End of training

README.md

@@ -1,4 +1,6 @@
 ---
+language:
+- en
 tags:
 - generated_from_trainer
 datasets:
@@ -12,7 +14,7 @@ model-index:
       name: Text Classification
       type: text-classification
     dataset:
-      name: glue
+      name: GLUE STSB
       type: glue
       config: stsb
       split: validation
@@ -20,7 +22,7 @@ model-index:
     metrics:
     - name: Spearmanr
       type: spearmanr
-      value: 0.04713596164794335
+      value: 0.05749928795262334
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -28,12 +30,12 @@ should probably proofread and complete it, then remove this comment. -->
 
 # add_BERT_no_pretrain_stsb
 
-This model is a fine-tuned version of [](https://huggingface.co/) on the glue dataset.
+This model is a fine-tuned version of [](https://huggingface.co/) on the GLUE STSB dataset.
 It achieves the following results on the evaluation set:
-- Loss: 2.4727
-- Pearson: 0.0540
-- Spearmanr: 0.0471
-- Combined Score: 0.0506
+- Loss: 2.3307
+- Pearson: 0.0719
+- Spearmanr: 0.0575
+- Combined Score: 0.0647
 
 ## Model description
 

all_results.json

@@ -1,16 +1,16 @@
 {
     "epoch": 11.0,
-    "eval_combined_score": 0.016057292835545172,
-    "eval_loss": 2.252106189727783,
-    "eval_pearson": 0.014231574810207417,
-    "eval_runtime": 2.3308,
+    "eval_combined_score": 0.06472344315426198,
+    "eval_loss": 2.3307483196258545,
+    "eval_pearson": 0.07194759835590063,
+    "eval_runtime": 2.8288,
     "eval_samples": 1500,
-    "eval_samples_per_second": 643.549,
-    "eval_spearmanr": 0.017883010860882925,
-    "eval_steps_per_second": 5.148,
-    "train_loss": 2.6108932957504734,
-    "train_runtime": 375.1993,
+    "eval_samples_per_second": 530.257,
+    "eval_spearmanr": 0.05749928795262334,
+    "eval_steps_per_second": 4.242,
+    "train_loss": 2.1222905785146384,
+    "train_runtime": 469.3728,
     "train_samples": 5749,
-    "train_samples_per_second": 766.126,
-    "train_steps_per_second": 5.997
+    "train_samples_per_second": 612.413,
+    "train_steps_per_second": 4.794
 }

eval_results.json

@@ -1,11 +1,11 @@
 {
     "epoch": 11.0,
-    "eval_combined_score": 0.016057292835545172,
-    "eval_loss": 2.252106189727783,
-    "eval_pearson": 0.014231574810207417,
-    "eval_runtime": 2.3308,
+    "eval_combined_score": 0.06472344315426198,
+    "eval_loss": 2.3307483196258545,
+    "eval_pearson": 0.07194759835590063,
+    "eval_runtime": 2.8288,
     "eval_samples": 1500,
-    "eval_samples_per_second": 643.549,
-    "eval_spearmanr": 0.017883010860882925,
-    "eval_steps_per_second": 5.148
+    "eval_samples_per_second": 530.257,
+    "eval_spearmanr": 0.05749928795262334,
+    "eval_steps_per_second": 4.242
 }

logs/events.out.tfevents.1686776529.serv-9219.1715196.13

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6775b3aff7f6b7b54d83856d616cbf0243a52ed1f00b9076a46216ead8a84c83
+size 473

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 11.0,
-    "train_loss": 2.6108932957504734,
-    "train_runtime": 375.1993,
+    "train_loss": 2.1222905785146384,
+    "train_runtime": 469.3728,
     "train_samples": 5749,
-    "train_samples_per_second": 766.126,
-    "train_steps_per_second": 5.997
+    "train_samples_per_second": 612.413,
+    "train_steps_per_second": 4.794
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 2.252106189727783,
+  "best_metric": 2.3307483196258545,
   "best_model_checkpoint": "add_BERT_no_pretrain_stsb/checkpoint-270",
   "epoch": 11.0,
   "global_step": 495,
@@ -9,199 +9,199 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "learning_rate": 0.0004915555555555556,
-      "loss": 6.3185,
+      "learning_rate": 3.9200000000000004e-05,
+      "loss": 2.4937,
       "step": 45
     },
     {
       "epoch": 1.0,
-      "eval_combined_score": -0.0010166901927902152,
-      "eval_loss": 2.764800786972046,
-      "eval_pearson": 0.0033488460201501727,
-      "eval_runtime": 2.3463,
-      "eval_samples_per_second": 639.317,
-      "eval_spearmanr": -0.005382226405730603,
-      "eval_steps_per_second": 5.115,
+      "eval_combined_score": 0.03484396955490256,
+      "eval_loss": 2.637270927429199,
+      "eval_pearson": 0.041015241553137224,
+      "eval_runtime": 2.7912,
+      "eval_samples_per_second": 537.4,
+      "eval_spearmanr": 0.028672697556667882,
+      "eval_steps_per_second": 4.299,
       "step": 45
     },
     {
       "epoch": 2.0,
-      "learning_rate": 0.0004815555555555556,
-      "loss": 2.2915,
+      "learning_rate": 3.8400000000000005e-05,
+      "loss": 2.22,
       "step": 90
     },
     {
       "epoch": 2.0,
-      "eval_combined_score": 0.015394997836546201,
-      "eval_loss": 2.2692453861236572,
-      "eval_pearson": 0.020749248741372028,
-      "eval_runtime": 2.337,
-      "eval_samples_per_second": 641.849,
-      "eval_spearmanr": 0.010040746931720372,
-      "eval_steps_per_second": 5.135,
+      "eval_combined_score": 0.06629060639069365,
+      "eval_loss": 2.4288177490234375,
+      "eval_pearson": 0.07402195261812716,
+      "eval_runtime": 2.8138,
+      "eval_samples_per_second": 533.087,
+      "eval_spearmanr": 0.058559260163260116,
+      "eval_steps_per_second": 4.265,
       "step": 90
     },
     {
       "epoch": 3.0,
-      "learning_rate": 0.0004715555555555556,
-      "loss": 2.2747,
+      "learning_rate": 3.76e-05,
+      "loss": 2.1554,
       "step": 135
     },
     {
       "epoch": 3.0,
-      "eval_combined_score": 0.010336013790419181,
-      "eval_loss": 2.362309217453003,
-      "eval_pearson": 0.016689666809067433,
-      "eval_runtime": 2.3411,
-      "eval_samples_per_second": 640.734,
-      "eval_spearmanr": 0.003982360771770928,
-      "eval_steps_per_second": 5.126,
+      "eval_combined_score": 0.055355126341372324,
+      "eval_loss": 2.3868918418884277,
+      "eval_pearson": 0.06089199058890605,
+      "eval_runtime": 2.8061,
+      "eval_samples_per_second": 534.558,
+      "eval_spearmanr": 0.04981826209383859,
+      "eval_steps_per_second": 4.276,
       "step": 135
     },
     {
       "epoch": 4.0,
-      "learning_rate": 0.00046155555555555555,
-      "loss": 2.2372,
+      "learning_rate": 3.680000000000001e-05,
+      "loss": 2.0556,
       "step": 180
     },
     {
       "epoch": 4.0,
-      "eval_combined_score": 0.006720541376488365,
-      "eval_loss": 2.883573293685913,
-      "eval_pearson": 0.009008075092391842,
-      "eval_runtime": 2.342,
-      "eval_samples_per_second": 640.473,
-      "eval_spearmanr": 0.004433007660584889,
-      "eval_steps_per_second": 5.124,
+      "eval_combined_score": 0.07587737081449714,
+      "eval_loss": 2.5739591121673584,
+      "eval_pearson": 0.0800115043320288,
+      "eval_runtime": 2.8092,
+      "eval_samples_per_second": 533.959,
+      "eval_spearmanr": 0.07174323729696545,
+      "eval_steps_per_second": 4.272,
       "step": 180
     },
     {
       "epoch": 5.0,
-      "learning_rate": 0.0004515555555555556,
-      "loss": 2.2573,
+      "learning_rate": 3.6e-05,
+      "loss": 2.0221,
       "step": 225
     },
     {
       "epoch": 5.0,
-      "eval_combined_score": 0.035108956445297795,
-      "eval_loss": 2.2528281211853027,
-      "eval_pearson": 0.03370956593774324,
-      "eval_runtime": 2.3442,
-      "eval_samples_per_second": 639.885,
-      "eval_spearmanr": 0.03650834695285234,
-      "eval_steps_per_second": 5.119,
+      "eval_combined_score": 0.07517103861970047,
+      "eval_loss": 2.465646266937256,
+      "eval_pearson": 0.08491360492393328,
+      "eval_runtime": 2.8092,
+      "eval_samples_per_second": 533.96,
+      "eval_spearmanr": 0.06542847231546765,
+      "eval_steps_per_second": 4.272,
       "step": 225
     },
     {
       "epoch": 6.0,
-      "learning_rate": 0.00044155555555555555,
-      "loss": 2.1979,
+      "learning_rate": 3.52e-05,
+      "loss": 2.0403,
       "step": 270
     },
     {
       "epoch": 6.0,
-      "eval_combined_score": 0.016057292835545172,
-      "eval_loss": 2.252106189727783,
-      "eval_pearson": 0.014231574810207417,
-      "eval_runtime": 2.3413,
-      "eval_samples_per_second": 640.679,
-      "eval_spearmanr": 0.017883010860882925,
-      "eval_steps_per_second": 5.125,
+      "eval_combined_score": 0.06472344315426198,
+      "eval_loss": 2.3307483196258545,
+      "eval_pearson": 0.07194759835590063,
+      "eval_runtime": 2.8052,
+      "eval_samples_per_second": 534.716,
+      "eval_spearmanr": 0.05749928795262334,
+      "eval_steps_per_second": 4.278,
       "step": 270
     },
     {
       "epoch": 7.0,
-      "learning_rate": 0.0004315555555555556,
-      "loss": 2.2244,
+      "learning_rate": 3.44e-05,
+      "loss": 2.1732,
       "step": 315
     },
     {
       "epoch": 7.0,
-      "eval_combined_score": 0.01728166633294926,
-      "eval_loss": 2.3162295818328857,
-      "eval_pearson": 0.015679956373000268,
-      "eval_runtime": 2.3435,
-      "eval_samples_per_second": 640.074,
-      "eval_spearmanr": 0.018883376292898248,
-      "eval_steps_per_second": 5.121,
+      "eval_combined_score": 0.06413160396195583,
+      "eval_loss": 2.5174152851104736,
+      "eval_pearson": 0.06991174129088629,
+      "eval_runtime": 2.8054,
+      "eval_samples_per_second": 534.68,
+      "eval_spearmanr": 0.058351466633025366,
+      "eval_steps_per_second": 4.277,
       "step": 315
     },
     {
       "epoch": 8.0,
-      "learning_rate": 0.0004215555555555556,
-      "loss": 2.1832,
+      "learning_rate": 3.3600000000000004e-05,
+      "loss": 2.0399,
       "step": 360
     },
     {
       "epoch": 8.0,
-      "eval_combined_score": 0.002281605635553319,
-      "eval_loss": 2.3739013671875,
-      "eval_pearson": 0.0006456617990473748,
-      "eval_runtime": 2.3491,
-      "eval_samples_per_second": 638.533,
-      "eval_spearmanr": 0.003917549472059263,
-      "eval_steps_per_second": 5.108,
+      "eval_combined_score": 0.06615269419943948,
+      "eval_loss": 2.5648128986358643,
+      "eval_pearson": 0.07178384296600182,
+      "eval_runtime": 2.8071,
+      "eval_samples_per_second": 534.35,
+      "eval_spearmanr": 0.06052154543287713,
+      "eval_steps_per_second": 4.275,
       "step": 360
     },
     {
       "epoch": 9.0,
-      "learning_rate": 0.0004115555555555556,
-      "loss": 2.3249,
+      "learning_rate": 3.28e-05,
+      "loss": 2.0765,
       "step": 405
     },
     {
       "epoch": 9.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.3828952312469482,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.3478,
-      "eval_samples_per_second": 638.891,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 5.111,
+      "eval_combined_score": 0.05558546651483845,
+      "eval_loss": 2.3373327255249023,
+      "eval_pearson": 0.06209781802846023,
+      "eval_runtime": 2.8097,
+      "eval_samples_per_second": 533.863,
+      "eval_spearmanr": 0.04907311500121666,
+      "eval_steps_per_second": 4.271,
       "step": 405
     },
     {
       "epoch": 10.0,
-      "learning_rate": 0.00040155555555555555,
-      "loss": 2.1956,
+      "learning_rate": 3.2000000000000005e-05,
+      "loss": 2.0538,
       "step": 450
     },
     {
       "epoch": 10.0,
-      "eval_combined_score": NaN,
-      "eval_loss": 2.3083081245422363,
-      "eval_pearson": NaN,
-      "eval_runtime": 2.3435,
-      "eval_samples_per_second": 640.08,
-      "eval_spearmanr": NaN,
-      "eval_steps_per_second": 5.121,
+      "eval_combined_score": 0.04471248717433318,
+      "eval_loss": 2.6402227878570557,
+      "eval_pearson": 0.04634878343613905,
+      "eval_runtime": 2.8018,
+      "eval_samples_per_second": 535.361,
+      "eval_spearmanr": 0.04307619091252732,
+      "eval_steps_per_second": 4.283,
       "step": 450
     },
     {
       "epoch": 11.0,
-      "learning_rate": 0.0003915555555555556,
-      "loss": 2.2148,
+      "learning_rate": 3.1200000000000006e-05,
+      "loss": 2.0147,
       "step": 495
     },
     {
       "epoch": 11.0,
-      "eval_combined_score": -0.027554660458158586,
-      "eval_loss": 2.2706003189086914,
-      "eval_pearson": -0.028324243354752993,
-      "eval_runtime": 2.361,
-      "eval_samples_per_second": 635.329,
-      "eval_spearmanr": -0.026785077561564178,
-      "eval_steps_per_second": 5.083,
+      "eval_combined_score": 0.05055556341697049,
+      "eval_loss": 2.4726502895355225,
+      "eval_pearson": 0.05397516518599763,
+      "eval_runtime": 2.8161,
+      "eval_samples_per_second": 532.645,
+      "eval_spearmanr": 0.04713596164794335,
+      "eval_steps_per_second": 4.261,
       "step": 495
     },
     {
       "epoch": 11.0,
       "step": 495,
       "total_flos": 9031970478620672.0,
-      "train_loss": 2.6108932957504734,
-      "train_runtime": 375.1993,
-      "train_samples_per_second": 766.126,
-      "train_steps_per_second": 5.997
+      "train_loss": 2.1222905785146384,
+      "train_runtime": 469.3728,
+      "train_samples_per_second": 612.413,
+      "train_steps_per_second": 4.794
     }
   ],
   "max_steps": 2250,