End of training

README.md

@@ -3,12 +3,16 @@ library_name: transformers
 license: apache-2.0
 base_model: alignment-handbook/zephyr-7b-sft-full
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - alignment-handbook
 - generated_from_trainer
 datasets:
-- generator
+- scottsuk0306/DepthQA
 model-index:
 - name: zephyr-7b-stem-half
   results: []
@@ -19,9 +23,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # zephyr-7b-stem-half
 
-This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the generator dataset.
+This model is a fine-tuned version of [alignment-handbook/zephyr-7b-sft-full](https://huggingface.co/alignment-handbook/zephyr-7b-sft-full) on the scottsuk0306/DepthQA dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.1558
+- Loss: 0.1468
 
 ## Model description
 

all_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 10.0,
     "eval_loss": 0.14684411883354187,
-    "eval_runtime": 2.6647,
+    "eval_runtime": 4.0784,
     "eval_samples": 424,
-    "eval_samples_per_second": 7.13,
-    "eval_steps_per_second": 0.375,
+    "eval_samples_per_second": 4.659,
+    "eval_steps_per_second": 0.245,
     "total_flos": 8375186227200.0,
     "train_loss": 0.0,
     "train_runtime": 1.6437,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
     "eval_loss": 0.14684411883354187,
-    "eval_runtime": 2.6647,
+    "eval_runtime": 4.0784,
     "eval_samples": 424,
-    "eval_samples_per_second": 7.13,
-    "eval_steps_per_second": 0.375
+    "eval_samples_per_second": 4.659,
+    "eval_steps_per_second": 0.245
 }