Model save

README.md

@@ -1,10 +1,8 @@
 ---
-datasets: HuggingFaceH4/Bespoke-Stratos-17k
 library_name: transformers
 model_name: Qwen2.5-1.5B-Open-R1-Distill
 tags:
 - generated_from_trainer
-- open-r1
 - trl
 - sft
 licence: license
@@ -12,7 +10,7 @@ licence: license
 
 # Model Card for Qwen2.5-1.5B-Open-R1-Distill
 
-This model is a fine-tuned version of [None](https://huggingface.co/None) on the [HuggingFaceH4/Bespoke-Stratos-17k](https://huggingface.co/datasets/HuggingFaceH4/Bespoke-Stratos-17k) dataset.
+This model is a fine-tuned version of [None](https://huggingface.co/None).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -28,7 +26,7 @@ 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/xt0521/huggingface/runs/eoh1k9sl) 
+[<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/xt0521/huggingface/runs/glyjssry) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,14 +1,9 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.7885702848434448,
-    "eval_runtime": 1.9396,
-    "eval_samples": 100,
-    "eval_samples_per_second": 65.994,
-    "eval_steps_per_second": 2.062,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8114958641797128,
-    "train_runtime": 1241.942,
+    "train_loss": 0.7337305870281874,
+    "train_runtime": 1235.0855,
     "train_samples": 16610,
-    "train_samples_per_second": 17.4,
-    "train_steps_per_second": 0.136
+    "train_samples_per_second": 17.497,
+    "train_steps_per_second": 0.137
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "/data/share/huggingface/models--Qwen--Qwen2.5-1.5B-Instruct/snapshots/989aa7980e4cf806f80c7fef2b1adb7bc71aa306",
+  "_name_or_path": "/data/share8/cdn/open-r1/data/Qwen2.5-1.5B-Open-R1-Distill",
   "architectures": [
     "Qwen2ForCausalLM"
   ],
@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.48.2",
-  "use_cache": true,
+  "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:436e31c0fdab3b43d08aa0f4ea6cf8eae8ff072525b206062dfd58064686980d
+oid sha256:7dddd44bff605045e709f6628512c5276aea5b9b04fe1662e05e295dabaee4f2
 size 3087467144

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 76916824473600.0,
-    "train_loss": 0.8114958641797128,
-    "train_runtime": 1241.942,
+    "train_loss": 0.7337305870281874,
+    "train_runtime": 1235.0855,
     "train_samples": 16610,
-    "train_samples_per_second": 17.4,
-    "train_steps_per_second": 0.136
+    "train_samples_per_second": 17.497,
+    "train_steps_per_second": 0.137
 }

trainer_state.json

@@ -10,251 +10,251 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 0.5857761225642123,
+      "grad_norm": 0.07426696918416174,
       "learning_rate": 5.882352941176471e-06,
-      "loss": 1.0993,
+      "loss": 0.7515,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 0.4995625436387945,
+      "grad_norm": 0.08040276233943516,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.0431,
+      "loss": 0.755,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 0.26485019625158435,
+      "grad_norm": 0.09568613129719651,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 0.9663,
+      "loss": 0.7553,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.19810328582136066,
+      "grad_norm": 0.11073070319325445,
       "learning_rate": 1.9980782984658682e-05,
-      "loss": 0.9073,
+      "loss": 0.7481,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.15545420301777027,
+      "grad_norm": 0.1085253681695973,
       "learning_rate": 1.9863613034027224e-05,
-      "loss": 0.8704,
+      "loss": 0.742,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.1251471825263471,
+      "grad_norm": 0.10317427695342557,
       "learning_rate": 1.9641197940012136e-05,
-      "loss": 0.8413,
+      "loss": 0.7317,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.10400226170447692,
+      "grad_norm": 0.1095919075191608,
       "learning_rate": 1.9315910880512792e-05,
-      "loss": 0.835,
+      "loss": 0.7394,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.09174900478609728,
+      "grad_norm": 0.10584984713775991,
       "learning_rate": 1.8891222681391853e-05,
-      "loss": 0.8337,
+      "loss": 0.7445,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.08878405648463664,
+      "grad_norm": 0.1052036771871902,
       "learning_rate": 1.8371664782625287e-05,
-      "loss": 0.8179,
+      "loss": 0.7382,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.08734426916492237,
+      "grad_norm": 0.10185995684228026,
       "learning_rate": 1.7762780887657576e-05,
-      "loss": 0.8078,
+      "loss": 0.7327,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
-      "grad_norm": 0.08654300161749975,
+      "grad_norm": 0.11605723315982408,
       "learning_rate": 1.7071067811865477e-05,
-      "loss": 0.7971,
+      "loss": 0.7265,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.09049709236672561,
+      "grad_norm": 0.09496481679546319,
       "learning_rate": 1.6303906161279554e-05,
-      "loss": 0.8071,
+      "loss": 0.7404,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.08688275236146101,
+      "grad_norm": 0.09399123086469602,
       "learning_rate": 1.5469481581224274e-05,
-      "loss": 0.7809,
+      "loss": 0.7183,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
-      "grad_norm": 0.07889773883483042,
+      "grad_norm": 0.09454325149914487,
       "learning_rate": 1.4576697415156818e-05,
-      "loss": 0.7826,
+      "loss": 0.7239,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
-      "grad_norm": 0.08580994154606769,
+      "grad_norm": 0.09665316804682329,
       "learning_rate": 1.3635079705638298e-05,
-      "loss": 0.7936,
+      "loss": 0.7374,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.08418237174588653,
+      "grad_norm": 0.0931576465014015,
       "learning_rate": 1.2654675551080724e-05,
-      "loss": 0.7749,
+      "loss": 0.7206,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.08456036679862641,
+      "grad_norm": 0.09445002041196521,
       "learning_rate": 1.164594590280734e-05,
-      "loss": 0.7787,
+      "loss": 0.727,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.08496080754943852,
+      "grad_norm": 0.08585910164532164,
       "learning_rate": 1.0619653946285948e-05,
-      "loss": 0.7781,
+      "loss": 0.7287,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.08860936983983704,
+      "grad_norm": 0.0926566404725236,
       "learning_rate": 9.586750257511868e-06,
-      "loss": 0.7852,
+      "loss": 0.7378,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.08415248626283894,
+      "grad_norm": 0.08672021822123627,
       "learning_rate": 8.558255959926533e-06,
-      "loss": 0.7609,
+      "loss": 0.7164,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.7962175607681274,
-      "eval_runtime": 1.9652,
-      "eval_samples_per_second": 65.135,
-      "eval_steps_per_second": 2.035,
+      "eval_loss": 0.7641129493713379,
+      "eval_runtime": 1.9602,
+      "eval_samples_per_second": 65.298,
+      "eval_steps_per_second": 2.041,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.08441236349679139,
+      "grad_norm": 0.08622854416758327,
       "learning_rate": 7.545145128592009e-06,
-      "loss": 0.7649,
+      "loss": 0.7216,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.08177150474959695,
+      "grad_norm": 0.08204894186576381,
       "learning_rate": 6.558227696373617e-06,
-      "loss": 0.7761,
+      "loss": 0.7346,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
-      "grad_norm": 0.07746543923576864,
+      "grad_norm": 0.08097299352958673,
       "learning_rate": 5.608034111526298e-06,
-      "loss": 0.7704,
+      "loss": 0.73,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
-      "grad_norm": 0.07544691836481318,
+      "grad_norm": 0.07682552342075682,
       "learning_rate": 4.704702977392914e-06,
-      "loss": 0.7594,
+      "loss": 0.7212,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
-      "grad_norm": 0.08158636138275724,
+      "grad_norm": 0.08086018117086849,
       "learning_rate": 3.857872873103322e-06,
-      "loss": 0.7619,
+      "loss": 0.7248,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.0768086707215383,
+      "grad_norm": 0.07606171025014277,
       "learning_rate": 3.0765795095517026e-06,
-      "loss": 0.7634,
+      "loss": 0.728,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.0838125766014346,
+      "grad_norm": 0.08257799296717885,
       "learning_rate": 2.369159318001937e-06,
-      "loss": 0.7665,
+      "loss": 0.7323,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.07598473377325965,
+      "grad_norm": 0.07889584661038457,
       "learning_rate": 1.743160500034443e-06,
-      "loss": 0.7579,
+      "loss": 0.7251,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.07918517584309687,
+      "grad_norm": 0.08171166394737774,
       "learning_rate": 1.2052624879351105e-06,
-      "loss": 0.7646,
+      "loss": 0.7324,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.0757111355601306,
+      "grad_norm": 0.07432343760979292,
       "learning_rate": 7.612046748871327e-07,
-      "loss": 0.7747,
+      "loss": 0.743,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
-      "grad_norm": 0.07559936201946106,
+      "grad_norm": 0.07790477411393834,
       "learning_rate": 4.1572517541747294e-07,
-      "loss": 0.7694,
+      "loss": 0.7386,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.07247200888081211,
+      "grad_norm": 0.07339780497225679,
       "learning_rate": 1.7251026952640583e-07,
-      "loss": 0.7687,
+      "loss": 0.7383,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.0703519415777988,
+      "grad_norm": 0.07176968893681754,
       "learning_rate": 3.4155069933301535e-08,
-      "loss": 0.7526,
+      "loss": 0.7225,
       "step": 165
     },
     {
       "epoch": 1.0,
       "step": 169,
       "total_flos": 76916824473600.0,
-      "train_loss": 0.8114958641797128,
-      "train_runtime": 1241.942,
-      "train_samples_per_second": 17.4,
-      "train_steps_per_second": 0.136
+      "train_loss": 0.7337305870281874,
+      "train_runtime": 1235.0855,
+      "train_samples_per_second": 17.497,
+      "train_steps_per_second": 0.137
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5da9a21d9c3f1b688b6f1d7f3e417cb01a4678dcb6fc905d7da7d23aa708536
+oid sha256:dbaa3b15f07865886766a303e293d83c8ac37c11d3cc809fc3bdc3ab96eb0eba
 size 7352