End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: apache-2.0
 base_model: mistralai/Mistral-7B-Instruct-v0.2
 tags:
+- alignment-handbook
 - trl
 - simpo
 - generated_from_trainer
+- trl
+- simpo
+- generated_from_trainer
+datasets:
+- princeton-nlp/mistral-instruct-ultrafeedback
 model-index:
 - name: simpo-exps
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # simpo-exps
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the princeton-nlp/mistral-instruct-ultrafeedback dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.7909
-- Rewards/chosen: -1.6908
-- Rewards/rejected: -1.8447
-- Rewards/accuracies: 0.6024
-- Rewards/margins: 0.1538
-- Logps/rejected: -0.7379
-- Logps/chosen: -0.6763
-- Logits/rejected: -2.7573
-- Logits/chosen: -2.7593
+- Loss: 0.7905
+- Rewards/chosen: -1.7087
+- Rewards/rejected: -1.8663
+- Rewards/accuracies: 0.6051
+- Rewards/margins: 0.1576
+- Logps/rejected: -0.7465
+- Logps/chosen: -0.6835
+- Logits/rejected: -2.7543
+- Logits/chosen: -2.7563
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 0.9987943737441393,
+    "eval_logits/chosen": -2.756258010864258,
+    "eval_logits/rejected": -2.754289388656616,
+    "eval_logps/chosen": -0.6834676861763,
+    "eval_logps/rejected": -0.7465164661407471,
+    "eval_loss": 0.7904670238494873,
+    "eval_rewards/accuracies": 0.605053186416626,
+    "eval_rewards/chosen": -1.7086691856384277,
+    "eval_rewards/margins": 0.15762200951576233,
+    "eval_rewards/rejected": -1.8662911653518677,
+    "eval_runtime": 102.313,
+    "eval_samples": 2994,
+    "eval_samples_per_second": 29.263,
+    "eval_steps_per_second": 1.837,
     "total_flos": 0.0,
     "train_loss": 0.7914171927476645,
     "train_runtime": 8402.7408,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 0.9987943737441393,
+    "eval_logits/chosen": -2.756258010864258,
+    "eval_logits/rejected": -2.754289388656616,
+    "eval_logps/chosen": -0.6834676861763,
+    "eval_logps/rejected": -0.7465164661407471,
+    "eval_loss": 0.7904670238494873,
+    "eval_rewards/accuracies": 0.605053186416626,
+    "eval_rewards/chosen": -1.7086691856384277,
+    "eval_rewards/margins": 0.15762200951576233,
+    "eval_rewards/rejected": -1.8662911653518677,
+    "eval_runtime": 102.313,
+    "eval_samples": 2994,
+    "eval_samples_per_second": 29.263,
+    "eval_steps_per_second": 1.837
+}