Training in progress, epoch 1

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7220216606498195,
-    "eval_loss": 0.5765948295593262,
-    "eval_runtime": 0.2028,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6930166482925415,
+    "eval_runtime": 0.2259,
     "eval_samples": 277,
-    "eval_samples_per_second": 1365.807,
-    "eval_steps_per_second": 14.792,
-    "train_loss": 0.5880965404021434,
-    "train_runtime": 74.518,
+    "eval_samples_per_second": 1226.394,
+    "eval_steps_per_second": 13.282,
+    "train_loss": 0.6997218401004106,
+    "train_runtime": 107.0767,
     "train_samples": 2490,
-    "train_samples_per_second": 334.147,
-    "train_steps_per_second": 10.467
+    "train_samples_per_second": 232.544,
+    "train_steps_per_second": 7.284
 }

config.json

@@ -1,5 +1,5 @@
 {
-  "_name_or_path": "../../../../plm_cache/roberta-base",
+  "_name_or_path": "../../../../plm_cache/roberta-large",
   "architectures": [
     "RobertaForSequenceClassification"
   ],
@@ -10,13 +10,13 @@
   "finetuning_task": "rte",
   "hidden_act": "gelu",
   "hidden_dropout_prob": 0.1,
-  "hidden_size": 768,
+  "hidden_size": 1024,
   "id2label": {
     "0": "entailment",
     "1": "not_entailment"
   },
   "initializer_range": 0.02,
-  "intermediate_size": 3072,
+  "intermediate_size": 4096,
   "label2id": {
     "entailment": 0,
     "not_entailment": 1
@@ -24,8 +24,8 @@
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 514,
   "model_type": "roberta",
-  "num_attention_heads": 12,
-  "num_hidden_layers": 12,
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
   "pad_token_id": 1,
   "position_embedding_type": "absolute",
   "problem_type": "single_label_classification",

eval_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 10.0,
-    "eval_accuracy": 0.7220216606498195,
-    "eval_loss": 0.5765948295593262,
-    "eval_runtime": 0.2028,
+    "eval_accuracy": 0.5270758122743683,
+    "eval_loss": 0.6930166482925415,
+    "eval_runtime": 0.2259,
     "eval_samples": 277,
-    "eval_samples_per_second": 1365.807,
-    "eval_steps_per_second": 14.792
+    "eval_samples_per_second": 1226.394,
+    "eval_steps_per_second": 13.282
 }

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a89c652f2d16b97921600702b2a9695bef86f5720d7a2384eaf6268f5196a9b2
-size 2704633
+oid sha256:bc154e75055ed4ffe4f6fee8eb3b2c83fd827575bcbab6fd3b65b47693d12588
+size 17106681

runs/Jan29_15-48-48_node4/events.out.tfevents.1643442552.node4

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4a9718edfa67d6d7154739ab7868ff2be8fa78bf35e6f7ba4849484bd95bde41
-size 6573
+oid sha256:61d693dfa0eb2f0e4882ba26a780134f70dae796a68b3cdf5bc8eaf2a81b26cc
+size 7250

runs/Jan29_15-48-48_node4/events.out.tfevents.1643442660.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a37a54639e8f663b9e48afc468172b55940869c6f34d503fa8f0e51752ff238d
+size 363

runs/Jan29_16-31-17_node4/1643445112.230297/events.out.tfevents.1643445112.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82d225b734cb18ede058b02137259711497603f297b73c1984d9fc069f368086
+size 4585

runs/Jan29_16-31-17_node4/events.out.tfevents.1643445112.node4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:583a5a78f47360afe9e409cd945eeec5395fbe3831f7bd1c7cec649c5bd82552
+size 3517

tokenizer_config.json

@@ -1 +1 @@
-{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "special_tokens_map_file": null, "name_or_path": "../../../../plm_cache/roberta-base", "tokenizer_class": "RobertaTokenizer"}
+{"unk_token": "<unk>", "bos_token": "<s>", "eos_token": "</s>", "add_prefix_space": false, "errors": "replace", "sep_token": "</s>", "cls_token": "<s>", "pad_token": "<pad>", "mask_token": "<mask>", "trim_offsets": true, "special_tokens_map_file": null, "name_or_path": "../../../../plm_cache/roberta-large", "tokenizer_class": "RobertaTokenizer"}

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 10.0,
-    "train_loss": 0.5880965404021434,
-    "train_runtime": 74.518,
+    "train_loss": 0.6997218401004106,
+    "train_runtime": 107.0767,
     "train_samples": 2490,
-    "train_samples_per_second": 334.147,
-    "train_steps_per_second": 10.467
+    "train_samples_per_second": 232.544,
+    "train_steps_per_second": 7.284
 }

trainer_state.json

@@ -1,6 +1,6 @@
 {
-  "best_metric": 0.7220216606498195,
-  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-780",
+  "best_metric": 0.5270758122743683,
+  "best_model_checkpoint": "outputs/lora/roberta-base/v2/rte/checkpoint-234",
   "epoch": 10.0,
   "global_step": 780,
   "is_hyper_param_search": false,
@@ -10,112 +10,112 @@
     {
       "epoch": 1.0,
       "eval_accuracy": 0.4729241877256318,
-      "eval_loss": 0.7042641639709473,
-      "eval_runtime": 0.2314,
-      "eval_samples_per_second": 1197.187,
-      "eval_steps_per_second": 12.966,
+      "eval_loss": 0.7130928039550781,
+      "eval_runtime": 0.2073,
+      "eval_samples_per_second": 1336.057,
+      "eval_steps_per_second": 14.47,
       "step": 78
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.5776173285198556,
-      "eval_loss": 0.6811862587928772,
-      "eval_runtime": 0.2009,
-      "eval_samples_per_second": 1378.756,
-      "eval_steps_per_second": 14.932,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6993798613548279,
+      "eval_runtime": 0.2002,
+      "eval_samples_per_second": 1383.767,
+      "eval_steps_per_second": 14.987,
       "step": 156
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.6028880866425993,
-      "eval_loss": 0.6594843864440918,
-      "eval_runtime": 0.2361,
-      "eval_samples_per_second": 1173.317,
-      "eval_steps_per_second": 12.707,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6930166482925415,
+      "eval_runtime": 0.2199,
+      "eval_samples_per_second": 1259.694,
+      "eval_steps_per_second": 13.643,
       "step": 234
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.6714801444043321,
-      "eval_loss": 0.6216505765914917,
-      "eval_runtime": 0.1961,
-      "eval_samples_per_second": 1412.672,
-      "eval_steps_per_second": 15.3,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6923182606697083,
+      "eval_runtime": 0.201,
+      "eval_samples_per_second": 1378.119,
+      "eval_steps_per_second": 14.925,
       "step": 312
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.6787003610108303,
-      "eval_loss": 0.6077037453651428,
-      "eval_runtime": 0.2048,
-      "eval_samples_per_second": 1352.477,
-      "eval_steps_per_second": 14.648,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.696506917476654,
+      "eval_runtime": 0.2195,
+      "eval_samples_per_second": 1262.144,
+      "eval_steps_per_second": 13.669,
       "step": 390
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.6642599277978339,
-      "eval_loss": 0.6047545075416565,
-      "eval_runtime": 0.1975,
-      "eval_samples_per_second": 1402.222,
-      "eval_steps_per_second": 15.187,
+      "eval_accuracy": 0.5270758122743683,
+      "eval_loss": 0.6918743252754211,
+      "eval_runtime": 0.2123,
+      "eval_samples_per_second": 1304.889,
+      "eval_steps_per_second": 14.132,
       "step": 468
     },
     {
       "epoch": 6.41,
-      "learning_rate": 0.00019099590723055935,
-      "loss": 0.6366,
+      "learning_rate": 0.001145975443383356,
+      "loss": 0.7031,
       "step": 500
     },
     {
       "epoch": 7.0,
-      "eval_accuracy": 0.7111913357400722,
-      "eval_loss": 0.5592861771583557,
-      "eval_runtime": 0.1967,
-      "eval_samples_per_second": 1408.382,
-      "eval_steps_per_second": 15.253,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6935069561004639,
+      "eval_runtime": 0.2336,
+      "eval_samples_per_second": 1185.748,
+      "eval_steps_per_second": 12.842,
       "step": 546
     },
     {
       "epoch": 8.0,
-      "eval_accuracy": 0.703971119133574,
-      "eval_loss": 0.5585260987281799,
-      "eval_runtime": 0.2088,
-      "eval_samples_per_second": 1326.752,
-      "eval_steps_per_second": 14.369,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6938801407814026,
+      "eval_runtime": 0.2086,
+      "eval_samples_per_second": 1327.87,
+      "eval_steps_per_second": 14.381,
       "step": 624
     },
     {
       "epoch": 9.0,
-      "eval_accuracy": 0.7111913357400722,
-      "eval_loss": 0.571499764919281,
-      "eval_runtime": 0.2055,
-      "eval_samples_per_second": 1347.626,
-      "eval_steps_per_second": 14.595,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6936783194541931,
+      "eval_runtime": 0.2108,
+      "eval_samples_per_second": 1313.785,
+      "eval_steps_per_second": 14.229,
       "step": 702
     },
     {
       "epoch": 10.0,
-      "eval_accuracy": 0.7220216606498195,
-      "eval_loss": 0.5765948295593262,
-      "eval_runtime": 0.1974,
-      "eval_samples_per_second": 1403.091,
-      "eval_steps_per_second": 15.196,
+      "eval_accuracy": 0.4729241877256318,
+      "eval_loss": 0.6938663125038147,
+      "eval_runtime": 0.2046,
+      "eval_samples_per_second": 1353.687,
+      "eval_steps_per_second": 14.661,
       "step": 780
     },
     {
       "epoch": 10.0,
       "step": 780,
-      "total_flos": 1638956800972800.0,
-      "train_loss": 0.5880965404021434,
-      "train_runtime": 74.518,
-      "train_samples_per_second": 334.147,
-      "train_steps_per_second": 10.467
+      "total_flos": 1639048592332800.0,
+      "train_loss": 0.6997218401004106,
+      "train_runtime": 107.0767,
+      "train_samples_per_second": 232.544,
+      "train_steps_per_second": 7.284
     }
   ],
   "max_steps": 780,
   "num_train_epochs": 10,
-  "total_flos": 1638956800972800.0,
+  "total_flos": 1639048592332800.0,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d942a12835e627c5832563e175e275a68f55113feef1e6c7484e8dc0409bd4fa
+oid sha256:d3459cc4e350849119f9397e1e235b5df88dabf546417d4c780d5756a1941640
 size 2991