Model save

README.md

@@ -7,8 +7,6 @@ tags:
 - trl
 - sft
 licence: license
-datasets:
-- bespokelabs/Bespoke-Stratos-17k
 ---
 
 # Model Card for Qwen2.5-1.5B-Open-R1-Distill
@@ -20,15 +18,16 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ```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="Mingsmilet/Qwen2.5-1.5B-Open-R1-Distill", device="cuda")
-question = "what is min value of x^2+3?"
-output = generator([{"role": "system", "content":"Your role as an assistant involves thoroughly exploring questions through a systematic long thinking process before providing the final precise and accurate solutions. This requires engaging in a comprehensive cycle of analysis, summarizing, exploration, reassessment, reflection, backtracing, and iteration to develop well-considered thinking process. Please structure your response into two main sections: Thought and Solution. In the Thought section, detail your reasoning process using the specified format: <|begin_of_thought|> {thought with steps separated with '\n\n'} <|end_of_thought|> Each step should include detailed considerations such as analisying questions, summarizing relevant findings, brainstorming new ideas, verifying the accuracy of the current steps, refining any errors, and revisiting previous steps. In the Solution section, based on various attempts, explorations, and reflections from the Thought section, systematically present the final solution that you deem correct. The solution should remain a logical, accurate, concise expression style and detail necessary step needed to reach the conclusion, formatted as follows: <|begin_of_solution|> {final formatted, precise, and clear solution} <|end_of_solution|> Now, try to solve the following question through the above guidelines:"},{"role": "user", "content": question}], max_new_tokens=2048, return_full_text=False)[0]
+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/1653401183-mingmingai/huggingface/runs/batphcqf) 
+[<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/1653401183-mingmingai/huggingface/runs/2nzizgw4) 
 
 
 This model was trained with SFT.

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 76973799899136.0,
-    "train_loss": 0.8025032683942445,
-    "train_runtime": 710.0203,
+    "train_loss": 0.802504962717993,
+    "train_runtime": 695.5303,
     "train_samples": 16610,
-    "train_samples_per_second": 30.454,
-    "train_steps_per_second": 0.238
+    "train_samples_per_second": 31.089,
+    "train_steps_per_second": 0.243
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:846816e48b86e2be4cecdd5778e99f02cb73c4f53705b0b48f09130fb97960a4
+oid sha256:cdde57df3dbf087870df00e8e1ec680c6c90cd70e09da0bfc89de0dc63069fa2
 size 3087467144

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 76973799899136.0,
-    "train_loss": 0.8025032683942445,
-    "train_runtime": 710.0203,
+    "train_loss": 0.802504962717993,
+    "train_runtime": 695.5303,
     "train_samples": 16610,
-    "train_samples_per_second": 30.454,
-    "train_steps_per_second": 0.238
+    "train_samples_per_second": 31.089,
+    "train_steps_per_second": 0.243
 }

trainer_state.json

@@ -10,286 +10,286 @@
   "log_history": [
     {
       "epoch": 0.029585798816568046,
-      "grad_norm": 2.2988141077652386,
+      "grad_norm": 2.297573765324108,
       "learning_rate": 5.882352941176471e-06,
       "loss": 1.1002,
-      "mean_token_accuracy": 0.7101159904452009,
+      "mean_token_accuracy": 0.7100953695790506,
       "step": 5
     },
     {
       "epoch": 0.05917159763313609,
-      "grad_norm": 1.596218927081988,
+      "grad_norm": 1.5969597101948139,
       "learning_rate": 1.1764705882352942e-05,
       "loss": 1.0327,
-      "mean_token_accuracy": 0.720176724225474,
+      "mean_token_accuracy": 0.7202142927855761,
       "step": 10
     },
     {
       "epoch": 0.08875739644970414,
-      "grad_norm": 0.8479104246833407,
+      "grad_norm": 0.8485756643128731,
       "learning_rate": 1.7647058823529414e-05,
       "loss": 0.9517,
-      "mean_token_accuracy": 0.7325662989711975,
+      "mean_token_accuracy": 0.7326373095769856,
       "step": 15
     },
     {
       "epoch": 0.11834319526627218,
-      "grad_norm": 0.6465483847858254,
+      "grad_norm": 0.6438282504171126,
       "learning_rate": 1.9980782984658682e-05,
       "loss": 0.8804,
-      "mean_token_accuracy": 0.7473196725120462,
+      "mean_token_accuracy": 0.7473626526463326,
       "step": 20
     },
     {
       "epoch": 0.14792899408284024,
-      "grad_norm": 0.5525325403311692,
+      "grad_norm": 0.5518035399856269,
       "learning_rate": 1.9863613034027224e-05,
       "loss": 0.8536,
-      "mean_token_accuracy": 0.7518004036197122,
+      "mean_token_accuracy": 0.7518618707778764,
       "step": 25
     },
     {
       "epoch": 0.17751479289940827,
-      "grad_norm": 0.44179450773629486,
+      "grad_norm": 0.4402012573426451,
       "learning_rate": 1.9641197940012136e-05,
       "loss": 0.8404,
-      "mean_token_accuracy": 0.7545674682549227,
+      "mean_token_accuracy": 0.7545927971969407,
       "step": 30
     },
     {
       "epoch": 0.20710059171597633,
-      "grad_norm": 0.39496081405164835,
+      "grad_norm": 0.3943698562073155,
       "learning_rate": 1.9315910880512792e-05,
       "loss": 0.8212,
-      "mean_token_accuracy": 0.7580853602544403,
+      "mean_token_accuracy": 0.7580720047648368,
       "step": 35
     },
     {
       "epoch": 0.23668639053254437,
-      "grad_norm": 0.38859302520022476,
+      "grad_norm": 0.38684017157532,
       "learning_rate": 1.8891222681391853e-05,
       "loss": 0.805,
-      "mean_token_accuracy": 0.7621072520755232,
+      "mean_token_accuracy": 0.7620874011199529,
       "step": 40
     },
     {
       "epoch": 0.26627218934911245,
-      "grad_norm": 0.4281961569580366,
+      "grad_norm": 0.4262882071831819,
       "learning_rate": 1.8371664782625287e-05,
       "loss": 0.7984,
-      "mean_token_accuracy": 0.7635130587409846,
+      "mean_token_accuracy": 0.7635187814772423,
       "step": 45
     },
     {
       "epoch": 0.2958579881656805,
-      "grad_norm": 0.393042490001915,
+      "grad_norm": 0.3931013124021272,
       "learning_rate": 1.7762780887657576e-05,
-      "loss": 0.7936,
-      "mean_token_accuracy": 0.7642980542104124,
+      "loss": 0.7937,
+      "mean_token_accuracy": 0.7642257038779341,
       "step": 50
     },
     {
       "epoch": 0.3254437869822485,
-      "grad_norm": 0.3719832470365729,
+      "grad_norm": 0.3739445931259617,
       "learning_rate": 1.7071067811865477e-05,
       "loss": 0.7894,
-      "mean_token_accuracy": 0.7649172633057038,
+      "mean_token_accuracy": 0.7648935589559052,
       "step": 55
     },
     {
       "epoch": 0.35502958579881655,
-      "grad_norm": 0.36967289202890546,
+      "grad_norm": 0.3711114588864771,
       "learning_rate": 1.6303906161279554e-05,
       "loss": 0.7999,
-      "mean_token_accuracy": 0.7619633834951797,
+      "mean_token_accuracy": 0.7619695483759025,
       "step": 60
     },
     {
       "epoch": 0.38461538461538464,
-      "grad_norm": 0.3965455034810465,
+      "grad_norm": 0.3949481992103986,
       "learning_rate": 1.5469481581224274e-05,
-      "loss": 0.7867,
-      "mean_token_accuracy": 0.7647511953755232,
+      "loss": 0.7868,
+      "mean_token_accuracy": 0.7646586436181606,
       "step": 65
     },
     {
       "epoch": 0.41420118343195267,
-      "grad_norm": 0.368068249126072,
+      "grad_norm": 0.3710782910651557,
       "learning_rate": 1.4576697415156818e-05,
       "loss": 0.7782,
-      "mean_token_accuracy": 0.7674480153107033,
+      "mean_token_accuracy": 0.7674472564108428,
       "step": 70
     },
     {
       "epoch": 0.4437869822485207,
-      "grad_norm": 0.3388950128403286,
+      "grad_norm": 0.34005674173322326,
       "learning_rate": 1.3635079705638298e-05,
       "loss": 0.777,
-      "mean_token_accuracy": 0.7675313738107395,
+      "mean_token_accuracy": 0.7675319726128873,
       "step": 75
     },
     {
       "epoch": 0.47337278106508873,
-      "grad_norm": 0.3430098886707366,
+      "grad_norm": 0.3468353725329877,
       "learning_rate": 1.2654675551080724e-05,
       "loss": 0.7783,
-      "mean_token_accuracy": 0.7666054602362825,
+      "mean_token_accuracy": 0.7666067092453227,
       "step": 80
     },
     {
       "epoch": 0.5029585798816568,
-      "grad_norm": 0.33478411969703326,
+      "grad_norm": 0.33465976050856655,
       "learning_rate": 1.164594590280734e-05,
       "loss": 0.7785,
-      "mean_token_accuracy": 0.7669639865984688,
+      "mean_token_accuracy": 0.7669678075017483,
       "step": 85
     },
     {
       "epoch": 0.5325443786982249,
-      "grad_norm": 0.34768462687825924,
+      "grad_norm": 0.3498974122332,
       "learning_rate": 1.0619653946285948e-05,
       "loss": 0.7723,
-      "mean_token_accuracy": 0.7682154932141926,
+      "mean_token_accuracy": 0.7681925898394718,
       "step": 90
     },
     {
       "epoch": 0.5621301775147929,
-      "grad_norm": 0.3445315216411646,
+      "grad_norm": 0.34393653080794817,
       "learning_rate": 9.586750257511868e-06,
       "loss": 0.7668,
-      "mean_token_accuracy": 0.7698424375606829,
+      "mean_token_accuracy": 0.7697901380963259,
       "step": 95
     },
     {
       "epoch": 0.591715976331361,
-      "grad_norm": 0.3412337472975462,
+      "grad_norm": 0.33899820007825987,
       "learning_rate": 8.558255959926533e-06,
       "loss": 0.7582,
-      "mean_token_accuracy": 0.7714304347299908,
+      "mean_token_accuracy": 0.7714266195965762,
       "step": 100
     },
     {
       "epoch": 0.591715976331361,
-      "eval_loss": 0.7885846495628357,
-      "eval_mean_token_accuracy": 0.7482494053174686,
-      "eval_runtime": 2.2356,
-      "eval_samples_per_second": 57.703,
-      "eval_steps_per_second": 2.237,
+      "eval_loss": 0.788552463054657,
+      "eval_mean_token_accuracy": 0.7482402433575445,
+      "eval_runtime": 2.2045,
+      "eval_samples_per_second": 58.517,
+      "eval_steps_per_second": 2.268,
       "step": 100
     },
     {
       "epoch": 0.621301775147929,
-      "grad_norm": 0.35299455458895956,
+      "grad_norm": 0.353900421817561,
       "learning_rate": 7.545145128592009e-06,
       "loss": 0.759,
-      "mean_token_accuracy": 0.7714738320259433,
+      "mean_token_accuracy": 0.7714586673704755,
       "step": 105
     },
     {
       "epoch": 0.650887573964497,
-      "grad_norm": 0.31616289960261323,
+      "grad_norm": 0.3166511156569925,
       "learning_rate": 6.558227696373617e-06,
-      "loss": 0.7775,
-      "mean_token_accuracy": 0.7664820794613972,
+      "loss": 0.7776,
+      "mean_token_accuracy": 0.7664817000327632,
       "step": 110
     },
     {
       "epoch": 0.6804733727810651,
-      "grad_norm": 0.307888028457379,
+      "grad_norm": 0.3074350676983976,
       "learning_rate": 5.608034111526298e-06,
       "loss": 0.7636,
-      "mean_token_accuracy": 0.7704197310032757,
+      "mean_token_accuracy": 0.7703227561153221,
       "step": 115
     },
     {
       "epoch": 0.7100591715976331,
-      "grad_norm": 0.2995694223287612,
+      "grad_norm": 0.299274177260631,
       "learning_rate": 4.704702977392914e-06,
       "loss": 0.7511,
-      "mean_token_accuracy": 0.7742426406160382,
+      "mean_token_accuracy": 0.7742873168311635,
       "step": 120
     },
     {
       "epoch": 0.7396449704142012,
-      "grad_norm": 0.31585565817231076,
+      "grad_norm": 0.31617489086098977,
       "learning_rate": 3.857872873103322e-06,
-      "loss": 0.7595,
-      "mean_token_accuracy": 0.7717561342325581,
+      "loss": 0.7594,
+      "mean_token_accuracy": 0.7717538547987106,
       "step": 125
     },
     {
       "epoch": 0.7692307692307693,
-      "grad_norm": 0.32069400376048846,
+      "grad_norm": 0.32025277899891924,
       "learning_rate": 3.0765795095517026e-06,
-      "loss": 0.7624,
-      "mean_token_accuracy": 0.7705784584578896,
+      "loss": 0.7623,
+      "mean_token_accuracy": 0.7706147324391592,
       "step": 130
     },
     {
       "epoch": 0.7988165680473372,
-      "grad_norm": 0.30337427328877,
+      "grad_norm": 0.30317891871785263,
       "learning_rate": 2.369159318001937e-06,
       "loss": 0.7684,
-      "mean_token_accuracy": 0.7688614612364428,
+      "mean_token_accuracy": 0.7688305448428647,
       "step": 135
     },
     {
       "epoch": 0.8284023668639053,
-      "grad_norm": 0.2931589273225087,
+      "grad_norm": 0.29377898935847374,
       "learning_rate": 1.743160500034443e-06,
       "loss": 0.7597,
-      "mean_token_accuracy": 0.7717145410630937,
+      "mean_token_accuracy": 0.771730188624508,
       "step": 140
     },
     {
       "epoch": 0.8579881656804734,
-      "grad_norm": 0.31239776071011455,
+      "grad_norm": 0.3122803287793864,
       "learning_rate": 1.2052624879351105e-06,
-      "loss": 0.7595,
-      "mean_token_accuracy": 0.7714656640175818,
+      "loss": 0.7594,
+      "mean_token_accuracy": 0.7715123408941931,
       "step": 145
     },
     {
       "epoch": 0.8875739644970414,
-      "grad_norm": 0.2909480842066185,
+      "grad_norm": 0.289656138067616,
       "learning_rate": 7.612046748871327e-07,
       "loss": 0.7505,
-      "mean_token_accuracy": 0.7739669991508189,
+      "mean_token_accuracy": 0.7739841847801008,
       "step": 150
     },
     {
       "epoch": 0.9171597633136095,
-      "grad_norm": 0.279458821658001,
+      "grad_norm": 0.27970392255541077,
       "learning_rate": 4.1572517541747294e-07,
       "loss": 0.741,
-      "mean_token_accuracy": 0.7767323569336231,
+      "mean_token_accuracy": 0.7767522035827779,
       "step": 155
     },
     {
       "epoch": 0.9467455621301775,
-      "grad_norm": 0.29648687488650516,
+      "grad_norm": 0.2966574688397028,
       "learning_rate": 1.7251026952640583e-07,
       "loss": 0.754,
-      "mean_token_accuracy": 0.7732043043845805,
+      "mean_token_accuracy": 0.7732214833563628,
       "step": 160
     },
     {
       "epoch": 0.9763313609467456,
-      "grad_norm": 0.2773995248389471,
+      "grad_norm": 0.276771819219271,
       "learning_rate": 3.4155069933301535e-08,
-      "loss": 0.7467,
-      "mean_token_accuracy": 0.7749679952859235,
+      "loss": 0.7468,
+      "mean_token_accuracy": 0.774979807844946,
       "step": 165
     },
     {
       "epoch": 1.0,
-      "mean_token_accuracy": 0.7705127247007673,
+      "mean_token_accuracy": 0.7703761237331678,
       "step": 169,
       "total_flos": 76973799899136.0,
-      "train_loss": 0.8025032683942445,
-      "train_runtime": 710.0203,
-      "train_samples_per_second": 30.454,
-      "train_steps_per_second": 0.238
+      "train_loss": 0.802504962717993,
+      "train_runtime": 695.5303,
+      "train_samples_per_second": 31.089,
+      "train_steps_per_second": 0.243
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5e7043d0dcec036604186a342c05b4b1b9b359afc64da2f56ce61db35f0a4ae1
+oid sha256:c5e6be9345e2f96cd68e61867c69873588bdbf3fae2f77193ddc524b45a7ad6a
 size 7416