Model save

README.md

@@ -1,11 +1,9 @@
 ---
 base_model: Qwen/Qwen2.5-1.5B-Instruct
-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
@@ -13,7 +11,7 @@ licence: license
 
 # Model Card for Qwen2.5-1.5B-Open-R1-Distill
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-1.5B-Instruct) on the [HuggingFaceH4/Bespoke-Stratos-17k](https://huggingface.co/datasets/HuggingFaceH4/Bespoke-Stratos-17k) dataset.
+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
@@ -29,7 +27,7 @@ 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/xiaojia-song12/huggingface/runs/c0mgxv5y) 
+[<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/xiaojia-song12/huggingface/runs/hc5miacf) 
 
 
 This model was trained with SFT.

all_results.json

@@ -5,9 +5,9 @@
     "eval_samples_per_second": 57.495,
     "eval_steps_per_second": 2.228,
     "total_flos": 6.966137809639834e+17,
-    "train_loss": 0.8680753425733577,
-    "train_runtime": 812.5535,
+    "train_loss": 0.8674715298872727,
+    "train_runtime": 815.3952,
     "train_samples": 16610,
-    "train_samples_per_second": 26.611,
-    "train_steps_per_second": 0.208
+    "train_samples_per_second": 26.518,
+    "train_steps_per_second": 0.207
 }

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.49.0.dev0",
-  "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 sha256:9e6600174797e7c283b02f903def90633fb7d09e9abc244922e750a53382b5c9
+oid sha256:d69e13d1e0a1431e99e002eb37cd9f5b7404337bcbd336a59d7f02b2e4bbadf7
 size 3087467144

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 6.966137809639834e+17,
-    "train_loss": 0.8680753425733577,
-    "train_runtime": 812.5535,
+    "train_loss": 0.8674715298872727,
+    "train_runtime": 815.3952,
     "train_samples": 16610,
-    "train_samples_per_second": 26.611,
-    "train_steps_per_second": 0.208
+    "train_samples_per_second": 26.518,
+    "train_steps_per_second": 0.207
 }

trainer_state.json

@@ -10,286 +10,286 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 2.40625,
+      "grad_norm": 5.03125,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.1011,
-      "mean_token_accuracy": 0.7099540950451799,
+      "mean_token_accuracy": 0.7099399748962237,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 1.46875,
+      "grad_norm": 1.484375,
       "learning_rate": 1.1764705882352942e-05,
-      "loss": 1.0735,
-      "mean_token_accuracy": 0.7144238503704025,
+      "loss": 1.0741,
+      "mean_token_accuracy": 0.7142575641587607,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 1.03125,
+      "grad_norm": 0.9296875,
       "learning_rate": 1.7647058823529414e-05,
-      "loss": 1.0258,
-      "mean_token_accuracy": 0.7210694118575746,
+      "loss": 1.025,
+      "mean_token_accuracy": 0.7212255669151615,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.640625,
+      "grad_norm": 0.61328125,
       "learning_rate": 1.9980782984658682e-05,
-      "loss": 0.9525,
-      "mean_token_accuracy": 0.735324240337934,
+      "loss": 0.9455,
+      "mean_token_accuracy": 0.7365584053166825,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.5234375,
+      "grad_norm": 0.515625,
       "learning_rate": 1.9863613034027224e-05,
-      "loss": 0.9139,
-      "mean_token_accuracy": 0.7412657006848294,
+      "loss": 0.9097,
+      "mean_token_accuracy": 0.7419796471746476,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.419921875,
+      "grad_norm": 0.42578125,
       "learning_rate": 1.9641197940012136e-05,
-      "loss": 0.898,
-      "mean_token_accuracy": 0.7441235047874912,
+      "loss": 0.8953,
+      "mean_token_accuracy": 0.744632902913685,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.41015625,
+      "grad_norm": 0.408203125,
       "learning_rate": 1.9315910880512792e-05,
-      "loss": 0.8794,
-      "mean_token_accuracy": 0.7469598871124087,
+      "loss": 0.8769,
+      "mean_token_accuracy": 0.7474371048323014,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.390625,
+      "grad_norm": 0.388671875,
       "learning_rate": 1.8891222681391853e-05,
-      "loss": 0.8652,
-      "mean_token_accuracy": 0.7504773681649235,
+      "loss": 0.8623,
+      "mean_token_accuracy": 0.7511180018113048,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.359375,
+      "grad_norm": 0.357421875,
       "learning_rate": 1.8371664782625287e-05,
-      "loss": 0.8584,
-      "mean_token_accuracy": 0.7515782132486059,
+      "loss": 0.8557,
+      "mean_token_accuracy": 0.7520890468920156,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.328125,
+      "grad_norm": 0.326171875,
       "learning_rate": 1.7762780887657576e-05,
-      "loss": 0.8541,
-      "mean_token_accuracy": 0.7522614136819363,
+      "loss": 0.8519,
+      "mean_token_accuracy": 0.7527647439023715,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
       "grad_norm": 0.34375,
       "learning_rate": 1.7071067811865477e-05,
-      "loss": 0.8514,
-      "mean_token_accuracy": 0.752774123806763,
+      "loss": 0.8499,
+      "mean_token_accuracy": 0.7530196620772068,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.34765625,
+      "grad_norm": 0.345703125,
       "learning_rate": 1.6303906161279554e-05,
-      "loss": 0.8633,
-      "mean_token_accuracy": 0.7493902791974734,
+      "loss": 0.8623,
+      "mean_token_accuracy": 0.7496151012147616,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.32421875,
+      "grad_norm": 0.322265625,
       "learning_rate": 1.5469481581224274e-05,
-      "loss": 0.85,
-      "mean_token_accuracy": 0.7522941829192423,
+      "loss": 0.8491,
+      "mean_token_accuracy": 0.7524191922540802,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
       "grad_norm": 0.33984375,
       "learning_rate": 1.4576697415156818e-05,
-      "loss": 0.8425,
-      "mean_token_accuracy": 0.7545225920790829,
+      "loss": 0.8418,
+      "mean_token_accuracy": 0.7547066041239219,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
       "grad_norm": 0.3125,
       "learning_rate": 1.3635079705638298e-05,
-      "loss": 0.8422,
-      "mean_token_accuracy": 0.7543091860268133,
+      "loss": 0.8416,
+      "mean_token_accuracy": 0.7544158237788706,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.322265625,
+      "grad_norm": 9.5625,
       "learning_rate": 1.2654675551080724e-05,
-      "loss": 0.8455,
-      "mean_token_accuracy": 0.7530642575258402,
+      "loss": 0.8451,
+      "mean_token_accuracy": 0.7531590392914497,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.330078125,
+      "grad_norm": 0.328125,
       "learning_rate": 1.164594590280734e-05,
-      "loss": 0.8456,
-      "mean_token_accuracy": 0.7532945419469173,
+      "loss": 0.8455,
+      "mean_token_accuracy": 0.7532937761410722,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.3359375,
+      "grad_norm": 0.337890625,
       "learning_rate": 1.0619653946285948e-05,
-      "loss": 0.8397,
-      "mean_token_accuracy": 0.7545404345362574,
+      "loss": 0.8399,
+      "mean_token_accuracy": 0.7544911735347861,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.353515625,
+      "grad_norm": 0.35546875,
       "learning_rate": 9.586750257511868e-06,
-      "loss": 0.8354,
-      "mean_token_accuracy": 0.755724879023685,
+      "loss": 0.8358,
+      "mean_token_accuracy": 0.7556337403583416,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.32421875,
+      "grad_norm": 13.75,
       "learning_rate": 8.558255959926533e-06,
-      "loss": 0.8265,
-      "mean_token_accuracy": 0.757549675432504,
+      "loss": 0.8269,
+      "mean_token_accuracy": 0.7575153177712692,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.8585203289985657,
-      "eval_mean_token_accuracy": 0.7337473206121845,
-      "eval_runtime": 2.291,
-      "eval_samples_per_second": 56.307,
-      "eval_steps_per_second": 2.182,
+      "eval_loss": 0.8590129017829895,
+      "eval_mean_token_accuracy": 0.7343020822802343,
+      "eval_runtime": 2.301,
+      "eval_samples_per_second": 56.063,
+      "eval_steps_per_second": 2.173,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.3203125,
+      "grad_norm": 0.326171875,
       "learning_rate": 7.545145128592009e-06,
-      "loss": 0.8284,
-      "mean_token_accuracy": 0.7576493237022721,
+      "loss": 0.8289,
+      "mean_token_accuracy": 0.7575234407641943,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.314453125,
+      "grad_norm": 21.25,
       "learning_rate": 6.558227696373617e-06,
-      "loss": 0.8496,
-      "mean_token_accuracy": 0.7518721351805249,
+      "loss": 0.8501,
+      "mean_token_accuracy": 0.7518221200713989,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
       "grad_norm": 0.322265625,
       "learning_rate": 5.608034111526298e-06,
-      "loss": 0.8355,
-      "mean_token_accuracy": 0.7557122291164196,
+      "loss": 0.8361,
+      "mean_token_accuracy": 0.7556435102369359,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
       "grad_norm": 0.328125,
       "learning_rate": 4.704702977392914e-06,
-      "loss": 0.8229,
-      "mean_token_accuracy": 0.7593972283565047,
+      "loss": 0.8234,
+      "mean_token_accuracy": 0.7592704800654426,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
       "grad_norm": 0.314453125,
       "learning_rate": 3.857872873103322e-06,
-      "loss": 0.8322,
-      "mean_token_accuracy": 0.7566853250706351,
+      "loss": 0.8328,
+      "mean_token_accuracy": 0.7565933236472202,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.326171875,
+      "grad_norm": 0.328125,
       "learning_rate": 3.0765795095517026e-06,
-      "loss": 0.8366,
-      "mean_token_accuracy": 0.7552665039102797,
+      "loss": 0.8372,
+      "mean_token_accuracy": 0.7552126694771518,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.3125,
+      "grad_norm": 0.314453125,
       "learning_rate": 2.369159318001937e-06,
-      "loss": 0.843,
-      "mean_token_accuracy": 0.7535886881654584,
+      "loss": 0.8437,
+      "mean_token_accuracy": 0.7534241447974153,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.31640625,
+      "grad_norm": 0.318359375,
       "learning_rate": 1.743160500034443e-06,
-      "loss": 0.8344,
-      "mean_token_accuracy": 0.7564495559624271,
+      "loss": 0.835,
+      "mean_token_accuracy": 0.7563463400384239,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.328125,
+      "grad_norm": 0.330078125,
       "learning_rate": 1.2052624879351105e-06,
-      "loss": 0.8346,
-      "mean_token_accuracy": 0.756153973007336,
+      "loss": 0.8351,
+      "mean_token_accuracy": 0.7560251329699866,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.310546875,
+      "grad_norm": 0.30859375,
       "learning_rate": 7.612046748871327e-07,
-      "loss": 0.8248,
-      "mean_token_accuracy": 0.7587105830501064,
+      "loss": 0.8254,
+      "mean_token_accuracy": 0.7586063543068307,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
       "grad_norm": 0.31640625,
       "learning_rate": 4.1572517541747294e-07,
-      "loss": 0.8152,
-      "mean_token_accuracy": 0.761315871287601,
+      "loss": 0.8158,
+      "mean_token_accuracy": 0.7611673621269803,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.322265625,
+      "grad_norm": 0.32421875,
       "learning_rate": 1.7251026952640583e-07,
-      "loss": 0.8283,
-      "mean_token_accuracy": 0.7578360411652272,
+      "loss": 0.8289,
+      "mean_token_accuracy": 0.7576611812663976,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.314453125,
+      "grad_norm": 0.3125,
       "learning_rate": 3.4155069933301535e-08,
-      "loss": 0.821,
-      "mean_token_accuracy": 0.7596401558672835,
+      "loss": 0.8216,
+      "mean_token_accuracy": 0.7594309628290168,
       "step": 165
     },
     {
       "epoch": 1.0,
-      "mean_token_accuracy": 0.7548530141188281,
+      "mean_token_accuracy": 0.7546930193447599,
       "step": 169,
       "total_flos": 6.966137809639834e+17,
-      "train_loss": 0.8680753425733577,
-      "train_runtime": 812.5535,
-      "train_samples_per_second": 26.611,
-      "train_steps_per_second": 0.208
+      "train_loss": 0.8674715298872727,
+      "train_runtime": 815.3952,
+      "train_samples_per_second": 26.518,
+      "train_steps_per_second": 0.207
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e99bb66c8beb3692b77eca2bc0dc7a0456ee44136dad588108c411f3ed131b65
+oid sha256:1fd4518eeebb19a79bf77cb8690174a32d07f567e7bd1ae9d9a31d0e6e6e1074
 size 6008