End of training

README.md

@@ -4,6 +4,7 @@ library_name: transformers
 model_name: llama_tuned
 tags:
 - generated_from_trainer
+- alignment-handbook
 - trl
 - sft
 licence: license

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 39.89333333333333,
+    "eval_loss": 2.1213064193725586,
+    "eval_runtime": 12.4292,
+    "eval_samples": 1124,
+    "eval_samples_per_second": 90.432,
+    "eval_steps_per_second": 15.126,
     "total_flos": 1.743333613371392e+18,
     "train_loss": 0.10288844243324496,
     "train_runtime": 9416.1259,

config.json

@@ -36,6 +36,6 @@
   "tie_word_embeddings": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.46.3",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 40.0,
-    "eval_loss": 0.0030517119448632,
-    "eval_runtime": 0.1724,
-    "eval_samples": 10,
-    "eval_samples_per_second": 58.01,
-    "eval_steps_per_second": 11.602
+    "epoch": 39.89333333333333,
+    "eval_loss": 2.1213064193725586,
+    "eval_runtime": 12.4292,
+    "eval_samples": 1124,
+    "eval_samples_per_second": 90.432,
+    "eval_steps_per_second": 15.126
 }