End of training

README.md

@@ -4,6 +4,7 @@ license: apache-2.0
 base_model: Qwen/Qwen2.5-7B-Instruct
 tags:
 - llama-factory
+- full
 - generated_from_trainer
 model-index:
 - name: ds_no_offload_liger_packing_torchcompile
@@ -15,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # ds_no_offload_liger_packing_torchcompile
 
-This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on an unknown dataset.
+This model is a fine-tuned version of [Qwen/Qwen2.5-7B-Instruct](https://huggingface.co/Qwen/Qwen2.5-7B-Instruct) on the mlfoundations-dev/wikipedia_seed_science dataset.
 
 ## Model description
 

all_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 6.208039628360909e+16,
-    "train_loss": 1.2118252846929762,
-    "train_runtime": 1204.5753,
-    "train_samples_per_second": 1.432,
-    "train_steps_per_second": 0.015
+    "total_flos": 6.177445717318042e+16,
+    "train_loss": 1.0301644206047058,
+    "train_runtime": 1348.7564,
+    "train_samples_per_second": 1.279,
+    "train_steps_per_second": 0.013
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "total_flos": 6.208039628360909e+16,
-    "train_loss": 1.2118252846929762,
-    "train_runtime": 1204.5753,
-    "train_samples_per_second": 1.432,
-    "train_steps_per_second": 0.015
+    "total_flos": 6.177445717318042e+16,
+    "train_loss": 1.0301644206047058,
+    "train_runtime": 1348.7564,
+    "train_samples_per_second": 1.279,
+    "train_steps_per_second": 0.013
 }

trainer_state.json

@@ -10,138 +10,138 @@
   "log_history": [
     {
       "epoch": 0.16666666666666666,
-      "grad_norm": 0.9121807009354492,
+      "grad_norm": 8.619477723670265,
       "learning_rate": 5e-06,
-      "loss": 1.2823,
+      "loss": 1.2879,
       "step": 1
     },
     {
       "epoch": 0.3333333333333333,
-      "grad_norm": 0.9764893050786124,
+      "grad_norm": 9.185120522478208,
       "learning_rate": 1e-05,
-      "loss": 1.3191,
+      "loss": 1.3268,
       "step": 2
     },
     {
       "epoch": 0.5,
-      "grad_norm": 0.9284384945644188,
+      "grad_norm": 6.222909291903273,
       "learning_rate": 9.903926402016153e-06,
-      "loss": 1.287,
+      "loss": 1.2078,
       "step": 3
     },
     {
       "epoch": 0.6666666666666666,
-      "grad_norm": 0.8582970971638989,
+      "grad_norm": 3.023033299660152,
       "learning_rate": 9.619397662556434e-06,
-      "loss": 1.2686,
+      "loss": 1.1302,
       "step": 4
     },
     {
       "epoch": 0.8333333333333334,
-      "grad_norm": 0.8635519627747134,
+      "grad_norm": 6.084361758324089,
       "learning_rate": 9.157348061512728e-06,
-      "loss": 1.294,
+      "loss": 1.1449,
       "step": 5
     },
     {
       "epoch": 1.0,
-      "grad_norm": 0.7560256587528488,
+      "grad_norm": 6.651003657867708,
       "learning_rate": 8.535533905932739e-06,
-      "loss": 1.2551,
+      "loss": 1.085,
       "step": 6
     },
     {
       "epoch": 1.1666666666666667,
-      "grad_norm": 0.6624523616926192,
+      "grad_norm": 6.829832541351538,
       "learning_rate": 7.777851165098012e-06,
-      "loss": 1.2267,
+      "loss": 1.07,
       "step": 7
     },
     {
       "epoch": 1.3333333333333333,
-      "grad_norm": 0.5840695781359019,
+      "grad_norm": 4.929064829376609,
       "learning_rate": 6.913417161825449e-06,
-      "loss": 1.2174,
+      "loss": 1.0365,
       "step": 8
     },
     {
       "epoch": 1.5,
-      "grad_norm": 0.5208348821330916,
+      "grad_norm": 3.1841423652622267,
       "learning_rate": 5.975451610080643e-06,
-      "loss": 1.2134,
+      "loss": 1.0099,
       "step": 9
     },
     {
       "epoch": 1.6666666666666665,
-      "grad_norm": 0.4477585736402009,
+      "grad_norm": 2.047054438760332,
       "learning_rate": 5e-06,
-      "loss": 1.1716,
+      "loss": 0.9637,
       "step": 10
     },
     {
       "epoch": 1.8333333333333335,
-      "grad_norm": 0.4210137832048049,
+      "grad_norm": 2.187454406819176,
       "learning_rate": 4.02454838991936e-06,
-      "loss": 1.1606,
+      "loss": 0.933,
       "step": 11
     },
     {
       "epoch": 2.0,
-      "grad_norm": 0.39480845230736955,
+      "grad_norm": 2.0046247169383773,
       "learning_rate": 3.0865828381745515e-06,
-      "loss": 1.1714,
+      "loss": 0.9329,
       "step": 12
     },
     {
       "epoch": 2.1666666666666665,
-      "grad_norm": 0.3736519264684104,
+      "grad_norm": 1.6402479695299617,
       "learning_rate": 2.2221488349019903e-06,
-      "loss": 1.1518,
+      "loss": 0.9097,
       "step": 13
     },
     {
       "epoch": 2.3333333333333335,
-      "grad_norm": 0.36453793415908753,
+      "grad_norm": 1.3205807080108158,
       "learning_rate": 1.4644660940672628e-06,
-      "loss": 1.1449,
+      "loss": 0.8906,
       "step": 14
     },
     {
       "epoch": 2.5,
-      "grad_norm": 0.35639694150571866,
+      "grad_norm": 1.1426770548487395,
       "learning_rate": 8.426519384872733e-07,
-      "loss": 1.167,
+      "loss": 0.9138,
       "step": 15
     },
     {
       "epoch": 2.6666666666666665,
-      "grad_norm": 0.37214055458414597,
+      "grad_norm": 0.9087724744338759,
       "learning_rate": 3.8060233744356634e-07,
-      "loss": 1.1749,
+      "loss": 0.9051,
       "step": 16
     },
     {
       "epoch": 2.8333333333333335,
-      "grad_norm": 0.34523665157571026,
+      "grad_norm": 0.7725544929729385,
       "learning_rate": 9.607359798384785e-08,
-      "loss": 1.1597,
+      "loss": 0.9021,
       "step": 17
     },
     {
       "epoch": 3.0,
-      "grad_norm": 0.34261372184850886,
+      "grad_norm": 0.7028118802519975,
       "learning_rate": 0.0,
-      "loss": 1.1474,
+      "loss": 0.8932,
       "step": 18
     },
     {
       "epoch": 3.0,
       "step": 18,
-      "total_flos": 6.208039628360909e+16,
-      "train_loss": 1.2118252846929762,
-      "train_runtime": 1204.5753,
-      "train_samples_per_second": 1.432,
-      "train_steps_per_second": 0.015
+      "total_flos": 6.177445717318042e+16,
+      "train_loss": 1.0301644206047058,
+      "train_runtime": 1348.7564,
+      "train_samples_per_second": 1.279,
+      "train_steps_per_second": 0.013
     }
   ],
   "logging_steps": 1.0,
@@ -161,7 +161,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6.208039628360909e+16,
+  "total_flos": 6.177445717318042e+16,
   "train_batch_size": 1,
   "trial_name": null,
   "trial_params": null