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: staqc-ot3-100k-math-subset-traces-terminus-2_save-strategy_steps_Qwen3-8B
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # staqc-ot3-100k-math-subset-traces-terminus-2_save-strategy_steps_Qwen3-8B
 
-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 DCAgent/staqc-ot3-100k-math-subset-traces-terminus-2 dataset.
 
 ## Model description
 

all_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 11.07456719317637,
+    "achieved_tflops_per_gpu_theoretical": 670.8051034389872,
+    "epoch": 5.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.4211060404777527,
+    "mfu_percent": 3.549540767043708,
+    "mfu_percent_theoretical": 215.00163571762408,
+    "total_flos": 1.1298365271930044e+19,
+    "train_loss": 0.22232198994150273,
+    "train_runtime": 31881.5091,
+    "train_samples_per_second": 12.696,
+    "train_steps_per_second": 0.198,
+    "valid_targets_mean": 15457.9,
+    "valid_targets_min": 697
+}

train_results.json

@@ -0,0 +1,16 @@
+{
+    "achieved_tflops_per_gpu": 11.07456719317637,
+    "achieved_tflops_per_gpu_theoretical": 670.8051034389872,
+    "epoch": 5.0,
+    "loss_nan_ranks": 0,
+    "loss_rank_avg": 0.4211060404777527,
+    "mfu_percent": 3.549540767043708,
+    "mfu_percent_theoretical": 215.00163571762408,
+    "total_flos": 1.1298365271930044e+19,
+    "train_loss": 0.22232198994150273,
+    "train_runtime": 31881.5091,
+    "train_samples_per_second": 12.696,
+    "train_steps_per_second": 0.198,
+    "valid_targets_mean": 15457.9,
+    "valid_targets_min": 697
+}