Model save

README.md

@@ -1,11 +1,9 @@
 ---
-base_model: Qwen/Qwen2.5-1.5B-Instruct
-datasets: flyingbugs/OpenR1-Math-220k-pruned-keep-0.5-end-start-0.5
+base_model: Qwen/Qwen2.5-Math-1.5B-Instruct
 library_name: transformers
 model_name: Qwen2.5-1.5B-Open-R1-Distill-eos-epic
 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-eos-epic
 
-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 [flyingbugs/OpenR1-Math-220k-pruned-keep-0.5-end-start-0.5](https://huggingface.co/datasets/flyingbugs/OpenR1-Math-220k-pruned-keep-0.5-end-start-0.5) dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-Math-1.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-Math-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/jjh233/huggingface/runs/371aa1gg) 
+[<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/jjh233/huggingface/runs/3r9dxsmv) 
 
 
 This model was trained with SFT.
@@ -38,7 +36,7 @@ This model was trained with SFT.
 
 - TRL: 0.16.0.dev0
 - Transformers: 4.54.0
-- Pytorch: 2.5.1
+- Pytorch: 2.7.1
 - Datasets: 4.0.0
 - Tokenizers: 0.21.2
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 962656789856256.0,
-    "train_loss": 0.551490586363908,
-    "train_runtime": 11763.4703,
+    "train_loss": 0.0,
+    "train_runtime": 1.0579,
     "train_samples": 93733,
-    "train_samples_per_second": 5.731,
-    "train_steps_per_second": 0.045
+    "train_samples_per_second": 63710.085,
+    "train_steps_per_second": 499.109
 }

chat_template.jinja

@@ -3,7 +3,7 @@
     {%- if messages[0]['role'] == 'system' %}
         {{- messages[0]['content'] }}
     {%- else %}
-        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+        {{- 'Please reason step by step, and put your final answer within \\boxed{}.' }}
     {%- endif %}
     {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
     {%- for tool in tools %}
@@ -15,7 +15,7 @@
     {%- if messages[0]['role'] == 'system' %}
         {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
     {%- else %}
-        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+        {{- '<|im_start|>system\nPlease reason step by step, and put your final answer within \\boxed{}.<|im_end|>\n' }}
     {%- endif %}
 {%- endif %}
 {%- for message in messages %}

config.json

@@ -39,7 +39,7 @@
     "full_attention",
     "full_attention"
   ],
-  "max_position_embeddings": 32768,
+  "max_position_embeddings": 4096,
   "max_window_layers": 21,
   "model_type": "qwen2",
   "num_attention_heads": 12,
@@ -47,12 +47,12 @@
   "num_key_value_heads": 2,
   "rms_norm_eps": 1e-06,
   "rope_scaling": null,
-  "rope_theta": 1000000.0,
+  "rope_theta": 10000.0,
   "sliding_window": null,
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.54.0",
-  "use_cache": true,
+  "use_cache": false,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

generation_config.json

@@ -1,14 +1,9 @@
 {
   "bos_token_id": 151643,
-  "do_sample": true,
   "eos_token_id": [
     151645,
     151643
   ],
   "pad_token_id": 151643,
-  "repetition_penalty": 1.1,
-  "temperature": 0.7,
-  "top_k": 20,
-  "top_p": 0.8,
   "transformers_version": "4.54.0"
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "total_flos": 962656789856256.0,
-    "train_loss": 0.551490586363908,
-    "train_runtime": 11763.4703,
+    "train_loss": 0.0,
+    "train_runtime": 1.0579,
     "train_samples": 93733,
-    "train_samples_per_second": 5.731,
-    "train_steps_per_second": 0.045
+    "train_samples_per_second": 63710.085,
+    "train_steps_per_second": 499.109
 }

trainer_state.json

@@ -748,10 +748,10 @@
       "epoch": 3.0,
       "step": 528,
       "total_flos": 962656789856256.0,
-      "train_loss": 0.551490586363908,
-      "train_runtime": 11763.4703,
-      "train_samples_per_second": 5.731,
-      "train_steps_per_second": 0.045
+      "train_loss": 0.0,
+      "train_runtime": 1.0579,
+      "train_samples_per_second": 63710.085,
+      "train_steps_per_second": 499.109
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:36c73442ca5e71438beadd5486fd837e9558129a2755e982f46a6a2dda4c8ae2
-size 7544
+oid sha256:4b92e82046467d743bafc37b78b07bdde765260981b7b2c7ed14341172f5ea31
+size 8017