End of training

README.md

@@ -1,9 +1,17 @@
 ---
 library_name: transformers
+license: apache-2.0
+base_model: Qwen/Qwen2.5-1.5B-Instruct
 tags:
+- alignment-handbook
 - trl
 - dpo
 - generated_from_trainer
+- trl
+- dpo
+- generated_from_trainer
+datasets:
+- YuchenLi01/MATH_Qwen2.5-1.5BInstruct_DPO_MoreUniqueResponseNoGTv3pair1
 model-index:
 - name: genv3pair1NoGT_1.5B_cdpo_lm1_ebs32_lr5e-07_beta0.4_epoch8.0_42
   results: []
@@ -14,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # genv3pair1NoGT_1.5B_cdpo_lm1_ebs32_lr5e-07_beta0.4_epoch8.0_42
 
-This model was trained from scratch on an unknown dataset.
+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 YuchenLi01/MATH_Qwen2.5-1.5BInstruct_DPO_MoreUniqueResponseNoGTv3pair1 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1826
-- Rewards/chosen: 4.3610
+- Loss: 0.1849
+- Rewards/chosen: 4.3318
 - Rewards/rejected: 0.0
 - Rewards/accuracies: 0.9250
-- Rewards/margins: 4.3610
-- Logps/rejected: -32.5126
-- Logps/chosen: -19.1676
-- Logits/rejected: -3.4642
-- Logits/chosen: -3.3984
+- Rewards/margins: 4.3318
+- Logps/rejected: -32.6177
+- Logps/chosen: -19.2407
+- Logits/rejected: -3.4640
+- Logits/chosen: -3.3979
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 8.0,
+    "eval_logits/chosen": -3.3978519439697266,
+    "eval_logits/rejected": -3.4640097618103027,
+    "eval_logps/chosen": -19.240652084350586,
+    "eval_logps/rejected": -32.617706298828125,
+    "eval_loss": 0.18487206101417542,
+    "eval_rewards/accuracies": 0.925000011920929,
+    "eval_rewards/chosen": 4.331756114959717,
+    "eval_rewards/margins": 4.331756114959717,
+    "eval_rewards/rejected": 0.0,
+    "eval_runtime": 27.7567,
+    "eval_samples": 310,
+    "eval_samples_per_second": 11.168,
+    "eval_steps_per_second": 0.36,
     "total_flos": 0.0,
     "train_loss": 0.05916496973182985,
     "train_runtime": 12200.893,

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.45.2",
-  "use_cache": false,
+  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151936
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 8.0,
+    "eval_logits/chosen": -3.3978519439697266,
+    "eval_logits/rejected": -3.4640097618103027,
+    "eval_logps/chosen": -19.240652084350586,
+    "eval_logps/rejected": -32.617706298828125,
+    "eval_loss": 0.18487206101417542,
+    "eval_rewards/accuracies": 0.925000011920929,
+    "eval_rewards/chosen": 4.331756114959717,
+    "eval_rewards/margins": 4.331756114959717,
+    "eval_rewards/rejected": 0.0,
+    "eval_runtime": 27.7567,
+    "eval_samples": 310,
+    "eval_samples_per_second": 11.168,
+    "eval_steps_per_second": 0.36
+}