End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: llama3
 base_model: meta-llama/Meta-Llama-3-8B-Instruct
 tags:
+- alignment-handbook
 - trl
 - simpo
 - generated_from_trainer
+- trl
+- simpo
+- generated_from_trainer
+datasets:
+- yakazimir/llama3-ultrafeedback-armorm
 model-index:
 - name: llama3_cpo_best_entropy
   results: []
@@ -16,18 +22,18 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3_cpo_best_entropy
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the yakazimir/llama3-ultrafeedback-armorm dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.7650
-- Rewards/chosen: -25.4120
-- Rewards/rejected: -32.9268
+- Loss: 1.7538
+- Rewards/chosen: -25.5611
+- Rewards/rejected: -33.0936
 - Rewards/accuracies: 0.8434
-- Rewards/margins: 7.5148
-- Logps/rejected: -3.2927
-- Logps/chosen: -2.5412
-- Logits/rejected: -1.1664
-- Logits/chosen: -1.2180
-- Semantic Entropy: 0.3711
+- Rewards/margins: 7.5325
+- Logps/rejected: -3.3094
+- Logps/chosen: -2.5561
+- Logits/rejected: -1.1508
+- Logits/chosen: -1.2004
+- Semantic Entropy: 0.3675
 
 ## Model description
 

all_results.json

@@ -1,5 +1,19 @@
 {
     "epoch": 0.9989071038251366,
+    "eval_logits/chosen": -1.200439691543579,
+    "eval_logits/rejected": -1.150753140449524,
+    "eval_logps/chosen": -2.5561068058013916,
+    "eval_logps/rejected": -3.309354782104492,
+    "eval_loss": 1.7538212537765503,
+    "eval_rewards/accuracies": 0.8433734774589539,
+    "eval_rewards/chosen": -25.561067581176758,
+    "eval_rewards/margins": 7.532484531402588,
+    "eval_rewards/rejected": -33.09355163574219,
+    "eval_runtime": 33.6933,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.118,
+    "eval_semantic_entropy": 0.36752766370773315,
+    "eval_steps_per_second": 2.463,
     "total_flos": 0.0,
     "train_loss": 2.585477126766347,
     "train_runtime": 5989.402,

config.json

@@ -24,6 +24,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -0,0 +1,17 @@
+{
+    "epoch": 0.9989071038251366,
+    "eval_logits/chosen": -1.200439691543579,
+    "eval_logits/rejected": -1.150753140449524,
+    "eval_logps/chosen": -2.5561068058013916,
+    "eval_logps/rejected": -3.309354782104492,
+    "eval_loss": 1.7538212537765503,
+    "eval_rewards/accuracies": 0.8433734774589539,
+    "eval_rewards/chosen": -25.561067581176758,
+    "eval_rewards/margins": 7.532484531402588,
+    "eval_rewards/rejected": -33.09355163574219,
+    "eval_runtime": 33.6933,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.118,
+    "eval_semantic_entropy": 0.36752766370773315,
+    "eval_steps_per_second": 2.463
+}