End of training

README.md

@@ -2,12 +2,13 @@
 license: llama3
 library_name: peft
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
 base_model: meta-llama/Meta-Llama-3-8B
 datasets:
-- generator
+- preference-data
 model-index:
 - name: downstream_0.1p_seed42_level2_syntax
   results: []
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # downstream_0.1p_seed42_level2_syntax
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) on the preference-data dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.0367
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 1.0366965532302856,
+    "eval_runtime": 2.4052,
+    "eval_samples": 10,
+    "eval_samples_per_second": 2.91,
+    "eval_steps_per_second": 0.832,
     "total_flos": 1.4408836278386688e+16,
     "train_loss": 1.131237539688396,
     "train_runtime": 17315.3957,

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.9993876301285977,
-    "eval_loss": 1.037530541419983,
-    "eval_runtime": 2.3375,
+    "epoch": 1.0,
+    "eval_loss": 1.0366965532302856,
+    "eval_runtime": 2.4052,
     "eval_samples": 10,
-    "eval_samples_per_second": 2.995,
-    "eval_steps_per_second": 0.856
+    "eval_samples_per_second": 2.91,
+    "eval_steps_per_second": 0.832
 }