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: llama3instruct_-l5-10-0_3-1e-6-2_best
   results: []
@@ -16,17 +22,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama3instruct_-l5-10-0_3-1e-6-2_best
 
-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: 1.2716
-- Rewards/chosen: -7.9331
-- Rewards/rejected: -19.2874
-- Rewards/accuracies: 0.8614
-- Rewards/margins: 11.3543
-- Logps/rejected: -1.9287
-- Logps/chosen: -0.7933
-- Logits/rejected: -1.2977
-- Logits/chosen: -1.3755
+- Loss: 1.2685
+- Rewards/chosen: -8.0711
+- Rewards/rejected: -19.7238
+- Rewards/accuracies: 0.8675
+- Rewards/margins: 11.6527
+- Logps/rejected: -1.9724
+- Logps/chosen: -0.8071
+- Logits/rejected: -1.3327
+- Logits/chosen: -1.4140
 
 ## Model description
 

all_results.json

@@ -1,5 +1,18 @@
 {
     "epoch": 1.9978142076502732,
+    "eval_logits/chosen": -1.4140465259552002,
+    "eval_logits/rejected": -1.3327398300170898,
+    "eval_logps/chosen": -0.8071117401123047,
+    "eval_logps/rejected": -1.9723843336105347,
+    "eval_loss": 1.2685352563858032,
+    "eval_rewards/accuracies": 0.8674699068069458,
+    "eval_rewards/chosen": -8.071117401123047,
+    "eval_rewards/margins": 11.652724266052246,
+    "eval_rewards/rejected": -19.723840713500977,
+    "eval_runtime": 33.7378,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.066,
+    "eval_steps_per_second": 2.46,
     "total_flos": 0.0,
     "train_loss": 1.4686054226084402,
     "train_runtime": 12099.7941,

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 +1,16 @@
+{
+    "epoch": 1.9978142076502732,
+    "eval_logits/chosen": -1.4140465259552002,
+    "eval_logits/rejected": -1.3327398300170898,
+    "eval_logps/chosen": -0.8071117401123047,
+    "eval_logps/rejected": -1.9723843336105347,
+    "eval_loss": 1.2685352563858032,
+    "eval_rewards/accuracies": 0.8674699068069458,
+    "eval_rewards/chosen": -8.071117401123047,
+    "eval_rewards/margins": 11.652724266052246,
+    "eval_rewards/rejected": -19.723840713500977,
+    "eval_runtime": 33.7378,
+    "eval_samples": 1318,
+    "eval_samples_per_second": 39.066,
+    "eval_steps_per_second": 2.46
+}