End of training

README.md

@@ -3,11 +3,15 @@ library_name: transformers
 license: llama3.1
 base_model: meta-llama/Llama-3.1-8B
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- pragsri8/RLHFlow_ultrafeedback_iter1_vanilla_bon-sft
 model-index:
 - name: llama-3.1-8b-sft-full_vanilla_bon-sft
   results: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama-3.1-8b-sft-full_vanilla_bon-sft
 
-This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Llama-3.1-8B](https://huggingface.co/meta-llama/Llama-3.1-8B) on the pragsri8/RLHFlow_ultrafeedback_iter1_vanilla_bon-sft dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.3875
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 1.3874934911727905,
+    "eval_runtime": 10.447,
+    "eval_samples": 2000,
+    "eval_samples_per_second": 19.144,
+    "eval_steps_per_second": 0.67,
     "total_flos": 12353399685120.0,
     "train_loss": 1.5075374740665242,
     "train_runtime": 421.8974,

config.json

@@ -31,6 +31,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.49.0",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.3869625329971313,
-    "eval_runtime": 10.7262,
+    "eval_loss": 1.3874934911727905,
+    "eval_runtime": 10.447,
     "eval_samples": 2000,
-    "eval_samples_per_second": 18.646,
-    "eval_steps_per_second": 0.653
+    "eval_samples_per_second": 19.144,
+    "eval_steps_per_second": 0.67
 }

runs/Apr10_09-35-29_multi-objective-preference-optimization4/events.out.tfevents.1744278450.multi-objective-preference-optimization4.261071.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b20f6e3333ad3f2d3d3317738c7db7208805397fcc2b7f1288ad840e3aa3aae1
+size 354