End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: Qwen/Qwen3-8B
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: magicoder-evol-instruct-110k-sandboxes-traces-terminus-2_overwrite-output-dir_True
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # magicoder-evol-instruct-110k-sandboxes-traces-terminus-2_overwrite-output-dir_True
 
-This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on an unknown dataset.
+This model is a fine-tuned version of [Qwen/Qwen3-8B](https://huggingface.co/Qwen/Qwen3-8B) on the mlfoundations-dev/magicoder-evol-instruct-110k-sandboxes-traces-terminus-2 dataset.
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 2.4923871495979912,
+    "achieved_tflops_per_gpu_theoretical": 460.7550205439509,
+    "epoch": 5.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.24144230782985687,
+    "mfu_percent": 0.17614043460056475,
+    "mfu_percent_theoretical": 32.562192264590166,
+    "total_flos": 4.577212452957061e+17,
+    "train_loss": 0.45931496092906365,
+    "train_runtime": 11477.9832,
+    "train_samples_per_second": 4.348,
+    "train_steps_per_second": 0.136,
+    "valid_targets_mean": 2344.4,
+    "valid_targets_min": 827
+}

train_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 2.4923871495979912,
+    "achieved_tflops_per_gpu_theoretical": 460.7550205439509,
+    "epoch": 5.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.24144230782985687,
+    "mfu_percent": 0.17614043460056475,
+    "mfu_percent_theoretical": 32.562192264590166,
+    "total_flos": 4.577212452957061e+17,
+    "train_loss": 0.45931496092906365,
+    "train_runtime": 11477.9832,
+    "train_samples_per_second": 4.348,
+    "train_steps_per_second": 0.136,
+    "valid_targets_mean": 2344.4,
+    "valid_targets_min": 827
+}