End of training

README.md

@@ -2,11 +2,15 @@
 license: apache-2.0
 base_model: Qwen/Qwen2-0.5B
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/ultrachat_200k
 model-index:
 - name: qwen2-0.5b-sft
   results: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhaoyang1/huggingface/runs/n1cgl3ti)
 # qwen2-0.5b-sft
 
-This model is a fine-tuned version of [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B) on the generator dataset.
+This model is a fine-tuned version of [Qwen/Qwen2-0.5B](https://huggingface.co/Qwen/Qwen2-0.5B) on the HuggingFaceH4/ultrachat_200k dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.5269
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9998676022772408,
+    "eval_loss": 1.526898980140686,
+    "eval_runtime": 437.3232,
+    "eval_samples": 23109,
+    "eval_samples_per_second": 61.154,
+    "eval_steps_per_second": 3.823,
     "total_flos": 106218135748608.0,
     "train_loss": 1.5447227579809852,
     "train_runtime": 15813.94,

config.json

@@ -23,7 +23,7 @@
   "tie_word_embeddings": true,
   "torch_dtype": "float16",
   "transformers_version": "4.42.0",
-  "use_cache": false,
+  "use_cache": true,
   "use_sliding_window": false,
   "vocab_size": 151646
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 0.999738425320429,
-    "eval_loss": 1.5079056024551392,
-    "eval_runtime": 417.8091,
+    "epoch": 0.9998676022772408,
+    "eval_loss": 1.526898980140686,
+    "eval_runtime": 437.3232,
     "eval_samples": 23109,
-    "eval_samples_per_second": 64.802,
-    "eval_steps_per_second": 4.052
+    "eval_samples_per_second": 61.154,
+    "eval_steps_per_second": 3.823
 }