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: glm-4_6-freelancer-traces-pm
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # glm-4_6-freelancer-traces-pm
 
-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 DCAgent2/glm-4.6-freelancer-traces dataset.
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 1.757655959240902,
+    "achieved_tflops_per_gpu_theoretical": 145.67989620139622,
+    "epoch": 7.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.2028736174106598,
+    "mfu_percent": 0.5633512689874686,
+    "mfu_percent_theoretical": 46.69227442352443,
+    "total_flos": 7.99498208344277e+17,
+    "train_loss": 0.28994313101414193,
+    "train_runtime": 56858.2694,
+    "train_samples_per_second": 0.684,
+    "train_steps_per_second": 0.043,
+    "valid_targets_mean": 2723.6,
+    "valid_targets_min": 1612
+}

train_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 1.757655959240902,
+    "achieved_tflops_per_gpu_theoretical": 145.67989620139622,
+    "epoch": 7.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.2028736174106598,
+    "mfu_percent": 0.5633512689874686,
+    "mfu_percent_theoretical": 46.69227442352443,
+    "total_flos": 7.99498208344277e+17,
+    "train_loss": 0.28994313101414193,
+    "train_runtime": 56858.2694,
+    "train_samples_per_second": 0.684,
+    "train_steps_per_second": 0.043,
+    "valid_targets_mean": 2723.6,
+    "valid_targets_min": 1612
+}