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_orpo_best_entropy
   results: []
@@ -16,18 +22,18 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3_orpo_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: 2.5608
-- Rewards/chosen: -12.8797
-- Rewards/rejected: -17.3972
+- Loss: 2.5561
+- Rewards/chosen: -12.9600
+- Rewards/rejected: -17.5108
 - Rewards/accuracies: 0.8072
-- Rewards/margins: 4.5175
-- Logps/rejected: -1.7397
-- Logps/chosen: -1.2880
-- Logits/rejected: -1.3473
-- Logits/chosen: -1.3813
-- Semantic Entropy: 0.7719
+- Rewards/margins: 4.5509
+- Logps/rejected: -1.7511
+- Logps/chosen: -1.2960
+- Logits/rejected: -1.3511
+- Logits/chosen: -1.3851
+- Semantic Entropy: 0.7683
 
 ## Model description
 

all_results.json

@@ -1,5 +1,19 @@
 {
     "epoch": 0.9989071038251366,
+    "eval_logits/chosen": -1.3851122856140137,
+    "eval_logits/rejected": -1.3511338233947754,
+    "eval_logps/chosen": -1.2959951162338257,
+    "eval_logps/rejected": -1.7510826587677002,
+    "eval_loss": 2.5561344623565674,
+    "eval_rewards/accuracies": 0.8072289228439331,
+    "eval_rewards/chosen": -12.959952354431152,
+    "eval_rewards/margins": 4.550872325897217,
+    "eval_rewards/rejected": -17.51082420349121,
+    "eval_runtime": 33.4053,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.455,
+    "eval_semantic_entropy": 0.7683338522911072,
+    "eval_steps_per_second": 2.485,
     "total_flos": 0.0,
     "train_loss": 3.4314852449513107,
     "train_runtime": 5934.8281,

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": -1.3851122856140137,
+    "eval_logits/rejected": -1.3511338233947754,
+    "eval_logps/chosen": -1.2959951162338257,
+    "eval_logps/rejected": -1.7510826587677002,
+    "eval_loss": 2.5561344623565674,
+    "eval_rewards/accuracies": 0.8072289228439331,
+    "eval_rewards/chosen": -12.959952354431152,
+    "eval_rewards/margins": 4.550872325897217,
+    "eval_rewards/rejected": -17.51082420349121,
+    "eval_runtime": 33.4053,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.455,
+    "eval_semantic_entropy": 0.7683338522911072,
+    "eval_steps_per_second": 2.485
+}