998,
+      "learning_rate": 0.00012546711202412287,
+      "loss": 0.5993,
+      "mean_token_accuracy": 0.7985120855271817,
+      "step": 65
+    },
+    {
+      "epoch": 0.5118829981718465,
+      "grad_norm": 0.11906760931015015,
+      "learning_rate": 0.00011283983551465511,
+      "loss": 0.5966,
+      "mean_token_accuracy": 0.7997283555567265,
+      "step": 70
+    },
+    {
+      "epoch": 0.5484460694698354,
+      "grad_norm": 0.10403923690319061,
+      "learning_rate": 0.0001,
+      "loss": 0.5929,
+      "mean_token_accuracy": 0.8001663401722908,
+      "step": 75
+    },
+    {
+      "epoch": 0.5850091407678245,
+      "grad_norm": 0.11625717580318451,
+      "learning_rate": 8.71601644853449e-05,
+      "loss": 0.5853,
+      "mean_token_accuracy": 0.801834324002266,
+      "step": 80
+    },
+    {
+      "epoch": 0.6215722120658135,
+      "grad_norm": 0.14587707817554474,
+      "learning_rate": 7.453288797587714e-05,
+      "loss": 0.5772,
+      "mean_token_accuracy": 0.8043950639665127,
+      "step": 85
+    },
+    {
+      "epoch": 0.6581352833638026,
+      "grad_norm": 0.09371322393417358,
+      "learning_rate": 6.232721063648148e-05,
+      "loss": 0.5775,
+      "mean_token_accuracy": 0.8033298641443253,
+      "step": 90
+    },
+    {
+      "epoch": 0.6946983546617916,
+      "grad_norm": 0.10421498864889145,
+      "learning_rate": 5.074519320461357e-05,
+      "loss": 0.5726,
+      "mean_token_accuracy": 0.8055274039506912,
+      "step": 95
+    },
+    {
+      "epoch": 0.7312614259597806,
+      "grad_norm": 0.10395624488592148,
+      "learning_rate": 3.997857194516319e-05,
+      "loss": 0.5734,
+      "mean_token_accuracy": 0.805304603278637,
+      "step": 100
+    },
+    {
+      "epoch": 0.7678244972577697,
+      "grad_norm": 0.10051889717578888,
+      "learning_rate": 3.0205584523365626e-05,
+      "loss": 0.5767,
+      "mean_token_accuracy": 0.8036030374467373,
+      "step": 105
+    },
+    {
+      "epoch": 0.8043875685557587,
+      "grad_norm": 0.09088820964097977,
+      "learning_rate": 2.1588019342328968e-05,
+      "loss": 0.5796,
+      "mean_token_accuracy": 0.8022675953805447,
+      "step": 110
+    },
+    {
+      "epoch": 0.8409506398537477,
+      "grad_norm": 0.09396003931760788,
+      "learning_rate": 1.426853719236676e-05,
+      "loss": 0.5678,
+      "mean_token_accuracy": 0.8059396557509899,
+      "step": 115
+    },
+    {
+      "epoch": 0.8775137111517367,
+      "grad_norm": 0.08524946868419647,
+      "learning_rate": 8.368309551299536e-06,
+      "loss": 0.5717,
+      "mean_token_accuracy": 0.8042348228394985,
+      "step": 120
+    },
+    {
+      "epoch": 0.9140767824497258,
+      "grad_norm": 0.08044712990522385,
+      "learning_rate": 3.985012632839824e-06,
+      "loss": 0.5676,
+      "mean_token_accuracy": 0.8066093817353248,
+      "step": 125
+    },
+    {
+      "epoch": 0.9506398537477148,
+      "grad_norm": 0.08666696399450302,
+      "learning_rate": 1.1912103908922945e-06,
+      "loss": 0.5657,
+      "mean_token_accuracy": 0.8065590254962445,
+      "step": 130
+    },
+    {
+      "epoch": 0.9872029250457038,
+      "grad_norm": 0.0783163458108902,
+      "learning_rate": 3.3153248568695835e-08,
+      "loss": 0.5627,
+      "mean_token_accuracy": 0.8076440550386905,
+      "step": 135
+    },
+    {
+      "epoch": 0.9945155393053017,
+      "mean_token_accuracy": 0.8013826087117195,
+      "step": 136,
+      "total_flos": 1.4186790468871782e+17,
+      "train_loss": 0.6605893557562548,
+      "train_runtime": 6552.0914,
+      "train_samples_per_second": 0.334,
+      "train_steps_per_second": 0.021
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 136,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 1.4186790468871782e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}