Training in progress, step 200

all_results.json

@@ -1,18 +1,18 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 97.5,
-    "eval_average_metrics": 97.5,
-    "eval_loss": 0.03714486584067345,
-    "eval_runtime": 5.7314,
-    "eval_samples_per_second": 174.477,
-    "test_accuracy": 93.34862385321101,
-    "test_average_metrics": 93.34862385321101,
-    "test_loss": 0.07569558918476105,
-    "test_runtime": 5.6934,
-    "test_samples_per_second": 153.159,
-    "train_loss": 0.058686515289424354,
-    "train_runtime": 1966.0928,
+    "eval_accuracy": 60.099999999999994,
+    "eval_average_metrics": 60.099999999999994,
+    "eval_loss": 0.2783910036087036,
+    "eval_runtime": 3.2954,
+    "eval_samples_per_second": 303.458,
+    "test_accuracy": 58.142201834862384,
+    "test_average_metrics": 58.142201834862384,
+    "test_loss": 0.2752912938594818,
+    "test_runtime": 3.1391,
+    "test_samples_per_second": 277.783,
+    "train_loss": 0.5804944922785328,
+    "train_runtime": 1254.7674,
     "train_samples": 66349,
-    "train_samples_per_second": 101.24,
-    "train_steps_per_second": 3.165
+    "train_samples_per_second": 158.633,
+    "train_steps_per_second": 4.959
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "eval_accuracy": 97.5,
-    "eval_average_metrics": 97.5,
-    "eval_loss": 0.03714486584067345,
-    "eval_runtime": 5.7314,
-    "eval_samples_per_second": 174.477
+    "eval_accuracy": 60.099999999999994,
+    "eval_average_metrics": 60.099999999999994,
+    "eval_loss": 0.2783910036087036,
+    "eval_runtime": 3.2954,
+    "eval_samples_per_second": 303.458
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:313a8aa2ad882fe8925fb538257ec103af71608407bbe84799eb3b904abaa5aa
+oid sha256:ca578edc26c679322afc4e794b748861809ce462f741454b4a5d1327e399320f
 size 335851

runs/Feb04_09-56-16_node1/events.out.tfevents.1643940000.node1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a091178aa1969a4b4a220994b83ce23c3e06f9259d106e3b6071afc59671639a
-size 15536
+oid sha256:19b60bae55358e79ee8cc62539e971324e17914693622f66af818a2d245fce4b
+size 16534

runs/Feb04_09-56-16_node1/events.out.tfevents.1643941260.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3fbc496fbb0b017946d09a318fa8b81191c585c2053d8ff1f9d5474c4e9c1776
+size 684

runs/Feb04_10-46-17_node1/1643942904.4796145/events.out.tfevents.1643942904.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e663a68a3ea6d4ba8bf840e42067622d741437998789d2407af56650d036e12f
+size 5026

runs/Feb04_10-46-17_node1/events.out.tfevents.1643942904.node1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f95558cca914d35629334a365c56bd84a73822d96a0efeb17cb5e14af2e46bb
+size 4312

test_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "test_accuracy": 93.34862385321101,
-    "test_average_metrics": 93.34862385321101,
-    "test_loss": 0.07569558918476105,
-    "test_runtime": 5.6934,
-    "test_samples_per_second": 153.159
+    "test_accuracy": 58.142201834862384,
+    "test_average_metrics": 58.142201834862384,
+    "test_loss": 0.2752912938594818,
+    "test_runtime": 3.1391,
+    "test_samples_per_second": 277.783
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.0,
-    "train_loss": 0.058686515289424354,
-    "train_runtime": 1966.0928,
+    "train_loss": 0.5804944922785328,
+    "train_runtime": 1254.7674,
     "train_samples": 66349,
-    "train_samples_per_second": 101.24,
-    "train_steps_per_second": 3.165
+    "train_samples_per_second": 158.633,
+    "train_steps_per_second": 4.959
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 97.5,
-  "best_model_checkpoint": "outputs/bitfit/t5-base/sst2/checkpoint-3000",
+  "best_metric": 60.099999999999994,
+  "best_model_checkpoint": "outputs/soft_prompt/t5-base/sst2/checkpoint-5800",
   "epoch": 3.0,
   "global_step": 6222,
   "is_hyper_param_search": false,
@@ -9,368 +9,368 @@
   "log_history": [
     {
       "epoch": 0.1,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.04369654133915901,
-      "eval_runtime": 5.1897,
-      "eval_samples_per_second": 192.69,
+      "eval_accuracy": 0.0,
+      "eval_average_metrics": 0.0,
+      "eval_loss": 1.9322336912155151,
+      "eval_runtime": 9.7016,
+      "eval_samples_per_second": 103.076,
       "step": 200
     },
     {
       "epoch": 0.19,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.042765092104673386,
-      "eval_runtime": 5.6341,
-      "eval_samples_per_second": 177.492,
+      "eval_accuracy": 51.0,
+      "eval_average_metrics": 51.0,
+      "eval_loss": 0.7396745681762695,
+      "eval_runtime": 3.3022,
+      "eval_samples_per_second": 302.832,
       "step": 400
     },
     {
       "epoch": 0.24,
       "learning_rate": 0.00027589199614271937,
-      "loss": 0.1281,
+      "loss": 2.5059,
       "step": 500
     },
     {
       "epoch": 0.29,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.037586040794849396,
-      "eval_runtime": 6.2353,
-      "eval_samples_per_second": 160.377,
+      "eval_accuracy": 55.2,
+      "eval_average_metrics": 55.2,
+      "eval_loss": 0.5560445785522461,
+      "eval_runtime": 3.2155,
+      "eval_samples_per_second": 310.993,
       "step": 600
     },
     {
       "epoch": 0.39,
-      "eval_accuracy": 96.7,
-      "eval_average_metrics": 96.7,
-      "eval_loss": 0.04588627442717552,
-      "eval_runtime": 5.966,
-      "eval_samples_per_second": 167.617,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.463209867477417,
+      "eval_runtime": 3.1314,
+      "eval_samples_per_second": 319.345,
       "step": 800
     },
     {
       "epoch": 0.48,
       "learning_rate": 0.0002517839922854387,
-      "loss": 0.0649,
+      "loss": 0.7329,
       "step": 1000
     },
     {
       "epoch": 0.48,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.040396977216005325,
-      "eval_runtime": 5.3919,
-      "eval_samples_per_second": 185.463,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.4116092920303345,
+      "eval_runtime": 3.3786,
+      "eval_samples_per_second": 295.977,
       "step": 1000
     },
     {
       "epoch": 0.58,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.03829416632652283,
-      "eval_runtime": 5.3561,
-      "eval_samples_per_second": 186.702,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.38179224729537964,
+      "eval_runtime": 3.2857,
+      "eval_samples_per_second": 304.351,
       "step": 1200
     },
     {
       "epoch": 0.68,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.039076462388038635,
-      "eval_runtime": 5.9241,
-      "eval_samples_per_second": 168.803,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.35022690892219543,
+      "eval_runtime": 2.8293,
+      "eval_samples_per_second": 353.443,
       "step": 1400
     },
     {
       "epoch": 0.72,
       "learning_rate": 0.00022767598842815813,
-      "loss": 0.0578,
+      "loss": 0.4828,
       "step": 1500
     },
     {
       "epoch": 0.77,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.04230912774801254,
-      "eval_runtime": 5.153,
-      "eval_samples_per_second": 194.06,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.3595990538597107,
+      "eval_runtime": 3.2567,
+      "eval_samples_per_second": 307.055,
       "step": 1600
     },
     {
       "epoch": 0.87,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.03823951631784439,
-      "eval_runtime": 5.1279,
-      "eval_samples_per_second": 195.011,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.3511555790901184,
+      "eval_runtime": 3.0816,
+      "eval_samples_per_second": 324.506,
       "step": 1800
     },
     {
       "epoch": 0.96,
       "learning_rate": 0.00020356798457087753,
-      "loss": 0.0582,
+      "loss": 0.421,
       "step": 2000
     },
     {
       "epoch": 0.96,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.0395108200609684,
-      "eval_runtime": 5.8042,
-      "eval_samples_per_second": 172.29,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.3314475417137146,
+      "eval_runtime": 2.522,
+      "eval_samples_per_second": 396.513,
       "step": 2000
     },
     {
       "epoch": 1.06,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.04088559374213219,
-      "eval_runtime": 5.4894,
-      "eval_samples_per_second": 182.168,
+      "eval_accuracy": 55.300000000000004,
+      "eval_average_metrics": 55.300000000000004,
+      "eval_loss": 0.3108561336994171,
+      "eval_runtime": 2.7063,
+      "eval_samples_per_second": 369.512,
       "step": 2200
     },
     {
       "epoch": 1.16,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.038739945739507675,
-      "eval_runtime": 4.6293,
-      "eval_samples_per_second": 216.018,
+      "eval_accuracy": 55.7,
+      "eval_average_metrics": 55.7,
+      "eval_loss": 0.2974016070365906,
+      "eval_runtime": 3.5599,
+      "eval_samples_per_second": 280.91,
       "step": 2400
     },
     {
       "epoch": 1.21,
       "learning_rate": 0.0001794599807135969,
-      "loss": 0.054,
+      "loss": 0.3931,
       "step": 2500
     },
     {
       "epoch": 1.25,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.04564524069428444,
-      "eval_runtime": 5.2048,
-      "eval_samples_per_second": 192.132,
+      "eval_accuracy": 55.7,
+      "eval_average_metrics": 55.7,
+      "eval_loss": 0.3047039210796356,
+      "eval_runtime": 3.0837,
+      "eval_samples_per_second": 324.282,
       "step": 2600
     },
     {
       "epoch": 1.35,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.03781759738922119,
-      "eval_runtime": 5.6194,
-      "eval_samples_per_second": 177.956,
+      "eval_accuracy": 55.900000000000006,
+      "eval_average_metrics": 55.900000000000006,
+      "eval_loss": 0.2895060181617737,
+      "eval_runtime": 3.9994,
+      "eval_samples_per_second": 250.039,
       "step": 2800
     },
     {
       "epoch": 1.45,
       "learning_rate": 0.00015535197685631627,
-      "loss": 0.0533,
+      "loss": 0.3777,
       "step": 3000
     },
     {
       "epoch": 1.45,
-      "eval_accuracy": 97.5,
-      "eval_average_metrics": 97.5,
-      "eval_loss": 0.03714486584067345,
-      "eval_runtime": 5.328,
-      "eval_samples_per_second": 187.688,
+      "eval_accuracy": 56.2,
+      "eval_average_metrics": 56.2,
+      "eval_loss": 0.28209826350212097,
+      "eval_runtime": 3.5924,
+      "eval_samples_per_second": 278.366,
       "step": 3000
     },
     {
       "epoch": 1.54,
-      "eval_accuracy": 97.5,
-      "eval_average_metrics": 97.5,
-      "eval_loss": 0.0382937453687191,
-      "eval_runtime": 4.0581,
-      "eval_samples_per_second": 246.419,
+      "eval_accuracy": 57.4,
+      "eval_average_metrics": 57.4,
+      "eval_loss": 0.2806238532066345,
+      "eval_runtime": 3.1947,
+      "eval_samples_per_second": 313.019,
       "step": 3200
     },
     {
       "epoch": 1.64,
-      "eval_accuracy": 97.5,
-      "eval_average_metrics": 97.5,
-      "eval_loss": 0.037162039428949356,
-      "eval_runtime": 5.3134,
-      "eval_samples_per_second": 188.202,
+      "eval_accuracy": 56.699999999999996,
+      "eval_average_metrics": 56.699999999999996,
+      "eval_loss": 0.2809857428073883,
+      "eval_runtime": 3.3046,
+      "eval_samples_per_second": 302.612,
       "step": 3400
     },
     {
       "epoch": 1.69,
       "learning_rate": 0.00013124397299903566,
-      "loss": 0.0539,
+      "loss": 0.372,
       "step": 3500
     },
     {
       "epoch": 1.74,
-      "eval_accuracy": 97.5,
-      "eval_average_metrics": 97.5,
-      "eval_loss": 0.03954707458615303,
-      "eval_runtime": 5.9646,
-      "eval_samples_per_second": 167.655,
+      "eval_accuracy": 56.89999999999999,
+      "eval_average_metrics": 56.89999999999999,
+      "eval_loss": 0.2799268066883087,
+      "eval_runtime": 3.5652,
+      "eval_samples_per_second": 280.491,
       "step": 3600
     },
     {
       "epoch": 1.83,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.037706729024648666,
-      "eval_runtime": 4.5321,
-      "eval_samples_per_second": 220.649,
+      "eval_accuracy": 56.599999999999994,
+      "eval_average_metrics": 56.599999999999994,
+      "eval_loss": 0.2795256972312927,
+      "eval_runtime": 3.5541,
+      "eval_samples_per_second": 281.363,
       "step": 3800
     },
     {
       "epoch": 1.93,
       "learning_rate": 0.00010713596914175504,
-      "loss": 0.0531,
+      "loss": 0.3629,
       "step": 4000
     },
     {
       "epoch": 1.93,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.03779396042227745,
-      "eval_runtime": 6.6618,
-      "eval_samples_per_second": 150.108,
+      "eval_accuracy": 57.099999999999994,
+      "eval_average_metrics": 57.099999999999994,
+      "eval_loss": 0.27900928258895874,
+      "eval_runtime": 3.42,
+      "eval_samples_per_second": 292.396,
       "step": 4000
     },
     {
       "epoch": 2.03,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.04000015929341316,
-      "eval_runtime": 6.7004,
-      "eval_samples_per_second": 149.245,
+      "eval_accuracy": 57.699999999999996,
+      "eval_average_metrics": 57.699999999999996,
+      "eval_loss": 0.27878034114837646,
+      "eval_runtime": 3.38,
+      "eval_samples_per_second": 295.859,
       "step": 4200
     },
     {
       "epoch": 2.12,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.0396127812564373,
-      "eval_runtime": 5.9828,
-      "eval_samples_per_second": 167.145,
+      "eval_accuracy": 57.699999999999996,
+      "eval_average_metrics": 57.699999999999996,
+      "eval_loss": 0.27947279810905457,
+      "eval_runtime": 3.1427,
+      "eval_samples_per_second": 318.196,
       "step": 4400
     },
     {
       "epoch": 2.17,
       "learning_rate": 8.302796528447444e-05,
-      "loss": 0.049,
+      "loss": 0.3597,
       "step": 4500
     },
     {
       "epoch": 2.22,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.0382530614733696,
-      "eval_runtime": 5.4179,
-      "eval_samples_per_second": 184.574,
+      "eval_accuracy": 57.99999999999999,
+      "eval_average_metrics": 57.99999999999999,
+      "eval_loss": 0.2773045599460602,
+      "eval_runtime": 3.1388,
+      "eval_samples_per_second": 318.593,
       "step": 4600
     },
     {
       "epoch": 2.31,
-      "eval_accuracy": 97.0,
-      "eval_average_metrics": 97.0,
-      "eval_loss": 0.04122977331280708,
-      "eval_runtime": 4.9766,
-      "eval_samples_per_second": 200.94,
+      "eval_accuracy": 59.8,
+      "eval_average_metrics": 59.8,
+      "eval_loss": 0.2791491448879242,
+      "eval_runtime": 3.8832,
+      "eval_samples_per_second": 257.519,
       "step": 4800
     },
     {
       "epoch": 2.41,
       "learning_rate": 5.891996142719383e-05,
-      "loss": 0.0484,
+      "loss": 0.3555,
       "step": 5000
     },
     {
       "epoch": 2.41,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.03976716473698616,
-      "eval_runtime": 5.1121,
-      "eval_samples_per_second": 195.615,
+      "eval_accuracy": 59.4,
+      "eval_average_metrics": 59.4,
+      "eval_loss": 0.2792259454727173,
+      "eval_runtime": 3.3994,
+      "eval_samples_per_second": 294.169,
       "step": 5000
     },
     {
       "epoch": 2.51,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.03887654095888138,
-      "eval_runtime": 5.3028,
-      "eval_samples_per_second": 188.578,
+      "eval_accuracy": 59.3,
+      "eval_average_metrics": 59.3,
+      "eval_loss": 0.27831217646598816,
+      "eval_runtime": 3.2184,
+      "eval_samples_per_second": 310.718,
       "step": 5200
     },
     {
       "epoch": 2.6,
-      "eval_accuracy": 97.3,
-      "eval_average_metrics": 97.3,
-      "eval_loss": 0.03931749612092972,
-      "eval_runtime": 6.019,
-      "eval_samples_per_second": 166.14,
+      "eval_accuracy": 59.199999999999996,
+      "eval_average_metrics": 59.199999999999996,
+      "eval_loss": 0.2772988975048065,
+      "eval_runtime": 3.2951,
+      "eval_samples_per_second": 303.485,
       "step": 5400
     },
     {
       "epoch": 2.65,
       "learning_rate": 3.481195756991321e-05,
-      "loss": 0.0478,
+      "loss": 0.3528,
       "step": 5500
     },
     {
       "epoch": 2.7,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.03971054032444954,
-      "eval_runtime": 5.7951,
-      "eval_samples_per_second": 172.558,
+      "eval_accuracy": 59.9,
+      "eval_average_metrics": 59.9,
+      "eval_loss": 0.2777373790740967,
+      "eval_runtime": 3.3995,
+      "eval_samples_per_second": 294.161,
       "step": 5600
     },
     {
       "epoch": 2.8,
-      "eval_accuracy": 97.1,
-      "eval_average_metrics": 97.1,
-      "eval_loss": 0.04016176983714104,
-      "eval_runtime": 5.5419,
-      "eval_samples_per_second": 180.443,
+      "eval_accuracy": 60.099999999999994,
+      "eval_average_metrics": 60.099999999999994,
+      "eval_loss": 0.2783910036087036,
+      "eval_runtime": 3.8322,
+      "eval_samples_per_second": 260.948,
       "step": 5800
     },
     {
       "epoch": 2.89,
       "learning_rate": 1.0703953712632592e-05,
-      "loss": 0.0444,
+      "loss": 0.3503,
       "step": 6000
     },
     {
       "epoch": 2.89,
-      "eval_accuracy": 97.0,
-      "eval_average_metrics": 97.0,
-      "eval_loss": 0.04050704091787338,
-      "eval_runtime": 6.1769,
-      "eval_samples_per_second": 161.894,
+      "eval_accuracy": 59.699999999999996,
+      "eval_average_metrics": 59.699999999999996,
+      "eval_loss": 0.27744776010513306,
+      "eval_runtime": 3.3726,
+      "eval_samples_per_second": 296.511,
       "step": 6000
     },
     {
       "epoch": 2.99,
-      "eval_accuracy": 97.2,
-      "eval_average_metrics": 97.2,
-      "eval_loss": 0.04039543867111206,
-      "eval_runtime": 5.4148,
-      "eval_samples_per_second": 184.68,
+      "eval_accuracy": 59.9,
+      "eval_average_metrics": 59.9,
+      "eval_loss": 0.27743667364120483,
+      "eval_runtime": 3.6434,
+      "eval_samples_per_second": 274.47,
       "step": 6200
     },
     {
       "epoch": 3.0,
       "step": 6222,
-      "total_flos": 1.3280955617597184e+16,
-      "train_loss": 0.058686515289424354,
-      "train_runtime": 1966.0928,
-      "train_samples_per_second": 101.24,
-      "train_steps_per_second": 3.165
+      "total_flos": 1.316605032958464e+16,
+      "train_loss": 0.5804944922785328,
+      "train_runtime": 1254.7674,
+      "train_samples_per_second": 158.633,
+      "train_steps_per_second": 4.959
     }
   ],
   "max_steps": 6222,
   "num_train_epochs": 3,
-  "total_flos": 1.3280955617597184e+16,
+  "total_flos": 1.316605032958464e+16,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f396571923132052a008693cef3a6acfe656e9be0d56e2e10ea73893ffe1b46
+oid sha256:011028b4522e6a361729bdf6448cb8ad949df04a35d5783ce734197da2e28eb4
 size 3183

training_config.json

@@ -1 +1 @@
-{"dataset_config_name": ["en"], "delta_type": "soft_prompt", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "sst2", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.0003, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 3, "output_dir": "outputs/soft_prompt/t5-base/sst2", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "soft_token_num": 100, "split_validation_test": true, "task_name": "sst2", "test_dataset_config_name": ["en"], "test_dataset_name": "sst2", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas"], "warmup_steps": 0}
+{"dataset_config_name": ["en"], "delta_type": "soft_prompt", "do_eval": true, "do_test": true, "do_train": true, "eval_dataset_config_name": ["en"], "eval_dataset_name": "sst2", "eval_steps": 200, "evaluation_strategy": "steps", "greater_is_better": true, "learning_rate": 0.03, "load_best_model_at_end": true, "max_source_length": 128, "metric_for_best_model": "average_metrics", "model_name_or_path": "../../../../plm_cache/t5-base", "num_train_epochs": 3, "output_dir": "outputs/soft_prompt/t5-base/sst2", "overwrite_output_dir": true, "per_device_eval_batch_size": 32, "per_device_train_batch_size": 32, "predict_with_generate": true, "push_to_hub": true, "save_steps": 200, "save_strategy": "steps", "save_total_limit": 1, "seed": 42, "soft_token_num": 100, "split_validation_test": true, "task_name": "sst2", "test_dataset_config_name": ["en"], "test_dataset_name": "sst2", "tokenizer_name": "../../../../plm_cache/t5-base", "unfrozen_modules": ["deltas"], "warmup_steps": 0}